Biotechnology Products of Secondary Metabolism

Biotechnology
Second Edition
Volume 7
Products of Secondary Metabolism
VCH 4b
A Wiley company
Biotechnology
Second Edition
Fundamentals Special Topics
Volume 1 Volume 9
Biological Fundamentals Enzymes, Biomass, Food and Feed
Volume 2 Volume 10
Genetic Fundamentals and Special Processes
Genetic Engineering
Volumes l l a and b
Volume 3 Environmental Processes
Bioprocessing
Volume 12
Volume 4 Legal, Economic and
Measuring, Modelling, and Control Ethical Dimensions
Products
Volume 5
Recombinant Proteins,
Monoclonal Antibodies, and
Therapeutic Genes
Volume 6
Products of Primary Metabolism
Volume 7
Products of Secondary Metabolism
Volume 8
Biotransformations
A Multi-Volume Comprehensive Treatise
Biotechnology
Second, Completely Revised Edition
Edited by
H.-J. Rehm and G. Reed
in cooperation with
A. Piihler and P. Stadler
Volume 7
Products of
Secondary Metabolism
Edited by
H. Kleinkauf and H. von Dohren
VCH 4b
A Wiley company
Series Editors: Volume Editors:
Prof. Dr. H.-J. R eh m Dr. G. R e e d Prof. Dr. H. Kleinkauf
Institut f u r Mikrobiologie 1914 N. Prospect Ave. #61 Dr. H. von D o h r en
Universitat Munster Milwaukee, WI 53202-1401 Institut f u r Biochemie
CorrensstraBe 3 USA Technische Universitat
D-48149 Munster Franklin-StraBe 29
FRG A-10587 Berlin
Prof. Dr. P. J. W. Stadler Germany
Prof. Dr. A. Piihler Bayer AG
Biologie VI (Genetik) Verfahrensentwicklung Biochemie
Universitat Bielefeld Leitung
P.O. Box 100131 Friedrich-Ebert-StraBe 217
D-33501 Bielefeld D-42096 Wuppertal
FRG FRG
This book was carefully produced. Nevertheless, authors, editors and publisher do not warrant the information
contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations,
procedural details or other items may inadvertently be inaccurate.
Executive Editor: Dr. Hans-Joachim Kraus

Editorial Director: Karin Dembowsky
Production Manager: Hans-Jochen Schmitt
Library of Congress Card No.: applied for
British Library Cataloguing-in-Publication Data:

A catalogue record for this book is available from the British Library
Die Deutsche Bibliothek - CIP-Einheitsaufnahme

Biotechnology : a multi volume comprehensive
treatise I ed. by H.-J. Rehm and G. Reed. In
cooperation with A. Piihler and P. Stadler. -
2., completely rev. ed. -VCH.
ISBN 3-527-28310-2 (Weinheim ...)
NE: Rehm, Hans J. [Hrsg.]
Vol. 7. Products of secondary metabolism I ed. by H. Kleinkauf and H. von Dohren - 1997
ISBN 3-S27-28317-X
OVCH Verlagsgesellschaft mbH, D-69451 Weinheim (Federal Republic of Germany), 1997
Printed on acid-free and chlorine-free paper.
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any
form -by photoprinting, microfilm, or any other means-nor transmitted or translated into a machine language without
written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifi-
cally marked as such, are not to be considered unprotected by law.
Composition and Printing: Zechnersche Buchdruckerei, D-67330 Speyer.
Bookbinding: J. SchSiffer, D-67269 Griinstadt.
Printed in the Federal Republic of Germany
Preface
In recognition of the enormous advances in series. Its members come from key institu-
biotechnology in recent years, we are pleased tions representing scientific input from about
to present this Second Edition of “Biotech- twenty countries.
nology” relatively soon after the introduction The volume editors and the authors of the
of the First Edition of this multi-volume com- individual chapters have been chosen for
prehensive treatise. Since this series was ex- their recognized expertise and their contribu-
tremely well accepted by the scientific com- tions to the various fields of biotechnology.
munity, we have maintained the overall goal Their willingness to impart this knowledge to
of creating a number of volumes, each de- their colleagues forms the basis of “Biotech-
voted to a certain topic, which provide scien- nology” and is gratefully acknowledged.
tists in academia, industry, and public institu- Moreover, this work could not have been
tions with a well-balanced and comprehensive brought to fruition without the foresight and
overview of this growing field. We have fully the constant and diligent support of the pub-
revised the Second Edition and expanded it lisher. We are grateful to VCH for publishing
from ten to twelve volumes in order to take “Biotechnology” with their customary excel-
all recent developments into account. lence. Special thanks are due to Dr. Hans-
These twelve volumes are organized into Joachim Kraus and Karin Dembowsky, with-
three sections. The first four volumes consid- out whose constant efforts the series could
er the fundamentals of biotechnology from not be published. Finally, the editors wish to
biological, biochemical, molecular biological, thank the members of the Scientific Advisory
and chemical engineering perspectives. The Board for their encouragement, their helpful
next four volumes are devoted to products of suggestions, and their constructive criticism.
industrial relevance. Special attention is given
here to products derived from genetically en- H.-J. Rehm
gineered microorganisms and mammalian G. Reed
cells. The last four volumes are dedicated to A. Puhler
the description of special topics. P. Stadler
The new “Biotechnology” is a reference
work, a comprehensive description of the
state-of-the-art, and a guide to the original
literature. It is specifically directed to micro-
biologists, biochemists, molecular biologists,
bioengineers, chemical engineers, and food
and pharmaceutical chemists working in indus-
try, at universities or at public institutions.
A carefully selected and distinguished
Scientific Advisory Board stands behind the
Scientific Advisory Board
Pro$ Dr. M. J. Beker Pro$ Dr. T. K. Ghose

August Kirchenstein Institute of Microbiology Biochemical Engineering Research Centre
Latvian Academy of Sciences Indian Institute of Technology
Riga, Latvia New Delhi, India
Pro$ Dr. J. D. Bu’Lock Pro$ Dr. I. Goldberg

Weizmann Microbial Chemistry Laboratory Department of Applied Microbiology
Department of Chemistry The Hebrew University
University of Manchester Jerusalem, Israel
Manchester, UK
Pro$ Dr. C. L. Cooney Pro$ Dr. G. Goma

Department of Chemical Engineering Departement de GCnie Biochimique et
Massachusetts Institute of Technology Alimentaire
Cambridge, MA, USA Institut National des Sciences AppliquCes
Toulouse, France
Pro$ Dr. H. W. Doelle Sir D. A. Hopwood

Department of Microbiology Department of Genetics
University of Queensland John Innes Institute
St. Lucia, Australia Norwich, UK
Prof Dr. J. Drews Pro$ Dr. E. H. Houwink

F. Hoffmann-La Roche AG Organon International bv
Basel, Switzerland Scientific Development Group
Oss. The Netherlands
Pro$ Dr. A. Fiechter Pro$ Dr. A. E. Humphrey

Institut fur Biotechnologie Center for Molecular Bioscience and
Eidgenossische Technische Hochschule Biotechnology
Zurich, Switzerland Lehigh University
Bethlehem, PA, USA
VIII Scientific Advisory Board
Prof. Dr. I. Karube Prof. Dr. K. Schiigerl

Research Center for Advanced Science Institut fur Technische Chemie
and Technology Universitat Hannover
University of Tokyo Hannover, Germany
Tokyo, Japan
Prof. Dr. M . A. Lachance Prof. Dr. P. Sensi

Department of Plant Sciences Chair of Fermentation Chemistry
University of Western Ontario and Industrial Microbiology
London, Ontario, Canada Lepetit Research Center
Gerenzano, Italy
Prof. Dr. Y. Liu Prof. Dr. Y. H. Tan

China National Center for Biotechnology Institute of Molecular and Cell Biology
Development National University of Singapore
Beijing, China Singapore
Prof. Dr. J. F. Martin Prof. Dr. D. Thomas

Department of Microbiology Laboratoire de Technologie Enzymatique
University of Leon UniversitC de Compibgne
Leon, Spain Compibgne, France
Prof. Dr. B. Mattiasson Prof. Dr. W . Verstraete

Department of Biotechnology Laboratory of Microbial Ecology
Chemical Center Rijksuniversiteit Gent
University of Lund Gent, Belgium
Lund, Sweden
Prof. Dr. M . Roehr Prof. Dr. E.-L. Winnacker

Institut fur Biochemische Technologie Institut fur Biochemie
und Mikrobiologie Universitat Munchen
Technische Universitat Wien Munchen, Germany
Wien, Austria
Prof. Dr. H. Sahm

Institut fur Biotechnologie
Forschungszentrum Julich
Julich, Germany
Contents
Introduction 9 Glycopeptide Antibiotics

H. von Dohren, H. Kleinkauf (Dalbaheptides) 369
G. Lancini, B. Cavalleri
General Aspects of Secondary 10 Aminoglycosides and Sugar
Metabolism 1 Components in Other Secondary
H. von Dohren, U. Grafe Metabolites 397
Regulation of Bacterial Antibiotic W. Piepersberg, J. Distler
Production 57 11 Products from Basidiomycetes 489
K. Chater, M. Bibb G. Erkel, T. Anke
Screening of Novel Receptor-Active 12 Cyclosporins: Recent Developments in
Compounds of Microbial Origin 107 Biosynthesis, Pharmacology and
H. Tanaka, S. Omura Biology, and Clinical Applications 535
Microbial Lipids 133 J. Kallen, V. Mikol, V. F. J. Quesniaux,
C. Ratledge M. D. Walkinshaw,E. Schneider-Scherzer,
Microbial Siderophores 199 K. Schorgendorfer, G. Weber, H. Fliri
G. Winkelmann, H. Drechsel 13 Secondary Products from Plant Cell
Advances in the Molecular Genetics of Cultures 593
PLactam Antibiotic Biosynthesis 247 J. Berlin
P. 15.Skacrud, T. Schwecke, 14 Biotechnical Drugs as Antitumor
H. v. Liempt, M. B. Tobin Agents 641
Peptide Antibiotics 277 U. Grafe, K . Dornberger, H.-P. Saluz
H. Kleinkauj H. von Dohren
Lantibiotics 323 Index 707
R. Jack, F. Gotz, G. Jung
Contributors
Prof. Dr. Timm Anke Prof. Keith Chater

Lehrbereich Biotechnologie John Innes Centre
Universitat Kaiserslautern Norwich Research Park
Postfach 3049 Colney Lane
D-67618 Kaiserslautern Colney, Norwich NR4 7UH
Germany UK
Chapter I1 Chapter 2
Dr. Jochen Berlin Jurgen Distler

Gesellschaft fur Biotechnologische Bergische Universitat G H
Forschung Mikrobiologie - FB 19
Mascheroder Weg 1 Gauss-StraSe 20
D-38124 Braunschweig D-42097 Wuppertal
Germany Germany
Chapter 13 Chapter 10
Dr. Mervin Bibb Dr. Hans von Dohren

John Innes Centre Institut fur Biochemie
Norwich Research Park Technische Universitat
Colney Lane Franklin-Str. 29
Colney, Norwich NR4 7UH D-10587 Berlin
UK Germany
Chapter 2 Chapters I, 7
Dr. Bruno Cavalleri Dr. Klausjiirgen Dornberger

MMDRI - Lepetit Research Center Hans-Knoll-Institut fur Naturstoff-Forschung
Via R. Lepetit, 34 Bereich Naturstoffchemie
1-21040 Gerenzano (Varese) BeutenbergstraBe 11
Italy D-07745 Jena
Chapter 9 Germany
Chapter 14
XI1 Contributors
Dr. Hartmut Drechsel Prof. Dr. Gunter Jung

Mikrobiologie und Biotechnologie Universitat Tubingen
Universitat Tubingen Institut fur Organische Chemie
Auf der Morgenstelle 1 Auf der Morgenstelle 18
D-72076 Tubingen D-72076 Tubingen
Germany Germany
Chapter 5 Chapter 8
Dr. Gerhard Erkel Dr. Jorg Kallen

Lehrbereich Biotechnologie Sandoz Pharma Ltd.
Universitat Kaiserslautern Preclinical Research
Postfach 3049 CH-4002 Basel
D-67618 Kaiserslautern Switzerland
Germany Chapter I2
Chapter I1
Dr. Hans Fliri Prof. Dr. Horst Kleinkauf

RhBne Poulonc Rorer S.A. Institut fur Biochemie
Centre de Recherche de Vitry-Alfortville Technische Universitat
13, quai Jules Guesde Franklin-Str. 29
F-94403 Vitry-sur-Seine Cedex D-10587 Berlin
France Germany
Dr. Friedrich Gotz Dr. Giancarlo Lancini

Lehrstuhl fur Mikrobielle Genetik MMDRI - Lepetit Research Center
Universitat Tubingen Via R. Lepetit, 34
Auf der Morgenstelle 18 1-21040 Gerenzano (Varese)
D-72076 Tubingen Italy
Germany Chapter 9
Chapter 8
Prof. Dr. Udo Grafe Dr. Henk van Liempt

Hans-Knoll-Institut fur Naturstoff-Forschung DRL, BT-FDG
Bereich Naturstoffchemie SudstraSe 125
BeutenbergstraSe 11 D-53175 Bonn
D-07745 Jena Germany
Germany Chapter 6
Chapters I, 14
Dr. Ralph Jack Dr. Vincent Mikol

Universitat Tubingen Sandoz Pharma Ltd.
Institut fur Organische Chemie Preclinical Research
Auf der Morgenstelle 18 CH-4002 Basel
D-72076 Tubingen Switzerland
Germany Chapter 12
Chapter 8
Contributors XI11
Prof. Dr. Satoshi Omura Dr. Kurt Schorgendorfer

School of Pharmaceutical Sciences Biochemie GmbH
Kitasato University A-6330 Kufstein-Schaftenau
The Kitasato Institute Austria
9-1, Shirokane 5-chome Chapter 12
Minato-ku, Tokyo 108
Japan
Chapter 3
Prof. Dr. Wolfgang Piepersberg Dr. Torsten Schwecke

Bergische Universitat G H Institute of Biochemistry
Mikrobiologie - FB 19 University of Cambridge
Gauss-Strafie 20 Tennis Court Road
D-42097 Wuppertal Cambridge, CB2 1QW
Germany UK
Dr. Valkrie F.J. Quesniaux Dr. Paul L. Skatrud

Sandoz Pharma Ltd. Infectious Diseases Research
Preclinical Research Eli Lilly and Company
CH-4002 Basel Lilly Corporate Center
Switzerland Indianapolis, IN 46285
Chapter 12 USA
Chapter 6
Prof. Colin Ratledge Dr. Haruo Tanaka

The University of Hull School of Pharmaceutical Sciences
Department of Applied Biology Kitasato University
Hull HU6 7RX The Kitasato Institute
UK Minato-ku, Tokyo 108
Chapter 4 Japan
Chapter 3
Prof. Dr. habil. Hans-Peter Saluz Dr. Matthew B. Tobin

Hans-Knoll-Institut fur Naturstoff-Forschung Infectious Diseases Research
Bereich Naturstoffchemie Eli Lilly and Company
BeutenbergstraSe 11 Lilly Corporate Center
D-07745 Jena Indianapolis, IN 46285
Germany USA
Dr. Elisabeth Schneider-Scherzer Dr. Malcolm D. Walkinshaw

Biochemie GmbH Sandoz Pharma Ltd.
A-6330 Kufstein-Schaftenau Preclinical Research
Austria CH-4002 Basel
Chapter 12 Switzerland
Chapter 12
XIV Contributors
Dr. Gerhard Weber Prof. Dr. Giinther Winkelmann

Biochemie GmbH Mikrobiologie und Biotechnologie
A4330 Kufstein-Schaftenau Universitat Tubingen
Austria Auf der Morgenstelle 1
Chapter 12 D-72076 Tubingen
Germany
Chapter 5
Biotechnology
Edited by H.-J. Rehm and G. Reed
OVCH Verlagsgesellschaft mbH, 1997
This volumes provides an overview of sec- Still largely unconnected to the back-
ondary metabolites illustrating most aspects ground of their producers secondary metabol-
of their discovery, formation, exploitation, ites generally are high-value compounds es-
and production. Compared to the first edition tablished mainly in pharmacology, veterinary
the focus when has clearly shifted towards the medicine, agriculture, and biochemical and
molecular genetic background of the produc- medical research. The introductory chapter
ing organisms. These efforts serve not only points to product fields and to the genetic in-
our understanding of the production proc- vestigation of biosynthetic unit operations.
esses to permit improvements by genetic ma- Regulatory mechanisms are then considered
nipulations, but also promote our apprecia- in the most advanced fields of the proka-
tion of the environmental significance of sec- ryotes. As the central field of present drug
ondary metabolites. discovery approaches target-based screenings
The term “secondary metabolite” has been are discussed. Compound groups considered
discussed widely, and a shift in perception are lipids siderophores, aminoglycosides, and
took place in the last years. From a play- peptides (p-lactams, dalbaheptides, cyclospo-
ground of nature leading to mostly disparable rins, lantibiotics). Producer groups presented
products ideas focus now on special purpose are basidiomycetes and plant cells. As a tar-
products promoting evolutionary advantages. get group antitumor drugs are evaluated.
This shift is connected to the impressive eluci- An updated chapter on macrolides as sec-
dation of the genetics of multistep synthetic ondary metabolites including reprogramming
processes of secondary metabolite formation. strategies will be included in Volume 10 of
Genes encoding biosynthetic reaction se- the Second Edition of Biofechnofogy(see also
quences have been found clustered together Volume 4 of the First Edition).
with resistance or export genes and are under Further chapters to be consulted are espe-
the control of specific signals. Biosynthetic cially on biopolymers and surfactants (Vol-
functions or unit operations reside on mod- ume 6), on the overproduction of metabolites
ules, and these modules in their functional and the treatment of producer organisms like
protein state interact to assure the fidelity of bacilli, streptomycetes and filamentous fungi
the multistep processes. The genetic burden (Volume 1) as well as on reactor modeling
for many of these processes seems remarka- (Volume 3). We thank our colleagues for
ble, and genes assembled from modules often their valuable contributions, the publisher for
display sizes of 10 to more than 45 kilobases. their patience and cooperativity, and the se-
Since some of the now established microbial ries editors for many helpful suggestions.
genomes are devoid of such multistep path-
ways, their unique placement in other ge-
nomes indicates important functions for their Berlin, March 1997 Hans von Dohren
producers. Horst Kleinkauf
Biotechnology
1 General Aspects of Secondary

Metabolism
HANSVON DOHREN
Berlin, Germany
UDOGRAFE
Jena, Germany
1 Introduction: The Importance of Secondary Metabolites as Drugs 2

2 Secondary Metabolism, an Expression of Cellular and Organismic Individuality 11
2.1 Roles of Secondary Metabolites in Producing Organisms 11
2.2 Regulation of Microbial Secondary Metabolism 17
2.2.1 Genetic Organization of Product Formation 17
2.2.2 Regulatory Mechanisms 23
2.2.3 Genetic Instability 26
2.2.4 Developmental Processes 27
2.2.5 The A-Factor and the Signal Cascade of Cytodifferentiation in Streptomyces 27
2.2.6 Overproduction of Microbial Secondary Metabolites and Precursor Pools 29
2.2.7 Biotechnical Production of Secondary Metabolites 31
3 The Biosynthetic Pathways 31
3.1 Precursors and the Main Biosynthetic Pathways 31
3.2 Secondary Metabolites Formed through Biosynthetic Modifications of a Single
Precursor 31
3.3 Polyketides 32
3.4 Terpenes 35
3.5 Sugar-Derived Oligomeric Structures 35
3.6 Oligo- and Polypeptides 36
3.7 Biosynthetic Modifications of Structures and Precursor-Directed Biosyntheses 37
4 Variability of Structures of Secondary Metabolites 38
4.1 Secondary Metabolites as Products of Biological “Unit Operations” 38
4.2 Structural Classifications of Secondary Metabolites 38
5 Future Perspectives: New Products of the Secondary Metabolism 40
6 References 41
2 1 General Aspects of Secondary Metabolism
1 Introduction: The crease in resistant nosocomial and opportu-

nistic pathogens (particularly dangerous to
Importance of Secondary immunosuppressed AIDS and tumor pa-
tients) requires both improvement of known
Metabolites as Drugs drugs and search for new drugs (GRAFE,
1992; LANCINIand LORENZETTI,1993; HUT-
CHINSON, 1994).
Today, bioactive secondary metabolites of Microbial products such as doxorubicin,
microorganisms and of plants, and their syn- bleomycin, and mitomycin C are indispensa-
thetic derivatives as well, are among the most ble as cancerostatics (Fox, 1991). The same is
frequently used therapeutics in human and true for plant metabolites such as the vinca al-
veterinary medicine (Scrip, 1993). The inven- kaloids, taxol, and their chemical derivatives
tion of antibiotic therapy contributed greatly which exert excellent antitumor activity by in-
to the successful control of most of the epi- teraction with the cellular mitotic system
demic infectious diseases and even promoted (NOBLE, 1990 Fox, 1991; HEINSTEINand
their disappearance. Moreover, it contributed CHANG,1994 POTIERet al., 1994).
to the general increase in the lifespan of man, However, even the non-therapeutic fields
not only in industrialized countries. New ap- of application, such as in animal husbandry
plications for bioactive biotechnical products and plant protection, contributed to a high
in medical care like their use as immunosup- degree to the continuing interest in secondary
pressants or antiatherosclerotics, and as ani- metabolite production. Last but not least, nat-
mal growth promoters and pesticides in agri- ural products of biotechnical and agricultural
culture rendered research on new secondary origin play an important role as “biochemical
metabolites an apparently endless story (SAN- tools” in molecular biology and in the investi-
GLIER and LARPENT, 1989; ComitC Editorial, gation of cellular functions.
1992; LANCINIand LORENZE-ITI,1993; VIN- More than loo00 antibiotics and similar
ING and STUTTARD, 1995). bioactive secondary metabolites have been
In the past, natural products supplied 5 4 % isolated so far from microbes, and a compara-
of the annual increase in the world’s total bly higher number of drugs was derived from
pharmaceutical market. The list of the 25 plants and even from animals (see, e.g., ma-
worldwide best-selling drugs for application rine tunicates, molluscs, toxic insects, snakes,
in humans in 1992 includes a series of drugs and toads) (BERDYet al., 1980; LAATSCH,
of microbial origin which are used either in 1994). Approximately 500 new representa-
their native structures or as chemical derivatives of low-molecular weight compounds are
tives (see, e.g., Mevacor, Cefaclor or other published every year.
cephalosporins, Augmentin, Sandimmun) In addition to this huge and still growing
(Scrip, 1993). number of bioactive molecules, more than
Many plant products, from digitalis glyco- 1OOOOO derivatives as representatives of some
sides and neuroactive alkaloids to the pyre- few basic structures (e. g., p-lactams, macro-
thrines, serve as therapeutics for human dis- lides, aminoglycosides, tetracyclines, anthra-
eases and as agricultural agents (Comitk Edi- cyclines) were obtained by means of synthetic
torial, 1992). Sometimes, the experiences derivatizations (LAATSCH,1994). Irrespective
made in folk medicine initiated the discovery of this plethora of drug molecules a little
of new plant-derived antitumor drugs, anti- more than a hundred basic structures gained
neuralgic, antihypertonic, antidepressant, in- practical importance.
secticidal, nematicidal, and other bioactive We owe much progress in the detection of
compounds. new drug structures to modern physico-
Antiinfective chemotherapy once was the chemical approaches such as mass-spectrome-
classical domain of biotechnical drug produc- try, high-field nuclear magnetic resonance
tion due to the discovery of p-lactam antibiot- spectroscopy and X-ray diffractometry. Com-
ics, such as penicillins, cephalosporins, clavu- pilations of the numerous structural data
lanates, and carbapenems. Even today, the in- (BERDY et al., 1980; BYCROFT, 1988;
I Introduction: The Importance of Secondary Metabolites as Drugs 3
LAATSCH,1994) provide indispensable assist- From the recently completed chemical syn-
ance in the identification of new drug mole- thesis of taxol it is evident that, as in bicyclic
cules. Thus, the enormous number of already plactams, classical approaches cannot com-
known metabolites from microbes and plants pete with natural producers. Instead, increas-
increased the detection and isolation of alrea- ing attention is given to the recruitment of
dy known structures dramatically. biocatalysts for certain key reactions in
A compilation of about 200 recently de- metabolite production. In addition, directed
scribed products illustrates the current trends biosynthesis in microbial cultures (THIER-
in screening efforts (Tab. 1). These have been ICKE and ROHR, 1993), production of plant
published during the last two years. It is evi- products in cell cultures (BERLIN,Chapter 14,
dent from these data that highly selective this volume), and cell free in vitro systems of
screens prevail and yet the majority of com- enzymatic synthesis and peptide and protein
pounds originate from the classical Actino- producing translation systems are considered
mycete pool. Rare bacteria and fungi, marine as complementary methods in structure-func-
microorganisms and plants now have a signifi- tion studies (ALAKHOVand VON DOHREN,
cant share. It is obvious that well-known or- unpublished data).
ganisms again contribute with newly isolated Only 30% of the total developmental ef-
substances to new, e. g., receptor targeted forts have been spent to the search for a new
screens. Strategies of such screens are dis- drug. However, for the estimation of its effi-
cussed in this volume in Chapter 3 by TANA- cacy and evaluation of safety often more than
KA and OMURA. 50% are needed. Taking into account a quota
The development of new drugs from natu- of approximately 1:15000 for a hit structure,
ral sources is common practice of the pharma- the challenges of modern pharmaceutical de-
ceutical industry. 6000 to loo00 chemicals velopment become visible. In general, natural
have to be tested in a given assay system to products seem to offer greater chances than
obtain one single compound suitable as a synthetically derived agents. Hence, a great
therapeutical agent (OMURA,1992; KROHN research potential is still dedicated to the dis-
et al., 1993). No wonder that research and de- covery of new natural drugs and their bio-
velopment for a new approved drug may cost technical production. Classical strategies of
up to one billion US$. In most cases, a new drug development are being more and more
natural “leading structure” is intensively supplemented by new biomedical approaches
modified by chemical means to improve its and ideas and by the use of genetically engi-
activity and to reduce side effects. Chemistry neered microbes and cells as screening organ-
is also extremely helpful if rather rare natural isms (TOMODAand OMURA,1990; ELDERet
products occurring in low amounts or in oral., 1993). These tools initiated a “renais-
ganisms from sensitive ecological areas have sance” in the search for new leading struc-
been proposed as drugs. For example, 40000 tures. New sources of bioactive material, such
yew trees, i. e., the whole population of as marine organisms, and new microbes from
Northern America, would be required to pro- ecological “niches” promoted the recent ad-
duce 25 kg of taxol, a new promising cancero- vances in the discovery of drugs (WILLIAMS
static drug, and even this amount would not and VICKERS1986; RINEHARTand SHIELD
be sufficient to treat every cancer patient. 1988; MONAGHAN and TKACZ,1990; JACOB
Fortunately, taxol derivatives of similar activ- and ZASLOFF,1994; JENSENand FENICAL,
ity (taxotere) can be obtained by chemical 1994) (Tab. 1).
derivatization of taxoid metabolites which are Present research activities were also stimu-
obtainable in large quantities from the dried lated by the discovery of block busters (Scrip,
leaves of European yews (HEINSTEINand 1993) such as cyclosporin A (KAHAN,1987),
CHANG,1994). Alternatively, cell cultures avermectins (CAMPBELL,1989), acarbose
(ELLISet al., 1996) or endophytic fungi such (MULLER, 1989), and monacolin (ENDO,
as Pestalotiopsis microspora (STIERLEet al., 1979) in microbial cultures. A series of very
1994, 1995; STROBELet al., 1996) of Taxus promising new screening drugs (zaragozic
species could be exploited for production. acid, squalestatins) (HASUMI,1993), erbstatin
4 I General Aspects of Secondary Metabolism
Tab. 1. Selected Natural Products Detected by Screening Efforts Published in 1995196
Compound Producing Organisms Structural Selected Research Group

Reference Type' Properties Involved
Antimicrobial Drugs:
Griseusin Actinomycete PK antibacterial Institute of Microbial
derivatives (unidentified) Chemistry
BE-24566B Streptomyces PK antibacterial Banyu Pharm. Co.
violaceus-niger
Amicenomycin Streptomyces sp. PK-GLYC antibacterial Institute of Microbial
Chemistry
Kalimantacins Alcaligenes sp. PK, mod. antibacterial, Yamanouchi Pharm. Co.
MDR strains and PT Kalbe Pharma
A21459 Actinoplanes sp. PEP antibacterial Lepetit
Epoxyquino- Amycolatopsis acyl AA antibacterial Institute of Microbial
mycins Chemistry
GE 37468 Streptomyces sp. PEP antibacterial Lepetit
Phencomycin Streptomyces sp. PK antibacterial Hoechst
Chrysoapermin Apiocrea chrysosperma PEP antibacterial Hans Knoll Institute
antifungal and Univ. Tiibingen
Bacillaene Bacillus subtilis PK antibacterial Bristol Myers Squibb
GE2270 Planobispora rosea PEP antibacterial Lepetit
AL072 Streptomyces sp. PK, mod. antilegionella Cheil Foods & Chem.
Inc and NIH Korea
Ripostatin Sorangium cellulosum PK antibacterial GBF
Sorangiolid Sorangium cellulosum PK antibacterial GBF
Thiomarinol Alteromonas rava antibacterial Sankyo
(marine)
Echinoserine Streptomyces tendae PEP + PK antibacterial Univ. Tiibingen and
Hans Knoll Institute
07F275 unidentified fungus PK antibacterial Lederle
Pyralomycins Actinomadura spiralis PK, mod. antibacterial Institute of Microbial
Chemistry
RS-22 Streptomyces PK antibacterial RIKEN
violaceusniger
Ochracenomy- Amycolatopsis sp. PK antibacterial Institute of Microbial
cins Chemistry
Azicemycins Amycolatopsis PK antibacterial Institute of Microbial
sulphurea Chemistry
Amythiamycin Amycolaotopsis sp. PEP antibacterial Institute of Microbial
Chemistry
APHE 31~ Streptoverticillium ALK antibacterial Univ. Alcala
griseocarnum
Aurantimycin Streptomyces PEP antibacterial, Hans Knoll Institute
aurantiacus cytotoxic
Cineromycins Streptomyces antibacterial Univ. Tlibingen, Univ.
griseoviridus Gattingen, Hans Knoll
Institute
Papyracon Lachnum payraceum TERP antibacterial Univ. Lund and Univ.
Kaiserslautern
Cephem Penicillium chryso- PEP, mod. antibacterial Panlabs
derivatives genum
Sorrentanone Penicillium chryso- PK antibacterial Bristol Myers Squibb
genum
Tab. 1. (Continued)

Reference Type ' Properties Involved
Benzastatin Streptomyces ALK antifungal, KRIBB
nitrosporeus antiviral free
radical scaven-
ger
Jerangolides Sorangium cellulosum PK antifungal GBF
BE29602 Fusarium sp. PK-GLYC antifungal Banyu Pharm. Co.
Dibefurin unidentified mushroom PK antifungal Abbott
Darlucins Sphaerellopsis filu PK-mod. antibacterial, Univ. Kaiserslautern
antifungal and Univ. Munich
Fusaricidin Bacillus polymyxa PEP-PK antifungal, Wakunaga Pharm. Co.
antibacterial and PT Kalbe Pharma
Helioferin Mycogone rosea PEP-PK antifungal, Hans Knoll Institute
antibacterial
Azalomycin Actinomycete PK antifungal Hoechst, AgrEvo
Liposidolide Streptomyces sp. PK antifungal RIKEN
Chirosazol Sorangium cellulosum PK antifungal GBF
cytotoxic
Ratjadon Sorangium cellulosum PK antifungal GBF
Chrysospermin Apiocrea chrysosperma PEP antifungal Hans Knoll Institute
Hydroxystro- Pterula sp. PK antifungal Univ. Kaiserslauern and
bilurin Univ. Munich
Fusacandin Fusarium sambucinum PK-GLYC antifungal Abbott
Favolon Favolaschia TERP antifungal Univ. Kaiserslautern
and Univ. Munich
Aureobasidins A ureobasidium PEP antifungal Takara Shuzo Co.
pullulans
Phosmidosine Streptomyces sp. NUC antifungal RIKEN and SynPhar
Lab. Inc.
NP-1OlA Streptomyces A-mod. antifungal Hokkaido Univ.
aurantiogriseus
YM-47522 Bacillus sp. PK antifungal Yamanouchi Pharm. Co
Australifungin Sporomiella australis PK antifungal Merck Sharp & Dohme
AKD-2 Streptomyces sp. PK antibacterial Univ. Osaka City
antifungal
UK-2AIBICID Streptomyces sp. PEP antifungal Osaka Univ. and
Suntory Ltd.
Prodimicin Actinomadura spinosa PK antifungal Meijo U, Toyama Pref.
Univ.
Pradimicin Actinomadura spinosa PK antifungal Toyama Pref. Univ. and
Bristol Myers Squibb
Ascosteroside Ascotricha amphitricha TERP-GLYC antifungal Bristol Myers Squibb
Epothilone Mucor hiemalis PK-AA antifungal, GBF
cytotoxic
Fusarielin Fusarium sp. PK antifungal Univ. Tokyo
Saricandin Fusarium sp. PK antifungal Abbott
Furanocandin Tricothecium sp. PK-GLYC antifungal Meiji Seika Kaishi Ltd
and Mitsubishi Chem.
Corp.
Siamycin Streptomyces sp. PEP antiviral, HIV Bristol Myers Squibb
L-671,776, Stachybotrys sp. PK-AA HIV protease Ciba-Geigy
derivatives inhib., endothe-
lin antag.
Tab. 1. (Continued)

Benzastatins Streptomyces ALK free radical KRIBB

nitrosporeus scavenger anti-
fungal, antiviral
Triterpene- Fusarium compactum TERP rhinovirus pro- Abbott
sulfates tease inhib.
Quinoxa- Betula papyrifera PEP-PK antiviral: Merck Sharp & Dohme
peptides HIV1,2,RT
Karalicin Pseudomonas PK antiviral: HSV Univ. Cagliari and Univ.
fluorescens Cattolin (Rome)
AH-758 Streptomyces sp. PK antiviral: HSV Kumamoto Univ.
Eulicin Streptomyces sp. PEP-mod. antiviral: HIVl Jikei Univ., Institute of
Microbial Chemistry
Sattabacin Bacillus sp. A-mod. antiviral: HSV Univ. Cagliari and Univ.
Rome
Sattazolin Bacillus sp. AA-mod. antiviral: HSV Univ. Cagliari and Univ.
Rome
GE20372 Streptomyces sp. PEP antiviral: HIV Lepetit
Isochromo- Penicillium sp. PK antiviral: HIV Kitasato
philones DGAT, AC-
TAT inhib.
Antitumor Drugs
Sch5290011 Gliocladiurn sp. PEP antitumor Schering-Plough
Rakicidins Micromonospora sp. PEP-PK cytotoxic Bristol Myers Squibb
Esperamicin Actinomadura PK-GLYC antitumor Bristol Myers Squibb
verrucosospora
Ossamycin Streptomyces PK cytotoxic Lilly
hygroscopicus
Acetophthalidin Penicillium sp. (marine) PK cell cycle RIKEN
inhibitor
Tryprostatins Aspergillus fumigatus PEP-TERP cell cycle inhib. RIKEN
Sparoxomycin Streptomyces NUC-mod. proliferation Toyama Pref. Univ.
sparsogenus mod.
Cochleamycins Streptomyces sp. PK antitumor Kirin Brewery Co.
Himastatin Streptomyces PEP antitumor Bristol Myers Squibb
hygroscopicus
Chondramide Chondromyces crocatu'S PEP cytotoxic GBF
Anguinomycin Streptomyces sp. PK cytotoxic Univ. Tokyo
Clovalicin Sporothrix sp. PK cytocidal Kitasato
Clecarmycin Streptomyces sp. PK antitumor Kyowa Hakko Kogyo
Piericidin Streptomyces sp. PK-GLYC antitumor Snow Brand Milk Co.
derivatives and Kamagawa Univ.
Hydroxymyco- Bacillus sp. PK cancerostatic Institute of Microbial
trienine Chemistry and Showa
College
FR901537 Bacillus sp. PEP-PK aromatase Fujisawa
inhib.
Medelamine Streptomyces sp. PK, mod. anticancer Nippon Kayaku
Naphthablin Streptomyces sp. PK oncogen Keio Univ. and Institute
function inhib. of Microbial Chemistry
Macquarimicin Micromonospora sp. PK cytotoxic Abbott
Tab. 1. (Continued)

Thiazinotrieno- Streptomyces sp. PK-PEP cytostatic Institute of Microbial

mycin (cancer) Chemistry and Showa
College
Cremeduycin Streptomyces cremeus A, mod. cytotoxic Univ. Illinois
Tryprostatin Aspergillus fumigatus PEP, mod. cell cycle inhib. RIKEN
Sch50673,6 Nattrassia mangiferae PK antitumor Schering-Plough
Terpentecin Streptomyces sp. PK antitumor: Kyowa Hakko Kogyo
topoisomerase
inhib.
FD-211 Myceliophthora lutea PK cytotoxic: MDR Taisho Pharm. Co.
Cytogenin Streptoverticillium PK antitumor Institute of Chemother-
eurocidium apy (Shizuoka) and In-
stitue of Microbial
Chemistry
Enaminedonin Streptomyces sp. PEP-PK detransforming RIKEN
tumor cells
Dihydroepi- Penicillium patulum PK-mod. IL-1 receptor Upjohn
epoformin antag.
EI-1507-1/2 Streptomyces sp. PK IL-1-converting Kyowa Hakko Kogyo
enz. inhib.
TAN-15 11 Streptosporangium PEP-PK induces Takeda
amethystogenes cytokines
CJ-12,371,2 unidentified fungus PK DNA gyrase Pfizer
inh.
Pharmacological Activities
FR901,483 Cladybotryum sp. ALK-P-ester immunosuppr. Fujisawa Pharm. Co.
PA-48,153 Streptomyces prunicolor PK immunosuppr. Shionogi
27-0-demethyl- Steptomyces PK-AA immunosuppr. Smith Kline Beecham
Rapamycin hygroscopicus
NFAT 68,133 Streptomyces sp. PK immnuosuppr. Abbott
Stevastatin Penicillium sp. PEP-PK immunosuppr. Nippon Kayaku
Trichstatin Streptomyces sp. PK immunosuppr., Kyowa Hakko Kogyo
histidine decar-
boxylase inhib.
Cytosporin Cytospora sp. PK angiotensin bdg. Merck Sharp & Dohme
inhib.
Leustroducsin Streptomyces platensis PK-P-ester t hrombocytosis Sankyo Co.
inhib.
Plactins Agonomycetales PEP stimulates fibri- Tokyo Noko Univ.
nolytic activity
TAN1323CID Streptomyces PK angiogenesis Takeda
purpurescens inhib.
Monamidocin Streptomyces sp. PEP fibrinogen rec. Nippon Roche
antag.
A-72363 Streptomyces nobilis GLYC heparanase Sankyo Co.
inhib.
Trachyspic acid Talaromyces PK hep a r a na se Sankyo and Univ.
trachyspermus inhib. Tokyo
Carbazo- Streptomyces violaceus AA-PK antioxidant Univ. Tokyo
quinocins
Tab. 1. (Continued)

Phenopyrazin , Penicillium sp. PK-AA radical Kitasato

scavenger
Balmoralmycin Streptomyces sp. PK protein kinase Ciba-Geigy
inhib.
Staurosporine Streptomyces ALK protein kinase Ciba Geigy
analogs longisporoflavus C inhib.
Paeciloquinones Paecilomyces carneus PK protein tyrosine Ciba Geigy and Panlabs
kinase inhib.
Factor AIC unidentified fungus PK, mod. myoinositol Lepetit
Pase inhib.
MS-444 Micromonospora sp. PK myosin light Kyowa Hakko Kogyo
chain kinase
inhib.
WS79089B Streptosporangium PK endothelin con- Fujisawa
roseum verting enzyme
inhib.
Stachybocin Stachybotrys sp. PK-AA endothelin rec. Asahi
antag.
RES-1149 Aspergillus sp. PK endothelin rec. Kyowa Hakko Kogyo
antag.
RES-701 Streptomyces sp. PEP endothelin rec. Kyowa Hakko Kogyo
antag.
L-671,776 Stachybotrys sp. PK-AA endothelin rec. Ciba-Geigy
derivatives antag.
Mer-A2026 Streptomyces pactum PK vasodilatory Mercian Corp.
ET Penicillium sclerotium PK endothelin rec. Xenova and Parke
antag. Davis
Drirnane-ses- Aspergillus ustus TERP-PK entothelin rec. Xenova
quiterpenes bdg.
Bassiatin Beauveria bassina PEP platelet aggr. Taisho Pharm.
inhib.
Herquline Penicillium herquei ALK platelet aggr. Kitasato
inhib.
Schizostatin Schizophyllum TERP squalene synth. Sankyo
commune inhib.
Macrosphelide Microsphaeropsis sp. PK cell adhesion Kitasato
inhib.
Sulfobacins Chryseobacter sp. PK-S Willebrand fac- Nippon Roche
tor rec. antag.
Lateritin Gibberella lateritium PEP ACAT inhib. Tokyo Noko Univ.
Isohalobacillin Bacillus sp. PEP-PK ACAT inhib. Tokyo Noko Univ.
GERI-BP002-A Aspergillus fumigatus PK ACAT inhib. KRIBB
Pyripyropenes Aspergillus furnigatus PK ACAT inhib. Kitasato and Pfizer
Amidepsine Humicola sp. DGAT inhib. Kitasato
Terpendole Albophoma TERP-mod. ACAT inhib. Kitasato
yamanashiensis
Epi-cochlio- Stachybotrys bisbyi TERP-PK ACAT inhib. Sankyo
quinone
F1839 Stachybotrys TERP-PK cholesterol Tokyo Tanabe Co. and
esterase inhib. Univ. Tokyo
CETPI Cytospora (insect PK cholesteryl ester Cornell Univ. and
associated) transfer protein Schering-Plough
inhib.
Tab. 1. (Continued)
~~

Fluvirucin Streptomcyces sp. PK-GLYC phospholipase Univ. Keio

inhib.
Thermorubin Thermoactinomyces sp. PK aldose reduct- UNITIKA Co. and
ase inhib. Univ. Osaka
Salfredins Crucibulum sp. PK-mod. aldose reduct- Shionogi
ase inhib.
Panosialins Streptomyces sp. PK-mod. glycosidase Kitasato
inhib.
Xenovulene Acremonium strictum TERP GABA-benzo- Xenova
diazepine re-
ceptor binding
Arisugacin Penicillium sp. TERP AChE-inhib. Kitasato
Nerfilin I Streptomyces halstedii PEP-PK neurite out- Somtech and Univ.
growth ind. Tokyo
Michigazones Streptomyces halstedii PEP neuronal cell Univ. Tokyo
protecting
Aestivophoerin Streptomy ces PK-mod. neuronal cell Univ. Tokyo
purpeofuscus protecting
Lavandu- Streptomyces neuronal cell Univ. Tokyo
quinocin virdochromogenes protecting
Epolactaene Penicillium sp. (marine) PK neuritogenic RIKEN and Kaken
Pharm. Co.
MQ-387 Streptomyces PEP aPase N inhib. KRIBB
nayagawaensis
YL-01869P Actinomadura PEP-mod. matrix metallo- Sankyo
ultramentaria proteinase
inhib.
YM 4714112 Flexibacter sp. PEP elastase inhib. Yamanouchi Pharm. Co.
Poststatin Streptomyces PEP Pro-endopeptid- Institute of Microbial
virdochromogenes ase inhib. Chemistry
Cathstatins Microascus longirostris PEP-mod. proteinase SynPhar Lab Inc. and
inhib. Institute of Marine
Bioscience (Halifax)
BE-40644 Actinoplanes sp. PK t hioredoxin Tsukuba Res. and
inhib. Banyo Pharm. Co.
RPR113228 Chrysosporium lobatum TERP farnesyl protein R h h e Poulenc Rorer
transferase
inhib.
Andrastin Penicillium sp. TERP-PK farnesyl protein Kitasato and Keio Univ.
transferase
inhib.
Saquayamycins Actinomycetes PK farnesyl protein Keio Univ. and Institute
transferase of Microbial Chemistry
inhib.
Agricultural Uses
Rotihibin Streptomy ces PEP-PK plant growth Univ. Tokyo and
graminofaciens regulator Ajinimoto
Pironetin Streptomyces sp. PK plant growth Nippon Kayaku
regulator
Phthoxazolin Streptomyces PK herbicidal Univ. Paul Sabatier
griseoaurantiacus (Toulouse)
Tab. 1. (Continued)

Methylstrept- Streptomyces sp. PK-mod. herbicidal Hoechst India

imidon-deri-
vatives
Fudecalone Penicillium sp. PK anticoccidial Kitasato
Arohynapene Penicillium sp. PK anticoccidial Kitasato
Xanthoquinodin Humicola sp. PK anticoccidial Kitasato
Hydrantomycin Streptomyces sp. PK herbicidal Kitasato
antibiotic
Iturins Bacillus subtilis PEP-PK phytopathogens USDA, Univ. Texas and
Univ. Purdue
Trichorzins Trichoderma harzianum PEP antifungal CNRS (Paris)
Azalom ycin Actinomycete PK antifungal Hoechst and AgrEvo
Streptomyces Merck Sharp and
hygroscopicus Dohme
Phthoxazolines Streptomyces sp. PK-mod. antifungal Kitasato
Phenamide Streptomyces albospinus AA-mod. antifungal Monsanto
Patulodin Penicillium urticae PK antifungal Osaka Univ.
Gualamycin Streptomyces sp. GLYC acaricidal Nippon Kayaku Co.
NK-374200 Taralomyces sp. NUC-PEP insecticidal Nippon Kayaku Co.
Melanoxadin Trichoderma sp. melanine bios. Teikyo Univ. and
inhib. Tokyo Univ.
Albocycline Streptomyces sp. PK melanogenesis Kitasato
inhib.
CI-4 Pseudomonas sp. PEP chitinase inhib. Shimizu Labs.
(marine)
Oligosperons Arthrobotyrys TERP nematocidal Australian National
oligospora Univ.
Isocoumarins Lachnum sp. (Ascomy- PK nematocidal Univ. Kaiserslautern
cete) (FRG) and Univ. Lund
(Sweden)
Milbemycins Streptomyces sp. PK antihelminthic Smith Kline Beecham
Sulfinemycin Streptomyces albus PK-mod. antihelminthic Lederle
Musacins Streptomyces antihelminthic Univ. Gottingen, Univ.
griseoviridis Tubingen, Hans Knoll
Institute
Lachnum- Lachnum papyraceum PK nematocidal, Univ. Lund and Univ.
lactone cytotoxic Kaiserslautern
' Structural type: PEP - peptide, PK - polyketide, TERP - terpenoid, GLYC - glycoside, A A - amino
acid, NUC - nucleoside, mod. - modified.
* Property: antag. - antagonist; bios. - biosynthesis; ind. - inducer; inhib. - inhibitor; rec. - receptor.
Group identification: Univ. - University of.
(AZUMA, 1987), bestatin (OCHIAI, 1987), to- out up to a volume of more than 300 m3. The
postins (SUZUKIet al., 1990), etc., are to be yield is sometimes more than 40 g L-' (VAN-
introduced into future therapy. DAMME, 1984), and up to 1OOgL-' in peni-
The large-scale biotechnical production of cillin fermentations. This demonstrates the ef-
bioactive compounds has been developed in a ficiency of strain selection which supported
highly effective manner. Fermentations of knowledge of biosynthesis and strain genetics.
high-producing microorganisms are carried Optimum bioprocess control and suitable fer-
mentation equipment were developed as fur- small, but systematically defined groups of or-
ther prerequisites of a highly efficienct pro- ganisms (e.g., special species and genera of
duction of biotechnical drugs. microbes, plants, animals) and point to the
As an introduction to this volume, this enormous variability of chemical structures
chapter summarizes some of the general as- (ComitC Editorial, 1992). In microbes, the ca-
pects of secondary metabolism in microorgan- pacity to generate secondary metabolites is
isms such as: frequently lost by genomic mutations, but this
feature misses any concomitant effect on the
- the biological role of bioactive compounds vegetative development of the pertinent
in the producer strains, strains (SHAPIRO,1989; OLESKIN, 1994). An
- the biosynthetic pathways and their organi- inverse correlation is usually observed be-
zation, tween specific growth rate and the formation
- natural and induced variations of second- of secondary metabolites such as antibiotics.
ary metabolite structures and problems of Particular features of morphological differen-
their structural classification. tiation in surface or submerged cultures, such
as the formation of spores and conidia, seem
Finally, future perspectives of drug screen- to be related to the production capacity of
ing from microbial sources are discussed. secondary metabolism. Moreover, a maxi-
mum production rate of antibiotics and other
secondary metabolites (pigments, alkaloids,
mycotoxins, enzyme inhibitors, etc.) has fre-
quently been observed when growth-promot-
ing substrates were depleted from the me-
2 Secondary Metabolism, dium (DEMAIN,1992). This phenomenon was
called “catabolite regulation” (DEMAIN,
an Expression of Cellular 1974). This may be one of the reasons for the
phase-dependency of biosynthesis of many
and Organismic microbial drugs.
Thus, during the microbial growth phase
Individuality (trophophase) secondary metabolism is often
suppressed, but increased later during the
2.1 Roles of Secondary “idiophase” (VINING,1986). Sometimes this
feature is not present and depends on the par-
Metabolites in Producing ticular strains and growth conditions. For in-
Organisms stance, the formation of phytotoxins by some
phytopathogenic microbes such as Alternaria
The majority of bioactive products of mi- and Fusarium strains is not a subject of catab-
croorganisms and plants is generated by sec- olite regulation and even occurs in a growth-
ondary metabolism. This part of the meta- associated manner (REUTER,1989). On the
bolic machinery of microbes, plants, and ani- other hand, the production of antifungal ef-
mals may play no essential role in the vegeta- fectors including peptaibol trichorzianine may
tive development of the producing organisms, be induced, as shown in Trichoderma har-
but seems to convey advantages to the perti- zianum by cell walls of the plant pathogen
nent species concerning its long-term survival Botrytis cinerea (SCHIRMBOCK et al., 1994).
in the biological community and environment Likewise, certain plant metabolites may in-
(LUCKNERet al., 1977; KLEINKAUF and VON duce the synthesis of peptide antibiotics in
DOHREN, 1986; WILLIAMSet al., 1989; the respective pathogenic Pseudomonas
LUCKNER,1989 VINING,1992; WILLIAMS et strains (MAZZOLAand WHITE,1994; MO et
al., 1992; CAVALIER-SMITH, 1992; OLESKIN, al., 1995). In general, the phase-dependency
1994; VININGand STUTTARD,1995) (Tab. 2). or specific inducibility indicates that the sec-
Further interpretations imply the formation ondary metabolism is strictly governed by in-
of certain secondary metabolites by relatively herent regulatory systems (see Sect. 2.2).
Tab. 2. Presumed “Roles” of Secondary Metabolites in Their Producer Organism

Exogenous “role” in the Endogenous “role” in the
environment producing organism
- -
endogenous regulatory
protection against competing
organisms signals triggering morpho-
- - genesis
regulation of commensalism endogenous signals regulating
and cohabitation mating processes such as
-
pheromones
protection against
physicochemical noxes (UV light) endogenous detoxification
of metabolites
-
acquisition of trace elements
-
supply of special building
detoxification of trace material of the cell wall
elements
endogenous reserve material
not accessible to other micro-
organisms
Most of the secondary metabolites are bio- LIFFE, 1992; JOHNSON and ADAMS, 1992).
synthesized in microbes and plants via com- Others act as chemoattractants or as repel-
plex multistep pathways involving many enzy- lents towards insects fructifying flowers or
matic and even non-enzymatic events. These damaging plant tissues. A series of plant hor-
appear to be integrated in a coordinated man- mones (cytokinins, gibberelic acid, jasmonic
ner into the global microbial processes of cy- acid, etc.) are similar in structure but per defi-
todifferentiation such as formation of spores, nitionem are not secondary metabolites. An-
conidia, and aerial mycelia (LUCKNER,1989), other function of secondary metabolites in
or in the processes of invasion or defense. plants is the detoxification of poisonous
The same is true for plants in which second- metabolites via an endogenous compartmen-
ary metabolite formation occurs in different tized storage (LUCKNER,1989). The role of
tissues, e. g., roots, leaves, flowers, and seeds. secondary metabolism in microbes is even
Hence, it seems obvious that secondary more difficult to understand. Cellular efforts
metabolism does not reflect an occasional needed for secondary pathways are rather
feature but is the result of a very long evolu- low in the wild-type strains (only a small
tionary development. As was shown for the amount of the overall substrate intake is con-
tetracycline antibiotics from Sfrepfomyces verted to bioactive secondary metabolites).
spp. more than 200 genes may affect the bio- This part of metabolism would possibly have
synthetic pathway (VANEKand HOSTALEK, been eliminated during phylogenesis without
1985). No wonder that speculation about the any selective advantage of secondary metab-
endogenous “function” and “roles” of sec- olite production. It appears to be a generally
ondary metabolites in the producing organ- accepted view that microbial secondary
isms themselves never came to an end (VA- metabolites play an important but not gener-
NEK et al., 1981; VINING, 1992; OLESKIN, alizable rote, at least in special situations,
1994; VINING and STUTTARD, 1995). e. g., in warranting the survival in particular
To maintain such a great number of genes, environmental systems, during limitation of
generally linked into clusters, during evolu- nutrient supply or even in the course of mor-
tion should be of advantage to the pertinent phological development (LUCKNERet al.,
organism. Obviously, in plants many second- 1977; KLEINKAUFand VON DOHREN,1986;
ary metabolites are involved in the protection VINING,1992; KELLet al., 1995; VINING and
against microorganims and animals (CUND- STUTTARD, 1995). From this point of view,
the formation of large amounts of antibiotics lated by secondary metabolites in hetero-

by high-producing strains (substrate conver- logous populations.
sion rates Yglucose,drug >0.1) would be consid-
ered as a “pathophysiological” problem (VA- The self-protecting mechanism in antibiot-
NEK et al., 1981). In order to better under- ic-producing microbes should be mentioned
stand the general roles of secondary metabol- as a further evidence of an ecological function
ites in microbes one could refer to the color of antibiotics, as a “weapon” against competi-
of hairs and feathers in animals, their odorous tors (ZAHNERet al., 1983; BRUCKNER et al.,
pheromones, and other metabolic products 1990; CUNDLIFFE, 1989,1992; WILLIAMS and
which do not contribute per se to the vegeta- MAPLESTONE, 1992). By this means the mi-
tive life of the pertinent species. But they crobe prevents suicide due to its own second-
could have outstanding importance during ary metabolite either by enzymatic modifica-
the adaptation to changing media, in the pro- tions of the drug, by alteration of its biologi-
tection against competing organisms, and in cal target, or by an active transport-directed
the regulation of sexual and asexual processes export (see, e. g., the tetracycline efflux)
of genetic exchange. General discussions of (JOHNSON and ADAMS, 1992; NIKAIDO,
secondary metabolite formation in microbes 1994). Usually, resistance mechanisms of the
consider four major fields of importance antibiotic-producing microorganisms are the
(LUCKNER et al., 1977; KLEINKAUFand VON same as in antibiotic-resistant bacteria. The
DOHREN,1986; LUCKNER,1989; WILLIAMS analysis of the gene sequences encoding re-
et al., 1989, 1992; VINING,1992; CAVALIER- sistance determinants support the idea that
SMITH,1992; OLESKIN,1994; VININGand the transfer of resistance occurs from the anti-
STU~TARD, 1995) (Tab. 2): biotic producers to the non-producing mi-
crobes (JOHNSONand ADAMS, 1992; SA-
(1) The formation of secondary metabolites LYERS et al., 1995; HIRAMATSU, 1995; DAV-
facilitates the adaptation to metabolic im- IES, 1994). In addition, the emergence of new
balances as a kind of a “metabolic valve”, types of resistance factors by the formation of
which is needed to remove an excess of mosaic genes has been analyzed in P-lactam-
toxic, endogenous metabolites that other- resistant pneumococci (SPRATT,1994; COF-
wise are accumulated during a partial lim- FEY et al., 1995).
itation of substrates. The great variation of both active and inac-
(2) Secondary metabolism could be a source tive secondary metabolites, that were ob-
of individual building blocks of cells or of served in microorganims and plants supplied
metabolic reserves which warrant the in- the main arguments against their determined
dividuality and particular functionality of function. Obviously, the formation of a bioac-
the given strain. tive secondary metabolite, such as an anti-
(3) Secondary metabolites could be regarded biotic, rather appears as an exception than as
as endogenous signals triggering particu- a rule. Frequently, many inactive “shunt-me-
lar stages of morphogenesis and the ex- tabolites” and congenors are produced in ad-
change of genetic material (see Fig. 1). dition to the few active metabolites. It is not
This hypothesis was particularly sup- reasonable to believe that all these metabo-
ported by the observation that the major- lites are needed in a single organism. It might
ity of the “good” producers (e.g., actino- be that a “function” of a secondary metabo-
mycetes, fungi, bacteria) display a life cy- lite could become apparent only in a particu-
cle involving several stages of morpholog- lar, exceptional situation or in special stages
ical differentiation. of development. Hence, the selection pres-
(4) Secondary metabolite formation is partic- sure on structures and secondary pathways is
ularly important in biosystems as a signal necessarily low (ZAHNERet al., 1983). As a
of interspecific “communication” be- consequence, spontaneously evolving variants
tween microbes and other microbes, and mutants could survive with the same
plants, and animals. Symbiosis, commen- probability as their parents, and modifications
salism, and antagonism could be regu- of secondary pathways and structures would
M0
A-factor
, , J ,
VB-factorS factorfrom
Str. vlrldochromugenes
differolide
0
autoinducer from Q
$?
Vibrio Rscheri
on
OH 0
Basidifferquinone Gennicidin
Butalactin
ChOH
no v v - 0
trlsporlc acid C sirenin
antheridiol oogonlol
Fig. 1. Structures of some representatives of signaling molecules from bacteria (streptomycetes) and fungi
(for references, see text).
be preserved (secondary metabolism as a tochromes, chlorophylls, sexual pheromones

“playground of evolution”) (ZAHNER et al., of fungi and bacteria, etc. might have been
1983). This might explain the existence of the evolved similarly. Some of them may be at-
numerous similar structures. According to tested to defined “functions” of microbial sec-
this hypothesis, the limited substrate specifici- ondary metabolites (Tab. 2, Fig. 1).
ty of some enzymes of secondary metabolism A role of secondary metabolism in the ad-
has to be mentioned (LUCKNER,1989). How- aptation to changing nutrient conditions is a
ever, it should be noted that in many multi- realistic position since an excessive supply of
step processes this limited specificity is re- metabolic intermediates (precursors) usually
stricted to certain steps and thus less re- induces or stimulates drug production (DE-
stricted structural regions of the compounds MAIN 1974,1984,1992). Growth may become
(KLEINKAUF and VON DOHREN,in press). A imbalanced and precursors are accumulated
few secondary metabolites, out the pool of during the limitation of a given substrate in
the many non-functional metabolites, have the medium, while others are still available in
apparently acquired an essential role in excess. Excessive precursors could be re-
growth and differentiation. The siderophores, leased into the medium or converted to hard-
e.g., are microbial vehicles of iron transport ly metabolizable products which would not
formed in variable structures as constitutive support the growth of competitors. Moreover,
parts of the iron uptake system (VON DER colored secondary metabolites, such as pig-
HELMand NEILANDS, 1987; WINKELMANN, ments, could protect cells and spores from
1991; WINKELMANN and DRECHSEL, Chapter damage by ultraviolet radiation or also could
5, this volume). Per definitionern, they should promote the acquisition of rare elements via
not be regarded as secondary metabolites. complex formation as, e. g., siderophores.
Highly specialized biomolecules such as cy- Complex formation could also protect the
cells from high concentrations of toxic heavy Rhizocfonia fungi on plant roots by products
metals. of cohabiting streptomycetes and bacteria. In-
The incorporation of secondary metabo- terspecific effects have also been postulated
lites into cellular structures has been sug- for volatile compounds which are formed,
gested to contribute to their individual char- e. g., by streptomycetes and cyanobacteria.
acteristics. Thus, streptomycin and its build- Geosmin, isoborneol, and mucidon are the
ing moiety, streptidine, were established as a constituents of the typical earthy odor. It has
constitutent of the cell wall of the producing been shown that sclerin and scleroid from the
Sfrepfomyces griseus (DEMAIN,1984; DIST- fungus Sclerofinia liberfiana stimulate the bio-
LER et al., 1992). Otherwise, the production synthesis of aminoglycosides by streptomy-
of secondary metabolites (so-called “idio- cetes, but also the growth of some plants
lites”) (DEMAIN,1992), could serve as a kind (KUBOTAet al., 1966; OXFORDet al., 1986).
of a metabolic reserve which cannot be The formation of phytotoxins by phytopa-
metabolized by other microbes. Some anti- thogenic microbes is mentioned as another in-
biotics (anthracyclines, tetracyclines, cyclos- terspecific communication system (KOHMO-
porins, etc.), e. g., are stored within the myce- TO and YODER, 1994). Constituents of the
lium and their complete degradation requires microbial cell wall (elicitors such as p1,3-1,6-
a series of specialized enzymatic steps. Other- glucans from Phytophfora megasperma) are
wise, bioconversions of antibiotics are a con- recognized by specific plant cell membrane
stitutive part of the self-protecting mecha- receptors. Subsequently, a series of protective
nisms of the producer strain. mechanisms is induced in the plant (e.g., hy-
Moreover, concentrations of several anti- persensitivity reactions, de novo synthesis of
biotics were shown to decrease in the course tissues, secretion of enzymes lysing microor-
of prolonged cultivation, thus indicating the ganisms, and formation of antimicrobial phy-
onset of degradative processes. Some fungi toalexins). On the other hand, some of the
are well-known to degradate their own poly- phytoalexins are inactivated by enzymes of
ketides such as, e.g., citrinin (BARBERet al., phytopathogenic microbes.
1988) and zearalenon and even to use them In the natural habitat genetic information
for additional syntheses. Active antibiotics can be transferred from one microbe to an-
were usually not detected in soil samples, al- other interspecifically. Both biosynthetic pro-
though recently sensitive procedures have cedures and resistance mechanisms thus can
permitted the detection of phenazines (COOK be spread among various heterologous spe-
et al., 1995). Their complete degradation un- cies and genera. Apparently this is also true
der natural conditions seems very likely. for genetic exchanges between plants and mi-
Most likely, a series of signaling molecules crobes. A recent intriguing example is the dis-
is supplied by the secondary metabolism that covery of a taxol producing fungus living in
possess interspecific (ecological) or species- taxol producing yew trees (STIERLEet al.,
dependent functions, e. g., as signals trigger- 1994). Typical plant hormones such as gibber-
ing morphogenesis and the exchange of ge- ellins and jasmonic acid are also produced by
netic material (Fig. l). By growth inhibition some microorganisms. Aflatoxins formed via
of competing microbes a producer strain complicated biosynthetic pathways in fungi,
could attain an advantage (c. f. the production such as Aspergillus, have been established in
of herbicidal antibiotics by phytopathogenic actinomycetes. Sequence analyses of the
bacteria which damage plant tissues and facil- genes encoding penicillin and cephalosporin
itate nutrient acquisition from the host) biosynthetic clusters (ACV synthase, isopeni-
(KOHMOTOand YODER, 1994; MAZZOLA cillin N-synthase, acyltransferase, deacetoxy-
and WHITE,1994; M o et al., 1995). Vice versa, cephalosporin C-synthase, and deacetoxy-
secondary metabolism could confer a particu- cephalosporin C-hydroxylase) in Penicillium
lar advantage in symbiotic systems, such as chrysogenum, Acremonium chrysogenum, and
Pseudomonaslplant roots, to both the produc- Streptomyces spp. strongly suggested that fun-
ing strain and the symbiont. An example is gi received the pertinent genes from the pro-
the control of phytopathogenic Fusarium or karyotic actinomycetes during evolution
(LANDANet al. 1990 MILLERand INGOLIA, tors of Streptomyces differentiation (Fig. 2)

1993; BUADESand MOYA, 1996). The pro- (KHOKHLOV,1982; GRAFE 1989; HORINOU-
duction of cephabacins, chitinovorins, clavu- CHI and BEPPU,l990,1992a, b, 1995; BEPPU,
lanates, olivanic acids, carbapenems, and 1995). They are required as microbial “hor-
thiopeptides by unicellular bacteria and strep- mone-like” substances in few species such as
tomycetes may indicate that an original bio- streptomycin, virginiamycin or anthracycline
synthetic pathway was spread horizontally producing strains. These effectors permit the
among different microbes, thus giving rise to formation of antibiotics and aerial mycelium
evolutionary variations of structures and by some blocked, asporogenous, antibiotic-
pathways. negative mutants even in very low concentra-
The evolution of secondary metabolism tions. Several other autoregulators of mor-
even appears to create hybrid structures by phogenesis have been investigated (see, e. g.,
the combination of genetic material originat- factor C) (SZESZAKet al., 1991). Otherwise,
ing from heterologous hosts. Recently, thio- germicidin B (PETERSENet al., 1993) from
marinol (SHIOZAWA et al., 1993) was isolated Streptomyces violaceusniger inhibits germina-
from the marine bacterium Alteromonas rava tion of its own spores by interference with en-
as a composite compound formed by the es- dogenous ATPase. Antibiotics such as hor-
terification of pseudomonic acid (found in maomycin (ROSSNERet al., 1990) and pama-
Pseudomonas fluorescens) and holomycin (a mycin (KONDOet al., 1986) were shown to
pyrrothine antibiotic, found in Streptomyces have autoregulatory functions. Moreover,
Spa). streptomycetes can produce interspecific in-
The involvement of secondary metabolism ducers such as anthranilic acid and basidiffer-
in the regulation of microbial cytodifferentia- quinone (Fig. 1) which affect basidiomycetes
tion seems to be important, at least in some and the formation of fruiting bodies (AZUMA
cases. The morphogenesis of antibiotic-pro- et al., 1980; MURAOet al., 1984).
ducing microorganisms (streptomycetes, fun- Moreover, regulatory molecules inducing
gi, Mycobacteria, etc.) is obviously mediated cytodifferentiation were isolated from fungi
by a plethora of biochemical steps, which dis- and molds confirming that morphogenesis
play a high specificity for the given organism. can be mediated by the aid of an agency of
The pathways are regulated by individual sig- specialized endogeneous factors (HAYASHIet
nals in a highly coordinated manner (Fig. 1) al., 1985). They can be regarded as secondary
(LUCKNER,1989). During morphogenesis, si- metabolites since they do not possess any
lent genes are activated that have not been function in vegetative development.
expressed during the growth phase. Accord- In addition, sexual factors from fungi and
ingly, several endogenous non-antibiotic reg- yeasts can be considered as functionalized
ulators of the cell cycle were discovered in secondary metabolites. They trigger zygo-
Streptomyces cultures, and their structure was spore formation by haploid cells belonging to
elucidated (see below) (KHOKHLOV,1982; different mating types (GOODAY,1974). Dur-
GRAFE, 1989 HORINOUCHIand BEPPU, ing the evolution of signal systems, from the
l990,1992a,b, 1995; BEPPU,1992,1995). Cor- simple pro- and eukaryotes up to the hor-
relations between the biogenesis of some pep- monal control in mammalians, some struc-
tidic antibiotics and morphogenesis were also tures and activities have been conserved. The
described for synchronously growing Bacillus alpha-factor of the yeast Saccharornyces cere-
cultures (MARAHIELet al., 1979). Tyrocidin, visiae as one of its sexual pheromones, e.g.,
gramicidin, and bacitracin are produced dur- appears to be partially homologous to the hu-
ing the onset of sporulation, suggesting that man gonadotropin releasing hormone (Lou-
their function concerns the control of tran- MAYE et al., 1982). Moreover, inducers of dif-
scription, spore permeability, dormancy of ferentiation of Friend leukemia cells were iso-
spores, and their temperature stability (MA- lated from soil organism such as Chaetomium
RAHIEL et al., 1979,1993). sp. These chlorine containing substituted di-
The y-butyrolactones represent a particu- phenols (Fig. 1) also induce morphogenesis
larly important group of endogenous regula- (stalk cell differentiation) of Dictyostelium
Fig. 2. Regulatory events suggested to be involved in morphogenesis and secondary metabolism of Strep-
tomyces griseus (P: promotor) (HORINOUCHI and BEPPU,1992a).
discoideum, suggesting the similarity of mam- 2.2 Regulation of Microbial

malian and fungal control of the cell cycle
(KUBOHARAet al., 1993). Recently, the oc-
Secondary Metabolism
currence of sexual pheromones was even es-
tablished for the prokaryote Streptococcus 2.2.1 Genetic Organization of
faecalis. Its pheromones stimulate or inhibit
the transfer of conjugative plasmids from do-
Product Formation
nor to recipient strains (WIRTHet al., 1990).
Peptides triggering competence in Bacillus A large number of biosynthetic genes were
subtilis have been characterized and were isolated and characterized and, in general,
termed pheromones (D’SOUZAet al., 1994; they have been found assembled in clusters
SOLOMONet al., 1995; HAMOEN et al., (Tab. 3). Such clusters may contain informa-
1995). tion for the biosynthesis of the basic structure
of the metabolite, its modification, resistance
determinants, e. g., promoting modification of
products, targets, altered targets, or export
systems, as well as regulatory elements; indi-
vidual gene products which might as well ex-
ert regulatory functions.
Tab. 3. Biosynthetic Clusters Identified
Compound Type 0rganism Selected References'

A54145 acylpeptidolactone Streptomyces fradiae BALTZet al., 1996'
Aflatoxins polyketide Aspergillus parasiticus, BROWNet al. 1996
Aspergillus fzavus MAHANTIet al., 1996
Actinomycin chromopeptidolactone Streptomyces chrysomallus KELLERet al., 19962
Anguibactin modified peptide Vibrio anguillarum CHENet al., 1996
Astaxanthin carotinoid Agrobacterium aurantiacum MISAWA et al., 1995
Avermectin polyketide Streptomyces avermitilis MACNEIL,1995
Avilamycin polyketide Streptomyces viridochromo- BECHTHOLD et al., 1996'
genes
Bacitracin branched cyclopeptide Bacillus licheniformis HERZOG-VELIKONJA et al.,
1994
Bialaphos peptide Streptomyces viridochromo- et al., 1996
SCHWARTZ
genes
Carbomycin polyketide Streptomyces ARISAWA
et al., 1995
thermotolerans
Carotinoids terpenoids Rhodobacter capsulatus ARMSTRONG, 1994
Carotinoids terpenoids Myxococcus xanthus ARMSTRONG, 1994
Carotinoids terpenoids Synecococcus PCC7942 ARMSTRONG, 1994
Clavulanic modified peptide Streptomyces clavuligerus HODGSONet al., 1995
acid
Cephalo- modified peptide Acremonium chrysogenum MART~N and GUTIERREZ,
sporin 1995
Cephamycin modified peptide Nocardia lactamdurans COQUEet al., 1993,
1995a,b; PETRICH et al.,
1994
Coronatin modified polyketide Pseudomonas syringae BENDERet al., 1996
Cyclosporin cyclopeptide Tolypocladium niveum WEBERet al., 1994
Daptomycin acylpeptidolactone Streptomyces roseosporus BALTZet al., 1996'
Daunomycin, polyketide Streptomyces C51'peucetius YE et al., 1994; GRIMMet
Daunoru- al., 1994; FILIPPINI et al.,
bicin, Do- 1995; MADDURI and HUT-
xorubicin CHINSON, 1995a, b;
DICKENS et al., 1996
Destruxin peptidolactone Metarhizium anisopliae BAILEYet al., 1996
Elloramycin polyketide Streptomyces olivaceus DECKERet al., 1995
Fatty acids polyketide Streptomyces glaucescens SUMMERS et al., 1995
Fatty acids polyketide Escherichia coli ROCKand CRONAN, 1996
Fengymycin peptide Bacillus subtilis Liu et al., 1996'
Ferrichrome cyclopeptide Ustilago maydis LEONGet al., 1996'
Frenolicin polyketide Streptomyces roseofulvus BIBBet al., 1994
Geldana- polyketide Streptomyces hygroscopicus ALLENand RITCHIE,1994
mycin
Gramicidin S cyclopeptide Bacillus brevis ATCC9999 TURGAY and MARAHIEL,
1995
Granaticin polyketide Streptomyces violaceoruber SHERMAN et al. 1989;
BECHTHOLD et al., 1995
Griseusin polyketide Streptomyces griseus Yu et al., 1994
HC-toxin cyclopeptide Helminthosporium PITKINet al., 1996
carbonum
HET? polyketide? Anabaena sp. BLACKand WOLK,1994
Immuno- modified polyketide Streptomyces sp. MOTAMEDI et al., 19962
mycin
Jadomycin B polyketide Streptomyces venezuelae YANGet al., 1995b, 1996b
Tab. 3. (Continued)
Compound Type Organism Selected References'
Me1anin Aspergillus nidulans TAKANO

et al.. 1995
Colletotrichum lagenarium
Landomycin poly ket ide, glycosylated Streptomyces sp. BECHTHOLD et al., 1996*
6-Methylsali- polyketide Penicillium patulum BECKet al., 1990
cylic acid
Microcystin cyclopeptide Microcystis aeruginosa MEISSNER et al., 1996
Mithramycin polyketide Streptomyces argillaceus LOMBd et al., 1996
unknown polyketide Streptomyces cinnamonensis ARROWSMITH et al., 1992
Nikkornycin modified peptide Streptomyces tendae BORMANN et al., 1996
Nodusmicin polyketide Saccharopolyspora hirsuta LE GOUILLet al., 1993
Nogalamycin polyketide Streptomyces nogalater YLIHONKO et al., 1996
Olean- polyketide Streptomyces antibioticus Q U I R ~and
S SALAS,1995
domycin
Oxytetra- polyketide Streptomyces rimosus KIM et al., 1994
cyclin
Penicillin modified peptide Aspergillus nidulans, SMITHet al., 1990,
Penicillium chrysogenum MACCABEet al., 1990;
DfEz et al., 1990
Phenazin heterocycle Pseudomonas aureofaciens PIERSONet al., 1995
Pristinamycin acylpeptidolactone Streptomyces DE CRECY-LAGARD,
A pristinaespiralis personal Communication
Pristinamycin polyketide/peptide Streptomyces sp. BECKet al.. 1990
M
Purornycin modified aminoglucoside Streptomyces alboniger TERCEROet al., 19%
Pyoverdin branched cycloacylpeptide Pseudomonas fluorescens STINTZIet al., 1996
Rapamycin modified polyketide Streptomyces hygroscopicus SCHWECKE et al., 1995
Saframycin modified peptide Myxococcus xanthus POSPIECHet al., 1996
Soraphen A modified polyketide Sorangium cellulosum SCHUPPet al., 1995
Sterigmato- polyketide Aspergillus nidulans BROWNet al., 1996
cystin
Streptomycin aminoglycoside Streptomyces glaucescens
Streptomyces griseus BEYERet al., 1996
Streptothricin modified aminoglucoside Streptomyces rochei FERNANDEZ-MORENO et
al., 1996
Surfactin peptidolactone Bacillus subtilis COSMINA et al., 1993
Tetraceno- polyketide Streptomyces glaucescens SHENand HUTCHINSON,
mycin 1994
Tylosin polyketide Streptomyces fradiae MERSON-DAVIES and
CUNDLIFFE, 1994
Urdamycin polyketide Streptomyces fradiae DECKERet al., 1995
Whi, spore polyket ide Streptomyces coelicolor DAVISand CHATER,1990
pigment
Zeaxanthin terpenoid (carotinoid) Erwinia herbicola, ARMSTRONG, 1994,
Erwinia uredovora HUNDLEet al., 1994
'Presented at the conference Genetics and Molecular Biology of Industrial Microorganisms. Bloomington
1996.
Presented at the symposium Enzymology of Biosynthesis of Natural Products. Berlin 1996.
Abstracts available from the authors on request.
The techniques employed include reverse ing key result, as shown for the industrial
genetics if sequence data of relevant enzymes penicillin producer (FIERRO et al., 1995;
is available, the use of homologous gene MARTfN and GUTIERREZ,1995). The main
probes or probes constructed from key se- findings with regard to sequencing of com-
quences, the generation by PCR of specific plete genomic fragments are as follows:
probes flanked by conserved key motifs, com-
plementation of idiotrophic mutants, expres- - The identification of biosynthetic genes fol-
sion of pathways or single step enzymes in lows by the detection of core sequences.
heterologous hosts, cloning of resistance de- Such sequences permit the recognition of
terminants followed by isolation of flanking types of biosynthetic unit operations like
sequences, identification and cloning of regu- polyketide condensation reactions, the spe-
latory genes or sequences (promoters, regula- cificities of the respective transferase sites
tory protein binding sites, pleiotropic genes, (HAYDOCKet al., 1995), the number of
“master” genes, etc.). elongation steps, amino acid activation
To improve product levels, the addition of sites; in the case of repetitive cycles where
extra copies of positive regulators (CHATER, certain sites are reused, as in type I1 poly-
1992; HOPWOOD et al., 1995; CHATERand ketide forming systems or, e. g., cyclodepsi-
BIBB,Chapter 2, this volume), extra copies of peptide synthetases, where the number of
biosynthetic genes possibly representing bot- steps remains uncertain.
tlenecks (SKATRUDet al., Chapter 6, this vol- - Additional genes for modification reactions
ume), or the alteration of promoters of key like oxygenases and transferases are readily
enzymes are under investigation. identifiable by standard structural align-
The analysis of clusters has revealed a ments as well as possible regulatory pro-
wealth of information including biosynthetic teins.
unit operations and their surprisingly com-
plex organization. The majority of large pro- At present, however, the unambiguous cor-
teins now known are multifunctional enzymes relation of product and biosynthetic machin-
involved in peptide and polyketide formation, ery is not possible without the support of var-
with sizes ranging from 165 kDa to 1.7 MDa. ious genetic techniques or, if not available
Other systems also forming polyketides, pep- due to the lack of transformation systems,
tides, aminoglycosides, etc., are comprised of structural details from protein chemistry of
non-integrated enzyme activities, still per- isolated enzymes or multienzymes.
forming the synthesis of highly complex struc- To illustrate a few concepts, we will point
tures. The details of various biosynthetic clus- to some recent examples of cluster analysis:
ters are described in the respective chapters PLactam antibiotics as classical examples of
on regulatory mechanisms (CHATER and modified peptides are still leading antibacter-
BIBB,Chapter 2, this volume), peptides (VON ial drugs. Some efforts have been directed to
DOHREN and KLEINKAUF,Chapter 7, this understand at the molecular level the per-
volume), plactams (SKATRUDet al., Chapter formance of industrial overproducers selected
6, this volume), lantibiotics (JACK et al., for decades (SKATRUDet al., Chapter 6, this
Chapter 8, this volume), and aminoglycosides volume). Following the reverse genetics ap-
(PIEPERSBERG and DISTLER,Chapter 10, this proach in isolation of the isopenicillin N syn-
volume). Recent highlights of the elucidation thase gene (SAMSON et al., 1985), which cata-
of such data have been the rapamycin and im- lyzes the formation of the penem bicycle from
munomycin clusters in Streptomyces, the ery- the tripeptide precursor ACV, the clustering
thromycin cluster in Succharopolysporu, the of biosynthetic genes was demonstrated in
surfactin and gramicidin S clusters in Bacillus, both pro- and eukaryotic producers (BARTON
various plactam clusters, and the sterigmato- et al., 1990). The two key enzymes, ACV syn-
cystin cluster in Aspergillus nidulans. An thetase and isopenicillin N synthase showed
overview of examples is presented in Tab. 3. extensive similarities in both bacteria and
The amplification of biosynthetic clusters fungi, and a horizontal intergenic transfer has
in highly selected strains has been a fascinat- been suggested (LANDANet al., 1990; MILL-
ER and INGOLIA,1993; BUADESand MOYA, ploying different reporter genes which al-
1996). The linkage of these adjacent genes il- lowed to measure the expression of both
lustrates well basic principles of cluster organ- genes simultaneously (BRAKHAGEet al.,
ization (Fig. 3) (AHARONOWITZ et al., 1992). 1992; BRAKHAGEand TURNER, 1995;
In bacteria both genes are transcribed unidi- BRAGKHAGE and VAN DEN BRULLE,1995;
rectionally within an operon linked to sets of THENBERGet al., 1996). Such results suggest
other genes the products of which are re- possible additional functions for the penicillin
quired for the modifying reactions of the ce- tripeptide precursor, besides its role in the
phem nucleus to cephamycin, and the forma- formation and the still unclear excretion of
tion of the plactamase inhibitor clavulanic penicillins. The 872 bp intergenic region be-
acid (WARDand HODGSON,1993). Such ex- tween the A . nidulans acvA (pcbAB) and
tensive linkages have been termed superclus- ipnA (pcbC) permits the complex and sensi-
ters. In fungi the encoding genes for ACVS tive regulation involving several protein fac-
and isopenicillin N synthase are bidirectional- tors (for P. chrysogenum, see FENG et al.,
ly transcribed, separated by intergenic regions 1995; CHU et al., 1995). The current knowl-
of about 1 kbp. A variety of environmental edge of regulatory factors and putative fac-
conditions are known to affect fungal plac- tors implied by the identification and charac-
tam production at the transcriptional level terization of trans-acting mutations specifica-
(ESPESOand PENALVA,1996; SUAREZand lly involved in the regulation of the A. nidu-
PENALVA,1996; BRAKHAGEand TURNER, lans biosynthetic genes is summarized in Fig.
1995). The bidirectionally oriented promoters 3b. One of these factors, designated PACC,
between acvA (pcbAB) and inpA (pcbC) may was shown to activate at least the ipnA gene
permit the asymmetrical expression of both transcription in response to shifts to alkaline
genes, and indeed different levels of expres- pH values (SHAHet al., 1991; ESPESOet al.,
sion have been obtained in constructs em- 1993;TILLBURN et al., 1995; ARST,1996). For
a
E F
&lactarns
AB C
' 10kbp ' aflatoxin
raparnycin
A P
c N ' H
acvA aat
Fig. 3%Organization of the biosyn-
thetic clusters of plactams, rapamy-
b
- cin, and sterigmatocystin, b regulato-
ry sites identified in the penicillin
biosynthetic cluster in Aspergillus
nidulans.
PACC seven binding sites with different af- der the regime of organizationally specific
finities have been mapped in this intergenic mechanisms of regulation. The respective reg-
region (SUAREZand PENALVA,1996). An- ulatory mechanisms will be evaluated compa-
other binding site containing a CCAAT motif ratively in a variety of pro- and eukaryotic
was detected, bound by a protein complex de- hosts.
signated PENRl (THENBERG et al., 1996). Regulation of the formation of secondary
PENRl also binds to a CCAAT-containing metabolites in eukaryotes, however, does not
DNA region in the promoter of the aat gene need to be this complex, as will be discussed
encoding acyl-CoA:isopenicillin N acyltrans- below in the case of sterigmacystin/aflatoxin
ferase which is located 3’ of the ipnA gene biosynthesis. As a second example for the or-
(LITZKAet al., 1996). Deletion analysis and ganization of biosynthetic information the PO-
mutagenesis experiments indicated that the lyketide immunosuppressant rapamycin has
binding of PENRl represses the expression of been selected (SCHWECKE et al., 1995). This
acvA and increases that of both ipnA and aat polyketide with an iminoacyl residue is of in-
(THEN BERG et al., 1996; LITZKA et al., terest as an immunosuppressor in autoim-
1996). PENRl thus represents the first exam- mune disease and transplantation. Its biosyn-
ple of a regulatory protein controlling the thesis proceeds by 16 successive condensation
regulation of the whole plactam biosynthesis and 21 modification reactions of 7 acetyl and
gene cluster in fungi. However, many promot- propionyl residues, respectively, followed by
ers of eukaryotic genes are known to contain pipecolate onto the cyclohexane carboxylic
CCAAT motifs which are bound by distinct acid starter unit. The respective cluster has
gene regulatory proteins (JOHNSON and been identified in Streptomyces hygroscopicus
MCKNIGHT,1989). At the time being, it is un- by LEADLAYet al. (SCHWECKE et al., 1995)
known what kind of CCAAT binding protein using polyketide synthase gene probes of ery-
PENRl represents and whether it is a global thromycin synthase from Saccharopolyspora
acting factor specific for the regulation of /3- erythrea). The sequence of 107.3 kbp has
lactam biosynthesis genes. been determined as well as the boundary se-
Using a genetic approach which is feasible quences, to assure the completeness of the ef-
for the ascomycete A. nidulans, three reces- fort. The key part of the cluster is represented
sive trans-acting mutations were identified de- by four genes encoding multifunctional en-
signated prgAllprgB1 for penicillin regula- zymes with sizes of 900 (A), 1070 (B), 660
tion (BRAKHAGEand VAN DEN BRULLE, (C), and 154.1 kDa (P) responsible for the
1995) and npeEl (P~REZ-ESTEBAN et al., formation of the macrolactam ring. These
1995). These mutations formally correspond four genes of 25.7, 30.7, 18.8, and 4.6 kb
to positively acting regulatory genes. Mutants unambiguously correlate with the structural
carrying one of the mutations mentioned pro- features of the product, however, module 3
duced reduced amounts of penicillin. For and 6 contain catalytic sites for the reduction
prgAl and prgBl it was shown that the ex- of the polyketide intermediates, which actual-
pression of both genes acvA and ipnA was af- ly are not found in rapamycin. The solution of
fected (BRAKHAGEand VAN DEN BRULLE, this problem remains to be found and plausi-
1995), whereas npeEl controls at least ipnA ble explanations are either non-functionality
expression (P~REz-ESTEBAN et al., 1995). due to, e.g., point mutations, or a possible
The major nitrogen regulatory protein NRE transient reduction of the intermediates to fa-
of Penicillium chrysogenum has also been cilitate folding, which is reversed later.
found to specifically attach to three GATA/ These key genes are flanked by additional
GATT pairs within this intergenic region 24 open reading frames, most of which have
(HAAS and MARZLUFF, 1995). The pairwise been assigned tentative functions including
attachment sites indicate a possible dimeric modification of the macrolactam, export, and
state of this GATA family transcription fac- regulation. Standard identification proce-
tor and as well connect this regulatory site dures are hampered by the non-availability of
with nitrogen assimilation. This example illus- genetic operations for this strain.
trates that similar biosynthetic genes are un-
The essential data in this case are the pres- product cluster. These types of genes have
ence of large polyfunctional genes in proka- been commonly referred to as primary path-
ryotic clusters and the surprising lack of strict way enzymes. The respective hexanoyl struc-
correlation of expected biosynthetic unit op- ture serves as a starter and is elongated by a
erations within the predicted modules with type I1 system forming an aromatic polyke-
the actual gene structures found. A similar tide. So far, such systems have been found
observation has also been made in the case of only in prokaryotes. Gene characteristics,
the avermectin biosynthetic cluster (MCNEIL however, do not suggest a horizontal transfer
et al., 1995). as in the plactam case (BROWNet al., 1996).
As a recent eukaryotic example the sterig- Finally, a specific transcription factor is a key
matocystin biosynthetic cluster in A. niduluns element in the expression of the enzyme sys-
is considered (BROWNet al., 1996). Sterigma- tem, and no evidence has yet been obtained
tocystin is the penultimate intermediate in the for complex timing and differential gene ex-
biosynthesis of aflatoxins. Both polyketides pression as in the penicillin pathway in A. ni-
are highly mutagenic and thus carcinogenic. duluns.
They spoil food upon fungal colonization, es- Inspection of other clusters included in
pecially by A. flavus and A. parasiticus. These Tab. 3 suggests extensive similarities of cer-
losses may be reduced by a detailed under- tain groups which, at first sight, look like
standing of the regulation of the biosynthetic structurally unrelated compounds. Certain
events. So, e. g., the induction of aflatoxin for- types of regulatory genes are implied in the
mation has been shown to be strongly sup- formation of various metabolites. There
pressed by jasmonate, a phytohormone seems to be a non-species-related separation
(GOODRICHTANRIKULU et al., 1995). De- of type I and type I1 systems, e.g., in polyke-
tailed genetic studies have confirmed the link- tide formation, but the various degrees of in-
age and coregulation of sterigmatocystin and tegration of biosynthetic modules catalyzing
aflatoxin biosynthesis (TRAILet al., 1995a, b; unit operations may be dictated by the chem-
KELLER and ADAMS, 1995; BROWNet al., istry of their products. Finally, the clustering
1996). The recent sequencing of the sterigma- of pathways also suggests their genetic trans-
tocystin biosynthetic cluster in A. niduluns refer between various hosts. Within the evolu-
vealed within a 60 kb region 25 transcripts, tionary frame, adaptation of pathways to var-
the expression of which is coordinated under ious targets has been proposed, e. g., for As-
conditions of toxin production. The cluster is pergilli adapting to insect colonization and
flanked by genes also expressed under non- perhaps moving to other target organisms
production conditions. The regulatory gene (WICKLOWet al., 1994). The structures of
aflR and its A. flavus homolog both specifical- metabolites with key roles in invasive pro-
ly induce gene expression within the cluster. cesses would then adapt to new targets by
Among the identified genes are a fatty acid evolutionary processes.
synthase, five monoxoygenases, four dehy-
drogenases, an esterase, an O-methyltransfer-
ase, a reductase, and an oxidase, all function-
ally implied in the proposed reaction se- 2.2.2 Regulatory Mechanisms
quence. Comparative evaluation of the re-
spective cluster in A. parasiticus shows con- Mechanisms involved in the regulation of
servation of clustering, but no strict conserva- secondary metabolite expression have been
tion of the gene order (TRAILet al., 1995a, b; reviewed recently, focussing on global control
Yu and LEONARD,1995). Conservation of in bacterial systems (DOULL and VINING,
clustering has been suggested to serve both 1995), bacterial mechanisms in detail (CHAT-
purposes of global regulation and horizontal ER and BIBB, Chapter 2, this volume), anti-
movement of biosynthetic activities among biotic formation in Streptomyces coelicolor
species. The striking features of the tremen- (HOPWOODet al., 1995), and autoregulators
dous efforts so far show the integration of a (HORINOUCHIand BEPPU, 1995; BEPPU,
specific fatty acid synthase into a secondary 1995). Eukaryotic systems except for P-lac-
tams have not been in focus regarding special wall composition and changes in the meta-
metabolites. Recent reviews cover plactams bolic spectrum. The changes may not be ob-
(BRAKHAGEand TURNER,1995; SKATRUD vious and some work has been conducted on
et al., Chapter 6, this volume; JENSENand model systems such as Escherichia coli, Bacil-
DEMAIN,1995). lus subtilis, and Aspergillus nidulans. Besides
A variety of stress conditions have been nutrient depletion as envisioned and studied
documented to lead to secondary metabolite in chemostate-like environments employed in
production (DEMAIN,1984; DOULLand VIN- fermentation, a generally neglected field is
ING,1995; VINING and STUTTARD,1995). Be- the response to environmental factors indicat-
sides physical parameters (temperature ing the presence of alike or competing organ-
shock, radiation) chemical signals will trigger isms. According to our understanding of the
the formation of various small response mole- basic role of many of the metabolites em-
cules, which are the subject of this volume. ployed in the control of invasive processes.
Such signals include both high and low con- Such approaches seem obvious. It has been
centrations of oxygen (oxidative stress, lack shown that cell density critically affects anti-
of oxygen, or shift to anaerobic growth), acid- biotic production (WILLIAMSet al., 1992;
ity (pH shift), but generally the response to FUCQUAet al., 1994; SANCHEZand BRANA,
nutrient alterations. Phase-dependency of 1996). The induction of nisin formation by ni-
secondary metabolite formation in microbial sin itself, as mediated by its cluster-inherited
cultures and its correlation to morphological signal system, is another intriguing example
changes suggest that secondary metabolism is (RA et al., 1996; DERUYTERet al., 1996).
subject to general regulatory mechanisms Likewise the presence of phytopathogenic
governing cellular development (BARABASet fungi induces responses, e. g., in rhizosphere
al., 1994). Only some of the regulatory fea- colonizing bacteria including the production
tures have been elucidated in the past and of antifungals (KAJIMURA et al., 1995; PIER-
many are still to be unraveled. Nutrient shift SON and PIERSON,1996). While the presence
regulation of growth is closely coupled to dif- of resistant microorganisms has been applied
ferention through a series of common meta- in selection processes for antimicrobial agents
bolic signals and regulations such as mediated the identification of response signals is still an
by sigma factors and transcriptional enhanc- open field.
ers. In this context, two major questions are
addressed:
(1) Why are microbial secondary metabolism Stress Conditions Related to Nutrient Limita-
and morphogenesis suppressed during tions
growth on media which are rich in car-
bon, nitrogen, or phosphorus and what is In connection with nutrient depletion car-
the cause of catabolite regulation? bon, nitrogen, and phosphate starvation are
(2) What is the nature of the general signals considered in general. The differential induc-
governing a plethora of metabolic events tion of metabolite forming processes has been
and how do they cooperate within the excellently demonstrated by BUSHELLand
cellular frame of developmental pro- FRYDAY(1983). Extensive studies of this as-
grams? pect have also been conducted in the antibiot-
ic fermentation of gramicidin S in Bacillus
There are indeed drastic variations in the brevis. Formation of this cyclopeptide has
extent of responses upon nutritional stress. been found in a variety of stress conditions,
Obvious morphological changes like sporula- including sporulation and non-sporulation
tion or formation of aerial mycelia are caused conditions and, surprisingly, two phosphate
by an undetermined number of respective concentration ranges (KLEINKAUFand VON
genes, reading to sets of proteins and media- DBHREN, 1986) differing from other phos-
tors promoting alterations in the cellular com- phate-effected systems (LIRASet al., 1990).
position. Such changes include altered cell Thus, in many cases less specific induction
and maintenance by interacting regulatory bon sources causing slow growth promote
devices are implied and manipulation may be production. This has been demonstrated nice-
exerted by growth rate control. ly in the case of bacitracin formation in Bucil-
Nutritional downshift in the media caused lus licheniformis (HANLONand HODGES,
by limitation of particular metabolites (amino 1981). Glucose-6-phosphate suppresses the
acids, ATP, sugars, etc.) promotes excessive synthetase enzymes in penicillin biosynthesis
formation of some metabolites due to an im- (JENSENand DEMAIN,1995). However, as
balanced metabolism (supra) (MART~N et al., was shown for ACV-synthase, IPN-synthase,
1986; LIRAS et al., 1990). Accumulation of and expandase in penicillin and cephalospo-
these “precursors” is known to induce sec- rin producing fungal and Streptomyces strains
ondary pathways (see, e.g., the induction of inhibition or repression by glucosed-phos-
ergotamin alkaloid formation by tryptophane phate, ammonium, and phosphate ions de-
in Cluviceps strains) (HOTTER,1986). On the pend on the given strain (AHARONOWITZ et
other hand, limitation of some endogenous al., 1992).
metabolites could be important which inhibit Carbon uptake systems have been studied
global regulatory mechanisms governing aer- in several organisms including enteric bacte-
ial mycelium and spore formation. In this re- ria in which the phosphoenolpyruvate<arbo-
pressing or inhibitory effects on the second- hydrate phosphotransferase system controls
ary pathways and on morphogenesis could be uptake and transport (POSTMAet al., 1993).
diminished. Both features, accumulation of This phosphorylation-controlled multistep
precursors and limitation of repressing process involves adenylate cyclase and
metabolites seem to be involved (DEMAIN, CAMP-mediated gene regulation. Other
1974, 1992; MART~N et al., 1986; HORINOU- mechanisms operate in gram-positive bacteria
CHI et al., 1990, LIRASet al., 1990). The perti- (STEWART,1993) and streptomycetes (CHA-
nent regulatory mechanism may be similar to TER and BIBB,Chapter 2, this volume).
those shown for other global microbial regul- Nitrogen depletion again is a determining
ations. factor in many antibiotic fermentations (SHA-
Metabolite formation has been studied in PIRO,1989). These effects are attributed to ni-
detail in model cases of surfactin (B. subtilis), trogen catabolite repression. A two-compo-
streptomycin (Streptomyces griseus), or peni- nent system sensing the glutamine and a-ke-
cillin (A. niduluns and P. chrysogenum). It is toglutarate levels activates transcription of
controlled by superimposed regulatory cas- catabolic enzymes releasing ammonia or oth-
cades or networks. Such networks include in- er nitrogen sources by autophosphorylation
tracellular and extracellular components and of a His protein kinase (DOULLand VINING,
might include regulators, transducers, signal- 1995). The activation of glutamine synthetase
ing systems, interacting repressors, and acti- is included in this process, the activity of
vators, as well as modification and expression which is as well controlled by several factors
systems. In the case of streptomycin the term including the glutamine level. Actinomycetes
“decision phase” has been coined as a model contain two types of glutamine synthetases. In
for a variety of production processes (PIE- process analysis ammonia has been found to
PERSBERG,1995; PIEPERSBERG and DIST- repress secondary metabolite formation.
LER, Chapter 10, this volume). Despite this Roles of various nitrogen sources have not
complexity, manipulations of single genes been evaluated in detail, but are discussed in
may have substantial effects on production the case of plactams (SKATRUDet al., Chap-
levels. ter 6, this volume). Ammonium ions are also
Most information on the respective deple- catabolite repressors of plactam biosynthesis
tion events have come from model organisms, (cephalosporin C, cephamycin C) in Acre-
but they proved to be useful in a variety of monium and some Streptomyces spp. (JENSEN
cases. Carbon sources are known but poorly and DEMAIN,1995; DEMAIN,1989). Deami-
understood tools in natural product pro- nation of L-valine in the biosynthesis of tylo-
cesses. Readily assimilated compounds, e. g., sin is subject to catabolite regulation by am-
glucose repress production while other car- monium ions (TANAKA,1986).
In bacteria, a stringent response is caused tors (AfsR protein) (HORINOUCHI and BEP-
by nitrogen limitation (CASHEL,1975) and PU, 1992b; BEPPU,1995).
the appearance of non-acylated tRNAs. A The response to exogenous phosphate has
concomitant increase of guanosine-3 ',5 '-te- been studied in E. coli and the involvement of
traphosphate (ppGpp) concentration switches more than 30 genes in the PHO regulon has
off unfavorable biosynthetic processes. Ri- been established (WANNER,1993). Respec-
bosomal protein synthesis is reduced, but the tive efforts in antibiotic production have been
degradation of amino acids continues. This reviewed (LIRASet al., 1990). So p-amino-
fact is due to the binding of ppGpp to RNA benzoate synthetase by S. griseus as a key en-
polymerase and the alteration of its promoter zyme of candicidin synthesis is negatively reg-
recognition. Thus, transcription of many ulated by inorganic phosphate (MART~N,
genes might be stimulated while the expres- 1989). An upstream promotor region of
sion of others declines in a coordinated man- 113 bp length and rich in AT was identified as
ner. The molecule of guanosine-3 ' , 5 '-tetra- a binding site of a general phosphate-depend-
phosphate might be involved in the regula- ent repressor protein. If phosphate-insensi-
tion of the secondary metabolism and also in tive genes such as the Pgalactosidase gene
sporulation of streptomycetes (OCHI, 1990). were coupled to this fragment and transferred
The heterogeneity of promotor structures in other Streptomyces hosts (such as S. livi-
and the complementation of bacterial RNA dam) they became subject to phosphate con-
polymerases by sigma factors could provide trol.
another rational basis for the understanding
of the developmental regulation of gene ex-
pression (CHATERand BIBB,Chapter 2, this
volume). RNA polymerase consists of a core 2.2.3 Genetic Instability
enzyme composed of each of two a- and two
Psubunits. Bacterial promotor recognition is The formation of secondary metabolites
regulated by sigma factors (a7, u43,etc.) at- often is genetically instable and many expla-
tached to the core enzyme. Depending on the nations for this phenomenon have been given
type of the individual sigma factor, either (DYSON and SCHREMPF, 1987; ALTEN-
general (e.g., the factors needed for vegeta- BUCHNER, 1994). The occurrence of extracel-
tive growth) or specialized genes (e. g., those lular plasmids containing transposon struc-
responsible for secondary metabolism and cy- tures and IS elements was discussed initially.
todifferentiation) can be transcribed. In These could be integrated into the genome
Streptomyces griseus MARCOSet al. (1995) and induce genomic rearrangements and gene
identified three sigma factors differentially disruptions (HORNEMANN et al., 1993).
expressed under specific nutritional condi- Streptomycetes contain only one single lin-
tions. The sigma factors whiG and sigF, each ear chromosome (8 Mb) (ALTENBUCHNER,
controlling certain events in the development 1994; CHENet al., 1994; REDENBACHet al.,
of spore chains in Streptomyces coelicolor, are 1996). Gene mapping experiments, comple-
controlled by transcriptional and posttran- mentation of blocked mutants, and heterolo-
scriptional events involving additional pro- gous expression of genes in different Strepto-
teins (KELEMENet al., 1996). myces hosts have shown that the genes of sec-
Recent approaches of molecular genetics ondary metabolite production are localized
showed that DNA-binding protein factors are on chromosomal gene clusters (HOPWOODet
crucial for the transcription of both eukaryot- al., 1983; LIU et al., 1992; STUTTARDand
ic and prokaryotic genes (HORINOUCHI and VINING,1995). Clusters which are responsible
BEPPU,1992b; CHATER,1992; THENBERGet for the polyketide and aminoglycoside syn-
al., 1996). They often occur as dimers and theses contain the genes of self-resistance
stimulate activity by binding to particular pro- protecting against the toxicity of the own sec-
motor regions. An example is the regulatory ondary metabolite (SENOand BALTZ,1989).
system of the y-butyrolactones (A-factor) in- Moreover, regulatory gene products are in-
volving proteinaceous transcriptional activa- volved which integrate secondary metabolism
into cellular developmental regulations thesis. Accordingly, leucyl-tRNA could signi-

(DISTLERet al., 1988; HORINOUCHI and BEP- fy a marker governing gene transcription of
PU, 1990 BEPPU,1992). differentiation-dependent pathways, at least
An organizational principle is the forma- in streptomycetes (DAVIESand CHATER,
tion of large “amplified” genomic structures 1992; CHATER,1992).
(DYSON and SCHREMPF,1987). Such se- As a summary, initial factors suppressing or
quences could be used as amplifying tools for stimulating cytodifferentiation and secondary
biosynthetic pathways in the future. A few ex- metabolism of the microbes are excessive nu-
amples demonstrate that the enlarged “gene trients converted to regulatory metabolites,
dosage” contributes to improved drug pro- nutrient downshifts, the accompanying
duction in high-yielding strains (TURNER, changes of general regulatory metabolites
1992; MARTfN and GUTIERREZ, 1995). COm- (such as ppGpp), and accumulations of pre-
mercial penicillin producing strains derived in cursors due to metabolic imbalances. Low
decades of random selections may contain and high molecular-weight mediators are
more than 20 copies of the biosynthetic clus- needed to trigger the coordination of numer-
ter linked by conserved hexanucleotide spac- ous pathways and cellular events in cytodif-
er elements (FIERROet al., 1995). The fre- ferentiation. The A-factor and similar y-buty-
quent deletion of the cluster has been related rolactones are mentioned here as a particular-
to these hot spots of recombination, since ly intriguing example of how the complex de-
non-producer mutants have lost the entire re- velopmental programs are organized in strep-
gion within these boundaries (FIERROet al., tomycetes (HORINOUCHIand BEPPU, 1990,
1996). 1992a, b; BEPPU,1995).
In the event of transpositions, frameshift
mutations may lead to disruptions of both
structural and regulatory genes (gene dele-
tions) (HORNEMANN et al., 1993). Activation 2.2.5 The A-Factor and the Signal
of “silent” gene sequences may occur in mu- Cascade of Cytodifferentiation in
tant strains because of the same reason. Thus,
regeneration of protoplasts or curing of plas- Streptomy ces
mids may yield mutants of streptomycetes ex-
periencing a completely altered pattern of The A-factor and its dihydro derivatives
secondary metabolism (see, e. g., the forma- (Fig. 1) are formed by numerous streptomy-
tion of curromycin, indolizomycin, and iremy- cetes. In some Streptomyces strains such as S.
cin) (OGARA et al., 1985; OKAMIet al., griseus and S. virginiae y-butyrolactone auto-
1988). regulators are required as a kind of a micro-
bial hormone for antibiotic production and
even for sporulation (PLINERet al., 1976; Ho-
2.2.4 Developmental Processes RINOUCHI and BEPPU,1990; ISHIZUKA et al.,
1992).
Another feature of regulation of gene ex- In 1975, KHOKHLOV and coworkers (PLIN-
pression in the development of streptomy- ER et al., 1976) found an idiotrophic mutant
cetes implies regulatory genes, such as whi, strain of S. griseus which neither produced
bld, afs, and abs in S. lividans and S. coelico- streptomycin (an aminoglycoside antibiotic)
lor (HOPWOOD,1988 DAVIESand CHATER, nor formed aerial mycelia and spores. It re-
1992). Thus, deletion of the whiB gene causes gained normal cytodifferentiation in the pres-
the concomitant loss of aerial mycelium for- ence of the culture liquid of the parental
mation. BldA was shown to specify the leu- strain. Later, the structure of the “autoregula-
tRNA (UUA codon). Further evidence sug- tory” A-factor and a series of 2‘-dihydro der-
gested that ‘ITA codons in the DNA are ab- ivatives was elucidated (PLINERet al., 1976;
sent from all genes involved in vegetative ISHIZUKAet al., 1992). The A-factor (2R-hy-
growth, but are present in the regulatory or droxymethyl - 3 - oxocaproyl) - y - butyrolactone,
resistance gene clusters of antibiotic biosyn- its homologs and analogs are capable not only
to induce streptomycin biosynthesis and aer- 1995). Gene disruption of afsK in S. coelico-
ial mycelium formation but also daunorubi- lor caused the reduction of actinorhodin for-
cin, virginiamycin, and carbapenem produc- mation without effecting growth. Residual
tion in other Streptomyces strains, biolumi- biosynthetic activities may be regulated by
nescence in Vibrio fischeri, nodulation of other kinases and the two-component system
plant associated bacteria, and toxin produc- ufsQllafsQ2 controlling actinorhodin produc-
tion in Pseudomonas aeruginosa (BEPPU, tion in S. coelicolor.
1995). So the phosphorylated afsR protein seems
For the dihydro derivatives of the A-factor to bind to regulatory DNA sequences near
isolated from Streptomyces viridochromo- the ufsA gene (in S. griseus), act genes (in S.
genes and S. bikiniensis, the 2R,3R- and the coelicolor), and red genes (in S. lividans) and
2S,3R-configuration was initially proposed to enhance their transcription. The afsA gene
(SAKUDAand YAMADA,1991) but later the encodes for the biosynthesis of A-factor-like
absolute stereochemistry was established as molecules, which is accomplished by the fu-
2S,3R,2' R and 2S,3R,2 S, respectively (YA- sion of phosphorylated glycerol and P-keto-
MADA et al., 1987; SAKUDAet al., 1992; LI et fatty acids (SAKUDAet al., 1992). Intracellu-
al., 1992). The latter structure, but not the A- lar recognition of the A-factor occurs via an
factor-type 2 '-0x0-butyrolactones, induce the A-factor binding protein acting as the repres-
production of the peptide antibiotic virginia- sor of the X-gene (HORINOUCHI and BEPPU,
mycin by S. virginae (virginiae butanolides) 1992a). Inactivation by A-factor thus permits
(YAMADAet al., 1987). Recently, 2'-deoxy formation of the X-protein acting as a tran-
derivatives (NFX factors) were even shown to scriptional enhancer of the strR and aphD
stimulate virginiamycin production in the genes. While AphD is responsible for the self-
same manner, and NFX-2 ((2R,3R,4S)-2-hex- resistance of the producer strain to strepto-
yl-3-hydroxy-4-pentanolide) proved to be mycin, strR appears as a transcriptional anti-
identical with blastomycinol lactole (a compo- terminator of streptomycin biosynthesis. Al-
nent of antimycin A l ) (YAMADAet al., 1987; though the scheme is still incomplete it sug-
KIMet al., 1990 OKAMOTO et al., 1992). gests that numerous events of sporulation and
Thus y-butyrolactones play an outstanding secondary metabolism could be governed by
role as regulatory signals inducing cytodiffer- afsR, X- and strR gene products in a con-
entiation and formation of quite different sec- certed manner.
ondary metabolite structures such as amino- In analogy to S. griseus, a virginiae butanol-
glycosides, polyketides, and peptides (HORI- ide (VB-C) binding protein (Mr 36000 Da)
NOUCHI and BEPPU,l990,1992a, b SAKUDA was isolated from S. virginiae which is sug-
et al., 1992; BEPPU,1995). gested to be involved in the mechanism of
Genetical and biochemical experiments pleiotropic signal transduction. The binding
contributed much to the present knowledge activity of this protein towards VB-C de-
of the regulatory cascade of cytodifferentia- creased by 40% in presence of DNA in a sim-
tion of S. griseus which involves the A-factor ilar manner as shown for other regulatory
and its congeners as signal transmitters proteins and transcriptional factors (KIM et
(Fig. 2). AfsR (a 100 kDa protein encoded by al., 1990; SAKUDAet al., 1992). The pertinent
the afsR gene containing ATP and DNA- gene was sequenced and displayed considera-
binding domains) represents an early event in ble homology (6244%) to the amino acid se-
the cytodifferentiation of S. griseus. It is ac- quences of ribosomal protein L ll of diverse
tive in its phosphorylated form AfsR-P as a origins (rpIK) and to the essential protein of
transcriptional activator of several other E. coli. This suggested it to be a part of an
genes and it can be phosphorylated by afsK, a essential gene cluster encoding general com-
respective kinase. The N-terminal region of ponents of the transcriptional and translation-
this kinase shows significant similarity to oth- al systems. Exogenously added factors have
er Ser/Thr kinases including the P-adrenergic been used here to improve metabolite pro-
receptor kinase, the Rous sarcoma oncogene duction in the case of virginiamycin (YANGet
product, and a Myxococcus enzyme (BEPPU, al., 1995a, 1996a).
Possibly, the afsR gene of S. griseus is also tors. They possess enzymic domains at the in-
controlled by other genes which have not ner site of the membrane (MEIGHEN,1991;
been identified so far. The whiB gene of S. FUQUA et al., 1994; GEIGER,1994).
coelicolor, e. g., is responsible for early sporu- Early evidence for autoregulatory functions
lation events due to the formation of a small of special metabolites in the differentiation
transcription factor-like protein which is dis- and diploidization was presented for a series
pensable for growth, but essential for sporula- of fungi (GOODAY,1974; ZAKELJMAVRIC et
tion (DAVIESand CHATER,1992; CHATER, al., 1995). In some molds there are sex hor-
1992). mones like antheridiol, sirenin, oogoniol, and
Moreover, S. griseus mutants which were trisporic acids (Fig. l), which trigger zygos-
recently investigated, produce the A-factor pore formation and the subsequent exchange
but nevertheless miss the normal sporulation of genetic material. In the aquatic fungus
behavior (MCCUEet al., 1992). An open cod- Achlya the signaling chain of the fungal sterol
ing gene sequence ( O W 1590) was identified antheridiol displays similarity to mammalian
which is possibly responsible for the synthesis cells. Here the response to steroidal sex hor-
of two polypeptidic transcription factors (P 56 mones is also mediated by membrane recep-
and P49.5). Dimerization of P5 6 was sug- tors (ZAKELJMAVRIC et al., 1995).
gested to induce the onset of sporulation, but
P49.5 prevents this event. In the above mu-
tant imbalanced regulation of the syntheses of
P 56 and P 49.5 have been proposed to cause 2.2.6 Overproduction of Microbial
lack of sporulation. Secondary Metabolites and
As another type of event, ADP-ribosyla-
tion of proteins catalyzed by NAD-glycohy- Precursor Pools
drolase and ADP-ribosyltransferase seems to
participate in cytodifferentiation of S. griseus. Much experience has been obtained in the
Failure to ADP-ribosylate certain cellular past with empirical selections of high-yielding
proteins in mutant strains was thought to strains of antibiotic producing microorgan-
cause impaired differentiation (SZESZAKet isms. Comparison of the high-producing mu-
al., 1991; OCHIet al., 1992). tants of streptomycetes and fungi with the
A y-butyrolactone derivative, phydroxy- low-yielding wild-type strains suggested that a
butyryl-homoserine lactone, is the autoinduc- series of heritable metabolic changes had
er of light emission by Vibrio hurveyi been introduced (OCHIet al., 1988; VANEK
(MEIGHEN,1991; WILLIAMS et al., 1992; Fu- and HOSTALEK,1988), for instance:
QUA et al., 1994; GEIGER,1994). Similar to
other photobacteria, luminescence is strongly - the elimination of “bottle-necks’’ in the
influenced by the density of the cell culture. production of biosynthetic precursors,
V. hurveyi synthesizes the above small extra- - the suppression of negative catabolite regu-
cellular molecule, which accumulates in the lations concomitant with increased produc-
growth medium and induces luminescence by tion of synthetases,
luciferase and FMNH2-coupled oxidation of a - improved resistance of the producer strain
long-chain fatty aldehyde. Vibrio fischeri against its own toxic product, and
forms a similar autoinducer, P-ketocaproyl - the absence of negative feedback regula-
homoserine lactone (FUQUAet al., 1994). tion of the formed secondary metabolite on
Previously, similar molecules have been re- its biogenesis.
ported to regulate carbapenem biosynthesis
by Erwiniu curotovora (BAINTON et al., To realize these prerequisites of high pro-
1992). ductivity, the natural regulatory mechanism
The signaling pathway of light emission of the wild-type strains, permitting only little
which is induced in presence of the above product formation had to be altered in a step-
mentioned butyrolactones seems to involve by-step selection procedure. The alterations
transmembrane signaling proteins as recep- concern both the genetic and the physiologi-
cal system of the pertinent strain, the second- inhibits and suppresses homocitrate synthet-
ary pathways, and the cellular morphology ase in the low-producing strains as a negative
(VANEKand MIKULIK, 1978). feedback regulator. The high-producing
Many of the high-producing strains over- strains display greatly reduced sensitivity to
produce the pertinent precursors. An exces- lysine (MARTIN and DEMAIN,1980). This
sive precursor supply thus appears to deter- “branched-pathway’’ model of regulation was
mine high secondary metabolite production. also reported for the biogenesis of candicidin
Moreover, when several alternative precur- by S. griseus. It is reduced by excessive tryp-
sors can be used by the same biosynthetic tophan in the medium due to the feedback in-
pathway the availability of the individual pre- hibition of the p-aminobenzoic acid synthe-
cursors governs the quality of formed prod- tase (MARTIN,1978).
ucts. Wild-type strains often produce a series L-cysteine needed for p-lactam production
of homologous structures due to the usage of can be produced either from sulfide and 0-
several intracellularly supplied precursors acetylserine or by reverse transsulfuration of
(SANGLIERand LARPENT, 1989). During 0-acetylhomoserine using L-methionine as a
strain improvement by mutagenesis and selec- donor of sulfur. In P. chrysogenurn (forming
tion empirical pathway engineering was done. penicillin G ) cysteine is produced mainly by
Sometimes, the selection promoted excessive the sulfate reduction pathway, in Acremon-
formation of a single precursor and, conse- ium chrysogenum (producing cephalosporin
quently, a single product was formed instead C) via transsulfuration (MARTIN,1978). In
of a series of homologous structures (CLAR- the latter strain, feeding of L-methionine
IDGE, 1983; THIERICKE and ROHR,1993). highly stimulates cephalosporin biosynthesis
“Precursor-directed biosynthesis”, “muta- concomitant with the formation of arthro-
tional” and “hybrid” biosyntheses signify mi- spores in submerged fermentations (MARTIN
crobiological techniques (CLARIDGE,1983; et al., 1986).
THIERICKE and ROHR, 1993), which have High-producing strains were shown to syn-
successfully been used in the past to alter thesize precursors by particular metabolic se-
product formation by excessive feeding of quences. Carboxylation of acetyl coenzyme A
precursors or biosynthetic intermediates to by oxaloacetate to yield malonyl coenzyme A
parental strains and their mutants. Even and the activation of D-glucose by polyphos-
when structural analogs of the special precur- phate glucokinase are characteristics of some
sor were fed to the medium they could be streptomycetes (QUEENERet al., 1986; VA-
used as a substitute of the natural structure. NEK et al., 1978). These peculiar pathways en-
In this manner, the formation of many new hance precursor supply in the biosyntheses of
and unusual secondary metabolites was de- tetracyclines, erythromycin, and macrolide
monstrated (SHIERet al., 1969). polyenes.
During the rapid (balanced) growth of mi- Compartmentation of the precursor- and
crobial cultures no excess of intermediary me- energy-generating metabolism plays an im-
tabolites is available, but when some sub- portant but yet incompletely understood role
strates become rate-limiting while others are in eukaryotic microorganisms. The biosynthe-
still available a metabolic imbalance arises sis of benzodiazepines by Penicillium cyclo-
which promotes the accumulation of precur- pium depends on precursor pools stored with-
sors (imbalanced growth) (DEMAIN,1974, in vacuoles. Their membranes become
1992). Apparently, the size of precursor pools permeable during the production phase due
is of regulatory importance in secondary me- to the appearance of a particular permeabiliz-
tabolite formation and determines the pro- ing factor (Roos and LUCKNER,1986).
duction rate.
Investigations of the plactam biosynthesis
illustrate well that penicillin formation by P.
chrysogenum is subject to negative feedback
control by L-lysine, and to a lesser extent by
L-valine (MARTINet al., 1986). The former
2.2.7 Biotechnical Production of 3 The Biosynthetic

Secondary Metabolites
Pathways
For more than 50 years semi-empirical
rules determined the scaling-up of microbial 3.1 Precursors and the Main
procedures for the production of secondary Biosynthetic Pathways
metabolites.
Maximum production rate of a given sec-
ondary metabolite usually is attained below Secondary metabolites are formed from
the maximum growth rate. Consequently, fer- few starter molecules acting as precursors.
mentations are carried out under partial sub- They will either be modified to yield new
strate limitation. Mostly, complex nutrient chemical derivatives of the initial molecule or
sources are employed or slow feeding of sub- they will be coupled to oligomeric material
strates such as glucose which cause a vigorous which is subsequently modified. An outstand-
development of biomass concomitant with ca- ing variability of structures arises from the
tabolite repression of secondary pathways. latter biosynthetic principle which combines
But, an optimized fermentation process is homologous and even heterologous building
characterized by moderate development of moieties in a polycondensation process.
biomass. Secondary metabolism thus occurs Moreover, oligomeric structures once formed,
parallel to submaximal but continuous such as the aglycones of macrolides, angucy-
growth. A major goal of the bioengineer is to clines, and anthracyclines, can be linked to
grow high concentrations of producing bio- other moieties like biosynthetically modified
mass in the fermenter. Finally, the available sugars.
oxygen concentration in the fermenter is the Only a few precursor structures are used in
critical value for high productivity (CALAM, secondary metabolite formation: coenzyme A
1987; FIECHTER, 1988). Oxygen intake is de- derivatives of lower fatty acids (acetyl-, pro-
pendent on fermenter geometry and impeller pionyl-, n- and isobutyryl-CoA, etc.), meva-
performance. lonate (also derived from acetyl-CoA), amino
Promotion of impeller speed increases acids and shikimate, sugars (preferably glu-
shear stress of the mycelia and causes frag- cose), and nucleosides (purines and pyrimid-
mentations and reduction of the mycelial pro- ines). These precursors are also needed in
duction rate. More than other microbial pro- primary metabolism to form cellular materials
cesses fermentations of secondary metabol- such as proteins, nucleic acids, cell wall con-
ites require producer strains displaying an op- stituents, and membrane lipids (ZAHNERand
timal morphological behavior under the given ZEECK,1987). Numerous biotransformations
technical conditions. Changes of the mycelial of single molecules are known, but oligomeri-
morphology not only cause alterations in the zations of the above mentioned basic struc-
rheological behavior of the fermentation tures only occur by three pathways: glycosyla-
broth, but also affect the intake of oxygen tion of activated sugars and polycondensa-
into the culture. Moreover, nutrient penetra- tions involving either activated fatty acids or
tion into the cells is affected by the formation amino acids.
of pellets and mycelial aggregations (STEELE
and STOWERS,1991).
Another serious problem in large-scale bio- 3.2 Secondary Metabolites Formed
technical production of secondary metabol- through Biosynthetic Modifications
ites is heat formation. Maintenance metabol-
ism of high biomass concentrations burns a of a Single Precursor
great part of substrate without product for-
mation. Hence, strains selected for low heat Structurally modified monosaccharides,
production appear particularly promising. such as valienamin in acarbose (MULLER,
1989), are derived from glucose via a series of
detectable intermediate of erythromycin bio- system is formed from the same intermediate
synthesis. Altered structures of polyketides polyketide (ROHR et al., 1993). Obviously,
may be engineered by point mutations within daunomycin, tetracyclines, tetracenomycines,
functional domains (KATZ and DONADIO,and some angucyclines arise from nonaketide
1995), by positional alterations of domains, precursors which are cyclicized in a quite dif-
e.g., the terminating thioesterase domain ferent manner in the course of polyketide
(WIESMANN et al., 1995), or by domain ex- processing (Fig. 6). Various successful at-
changes (BEDFORDet al., 1996; OLIYNYKet tempts have been made to deduce the func-
al., 1996). tions of proteins detected in type I1 polyke-
The genes of the aromatic type I1 polyketide biosynthetic clusters (KIM et al., 1995).
tide synthases from different streptomycetes This has led to the concept of a minimal poly-
display extensive sequence homology suggest- ketide forming system containing the con-
ing only minor differences in the substrate densing enzyme, the acyl carrier protein, and
specificity and in the sequence of reactions a malonyl-CoA transferase (MCDANIELet
(O'HAGAN, 1991; DONADIOet al., 1991; al., 1994). Additional proteins may then func-
HOPWOODand KHOSLA,1992). But the indi- tion as chain length factors determining the
vidual manner of folding of the intermediate number of elongation steps and as cyclases di-
enzyme-bound polyketides determines in a recting the mode of cyclization (HUTCHINSON
large measure what kind of cyclic aromatic and FUJII,1995). A number of new polyke-
one sugar
t 1 to 3 S U M R
OM.
Ho 0 O H 0 rug.r
E
r
tetracyclines anthracyclines
additiorurl
lactone structures
In the macro-
oligdides \ 0 -to 3
integratedrings
and hemlketd
glycosylation by 0 to 2 sugars
* Ito 8 tetrahydro-
structures pyranyi and tetra-
n alkyls
10 to 80 memberedring
c= 0
'
c=c -W
/
up to seven
conjugated
doublebonds
Ito 3 suga1 " insteadof0
macrolides and polyenes polyethers
Fig. 6. Variations of polyketide structures (tetracyclines, anthracyclines, macrolides, polyethers) occurring
in microorganisms. The substituents may vary in dependence of the given compound. The polyether struc-
ture shown above is highly variable with regard to the arrangement of the structural elements (rings,
hydroxyketo structure, substituents).
tides have been formed by new strains with as 1-0-dTDP and 1-0-dUDP derivatives and
various combinations of minimal systems and mutual coupling to other activated sugars
factors leading to first combinatorial biosyn- generate more than 200 oligosaccharide struc-
thetic approaches (TSOIand KHOSLA,1995; tures in actinomycetes (BERDYet al., 1980
KAO et al., 1995). Without detailed structural BYCROFT,1988; LAATSCH,1994). Aminocy-
knowledge of the proteins involved the re- clitols and other secondary metabolites thus
sults remain highly unpredictable (MEURER originate from a few sugar moieties (HOITA
and HUTCHINSON, 1995), but exciting proce- et al., 1995). The biosynthetic pathways lead-
dures for the generation of new compounds ing to some therapeutically important repre-
have been opened up (HUTCHINSON, 1994). sentatives of sugar-derived structures such as
streptomycin, kanamycin, and lincomycin
have been investigated in detail (WRIGHT,
3.4 Terpenes 1983; PIEPERSBERG,1994, 1995; PIEPERS-
BERG and DISTLER,Chapter 10, this volume).
A plethora of mono-, sesqui-, di-, and tri- L-Glucosamine, streptidine, and L-streptose
terpenoid structures of secondary metabolites as constitutive parts of the streptomycin mol-
are formed from acetyl coenzyme A via me- ecule are formed via three independent, mul-
valonate and isopentenyl pyrophosphate. The tistep pathways. Thus, dTDP-L-dihydrostrep-
initial steps of their biosynthesis (e. g., forma- tose formation is started from 1-0-dTDP-glu-
tion of Phydroxy methylglutaryl coenzyme cose followed by dehydratation, 35-epime1-i-
A, isopentenyl pyrophosphate, geranyl pyro- zation, and reduction in an initial series of
phosphate, farnesyl pyrophosphate) are the reactions (WRIGHT, 1983; PIEPERSBERG,
same as in the formation of triterpenoid ste- 1994, 1995). Streptidine is synthesized by S.
roids and hopanoids as essential cellular con- griseus from glucose via a series of at least
stituents of fungi and bacteria (CANE et al. twelve enzymic steps. By linkage of the three
1992; CANE,1992, 1995). Terpenoid second- subunits hydrostreptomycin-6-0-phosphate is
ary metabolites frequently occur as secondary formed intracellularly which is inactive as an
metabolites in plants and fungi, but they are antibiotic (PIEPERSBERG,1994). During its
rather unusual in bacteria (see, e. g., pentale- transport through the cytoplasmic membrane
nolacton, arenaemycin) (BERDYet al., 1980). outside the cells oxidation occurs and phos-
Final steps of fungal terpenoid biosynthesis phate is split off to yield the active streptomy-
(e. g., trichothecens, germacrine, aristolo- cin (WALKERand WALKER,1978). Strepto-
chene, etc.) are carried out by specialized mycin biosynthesis was studied in more detail
cyclases (CANE,1992). Many cyclizations in- by the investigation of the pertinent genes
volve the protonation or alkylation of a dou- and the corresponding enzymes (PIEPERS-
ble bond or an epoxide and the ionization of BERG, 1994, 1995; PIEPERSBERG and DIST-
an allylic diphosphate ester. Thereafter, car- LER, Chapter 10, this volume). It provides a
bocationic intermediates are formed by the nice example of the formation of sugar-de-
electrophilic attack of the resulting species to rived secondary metabolites. Moreover, the
an olefinic bond followed by proton elimina- same biosynthetic mechanism, steps of sugar
tion and a reaction with water as a nucleo- activation and transformation, are suggested
phile. A series of terpenoid cyclases have to be involved in the formation of mixed-type
been investigated recently by labeling and structures such as, e.g., macrolides (KATZ
gene cloning experiments (CANEet al., 1992; and DONADIO,1995), anthracycline antibio-
CANE,1992, 1995). tics (HUTCHINSON, 1995), and glycopeptides
(ZMIJEWSKI and FAYERMAN, 1995; LANCINI
and CAVALLERI, Chapter 9, this volume).
3.5 Sugar-Derived Oligomeric
Structures
Biotransformations of simple monosaccha-
rides, their activation as 1-0-nucleosides such
3.6 Oligo- and Polypeptides tion of the peptidic bonds occurs through
translocation of the growing nascent peptide
Three ways of peptide bond formation are chain involving a phosphopantothenoyl car-
known in secondary metabolism (KLEINKAUF rier moiety (Fig. 7).
and VON DOHREN, 1990, 1996): The terminating reactions are carried out
by specified enzymic subunits of the same
- coupling of amino acids by single enzymes multienzyme complex. Cyclizations can occur
to form small peptides with up to five ami- to form cyclo- and depsipeptides as well as re-
no acids (e. g., glutathione, peptidoglycan), ductions, oxidations, and methylations which
- non-ribosomal biosynthesis of larger pep- introduce, e.g., a disulfide bond (see, e.g.,
tides (containing up to about 50 amino triostins) (VON DOHREN, 1990 BERDYet al.,
acids) by multienzyme complexes, and 1980) or reduce a carboxylic acid to the perti-
- ribosomal mechanisms. nent aldehyde (see, e. g., pepstanone) (BER-
DY et al., 1980).
Oligopeptide biosynthesis on multienzyme A major difference of template-directed
complexes (as the most important mecha- mechanisms as compared to the ribosomal
nism) has been described for many bacterial formation of peptidic bonds is the acceptance
products such as gramicidins, bacitracin, tyro- of non-proteinogenic amino acids and even of
cidin, and fungal secondary metabolites such hydroxy acids and fatty acids either as build-
as enniatins and cyclosporins (KLEINKAUF ing blocks of the oligomer formation or as
and VON DOHREN, 1987,1990,1996). carbon and nitrogen terminal substituents
The individual amino acids are first acti- (KLEINKAUF and VON DOHREN,1987; VON
vated via adenylate formation and thereafter DOHREN, 1990). This peculiarity of the
are bound as thioesters to the non-ribosomal non-ribosomal mechanism contributes in a
synthase multienzyme complex. Subsequent- particular manner to the structural diversity
ly, they are coupled in a step-by-step proce- of low-molecular weight peptides produced as
dure to form large polypeptides which are secondary metabolites by so many microor-
sometimes composed of several subunits. The ganisms (BERDYet al., 1980 BYCROFT,1988;
sequence of the amino acids in the peptide is LAATSCH,1994).
exactly the same as that of the amino acids Genetic analysis of peptide forming en-
activated on the multienzyme complex (“thio- zyme systems has revealed a modular struc-
template-directed non-ribosomal peptide syn- ture of the enzymes involved. As in the case
thesis on a protein matrix”). Stepwise forma- of polyketides various degrees of integration
Condensation domain (optional)
Carrier domain 1
ondensation domain 1
start site (P)
Z L,A-site~ { Carrier
T ~ domaindomain
2 2
P-site
-&4
t-* -site
domain 3
Fig. 7. Schematic view of

limeization domain the multiple carrier pro-
tein model of enzymatic
I I nidsterase domain peptide formation (thio-
Exit site template model).
are found, with eukaryotic systems generally which are unable to carry out the complete
being fully integrated (KLEINKAUF and VON biosynthetic pathway (SHIER et al., 1969;
DOHREN,1996, and Chapter 7, this volume). CLARIDGE,1983; THIERICKE and ROHR,
Genetic exchange of modules specifying ami- 1993). Biosynthesis is initiated, again, when
no acids or related substrates in the protein the missing intermediate is fed to the me-
code may lead to new peptides of altered dium. Feeding of chemically derived analogs
composition (STACHELHAUSet al., 1995a, of the pertinent intermediate can yield new
b). structural variants of the initial products. This
Polypeptide-type secondary metabolites technique was invented already in 1969 by
such as, e.g., microcins, tendamistat, subtilin, SHIER(SHIERet al., 1969), and in a few cases
and lantibiotics (epidermin, gallidermin, ni- (avermectins, cyclosporins) (see, e. g., DUT-
sin) are biosynthesized in microorganisms on TON et al., 1991) more powerful compounds
the ribosomes as larger prepeptides. During were obtained. Similar to the mutational bio-
their export into the medium, proteolytic synthesis the “hybrid biosynthesis” employs
processing occurs to yield the bioactive struc- idiotrophic mutants which are blocked in a
tures. A series of posttranslational altera- particular step of the secondary biosynthetic
tions, such as the linkage to chromogenic and pathway (SADAKANE et al., 1983). Some of
other groups, the formation of lanthionine, them accumulate intermediates of the inter-
methyl lanthione, and disulphide units in- rupted biosynthetic chain due to the lack of a
creases the number of possible homologs and transforming enzymic step. Such kinds of in-
creates the bioactive structures (SAHLet al., termediates (e.g., the protylonolide from
1995; GASSON,1995; JACKet al., Chapter 8, Streptomyces fradiae) were fed to blocked
this volume; MORENOet al., 1995). mutants of another strain missing the forma-
tion of a similar intermediate (e. g,. spiramyci-
no lactone in idiotrophs of Streptomyces am-
3.7 Biosynthetic Modifications of bofaciens forming spiramycin). Sometimes
Structures and Precursor-Directed the fed heterologous metabolite (e. g., proty-
lonolide) can be used in the same manner as
Biosyntheses the native metabolite (spiramycino lactone).
In this way, chimeramycins were formed as
The secondary metabolism is carried out by hybrids of secondary metabolite structures
specified enzymes acting within the frame of from two different Streptomyces strains (SAD-
long biosynthetic chains. Modified structures AKANE et al., 1983).
can frequently be obtained due to the com- As was demonstrated with biosynthetic en-
parably low substrate specificity of some enzymes, e. g., acyltransferase and isopenicillin-
zymes (LUCKNER, 1989). In many cases, feed- N-synthase of Penicillium chrysogenum, cy-
ing of a tentative precursor molecule or inter- closporin synthase of Beauveria niveum, en-
ruption of its biosynthesis, e.g., by the addi- niatin synthase of Fusarium oxysporum, and
tion of metabolic inhibitors, has been used gramicidin S synthases of Bacillus brevis, di-
successfully to direct the secondary metabol- rected biosyntheses can also be carried out
ism toward the formation of one single com- very efficiently by cell-free enzymes (BALD-
ponent of a mixture of naturally occurring WIN et al., 1991; MARTINEZ-BLANCO, 1991;
metabolites (precursor-directed biosynthesis) LAWENand TRABER,1993; KLEINKAUF and
(SADAKANEet al., 1983; DUTTON et al., VON DOHREN,1996). The above biocatalysts
1991). Some of the producer strains even ac- convert a series of synthetic acyl coenzyme-A
cept structural homologs of the natural pre- derivatives and homologs of the ACV-tripep-
cursor to form unusual derivatives of the ori- tide to form novel penicillins which have not
ginal molecule(s) (SADAKANEet al., 1983; occurred as microbial products so far. Refer-
BALDWIN et al., 1991; MARTINEZ-BLANCO et ence should also be given here to the use of
al., 1991; LUENGO,1995). enzymes in biotransformations of secondary
The term “mutational biosynthesis” signi- metabolites (see, e. g., the enzymatic hydro-
fies the use of blocked “idiotrophic” mutants lyses of the side chains of penicillins and ce-
phalosporin C). The total synthesis of various terpenes by some plants). This is the reason
cyclopeptides and depsipeptides has been car- why the search for new structures turns to
ried out up to the milligram scale. unusual sources such as plants, animals, and
Moreover, growing evidence attests to the microorganismsfrom special ecosystems (e.g.,
outstanding possibilities of molecular genetics marine animals and bacteria, special fungi,
in the modification of already known struc- lichens, algae). Plants referred to in folk med-
tures, and in the generation of new structures icine and marine tunicates, toxic snakes, and
(HOPWOOD, 1989; HOPWOOD and SHERMAN, toads offer an advantageous field of research
1990; HUTCHINSON, 1994; HUTCHINSON and on new “leading” structures. Moreover, the
FUJII,1995). Genes of Sfrepfomyces type I1 biosynthetically available modifications of ba-
polyketide synthases have recently been sic structures such as macrolides, peptides,
transferred to other Sfreptomyces hosts, and polyethers, etc. follow distinct rules: some
the biosynthesis of new and modified aromat- derivatives occur frequently, but others are
ic structures (hybrid antibiotics) is being ex- very rare. In general, the anthracyclines, e. g.,
ploited (HOPWOOD,1989; HUTCHINSON and occur as glycosylated derivatives, whereas the
FUJII,1995). tetracyclines are usually non-glycosylated.
But previously, the dactylocyclins were de-
tected in cultures of Ducfylosporungiumsp. as
the first glycosylated representatives of the
tetracycline family (TYMIAK et al., 1993).
Otherwise, small structural changes of a
4 Variability of Structures given basic structure will often cause major
of Secondary Metabolites changes in biological activities. The macrolide
antibiotics from streptomycetes are an exam-
ple which are similar in structure but possess
4.1 Secondary Metabolites as antibacterial, antifungal, insecticidal, nemato-
Products of Biological “Unit cidal, immunosuppressant, and cytotoxic
properties. Traditional rescreening of com-
Operations” pounds in newly established biological
screens leads to the detection of unsuspected
Starting from a few molecular structures as biological activities. In addition, chemical
precursors, the secondary pathways of the mi- derivatization of side chains is an established
crobial kingdom produce much more than and especially effective procedure to arrive at
10000, and the secondary pathways of plants functionally improved structures.
produce more than 100OOO different chemical
individuals (VERALL,1985; GROOMBRIDGE,
1992). At first glance, this huge number ap- 4.2 Structural Classifications of
pears to be incredibly high, but the observer
soon recognizes that the majority of struc- Secondary Metabolites
tures are representatives of some few struc-
tural classes. Many homologs of a basic struc- The large number of known secondary me-
ture have been disclosed, not only in a given tabolites needs classification. This could be
strain but also in different species and genera achieved by considering their biosynthesis,
(BERDY et al., 1980; BYCROFT, 1988; the producing organisms (bacteria, fungi,
LAATSCH,1994). The detection of a novel plants, animals, etc.), their biological activi-
structural class of natural drugs structurally ties, and also their chemical structures. Few
unrelated to the already known compounds examples can be mentioned here to show how
appears to be rather rare. Some organisms the structural variability of secondary metab-
are characterized by the preferred production olites is channeled by classifications according
of a particular secondary class of metabolites to biosynthetic origin and chemical nature
(c. f., the frequent formation of polyene mac- (BERDYet al., 1980; LANCINIand LOREN-
rolides by streptomycetesor of sesqui- and di- ZE’ITI, 1993).
4 Variability of Structures of Secondary Metabolites 39
Peptides Moreover, one to three sugars are attached

to the non-polyene macrolide aglycones
Peptidic drugs are produced by numerous which are excessively substituted by hydroxy,
bacteria, fungi, plants and even animals (c.f. methoxy, methyl, and epoxy functions.
the magainins and other skin antibiotics of Even open-chain polyenic fatty acids (e. g.,
toxic toads) (JACOB and ZASLOFF, 1994). enacyloxin) are produced by strains which
Peptide antibiotics from microbial sources oc- cannot carry out the final step of lactonization
cur as linear homopeptides so far composed (WATANABEet al., 1992).
of a maximum of 45 amino acids (KLEINKAUF Structures of the antitumor anthracyclines
and VON DOHREN,1996). Substitutions by also demonstrate the diversity which has been
fatty acids are a characteristic feature of the introduced by a few modifications into a basic
lipopeptides. Many cyclic peptides are known structure of a tetrahydro naphthacenequi-
(c. f., the cyclosporins as undecapeptides) and none backbone. Up to ten sugars are linked
the amino acids are often replaced by a-and to several molecule positions. In addition, the
Phydroxy acids in an irregular or even a reg- number of hydroxy, carboxymethyl, and keto
ular manner (peptidolactones and depsipep- groups varies in the individual representatives
tides). to form approximately 300 different struc-
Even non-proteinaceous amino acids can tures.
be constitutive parts of peptidic drugs (KEIN- Many aromatic polycyclic compounds are
KAUF and VON DOHREN,1986, 1987, 1990). also derived from the polyketide pathway.
Small peptide chains can be linked to other During their biosynthesis ring closures involv-
unique structures such as fatty acids and chro- ing nitrogen and oxygen substituents are fre-
mophoric groups whereas combinations with quent features. In this way, even heterocyclic
sugars, macrolides, and anthracyclines seem structures such as carbazols, phenoxazins, and
to be unusual. phenazines are formed (BERDYet al., 1980
Variable structures have also been unrav- BYCROFT,1988; LAATSCH1994).
eled in the high-molecular weight peptide an- In fungi mycotoxins such as the aflatoxins
tibiotics from streptomycetes such as and ochratoxins are likewise polyketide
lantibiotics (SAHLet al., 1995; GASSON,1995; metabolites (TURNERand ALDRIDGE,1983;
JACKet al., Chapter 8, this volume) and en- BROWNet al., 1996).
zyme inhibitors (subtilin, streptinoplasmin,
etc.) (BERDYet al., 1980). The peptide chains
are formed by ribosomal mechanisms and Terpenoid Structures
posttranslational modifications create the in-
dividual bioactive structures. Rich sources are plants and fungi, while
terpenoid structures rarely occur in bacteria.
Characteristic fungal terpenoids are mycotox-
ins such as trichothecens (BERDYet al., 1980
Polyketide Drugs TURNERand ALRIDGE, 1983; LUCKNER,
1989). Important bioactive terpenoid struc-
Actinomycetes are rich sources of polyke- tures from plants are the vinca alkaloids (No-
tide metabolites like macrolides, polycyclic BLE, 1990 Fox, 1991) and taxol (HEINSTEIN
aromatic and semi-aromatic compounds like and CHANG,1994). Moreover, the triterpen-
tetracyclines, anthracyclines and angucy- oid steran backbone is widely distributed in
clines, polyethers, and ansamycins. natural structures such as digitalis glycosides,
The variability of macrolide structures in- saponins, etc. Fungal antibiotics such as fu-
volves ring sizes ranging from 10 to 60 (as re- sidic acid, cephalosporin P, azasterols, and
cently found in quinolidomycin) (HAYAKA- toad toxins (see, e. g., bufadienolides) are also
WA et al., 1993). Up to seven conjugated and representatives of the same structural class.
additionally isolated double bonds can be Marine tunicates offer a rich source of unique
present in the macrocycle (BERDY et al., steroid-type molecules (CAVALIER-SMITH,
1980; B Y C R O ~1988;
, LAATSCH,1994). 1992).
Oligoglycosides a growing business stimulating other fields of

biotechnology and biomedicine (VERALL,
Up to several hundred sugars (see, e. g., shi- 1985; MONAGHAN and TKACZ, 1990)
zophyllan, lentinan, avilamycins) can be (Tab. 1).
linked in a linear or even a cyclic manner. As far as the sources of new drugs are con-
Therapeutically useful oligomeric representa- cerned, only a small percentage of the pre-
tives of the oligosaccharides are the aminocy- sumed microbial world population has been
clitols which contain amino inositols such as explored so far, and only a minor part of the
streptidine (in streptomycin) and 2-deoxy- existing microbial strains has been deposited
streptamin (in neomycins, gentamicins, kana- in strain collections (CHICARELLI-ROBINSON
mycins, istamycins, etc.) (UMEZAWAand et al., 1994). The traditional sources of bioac-
HOOPER,1982; DIMITRIU, 1996). tive microbial metabolites, actinomycetes, ba-
cilli, and sporulating fungi, still appear prom-
ising. Special genera such as the Myxobacteria
Nucleoside Antibiotics (REICHENBACH and HOFLE,1993) have been
demonstrated as particularly rich producers
Approximately 150 nucleoside analogs are of unique structures. Future interest is fo-
known from natural sources like actinomy- cused on ecological “niches” and microhabi-
cetes and fungi. They are characterized by the tats which might harbor peculiar organisms.
presence of “false” nucleobases as aglycones They look promising because they could miss
and/or by “false” sugars (ISONO,1990, 1995). special metabolic control due to their particu-
The biosynthetic strategies imply their forma- lar adaptation to the natural environment. In
tion from a normal nucleoside (see, e.g., gua- this context, microorganisms from extremly
nosine in the biogenesis of tomaymycin) and/ poor grounds, marine systems, plant rhizo-
or sugars such as ribose. and endospheres became subject to detailed
These few examples mentioned above de- investigations (OMURA, 1992). Plants still
monstrate that the structural characteristics of provide an apparently inexhaustible reservoir
secondary metabolites can serve as a tool for of new bioactive structures. More than in the
their classification despite their outstanding past, increasing knowledge on the molecular,
diversity. Many compounds combine structur- cellular, and organismic causes of diseases
al elements of several basic classes as can be promotes the search for new drugs possessing
seen in Tab. 1. Thus, biosynthetic pathway more specific activity (TOMODAand OMURA,
enzymes of different types interact in a highly 1990; OMURA,1992; TANAKA and OMURA,
specific manner. Chapter 3, this volume). Today, the screening
assay determines what kind of novel natural
product will be detected. In the last decade,
an increasing number of publications on en-
zyme inhibitors and receptor antagonists at-
5 Future Perspectives: tests that classical screening for “simply” anti-
biotic molecules has been extended and ra-
New Products of the tionalized on the basis of modern and bio-
chemical pharmacy and molecular biology
Secondary Metabolism (see Tab. 1). Even viral targets such as, e.g.,
viral proteinase and adhesive proteins (GP
As DAVIDPERLMAN,one of the pioneers 120) became amenable to the search for new
of modern industrial microbiology once inhibitors.
stated microbial capacity is rather unlimited, The following screening assays are now
and if mankind asks the microbes the proper commonplace: Mammalian cell cultures to
questions they will truly answer. This idea ap- detect receptor agonists and antagonists, in-
plies to all of the living organisms when they ducers and inhibitors of cytodifferentiation,
are considered as a source of new bioactive and effectors of cell-to-cell and cell-to-virus
structures. Screening for new structures is still interactions. Mammalian cell lines trans-
6 References 41
formed by the expression of oncogenes are Acknowledgements

used in the search for new antitumor agents
promoting redifferentiation of cancer cells. A We are gratefully indebted to Dr. HANS-
particular advantage of these assays is that PETERSALUZfor helpful comments and crit-
they reduce animal trials which otherwise ical manuscript revision and to Dr. AXEL
would be necessary in the development of BRAKHAGE for helpful comments and part of
new pharmacological agents. In the same Fig. 3.
manner the use of plant and insect cell cul-
tures permits more rational and time-saving
drug discovery in screening for new phytoef-
fectors and insecticides.
Cloning of genes encoding for enzymes, re-
ceptors, protein factors, etc. and their expres- 6 References
sion in heterologous organisms supplies an-
other promising approach to drug discovery. AHARONOWITZ, Y., COHEN,G., MART~N, J. F.
The insertion of regulatory and reporter gene (1992), Penicillin and cephalosporin biosynthetic
sequences into microorganisms and cells may genes: structure, organization, regulation, and
be useful for the detection of compounds evolution, Annu. Rev. Microbiol. 46, 461495.
which interfere with DNA-binding proteins ALLEN,I. W., RITCHIE,D. A. (1994), Cloning and
and transcriptional regulators. In this way, analysis of DNA sequences from Streptomyces
hygroscopicus encoding geldanamycin biosyn-
new specific inhibitors of oncogenesis and vi- thesis, MoC. Gen. Genet. 243, 593-599.
ral replication may be uncovered. ALTENBUCHNER, J. (1994), Hohe genetische Insta-
As a conclusion, secondary metabolism in bilitat von Streptomyceten durch chromosomale
microbes, plants, and animals still promises Deletionen und DNA-Amplifikationen, Bio/En-
new leading structures for future drug devel- gineering 3, 33-46.
opment. This promise is due to the apparently ARISAWA,A., TSUNEKAWA, H., OKAMURA, K.,
inexhaustible pool of organisms and struc- OKAMOTO,R. (1995), Nucleotide sequence
tures and the rapid development in biomedi- analysis of the carbomycin biosynthetic genes in-
cal and biochemical disease research. More- cluding the 3-0-acyltransferase gene from Strep-
tomyces thermotolerans, Biosci. Biotech. Bio-
over, secondary metabolism supplies biochem. 59,582-588.
chemical tools which allow deeper insights ARMSTRONG, G . A. (1994), Eubacteria show their
into cellular processes (see, e.g., the previous true colors: genetics of carotinoid pigment bio-
discovery of inhibitors of protein kinases and synthesis from microbes to plants, J. Bacteriol.
protein phosphatases as effectors of the mam- 176,47954802.
malian cell cycle). It seems reasonable to be- ARROWSMITH, T. J., MALPARTIDA, F., SHERMAN,
lieve that the discovery of new leading struc- D. H., BIRCH,A., HOPWOOD, D. A., ROBINSON,
tures of antibiotics, anticancer and antiviral J. A. (1992), Characterization of actl-homolo-
drugs, pharmacological agents, crop protect- gous DNA encoding polyketide synthase genes
from the monensin producer Streptomyces cin-
ing and insecticidal compounds, etc. will con- namonensis, Mol. Gen. Genet. 234,256264.
tinue to promote the future development of ARST,H. N., JR. (1996), Regulation of gene expres-
biotechnology and medicine. Every new sion by pH, in: The Mycota III (ESSER, K.,
structure provides a challenge to the bio- LEMKE,P. A,, Eds.), Biochemistry and Molecu-
chemist exploring its mode of action, to the Car Biology (BRAMBLE, R., MARZLUFF,G. A.,
chemist wanting to disclose the structure-ac- Eds.), pp. 235-240. Berlin, Heidelberg: Springer-
tivity relationships, to the pharmacologist Verlag.
studying the activity in macroorganisms, to AZUMA,I. (1987), Developments of immunostimu-
the molecular biologist investigating the lants in Japan, in: Immunostimulants Now and
Tomorrow (AZUMA,I., JOLLES,G., Eds.), pp.
genes of the biosynthetic pathway, to the fer- 41-56. Tokyo, Berlin: Japan Scientific Press/
mentation engineer developing a new bio- Springer.
technical procedure, and last but not least, to AZUMA,M., HORI, K., HASHI,Y.-O., YOSHIDA,
those who could benefit for their health from M., HORINOUCHI, S., BEPPU,T. (1980), Basidif-
new and better drugs. ferquinone, a new inducer of fruiting-body for-
42 I General Aspects of Secondary Metabolisn
mation of a basidiomycete Farvolus arcularius BEPPU,T. (1992), Secondary metabolites as chemi-

from a Streptomyces strain. 11. Structure of ba- cal signals for cellular differentiation, Gene 115,
sidifferquinone, Agric. Biol. Chem. 54, 1447- 159-165.
1452. BEPPU,T. (1995), Signal transduction and second-
BAILEY,A. M., KERSHAW, M. J., HUNT,B. A., PA- ary metabolism: prospects for controlling pro-
TERSON, I. C., CHARNLEY, A. K., REYNOLDS, S. ductivity, Trends Biotechnol. 13, 264-269.
E., CLARKSON, J. M. (1996), Cloning and se- BERDY,J., ASZALOS,A., BOSTIAN,M., MCNITT,
quence analysis of an intron-containing domain K. L. (1980), CRC Handbook of Antibiotic
from a peptide synthetase of the entomopatho- Compounds, 1980-1984. Boca Raton, FL: CRC
genic fungus Metarhizium anisopliae, Gene 173, Press.
195-197. BEYER,S., DISTLER,J., PIEPERSBERG, W. (1996),
BAINTON,N. J., STEAD,P., CHABRA,S. R., BY- The Str cluster for the biosynthesis of 5'-hydroxy-
CROFT,B. W., SALMOND, G. P. C., STEWARD, streptomycin in Streptomyces glaucescens
G. S. A. B., WILLIAMS, P. (1992), N-(3-Oxohex- Gla.O(ETH 22794) - new operons and evidence
anoyl)-L-homoserine lactone regulates carba- for pathway specific regulation by StrR, Mol.
penem antibiotic production in Envinia caroto- Gen. Genet. 250,775-784.
vora, Biochem. J. 288, 977-1004. BIBB,M. J., SHERMAN, D. H., OMURA,S., HOP-
BALDWIN,J. E., BRADLEY,M., ABBOTT,S. D., WOOD,D. A. (1994), Cloning, sequencing and
ADLINGTON, R. M. (1991), New penicillins from deduced functions of a cluster of Streptomyces
isopenicillin N-synthase, Tetrahedron 47, 5309- genes probably encoding biosynthesis of the po-
5328. lyketide antibiotic frenolicin, Gene 142, 31-39.
BARABAS,G., PENYIGE,A., HIRANO,T. (1994), BIRCH, A. W., ROBINSON, J. A. (1995), Polyethers,
Hormone-like factors influencing differentiation in: Genetics and Biochemistry of Antibiotic Pro-
of Streptomyces cultures, FEMS Microb. Rev. 14, duction (VINING,L. C., STUITARD, C., Eds.),
75-82. pp. 443476. Boston, M A Butterworth-Heine-
mann.
BARBER,J., CHAPMAN,A. C., HOWARD,T. D.,
BIRCH,A. W., LEISER,A., ROBINSON, J. A. (1993),
TEBB, G. (1988), Recycling of polyketides by
Cloning, sequencing, and expression of the gene
fungi: the degradation of citrinin by Penicillium
encoding methylmalonyl coenzyme A mutase
citrinum, Appl. Microb. Biotechnol. 29, 387-
from Streptomyces cinnamonensis, J. Bacteriol.
391.
175,3511-3519.
BEDFORD,D., JACOBSEN, J. R., Luo, G., CANE,D. BLACK,T. A., WOLK,C. P. (1994), Analysis of a
E., KHOSLA,C. (1996), A functional chimeric Het(-) mutation in Anabaena Sp strain PCC
modular polyketide synthase generated via do- 7120 implicates a secondary metabolite in the
main replacement, Chemistry & Biology 3, 827- regulation of heterocyst spacing, J. Bacteriol.
831. 176,2282-2292.
BECHTHOLD, A., SOHNG,J. K., SMITH,T. M., CHU, BORMANN,C., MOHRLE, V., BRUNTNER,C.
X., FLOSS,H. G. (1995), Identification of Strep- (1996), Cloning and heterologous expression of
tomyces violaceoruber Tii22 genes involved in the entire set of structural genes for nikkomycin
the biosynthesis of granaticin, Mol. Gen. Genet. synthesis from Streptomyces tendae Tii901 in
248,610-620. Streptomyces lividans, J. Bacteriol. 178, 1216-
BECK,J., RIPKA,S., SIEGNER,A., SCHILTZ,E., 1218.
SCHWEIZER, E. (1990), The multifunctional 6- BRAKHAGE, A. A., TURNER, G. (1995), Biotechni-
methylsalicylic acid synthase gene of Penicillium cal genetics of antibiotic biosynthesis, in: The
patulum. Its gene structure relative to that of Mycota IZ (ESSER,K., LEMKE,P. A., Eds.), Ge-
other polyketide synthases, Eur. J. Biochem. netics and Biotechnology (KUCK, U., Ed.), pp.
192,487498. 263-385. Berlin, Heidelberg: Springer-Verlag.
BBHAL,V., HUNTER,I. S. (1995) Tetracyclines, in: BRAKHAGE, A. A., VAN DEN BRULLE,J. (1995),
Genetics and Biochemistry of Antibiotic Produc- Use of reporter genes to identify recessive trans-
tion (VINING,L. C., STUTTARD,C., Eds.), pp. acting mutations specifically involved in the reg-
359-384. Boston: Butterworth-Heinemann. ulation of Aspergillus nidulans penicillin biosyn-
BENDER,C., PALMER,D., PENALOZAVAZQUEZ, thetic genes, J. Bacteriol. 177, 2781-2788.
A., RANGASWAMY, V., ULLRICH,M. (1996), BRAKHAGE,A. A., BROWNE,P., TURNER,G.
Biosynthesis of coronatine, a thermoregulated (1992), Regulation of Aspergillus nidulans peni-
phytotoxin produced by the phytopathogen cillin biosynthesis and penicillin biosynthesis
Pseudomonas syringae, Arch. Microbiol. 166, genes acvA and ipnA by glucose, J. Bacteriol.
71-75. 174,3789-3799.
6 References 43
BROWN,D. W., Yu, J.-H., KELKAR,H. S., FER- CHEN,C. W., LIN, Y.-S., YANG,Y.-L., Tsou, M.-
NANDES, M., NESBITT,T. C., KELLER,N. P., F., CHANG,H.-M., KIESER,H. M., HOPWOOD,
ADAMS,T. H., LEONARD, T. J. (1996), Twenty- D. A. (1994), The linear chromosomes of Strep-
five coregulated transcripts define a sterigmato- tomyces: structure and dynamics, Actinomyceto-
cystin gene cluster in Aspergillus nidulans, Proc. logica 8, 103-112.
Natl. Acad. Sci. USA 93, 1418-1422. CHEN,Q., WERTHEIMER, A. M., TOLMASKY, M.
BRUCKNER,B., BLECHSCHMIDT, D., SEMBDNER, E., CROSA,J. H. (1996), The AngR protein and
G. SCHNEIDER, G. (1990), Fungal gibberellin the siderophore anguibactin positively regulate
production, in: Biotechnology of Vitamins, Pig- the expression of iron-transport genes in Vibrio
ments, and Growth Factors (VANDAMME, E., anguillarum, Mol. Microbiol. 22, 127-134.
Ed.), pp. 383-429. London, New York: Else- CHIAO,J.-S., XIA, T.-H., MEI, B.-G., JIN, Z. K.,
vier. Gu, W.-L. (1995), Ansamycins, in: Genetics and
BUADES,C., MOYA,A. (1996), Phylogenetic analy- Biochemistry of Antibiotic Production (VINING,
sis of isopenicillin N synthetase horizontal gene L. C., STUTTARD,C., Eds.), pp. 477-498. Bos-
transfer, J. Mol. Evol. 42, 537-542. ton, MA: Butterworth-Heinemann.
BUSHELL, M. E., FRYDAY, A. (1983), The applica- CHICARELLI-ROBINSON, M. I., Fox, F., NISBETH,
tion of materials balancing to the characteriza- C. (1994), The value of microbial screening in
tion of sequential secondary metabolite forma- drug discovery (abstract), in: New Strategies in
tion in Streptomyces cattleya NRRL 8057, J. Gen. Searching Drug Leads, pp. 1-3. Airport Sympo-
Microbiol. 129, 1733-1741. sium, Frankfurt, 28 February.
BYCROFT,B. J. (1988), Dictionary of Antibiotics. CHU, Y. W., RENNO,D., SAUNDERS, G. (1995),
London, New York: Chapman and Hall. Detection of a protein which binds specificallyto
CALAM,C. T. (1987), Process Design and Control the upstream region of the pcbAB gene in Peni-
in Antibiotic Fermentations. Cambridge: Cam- cillium chrysogenum, Curr. Genet. 28, 184-189.
bridge University Press. CLARIDGE,C. A. (1983), Mutasynthesis and di-
CAMPBELL, W. C. (Ed.) (1989). Ivermectin and Av- rected biosynthesis for the production of new
ermectin. New York, Heidelberg: Springer-Ver- antibiotics, in: Basic Biology and New Develop-
lag. ments in Biotechnology (HOLAENDER, A., LAS-
CANE,D. E. (1992), Terpenoid cyclases: design and KIN,A. I., ROGERS,P., Eds.), pp. 231-269. New
function of electrophilic catalysts, in: Secondary York: Plenum Press.
Metabolites: Their Function and Evolution, Ciba COFFEY,T. J., DOWSON,C. G., DANIELS,M.,
Foundation Symposium 171 (CHADWICK, D. J., SPRATT,B. G. (1995), Genetics and molecular
WHELAN,J., Eds.), pp. 163-168. Chichester, biology of beta-lactam-resistant Pneumococci,
New York: J. Wiley & Sons. Microb. Drug Res. - Mechanisms, Epidemiology
CANE,D. E. (1995), Isoprenoid antibiotics, in: Ge- and Disease 1,29-34.
netics and Biochemistry of Antibiotic Production COMITBEDITORIAL(1992), Medicaments antibio-
(VINING,L. C., STUTTARD,C., Eds.), pp. 633- tiques. Traite de chimie therapeutique, Vol. 2.
656. Boston, MA: Butterworth-Heinemann. Paris: Tec & Doc Lavosier.
CANE,D. E., YANG,G., COATES,R. M., PYNN,H., COOK,R. J., THOASHOW, L. S., WELLER,D. M.,
HOLM,T. M. (1992), Trichodiene synthase. Syn- FUJIMOTO, D., MAZZOLA,M., BANGERA,B.,
ergistic inhibition by inorganic phosphate and KIM, D. (1995) Molecular mechanisms of de-
aza-analogs of the bisabolyl cation, J. Org. fense by rhizobacteria against root disease, Proc.
Chem. 51, 3454-3462. Natl. Acad. Sci. USA 92, 4197-4201.
CASHEL,M. (1975), Regulation of bacterial ppGpp COQUE,J. J. R., LIRAS,P., MART~N, J. F. (1993),
and pppGpp, Ann. Rev. Microbiol. 29, 301-335. Genes for a p-lactamase, a pencillin binding pro-
CAVALIER-SMITH, T. (1992), Origins of secondary tein and a transmembrane protein are clustered
metabolism, in: Secondary Metabolites: Their with the cephamycin biosynthetic genes in No-
Function and Evolution, Ciba Foundation Sym- cardia lactamdurans, EMBO J. 12, 631-639.
posium 171 (CHADWICK,D. J., WHELAN,J., COQUE,J. J. R., ENGUITA, F. J., MARTIN,J. F., LI-
Eds.), pp. 64-87. Chichester, New York: J. Wiley RAS,P. (1995a), A two protein component 7-ce-
& Sons. phem-methoxylase encoded by two genes of the
CHATER,K. F. (1992). Genetic regulation of sec- cephamycin C cluster converts cephalosporin to
ondary metabolic pathways in streptomyces, in: 7-methoxycephalosporin C, J. Bacteriol. 177,
Secondary Metabolites: Their Function and Evo- 2230-2235.
lution, Ciba Foundation Symposium I71 (CHAD- COQUE,J. J. R., PEREZ-LLARENA, F. J., ENGUITA,
WICK, D. J., WHELAN,J., Eds.), pp. 144-162. F. J., FUENTE,J. L., MARTfN, J. F., LIRAS,P.
Chichester, New York J. Wiley & Sons. (1995b), Characterization of the cmcH genes of
44 1 General Aspects of Secondary Metabolism1
Nocardia lactamdurans encoding a functional 3’- DEVOS,W. M. (1996), Functional analysis of

hydroxymethylcephem 0-carbamoyltransferase promoters in the nisin gene cluster of Lactoba-
for cephamycin biosynthesis, Gene 162,21-27. cillus lactis, J. Bacteriol. 178, 3434-3439.
COSIMA,P., RODRIGUEZ,F., DE FERRA, F., DICKENS, M. L., YE, J. S., STROHL,W. R. (1996),
GRANDI,G., PEREGO,M., VENEMA,G., VAN Cloning, sequencing, and analysis of aklavike-
SINDEREN, D. (1993), Sequence and analysis of tone reductase from Streptomyces sp. strain C5,
the genetic locus responsible for surfactin syn- J. Bacteriol. 178, 3384-3388.
thesis in Bacillus subtilis, Mol. Microbiol. 8, 821- D~Ez,B., GUTIBRREZ,s., BARREDO,J. L., VAN
831. SOLINGEN,P., VAN DER VOORT, L. H. M.,
CUNDLIFFE, E. (1989), How do antibiotic-produc- MARTfN, J. F. (1990), The cluster of penicillin
ing organisms avoid suicide, Annu. Rev. Micro- biosynthetic genes. Identification and characteri-
biol. 43, 207-233. zation of the pcbAB gene encoding a-aminodi-
CUNDLIFFE, E. (1992), Self-protection mechanisms pyl-cysteinyl-valine synthetase and linkage to
in antibiotic producers, in: Secondary Metabol- the pcbC and penDE genes, J. Biol. Chem. 265,
ites: Their Function and Evolution, Ciba Founda- 16358-16365.
tion Symposium 171 (CHADWICK, D. J., WHE- DIMITRIU, S. (Ed.) (1996), Polysaccharides in Med-
LAN,J., Eds.), pp. 199-214. Chichester, New ical Applications. New York, Basel, Hongkong:
York: J. Wiley & Sons. Marcel Dekker.
DAVIES,J. (1994), Inactivation of antibiotics and DISTLER,W., EBERT,A., HEINZEL,P., MANSOU-
the dissemination of resistance genes, Science RI,K., MAYER,G., PISSOWOTZKI, K. (1988), Ex-
264,375-382. pression of genes for streptomycin biosynthesis,
DAVIS,N. K., CHATER,K. F. (1990), Spore colour in: Biology of Actinomycetes ’88 (OKAMI,Y.,
in Streptomyces coelicolor A3(2) involves the de- BEPPU,T., OGAWARA,H., Eds.), pp. 86-91.
velopmentally regulated synthesis of a com- Toyko: Japan Scientific Society Press.
pound biosynthetically related to polyketide an- DISTLER,J., MANSOURI, K., STOCKMANN, M., PIE-
tibiotics, Mol. Microbiol. 4, 1679-1691. PERSBERG, W. (1992), Streptomycin biosynthe-
DAVIS,N. K., CHATER,K. F. (1992), The Strepto- sis and its regulation in Streptomycetes, Gene
myces coelicolor whiB gene encodes a small 115,105-111.
transcription factor-like protein dispensable for DONADIO,S., STARER,M. J., MCALPINE,J. B.,
growth but essential for sporulation, Mol. Gen. SWANSON, S. J., KATZ,L. (1991), Modular or-
Genet. 232,351-358. ganization of genes required for complex poly-
DECKER,H., ROHR,J., MOTAMEDI, H., ZAHNER, ketide biosynthesis, Science 252,675-679.
H., HUTCHINSON, C. R. (1995), Identification of DOULL,J. L., VINING,L. C. (1995), Global physio-
Streptomyces olivaceus Tii2353 genes involved in logical controls, in: Genetics and Biochemistry of
the production of the polyketide elloramycin, Antibiotic Production (VINING,L. C., STUT-
Gene 166,121-126. TARD,C., Eds.), pp. 9-64. Boston, MA: Butter-
DEMAIN,A. L. (1974), Nutrition and the function worth-Heinemann.
of secondary metabolites, Adv. Appl. Microbiol. D’SOUZA,C., NAKANO, M. M., ZUBER,P. (1994),
16, 177-202. Identification of comS, a gene of the srfA ope-
DEMAIN,A. L. (1984), Biology of antibiotic forma- ron that regulates the establishment of genetic
tion, in: Biotechnology of Industrial Antibiotics competence in Bacillus subtilis, Proc. Natl. Acad.
(VANDAMME, E. J., Ed.), pp. 33-39. New York Sci. USA 91, 9397-9401.
Marcel Dekker. DUITON, C. J., GIBSON,S. P., GOODIE,A. C.,
DEMAIN,A. L. (1989), Carbon source regulation of HOLDOM,K. S., PACEY,M. S., RUDDOCK,J.,
idiolite biosynthesis in actinomycetes, in: Regul- BU’LOCK,J. D., RICHARDS, M. K. (1991), Novel
ation of Secondary Metabolism in Actinomycetes avermectins produced by mutational biosynthe-
(SHAPIRO,S., Ed.), pp. 127-134. Boca Raton, sis, J. Antibiot. 44, 357-365.
F L CRC Press. DYSON,P., SCHREMPF, H. (1987), Genetic instabili-
DEMAIN, A. L. (1992), Microbial secondary metab- ty and DNA-amplification in Streptomyces livi-
olism: a new theoretical frontier for academia, a duns 66,J. Bacteriol. 169, 47694803.
new opportunity for industry, in: Secondary Me- ELDER, J. H., SCHNOLZER,M., HASSELKUS-
tabolites: Their Function and Evolution, Ciba LIGHT, C. S., HENSON,M., LERNER,D. A.,
Foundation Symposium 171 (CHADWICK, D. J., PHILLIPS,T. R., WAGAMAN, P. C., KENT,S. B.
WHELAN,J., Eds.). Chichester, New York: J. H. (1993), Identification of proteolytic process-
Wiley & Sons. ing sites within the gag and pol polyproteins of
DE RUYTER,P. G. G. A., KUIPERS,0. P., BEER- feline immunodeficiency virus, J. Virol. 67,1869-
THUYZEN, M. M., VAN ALENBOERRIGTER, I., 1876.
6 References 45
ELLIS, D. D., ZELDIN,E. L., BRODHAGEN,M., L. C., STUTTARD,C., Eds.), pp. 283-306. Bos-
RUSSIN,W. A., M c C o w , B. H. (1996), Taxol ton, MA: Butterworth-Heinemann.
production in nodule cultures of Taxus, J. Nat. GEIGER,0. (1994), N-Acyl-homoserinlacton-Au-
Prod. 59, 246-250. toinduktor-Signalmolekule bei gramnegativen
ENDO,A. (1979), Monacolin K, a new hyperchole- Bakterien, BioEngineering 5, 40-46.
sterolemic agent produced by Monascus species, GOODAY,G . W. (1974), Fungal sex hormones,
J. Antibiot. 32, 852-854. Ann. Rev. Biochem. 1974,3549.
ESPESO,E. A., PENALVA,M. A. (1996), Three GOODRICHTANRIKULU, M., MAHONEY,M. E.,
binding sites for the Aspergillus nidulans PacC RODRIGUEZ,S. B. (1995), The plant growth reg-
zinc-finger transcription factor are necessary and ulator methyl jasmonate inhibits aflatoxin pro-
sufficient for regulation by ambient pH of the duction by Aspergillus flavus, Microbiology 141,
isopenicillin N synthase gene promoter, J. Biol. 2831-2837.
Chem. 271,28825-28830. GRAFE,U. (1989), Autoregulatory secondary me-
ESPESO,E. A., TILBURN,J., ARST, H. N., JR., tabolites from actinomycetes, in: Regulation of
PENALVA,M. A. (1993), pH regulation is a ma- Secondary Metabolism in Actinomycetes (SHAPI-
jor determinant in expression of a fungal biosyn- RO, S., Ed.), pp. 75-126. Boca Raton, F L CRC
thetic gene, EMBO J. 12,3947-3956. Press.
FENG,B., FRIEDLIN, E., MARZLUFF,G. A. (1995), GRAFE,U. (1992), Biochemie der Antibiotika. Hei-
Nuclear DNA-binding proteins which recognize delberg: Spektrum Verlag.
the intergenic control region of penicillin biosyn- GRIMM,A., MADDURI, K., ALI, A., HUTCHINSON,
thetic genes, Curr. Genet. 27, 351-358. C. R. (1994), Characterization of the Strepto-
FERNANDEZ-MORENO, M. A., VALL~N, C., MAL- myces peucetius ATCC 29050 genes encoding
PARTIDA,F. (1996), Isolation and characteriza- doxorubicin polyketide synthase, Gene 151, 1-
tion of streptothricin biosynthetic genes from a 10.
newly Streptomyces isolated strain, Proc. Con$ GROOMBRIDGE, B. (Ed.) (1992), Global Biodiver-
Biol. Streptomycetes, Ohrbeck, Abstract V51. sity. Status of the Earth’s Living Resources. Lon-
FIECHTER,A. (1988), Bioprocess development, in: don: Chapman and Hall.
Overproduction of Macrolide Metabolites (VA- HAAS,H., MARZLUFF,G. A. (1995), NRE, the ma-
NEK, Z., HOSTALEK, Z., Eds.), pp. 231-259. Bos- jor nitrogen regulatory protein of Penicillium
ton, London: Buttenvorth. chrysogenum, binds specifically to elements in
FIERRO,F., BARREDO,J. L., DIEZ, B., GUTIER- the intergenic promoter regions of nitrate as-
REZ, S., FERNANDEZ, F. J., MARTIN, J. F. similation and penicillin biosynthetic gene clus-
(1995), The penicillin gene cluster is amplified in ters, Curr. Genet. 28, 177-183.
tandem repeats linked by conserved hexanucleo- HAMOEN,L. W., ESHUIS,H., JONGBLOED, J., VEN-
tide sequences, Proc. Natl. Acad. Sci. USA 92, EMA, G., VAN SINDEREN,D. (1995), A small
62004204. gene, designated comS, located within the cod-
FIERRO,F., MONTENEGRO, E., GUTIERREZ,S., ing region of the fourth amino acid-activation
MART~N, J. F. (1996), Mutants blocked in peni- domain of srfA, is required for competence de-
cillin biosynthesis show a deletion of the entire velopment in Bacillus subtilis, Mol. Microbiol.
penicillin gene cluster at a specific site within a 15, 55-63.
conserved hexanucleotide sequence, Appl. Mi- HANLON, G. W., HODGES,N. A. (1981), Bacitracin
crobiol. Biotechnol. 44, 597-604. and protease production in relation to sporula-
FILIPPINI,S., SOLINAS, M. M., BREME, U., tion during exponential growth of Bacillus liche-
SCHLUTER,M. B., GABELLINI, D., BIAMONTI, nijormis on poorly utilized carbon and nitrogen
G., COLOMBO,A. L., GAROFANO,L. (1995), source, J. Bacteriol. 147, 427-431.
Streptomyces peucetius daunorubicin biosynthe- HASUMI,K. (1993), Competitive inhibition of
sis gene, dnrR sequence and heterologous ex- squalene synthetase by squalestatin-I, J. Anti-
pression, Microbiology 141, 1007-1016. biot. 46,689-691.
Fox, B. W. (1991), Natural products in cancer HAYAKAWA, Y., MATSUOKA,M., SHIN-YA,K.,
treatment from bench to the clinic, Trans. R. SETO,H. (1993), Quinolidomycins, A,, AZ, and
SOC. Trop. Med. Hyg. 85,22-85. B1, novel 60-membered macrolide antibiotics. I.
FUQUA,W. C., WINANS,S. C., GREENBERG, E. P. Taxonomy, fermentation, isolation, physico-
(1994), Quorum sensing in bacteria: the LuxR- chemical properties and biological activity, J.
Lux I family of cell-density-responsive transcrip- Antibiot. 46, 1557-1562.
tional regulators, J. Bacteriol. 176, 269-275. HAYASHI,H., TARUI,N., MURAO,S. (1985), Isola-
GASSON,M. J. (1995), Lantibiotics, in: Genetics and tion and identification of cyclooctasulfur, a fruit-
Biochemistry of Antibiotic Production (VINING, ing-body-inducing substance, produced by Strep-
tomyces albulus T0447, Agric. Biol. Chem. 49, HORINOUCHI, S., BEPPU,T. (1990), Autoregulato-
101. ry factors of secondary metabolism and morpho-
HAYDOCK,S. F., APARICIO,J. F., MOLNAR,I., genesis in actinomycetes, CRC Crit. Rev. Bio-
SCHWECKE,T., KHAW, L. E., K ~ N I GA., , technol. 10,191-204.
MARSDEN, A. F. A., GALLOWAY, I. S., STAUN- HORINOUCHI, S., BEPPU,T. (1992a), Regulation of
TON,J., LEADLAY, P. F. (1995), Divergent se- secondary metabolism and cell differentiation in
quence motifs correlated with the substrate spe- streptomycetes:A-factor as a microbial hormone
cificity of (methy1)malonyl-CoA-acyl carrier and the AfsR protein as a component of a two-
protein transacylase domains in modular polyke- component regulatory system, Gene 115, 167-
tide syntheses, FEBS Lett. 374, 246-248. 172.
HEINSTEIN, P. F., CHANG, C. J. (1994), Taxol, Ann. HORINOUCHI, S., BEPPU,T. (1992b), Autoregula-
Rev. Plant. Physiol. Plant Mol. Biol. 45, 663- tory factors and communication in actinomy-
674. cetes, Annu. Rev. Microbiol. 46, 377-393.
HERZOG-VELIKONJA, B., PODLESEK,Z., GRAB- HORINOUCHI, S., BEPPU,T. (1995), Autoregula-
NAR, M. (1994), Isolation and characterization tors, in: Genetics and Biochemistry of Antibiotic
of Tn917-generated bacitracin deficient mutants Production (VINING,L. C., STUTTARD,C.,
of Bacillus licheniformis, FEMS Microbiol. Lett. Eds.), pp. 103-120. Boston, MA: Butterworth-
121,147-152. Heinemann.
HIRAMATSU, K. (1995), Molecular evolution of HORINOUCHI, S., KITO, M., NISHIYAMA, M., Fu-
MRSA, Microbiol. Immunol. 39, 531-543. RUYA, K., HONG,S. K., MIYAKE,K., BEPPU,T.
(1990), Primary structure of AfsR, a global regu-
HODGSON,J. E., FOSBERRY, A. P., RAWLINSON,
latory protein for secondary metabolite forma-
N. S., Ross, H. N. M., NEAL,R. J., ARNELL, J.
tion in Streptomyces coelicolor A3(2), Gene 95,
C., EARL,A. J., LAWLOR,E. J. (1995), Clavu-
49-56.
lanic acid biosynthesis in Streptomyces clavulige-
rus: gene cloning and characterization, Gene 166, HORNEMANN, U., ZHANG,X. Y., OTTO, J. C.
49-55. (1993), Transferable Streptomyces DNA amplifi-
cation and coamplification of foreign DNA se-
HOPWOOD, D. A. (1988), Towards an understand- quences, J. Bacteriol. 175, 1126-1133.
ing of gene switching in Streptomyces, the basis HOTTA, K., DAVIES,J., YAGISAWA,M. (1995),
of sporulation and antibiotic production, Proc. Aminoglycosides and aminocyclitols, in: Ge-
R. SOC. London B: Biological Sciences 235, 121- netics and Biochemistry of Antibiotic Production
138. (VINING,L. C., STU-ITARD,C., Eds.), pp. 571-
HOPWOOD,D. A. (1989), Antibiotics: opportuni- 596. Boston, MA: Butterworth-Heinemann.
ties for genetic manipulation, Philos. Trans. R. HUNDLE,B., ALBERTI,M., NIEVELSTEIN,V.,
SOC. (London) 324,549-562. BEYER,P., KLEINIG,H., ARMSTRONG, G. A.,
HOPWOOD,D. A., KHOSLA,C. (1992), Genes for BURKE,D. H., HEARST,J. E. (1994), Functional
polyketide secondary metabolic pathways in mi- assignment of Erwinia herbicola EholO caroti-
croorganisms and plants, in: Secondary Metabol- noid genes expressed in Escherichia coli, Mol.
ites: Their Function and Evolution, Ciba Founda- Gen. Genet. 245, 406-416.
tion Symposium 171 (CHADWICK, D. J., WHE- HUTTER,R. A. (1986), Overproduction of micro-
LAN, J., Eds.), pp. 88-112. Chichester, New bial metabolites, in: Biotechnology, 1st Edn.,
York J. Wiley & Sons. Vol. 4 (REHM,H. J., REED,G., Eds.), pp. 3-14.
HOPWOOD, D. A., SHERMAN, D. A. (1990), Molec- Weinheim: VCH.
ular genetics of polyketides and its comparison HUTCHINSON, C. R. (1994), Drug synthesis by ge-
to fatty acid biosynthesis, Annu. Rev. Genet. 24, netically engineered microorganisms, BiolTech-
37-66. nology 12,375-380.
HOPWOOD,D. A., KIESER,T., WREIGHT,E. M., HUTCHINSON, C. R. (1995), Anthracyclines, in: Ge-
BIBB,M. J. (1983), Plasmids, recombination and netics and Biochemistry of Antibiotic Production
chromosome mapping in Streptomyces lividans (VINING,L. C., S T U ~ A R D C.,, Eds.), pp. 331-
66, J. Gen. Microb. 129, 2257-2269. 358. Boston, MA: Butterworth-Heinemann.
HOPWOOD,D. A., CHATER,K. F., BIBB, M. J. HUTCHINSON, C. R., FUJII,I. (1995), Polyketide
(1995), Genetics and antibiotic production in synthase gene manipulation: a structure-func-
Streptomyces coelicolor A3(2), a model Strepto- tion approach in engineering novel antibiotics,
mycete, in: Genetics and Biochemistry of Anti- Annu. Rev. Microbiol. 49,201-238.
biotic Production (VINING, L. c., STUTTARD, c., ISHIZUKA,H., HORINOUCHI, S., KIESER,H. M.,
Eds.), pp. 65-102. Boston, MA: Butterworth- HOPWOOD,D., BEPPU,T. (1992), A putative
Heinemann. two component regulatory system involved in
6 References 47
secondary metabolism in Streptomyces species, J. KELLER,N. P., ADAMS,T. H. (1995), Analysis of a

Bacteriol. 174, 7585-7594. mycotoxin gene cluster in Aspergillus nidulans,
ISONO,K. (199O), Nucleoside antibiotics: Structure, SAAS Bull. Biochem. Biotechnol. 8, 14-21.
biological activity, and biosynthesis, J. Antibiot. KHOKHLOV, A. S. (1982), Low molecular weight
41, 1711-1752. microbial bioregulators of secondary metabol-
ISONO,K. (1995), Polyoxins and related nucleo- ism, in: Overproduction of Microbial Products
tides, in: Genetics and Biochemistry of Antibiotic (KRUMPHANZ, V., SIKYTA,B., VANEK, Z.,
Production (VINING, L. C., STUITARD, C., Eds.), pp. 97-109. London: Academic Press.
Eds.), pp. 597418. Boston, M A Butterworth- KIM, H. S., TADA,H., NIHIRA,T., YAMADA,Y.
Heinemann. (1990), Purification and characterization of vir-
JACOB,L., ZASLOFF,M. (1994), Potential thera- giniae butanolide C-binding protein, a possible
peutic applications of magainins and other an- pleotropic signal-transducer in Streptomyces vir-
timicrobial agents of animal origin, in: Antimi- giniae, J. Antibiot. 43, 692-706.
crobial Peptides, Ciba Foundation Symposium KIM,E. S., BIBB,M. J., BUTLER,M. J., HOPWOOD,
186 (BOMAN,H. G., MARSH,J., GOODE,J. A., D. A., SHERMAN, D. H. (1994), Sequences of
Eds.), Chichester: J. Wiley & Sons. the oxytetracycline polyketide synthase-encod-
JENSEN,S. E., DEMAIN, A. L. (1995), Beta-lactams, ing otc genes from Streptomyces rimosus, Gene
in: Genetics and Biochemistry of Antibiotic Pro- 141, 141-142.
duction (VINING,L. C., STUTTARD,C., Eds.), KIM,E.-S., CRAMER, K. D., SHREVE,A. L., SHER-
pp. 239-268. Boston, MA: Butterworth-Heine- MAN, D. H. (1995), Heterologous expression of
mann. an engineered biosynthetic pathway: functional
JENSEN,P. R., FENICAL,W. (1994), Strategies for dissection of type I1 polyketide synthase compo-
the discovery of secondary metabolites from nents in Streptomyces species, J. Bacteriol. 177,
bacteria: ecological perspectives, Annu. Rev. Mi- 1202-1207.
crobiol. #, 559-584. KLEINKAUF,H., VON DOHREN,H. (1986), Peptide
JOHNSON, R., ADAMS,J. (1992), The ecology and Antibiotics, in: Regulation of Secondary Metab-
evolution of tetracycline resistance, TREE 7 , olite Formation (KLEINKAUF, H., VON DOHREN,
295-299. H., DORNAUER, H., NESEMANN, G., Eds.), pp.
JOHNSON, P. F., MCKNIGHT,S. L. (1989), Euka- 173-207. Weinheim: VCH.
ryotic transcriptional regulatory proteins, Annu. KLEINKAUF, H., VON DOHREN,H. (1987), Biosyn-
Rev. Biochem. 58,799-839. thesis of peptide antibiotics, Annu Rev. Micro-
KAHAN,B. D. (Ed.) (1987), Cyclosporins. London: biol. 41, 259-289.
Grune & Stratton. KLEINKAUF, H., VON DOHREN,H. (Eds.) (199O),
KAJIMURA, Y., SUGUYAMA,M., KANDEA,M. Biochemistry of Peptide Antibiotics. Recent Ad-
(1999, Bacillopeptins, new cyclic lipopeptide vances in the Biochemistry of o-Lactams and Mi-
antibiotics from Bacillus subtilis FR-2, J. Anti- crobial Bioactive Peptides. Berlin: DeGruyter.
biot. #, 1095-1103. KLEINKAUF, H., VON DOHREN,H. (1996), A non-
KAO, C. M., Luo, G., KATZ, L., CANE,D. E., ribosomal system of peptide biosynthesis, Eur. J.
KHOSLA,C. (1999, Manipulation of macrolide Biochem. 236, 335-51.
ring size by directed mutagenesis of a modular KLEINKAUF, H., VON DOHREN,H. (in press), Bio-
polyketide synthase, J. Am. Chem. SOC. 117, synthesis of cyclosporins and related peptides,
9105-91 06. in: Fungal Biotechnology (ANKE,T., Ed.). Lon-
KATZ,L., DONADIO, S. (1995), Macrolides, in: Ge- don: Chapman and Hall.
netics and Biochemistry of Antibiotic Production KOHMOTO, K., YODER,0. C. (Eds.) (1994), Host-
(VINING,L. C., STUITARD,C., Eds.), pp. 385- Specific Toxin. Biosynthesis, Receptor and Mo-
420. Boston, M A Butterworth-Heinemann. lecular Biology. Tottori University Press.
KELEMEN,G. H., BROWN,G. L., KORMANEC, J., KONDO,S., YASUO,K., MATSUMA, M., KATAYA-
POTUCKOVA, L., CHATER,K. F., BUTTNER, M. MA,M., MARUMO,S. (1986), Pamamycin-6@7, a
J. (19%), The positions of the sigma factor new macrodiolide antibiotic from Streptomyces
genes, whiG and sigF, in the hierarchy controll- alboniger, J. Antibiot. 41, 1196-1204.
ing the development of spore chains in the aerial KROHN,K., KIRST,H. A., MAAG,H. (Eds.) (1993),
hyphae of Streptomyces coelicolor A3(2), Mol. Antibiotics and Antiviral Compounds. Wein-
Microbiol. 21, 593-603. heim: VCH.
KELL,D. B., KAPRELYANTS, A. S., GRAFEN,A. KUBOHARA, Y., OKAMOTO,K., TANAKA, Y., As-
(1995). Pheromones, social behaviour and the ALIN,K., SAKURAI, A., TAKAHASHI, N. (1993),
functions of secondarv metabolism in bacteria. Differanisol A. an inducer of the differentiation
Trends Ecol. Evol. lO.’126-129. of Friend leukemic cells induces stalk cell differ-
48 1 General Aspects of Secondary Metabolisnrl
entiation in Dictyostelium discoideum, FEBS New York, Heidelberg, Berlin, Tokyo: Springer-
Lett. 322, 73-75. Verlag.
KUBOTA,T., TOKOROYAMA, KAMIKAWA, T., SA- LUENGO,J. M. (1995), Enzymatic synthesis of hy-
TAMURA,Y. (1966), The structures of sclerin drophobic penicillins, J. Antibiot. 48, 1195-
and scleroide, metabolites of Sclerotinia libertia- 1212.
nu, Tetrahedron Lett. 1966, 5205. MACCABE,A. P., RIACH,M. B. R., UNKLES,S. E.,
LAATSCH,H. (1994), Antibase, Database. Heidel- KINGHORN, J. R. (1990), The Aspergillus nidu-
berg: Chemical Concepts. lam npeA locus consists of three contiguous
LANCINI, G., LORENZETTI, R. (1993), Biotechnolo- genes required for penicillin biosynthesis,
gy of Antibiotics and Other Bioactive Microbial EMBO J. 9,279-287.
Metabolites. New York, London: Plenum Press. MADDURI, K., HUTCHINSON, C. R. (1995a), Func-
LANDAN,G., COHEN,G., AHARONOWITZ, Y., tional characterization and transcriptional analy-
SHUALI, Y., GRAUR,D., SHIFFMAN,D. (1990), sis of the dnrR, locus, which controls daunorubi-
Evolution of isopenicillin N synthase may have cin biosynthesis in Streptomyces peucetius, J.
involved horizontal gene transfer, Mol. Biol. Bacteriol. 177, 1208-1215.
Evol. 7,399406. MADDURI,K., HUTCHINSON, C. R. (1995b), Func-
tional characterization and transcriptional analy-
LAWEN,A., TRABER,R. (1993), Substrate specifi-
cities of cyclosporin synthetase and peptolide sis of a gene cluster governing early and late
steps in daunorubicin biosynthesis in Strepto-
SDZ 214-103 synthetase, J. Biol. Chem. 268,
20452-20465. myces peucetius, J. Bacteriol. 177, 3879-3884.
MAHANTI,N., BHATNAGAR,D., CARY,J. W.,
LE GOUILL,C., DESMARAIS,D., DERY, C. V.
JOUBRAN,J., LINZ,J. E. (19%), Structure and
(1993), Saccharopolyspora hirsuta 367 encodes function of FAS-lA, a gene encoding a putative
clustered genes similar to ketoacyl synthase, ke- fatty acid synthetase directly involved in aflatox-
toacyl reductase, acyl carrier protein, and biotin in biosynthesis in Aspergillus parasiticus, Appl.
carboxyl carrier protein, Mol. Gen. Genet. 240, Environ. Microbiol. 62, 191-195.
146-150.
MARAHIEL, M. A., DANDERS,W., KRAUSE,M.,
LI, W., NIKIRA,T., SAKUDA, S., HISHIDA, T. YA- KLEINKAUF, H. (1979), Biological role of grami-
MADA,Y. (1992), New inducing factors for vir- cidin S in spore function, Eur. J. Biochem. 99,
giniamycin production from Streptomyces anti- 49-58,
bioticus, J. Ferment. Biotechnol. 14, 214-217. MARAHIEL,M. A,, NAKANO,M. M., ZUBER,P.
LIRAS,P., ASTURIAS, J. A., MARTIN, J. F. (1990), (1993), Regulation of peptide antibiotic produc-
Phosphate sequences involved in transcriptional tion in Bacillus, Mol. Microbiol. I , 631-636.
regulation of antibiotic biosynthesis, Trends Bio- MARCOS,A. T., GUTIERREZ,S., DIEZ, B., FER-
technol. 8, 184-189. NANDEZ,F. J., OGUIZA,J. A., MARTIN,J. F.
LITZKA,O., THENBERG, K., BRAKHAGE, A. A. (1995), Three genes hrdB, hrdD and hrdT of
(1996), The Aspergillus nidulans penicillin bio- Streptomyces griseus IMRU 3570, encoding sig-
synthesis gene aat (penDE) is controlled by a ma-factor-like proteins, are differentially ex-
CCAAT containing DNA element, Eur. J. Bio- pressed under specific nutritional conditions,
chem. 238,675-682. Gene 153,41-48.
LIU, Y. S., KIESER,H. M., HOPWOOD,D. A., MART~N, J. F. (1978), Manipulation of gene expres-
CHEN,C. W. (1992), The chromosomal DNA of sion in the development of antibiotic produc-
Streptomyces lividans 66 is linear, Mol. Micro- tion, in: Antibiotics and Other Secondary Metab-
biol. 10, 923-933. olites (HOTTER,R., LEISINGER, J., NOESCH, J.,
LOMBb, F., BLANCO,G., FERNANDEZ, E., MEN- WEHRLI,W., Eds.), pp. 19-38. London, New
DEZ,C., SALAS,J. A. (1996), Characterization of York: Academic Press.
Streptomyces argillaceus genes encoding a poly- MARTIN,J. F. (1989), Molecular mechanisms for
ketide synthase involved in the biosynthesis of the control by phosphate of the biosynthesis of
the antitumor mithramycin, Gene 172, 87-91. antibiotics and other secondary metabolites, in:
LOUMAYE, E. THORNER, J., GAIT, K. J. (1982), Regulation of Secondary Metabolism in Actino-
Yeast mating pheromone activates mammalian mycetes (SHAPIRO,S., Ed.), pp. 213-233. Boca-
gonadotrophs: evolutionary conservation of a re- Raton, FL: CRC Press.
productive hormone, Science 218,1323-1325. MART~N, J. F., DEMAIN,A. L. (1980), Control of
LUCKNER, M. (1989), Secondary Metabolism in Mi- antibiotic biosynthesis, Microbiol. Rev. 44, 230-
crobes, Plants and Animals. Jena: G. Fischer. 251.
LUCKNER, M., NOVER,L., BOHM,H. (1977), Sec- MART~N, J. F., GUTIERREZ, S. (1995), Genes for
ondary Metabolism and Cell Differentiation. beta-lactam antibiotic biosynthesis, Antonie van
6 References 49
Leeuwenhoek Int. J. Gen. Mol. Microbiol. 67, MISAWA,N., SATOMI,Y., KONDO,K., YOKOYA-
181-200. MA, A., KAJIWARA, s., SAITO,T., OHTANI,T.,
MARTIN, J. F., LOPEZ-NIETO,M. J., CASTRO,J. MIKI,W. (1995), Structure and functional analy-
M., CORTES,J., RAMOS,F. R., CANTORAL,J. sis of a amrine bacterial carotinoid biosynthesis
N., ALVAREZ,E., RAMIREZ,M. G., BARREDO, gene cluster and astaxanthin biosynthetic path-
J. C., LIRAS,P. (1986), Enzymes involved in p- way proposed at the gene level, J. Bacteriol. 177,
lactam biosynthesis controlled by carbon and ni- 65754584.
trogen regulation, in: Regulation of Secondary Mo, Y. Y., GEIBEL,M., BONSALL,R. F., GROSS,
Metabolite Formation (KLEINKAUF,H., VON D. C. (1995), Analysis of sweet cherry (Prunus
DOHREN,H., DORNAUER H., NESEMANN, G., avium L.) leaves for plant signal molecules that
Eds.), pp. 41-47. Weinheim: VCH. activate the syrB gene required for synthesis of
MARTINEZ-BLANCO, H., REGLERS,A., LUMENGO, the phytotoxin, syringomycin, by Pseudomonas
J. M. (1991), “ I n vifro”-syntheses of different syringae pv. syringae, Plant Physiol. 107, 603-
naturally-occurring semisynthetic and synthetic 612.
penicillins using a new effective enzymatic cou- MONAGHAN, R. L., TKACZ,J. S. (1990), Bioactive
pled system, J. Antibiot. 44, 1252-1254. microbial products: focus upon mechanism of ac-
MAZZOLA,M., WHITE,F. F. (1994), A mutation in tion, Annu. Rev. Microbiol. 44,271-301.
the indole-3-acetic acid biosynthesis pathway of MORENO,F., MILLAN,J. L. S., HERNANDEZ-CHI-
Pseudomonas syringae PV syringae affects c o , C., KOLTER,R. (1995) Microcins, in: Ge-
growth in Phaseolus vulgaris and syringomycin netics and Biochemistry of Antibiotic Production
production, J. Bacteriol. 176, 1374-1382. (VINING,L. C., STUITARD,C., Eds.), pp. 307-
MCCUE,L. A., KWAK,J., BABCOCK,M. J, KEN- 322. Boston, MA: Butterworth-Heinemann.
DRICK,K. E. (1992), Molecular analysis of spo- MOLLER,L. (1989), Chemistry, biochemistry and
rulation in Streptomyces griseus, Gene 115, 173- therapeutic potential of microbial a-glucosidase
179. inhibitors, in: Novel Microbial Products for Med-
MCDANIEL,R., EBERT-KOHOSLA, S., Fu, H., icine and Agriculture (DEMAIN,A. L., SOMKUTI,
HOPWOOD,D. A., KHOSLA,C. (1994), Engi- G . A., HUNTER-CEVERA, J. C., ROSSMORE,H.
neered biosynthesis of novel polyketides: in- W., Eds.), pp. 109-116. Amsterdam: Elsevier.
fluence of a downstream enzyme on the catalytic MURAO,S., HAYASHI, H., TARUI,H. (1984), An-
specificity of a minimal aromatic polyketide thranilic acid, as a fruiting body inducing sub-
synthase, Proc. Natl. Acad. Sci. USA 91, 11542- stance in Favolus arcularius, from a strain TA7
11346. of actinomycetes, Agric. Biol. Chem. 48, 1669-
MCNEIL,D. J. (1995), Avermectins, in: Genetics 1677.
and Biochemistry of Antibiotic Production (VIN- NIKAIDO,H. (1994), Prevention of drug access to
ING, L. c., STUTTARD,C., Eds.), pp. 421442. bacterial targets: permeability barriers and ac-
Boston, MA: Butterworth-Heinemann. tive efflux, Science 264, 382-388.
MEIGHEN, E. A. (1991), Molecular biology of bac- NOBLE,R. L. (1990), The discovery of the vinca al-
terial luminescence, Microbiol. Rev. 55, 123- kaloids - chemotherapeutic agents against can-
142. cer, Biochem. Cell. Biol. 68, 1344-1351.
MEISSNER,K., DITTMANN, E., BORNER,T. (1996), OCHI,K. (1990), Streptomyces relC mutants with an
Toxic and nontoxic strains of the cyanobacter- altered ribosomal protein ST-Ul and genetic
ium Microcystis aeruginosa contain sequences analysis of a Streptomyces griseus relC mutant, J.
homologous to peptide synthetase genes, FEMS Bacteriol. 172, 4008-4016.
Microbiol. Lett. 135, 295-303. OCHI,K., TSURUMI, Y., SHIGEMATSU, N., IWANI,
MERSON-DAVIES, L. A., CUNDLIFE,E. (1994), M., UMEHARA, K., OKUHARA, M. (1988), Phys-
Analysis of five tylosin biosynthetic genes from iological analysis of bicozamycin high-producing
the tyIIBA region of the Streptomyces fradiae Streptomyces griseoflavus used at industrial lev-
genome, Mol. Microbiol. 13, 349-355. el, J. Antibiot. 41, 1106-1115.
MEURER,G., HUTCHINSON, C. R. (1995), Dauno- OCHI,K., PENYIGE,A., BARABAS, G. (1992), The
rubicin type I1 polyketide synthase enzymes possible role of ADP-ribosylation in sporulation
DpsA and DpsB determine neither the choice of and streptomycin production by Streptomyces, J.
starter unit nor the cyclization pattern of aro- Gen. Microbiol. 138, 1745-1750.
matic polyketides, J. Am. Chem. SOC.117, 5899- OCHIAI,T. (1987), Critical view on the use of im-
5900. munostimulants in cancer treatment, in: Immu-
MILLER,J. R., INGOLIA, T. D. (1993), Cloning and nostimulants: Now and Tomorrow (AZUMA,I.,
characterization of &lactam biosynthetic genes, JOLLES,G., Eds.), pp. 167-172. TokyolBerlin:
Mol. Gen. Genet. 3,689495. Japan Scientific Societies PresslSpringer.
OGARA,M., NAKAYAMA,H., FURIHATA, K., SHI- suppression, FEMS Microbiol. Lett. 136, 101-
MADZU,A., SETO,H., OTAKE,N. (1985), Struc- 108.
ture of a new antibiotic curromycin A, produced PIERSON, L. S., GAFFNEY, T., LAM,S., GONG,F. C.
by a genetically modified strain of Streptomyces (1995), Molecular analysis of genes encoding
hygroscopicus, a polyether antibiotic producing phenazine biosynthesis in the biological control
organism, J. Antibiot. 38, 669-673. bacterium Pseudomonas aureofaciens 30-84,
O’HAGAN,D. (Ed.) (1991), The Polyketide Metab- FEMS Microbiol. Lett. 134, 299-307.
olites. Chichester: Ellis Horwood. PITKIN, J. W., PANACCIONE, D. G., WALTON,J. D.
OKAMOTO, S., NIHIRA,T., KATAOKA, H., SUZUKI, (1996), A putative cyclic peptide efflux pump en-
A., YAMADA,Y. (1992), Purification and molec- coded by the toxA gene of the plant pathogenic
ular cloning of a butyrolactone autoregulator re- fungus Cochliobolus carbonum, Microbiology
ceptor from Streptomyces virginiae, J. Biol. 142, 1557-1565.
Chem. 267, 1093-1098. PLINER,S. A., KLEINER,E. M., KORNITSKAYA, E.
OLESKIN,A. V. (1994), Social behaviour of micro- Y., TOVAROVA, I. I., ROZYNOV,B. V., SMIR-
bial populations, J. Basic Microbiol. 34, 425- NOVA,G. M., KHOKKLOV, A. S . (1976), Isola-
439. tion and primary characteristics of the A-factor,
OLIYNYK, M., BROWN,M. J. B., CORTES, J., Sov. J. Bioorg. Chem. (English translation) 2,
STAUNTON, J., LEADLAY, P. F. (1996), A hybrid 825-832.
modular polyketide synthase obtained by do- POSPIECH,A., BIETENHADER, J., SCHUPP, T.
main swapping, Chemistry & Biology 3, 833- (1996), Two multifunctional peptide synthetases
839. and an 0-methyltransferase are involved in the
OMURA,S. (Ed.) (1992), The Search of Bioactive biosynthesis of the DNA-binding antibiotic and
Metabolites. New York: Springer-Verlag. antitumor agent saframycin Mxl from Myxococ-
OXFORD, J. S., FIELDS, H. J., REEVES,D. S. (Eds.) cus xanthus, Microbiology 142, 741-746.
(1986), Drug Resistance in Viruses, Other Mi- POSTMA,P. W., LENGELER, J. W., JACOBSON, G.
crobes and Eukaryotes. London, New York: Ac- R. (1993), Phosphoeno1pyruvate:carbohydrate
ademic Press. phosphotransferase systems of bacteria, Micro-
PJ~REZ-ESTEBAN, B., G6MEZ-PARD0, E., PENAL- biol. Rev. 57, 543-594.
VA, M. A. (1995), A lacZ reporter fusion meth-
od for the genetic analysis of regulatory muta- POTIER,P., GUERI~TE-VOEGELEIN, F., GUENARD,
tions in pathways of fungal secondary metabol- D. (1994), Taxoids, a new class of antitumor
ism and its application to the Aspergillus nidu- agents of plant origin: recent results, Nouv. Rev.
lans penicillin pathway, J. Bacteriol. 177, 6069- Fr. Hematol. 36,521-523.
6076. QUEENER, S . W., WILKERSON, S., TUNIN,D. R.,
PETERSEN,F., ZAHNER, H., METZGER,J. W., MCDERMO~T, J. P., CHAPMAN, J. K., NASH,C.,
FREUND,S., HUMMEL, R. P. (1993). Germicidin, WESTPHELING, J. (1986), Cephalosporin C fer-
an autoregulative germination inhibitor of Strep- mentation biochemical and regulatory aspects of
tomyces viridochromogenes NRRL B-1551, J. sulfur metabolism, in: Biotechnology of Indus-
Antibiot. 46, 1126-1138. trial Antibiotics (VANDAMME, E. J., Ed.), pp.
PETRICH,A. K., LESKIW,B. K., PARADKAR, A. S., 141-170. New YorkBasel: Marcel Dekker.
JENSEN,S. E. (1994), Transcriptional mapping QUIROS, L. M., SALAS,J. A. (1995), Biosynthesis
of the genes encoding the early enzymes of the of the macrolide oleandomycin by Streptomyces
cephamycin biosynthestic pathway of Strepto- antibioticus, J. Biol. Chem. 270, 18234-18239.
myces clavuligerus, Gene 142,4143, RA, S. R., QIAO, M. Q., IMMONEN, T. I., SARIS,P.
PIEPERSBERG, W. (1994), Biosynthesis of second- E. J. (1996), Genes responsible for nisin synthe-
ary carbohydrate components in actinomycetes sis, regulation and immunity from a regulon of
and other bacteria. Perspectives of a pathway two operons and are induced by nisin in Lacto-
engineering (in German), BioEngineering 6, 27- coccus lactis N8,Microbiology 142, 1281-1288.
34. REDENBACH, M., KIESER,H. M., DENAPAITE, D.,
PIEPERSBERG, W. (1995), Streptomycin and related EICHNER,A., CULLUM,J., KINASHI, H., HOP-
aminoglycosides, in: Genetics and Biochemistry WOOD,D. A. (1996), A set of ordered cosmids
of Antibiotic Production (VINING,L. C., STUT- and a detailed genetic and physical map for the
TARD,c., Eds.), pp. 531-570. Boston, MA: But- 8 Mb Streptomyces coelicolor A3(2) chromo-
terworth-Heinemann. some, Mol. Microbiol. 21, 77-96.
PIERSON,L. S., PIERSON,E. A. (1996), Phenazine REICHENBACH, H., HOFLE,G. (1993), Biologically
antibiotic production in Pseudomonas aureofa- active secondary metabolites from Myxobacter-
ciens - role in rhizosphere ecology and pathogen ia, Biotechnol. Adv. 11, 219-277.
6 References 51
REUTER,G. (1989), Enzymatic regulation of micro- SANGLIER, J. J., LARPENT, J. P. (1989), Biotechno-
bial phytoeffector biosynthesis, Progr. Ind. Mi- logie des Antibiotiques. Paris, Milano, Barcelona,
crobiol. 27, 271-282. Mexico: Masson.
RINEHART, K. L., SHIELD,L. S. (1988), Biological SCHIRMBOCK, M., LORITO,M., WANG, Y. L.,
active marine natural products, in: Horizons on HAYES,C. K., ARISAN-ATAC, I., SCALA,F.,
Antibiotic Research (DAVIES,B. D., ICHIKAWA, HARMAN, G. E., KUBICEK, C. P. (1994), Parallel
T., MAEDA,K., MITCHER,L. A., Eds.), pp. 194- formation and synergism of hydrolytic enzymes
227. Tokyo: Japan Antibiotics Association. and peptaibol antibiotics, molecular mechanisms
ROCK,C. 0.. CRONAN,J. E. (1996), Escherichia involved in the antagonistic action of Trichoder-
coli as a model for the regulation of dissociable ma harzianum against phytopathogenic fungi,
(type 11) fatty acid biosynthesis, Biochim. Bio- Appl. Environ. Microbiol. 60, 43644370.
phys. Acta 1302, 1-16. SCHUPP,T., TUOPET,C., CLUZEL,B., NEFF, S.,
ROHR, J., SCHONEWOLF, M., UDVARNOKI, G., HILL,S., BECK,J. J., LIGON,J. M. (1995), A So-
ECKARDT,K. SCHUMANN, G., WAGNER,C., rangium cellulosum (Myxobacterium) gene clus-
BEALE,J. M., SOREY,S. D. (1993), Investiga- ter for the biosynthesis of the macrolie antibiotic
tions on the biosynthesis of the angucycline sorphen A-cloning, characterization, and homo-
group antibiotics, J. Org. Chem. 58, 2547-2551. logy to polyketide synthase genes from Actino-
Roos, W., LUCKNER, M. (1986). The spatial organ- mycetes, J. Bacteriol. 177, 3637-3679.
ization of secondary metabolism in microbial SCHWARTZ,D., ALIJAH,R., NUSSBAUMER, B.,
and plant cells, in: Cell Growth and Management PELZER, S., WOHLLEBEN, W. (1996), The pep-
(SUBRAMANIAN, A. V. (Ed.), pp. 45-73. Boca tide synthetase gene phsA from Streptomyces
Raton, F L CRC Press. viridochromogenes is not juxtaposed with other
genes involved in nonribosomal biosynthesis of
ROSSNER,E., ZEECK,A., KONIG,W. A. (1990),
peptides, Appl. Environ. Microbiol. 62, 570-
Structure elucidation of hormaomycin (in Ger-
577.
man), Angew. Chem. 102, 84-85.
SCHWECKE, T., APARICIO, J. F., MOLNAR,I., KO-
SADAKANE, N., TANAKA, Y., OMURA, S. (1983), NIG, A., KHAW,L. E., HAYDOCK, S. F., OLIY-
Hybrid biosynthesis of a new macrolide antibiot- NYK, M., CAFFREY, P., CORTES,J., LESTER,J.
ic by a daunomycin-producing microorganism, J. B., BOHM,G. A., STAUNTON, J., LEADLAY, P.
Antibiot. 36, 921-922. F. (1995), The biosynthetic gene cluster for the
SAHL,H.-G., JACK,R. W., BIERBAUM, G. (1995), polyketide immunosuppressant rapamycin, Proc.
Biosynthesis and biological activities of Natl. Acad. Sci. USA 91, 7839-7843.
lantibiotics with unique post-translational modif- Scrip, World Pharmacentical News (London)
ications, Eur. J. Biochem. WO,827-853. (1993), 1825,23.
SAKUDA, S., YAMADA, Y. (1991), Stereochemistry SENO,E. T., BALTZ,R. H. (1989), Structural organ-
of butyrolactone autoregulators from streptomy- ization and regulation of antibiotic biosynthesis
cetes, Tetrahedron Lett. 32, 1817-1820. and resistance genes in actinomycetes, in: Regul-
SAKUDA,S., HIGASHI,A., TANAKA,S., NIHIRA, ation of Secondary Metabolism in Actinomycetes
T., YAMADA,Y. (1992), Biosynthesis of virgin- (SHAPIRO,S., Ed.), pp. 243. Boca Raton, FL:
iae butanolide A, a butyrolactone autoregulator CRC Press.
from Streptornyces, J. Am. Chem. SOC.114,663- SHAH,A. J., TILBURN, J., ADLARD,M. W., ARST,
668. H. N., JR. (1991), pH regulation of penicillin
SALYERS, A, A., SHOEMAKER, N. A., STEVENS, A. production in Aspergillus nidulans, FEMS Mi-
M., LI, L.-Y. (1995), Conjugative transposons: crobiol. Lett. 77, 209-212.
an unusual and diverse set of integrated gene SHAPIRO, S. (Ed.) (1989), Regulation of Secondary
transfer elements, Microbiol. Rev. 59, 579-590. Metabolism in Actinomycetes. Boca Raton, F L
SAMSON, S. M., BELAGAJE, R., BLANKENSHIP, D. CRC Press.
T., CHAMPMAN, J. L., PERRY,D., SKATRUD, P. SHEN,B., HUTCHINSON, C. R. (1994), Triple hy-
L., VAN FRANK,R. M., ABRAHAM,E. P., droxylation of tetracenomycin A2 to tetraceno-
BALDWIN, J. E., QUEENER, S. W., INGOLIA, T. mycin C in Streptomyces glaucescens, J. Biol.
D. (1985), Isolation, sequence determination Chem. 269,30726-30733.
and expression in Escherichia coli of the isopeni- SHERMAN, D. H., MALPARTIDA, F., BIBB,M. J.,
cillin N synthetase gene from Cephalosporium KIESER,H. M., HOPWOOD, D. A. (1989), Struc-
acremonium, Nature 318, 191-194. ture and deduced function of the granaticin pro-
SANCHEZ, L., BRANA,A. F. (1996), Cell density inducing polyketide synthase gene cluster of Strep-
fluences antibiotic biosynthesis in Streptomyces tomyces violaceoruber Tii22, EMBO J. 8, 2717-
clavuligerus, Microbiology 142, 1209-1220. 2725.
SHIER,W. T., RINEHART, K. L., JR., GO~TLIEB, D. STROBEL,G., YANG,X. S., SEARS,J., KRAMER,
(1969), Preparation of four new antibiotics from R., SIDHU,R. S., HESS, W. M. (1996), Taxol
a mutant of Streptomyces fradiae, Proc. Natl. from Pestalotiopsis microspora, an endophytic
Acad. Sci USA 63, 198-204. fungus of Taxus wallachiana, Microbiology 142,
SHIOZAWA,H., KAGASAKI,T., KINOSHITA,T., 435-440.
HARUYAMA, H., DOMON,H., UTSUI,Y., Ko- SUAREZ, T., PENALVA, M. A. (1996), Characteriza-
DAMA, K., TAKAHASHI, s. (1993), Thiomarinol, tion of a Penicillium chrysogenum gene encod-
a new hybride antimicrobial antibiotic produced ing a p a c e transcription factor and its binding
by a marine bacterium, J. Antibiot. 46, 1834- sites in the divergent pcbAB-pcbC promoter of
1842. the penicillin biosynthetic cluster, Mol. Micro-
SMITH,D. J., BURNHAM, M. K. R., BULL,J. H., biol. 20, 529-540.
HODGSON,J. E., WARD, J. M., BROWNE,P., SUMMERS, R. G., ALI, A., SHEN,B., WESSEL,W.
BROWN, J., BARTON,B., EARL,A. J., TURNER, A., HUTCHINSON, C. R. (1995), Malonyl-coen-
G. (1990), PLactam antibiotic biosynthetic zyme A-acyl carrier protein acyltransferase of
genes have been conserved in clusters in proka- Streptomyces glaucescens - a possible link be-
ryotes and eukaryotes, EMBO J. 9,741-747. tween fatty acid and polyketide biosynthesis,
SOLOMON, J. M., MAGNUSON, R., SRIVASTAVA, Biochemistry 34, 9389-9402.
A., GROSSMAN, A. D. (1995), Convergent sens- SUZUKI,K., YAMAGUCHI, H., MIYAZAKI, S., NA-
ing pathways mediate response to two extracel- GORI, K. (1990), Topostin, a novel inhibitor of
lular competence factors in Bacillus subtilis, mammalian DNA topoisomerase I from Flexi-
Genes Dev. 9,547-558. bacter topostinus sp. nov. I. Taxonomy, and fer-
SPRA-IT, B. G. (1994), Resistance to antibiotics me- mentation of producing strain, J. Antibiot. 43,
diated by target alterations, Science 264, 388- 154-157.
393. SZESZAK,F., VITALIS,S., BEKESI,I., SZABO,G.
STACHELHAUS, T., SCHNEIDER, R., MARAHIEL, (1991), Presence of factor C in streptomycetes
M. A. (1995a), Rational design of peptide anti- and other bacteria, in: Genetics and Product For-
biotics by targeted replacement of bacterial and mation in Streptomycetes (BAUMBACH, S., KRU-
fungal domains, Science 269, 69-72. GEL, H., NOACK,D., Eds.), pp. 11-18. New
STACHELHAUS, T., SCHNEIDER,R., MARAHIEL, York: Plenum Press.
M. A. (1995b), Engineered biosynthesis of pep- TAKANO,Y., KUBO,Y., SHIMIZU,K., MISE, K.,
tide antibiotics, Biochem. Pharmacol. 52, 177- OKUNO,T., FURUSAWA, I. (1995), Structural
186. analysis of PKSl, a polyketide synthase gene in-
STEELE,D. B., STOWERS,M. D. (1991), Tech- volved in melanin biosynthesis in Colletotrichum
niques for selection of industrially important mi- lagenarium, Mol. Gen. Genet. 249, 162-167.
croorganisms, Annu. Rev. Microbiol. 45, 89- TANAKA, Y. (1986), Biosynthesis of tylosin and its
106. regulation by ammonium and phosphate, in:
STEWART,G. C. (1993), Catabolite repression in Regulation of Secondary Metabolite Formation
the gram-positive bacteria: generation of nega- (KLEINKAUF,H., VON DOHREN, H., DOR-
tive regulators of transcription, J. Cell Biochem. NAUER, H., NESEMANN, G., Eds.), pp. 305-337.
51,25-28. Weinheim: VCH.
STIERLE,H., STIERLE,D., STROBEL,G., BIGNANI, TERCERO,J. A., ESPINOSA, J. C., LACALLE, R. A.,
G., GROTHANS, P. (1994), Endophytic fungi of JIMBNEZ,A. (1996), The biosynthetic pathway
pacific yew (Taxus brevifolia) as a source of tax- of the aminonucleoside antibiotic puromycin, as
01, taxanes and other pharmacophores, ACS deduced from the molecular analysis of the pur
Symp. Ser. 557 (Bioregulators for crop protec- cluster of Streptomyces alboniger, J. Biol. Chem.
tion and pest control), 64-77. 271,1579-1590.
STIERLE,A., STROBEL,G., STIERLE,D., GROT- THENBERG, K., LITZKA,O., BRAKHAGE, A. A.
HAUS,P., BIGNAMI, G. (1995), The search for a (1996), Identification of a major cis-acting ele-
taxol-producing microorganism among the en- ment controlling bidirectionally transcribed pen-
dophytic fungi of the pacific yew, Taxus brevifol- icillin biosynthesis genes acvA (pcbAB) and
ia, J. Nat. Prod. 58, 1315-1324. ipnA (pcbC) of Aspergillus nidulans, J. Bacteriol.
STINTZI, A., CORNELIS,P., HOHNADEL,D., 178,3908-3916.
MEYER,J.-M., DEAN,C., POOLE,K., KOURAM- THIERICKE, R., ROHR,J. (1993), Biological varia-
BAS, S., KRISHNAPILLAI,V. (19%), Novel PYOV- tion of microbial metabolites by precursor-di-
erdine biosynthesis gene(s) of Pseudornonas ae- rected biosynthesis, Nat. Prod. Rep. 12, 265-
ruginosa PAO, Microbiology 142, 1181-1190. 289.
6 References 53
TILBURN, J., SARKAR, S., WIDDICK,D. A., ESPE- provement and Process Control Strategies. Bos-
so, E. A., OREJAS,M., MUNGROO,J., PENAL- tonllondon: Butterworth.
VA,M. A., ARST,H. N., JR. (1995), The Asper- VANEK,Z., MAJOR,J. (1967), Macrolide antibiot-
gillus PacC zinc finger transcription factor me- ics, in: Antibiotics, Vol. 2 (GOTTLIEB, D., SHAW,
diates regulation of both acidic- and alkaline-ex- P. D., Eds.), pp. 154-188. New York, Heidel-
pressed genes by ambient pH, EMBO J. 14,779- berg, Berlin, Tokyo: Springer-Verlag.
790. VANEK,Z., MIKULIK, K. (1978), Microbial growth
TOMODA,H., OMURA,S. (1990), New strategies and production of antibiotics, Folia Microbiol.
for discovery of enzyme inhibitors: screening 23,309-328.
with intact mammalian cells or intact microor- VANEK,Z., BEHAL,V., JECHOVA,V., CURDOVA,
ganisms having special functions, J. Antibiot. 38, E., BLUMAUEROVA, M., HOSTALEK, Z. (1978),
1207-1222. Formation of tetracycline antibiotics, in: Anti-
TRAIL,F., MAHANTI, N., LINZ,J. E. (1995a), Mo- biotics and Other Secondary Metabolites (HUT-
lecular biology of aflytoxin biosynthesis, Micro- TER, R., LEISINGER,I., NOESCH,J., WEHRLI,
biology 141, 755-765. W., Eds.), pp. 101-112. London, New York, San
TRAIL,F., MAHANTI,N., RARICK,M., MEHIGH, Francisco: Academic Press.
R.,LIANG,S. H., ZHOU,R., LINZ,J. E. (1995b), VANEK,Z., CUDLIN,J., BLUMAUEROVA, H., Hos-
Physical and transcriptional map of an aflatoxin TALEK,Z. (1981), Physiology and Pathophysio-
gene cluster in Aspergillus parasiticus and func- logy of the Production of Excessive Metabolites.
tional disruption of a gene involved early in the Institute of Microbiology, CS Academy of
aflatoxin pathway, Appl. Environ. Microbiol. 61, Sciences, Prague.
2665-2673. VERALL,M. S. (Ed.) (1985), Discovery, and Isola-
Tsor, C. J., KHOSLA, C. (1995), Combinatorial bio- tion of Microbial Products. Chichester: Ellis
synthesis of "unnatural" natural products: the Horwood.
polyketide example, Chem. Biol. 2, 355-362. VINING,L. C. (1986), Secondary metabolism, in:
TURGAY,K., MARAHIEL, M. A. (1995), The gtcRS Biotechnology, 1st Edn., Vol. 4 (REHM,H.-J.,
operon coding for two-component system regul- REED,G., Eds.), pp. 19-38. Weinheim: VCH.
atory proteins is located adjacent to the grs ope- VINING,L. C. (1992), Role of secondary metabol-
ron of Bacillus brevis, DNA Sequence 5, 283- ites from microbes, in: Secondary Metabolites:
290. Their Function and Evolution, Ciba Foundation
TURNER, G. (1992), Genes for the biosynthesis of Symposium 171 (CHADWICK, D. J., WHELAN,J.,
plactam compounds in microorganisms, in: Sec- Eds.), pp. 184-198. Chichester, New York: J. Wi-
ondary Metabolites: Their Function and Evolu- ley & Sons.
tion, Ciba Foundation Symposium 171 (CHAD- VINING, L. C., STUTTARD,C. (1995), Chlor-
WICK,D. J., WHELAN,J., Eds.), pp. 113-128. amphenicol, in: Genetics and Biochemistry of
Chichester, New York: J. Wiley & Sons. Antibiotic Production (VINING,L. C., STUT-
TURNER, W. B., ALDRIDGE, D. C. (1983), Fungal TARD,C., Eds.), pp. 505-530. Boston, MA: But-
Metabolites II. London: Academic Press. terworth-Heinemann.
TYMIAK,A. A., Ax, H. A., BOLGAR,M. S., VINING,L. C., SHAPIRO,S., AHMED,Z., VATS, S.,
KAHLE,A. D., PORUBCAN, M. A., ANDERSEN, DOULL,J. STUTTARD,C. (1986), Genetic and
N. H. (1993), Dactylocyclines, novel tetracycline physiological control of chloramphenicol pro-
derivatives produced by Dactylosprangium sp. duction, in: Regulation of Secondary Metabolite
11. Structure elucidation, J. Antibiot. 45, 1899- Formation (KLEINKAUF, H., VON DOHREN,H.,
1%. DORNAUER, H., NESEMANN, G., Eds.), pp. 209-
UMEZAWA, H., HOOPER,I. R. (Eds.) (1982), Ami- 224. Weinheim: VCH.
noglycoside Antibiotics. New York, Heidelberg, VOGEL, R. (1984), Natiirliche Enzyminhibitoren.
Berlin, Tokyo: Springer-Verlag. Stuttgart: Thieme.
VANDAMME, E. J. (Ed.) (1984), Biotechnology of VON DER HELM,P., NEILANDS, J. B. (Eds.) (1987),
Industrial Antibiotics. New York, Basel: Marcel Iron Transport in Microorganisms, Plants and
Dekker. Animals. Weinheim: VCH.
VANEK,Z., HOSTALEK,Z. (1985), Tetracycline VON DOHREN,H. (1990), Compilation of peptide
biosynthesis, in: The Tetracyclines (HLAVKA,J. structures: a biogenetic approach, in: Biochemis-
J., BROTHE,J. H., Eds.), pp. 137-138. New try of Peptide Antibiotics (KLEINKAUF, H., VON
York, Heidelberg, Berlin, Tokyo: Springer-Ver- D ~ H R E NH.,, Eds.), pp. 411-507. Berlin, New
lag. York: de Gruyter.
VANEK,Z., HOSTALEK, Z. (Eds.) (1988), Overpro- VON D ~ H R E N H., (1995), Peptides, in: Genetics and
duction of Microbial Metabolites: Strain Im- Biochemistry of Antibiotic Production (VINING,
L., STUTTARD,C., Eds.), pp. 129-172. Boston, (1992), Small-molecule dependent density-con-
MA: Butterworth-Heinemann. trol of gene expression in prokaryotes. Biolumi-
WALKER,J. B., WALKER,M. S. (1978), Enzymatic nescence and the biosynthesis of carbapenem
synthesis of streptomycin as a model system for antibiotics, FEMS Microbiol. Lett. 100, 161-167.
study of the regulation and evolution of antibiot- WINKELMANN, G. (Ed.) (1991), Handbook of Mi-
ic biosynthesis, in: Overproduction of Microbial crobial Iron Chelates. Boca Raton, FL: CRC
Products (KRUMPHANZL, V., SIKYTA,B., VA- Press.
NEK,Z., Eds.), pp. 423-438. London, New York: WIRTH,R., WANNER,G., GALLI,D. (1990), The
Academic Press. sex pheromone system of Enterococcus faecalis:
WANNER,B. L. (1993), Gene regulation by phos- a unique mechanism of plasmid uptake (in Ger-
phate in enteric bacteria, J. Cell. Biochem. 51, man), Forum Mikrobiol. 6, 321-332.
47-54. WRIGHT,D. C. (1983), The leucomycin-celesticid-
WARD,J. M., HODGSON.J. E. (1993), The biosyn- in-anthramycin group, in: Biochemistry and Ge-
thetic genes for clavulanic acid and cephamycin netic Regulation of Commercially Important An-
production occur as “super-cluster” in three tibiotics (VINING, C. C., Ed.), pp. 311-328. Read-
Streptomyces, FEMS Microbiol. 110, 239-242. ing: Addison-Wesley.
WATANABE, T., SHIRMA,K., IZAKI,K., SUGIYA- YAMADA,Y., SUGAMURA, K., KONDO,K., YA-
MA, T. (1992), New polyenic antibiotic active
NAGIMOTO, M., OKADA,H. (1987), The struc-
against gram-positive and gram-negative bacte- tures of inducing factors €or virginiamycin pro-
ria. VII. lsolation and structure of enacyloxin duction in Streptomyces virginiae, J. Antibiot. #,
IVa, possible biosynthetic intermediate of en- 496-504.
acyloxin IIa, J. Antibiot. 45, 575-576.
WEBER,G., SCHORGENDORFER, K., SCHNEIDER- YANG,Y. K., SHIMIZU, H., SHIOYA,S., SUGA,K.-
SCHERZER, E., LEITNER,E. (1994), The peptide I., NIHIRA,T., YAMADA,Y. (1995a), Optimum
synthetase catalyzing cyclosporine production in autoregulator addition strategy for maximum
Tolypocladium niveum is encoded by a giant virginiamycin production in batch cultivation of
45.8 kilobase open reading frame, Curr. Genet. Streptomyces virginiae, Biotechnol. Bioeng. 46,
26, 120-125. 437-442.
WICKLOW,D. T., DOWD, P. F., GLOER,J. B. YANG,K. Q., HAN,L., VINING,L. C. (1995b), Reg-
(1994), Antiinsectian effects of Aspergillus metab- ulation of jadomycin B production in Strepto-
olites, in: The Genus Aspergillus from Taxono- myces venezuelae ISP5230 - involvement of a re-
my and Genetics to Industrial Applications pressor gene, jadR(2), J. Bacteriol. 177, 6111-
(POWELL,K., RENWICK, A., PEBERDY, J., Eds.), 6117.
pp. 93-114. New York: Plenum Press. YANG, Y. K., MORIKAWA,M., SHIMIZU,H.,
WIESMANN, K. E. H., CORTBS,J., BROWN,M. J. SHIOYA,S., SUGA, K.-I., NIHIRA,T., YAMADA,
B., CUTTER,A. L., STAUNTON, J., LEADLEY,P. Y. (1996a), Maximum virginiamycin production
F. (1995), Polyketide synthesis in vitro on a mod- by optimization of cultivation conditions in
ular polyketide synthase, Chemistry & Biology batch culture with autoregulator addition, Bio-
2,583-589. technol. Bioeng. 49, 437444.
WILLIAMS,D. H., MAPLESTONE,R. A. (1992), YANG,K. Q., HAN,L., AYER,S. W., VINING,L. C.
Why are secondary metabolites synthesized? So- (1996b), Accumulation of the angucycline anti-
phistication in the inhibition of cell wall biosyn- biotic rabelomycin after disruption of an oxy-
thesis by vancomycin group antibiotics, in: Sec- genase gene in the jadomycin B biosynthetic
ondary Metabolites: Their Function and Evolu- gene cluster of Streptomyces venezuelae, Micro-
tion, Ciba Foundation Symposium I71 (CHAD- biology 142,123-132.
WICK, D. J., WHELAN,J., Eds.), pp. 45-63. Chi- YE, J., DICKENS,M. L., PLATER,R., LI, Y., LAW-
Chester, New York: J Wiley & Sons. RENCE, J., STROHL, w. R. (1994), Isolation and
WILLIAMS, S. T., VICKERS,J. L. (1986), The ecolo- sequence analysis of polyketide synthase genes
gy of antibiotic production, Microb. Ecol. U,43- from the daunomycin-producing Streptomyces
52. sp. strain C5, J. Bacteriol. 176, 6270-6280.
WILLIAMS,D. H., STONE,M. J., HAUCK,P. R., YLIHONKO, K., TUIKKANEN, J., JUSSILA,S., CONG,
RAHMAN,S. R. (1989), Why are secondary me- L. N., MANTSALA,P. (1996), A gene cluster in-
tabolites (natural products) biosynthesized? J. volved in nogalamycin biosynthesis from Strep-
Nut. Prod. 52, 1189-1208. tomyces nogalater - sequence analysis and com-
WILLIAMS,P., BAINTON,N. J., SWIFT,S., SCHA- plementation of early block mutations in the an-
BRA,C. R., WILSON,M. K., STEWARD,G., MI- thracycline pathway, Mol. Gen. Genet. 251, 113-
CHAEL, K., SALMAND, G. P. c.,BYROFT,B. w. 120.
6 References 55
Yu, J. H., LEONARD,T. J. (1995), Sterigmatocystin tion and secondary pathways, in: Secondary Me-
biosynthesis in Aspergillus nidulans requires a tabolism and Cell Differentiation in Fungi (BEN-
novel type I polyketide synthase, J. Bacteriol. N E ~J. W.,
, CIEGLER, A., Eds.), pp. 51-70. Lon-
177,4792-4800. don, New York: Academic Press.
Yu, T . W., BIBB,M. J., REVILL,W. P., HOPWOOD, ZAKELJMAVRIC, M., KASTELICSUHADOLC, T.,
D. A. (1994), Cloning, sequencing and analysis PLEMENITAS,A., RIZNER,T. L., BELIC, I.
of the griseusin polyketide synthase gene cluster (19959, Steroid hormone signalling system and
from Streptomyces griseus, J. Bacteriol. 176, fungi, Comp. Biochem. Physio1.-B. Comp. Bio-
2627-2634, chem. 112,6374142.
ZAHNER,H., ZEECK,A. (1987), Mikrobieller Se- ZMIJEWSKI, M. J., JR., FAYERMAN, J. T. (1995),
kundarstoffwechsel, in: Jahrbuch der Biotechno- Glycopeptides, in: Genetics and Biochemistry of
logie (PRAvE, P., Ed.), pp. 93-113. Miinchen, Antibiotic Production (VINING,L. C., STUT-
Wien: Oldenbourg. TARD, C., Eds.), pp. 269-282. Boston, MA: But-
ZAHNER,H., ANKE,H., ANKE,T. (1983), Evolu- terworth-Heinemann.
Biotechnology
2 Regulation of Bacterial Antibiotic

Production
KEITH F. CHATER
MERVYNJ. BIBB
Norwich, UK
1 Introduction 59
1.1 The Scope of this Chapter 59
1.2 Cellular Efficiency Involves Extensive Regulation of Metabolism in Response to
Growth Conditions 59
1.3 Antibiotic Production Does Not Usually Occur in Rapidly Growing Cultures 59
2 Themes in the Regulation of Antibiotic Production Illustrated by Examples from
Unicellular Bacteria 61
2.1 Intracellular Signals Associated with Starvation and Low Growth Rate Activate
Microcin C7 Synthesis in E. coli 61
2.2 A Critical Cell Population Density Signaled by an Autogenous Extracellular Signal
Molecule Triggers Carbapenem Synthesis by Erwinia carotovora 65
2.3 The Non-Ribosomal Production of Peptide Antibiotics in Various Bacillus spp. Is One
of a Number of Alternative Stationary-Phase Fates Determined by a Network of
Transition State Regulators Involving Protein Phosphorylation 67
2.4 What Has Been Learned from Studies of Antibiotic Production in Unicellular
Bacteria? 70
3 Regulation of Antibiotic Production in Streptomycetes and their Relatives 70
3.1 Introduction to the Organisms 70
3.2 General Physiological Aspects of the Regulation of Antibiotic Production in
Streptomyces 71
3.2.1 Metabolite Interference with Antibiotic Production in Streptomycetes 72
3.2.2 Antibiotic Production and Imbalances in Metabolism 74
3.2.3 The Possible Role of Growth Rate and ppGpp in Antibiotic Production 75
3.2.4 Antibiotic Production and the Accumulation of Small Diffusible Signaling
Compounds 78
3.2.5 Summary 80
58 2 Regulation of Bacterial Antibiotic Production
3.3 Genetics of Antibiotic Production 80

3.3.1 Organization of Antibiotic Biosynthetic Genes and Clusters 80
3.3.2 Genes that Pleiotropically Affect Antibiotic Production in Streptomyces
coelicolor: Introduction and Overview 81
3.3.2.1 afsB, afsR, and afsK - A Role for Protein Phosphorylation in Triggering
the Onset of Antibiotic Production? 81
3.3.2.2 afsQZ and afsQ2 - A Two-Component Regulatory System that Can
Influence Antibiotic Production 83
3.3.2.3 abaA Influences the Production of Three of the Four Antibiotics Made
by Streptomyces coelicolor 83
3.3.2.4 absA and absB - Mutants Isolated on the Basis of a Pleiotropic Defect in
Antibiotic Production 83
3.3.2.5 mia - Multicopy Inhibition of Antibiotic Production 84
3.3.2.6 Genes that Affect Both Antibiotic Production and Morphological
Differentiation 84
3.3.2.7 An Outline Scheme for the Interactions of Pleiotropic Antibiotic
Regulatory Genes in Streptomyces coelicolor 85
3.3.3 Streptomyces griseus - The A-Factor Cascade 87
3.3.4 Pathway-Specific Regulatory Genes 88
3.3.4.1 The actZZ-ORF4, redD, and dnrl Family of Pathway-Specific Activator
Genes 89
3.3.4.2 srmR - A Regulatory Gene for Spiramycin Production in Streptomyces
ambofaciens 90
3.3.4.3 strR Encodes a DNA-Binding Protein that Regulates at Least One of the
Streptomycin Biosynthetic Genes in Streptomyces griseus 90
3.3.4.4 brpA and dnrN - Regulatory Genes for Bialaphos Production in
Streptomyces hygroscopicus and for Daunorubicin Production in
Streptomyces peucetius which Show Different Degrees of Similarity to
Response Regulator Genes of Two-Component Systems 91
3.3.4.5 Negative Regulation of Methylenomycin Production in Streptomyces
coelicolor 91
3.3.5 Induction of Antibiotic Resistance in Antibiotic Producing Streptomycetes -
Antibiotics as Inducers of Gene Expression 92
3.3.5.1 actZZ-ORF1/2 of Streptomyces coelicolor and tcmWA of Streptomyces
glaucescens GLA.0 - Regulatory Cassettes for Antibiotic Export 92
3.3.5.2 srmB of Streptomyces ambofaciens - A Probable ATP-Dependent Efflux
System Induced by its Antibiotic Substrate 92
3.3.5.3 Induction of tlrA in Streptomyces fradiae - A Role for Transcriptional
Attenuation? 93
3.3.5.4 Induction of Resistance to Novobiocin in Streptomyces sphaeroides and
Streptomyces niveus - Roles for DNA Supercoiling and a Diffusible
Signaling Molecule 93
3.3.5.5 Regulation of Isoforms of the Target for Pentalenolactone Inhibition in
the Producing Organism 93
4 Concluding Remarks 94
5 References 96
I introduction 59
1 Introduction steps in the pathways and also cause repres-

sion of synthesis of many of the biosynthetic
enzymes, mostly by mechanisms involving re-
1.1 The Scope of this Chapter pressor proteins or transcriptional or transla-
tional attenuation. (We are not aware of any
Study of the molecular basis for the regula- such feedback regulatory loops in antibiotic
tion of antibiotic biosynthesis has gradually synthesis.) Likewise, the assimilation of car-
expanded since production genes were first bon usually involves specific induction of
cloned from Streptomyces spp. more than 10 genes for the relevant enzymes, usually by in-
years ago (see HOPWOOD et al., 1983 and teraction of a specific transcriptional repres-
MART~N and GIL,1984 for reviews of the ear- sor or activator protein with the substrate or a
ly history of this subject). This has led to a simple derivative of it generated in the cell by
plethora of recent reviews (e.g., MART~N and a constitutive low level of the pathway en-
LIRAS,1989; SENOand BALTZ,1989; CHAT- zyme(s). More global regulation is also neces-
ER, 1990, 1992; CHAMPNESS and CHATER, sary. For example, most free-living microbes
1994; HUTCHINSON et al., 1994) which have possess an integrated system of carbon cata-
dealt almost exclusively with Streptomyces bolite repression which ensures that the most
and related actinomycetes. Recently, knowl- readily utilized and ergogenic substrate (often
edge of the regulation of antibiotic biosynthe- glucose) is used in preference to less favora-
sis in non-actinomycete bacteria has emerged ble substrates. In E. coli, such repression op-
almost as a by-product of studies of the switch erates in two ways: via interference with up-
from rapid growth to stationary phase in take of the less favorable compound, prevent-
model organisms such as Escherichia coli and ing it from participating in induction of the
Bacillus subtilis and of mechanisms of plant enzymes needed for its assimilation (“inducer
pathogenicity in Erwinia carotovora. This has exclusion”) and through more direct in-
given a new opportunity, in this review, to ex- fluences on transcription of the genes for
amine both general themes common to the these enzymes (SAIER,1989; POSTMAet al.,
regulation of production of diverse antibiotics 1993; Fig. 1). In the latter case, the pattern of
by diverse bacteria, and the idiosyncrasies of available carbon sources determines the in-
different bacterial genera or different path- tracellular pool of CAMP, which in turn di-
ways within particular bacteria. rectly interacts with a CAMP-binding protein,
and modulates its ability to interact with the
transcriptional initiation complex at promot-
ers of various gene sets for utilization of car-
1.2 Cellular Efficiency Involves bon sources (Fig. 1). Carbon catabolite re-
Extensive Regulation of pression may, however, be exerted by differ-
Metabolism in Response to ent mechanisms in different microbes (SAIER,
1991).
Growth Conditions
When nutrients are abundant and readily 1.3 Antibiotic Production Does
available, microorganisms grow fast. In mixed Not Usually Occur in Rapidly
communities, rapid conversion of nutrients to
biomass is the overriding theme of metabol- Growing Cultures
ism and its efficiency is maximized by the
well-known regulatory systems that govern In bacteria, antibiotic production nearly al-
such assimilation (many of which are re- ways takes place only after rapid growth has
viewed by NEIDHARDT et al., 1987). Some of ceased. To a large extent, this chapter reviews
these are pathway-specific. Thus, in the feed- the molecular and physiological mechanisms
back loops of amino acid biosynthetic path- that underpin this general observation, but in
ways of E. coli, the end products typically in- this section we briefly reflect on the possible
hibit the activity of enzymes for the earliest adaptive significance of this regulatory pat-
Mtl jout)
Glucose
(out)
Fig. 1. Global regulation of carbon catabolite-repressible operons in E. coli involves multiple influences of
membrane-bound and cytoplasmic components. During uptake of glucose by EIIBCG1' (components B
and C of FTSG'', the glucose phosphotransferase system), the sugar is phosphorylated by the phosphory-
lated form of the cytoplasmic EIIAG" which is thereby itself dephosphorylated. This has three important
regulatory consequences. (1) The activity of adenylate cyclase is reduced, because it depends on interac-
tion with EIIG1'-P. This causes a drop in the level of CAMPavailable to bind to and activate the catabolite
repression protein (CRP). The CAMP-CRP complex is necessary for efficient transcription of many glu-
cose-repressible promoters. (2) Unphosphorylated EIIAG'' directly inhibits non-PTS sugar permeases, ex-
cluding the relevant sugars from the cytoplasm and, therefore, preventing them from inducing the relevant
genes for sugar catabolism. (3) The unphosphorylated EIIAG" competes with other PTS systems such as
-
EIIABCM", the uptake system for Mfl, for the phosphate-donating protein HPr P, thereby indirectly
inhibiting uptake of other PTS sugars EI, part of the enzyme cascade responsible for the transfer of phos-
phate from PEP to PTS sugars.
tern. We assume that antibiotic activities ob- olism being devoted to secondary metabo-
served experimentally have actual evolution- lism. Even at this late stage in exploiting an
ary significance: i.e., that antibiotics help the environment there is still potential advantage
producing organisms by inhibiting their com- to be gained from inhibiting competitors. This
petitors in natural environments. Why, then, could take several forms: inhibiting the devel-
do organisms not produce antibiotics opment of more persistent resting stages of
throughout growth to maximize this competi- competitors; greater competitiveness in the
tive advantage? Perhaps the answer lies in the hidden population dynamics of stationary
comparatively high diversion of resources phase (during which minor subpopulations of
away from biomass accumulation that might cells grow at the expense of the majority of
be required if the few cells present during the population) (ZAMBRANO et al., 1993); or,
early growth are to produce an inhibitory levin the case of developmentally complex or-
el of antibiotic. This might conflict with the ganisms such as streptomycetes, to prevent in-
need to grow as rapidly as possible in the vasion of colonies by competitors after the ly-
competition for the nutritional resources of a sis of some of the cells within colonies, which
new environment. On the other hand, the ef- may provide nutrition for spore development
fective production of chemical weapons at re- (CHATERand MERRICK,1979; MBNDEZet
latively high population density (i.e., no ear- al., 1985; but see also O'CONNORand Zus-
lier than the last few cell divisions before nu- MAN, 1988). Viewed in this way, one impor-
trient exhaustion) can be achieved with a tant aspect of the regulation of antibiotic pro-
much smaller proportion of each cell's metab- duction should be the mechanisms by which
2 Themes in the Regulation of Antibiotic Production Illustrated 61
information about population density or nu- (2)responsible for expression of many sta-
trient availability is perceived by cells. This tionary-phase genes (MULVEYand LOEWEN,
information must then be interpreted by the 1989; TANAKA et al., 1993; NGUYENet al.,
cell and ultimately used to activate the specif-1993; see Fig. 2 for a summary of u factor
ic relevant pathways of antibiotic biosynthe- structure and function). (The variation in mcc
sis. transcription among E. coli strains is consis-
tent with the finding that cultures left in sta-
tionary phase are often taken over by mu-
tants in which the C-terminus of 6s is deleted
ZAMBRANO et al., 1993.) Thus microcin C7 is
2 Themes in the produced only during stationary phase be-
cause, directly or indirectly, its production
Regulation of Antibiotic genes are activated via #. The nature of mcc
promoters and the conserved features of us-
Production Illustrated by dependent promoters in general remain to be
Examples from Unicellular fully elucidated. There is, however, some
overlap between the promoter class recog-
Bacteria nized by 6s and that recognized by the princi-
pal sigma factor, u70 (TANAKAet al., 1993;
NGUYENet al., 1993). This is consistent with
2.1 Intracellular Signals the close similarities between the regions of
Associated with Starvation and usand u70expected to make sequence-specif-
Low Growth Rate Activate ic DNA contacts (regions 2.4 and 4.2 in Fig.
2). Some promoters that are u7O-dependent
Microcin C7 Synthesis in E. coli and require activators during rapid growth
may perhaps be utilized during stationary
Bacterial antibiotic production is generally phase in an activator-independent manner by
found in organisms such as Streptomyces spp. RNA polymerase containing 2 (KOLTERet
that undergo complex differentiation (CHAT- al., 1993).
ER and MERRICK,1979), but many simple How is 2 activity increased on entry into
unicellular bacteria are also producers. Thus, stationary phase? GENTRY et al. (1993)
some E. coli strains produce microcins, a het- showed that uslevels respond to intracellular
erogeneous collection of inhibitory com- changes in the concentration of the important
pounds many (but not all) synthesized non- signaling molecule ppGpp, best known for its
ribosomally. The regulation of production of role in mediating the stringent response
one such compound, microcin C7, provides a (Fig.3). Levels of ppGpp increase in E. coli
nice example of the activation of antibiotic when cultures are limited for amino acids,
synthesis in response to a shift-down in cellu- inorganic nitrogen, or carbon (IRR,1972; CA-
lar metabolism. SHEL and RUDD,1987), and probably also for
Microcin C7 is a ca. loo0 Da oligopeptide phosphate (GENTRYet al., 1993), so ppGpp is
antibiotic whose production, in post-exponen- probably a regulator of entry into stationary
tial phase, is specified by genes (mcc) located phase. Consistent with this view, increasing
on the plasmid pMccC7 (NOVOAet al., 1986). the steady-state level of ppGpp either by the
Only certain laboratory strains of E. coli K-12 use of mutants deficient in ppGpp degrada-
support transcription of mcc-lac2 fusions tion or by manipulating the expression of a
(DfAZ-GUERRA et al., 1989), other Strains truncated ppGpp synthetase gene leads to a
bearing mutations in a locus that turned out reduction in growth rate under conditions of
to coincide with appR, initially studied as a nutritional sufficiency (SARUBBIet al., 1988
regulatory gene for acid phosphatase synthe- SCHREIBERet al., 1991). Interestingly an E.
sis. The appR gene, formerly also referred to coli strain unable to make ppGpp has a phe-
as nur and katF, is now known as rpoS and notype somewhat like that of an rpoS mutant
encodes an RNA polymerase sigma factor (GENTRYet al., 1993).
a promoter b /
Primaryandclosely Heat shock
Bacillus
sporulation
flagellar
SigB Idarillus)
5 elongation
complex
\ /6factorrelease /
Fig. 2. (a) The role of u factors in initiation of transcription. Nearly all u factors are related to each other
and share certain features. At (1) we emphasize two regions that interact with promoters; region 2.4 of any
particular uinteracts about 10 bp upstream of the transcription start point ( + 1)at sequences characteristic
of promoters dependent on that u,and region 4.2 interacts with DNA about two helical turns further
upstream (-35). (2) The u-DNA interaction is largely dependent on association of u with RNA polymer-
ase core enzyme to give RNA polymerase holoenzyme which is thereby directed to appropriate promoters.
(3) Initially a closed promoter complex is formed in which the DNA remains double-stranded. RNA poly-
merase-promoter interactions may be significantly affected at this and the next stage by contacts with
regulatory proteins, especially transcriptional activators, bound a short distance upstream. (4) The u factor
plays an important role in melting the DNA around +1 to give an open complex. (5) RNA polymerase
begins to transcribe, and the u factor is ejected and may become associated with another core enzyme
particle to reinitiate the cycle. (6) Eventually the completed mRNA and the RNA polymerase core enzyme
are released. The core enzyme may now potentially associate with a different u factor to initiate transcrip-
tion from a promoter of a different class. (b) Phylogeny of u factors. The familial relationships of most
known u factors from diverse bacteria are shown here. Arrows indicate u factors referred to in this chap-
ter. The diagram is basically that of LONETTOet al. (1994).
It is not clear how the increased ppGpp lev- nine biosynthesis). These intermediates could
els might cause increases in the level of 2 then be cyclized to homoserine lactone via a
(indeed, the mechanism of the stringent re- known interaction with tRNA synthetases.
sponse is still elusive). HUISMANand KOL- Homoserine lactone is proposed to be the
TER (1994b) suggested a speculative model critical intracellular regulator for induction of
for the effect of ppGpp, contingent on the rpoS. The key pieces of evidence to support
well-known association of increased ppGpp this model are: (1) the discovery of an E. coli
levels with increased expression of genes for gene, rspA, encoding a product resembling a
amino acid biosynthesis (Fig. 4). It predicts known lactonizing enzyme, which switches off
that the resultant increase in the threonine rpoS transcription when present at high copy
pool would cause feedback inhibition of number (RspA may degrade the homoserine
threonine biosynthesis, and so an increase in lactone); and (2) the elimination of rpoS ex-
the pool of homoserine and homoserine pression by mutations blocking homoserine
phosphate (which normally feed into threo- synthesis. The model was proposed following
Amino acid
stwation
Carbon
\\
starvation
\
tRNA
A
Uncharged
\1 .
Occupation of
ribosome A site
\f
Activity of RelA
( ribosome-bound
ppGpp synthetase)
Active
ribosomes
\
\
Re A
inactive
-
2 Themes in the Regu‘lation of Antibiotic Production Illustrated
Nutrient limitation
THREONINE
‘u
Stringent response
Increased synthesis
o f amino acids
Binding t o tRNA synthetases

/
LACTONE
THREONINE
63
Fig. 3. Activating ppGpp synthesis in E. coli. Most

is known about how amino acid starvation causes
ppGpp synthesis. The effects of carbon and espe-
cially phosphate starvation are comparatively little
8
0
Increased expression o f rpoSls) bs
Fig. 4. Speculative model connecting increased

studied. SpoT can apparently cause ppGpp synthe- ppGpp levels with increased production of d in E.
sis in a relA null mutant (HERNANDEZ and BRE- coli. The diagram represents the model formulated
MER, 1991; XIAOet al., 1991), but its major physio- by HUISMAN and KOLTER(1994b).
logical role is its ppGpp-degrading activity which is
inhibited under carbon-starved conditions (GEN-
TRY et al., 1993).
(Fig. 5). It has even turned out that a segment
of the rpoS mRNA responds to osmotic
stress, and a model has been proposed in
the discovery that homoserine lactones are which (by analogy with the situation for the
widespread as regulatory molecules (see mRNA for the heat shock sigma factor d2of
Sect. 2.2). E. colz) an internal segment of the mRNA
The regulation of us is proving to be very binds to the translational initiation region,
intricate (see HUISMAN and KOLTER,1994a, acting as an antisense inhibitor (LANCEand
for a review). Not only is its transcription ap- HENGGE-ARONIS, 1994). The stability of the
parently susceptible to subtle regulation by sense-antisense interaction is postulated to
various metabolic influences, but there is also depend on an osmotically sensitive protein.
regulation at the levels of translation and pro- Also, us half-life increases from about 2 min
tein stability (HECKER and SCHROETER, during rapid growth to 10-20 min during sta-
1985; LOEWENet al., 1993; MCCANNet al., tionary phase (LANCEand HENGGE-ARONIS,
1993; TAKAYANAGI et al., 1994; LANGEand 1994; TAKAYANAGI et al., 1994). Increased
HENGGE-ARONIS, 1994): indeed, posttrans- translation of rpoS mRNA at late exponential
criptional regulation is the overriding in- phase also implies that cell density informa-
fluence on us activity in minimal medium tion may be a component of 2 regulation,
Stationary phase,
/--
.Degradatioy
. I .
Highcell density,
Osmotic shock
Fig. 5. Complex control of the 2 subunit of E. coli RNA polymerase. The diagram is based on the data
and models of LANGEand HENGGE-ARONIS (1994) and TAKAYANAGI et al. (1994). Transcription of rpoS
is initiated from as many as four promoters. The CAMP-CRP complex (Fig. 1) inhibits use of one (or
more) of these promoters, and ppGpp stimulates use of one or more of them, possibly by causing increased
intracellular concentrations of homoserine lactone (HSL) (Fig. 4). Translation of the rpoS mRNA is
thought to be limited by a (protein-stabilized?) secondary structure that sequesters the ribosome-binding
site (RBS). This secondary structure is destabilized under some conditions, such as during osmotic shock,
releasing the RBS for translation. The resulting usprotein is rapidly degraded during the growth phase,
but is more stable in stationary phase, allowing it to direct RNA polymerase to transcribe stationary phase-
associated genes such as those determining microcin C biosynthesis.
perhaps involving extracellular substances stationary phase genes (ROMEOet al., 1993).
(LANCEand HENGGE-ARONIS, 1994). Exam- Interestingly, one of the genes needed for gly-
ples of such situations in other bacteria are cogen synthesis, glgS, maps away from the
discussed in the following sections. csrAlcAMP-CRPlppGpp-regulatedglg genes
Not all stationary-phase activities in E. coli and shows a clear dependence on us
are regulated by us (LANCEand HENGGE- (HENGGE-ARONIS and FISCHER,1992). The
ARONIS,1991a; MCCANNet al., 1991). For intricacy of stationary phase regulation is fur-
example, induction of many proteins by glu- ther illustrated by the finding that CAMP/
cose starvation is $-independent and re- CRP may also influence the expression of
quires the cAMP/CRP system (Fig. 1). A case rpoS (LANCEand HENGGE-ARONIS, 1994).
in point is provided by some of the E. coli Most of these stationary-phase regulatory
genes (notably the &CAY operon) for sta- devices are trans-acting, but at least one cis-
tionary phase-associated synthesis of glyco- acting mechanism is known: the “gearbox”
gen under conditions of nitrogen limitation promoter. Such promoters may be defined by
(ROMEOand PREISS,1989). Transcription of their property of enhanced relative expres-
glgCA Y is unaffected by rpoS mutations sion at low growth rates, coupled with resem-
(HENGGE-ARONIS and FISCHER,1992), and blance to a particular unusual - 10 consensus
it is not clear what form of RNA polymerase sequence (VICENTEet al., 1991). Gearbox
is involved in glgCA Y promoter recognition promoters have been discussed mostly in the
(ROMEOand PREISS,1989). Both ppGpp and context of some cell cycle-related genes, the
cAMP/CRP have significant regulatory im- expression of which may have a specially im-
pact on glgCA Y , particularly when the carbon portant relationship to growth rate, as well as
source is not glucose. The &CAY operon is in relation to stationary phase. The promoter
also subject to repression during growth by a of the mcbA-G operon, which specifies the
6.8 kDa protein encoded by the csrA gene ribosomally synthesized antibiotic microcin
which may also be an important regulator of B17, has gearbox kinetics and a gearbox-like
-10 region (VICENTEet al., 1991). There is unable to make OHHL (BAINTONet al.,
little information about what gives these pro- 1992a, b). Group 2 mutants are also pleio-
moters their property of gearbox kinetics, but tropically defective in the production of var-
it is not the result of recognition by a particu- ious exoenzymes associated with the degrada-
lar u factor (LANGEand HENGGE-ARONIS, tion of plant tissues during the disease pro-
1991b). cess, and this entire phenotype can be re-
versed by OHHL.
OHHL had been identified earlier as an ex-
2.2 A Critical Cell Population tracellular signaling molecule in a different
context: it is required to trigger light emission
Density Signaled by an by the marine organism Vibrio fischeri
Autogenous Extracellular Signal (MEIGHEN,1991), as a very effective signal of
Molecule Triggers Carbapenem increasing cell density (WILLIAMS, 1994;
Fig. 7). OHHL synthesis requires the action
Synthesis by Erwinia carotovora of the luxl gene product in a single-step bio-
synthesis from intermediary metabolism (pos-
PLactam antibiotics are produced by div- sibly from S-adenosyl methionine and 3-0x0-
erse bacteria including species of Strepto- hexanoyl coenzyme A) (EBERHARD et al.,
myces, Nocardia, Flavobacterium, and Erwi- 1981). During growth at low cell densities,
nia, as well as by fungi. There is surprisingly low-level expression of luxl results in a slow
little information about the genetic regulation accumulation of OHHL in the environment.
of plactam biosynthesis. The extreme amen- As cultures become denser, so the concentra-
ability of many purple gram-negative bacte- tion of OHHL builds up. Since OHHL is pre-
ria, including the carbapenem producer Er- dicted to be freely diffusible through mem-
winia carotovora, to rapid genetic manipula- branes (because of its lipophilic side chain),
tion has provided the most penetrating infor- the intracellular levels also increase until they
mation so far. However, the different life- are high enough (lo-’ M) for effective bind-
styles and ecologies of the different producers ing to a specific cytoplasmic receptor protein
may mean that their regulatory systems have (LuxR). The LuxR-OHHL complex can
diverged much more than the structural stimulate transcription of luxl, thereby caus-
genes. ing increased OHHL synthesis and reinforc-
In apparent contrast to the situation de- ing the signal that activates luminescence.
scribed above for microcin C7 production by Since luxl is part of an operon that also en-
E. coli, production of carbapenem by E. caro- codes luciferase, light emission is strongly ac-
tovora is regulated by a specialized extracel- tivated in response to a threshold level of
lular signal. Carbapenem non-producing OHHL. Regulatory systems that recognize
(Car-) mutants fall into two classes: group 1 critical levels of population density have been
mutants produce N-(3-oxohexanoyl)-~-homo-called “quorum sensors” (FUQUA et al.,
serine lactone (OHHL) (Fig. 6; EBERHARD1994). In the model described earlier impli-
et al., 1981) which restores the Car+ pheno- cating homoserine lactone as a hypothetical
type to group 2 mutants which are themselves intra- (rather than inter-) cellular regulatory
factor in E. coli it is supposed that the lactone
has no lipophilic side chain, so it is not re-
leased from the cell (HUISMAN and KOLTER
1994b).
OHHL and functional homologs of it have
b& 0 ”
turned out to be widespread. This has been
demonstrated by introducing a V. fischeri lux
gene set deleted for luxl into E. coli, and ex-
Fig. 6. Structure of the luminescence autoinducer posing the transformed strain to culture su-
N-(3-oxohexanoyl)-~-homoserine lactone (OHHL) pernatants of different bacteria. Lumines-
of Vibrio fischeri. cence was induced by samples from 18 strains
Fig. 7. Cell population-dependent expres-

sion of light emission mediated by the au-
toinducer OHHL of Vibrio fischeri. As
the cell density increases, the concentra-
tion of freely permeating OHHL be-
comes high enough to bind to LuxR
which then becomes an activator of the
1uxICDABE operon encoding a protein
(LuxI) that causes OHHL synthesis
(hence giving an autoinducing regulatory
loop) and enzymes required for lumines-
cence. The structure of OHHL is shown
in Fig. 6.
of gram-negative bacteria of 13 genera and 6 least two other OHHL-regulated promoters,

strains of gram-positive bacteria of 5 genera for traA in Agrobacterium tumefaciens and
(WILLIAMS, 1994). In several of the gram-ne- lasB of Pseudomonus aeruginosa (FUQUAet
gatives OHHL (or a related molecule) acts as al., 1994).
an indicator of cell density, and each organ- Carbapenem production is the only exam-
ism contains a gene that can substitute for ple of antibiotic production known to be
luxl (SWIFTet al., 1993). The luxl homologs “quorum-regulated’’ by OHHL-like regula-
form a rather widely diverged family of genes, tors in non-differentiating bacteria (though
but many are accompanied by luxR-like genes quorum regulation is a feature of production
encoding specific lactone-binding regulatory of some antibiotics by some differentiating
proteins (the LuxR family; FUQUAet al., bacteria including Bacillus subtilis and per-
1994), each of which is presumably specific haps Streptomyces griseus; see Sects. 2.3,3.2.4,
for a set of genes that determine a property and 3.3.3). However, existing screening sys-
responsive to cell population density. The tems may well miss some OHHL-related
emerging picture of these regulatory proteins compounds because there is considerable spe-
is of an N-terminal domain for lactone bind- cificity for binding. This has been revealed
ing and multimerization and a C-terminal do- both by analyzing the efficacy of synthetic
main with DNA-binding and transcription-ac- analogs of OHHL (WILLIAMS,1994) and by
tivating regions. The C-terminal domain re- the finding that the lux genes of another lumi-
sembles those of some other transcriptional nescent species, Vibrio hurveyi, respond to a
activators unconnected with quorum respon- closely related molecule, N-(3-hydroxybuty-
siveness, including the UhpA subfamily of re- ryl) homoserine lactone (CAO and MEIGHEN,
sponse regulators (Sect. 2.3), and even of 1989), instead of OHHL.
most (T factors (the LuxR superfamily; Fu-
QUA et al., 1994). LuxR probably binds to a
20 bp inverted repeat centered at about -40
from the transcription start point, and similar
sequences occupy equivalent positions in at
2.3 The Non-Ribosomal ways. All these influences appear to be trans-

mitted through a single transcriptional activa-
Production of Peptide Antibiotics tor, ComA, which also plays a key role in the
in Various Bacillus spp. Is One of onset of competence for transformation
a Number of Alternative (DUBNAU,1993; Fig. 8). ComA belongs to
the response regulator class of transcriptional
Stationary-Phase Fates activators (WEINRAUCH et al., 1989; Fig. 9),
Determined by a Network of activity of which is generally controlled by
phosphorylation of a conserved aspartate re-
Transition State Regulators sidue. This residue, and others involved in
Involving Protein Phosphorylation forming a phosphorylation pocket, are pres-
ent in ComA, so it is not surprising that its
The dissection of stationary-phase func- ability to bind to the srfA promoter in vifro is
tions has been most extensively analyzed in greatly enhanced by phosphorylation (ROG-
the gram-positive Bacillus subtilis 168. This GIANO and DUBNAU, 1993). In its active con-
reflects widespread interest in two of these figuration, ComA binds to each of two similar
functions, the formation of resistant endo- dyad elements upstream of the srfA promot-
spores and the occurrence of natural compe- er, and the bound proteins are presumed to
tence for genetic transformation. Bacillus spp. interact with each other bringing about DNA
often produce small peptide antibiotics dur- bending (ROGGIANOand DUBNAU,1993;
ing stationary phase, either through the ac- NAKANOand ZUBER, 1993) (Fig. 8). This
tion of large peptide synthetase enzymes or multiple interaction stimulates srfA transcrip-
by the posttranslational modification of ribo- tion by RNA polymerase bound to the pro-
somally synthesized propeptides. The non-ri- moter (probably via the major (+ factor, &;
bosomally synthesized peptides include anti- NAKANOet al., 1991). Phosphorylation of re-
biotics such as bacitracin (B. licheniformis), sponse regulators such as ComA typically oc-
gramicidin and gramicidin S (B. brevis), poly- curs when appropriate environmental condi-
myxins (B. polymyxa), tyrocidine (B. brevis tions are sensed by a membrane-located his-
ATCC 8185), and surfactin (B. subfilis ATCC tidine protein kinase, and ComA is no excep-
21 332). Understanding of the regulation of tion (WEINRAUCHet al., 1990). Its partner
production of these antibiotics was reviewed kinase, ComP, appears to behave as a “quo-
by MARAHIEL et al. (1993) and has recently rum sensor”, being activated by binding of an
been extended in the case of surfactin which extracellular competence “pheromone” pro-
is the main subject of this section. duced by the B. subtilis cells themselves. This
Surfactin production requires three large pheromone, a 9-10 amino acid oligopeptide
peptide synthetases encoded by three of the with a lipophilic modification, is formed from
four consecutive genes in the 27 kb srfA ope- the C-terminus of the 55 amino acid product
ron (COSMINAet al., 1993). The classical ge- of the comX locus, in a process dependent on
netic strain 168 of B. subtilis does not make the product of the immediately adjacent up-
surfactin, even though it contains an intact stream gene, comQ (come and comX are
srfA operon. This deficiency is due to a themselves immediately upstream of the
frameshift mutation (the sfp” allele) of the genes encoding the ComP-ComA proteins in-
adjacent sfp gene (NAKANOet al., 1992) volved in sensing and responding to the
which may be involved in the export of small ComX pheromone; Fig. 8).
peptides (GROSSMAN et al., 1993). Remarkably, a segment of the srfA operon
Transcription of the srfA operon is the ma- plays an additional role, as part of a regulato-
jor point of regulation of surfactin produc- ry cascade leading to competence (D’SOUZA
tion, and it has turned out to involve a com- et al., 1993) (Fig. 8). Surfactin itself is not in-
plex array of controlling elements including volved in this cascade, since srfA mutations
autoregulation, phosphorylation cascades, ex- eliminating production do not all eliminate
tracellular signaliig, and interplay with the competence; and indeed, B. subtilis 168 which
regulation of different developmental path- usually makes no surfactin is used as a labora-
Fig. 8. Signal cascades and networks leading to surfactin synthesis and other stationary-phase processes in
Bacillus subtilis. The onset of surfactin synthesis is transcriptionally dependent on the phosphorylated form
of the ComA response regulator. ComA phosphorylation is carried out partly by a cognate histidine pro-
tein kinase, ComP, which autophosphorylates a conserved histidine residue (H) when it interacts with an
extracellular oligopeptide pheromone, ComX, which is a processed and modified form of the comX gene
product. This pheromone and a second one, CSF, accumulate to effective concentrations only under con-
ditions of high cell density. CSF also stimulates ComA phosphorylation by an unknown route involving the
SpoOK transporter complex. Production of CSF is autoregulated via the sporulation phosphorelay which is
activated partially by S p d K and partially by the histidine protein kinase KinB and another protein kinase,
KinA, in response to unknown signals. The phosphorelay passes a phosphoryl group via SpoOF and S p d B
to SpoOA, a response regulator protein that is both an activator of sporulation genes and a repressor of
abrB, itself encoding a repressor of some sporulation genes and of CSF synthesis. Thus, activation of the
phosphorelay leads to relief of CSF synthesis from repression (the figure also shows that the activation of
surfactin synthesis is accompanied by the activation of competence; see text for further details).
tory organism because of its competence. frame, comS, within srfA encodes a critical
Studies of constructed srfA partial deletions regulatory element for competence (D’Sou-
had shown that regulation of competence re- ZA et al., 1994) (Fig. 8). comS is translated in
quires only the DNA region encoding the val- a different reading frame from the sequence
ine-activating domain of one of the peptide encoding the valine-activating domain of
synthetases (VANSINDEREN et al., 1993), and srfA,giving a protein which appears to acti-
that aminoacylation activity itself is not vate competence by interfering with a pro-
needed (D’SOUZAet al., 1993). This unusual tein-mediated inhibition of ComK, a critical
situation has been clarified by the recent dis- activator of the late competence genes comC,
covery that a distinct small open reading comG, and comDE (MSADEKet al., 1994;
SIGNAL
Transcriptional activation o f
target promoters
Fig. 9. Signal transduction in a bacterial two-component regulatory system. The conserved core elements of
such systems are a protein kinase module capable of autophosphorylation at a conserved histidine residue
(H), and a target response regulator module to which the activated phosphoryl group is transferred via an
acidic pocket consisting of conserved aspartate residues, one of which (Asp5’) becomes phosphorylated.
These modules are usually on separate proteins though they can also be found within single proteins
(ALEXand SIMON,1994). In the example shown, the histidine kinase module forms the C-terminal, cyto-
plasmic domain of a protein whose N-terminal domain is a surface receptor for an unspecified extracellular
signal. Binding of the ligand activates the kinase which then phosphorylates the N-terminal domain of a
cytoplasmic response regulator. This activates a C-terminal DNA-binding domain of the regulator. In this
example, binding to a specific DNA target sequence leads to transcriptional activation by contact with
RNA polymerase at an adjacent promoter. The DNA-bindingkranscriptional activation modules of such
response regulators themselves form subfamilies, some of which are found in regulatory proteins that are
not part of two-component systems (e.g., LuxR; Sect. 2.2).
KONG and DUBNAU,1994). It is not known and biosynthetic origin of CSF are not
why surfactin production has evolved this known, but its production is regulated by a
connection with competence. route different from that of ComX phero-
Activation of ComA can be detected, albeit mone: it is repressed by the AbrB protein, a
at a reduced level, in comP mutants that lack repressor of several stationary-phase path-
the cognate protein kinase. This is accounted ways including sporulation. Relief of these
for by an as yet incompletely defined pathway pathways from AbrB-mediated repression at
from a second membrane-bound signal recep- the onset of stationary phase is usually
tor. This receptor, encoded by the complex brought about by phosphorylation of SpoOA,
SPOOK locus, is an aggregate of several pro- the central regulator of transition stage genes,
teins and resembles various oligopeptide via a phosphorelay system that is best known
transport systems (hence the use of the acro- for its role in initiating sporulation (HOCH,
nym Opp to describe the SpoOK proteins). It 1993). The Opp complex is also one of several
is a member of the ATP-binding cassette routes by which largely undefined physiologi-
(ABC) family of transporters (HIGGINS, cal signals lead to transmission of phosphoryl
1992). The Opp complex responds to a sec- groups to SpoOA via the SpoOF and S p d B
ond oligopeptide pheromone, CSF (compe- proteins (RUDNERet al., 1991; HOCH, 1993).
tence stimulating factor), quite distinct from SpoOA has an NH2-terminal portion homolo-
the ComX pheromone. The precise structure gous to the phosphorylated domain of re-
sponse regulators (HOCH, 1993; Fig. 8). The duction with stationary phase is brought
-
SpoOA P protein represses ubrB, thereby about by an almost bewildering variety of
mechanisms. The initial switches typically re-
derepressing CSF production in an autoregu-
latory loop that biases further development in sult from a change in a constitutively synthe-
the direction of sporulation, surfactin pro- sized regulatory protein, brought about by
duction, and competence development. either intracellular or extracellular chemical
-
SpoOA P also directly activates some genes, signals (e.g., OHHL or oligopeptide phero-
mones). Extracellular signals may be mem-
notably some involved in sporulation. This
-
constellation of SpoOA P activities explains brane-diffusible and recognized by cytoplas-
mic binding proteins, or membrane-non-dif-
the pleiotropic sporulation, antibiotic produc-
tion, and competence deficiencies of spoOA fusible and recognized by membrane-bound
mutants, and of SPOOK,spoOB, and SPOOFmu- proteins able to initiate an intracellular phos-
tants deficient in the phosphorelay. The full phorylation cascade. In all cases described so
expression of spoOA, and hence the expres- far, the end result of transmission of the ini-
sion of srfA, also depends on a minor RNA tial signal is activation of transcription of
polymerase 5 factor, &', which directs tran- structural genes for antibiotic biosynthesis. In
scription of spoOA from an alternative, sta- the best characterized cases, this activation
tionary phase-specific promoter. The regula- may arise either because of increased levels of
tion of &' is itself highly complex, with in- a minor form of RNA polymerase containing
creased &' activity during entry into stationa- a sigma factor such as 2 of E. coli that can
ry phase involving both transcriptional and recognize the promoters of the structural
posttranscriptional control (HEALY et al., genes, or by the posttranscriptional activation
1991). of transcription factors needed for the major
The stimuli for activation of the SpoO phos- form of RNA polymerase to initiate tran-
phorelay are not yet well understood. The scription at the relevant promoters (e.g.,
multiple steps possibly provide the means to phosphorylation of ComA in B. subtilis or
integrate sensory input from diverse sources OHHL-mediated conformational change of a
(HOCH, 1993). For example, SpoOB is speci- LuxR homolog in carbapenem-producing E.
fied by the first gene in an operon that also curotovoru). It is abundantly clear from stud-
encodes an essential GTP-binding protein, ies in E. coli and B. subtilis that antibiotic
Obg, and Obg might cause phosphorylation production requires the integration of diverse
of SpoOB in response to intracellular informa- information through complex regulatory net-
tion about the cell cycle or the levels of GTP, works.
which are critical in signaling the onset of
sporulation (HOCH,1993).
In a further twist of this increasingly com-
plex system, srfA expression can be in-
fluenced by another regulatory protein, 3 Regulation of Antibiotic
DegU, which belongs to the same subfamily
of response regulators as ComA (HENNERet Production in
al., 1988; WEINRAUCHet al., 1989): hyper-
phosphorylated mutant forms of DegU re-
Streptomycetes and their
press srfA by an unknown mechanism (HAHN
and DUBNAU,1991).
Relatives
3.1 Introduction to the Organisms
2.4 What Has Been Learned from
Studies of Antibiotic Production in Streptomyces spp. are the most versatile
Unicellular Bacteria? and commercially important producers of an-
tibiotics and have traditionally been central to
The examples discussed so far have rescreening programs for new chemotherapeut-
vealed that the association of antibiotic pro- ic compounds. They are morphologically and
phylogenetically distinct from the bacteria netic linkage map of the chromosome (KIE-
dealt with in the preceding sections. Their SER et al., 1992), and the fact that the strain
branching, mycelial growth habit is probably produces at least four antibiotics, production
an adaptation that allows them to grow effi- of one of which (methylenomycin A) is plas-
ciently on the surface of insoluble organic de- mid-specified (Sect. 3.3.4.5); two others are
bris in soil. Dispersal is by means of spores, conveniently pigmented (actinorhodin is blue
typically borne on aerial hyphae, and in the at high pH and red at low pH while undecyl-
laboratory mature Strepfomyces colonies have prodigiosin is red). Moreover, diverse aspects
a furry appearance because of this aerial my- of the physiology and developmental biology
celium. The Acfinomycefulesto which strepto- of S. coelicolor have been studied providing
mycetes belong form a division of the gram- important information to relate to studies of
positive bacteria characterized by a high pro- secondary metabolism. Much of the informa-
portion of G + C in their DNA (on average tion reviewed below is drawn from S. griseus
74mol% G + C ) . B. subtilis, on the other and S. coelicolor.
hand, belongs to the division with low G + C
content. These divisions resulted from a very
ancient evolutionary separation. Striking pro- 3.2 General Physiological Aspects
gress has been made in the last decade in the of the Regulation of Antibiotic
molecular analysis of antibiotic production by
streptomycetes, notably in two model species, Production in Streptomyces
Streptomyces griseus, the producer of strepto-
mycin, and Strepfomyces coelicolor A3(2), ge- Like the organisms already described,
netically the most-studied strain (reviewed by streptomycetes grown in liquid media gener-
CHATERand HOPWOOD,1993, and HOP- ally produce antibiotics during stationary
WOOD et al., 1995). Studies in S. coelicolor phase or at low growth rates. (In the latter
have been helped by the extensive availability case, this may reflect production by cells in-
of natural and artificial genetic systems, the side mycelial pellets that may be nutritionally
l
development of a combined physical and ge- limited and that have, therefore, entered sta-
1.2 10. 0.10

60
0.08
?J
40 E
0.06 &
E
a
Y
0.04 0
z
20
2. 0.02
0 0. 0 0
Fig. 10. Growth phase-dependent production of candicidin by S. griseus in liquid culture. A Candicidin;
A dry weight; 0 glucose; 0 DNA. Redrawn from MART~N and MCDANIEL(1975).
ot Spores
Fig. 11. Growth phase-dependent pro-

duction of oleandomycin by surface-
grown cultures of s. antibioticus. 0 dry
weight; 0 dry weight minus glycogen;
8 24 40 56 72 88 104 A oleandomycin. Redrawn from MEN-
Time [h] DEZ et al. (1985).
tionary phase.) For example, candicidin was In Sects. 3.2.1-3.2.5 we review current un-
produced by S. griseus in liquid culture only derstanding of the physiological factors that
after net DNA synthesis had ceased (Fig. 10; might play a general role in triggering the on-
MARTIN and MCDANIEL, 1975). Biomass set of antibiotic synthesis, before moving to
continued to increase slowly, presumably consider the genetics of antibiotic production
through the accumulation of storage com- in Sect. 3.3.
pounds such as glycogen (BRANAet al., 1986)
or triacyl glycerol (OLUKOSHI and PACKTER,
1994).
Stationary phase antibiotic production 3.2.1 Metabolite Interference with
could be viewed as a physiological abnormal- Antibiotic Production in
ity that results from growing soil organisms in
submerged culture, particularly since most Streptomycetes
streptomycetes do not differentiate normally
in liquid culture. However, antibiotic produc- Growth is most rapid when readily utilized
tion on solid media is also growth phase-de- carbon, nitrogen, and phosphate sources are
pendent; thus, production of oleandomycin abundant, in part because of repression or in-
by Streptomyces antibioticus on agar began hibition by these metabolites of most inessen-
only after growth had ceased (Fig. 11; M ~ N - tial processes. In principle, antibiotic produc-
DEZ et al., 1985). Interestingly, oleandomycin tion could be subject to the same regulatory
production appeared to be confined largely to controls, and its occurrence during stationary
the substrate mycelium since synthesis of the phase might reflect relief from metabolite in-
antibiotic was apparently completed before terference after nutrient depletion. While
aerial hyphae appeared. In other cases, anti- there are many examples of metabolite inter-
biotic production coincides approximately ference with antibiotic biosynthesis, especial-
with the onset of morphological differentia- ly by glucose, ammonium, and phosphate
tion, and the isolation from both S. coelicolor (Tab. l),the underlying mechanisms are gen-
and S. griseus of bld mutants defective in both erally not known. In a few cases, repression of
processes suggests at least some common ele- transcription appears to be involved, as in
ments of genetic control (Sect. 3.3.2.6). glucose repression of phenoxazinone synth-
Tab. 1. Metabolites that Interfere with Antibiotic Production in Streptomycetes
Interfering metabolite Antibiotic
Carbon sources:
Citrate Novobiocin
Glucose Actinomycin, chloramphenicol, chlortetracycline, kanamycin, mitomy-
cin, neomycin, oleandomycin, puromycin, siomycin, streptomycin, tetra-
cycline, tylosin
Glycerol Actinomycin, cephamycin
Nitrogen sources:
Ammonium ions Actinorhodin, chloramphenicol, leucomycin, streptomycin, streptothri-
cin, tetracycline, tylosin, undecylprodigiosin
L-Glu, L-Ala, L-Phe, D-Val Actinomycin
L-Tyr, L-Phe, ~ - T r p PABA
, Candicidin
Inorganic phosphate:
Actinorhodin, candicidin, cephamycin, nanaomycin, nourseothricin,
streptomycin, tetracycline, tylosin, undecylprodigiosin, vancomycin
Information from DEMAIN

et al. (1983). DEMAIN
(1992), and HOBBSet al. (1992).
Fig. 12. Effect of glucose (Gluc) on activity (left) and mRNA levels (right) of phenoxazinone synthase
(PHS) in S. antibioticus (Gal, galactose). Redrawn from JONES(1985).
ase, the final enzyme in actinomycin biosyn- ponent(s) during culture have rarely been re-
thesis in S. antibioticus (JONES,1985; Fig. 12). ported, so the extent of the contribution of
Phosphate also appears to repress transcrip- metabolite interference to growth phase-de-
tion of genes required for candicidin biosyn- pendent antibiotic production is difficult to
thesis by S. griseus (Fig. 13; ASTURIAS et al., assess. Furthermore, although metabolite in-
1990) and of those for actinorhodin produc- terference might account for some examples
tion in S. coelicolor (HOBBSet al., 1992). of growth phase dependence, there appears
Unfortunately, the levels of nutrients and to be no general pattern. Even within one
the identification of the growth-limiting com- species, the production of different secondary
74 2 Regulation of Bacterial Antibiotic Produc:tion
100
3.2.2 Antibiotic Production and
Imbalances in Metabolism
E 75
f An alternative or additional possibility is
% 50 that antibiotic production is triggered by an
3 imbalance in metabolism. Undecylprodigio-
25 sin, the major component of the red antibiotic
of S. coelicolor (TSAOet al., 1985), is derived
0 partly from proline (WASSERMAN et al., 1974;
GERBERet al., 1978). To determine whether
the amino acid incorporated into the antibiot-
ic was synthesized internally or taken up from
outside, HOODet al. (1992) isolated and char-
acterized put mutants deficient in proline
transport, which turned out to be defective
also in proline catabolism. Since proline bio-
synthesis appears to be constitutive in S. coe-
licolor, such mutants might be expected to ac-
cumulate proline intracellularly. While this
has not been determined experimentally, the
mutants markedly overproduce the red anti-
biotic suggesting that undecylprodigiosin
serves as a sink for excess proline. The need
to remove surplus proline might reflect the
role that it plays as an osmoregulant in other
bacteria (KILLHAMand FIRESTONE,1984a,
b). It will be interesting to see whether the
Time [h] put mutants produce undecylprodigiosin ear-
lier than the parental strain, as predicted by
Fig. 13. Phosphate repression of candicidin produc- this hypothesis, and how this is mediated at
tion in S. griseus. pabS encodes PABA synthase, the level of gene regulation.
and plays an early role in candicidin production. An imbalance in carbon metabolism may
SPG, soya peptone-glucose medium. 0 SPG; W be responsible for triggering the production
SPG+7.5 mM phosphate. Redrawn from ASTU-
RIAS et al. (1990).
of methylenomycin A, another S. coelicolor
antibiotic. Methylenomycin biosynthesis be-
gan at the same time as a rapid drop in the
pH of a S. coelicolor fermentation (Fig. 14)
metabolites appears to be triggered by the de- caused by the efflux of a-ketoglutarate and
pletion of different nutrients. In Sfrepfomyces pyruvate from the mycelium (HOBBSet al.,
cuttleyu, the production of melanin, cephamy- 1992). Indeed, acid shock alone caused tran-
cin C, and thienamycin in batch culture oc- sient methylenomycin biosynthesis, suggest-
curred on depletion of glucose, ammonia, and ing that this might be a stress response to the
phosphate, respectively. In a chemostat, ceph- change in pH that presumably reflects the im-
amycin C production occurred at low growth balance in carbon metabolism (G. HOBBSand
rates that could be brought about by limiting S. G. OLIVER,personal communication).
for carbon, nitrogen, or phosphate, whereas
the production of thienamycin required a low
growth rate specifically associated with phos-
phate deficiency, consistent with the observa-
tions made in batch culture (LILLEYet al.,
1981).
300 -
4
3
200 -
T
g8 *
i
a
z
Y
n
100- 5 1
0 -
Fig. 14a-c. Methylenomycin pro- 0

250
duction during growth of S. coeli-
color A3(2) on minimal medium
with alanine as nitrogen source.
2
200
a Growth and change in extracel-
lular pH. I3 DNA; - Ln% COz Y
(logarithm of the percentage car- 150

3
'
bon dioxide concentration in the 8
fermenter exhaust gases); - - - .-0
pH. b Glucose assimilation and E
100
methylenomycin production.
A glucose; 0 methylenomycin. 50
c Production of pyruvate and
a-ketoglutarate.
0 a-ketoglutarate; A pyruvate. 0
Redrawn from HOBBSet al.
(1992). Time [h]
3.2.3 The Possible Role of Growth earlier (Sect. 2.1), ppGpp is believed by many
Rate and ppGpp in Antibiotic to play a central roie &-the growth rate con--
trol of gene expression in E. coli (SARUBBI et
Production al.. 1988: HERNANDEZ and BREMER.1990.
1993; SCHREIBER et al., 1991). A possible role
Most of the published data are consistent for ppGpp in triggering the onset of antibiotic
with a role for growth rate, or the cessation of production was first addressed in S. griseus by
growth, in determining the onset of antibiotic AN and VINING(1978). They found that
production in streptomycetes. As discussed streptomycin production occurred only after
c
0
gs
Fig. 15a-d. ppGpp and antibiotic

production in S. coelicolor A3(2).
a Growth (O),antibiotic, and
ppGpp (A) production in S. coe-
licolor A3(2). The shaded boxes
b labeled ACT or RED denote the
presence of actinorhodin or un-
decylprodigiosin, respectively. tD,
doubling time. b Transcription of
actll-ORF4 was monitored by S1
nuclease protection studies using
RNA isolated at the times indi-
cated from the culture shown in
(a). “Probe”, the position of the
full-length probe. EXP and
STAT, exponential and stationa-
ry phases, respectively, and the
shaded area between them indi-
cates the transition phase. SM,
size marker. Similar results were
observed for redD transcription.
c Growth curve of S. coelicolor
A3(2) with (A) and without (0)
C
nutritional shiftdown. The
shaded boxes labeled (0RED)
or (0ACT) denote the presence
of undecylprodigiosin or actinor-
hodin, respectively in the control
culture. SD indicates when the
cultures were subjected to shift-
down, the shaded box labeled (A
ACT) denotes the presence of
actinorhodin in the culture sub-
jected to shiftdown, and (0)the
ppGpp level in the shifted cul-
ture. d Transcription of actll-
ORF4 after nutritional shift-
down. S1 nuclease protection
analysis of actll-ORF4 transcripts
in RNA isolated at the times in-
0 10 20 30 dicated from the culture shown
Time [h] in (c).
d and HOLT et al. (1992) noted a peak of
Hours after shiftdown ppGpp accumulation that coincided with
transcription of the pathway-specific activator
gene, brpA, before production of bialaphos
by Streptomyces hygroscopicus. However,
BASCARAN et al. (1991) concluded that there
was no obligate relationship between ppGpp
and antibiotic production in Streptomyces cla-
vuligerus. Production of cephalosporins
(mainly cephamycin C) occurred during a
phase of slow exponential growth (doubling
Fig. 1Sd times of ca. 7 h and 18 h in rich and minimal
media, respectively) and increased in station-
ary phase, while ppGpp levels remained con-
a drop in ppGpp levels, so it seemed that stant from the beginning of growth. relC-like
ppGpp did not stimulate the initiation of anti- mutants accumulated lower levels of ppGpp
biotic synthesis. Subsequently, ppGpp was than the wild-type strain after nutritional
detected in several Streptomyces spp. (HA- shiftdown (varying between 8% and 85% of
MAGISHI et al., 1980; SIMUTH et al., 1979; HA- the wild-type levels), but there was no simple
MAGISHI et al., 1981; NISHINOand MURAO, correlation between the levels of ppGpp and
1981; STASTNAand MIKULIK, 1981), and po- cephalosporin production (some produced
sitive correlations were observed between more, and others less, cephalosporin than the
ppGpp and antibiotic biosynthesis in Strepto- wild-type strain).
myces aureofaciens (SIMUTHet al., 1979) and Recently, the relationship between ppGpp
Streptomyces galifaeus (HAMAGISHIet al., synthesis and the production of undecylprodi-
1981). OCHI(1986) found that an accumula- giosin and actinorhodin has been examined
tion of ppGpp following nutritional shiftdown in S. coelicolor (Fig. 15). Transcription of
was accompanied by an 8-fold increase in for- pathway-specific activator genes for each anti-
mycin production by Streptomyces lavendulae biotic (redD for undecylprodigiosin and
MA406-A-1; moreover, a mutant deficient in actll-ORF4 for actinorhodin: see Sect. 3.3.4.1)
ppGpp accumulation after amino acid deple- conspicuously increased as the culture was
tion (the “relaxed” phenotype) was impaired in transition between growth and stationary
in formycin production. Similar relC mutants phase, coinciding with the onset of ppGpp
were isolated from S. antibioticus 3720 (OCHI, production; this was followed by transcription
1987), Streptomyces griseojlavus 1805 (OCHI, of representative red and act biosynthetic
1988), S. griseus 13189 (OCHI, 1990a), and S. structural genes and production of the anti-
coeficolor (OCHI, 1990b). The mutants accu- biotics (STRAUCHet al., 1991; TAKANO et al.,
mulated low levels of ppGpp after nutritional 1992). In contrast, transcription of a typical
shiftdown (on average ca. 15% of wild-type rRNA gene set (rrnD) decreased markedly
levels) and were deficient in antibiotic pro- (E. TAKANO and M. J. BIBB,unpublished re-
duction, leading to the idea that ppGpp plays sults). Thus a positive correlation was ob-
a central role in triggering antibiotic produc- served between ppGpp accumulation and
tion (OCHI, 1990b). A streptomycin-produc- transcription of actll-ORF4 and redD at the
ing pseudorevertant of a relC mutant of S. gri- end of exponential growth. ppGpp synthesis
seus remained defective in (p)ppGpp accumu- could also be induced by a nutritional shift-
lation. This suppressor mutation was pro- down during rapid growth, and again resulted
posed to identify a gene activated by ppGpp in a rapid and marked decrease in rrnD tran-
and involved in triggering streptomycin pro- scription (STRAUCHet al., 1991). Transcrip-
duction. Consistent with OCHI’S proposal, tion of actZZ-ORF4 was detected within
KELLYet al. (1991) found reduced levels of 30min of shiftdown (Fig. 15) and transcrip-
actinomycin biosynthetic enzymes and tion of the act biosynthetic genes followed
mRNA in a relC mutant of S. antibioticus, 30 min later (TAKANO and BIBB,1994). How-
ever, there was no immediate stimulation of cell population density or they might be syn-
redD transcription (TAKANO and BIBB, thesized in response to physiological condi-
1994), suggesting that if ppGpp does play a tions under which antibiotic production
role in triggering the onset of antibiotic pro- would be favorable to the organism.
duction, it is not always sufficient: the activa- yButyrolactone compounds whose acyl-
tion of at least some biosynthetic pathways ated lactone structures somewhat resemble
may depend on additional factors. Such varia- the homoserine lactones found in gram-nega-
tions in regulatory responses of different sets tive bacteria (see Fig. 6) have been detected
of pathway genes in Streptomyces spp. are in many streptomycetes, and have been impli-
perhaps not surprising in view of comparable cated in antibiotic production and morpho-
variations observed in E. coli and Bacillus logical differentiation in several species (Fig.
spp. (see Sects. 2.1 and 2.3). 16). The most intensively studied example is
Correlations between ppGpp synthesis and A-factor (2-isocapryloyl-3R-hydroxymethyl-
the onset of antibiotic production, where they ybutyrolactone) which is required for strep-
occur, do not establish a causal relationship, tomycin production and morphological differ-
and although the relC mutants isolated by entiation in s. griseus (Sect. 3.3.3). Five re-
OCHI (1986, 1987, 1988, 1990a, b) showed a lated compounds, virginiae butanolides A-E
marked reduction in antibiotic production, (Fig. 16a; VB-A-E), are inducers of virginia-
they also grow at about half the maximal rate mycin production by Streptomyces virginiae
of the parental strain, presumably reflecting (YAMADAet al., 1987), and compounds with
impaired protein synthesis. Thus it is difficult the same biological activity are found in other
to assess whether the effect on antibiotic pro- streptomycetes (OHASHIet al., 1989). Such
duction is a direct consequence of reduced butyrolactones probably diffuse readily
levels of ppGpp or an indirect effect of the through membranes, so extracellular and in-
refC mutation on protein synthesis. It is inter- tracellular levels are likely to be the same:
esting to note that mutants of E. coli unable hence, just as with LuxR and OHHL, moni-
to make ppGpp (ppGpp" mutants) do not toring of extracellular concentrations is done
show the marked reduction in growth rate by cytoplasmically located binding proteins
characteristic of relC mutants (XIAO et al., with very high affinity and specificity for their
1991). Attempts to establish a role for ppGpp ligands (KD ca. 10-9M). To illustrate this
in triggering the onset of antibiotic produc- specificity, virginiae butanolide C shows no
tion in Streptomyces spp. will require the iso- biological activity in vivo with A-factor-defi-
lation of the corresponding ppGpp" mutants cient mutants of S. griseus and the A-factor
and possibly the ability to regulate ppGpp receptor does not bind it in vitro (MIYAKEet
levels in the absence of the physiological trau- al., 1989). However, ligand specificity can
ma associated with nutritional shiftdown, as vary between strains, and A-factor analogs
was done by GENTRYet al. (1993) in their with acyl chains of different lengths, or with a
analysis of the effects of ppGpp on rpoS ex- hydroxyl group rather than a carbonyl group
pression in E. coli (see Sect. 2.1). at position 6, had some activity in an anthra-
cycline-producing S. griseus strain that re-
quires A-factor for sporulation and antibiotic
3.2.4 Antibiotic Production and production (GRAFE et al., 1982, 1983). The
the Accumulation of Small receptor proteins have not been characterized
- an earlier report that a cytoplasmic binding
Diffusible Signaling Compounds protein for the S. virginiae factor VB-C
showed significant sequence homology to
The synthesis of threshold levels of small NusG (OKAMOTO et al., 1992), an E. coli pro-
diffusible signaling molecules appears to play tein believed to play a role in transcriptional
a central role in triggering production of at antitermination, appears to have been ill-
least some antibiotics in some streptomycetes founded (PUTTIKHUNT et al., 1993). How A-
(HORINOUCHI and BEPPU,1992, 1995). Such factor binding protein influences antibiotic
molecules might act simply as indicators of production is addressed in Sect. 3.3.3.
S. coelicolor does not make A-factor. How- 3.3 Genetics of Antibiotic

ever, it does make a series of structurally very
similar compounds (ANISOVAet al., 1984;
Production
EFREMENKOVA et al., 1985; Fig. 16b), and at
least some of these can apparently stimulate 3.3.1 Organization of Antibiotic
streptomycin production by S. griseus mu-
tants deficient in A-factor biosynthesis
Biosynthetic Genes and Clusters
(HARAet al., 1983). afsA mutants of S. coeli-
color that had lost the ability to cross-stimu- In order to clarify discussion of the genetic
late the S. griseus A-factor-deficient mutants regulation of antibiotic biosynthesis, we first
were unaffected in antibiotic production and briefly summarize the organization of the bio-
morphological differentiation. However, it synthetic genes themselves. Genes specifically
may be that the afsA mutants lack only some involved in the production of a particular an-
of the A-factor-like compounds made by the tibiotic are invariably found clustered togeth-
parental strain, so y-butyrolactones might er, and only one set, for methylenomycin pro-
conceivably play a role in triggering antibiotic duction (Sect. 3.3.4.9, is known to be plas-
production in S. coelicolor. Alternatively, an- mid-located rather than chromosomal. Thus,
tibiotic production in S. coelicolor might per- all of the act and red genes of S. coelicolor oc-
haps have lost a requirement for such A-fac- cur in chromosomal segments of ca. 23 kb and
tor-like compounds, a situation reminiscent of 35 kb, respectively. Transfer of these seg-
the A-factor receptor-deficient mutants of S. ments to Streptomyces parvulus, a host not
griseus (Sect. 3.3.3); the failure to detect any known to make any structurally similar com-
A-factor-binding protein in cell extracts of S. pounds, caused synthesis of the antibiotics
coelicolor is consistent with this hypothesis. (MALPARTIDA and HOPWOOD,1984; MAL-
PARTIDA et al., 1990). Similarly, expression in
surrogate hosts has led to the demonstration
3.2.5 Summary that the entire biosynthetic pathways for pro-
duction of tetracenomycin by Streptomyces
The various physiological factors that in- glaucescens and puromycin by Streptomyces
fluence the onset of antibiotic production in alboniger are located in DNA segments of
streptomycetes do not fit into a simple unify- 12.6 kb and 15 kb, respectively (DECKERand
ing model. However, it seems reasonable to HUTCHINSON, 1993; LACALLEet al., 1992).
propose an overall regulatory influence of The biosynthetic genes generally seem to be
growth rate with superimposed pathway-spe- organized into several transcription units of
cific regulatory effects influencing the produc- varying complexity. Pathway-specific regula-
tion of individual antibiotics. Such effects tory genes have been identified in several of
could include responsiveness to catabolite re- the clusters (Sect. 3.4), although there are ex-
pression or inhibition, imbalances in metabol- ceptions that seem so far to lack such genes,
ism, and environmental signals and stresses. It e.g., the tetracenomycin pathway and the very
is clear that low molecular-weight effectors, large biosynthetic clusters (ca. 45 kb and
like A-factor, play essential roles for some an- 95 kb, respectively) for the macrolides ery-
tibiotics. Whether these are produced as a thromycin (made by Saccharopolyspora ery-
consequence of a reduction in growth rate, or thraea; DONADIOet al., 1993, and references
in response to some extrinsic factor, or via therein) and avermectin (produced by Strep-
some autoregulatory circuit such as that de- tomyces avermitilis; MACNEIL et al., 1992).
scribed for LuxR and OHHL remains to be The biosynthetic clusters usually also contain
determined, as does the potential role of one or more genes that confer immunity to
ppGpp as an intracellular signaling mole- the antibiotic, which vary considerably in
cule. their mode of action. For example, resistance
to erythromycin in S. erythraea is accom-
plished by a 23s rRNA methylase that modif-
ies the target of the antibiotic by N6 dimethy-
a
A-factor
Q o0 0 S. griseus
OH
0 OH S. bikiniensis and
S. cyaneofuscatus
VB-D
/ 1 " 4 o*AoH
H Factorl
n
I
w
OH
S. viridochromogenes Fig. 16a, b. y-Butyrolactones

produced by streptomycetes.
a A-factor-like compounds
OH IM-2 in streptomycetes. b A-fac-
tor-like compounds from S.
0 OH coelicolor. Structures taken
0 from HODGSON(1992) and
Virginiae butanolides OH
EFREMENKOVA et al.
from S. virginiae Streptomyces sp. FRI-5 (1985).
lation of a specific adenine residue (THOMP- ORF2 (FERNANDEZ-MORENO et al., 1991)

SON et al., 1982). Different types of efflux sys- genes of S. gluucescens and S. coelicolor are
tems appear to confer resistance to different presumed to be driven by transmembrane
antibiotics; thus srmB, one of four spiramycin electrochemical gradients. In some cases, ex-
resistance genes in Streptomyces umbofuciens, pression of the resistance gene appears to be
appears to encode an ATP-dependent trans- induced by the corresponding antibiotic (Sect.
port system for this macrolide antibiotic 3.3.5).
(SCHONERet al., 1992), whereas the efflux
systems for tetracenomycin, methylenomycin,
and probably actinorhodin encoded by the
tcmA (GUILFOILEand HUTCHINSON, 1992a),
mmr (NEAL and CHATER,1987) and uctZZ-
OH 0 OH
Q-2a 0aA&-2b
G-2c
0 OH 0 OH
0Q2-d
Fig. 16b. OH
3.3.2 Genes that Pleiotropically 3.3.2.1 afsB, afsR, and afsK - A

Affect Antibiotic Production in Role for Protein Phosphorylation
Streptornyces coelicolor: in Triggering the Onset of
Introduction and Overview Antibiotic Production?
Many genes have been identified that ufsB mutants resemble ufsA mutants (Sect.
pleiotropically affect antibiotic production in 3.2.4) in that they cannot induce streptomycin
S. coelicolor, and several of these are likely to production and sporulation of A-factor-defi-
play a global role in regulating the onset (and cient mutants of S. griseus grown near them.
perhaps maintenance) of antibiotic synthesis. However, unlike ufsA mutants, they are de-
Mutants in about half of these pleiotropic fective in actinorhodin and undecylprodigio-
genes also show deficiencies in morphological sin synthesis (HARAet al., 1983). (Production
differentiation. These "bld" mutants are dis- of the other two antibiotics known to be
cussed in Sect. 3.3.2.6. Some of the work on made by S. coelicolor, methylenomycin and a
regulatory genes has involved use of a very calcium-dependent antibiotic (CDA), appears
close relative of S. coelicolor, S. lividuns 66, to be normal; ADAMIDISand CHAMPNESS,
which also contains act and red gene sets, but 1992.) Northern analysis failed to detect tran-
generally expresses them rather poorly. S. Ziv- scripts corresponding to uctl, uctZZ (including
iduns is a slightly more convenient recipient actZZ-ORF4, the pathway-specific activator
strain for transformation than S. coelicolor. gene), uctZZZ, and uctVZ in the ufsB mutant
BH5 (HORINOUCHI et al., 1989a). Attempts
to clone ufsB, which was presumed to regul-
ate production of the pigmented antibiotics
and the A-factor-like compounds, have so far
failed. However, they yielded ufsR, which was
obtained by screening of a library of S. coeli-
color DNA for overproduction of undecyl- inhibitors of eukaryotic protein kinases

prodigiosin and actinorhodin in an ufsB-like (HONGet al., 1993). Disruption of ufsK, like
mutant of S. lividuns (HH21) that could not that of ufsR, resulted in reduced levels of acti-
cross-feed an A-factor-deficient mutant of S. norhodin production, though the reason for
griseus (HORINOUCHI et al., 1983; STEINand this is not clear, since AfsR could still under-
COHEN,1989). At high copy number, a DNA go phosphorylation at serine and threonine
fragment containing ufsR and ufsR2 (see be- residues in the afsK mutant (MATSUMOTO et
low) suppresses the afsB phenotype in S. coe- al., 1994). Clearly some other protein kinase
licolor (HORINOUCHI et al., 1983) and stimu- can phosphorylate AfsR, a conclusion consis-
lates transcription of act genes in S. coelicolor tent with emerging evidence of multiple pro-
and S. lividuns (HORINOUCHI et al., 1989a). tein kinases in S. coelicolor (WATERSet al.,
ufsR encodes a protein of 933 amino acids 1994). Interestingly, the N-terminal region of
that contains putative DNA-binding helix- AfsR resembles the pathway-specific regula-
turn-helix motifs towards the C-terminus and tory proteins RedD, ActII-ORF4, and DnrI
potential ATP-binding sites towards the mid- (Sect. 3.3.4.1). Unlike redD and actll-ORF4,
dle of the protein (HORINOUCHI et al., 1986, whose transcription increases dramatically
1990); site-directed mutagenesis of the latter during transition phase, transcription of ufsR
resulted in a 4-fold reduction (though not an occurs throughout exponential growth and
elimination) of stimulatory activity in S. livi- declines slowly on entry into stationary phase
duns (HORINOUCHI et al., 1990). A similar re- (E. TAKANOand M. J. BIBB, unpublished).
duction was observed when fragments corre- Thus, any major role of ufsR in activating
sponding to the N-terminal264 or 510 amino genes expressed in stationary phase may de-
acids of AfsR, rather than the entire coding pend on posttranscriptional regulation of
region, were expressed in S. lividuns; even ufsR, a deduction consistent with the modifi-
cloned fragments containing the C-terminal cation of AfsR by phosphorylation.
half of AfsR elicited a stimulatory effect in S. A previously undetected gene, ufsR2, has
lividuns and restored actinorhodin, undecyl- recently been identified in S. lividans (VOGT-
prodigiosin, and A-factor production to afsB LI et al., 1994). afsR2, which appears to occur
mutants of S. coelicolor (HORINOUCHI et al., in a similar location in S. coelicolor (MATSU-
1983, 1990), although this effect could also be MOTO et al., 1994; VOGTLIet al., 1994), is
due to ufsR2, a recently identified small gene transcribed in the same direction as ufsR but
located immediately downstream of ufsR and from its own promoter and encodes a 63 ami-
also capable of stimulating actinorhodin and no acid protein. When cloned at high copy
undecylprodigiosin production (VOGTLI et number, afsR2 suppresses an ufsB mutation
al., 1994; see below). Disruption of ufsR in S. in S. lividuns and stimulates actinorhodin and
coelicolor using DNA fragments from the 5’, undecylprodigiosin production in S. coelico-
middle, and 3’ regions of the ufsR coding se- lor. The stimulatory effect on actinorhodin
quence led to only a 4-fold reduction in acti- production in both species appears to be me-
norhodin production, which was also delayed diated through transcription of uctll-ORF4,
(HORINOUCHI et al., 1990), suggesting that and is retained after deletion of most of the
ufsR is not essential for antibiotic biosynthe- C-terminal domain of the chromosomal copy
sis. AfsR is phosphorylated in vitro by the of ufsR in S. lividans. Fragments of S. coelico-
membrane-bound product of afsK which lies lor DNA containing the C-terminal portion of
downstream of, and in the opposite orienta- afsR stimulated actinorhodin production
tion to, ufsR (HONGet al., 1991; HORINOU- (HORINOUCHI et al., 1983, 1990; see above);
CHI, 1993). AfsK (799 amino acids) shows sig- since these fragments were likely to have con-
nificant similarity to eukaryotic serine- tained ufsR2, it is possible that the latter,
threonine protein kinases, and phosphoryla- rather than the C-terminus of AfsR, was re-
tion of AfsR occurs at serine and threonine sponsible for the increase in antibiotic pro-
residues (MATSUMOTOet al., 1994). More- duction. Since deletion of a chromosomal seg-
over, phosphorylation of AfsR in vitro is se- ment of S. lividuns that included afsR2 (and
verely reduced by K-252a and staurosporine, part of ufsR) did not completely block blue
pigment production, ufsR2 does not appear to 3.3.2.3 abaA Influences the
play an essential role in actinorhodin produc-
tion. Production of Three of the Four
Antibiotics Made by Streptomyces
coelicolor
3.3.2.2 afsQl and afsQ2 - A ubuA of S. coelicolor was isolated by virtue
Two-Component Regulatory of its ability to stimulate actinorhodin produc-
System that Can Influence tion in S. lividuns when cloned on a high-copy
number plasmid; the effect on undecylprodi-
Antibiotic Production giosin production was not reported (FERNAN-
DEZ-MORENOet al., 1992). Sequencing of a
ufsQl and ufsQ2 were isolated in the same 2 kb Pstl fragment revealed five short ORFs,
way as ufsR (ISHIZUKAet al., 1992). Se- with ORFs A, B, and C transcribed divergent-
quence analysis of a 1.3 kb Kpnl-Pstl frag- ly from ORFs D and E. ORFB and 137 nu-
ment of S. coelicolor DNA that stimulated ac- cleotides of downstream sequence were suffi-
tinorhodin, undecylprodigiosin, and A-factor cient to give the same stimulatory phenotype
production in S. lividuns HH21 identified in S. lividuns, and disruption of the chromoso-
ufsQl whose predicted product is homolo- mal copy of ORFB in S. coelicolor resulted in
gous to bacterial response regulator genes loss of actinorhodin production, almost com-
(Fig. 9): AfsQl belongs to the OmpR sub- plete loss of undecylprodigiosin synthesis, a
family (VOLZ, 1993). ufsQ2, which was subse- reduction in CDA production, but no effect
quently discovered downstream of ufsQl, ap- on methylenomycin. When cloned at high
pears to be translationally coupled to it. copy number, ubuA was unable to confer acti-
AfsQ2 belongs to the family of sensory histi- norhodin production on a mutant deficient in
dine protein kinases; thus the genes appear to the pathway-specific activator gene uctll-
constitute a two-component regulatory sys- ORF4, consistent with a location “higher up”
tem (Fig. 9 STOCKet al., 1990 ALEXand SI- in any putative regulatory cascade.
MON,1994). AfsQ2 is presumed to be a mem-
brane protein (it has putative membrane
spanning domains towards its N-terminus)
which is thought to be autophosphorylated at 3.3.2.4 absA and absB - Mutants
His294in response to an unknown signal; the Isolated on the Basis of a
phosphate group may then be transferred to Pleiotropic Defect in Antibiotic
Asp5*of AfsQl. AfsQl is, or interacts with, a
transcriptional activator (ISHIZUKAet al., Production
1992). Evidence for this model of AfsQ2 ac-
tion was obtained by changing His294 to An extensive screen for UV-induced mu-
G ~ uwhich ~ , in a loss of stimulatory
~ ~ resulted tants deficient in both actinorhodin and unde-
activity in S. lividuns. Cloned fragments con- cylprodigiosin production led to the identifi-
taining only ufsQl gave the same level of cation of ubsA and ubsB (ADAMIDISet al.,
stimulation as those containing both genes. 1990; ADAMIDIS and CHAMPNESS, 1992); mu-
However, disruption of both genes in S. coeli- tants of both classes were also defective in
color had no obvious phenotypic effect. Thus CDA and methylenomycin synthesis (though
ufsQl and ufsQ2 either are inessential for an- they expressed methylenomycin resistance).
tibiotic production or operate under as yet The rarity of ubsA mutants (5.10-6 per survi-
undefined physiological conditions. Alterna- vor) suggested that they may represent a par-
tively, the stimulatory effects of ufsQl may ticular allelic form (CHAMPNESS et al., 1992).
reflect the ability of AfsQ1, when present at Pseudorevertants of an ubsA mutant were ob-
high levels, to substitute for AbsA, a response tained that fell into two classes: sub(1) pseu-
regulator that clearly does play a role in anti- dorevertants which overproduced actinorhod-
biotic production (Sect. 3.3.2.4). in and undecylprodigiosin, and sub( 11) which
made wild-type levels of antibiotic. Both 3.3.2.6 Genes that Affect Both
classes contain suppressor mutations mapping
close to the starting absA mutation, but re-
Antibiotic Production and
combining with it. Sequencing of a DNA frag- Morphological Differentiation
ment that complements the absA mutation
showed that its product is homologous to bac- In surface-grown cultures of streptomy-
terial sensory histidine kinases, and prelimi- cetes, antibiotic production generally coin-
nary data suggest that a homolog of response cides with the onset of morphological differ-
regulators is located immediately downstream entiation (Sect. 3.2). The isolation of bld mu-
(P. BRIAN and W. CHAMPNESS,personal tants defective in both processes points to at
communication). Interestingly, additional co- least some common elements of genetic con-
pies of afsQl restored actinorhodin produc- trol. It is important to note that the morpho-
tion to an absA (but not absB) mutant (ISHI- logical deficiencies of some classes of bld mu-
ZUKA et al., 1992), possibly indicating cross- tants can be suppressed nutritionally or in
talk between the two systems (Sect. 3.3.2.2). some cases by cross-stimulation by diffusible
absB mutants sporulate less well than their factors. Phenotypic suppression of the pleio-
progenitor and produce low levels of actino- tropic defect in antibiotic production has
rhodin, undecylprodigiosin, and methyleno- been observed in only a few cases and gener-
mycin on some media. Attempts to clone ally not under conditions that suppress the
absB on a low copy number vector by screen- morphological deficiency (an exception to
ing for restoration of actinorhodin production this is provided by bldH mutants; see be-
led to the isolation, in addition to actZZ-ORF4 low).
and afsR, of a cloned fragment which fully The most extensive studies of bld mutants
complements the absB mutation (T. ADAM- have been made in S. coelicolor in which at
IDIS and W. CHAMPNESS, personal communi- least 10, and perhaps 11, different classes of
cation). bld mutants with deficiencies in antibiotic
production - bldA, B, D, E, F, G, H, Z, -17,
-21, -830 - have been identified (reviewed by
3.3.2.5 miu - Multicopy Inhibition classes CHAMPNESS and CHATER,1994). All mutant
except bldE and bldF (which both
of Antibiotic Production produce abundant undecylprodigiosin) are
deficient in both actinorhodin and undecyl-
Attempts to clone absA on a high copy prodigiosin synthesis and most are also defi-
number plasmid led to the identification of S. cient in methylenomycin and CDA produc-
coelicolor DNA that inhibited the production tion. Antibiotic production is restored to
of all four antibiotics. The DNA fragment, bldH mutants grown on mannitol instead of
which was isolated repeatedly, had no inhibi- glucose, and undecylprodigiosin is produced
tory effect at low copy number (CHAMPNESS by bldA mutants grown at low phosphate
et al., 1992). Transcription of redD and actZZ- concentrations. Four of the bld genes ( A , B,
O W 4 is undetectable in strains containing D, and G ) have been cloned, and detailed
the fragment on a high copy number plasmid characterization has been reported for bldA.
(W. CHAMPNESS, personal communication). Remarkably, bldA encodes the only tRNA in
Subcloning localized the inhibitory function, S. coelicolor and S. lividans that can translate
termed mia, to a 363 bp Sau3Al fragment the rare leucine codon UUA efficiently
that does not appear to be protein-coding. It (LAWLORet al., 1987; LESKIWet al., 1991a).
is not known whether the inhibitory effect re- The lack of expression of the xylE reporter
sults from the DNA itself or its transcript. gene when fused to act, red, or mmy tran-
scription units in bldA mutants suggests that
the defect in antibiotic production reflects a
failure to transcribe the biosynthetic structur-
al genes even though bldA encodes a compo-
nent of the translational apparatus (GUTHRIE
and CHATER,1990 BRUTONet al., 1991; A. in surface-grown cultures; furthermore, the
WIETZORREKand K. F. CHATER, unpub- efficiency of translation of seven UUA co-
lished). An explanation for the failure to tran- dons of a heterologous reporter gene appar-
scribe act genes is to be found in the presence ently increased in older cultures. The differ-
of a TTA codon in the pathway-specific regu- ences between the two sets of results may re-
latory gene acrZZ-ORF4. If this codon is flect the different growth conditions used
changed to the synonymous codon TTG, acti- possibly the liquid culture conditions of GRA-
norhodin production takes place even in a MAIO et al. (1993) overrode a regulatory role
bldA mutant (FERNANDEZ-MORENO et al., of bldA adapted for surface growth.
1991). Phenotypically similar bldA mutants
have also been isolated in the phylogenetical-
ly more distant S. griseus (MCCUE et al.,
1992), suggesting that the role of bldA in sec- 3.3.2.7 An Outline Scheme for the
ondary metabolism and differentiation is Interactions of Pleiotropic
widespread among streptomycetes. The unim- Antibiotic Regulatory Genes in
paired vegetative growth of bldA mutants in-
dicates that TTA codons are absent from Streptomyces coelicolor
genes essential for primary metabolism and
growth, but l'TA codons have been found in No satisfactory integrated model has yet
several genes likely to be expressed late in emerged for the roles of the various pleio-
growth. This, coupled with evidence that tropic regulatory genes in antibiotic produc-
bldA-specific RNA is more abundant late in tion (probably because not enough of the
surface growth (LAWLORet al., 1987), pro- pieces of the jigsaw are yet available), but
vided support for the idea that bldA regulates here we summarize some of the key features
antibiotic production by allowing the transla- that must be taken into account. For actino-
tion of UUA codon-containing mRNA only rhodin and undecylprodigiosin, expression of
under appropriate conditions (LESKIWet al., the pathway-specific activator genes actZZ-
1991b). However, caution is necessary in as- ORF4 and redD, respectively, appears to play
suming an active regulatory role for bldA, a major limiting role in determining the onset
since one series of detailed experiments on of antibiotic production (TAKANO et al.,
liquid-grown cultures of S. coelicolor failed to 1992; GRAMAJOet al., 1993), so it is attractive
reveal any limitation of translation of the to propose that all the pleiotropic genes (ufs,
acfll-ORF4 UUA codon during exponential aba, ubs, mia, bld) influence the synthesis of
growth (GRAMAJO et al., 1993). Together these two antibiotics via acrll-ORF and redD.
with the transition phase activation of acrZZ- The generalized and simplified scheme in Fig.
ORF4 transcription, these results were consis- 17 is built from the following observations
tent with a more prosaic possibility that the and deductions.
absence of TTA codons from vegetatively ex- (1) Transcription of acrll-ORF4 is virtually
pressed genes might reflect selection against undetectable in an afsB mutant, at least under
codons that were inefficiently translated dur- certain culture conditions, suggesting that the
ing growth, rather than a role for bfdA in the (still uncharacterized) AfsB gene product
temporal regulation of actinorhodin produc- may be higher than ActII-ORF4 in a tran-
tion. On the other hand, the observations of scriptional cascade (HORINOUCHIet al.,
LESKIWet al. (1993) tend to support a regula- 1989a). (It should be noted that afsB mutants
tory role for bldA. Northern analysis of RNA are noticeably leaky in their actinorhodin de-
from surface-grown cultures indicated that ficiency on a variety of different media.)
the amount of the bldA transcript increased (2) absA and absB, whose mutant phenotype
with growth, and S1 nuclease protection as- proves their importance for antibiotic biosyn-
says revealed an increase in the level of the 5 ' thesis, may perhaps play a role in maximizing
end of the mature bldA transcript late in expression of the pathway-specific activator
growth, both in rich liquid media (this was genes, since the introduction of acfll-ORF4
not observed by GRAMAJOet al., 1993) and and redD on high copy number plasmids res-
Fig. 17. Factors potentially

determining the onset of an-
tibiotic production in strep-
tomycetes. Thinner lines
present plausible interactions
for which there is currently
no direct evidence.
tores production of the relevant antibiotic in afsQ takes the activity of its product above a
absA and absB mutants (T. ADAMIDISand threshold level. If afsQl has such a role, it is
W. CHAMPNESS,personal communication). redundant in the wild-type strain under labo-
Alternatively, the abs genes may encode ac- ratory conditions; perhaps this role can also
cessory elements normally required for acti- be filled by afsR, a question that should be
norhodin and undecylprodigiosin synthesis resolved by isolating afsR and afsR afsQ null-
which are rendered unnecessary by overpro- mutants. It remains possible that the high
duction of ActII-ORF4 and RedD. copy number effects of afsQ result from arti-
(3) Multiple copies of afsR and afsR2 stimu- ficially induced cross-talk between normally
late actinorhodin production, but appear to separated regulatory elements.
depend on actZZ-ORF4 for this effect (B. FLO- (6) The recent sequence analysis of absA, to-
RIANO and M. J. BIBB, unpublished results; gether with the published data on afiRlafsK
T. ADAMIDISand W. M. CHAMPNESS, per- and afsQllafsQ2, strongly suggest that pro-
sonal communication; VOGTLI et al., 1994), tein phosphorylation, and potentially phos-
while multiple copies of segments encoding phorylation cascades, play a role in triggering
the C-terminal portion of AfsR, and in retro- antibiotic production; presumably AfsK,
spect containing afsR2, confer actinorhodin AfsQ2, and AbsA sense external signals that
and undecylprodigiosin production on absA cause phosphorylation of their regulatory
and absB mutants (CHAMPNESS et al., 1992). counterparts (AfsR, AfsQ1, and the product
(4) Taken together, observations (1)-(3) sug- of a gene located downstream of absA) which
gest a working model in which expression of can then stimulate transcription of the anti-
afsR and afsR2 or the activities of their prod- biotic biosynthetic pathways, perhaps via the
ucts, depend on absA and absB, and in which pathway-specific regulators.
AfsR2 and AfsR (perhaps in its phosphory- (7) None of the mutants described above are
lated form) stimulate expression of actll- unconditionally and completely defective in
ORF4 (and possibly redD). production of all four antibiotics; even absA
( 5 ) Extra copies of afsQl restore actinorhod- produces actinorhodin on some media (W.
in and undecylprodigiosin production to absA CHAMPNESS, unpublished data), and antibiot-
(but not absB) mutants (ISHIZUKAet .al., ic production in several of the others shows
1992), suggesting that afsQ may depend on media dependence. This may indicate a com-
absA for a role in enhancing acrll-ORF4 and plex regulatory network in which there are
redD activity, unless a high copy number of several different routes to activation of a par-
ticular pathway. (Alternatively, the available specific regulators, nor whether they act in a
mutants may not be truly null.) linear cascade or by convergence. Further-
(8) Little is known of how abaA fits into this more, there is little information about the
interactive scheme, but in affecting three of regulation of expression of the characterized
the four antibiotics it differs from both the afs pleiotropic regulatory genes, and several rele-
loci (which appear to affect only actinorhodin vant bfd genes remain uncharacterized.
and undecylprodigiosin) and the abs loci
(which affect all four).
(9) bfdA is the only bfd gene whose mode of 3.3.3 Streptomyces griseus - The
action is (partially) understood. bldA depend-
ence of actinorhodin production appears to A-Factor Cascade
be exerted entirely at the level of translation
of the unique UUA codon in the actll-ORF4 Although diffusible factors have been im-
transcript. Undecylprodigiosin production, plicated in the production of several antibio-
except at low phosphate levels, also requires tics in streptomycetes (see Sect. 3.2.4), the
bldA, though neither redD nor, apparently, role of A-factor in the production of strepto-
any of the red biosynthetic structural genes mycin in S. griseus is by far the best character-
contain TTA codons (NARVAand FEITEL- ized. A-factor was discovered by KHOKHLOV
SON, 1990 GUTHRIEand CHATER,1990). In et al. (1967). It was found to be required for
contrast to actll-ORF4 whose transcription is both streptomycin production and sporula-
not bldA-dependent, transcription of redD tion in S. griseus (KHOKHLOV, 1982) and also
could not be detected in a bldA mutant (J. for streptomycin resistance (HARAand BEP-
WHITEand M. J. BIBB,unpublished results). PU, 1982a). Subsequent genetic and molecular
Presumably there is at least one other gene analyses have provided considerable insights
required for redD transcription whose tran- into its mode of action. A-factor accumulates
script does contain a UUA codon. The pwb to detectable levels in the culture medium just
mutations which restore undecylprodigosin, before the onset of streptomycin production
but not actinorhodin or aerial mycelium for- (HARAand BEPPU, 1982b). It appears to be
mation, to bldA mutants and which may map freely diffusible across the cytoplasmic mem-
within the red cluster (E. P. GUTHRIEand K. brane and binds to a cytoplasmic A-factor-
F. CHATER,unpublished) could be relevant binding protein of approximately 26 kDa with
here. a stoichiometry of 1:1 and a dissociation con-
(10) Antibiotic production is associated with stant of 0.7 nM, consistent with the similar
reduced growth rate. One of the signals impli- low concentrations of A-factor required for
cated is an increased ppGpp level. No other biological activity. The receptor protein is
candidate signal, intracellular or extracellular, present at about 30-40 copies per genome. It
has been described that might have pleiotrop- is thought that binding of A-factor prevents
ic activity, though extracellular acidification the receptor protein from acting as a repres-
can activate methylenomycin production (see sor of a hypothetical gene (X) required for
Sect. 3.2.2). Whatever the signals, evidence is both sporulation and streptomycin produc-
growing that they may lead, directly or indi- tion (Fig. 18). A negative regulatory role for
rectly, to phosphorylation of several regulato- the binding protein is indicated by the discov-
ry proteins such as AfsR, AfsQ1, and AbsA ery that S. griseus mutants unable to make A-
by specific kinases and thence, via pathway- factor can undergo further mutations that res-
specific regulatory genes and their products, tore streptomycin production and sporulation
to transcription of genes encoding antibiotic by eliminating A-factor-binding protein.
pathways. Gene X is believed to activate transcription of
(11) Model building is limited by major gaps a gene that in turn encodes an activator of
in knowledge. It is not yet possible to deduce strR, a streptomycin pathway-specific activa-
at what level (transcriptional, translational, or tor gene (Sect. 3.3.4.3). In support of this, a
posttranslational) the phosphorylated pleio- protein in extracts of A-factor-producing
tropic regulators interact with the pathway- strains, but absent from A-factor-deficient
0
A-factor
0
binding
protein
A-factor A-factor
dependent
protein
c
I
Sporuhtion
I
Streptomycin
Fig. 18. Model for the regulation of streptomycin biosynthesis in S. griseus. P, promoter; Protein X, regul-
atory protein derived from unidentified gene X required for both sporulation and streptomycin produc-
tion; adp, regulatory gene encoding the A-factor-dependent protein that binds to the promoter region of
strR, the pathway-specific activator gene for streptomycin production; aphD and strB encode a resistance
determinant and biosynthetic enzyme, respectively; Sm, streptomycin. Redrawn from HORINOUCHI
(1993).
mutants, could bind to nucleotide sequences both processes; they also suggest that the A-
just upstream of strR. Little is known about factor-binding protein is present during early
how A-factor synthesis takes place or is con- growth. In view of the positive autoregulation
trolled. A putative A-factor biosynthetic of OHHL synthesis in Vibrio fischeri (see
gene, @A, was cloned from S. griseus, but its Sect. 2.2), one might anticipate that the A-fac-
predicted translation product did not resem- tor-binding protein also represses - directly
ble any other known protein (HORINOUCHIor indirectly - the expression of &A.
et al., 1989b). Surprisingly, ufsA did not hy-
bridize to DNA from some streptomycetes
that produce structurally similar y-butyrolac- 3.3.4 Pathway-Specific Regulatory
tones (HORINOUCHI et al., 1984). The earlier
onset of streptomycin production and sporu- Genes
lation in mutants lacking the A-factor-binding
protein (MIYAKEet al., 1990), and earlier We have already made frequent references
production of streptomycin on elevation of to the important role of pathway-specificreg-
A-factor levels, either by exogenous addition ulatory genes. Here we review these genes
(BEPPU,1992) or by cloning ufsA on a multi and their (deduced) products in more detail.
copy plasmid (HORINOUCHI et al., 1984), are
clearly consistent with a role for ufsA and the
y-butyrolactone in determining the timing of
3.3.4.1 The actll-ORF4, redD, and vators causes increased transcription of the
corresponding biosynthetic structural genes
dnrl Family of Pathway-Specific and can be used to cause antibiotic produc-
Activator Genes tion prematurely during rapid growth.
The redD and actll-ORF4 genes are homo-
Early genetic analyses identified putative logous to each other and to the positively-act-
pathway-specific activator genes for the unde- ing regulatory gene dnrl required for the pro-
cylprodigiosin (redD) and actinorhodin duction of daunorubicin in Streptomyces peu-
(actll-ORF4) biosynthetic pathways of S. coe- cetius (STUTZMAN-ENGWALL et al., 1992). In-
ficofor (reviewed by CHATER,1992). The fail- sertional inactivation of dnrl blocks produc-
ure of redD and actll-ORF4 mutants to co- tion of daunorubicin and all of its biosynthet-
synthesize with representatives of any other ic intermediates and prevents transcription of
red or act mutant class, the lack of expression putative operons containing daunorubicin
of red and act biosynthetic structural genes in biosynthetic and resistance genes. The pre-
redD and actll-ORF4 mutants, and the ability dicted redD, actll-ORF4, and dnrl gene prod-
of extra cloned copies of redD and actll- ucts show 33-37% amino acid sequence iden-
ORF4 to elicit overproduction of undecyl- tity in pairwise alignments (Fig. 19; STUTZ-
prodigiosin and actinorhodin, respectively, all MAN-ENGWALL et al., 1992). dnrl can com-
suggest that redD and actll-ORF4 are path- plement mutations in actll-ORF4, and actll-
way-specific activator genes. Furthermore, ORF4 can stimulate daunorubicin production
the stationary-phase production of undecyl- in S. peucetius (STUTZMAN-ENGWALL et al.,
prodigiosin and actinorhodin appears to re- 1992), but redD and actll-ORF4 do not show
sult from transcriptional activation of redD cross-complementation. Since computer anal-
(TAKANO et al., 1992) and actll-ORF4 (GRA- ysis using the algorithm of DODD and EGAN
MAJO et al., 1993), respectively. Production of (1990) failed to reveal likely helix-turn-helix
the antibiotics in rapidly growing cultures ap- DNA-binding motifs in these proteins, they
pears to be limited only by the absence of the may represent a novel family of DNA-bind-
relevant pathway-specific activator protein, ing regulatory proteins. Perhaps more likely,
because overproduction of the putative acti- they may need to interact with other proteins
Fig. 19. Alignment of the amino acid sequences of RedD, ActII-ORF4, DnrI, and the N-terminal region of
AfsR. The alignment was made using the PILEUP and PRETTYBOX programs contained in the UWG
sequence analysis package (DEVEREUX et al., 1984).
to effect activation of biosynthetic structural may be binding sites for the 65 kDa putative
gene promoters. transcriptional activator (SrmR) encoded by
Intriguingly, the N-terminal region of srmR. SrmR shows no significant sequence
AfsR, excluding the putative DNA-binding similarity to any other known protein. srmR
motifs in the C-terminal region (HORINOU- homologs have not been reported in gene
CHI et al., 1990), also shows significant identi- clusters for the production of other macrolide
ty to the RedD-ActII-ORF4-DnrI family antibiotics. The srmR gene contains a single
(Fig. 19), raising the possibility that the stimu- TTA codon, so it may be a target for regula-
lation of actinorhodin and undecylprodigiosin tion by a bldA homolog (Sect. 3.3.2.6).
production by multiple copies of afsR may re-
flect partial functional interchangeability of
AfsR with ActII-ORF and RedD. However,
since strong stimulatory effects were also ob- 3.3.4.3 strR Encodes a
served with segments of AfsR (notably the C-
terminal half) that are not homologous to DNA-Binding Protein that
RedD and ActII-ORF4, models that rely Regulates at Least One of the
solely on functional substitution are at best an Streptomycin Biosynthetic Genes
oversimplification.
in Streptomyces griseus
Production of streptomycin and 5 '-hy-
3.3.4.2 srmR - A Regulatory Gene droxy-streptomycin has been studied in S. gri-
for Spiramycin Production in sew and S. glaucescens GLA.0, respectively.
In S. griseus, strR appears to be a positive reg-
Streptomyces ambofaciens ulator of at least one of the biosynthetic struc-
tural genes, strBl, which encodes amidino-
Cloning and gene disruption revealed a pu- transferase I, and the strR homolog of S. gluu-
tative regulatory gene, srmR, for spiramycin cescens GLA.0 is presumed to perform the
production in Streptomyces ambofaciens. same function. StrR contains a potential he-
srmR mutants fail to make spiramycin and do lix-turn-helix DNA-binding motif, and the
not cosynthesize the antibiotic with srm mu- protein binds to at least two specific sites in-
tants that accumulate intermediates in the side the str clusters of both species. Analysis
biosynthetic pathway. srmR was required not of the binding sites reveals 11bp inverted re-
only for transcription of srmG, which encodes peats separated by 11 bp. These results make
the polyketide synthase that produces the it more likely that StrR acts as a conventional
aglycone of spiramycin, but also for expres- transcriptional activator, rather than through
sion of the resistance gene srmB (GEISTLICH transcriptional antitermination (RETZLAFFet
et al., 1992). Lack of expression of srmB in al., 1993). The strR genes of both species con-
srmR mutants proved to be an indirect effect tain single TTA codons and bldA mutants of
of the failure of srmR mutants to produce spi- S. griseus do not make streptomycin (MCCUE
ramycin, which is an inducer of its own resist- et al., 1992). Single 7 T A codons are also pres-
ance gene. srmR was also required for the ent in strN, encoding a biosynthetic enzyme
transcription of another flanking gene, srmX, of both species and in strA, the streptomycin
that is also likely to play a role in spiramycin resistance gene of S. glaucescens (DISTLERet
production. Multicopy cloning of srmR in the al., 1992).
wild-type strain led to a 4-fold increase in spi-
ramycin production. The srmG and srmX
promoters are strikingly similar to each
other, with three blocks of conserved se-
quences, centered at about position -39
(CCNGNCGTTCCT), -27 (CCCGGC), and
- 10 (CTGTNN-GNT), one or more of which
3.3.4.4 brpA and dnrN - S. peucetius encodes at least one other puta-
tive regulatory gene, dnrN. DnrN shows sig-
Regulatory Genes for Bialaphos nificant sequence similarity throughout its
Production in Streptomyces length to the UhpA subfamily of two-compo-
hygroscopicus and for nent response regulator proteins, including a
likely site for phosphorylation, although dnrN
Daunorubicin Production in does not appear to be closely linked to a sen-
Streptomyces peucetius which Show sory histidine protein kinase gene. Transcrip-
tion of dnrZ is reduced in dnrN mutants
Different Degrees of Similarity to (HUTCHINSON et al., 1994), and this may be
Response Regulator Genes of the cause of their daunorubicin deficiency. In-
Two-Component Systems deed, production is restored by adding extra
copies of the cloned dnrl gene. (On the other
Bialaphos is made by Streptomyces hygro- hand, extra copies of dnrN do not restore
scopicus at the approach of stationary phase production to a dnrl mutant.) A non-sporu-
(HOLT et al., 1992). Early studies identified lating derivative (H6101) of Srreptomyces
brpA as a likely pathway-specific activator peuceticus var. caesius that is deficient in dau-
gene for bialaphos production; brpA mutants norubicin production has been described.
were defective in at least 6 of the 13 steps Cloned copies of either dnrZ or dnrN restored
leading to bialaphos production, lacked at daunorubicin production in H6101 suggesting
least 7 of the bialaphos biosynthetic tran- that the mutant is defective in dnrN expres-
scripts, and showed reduced levels of biala- sion (presumably as a secondary result of the
phos resistance (ANZAIet al., 1987). Further- pleiotropic mutation) (HUTCHINSONet al.,
more, a brpA mutant lacked 27 proteins im- 1994; STUTZMAN-ENGWALL et al., 1992).
plicated in bialaphos production (HOLTet al.,
1992). brpA encodes a predicted product of
28 kDa whose C-terminal region resembles a 3.3.4.5 Negative Regulation of
region located towards the C-terminus of the Methylenomycin Production in
response regulators of the UhpA subfamily of
two-component regulatory systems (RAI- Streptomyces coelicolor
BAUD et al., 1991; GROSSet al., 1989). This
region includes a putative helix-turn-helix The biosynthetic genes for methylenomycin
motif, but does not extend to the conserved production in S. coelicolor reside on the
region that includes the site of phosphoryla- 350 kb linear plasmid SCPl (KINASHIet al.,
tion of the regulatory components. BrpA con- 1987). mmy genes were initially isolated by
tains three hydrophobic regions towards its mutational cloning yielding over 20 kb of con-
N-terminus, leading to suggestions that these tiguous DNA (CHATERand BRUTON,1983,
might represent transmembrane domains or 1985). Insert-directed prophage insertions
regions of hydrophobic interaction with other into the leftmost 3 kb of the cluster caused
proteins (RAIBAUD et al., 1991). brpA is tran- overproduction of methylenomycin. Se-
scribed from three promoters expressed at a quence analysis of this region revealed a gene
low level early in exponential growth but (mmyR) whose predicted product resembles
more strongly during a pause in growth short- the TetR family of repressor proteins (C. J.
ly before stationary phase, and the activity of BRUTONand K. F. CHATER,unpublished re-
one of them (brpAp3) continued to increase sults; Sect. 3.3.5.1). Disruption or deletion of
on entry into stationary phase. brpA contains mmyR resulted in overproduction of methyle-
a single TTA codon located towards the C- nomycin, providing the only known example
terminus of the coding region making it po- of pathway-specific negative regulation of an-
tentially bldA-dependent (bldA mutants of S. tibiotic production. The absence of methyle-
hygroscopicus have not been described). nomycin production from various pleiotropic
In addition to the actZZ-ORF4-like gene mutants (see Sect. 3.3.2) suggests that there
dnrl, the daunorubicin biosynthetic cluster of may also be a positively acting pathway-spe-
cific regulatory gene, and the absence of mmy al., 1991) demonstrated that ActII-ORF1 re-
gene transcription in a bldA mutant (A. pressed transcription of ucfll-ORF2 and of it-
WIETZORREK and K. F. CHATER,unpub- self (and indicated that both genes could be
lished results) leads to the prediction that this expressed in the absence of the pathway-spe-
gene should contain a TTA codon. cific activator ActII-ORF4; Sect. 3.3.4.1).
Both promoters were most active in S. coeli-
color cultures that were making actinorhodin.
GUILFOILE and HUTCHINSON(1992a, b)
3.3.5 Induction of Antibiotic showed that transcription of fcmA in S. gluu-
Resistance in Antibiotic Producing cescens was induced by tetracenomycin C and
Streptomycetes - Antibiotics as that inactivation of fcmR resulted in constitu-
tive tcmA expression; furthermore, in vitro
Inducers of Gene Expression binding of TcmR to the tcmWA intergenic re-
gion was inhibited in the presence of tetrace-
Resistance towards an antibiotic made by a nomycin C. It thus seems likely that export of
streptomycete often develops only at the on- tetracenomycin C is induced by the antibiotic
set of antibiotic production. In some cases, re- once production begins (GUILFOILEand
sistance may be a consequence of export HUTCHINSON, 1992b).
alone (and may therefore be regarded as a Resistance of S. coelicolor to methyleno-
late step in antibiotic production). In others, mycin is conferred by mmr which encodes a
specific resistance mechanisms operate, in ad- protein with significant sequence similarity to
dition to efflux, to ensure continued viability. the same family of transporter proteins as
Below, we consider examples of resistance ActII-ORE and TcmA (NEALand CHATER,
mechanisms that appear to be induced by 1987). HOBBSet al. (1992) found that tran-
their antibiotic substrates or by intermediates scripts corresponding to at least one of the
in the pathway. methylenomycin biosynthetic genes appeared
before that of mmr, suggesting that mmr ex-
pression might be induced by methylenomy-
3.3.5.1 actZZ-ORFU2 of cin or by an intermediate in the pathway.
Streptomyces coelicolor and
tcmWA of Streptomyces 3.3.5.2 srmB of Streptomyces
glaucescens GLA.0 - Regulatory ambofaciens - A Probable
Cassettes for Antibiotic Export
ATP-Dependent Efflux System
ucfZZ-ORF1/2 of S. coelicolor (CABALLERO Induced by its Antibiotic Substrate
et al., 1991) and fcmWA of S. gluucescens
GLA.0 (GUILFOILEand HUTCHINSON, The srmB, flrC,and drrA gene products re-
1992a) are divergently transcribed gene pairs spectively involved in the export of spiramy-
for export of actinorhodin and tetracenomycin from S. umbofuciens, tylosin from S. fru-
cin C, respectively. Both ActII-ORE and diue, and daunorubicin from S. peucetius
TcmA are similar to tetracycline transport show a high degree of amino acid sequence
proteins from several other gram-positive similarity (SCHONERet al., 1992; GUILFOILE
bacteria (including Tet347 of Sfrepfomycesri- and HUTCHINSON, 1991). The proteins each
mosus) and gram-negative organisms, and possess a putative ATP-binding motif, sug-
ActII-ORF1 and TcmR are clearly members gesting that they are components of ATP-de-
of the TetR family of repressor proteins. pendent efflux systems. Although details of
Inactivation of ucfll-ORE appeared to pre- the regulation of flrC and drrA have not been
vent the export of actinorhodin (FERNAN- published, expression of srmB in S. frudiue is
DEZ-MORENO et al., 1991). Studies of ucrZZ- induced by spiramycin (GEISTLICHet al.,
ORF1/2 expression in E. coli (CABALLERO et 1992; Sect. 3.3.4.2); increased levels of srmB
transcription occurred on addition of spira- that the gyrBR promoter responds to changes
mycin to mutants blocked in production of in DNA supercoiling. Transcription of gyrBR
the antibiotic (no increase was observed with increased when DNA gyrase was inhibited by
the wild-type strain). novobiocin or ciprofloxacin (an inhibitor of
the A subunit), i.e., under conditions that re-
duce negative supercoiling, and decreased
3.3.5.3 Induction of tZrA in during growth in a medium of high osmotic
Streptomyces fradiae - A Role for strength that should increase negative super-
coiling (HIGGINS et al., 1988). Thus resistance
Transcriptional Attenuation? to novobiocin in S. sphaeroides probably oc-
curs, at least in part, by production of the re-
tlrA and tlrD are two of at least four S. fra- sistant gyrase following the reduction in nega-
diae genes that confer tylosin resistance. They tive supercoiling which results from inhibition
cause di- and monomethylation, respectively, of the sensitive enzyme by the antibiotic.
of residue A-2058 of 23s rRNA. Unlike tlrD, Two different, uncharacterized genes that
which is expressed constitutively, expression confer novobiocin resistance on S. lividans
of rlrA is induced by tylosin or its biosynthetic have been isolated from S. niveus (HOG-
intermediates (KELEMENet al., 1994). In the GARTH et al., 1994); one hybridizes with a
absence of inducer, transcription terminates second resistance determinant from S. sphaer-
at the beginning of the coding region of tlrA oides (THIARAand CUNDLIFFE,1988). The
through the adoption of a particular second- isolation of multiple resistance genes and a
ary structure in the RNA; the presence of an marked increase in the level of novobiocin re-
inducer is thought to cause sensitive ribo- sistance (from 25 to over 200 pg mL-') dur-
somes to stall in the non-translated leader re- ing growth of S. niveus suggest that resistance
gion, preventing transcriptional termination is determined by several mechanisms that
and allowing production of the methylase. may be subject to different regulatory con-
trols. Recent studies on s. niveus (HOG-
GARTH et al., 1994) identified a diffusible 7-
butyrolactone signaling molecule that induces
3.3.5.4 Induction of Resistance to high-level resistance to novobiocin well be-
Novobiocin in Streptomyces fore the onset of production.
sphaeroides and Streptomyces
niveus - Roles for DNA 3.3.5.5 Regulation of Isoforms of
Supercoiling and a Diffusible the Target for Pentalenolactone
Signaling Molecule Inhibition in the Producing
Novobiocin, produced by Streptomyces Organism
sphaeroides and Streptomyces niveus, is an in-
hibitor of bacterial DNA gyrase. Gyrase ex- Pentalenolactone (PL) is a potent inhibitor
ists as a tetramer (A2B2) in which the two of glyceraldehyde-3-phosphate dehydrogen-
subunits have different functions that can be ase (GAPDH). The producer, Streptomyces
blocked by different groups of antibiotics. arenae, has two distinct isoforms of GAPDH:
The B subunit, encoded by gyrB, is the target a PL-sensitive enzyme produced before anti-
for novobiocin. S. sphaeroides has two gyrB biotic production and a PL-resistant form
genes (THIARAand CUNDLIFFE,1989,1993): produced on induction of PL synthesis. The
one encoding a novobiocin-sensitive B sub- sensitive isoform rapidly disappears when PL
unit, GyrBS, that is produced constitutively, is produced (FROHLICHet al., 1989). The two
and the other encoding a resistant B subunit, isoforms are encoded by two distinct genes,
GyrBR, that is produced in the presence of but the mechanisms responsible for their reg-
novobiocin. Transcriptional fusions showed ulation are unknown.
4 Concluding Remarks further example of this, beyond that de-

scribed above for B. subfilis, regulates pro-
Studies of the regulation of antibiotic production of three different secondary metabol-
duction in diverse prokaryotes are revealing ites (2,4-diacetyl phloroglucinol, HCN, and
several common themes. Typically, antibiotic pyoluteorin) in Pseudomonas aeruginosa
biosynthetic genes appear to be regulated (LAVILLEet al., 1992). Such two-component
principally at the level of transcription, and systems are clearly important in Streptomyces
this involves rather large numbers of appar- spp., but it is interesting that serine-threonine
ent regulatory genes, most of them acting kinases, previously described only in euka-
more or less globally on secondary metabol- ryotes, are also involved (there is also evi-
ism. The regulatory genes might exert their dence of multiple tyrosine kinases in strepto-
effects by direct interaction with promoters of mycetes, though their roles are unknown;
antibiotic production genes, though it seems WATERSet al., 1994).
more likely that they are mostly involved with While specialized minor u factors are often
producing, detecting, transmitting, and inte- required for transcription of genes for anti-
grating information relevant to deciding the biotic production, there is no evidence that a
appropriateness of commitment to produc- particular sub-branch of u factors is impli-
tion. Different organisms have different ecol- cated (Fig. 2b). Thus, the stationary-phase
ogies, and because different antibiotics differ factor 2 in E. coli is sometimes involved, and
in their biosynthetic origins and modes of ac- in Sfrepfomycesat least one antibiotic produc-
tion, and thus in their potential effectiveness tion gene is transcribed in vifro by RNA poly-
in different ecological situations, activation of merase containing uE (G. H. JONESand M. J.
any particular pathway of any particular or- BUTTNER,personal communication), a u fac-
ganism might be expected to require its own tor that was used as the paradigm of a new
combination of signals; it is, therefore, not subfamily of u factors (the ECF family; LON-
surprising to find great diversity in the sys- ETTO et al., 1994). Sfreptomyces spp. appear
tems analyzed so far. Nevertheless, there are to contain a rather large number of different
recurrent themes. One is the use of quorum u factors (BUTTNER,1989 LONETTOet al.,
sensing. Lipid-soluble lactones play such a 1994), and Sfrepfomycespromoters are very
role in widely different organisms, and in diverse in their sequences and complexity
each case the receptor is cytoplasmic and cap- (STROHL,1992), so it is quite likely that other
able of direct interaction with DNA to in- minor u factors may prove to be involved in
fluence gene expression. However, in B. sub- secondary metabolism.
filis the most well-defined extracellular signal- As knowledge of the regulation of antibiot-
ing is done by a small peptide probably mod- ic synthesis increases, new genetic approaches
ified with a hydrophilic side chain (MAGNU- to industrial overproduction will probably be
SON et al., 1994), and the receptor is a mem- devised which may be especially useful when
brane-located protein kinase that does not in- a new product is being developed. A particu-
teract with DNA directly but phosphorylates larly attractive approach is to add extra copies
and thereby activates a transcription factor. of positive regulatory genes which are found
This solution adopted by B. subtilis to the in many pathways (CHATER, 1990). This
problem of quorum sensing resembles the could be done by self-cloning, but an alterna-
mating pheromone systems of Enferococcus tive strategy is to collect a panel of cloned
faecalis (CLEWELL,1993). It is, therefore, in- DNA fragments (isolated from various spe-
teresting that the B. subfilis pheromone also cies) capable of heterologous stimulation of
controls the competence regulator, the func- production of some antibiotics, such as are
tion of which (like mating) is to permit the in- being discovered by F. MALPARTIDA et al.
gress of DNA. (personal communication) and to introduce
Protein phosphorylation is widely impli- these into any interesting new species. A sim-
cated in activating antibiotic synthesis, nota- ilar approach might also lead to the discovery
bly involving members of the two component of new antibiotics in old strains (HOPWOOD
histidine kinase-response regulator family. A et al., 1983).
Note added in proof approach was used to clone the ppGpp syn-
thetase gene (relA) of S. coelicolor (CHA-
Recently, several papers have been pub- KRABURTTY et al., 1996). The cloned gene
lished on the regulation of antibiotic produc- was used to create a null-mutant that is totally
tion in streptomycetes, mostly in S.coelicolor, deficient in ppGpp synthesis upon amino acid
and are discussed briefly here. The absA lo- starvation (CHAKRABURTTY, 1996). The re-
cus of S. coelicolor (Sect. 3.3.2.4) has been sulting mutant grows at the same rate as the
shown to encode a two-component regulatory relA strain but fails to make Act or Red on
+
system, absAIIA2, which acts as a negative some media, but does so on others; similar re-
regulator of antibiotic production (BRIANet sults were obtained by MART~NEZ-COSTA et
al., 1996). Disruption of absA results in early al. (1996). This indicates an obligatory role
hyperproduction of both actinorhodin (Act) for ppGpp in antibiotic biosynthesis, and to-
and undecylprodigiosin (Red). All four pre- gether with the conditional phenotype of the
viously isolated absA mutations lie in absAl afsR null-mutant (FLORIANOand BIBB,
encoding the predicted sensor histidine ki- 1996), indicates the presence of multiple sig-
nase; these mutations may lock the kinase in nal transduction pathways for the activation
an active conformation preventing the relief of antibiotic production. Further evidence for
of the negative influence of the phosphory- this stems from the isolation and characteriza-
lated form of AbsA2 on antibiotic synthesis. tion of an extracellular signaling molecule, a
In a potentially similar fashion, the cutRS lo- novel y-butyrolactone, that elicits the preco-
cus also acts to negatively regulate Act pro- cious production of both Act and Red when
duction in S. lividans and in S. coelicolor added to the wild-type strain (E. TAKANO,T.
(CHANGet al., 1996). Thus, protein phospho- NIHIRA,and M. J. BIBB,unpublished results).
rylation mediated by absAIIA2 and cutRS The relA and afsR mutants produce, but do
acts to negatively regulate antibiotic produc- not respond to, this factor. Genes encoding
tion, in contrast to the positive effects of afsKI the binding proteins for the y-butyrolactones
afsR (Sect. 3.3.2.1) and afsQlIQ2 (Sect. made by S.griseus (A-factor) and S. virginiae
3.3.2.2). Recent studies on afsR (FLORIANO (the virginae butanolides VB-A-E) (Sect.
and BIBB, 1996) revealed that while it is ho- 3.2.4) have been cloned and sequenced (ON-
mologous to actII-ORF4 and redD, pathway- AKA et al., 1995; OKAMOTO et al., 1995). An-
specific regulatory genes for Act and Red other pleiotropic regulatory gene for antibiot-
production, respectively, it cannot substitute ic production in S. coelicolor is afsB (Sect.
for them. Moreover, an in-frame deletion that 3.3.2.1). Attempts to complement afsB using
removed most of the afsR coding sequence a genomic library made in a low copy-number
resulted in loss of Act and Red production, plasmid led to the discovery that additional
and a marked reduction in the synthesis of copies of hrdB, which encodes the major u
the calcium-dependent antibiotic (CDA), but factor of S. coelicolor (BROWNet al., 1992),
only under some (non-permissive) nutritional restored Red and Act production in afsB mu-
conditions. Although additional copies of tants (WIETZORREK,1996). The effect of
afsR resulted in elevated levels of the actII- hrdB resembles a recent report in Pseudo-
ORF4 and redD transcripts, transcription of monas fluorescens (SCHNIDER et al., 1995), in
the pathway-specific regulatory genes under which production of the antibiotics pyoluteo-
non-permissive conditions was unaffected by rin and 2,4-diacetylphloroglucinol,which are
deletion of afsR. While afsR may operate in- made during stationary phase, was stimulated
dependently of the pathway-specific regulato- by the presence of additional copies of rpoD,
ry proteins to influence antibiotic production, which encodes the major and essential u fac-
the activity of ActII-ORF4 and of RedD un- tor of that organism. This may reflect a role
der non-permissive conditions could depend for the major u factor of both organisms in
on interaction with, or modification by, AfsR. the transcription of antibiotic biosynthetic
To assess whether there might be a causal re- genes, or may result from an indirect effect
lationship between ppGpp synthesis and anti- (e.g., provision of precursors). Consistent
biotic production (Sect. 3.2.3), a PCR-based with the former notion, in vitro transcription
of the pathway-specific regulatory gene redD lowed us to cite their unpublished results. We
was observed upon addition of a protein cor- also thank MEREDYTHLIMBERGand ANNE
responding in size to drdB to core RNA poly- WILLIAMSfor their patience in typing succes-
merase (FUJII et al., 1996). Recent studies sive versions of the manuscript. Our laborato-
have further elucidated the way in which ries' work in this area was funded by the Bio-
bldA (Sect. 3.3.2.6) influences the activation technology and Biological Research Council
of individual biosynthetic pathways. Analysis and the European Community.
of the Pwb mutations (Sect. 3.3.2.7) has iden-
tified an additional regulatory gene, red2
(GUTHRIE,E. P., FLAXMAN, C. S., WHITE,J.,
HODGSON,D. A., BIBB,M. J., and CHATER, References
K. F., manuscript in preparation), and re-ADAMIDIS, T., CHAMPNESS, W. (1992), Genetic
vealed a pathway-specific regulatory cascade.
analysis of absB, a Streptomyces coelicolor locus
red2 is located approximately 4 kb down- involved in global antibiotic regulation, J. Bacte-
stream of redD and contains a single UUA rial. 114,46224628.
codon. Disruption of red2 results in loss of
ADAMIDIS, T., RIGGLE,P., CHAMPNESS, W. C.
Red production and loss of redD transcrip- (1990), Mutations in a new Streptomyces coelico-
lor locus which globally block antibiotic biosyn-
tion, suggesting that RedZ is a transcriptional
activator of redD (WHITEand BIBB,in press). thesis but not sporulation, J. Bacteriol. 112,
RedZ shows end-to-end similarity to mem- 2962-2969.
ALEX,L. A., SIMON, M. I. (1994), Protein histidine
bers of the response regulator family of pro-
kinases and signal transduction in prokaryotes
teins, and possesses a putative DNA-binding and eukaryotes, Trends Genet. 10,133-138.
a-helix-turn-a-helix motif towards its C-ter-
AN, G., VINING, L. C. (1978), Intracellular levels of
minus but lacks the charged amino acids nor-guanosine 5 '-diphosphate 3 '-diphosphate
mally essential for phosphorylation of a re-(ppGpp) and guanosine 5 '-triphosphate 3 '-di-
phosphate (pppGpp) in cultures of Streptomyces
sponse regulator by its cognate sensory histid-
griseus producing streptomycin, Can. J. Micro-
ine protein kinase. The existence of two path-
biol. 24,502-511.
way-specific regulatory genes for Red produc-
ANISOVA, L. N., BLINOVA, I. N., EFREMENKOVA,
tion in S. coelicolor parallels the situation for
0. V., KOZ'MIN, Yu. P., ONOPRIENKO, V. V.,
daunorubicin synthesis in S. peuceticus, in SMIRNOVA, G. M., KHOKHLOV,A. S. (1984),
which dnrl and dnrN are homologues of redD Regulators of the development of Streptomyces
and red2, respectively. In S. peuceticus, tran-
coelicolor A3(2), Izv. Akad. Nauk SSSR, Ser.
scription of dnrl depends on dnrN (Sect. Biol. 1, 98-108.
3.3.4.4), and DnrN has been shown recentlyANZAI,H., MURAKAMI, T., IMAI,S., SATOH,A.,
to bind to the dnrl promoter region (FURUYA NAGAOKA, K., THOMPSON, C. J. (1987), Tran-
and HUTCHINSON, scriptional regulation of bialaphos biosynthesis
1996). Moreover, DnrI has
also been shown to bind to the promoters ofin Streptomyces hygroscopicus,J. Bacteriol. 169,
daunorubicin biosynthetic structural genes 3482-3488.
ASTURIAS, J. A., LIRAS,P., MARTiN, J. F. ( l m ) ,
(TANGet al., in press), providing the elusive
Phosphate control of pabS gene transcription
evidence that this family of pathway-specific
during candicidin biosynthesis, Gene 93, 79-84.
regulatory genes are indeed likely to act di-
BAINTON, N. J., BYCROFI-, B. W., CHHABRA, S. R.,
rectly as transcriptional activators (Sect. STEAD,P., GLEDHILL, L., HILL,P. J., REES,C.
3.3.4.1). E. D., WINSON, M. K., SALMOND, G. P. C., STE-
WART, G . S. A. B., WILLIAMS, P. (1992a), A
general role for the lux autoinducer in bacterial
Acknowledgements cell signalling: control of antibiotic biosynthesis
in Erwinia, Gene 116, 87-91.
BAINTON,N. J., STEAD,P., CHHABRA, S. R., BY-
We are grateful to WENDY CHAMPNESS, CROFT, B. W., SALMOND, G . P. C., STEWART,
ALAN GROSSMAN, DAVIDHOPWOOD,MI- G. S. A. B., WILLIAMS, P. (1992b), N-(3-Oxo-
CHIKO NAKANO,GEORGE SALMOND,and hexanoy1)-L-homoserine lactone regulates car-
PETERZUBERfor comments on parts of the bapenem antibiotic production in Erwinia caro-
manuscript, and to all those who have al- tovora, Biochem. J. 288,997-1004.
5 References 97
BASCARAN,V., SANCHEZ,L., HARDISSON,C., spp., in: Regulation of Bacterial Differentiation

B R A ~ AA., F. (1991), Stringent response and in- (PIGGOT,P., MORAN,C. P., YOUNGMAN, P.,
itiation of secondary metabolism in Streptomyces Eds.), pp. 61-93. Washington, DC: American
clavuligerus, J. Gen. Microbiol. 137, 1625-1634. Society for Microbiology.
BEPPU,T. (1992), Secondary metabolites as chemi- CHAMPNESS, W. C., RIGGLE,P., ADAMIDIS,T.,
cal signals for cellular differentiation, Gene 115, VANDERVERE, P. (1992), Identification of Strep-
159-1 65. tomyces coelicolor genes involved in regulation
B R A ~ AA., F., M ~ N D E ZC.,, DIAZ, L. A, MANZA- of antibiotic synthesis, Gene 115, 55-60.
NAL,M. B., HARDISSON, C. (1986), Glycogen CHANG,H.-M., SHIEH,Y.-T., BIBB,M. J., CHEN,C.
and trehalose accumulation during colony devel- W. (1996), The cutRS signal transduction system
opment in Streptomyces antibioticus, J. Gen. Mi- of Streptomyces lividans represses the biosynthe-
crobiol. 132, 1319-1326. sis of the polyketide antibiotic actinorhodin,
BRIAN,P., RIGGLE,P. J., SANTOS,R. A., CHAMP- Mol. Microbiol. 21, 1075-1085.
NESS, W. C. (1996), Global negative regulation CHATER,K. F. (1990), The improving prospects for
of Streptomyces coelicolor by an absA-encoded yield increase by genetic engineering in antibiot-
putative signal transduction system, J. Bacteriol. ic-producing streptomycetes, Bio/Technology 8,
178,3221-3231. 115-121.
BROWN,K. L., WOOD,S., BUTTNER,M. J. (1992), CHATER,K. F. (1992), Genetic regulation of sec-
Isolation and characterization of the major vege- ondary metabolic pathways in Streptomyces, in:
tative RNA polymerase of Streptomyces coelico- Secondary Metabolites. Their Function and Evo-
lor A3(2)-renaturation of a sigma-subunit using lution (CHADWICK,D. J., WHELAN,J., Eds.),
GroEL, Mol. Microbiol. 6, 1133-1139. pp. 144-162, Ciba Foundation Symposium 171.
BRUTON,C. J., GUTHRIE,E. P., CHATER,K. F. Chichester: Wiley.
(1991), Phage vectors that allow monitoring of CHATER,K. F., BRUTON,C. J. (1983), Mutational
transcription of secondary metabolism genes in cloning in Streptomyces and the isolation of anti-
Streptomyces, Bio/Technology 9, 652-656. biotic production genes, Gene 26, 67-78.
BUTTNER,M. J. (1989), RNA polymerase hetero- CHATER,K. F., BRUTON,C. J. (1985), Resistance,
geneity in Streptomyces coelicolor A3(2), Mol. regulatory and production genes for the anti-
Microbiol. 3, 1653-1659. biotic methylenomycin are clustered, EMBO J.
CABALLERO, J. L., MALPARTIDA, F., HOPWOOD, 4, 1892-1893.
D. A. (1991), Transcriptional organization and
regulation of an antibiotic export complex in the CHATER,K. F., HOPWOOD,D. A (1993), Strepto-
producing Streptomyces culture, Mol. Gen. Ge- myces, in: Bacillus subtilis and Other Gram-Posi-
net. 228, 372-380. tive Bacteria: Biochemistry, Physiology, and Mo-
lecular Genetics (SONENSHEIN, A. L., HOCH,J.
CAO, J. G., MEIGHEN,E. A. (1989), Purification A., LOSICK,R., Eds.), pp. 83-89. Washington,
and structural identification of an autoinducer D C American Society for Microbiology.
for the luminescence system of Vibrio harveyi, J.
Biol. Chem. 264,21670-21 676. CHATER,K. F., MERRICK,M. J. (1979), Strepto-
CASHEL,M., RUDD,K. E. (1987), The stringent re- myces, in: Developmental Biology of Prokaryotes
sponse, in: Escherichia coli and Salmonella ty- (PARISH,J. H., Ed.), pp. 93-114. Oxford: Black-
phimurium: Cellular and Molecular Biology well.
(NEIDHARDT, F., INGRAHAM, J. L., Low, K. B., CLEWELL,D. B. (1993), Bacterial sex pheromone-
MAGASANIK,B., SCHAECHTER, M., UMBAR- induced plasmid transfer, Cell 73, 9-12.
GER, H. E., Eds.), pp. 1410-1438. Washington, COSMINA,P., RODRIGUEZ,F., DE FERRA, F.,
D C American Society for Microbiology. GRANDI,G., PEREGO,M., VENEMA,G., VAN
CHAKRABURTTY, R. N. (1996), The (p)ppGpp syn- SINDEREN, D. (1993), Sequence and analysis of
thetase gene (relA) of Streptomyces coelicolor, the genetic locus responsible for surfactin syn-
A3(2). Ph D Thesis, University of East Anglia, thesis in Bacillus subtilis, Mol. Microbiol. 8,821-
Nonvich, UK. 831.
CHAKRABURTTY, R., WHITE, J., TAKANO,E., DECKER,H., HUTCHINSON,C. R. (1993), Tran-
BIBB,M. J. (1996), Cloning, characterization and scriptional analysis of the Streptomyces glauces-
disruption of a (p)ppGpp synthetase gene (relA) cens tetracenomycin C biosynthesis gene cluster,
from Streptomyces coelicolor A3(2), Mol. Micro- J. Bacteriol. 175, 3887-3892.
biol. 19, 357-368. DEMAIN, A. L. (1992), Microbial secondary metab-
CHAMPNESS, W. C., CHATER,K. F. (1994), Regula- olism: a new theoretical frontier for academia, a
tion and integration of antibiotic production and new opportunity for industry, in: Secondary Me-
morphological differentiation in Streptomyces tabolites. Their Function and Evolution (CHAD-
WICK,D. J., WHELAN,J., Eds.), pp. 3-23, Ciba SHEVICH, Y. E. (1985), Regulators of differen-
Foundation Symposium 171. Chichester: Wiley. tiation in actinomycetes, Antibiotiki i Meditsins-
DEMAIN, A. L., AHARONOWITZ, Y., MART~N, J.-F. kaya Biotekhnologiya 9, 687-707.
(1983), Metabolic control of secondary biosyn- FERNANDEZ-MORENO, M. A, CABALLERO, J. L.,
thetic pathways, in: Biochemistry and Genetic HOPWOOD,D. A., MALPARTIDA, F. (1991), The
Regulation of Commercially Important Antibio- act cluster contains regulatory and antibiotic ex-
tics (VINING,L. C., Ed.), pp. 49-72. London: port genes, direct targets for translational con-
Addison-Wesley. trol by the bldA transfer RNA gene of Strepto-
DEVEREUX,J., HAEBERLI,P., SMITHIES, 0. myces, Cell 66, 769-780.
(1984), A comprehensive set of sequence analy- FERNANDEZ-MORENO, M. A., MART~N-TRIANA,
sis programs for the VAX, Nucleic Acids Res. 12, A. J., MARTINEZ,E., NIEMI,J., KIESER,H. M.,
387-395. HOPWOOD, D. A., MALPARTIDA,F. (1992),
DfAZ-GUERRA, L., MORENO,F., SANMILLAN,J. abaA, a new pleiotropic regulatory locus for an-
L. (1989), appR gene product activates transcrip- tibiotic production in Streptomyces coelicolor, J.
tion of microcin C7 plasmid genes, J. Bacteriol. Bacteriol. 174, 2958-2967.
171,2906-2908. FLORIANO,B., BIBB,M. J. (1996), afsR is a pleio-
DISTLER,J., MANSOURI,K., MAYER,G., STOCK- tropic but conditionally required regulatory
MA”, M., PIEPERSBERG, w. (1992), Streptomy- gene for antibiotic production in Streptomyces
cin biosynthesis and its regulation in streptomy- coelicolor A3(2), Mol. Microbiol. 21, 385-396.
cetes, Gene 115, 105-111. FROHLICH,K. U., WIEDMANN, M., LOTTSPEICH,
DODD,I. B., EGAN,J. B. (1990), Improved detec- F., MECKE,D. (1989), Substitution of a pentale-
tion of helix-turn-helix DNA-binding motifs in nolactone-sensitive glyceraldehyde-3-phosphate
protein sequences, Nucleic Acids Res. 18, 5019- dehydrogenase by a genetically distinct resistant
5026. isoform accompanies pentalenolactone produc-
DONADIO,S., STASSI,D., MCALPINE,J. B., STAV- tion in Streptomyces arenae, J. Bacteriol. 171,
ER,M. J., SHELDON, P. J., JACKSON, M., SWAN- 6696-6702.
SON,S. J., WENDT-PIENKOWSKI, E., YI-GUANG, FUJII,T., GRAMAJO, H. C., TAKANO, E., BIBB,M.
W., JARVIS,B., HUTCHINSON, C. R., KATZ, L. J. (1996), redD and actII-ORF4, pathway-specif-
(1993), Recent developments in the genetics of ic regulatory genes for antibiotic production in
erythromycin formation, in: Industrial Microor- Streptomyces coelicolor A3(2), are transcribed in
ganisms: Basic and Applied Molecular Genetics vitro by an RNA polymerase holoenzyme con-
(BALTZ,R. H., HEGEMAN, G. D., SKATRUD,P. taining the non-essential u factor, ghrdD,J. Bac-
L., Eds.), pp. 257-265. Washington, DC: Ameri- teriol. 178, 3402-3405.
can Society for Microbiology. FUQUA,W. C., WINANS,S. C., GREENBERG, E. P.
D’SOUZA,C., NAKANO,M. M., CORBELL,N., Zu- (1994), Quorum sensing in bacteria: the LuxR-
BER,P. (1993), Amino-acylation site mutations Lux1 family of cell density-responsive transcrip-
in amino acid-activating domains of surfactin tional regulators, J. Bacteriol. 176, 269-275.
synthetase: effects on surfactin production and FURUYA,K., HUTCHINSON,C. R. (1996), The
competence development in Bacillus subtilis, J. DnrN protein of Streptomyces peuceticus, a pseu-
Bacteriol. 175, 3502-3510. do-response regulator, is a DNA-binding protein
D’SOUZA,C., NAKANO,M. M., ZUBER,P. (1994), involved in the regulation of daunorubicin bio-
Identification of comS, a gene of the srfA ope- synthesis, J. Bacteriol. 178, 6310-6318.
ron that regulates the establishment of genetic GEISTLICH,M., LOSICK,R., TURNER,J. R., RAO,
competence in Bacillus subtilis, Proc. Natl. Acad. R. N. (1992), Characterization of a novel regula-
Sci. USA 91,9397-9401. tory gene governing the expression of a polyke-
DUBNAU, D. (1993), Genetic exchange and homo- tide synthase gene in Streptomyces ambofaciens,
logous recombination, in: Bacillus subtilk and Mol. Microbiol. 6, 2019-2029.
Other Gram-Positive Bacteria: Biochemistry and GENTRY,D. R., HERNANDEZ, V. J., NGUYEN,L.
Molecular Genetics (SONENSHEIN, A. L., HOCH, H., JENSEN,D. B., CASHEL,M. (1993), Synthesis
J. A., LOSICK,R., Eds.), pp. 555-584. Washing- of the stationary-phase sigma factor 2 is posi-
ton, DC: American Society for Microbiology. tively regulated by ppGpp, J. Bacteriol. 175,
EBERHARD, A., BURLINGAME, A. L., KENYON,G. 7982-7989.
L., NEALSON,K. H., OPPENHEIMER,N. J. GERBER,N. N., MCINNES,A. G., SMITH,D. G.,
(1981), Structural identification of autoinducer WALTER,J. A., WRIGHT,J. L. C. (1978), Bio-
of Photobacterium fischeri luciferase, Biochemis- synthesis of prodiginines. 13C resonance assign-
try 20, 2444-2449. ments and enrichment patterns in nonyl-, cyclon-
EFREMENKOVA, 0. V., ANISOVA,L. N., BARTO- onyl-, methylcyclodecyl-, and butylcycloheptyl-
5 References 99
prodiginine produced by actinomycete cultures HAMAGISHI, Y., YOSHIMOTO, A., OKI, T. (1981),
supplemented with 13C-labeledacetate and "N- Determination of guanosine tetraphosphate
labeled nitrate, Can. J. Chem. 56, 1155-1163. (ppGpp) and adenosine pentaphosphate
GRAFE, U., SCHADE,W., ERITT,I., FLECK,W. F. (pppApp) in various microorganisms by ra-
(1982), A new inducer of anthracycline biosyn- dioimmunoassay, Arch. Microbiol. 130, 134-
thesis from Streptomyces viridochromogenes, J. 137.
Antibiot. 35, 1722-1723. HARA,O., BEPPU,T. (1982a), Induction of strepto-
GRAFE,U., REINHARDT, G., SCHADE,W., ERITT, mycin-inactivating enzyme by A-factor in Strep-
I., FLECK,W. F., RADICS,L. (1983), Interspe- tomyces griseus, J. Antibiot. 35, 1208-1215.
cific inducers of cytodifferentiation and anthra- HARA,O., BEPPU,T. (1982b), Mutants blocked in
cycline biosynthesis from Streptomyces bikinien- streptomycin production in Streptomyces griseus
sis and Streptomyces cyaneofuscatus,Biotechnol. - the role of A-factor, J. Antibiot. 35, 349-358.
Lett. 5, 591-596. HARA,O., HORINOUCHI, S., UOZUMI,T., BEPPU,
GRAMAJO, H. C., TAKANO, E., BIBB,M. J. (1993), T. (1983), Genetic analysis of A-factor synthesis
Stationary-phase production of the antibiotic ac- in Streptomyces coelicolor A3(2) and Strepto-
tinorhodin in Streptomyces coelicolor A3(2) is myces griseus, J. Gen. Microbiol. l29, 2939-
transcriptionally regulated, Mol. Microbiol. 7, 2944.
837-845. HEALY,J., WEIR,J., SMITH,I., LOSICK,R. (1991),
GROSS,R., ARICO,B., RAPPUOLI,R. (1989), Fami- Post-transcriptional control of a sporulation reg-
lies of bacterial signal-transducing proteins, Mol. ulatory gene encoding transcription factor # in
Microbiol. 3, 1661-1667. Bacillus subtilis, Mol. Microbiol. 5, 477487.
GROSSMAN, T. H., TUCKMAN, M., ELLESTAD,S., HECKER,M., SCHROETER, A. (1985), Synthese der
OSBURNE, M. S. (1993), Isolation and character- alkalischen Phosphatase in einem stringent und
ization of Bacillus subtilis genes involved in side- relaxed kontrollierten Stamm von Escherichia
rophore biosynthesis: relationship between B. coli nach Aminosauren- und Phosphatlimitation,
subtilis sfp" and Escherichia coli entD genes, J. J. Basic Microbiol. 25, 341-347.
Bacteriol. 175, 6203-6211. HENGGE-ARONIS, R., FISCHER,D. (1992), Identifi-
cation and molecular analysis of glgS, a novel
GUILFOILE, P. G., HUTCHINSON, C. R. (1991), A growth-phase regulated and rpoS-dependent
bacterial analog of the mdr gene of mammalian gene involved in glycogen synthesis in Esche-
tumor cells is present in Streptomyces peucetius, richia coli, Mol. Microbiol. 6, 1877-1886.
the producer of daunorubicin and doxorubicin,
HENNER,D. J., YANG,M., FERRARI,E. (1988),
Proc. Natl. Acad. Sci. USA 88,8553-8557. Localization of Bacillus subtilis sacU(hy) muta-
GUILFOILE, P. G., HUTCHINSON, C. R. (1992a), Se- tions to two linked genes with similarities to the
quence and transcriptional analysis of the Strep- conserved procaryotic family of two-component
tomyces glaucescens tcmAR tetracenomycin-C signalling systems, J. Bacteriol. 170, 5102-5109.
resistance and repressor gene loci, J. Bacteriol. HERNANDEZ, V. J., BREMER,H. (1990), Guano-
174,3651-3658. sine tetraphosphate (ppGpp) dependence of the
GUILFOILE,P. G., HUTCHINSON, C. R. (1992b), growth rate control of rrnA P1 promoter activity
The Streptomyces glaucescens TcmR protein rein Escherichia coli, J. Biol. Chem. 265, 11605-
presses transcription of the divergently oriented 11614.
tcmR and tcmA genes by binding to an intergen- HERNANDEZ, V. J., BREMER, H. (1991), Escherich-
ic operator region, J. Bacteriol. 174, 3659-3666. iu coli ppGpp synthetase I1 activity requires
GUTHRIE,E. P., CHATER,K. F. (1990), The level SPOT,J. Biol. Chem. 266,5991-5999.
of a transcript required for production of a HERNANDEZ, V. J., BREMER,H. (1993), Charac-
Streptomyces coelicolor antibiotic is conditional- terization of RNA and DNA synthesis in Esch-
ly dependent on a transfer RNA gene, J. Bacteri- erichia coli strains devoid of ppGpp, J. Biol.
01. 172,61894193. Chem. 268, 10851-10862.
HAHN,J., DUBNAU, D. (1991), Growth stage signal HIGGINS,C. F. (1992), ABC transporters: from mi-
transduction and the requirements for srfA in- croorganisms to man, Annu. Rev. Cell Biol. 8,
duction in development of competence, J. Bacte- 67-113.
riol. 173, 7275-7282. HIGGINS, C. F., DORMAN, C. J., STIRLING, D. A,
HAMAGISHI, Y., YOSHIMOTO, A., OKI,T., INUI,T. WADDELL,L., BOOTH,I. R., MAY, G., BRE-
(1980), Occurrence of guanosine-5 '-diphos- MER,E. (1988), A physiological role for DNA
phate-3 '-diphosphate and adenosine-5'-triphos- supercoiling in the osmotic regulation of gene
phate-3 '-diphosphate in Streptomyces galilaeus, expression in S. typhimurium and E. coli, Cell
Agric. Biol. Chem. 44, 1003-1007. 52, 569-584.
HOBBS,G., OBANYE, A. I. C., PETTY,J., MASON, HORINOUCHI, S. (1993), “Eukaryotic” signal trans-
J. C., BARRATT,E., GARDNER, D. C. J., FLETT, duction systems in the bacterial genus Strepto-
F., SMITH,C. P., BRODA,P., OLIVER,S. G. myces, Actinomycetologica 7, 68-81.
(1992), An integrated approach to studying reg- HORINOUCHI, S., BEPPU,T. (1992), Autoregulato-
ulation of production of the antibiotic methyle- ry factors and communication in actinomycetes.
nomycin by Streptomyces coelicolor A3(2), J. Annu. Rev. Microbiol. 46,317-398.
Bacteriol. 174, 1481-1494. HORINOUCHI, S., BEPPU,T. (1995), Autoregula-
HOCH,J. A. (1993), sp00 genes, the phosphorelay, tors, in: Genetics and Biochemistry of Antibiotic
and the initiation of sporulation, in: Bacillus sub- Production (VINING,L., STUTTARD,C., Eds.),
tilis and Other Gram-positive Bacteria: Biochem- pp. 103-119, Newton, MA: Butterworth-Heine-
istry, Physiology, and Molecular Genetics (SON- mann.
ENSHEIN, A. L., HOCH,J. A., LOSICK,R., Eds.), HORINOUCHI, S., HARA, O., BEPPU, T. (1983),
pp. 747-755. Washington, D C American Society Cloning of a pleiotropic gene that positively con-
for Microbiology. trols biosynthesis of A-factor, actinorhodin, and
HODGSON, D. A. (1992), Differentiation in actino- prodigiosin in Streptomyces coelicolor A3(2) and
mycetes, in: Prokaryotic Structure and Function. Streptomyces lividans, J. Bacteriol. 155, 1238-
A New Perspective (MOHAN,S., DOW,C., COLE, 1248.
J. A., Eds.), SGM Symposium Vol. 47, pp. 407- HORINOUCHI, S., KUMADA,Y., BEPPU,T. (1984),
440. Cambridge: Society for General Microbiol- Unstable genetic determinant of A-factor bio-
ogy. synthesis in streptomycin-producing organisms:
HOGGARTH, J. H., CUSHING,K. E., MITCHELL, J. Cloning and characterization, J. Bacteriol. 158,
I., RITCHIE,D. A. (1994), Induction of resist- 481-487.
ance to novobiocin in the novobiocin-producing HORINOUCHI, S., SUZUKI,H., BEPPU,T. (1986),
organism Streptomyces niveus, FEMS Microbiol. Nucleotide sequence of afsB, a pleiotropic gene
Lett. 116, 131-136. involved in secondary metabolism in Strepto-
HOLT, T. G., CHANG,C., LAURENTWINTER, C., myces coelicolor A3(2) and “Streptomyces livi-
MURAKAMI, T., GARRELS,J. I., DAVIES,J. E., dam”, J. Bacteriol. 168, 251-269.
THOMPSON, C. J. (1992), Global changes in gene
expression related to antibiotic synthesis in HORINOUCHI, S., MALPARTIDA, F., HOPWOOD, D.
Streptomyces hygroscopicus, Mol. Microbiol. 6, A., BEPPU,T. (1989a), afsB stimulates transcrip-
969-980. tion of the actinorhodin biosynthetic pathway in
Streptomyces coelicolor A3(2) and Streptomyces
HONG,S. K, KITO,M.,BEPPU,T., HORINOUCHI, S. lividans, Mol. Gen. Genet. 215, 355-351.
(1991), Phosphorylation of the AfsR product, a
global regulatory protein for secondary-meta- HORINOUCHI, S., SUZUKI,H., NISHIYAMA, M.,
bolite formation in Streptomyces coelicolor BEPPU,T. (1989b), Nucleotide sequence and
A3(2), J. Bacteriol. 173,2311-2318. transcriptional analysis of the Streptomyces gri-
HONG,S . K., MATSUMOTO, A., HORINOUCHI, S.,
seus gene (afsA) responsible for A-factor bio-
synthesis, J. Bacteriol. 171, 1206-1210.
BEPPU,T. C. (1993), Effect of protein kinase in-
hibitors on in vitro protein phosphorylation and HORINOUCHI, S., KITO, M., NISHIYAMA, M., Fu-
cellular differentiation of Streptomyces griseus, RUYA, K., HONG, S. K., MIYAKE, K., BEPPU, T.
Mol. Gen. Genet. 236,341-354. (1990), Primary structure of AfsR, a global regu-
HOOD,D. W., HEIDSTRA,R., SWOBODA,U. K., latory protein for secondary metabolite forma-
HODGSON,D. A. (1992), Molecular genetic tion in Streptomyces coelicolor A3(2), Gene 95,
analysis of proline and tryptophan biosynthesis 49-56.
in Streptomyces coelicolor A3(2) - interaction HUISMAN, G. W., KOLTER,R. (1994a), Regulation
between primary and secondary metabolism - a of gene expression at the onset of stationary
review, Gene 115,5-12. phase in Escherichia coli, in: Regulation of Bac-
HOPWOOD,D. A., BIBB, M. J., BRUTON,C. J., terial Differentiation (PIGGOT,P., MORAN,C. P.,
CHATER,K. F., FEITELSON,J. S., GIL, J. A. YOUNGMAN, P., Eds.), pp. 21-40. Washington,
(1983), Cloning Streptomyces genes for antibiot- D C American Society for Microbiology.
ic production, Trends Biotechnol. 1, 4248. HUISMAN,G. W., KOLTER, R. (1994b), Sensing
HOPWOOD,D. A., CHATER,K. F., BIBB, M. J. starvation: a homoserine lactone-dependent sig-
(1995), Genetics of antibiotic production in naling pathway in Escherichia coli, Science 265,
Streptomyces coelicolor A3(2), in: Genetics and 537-539.
Biochemistry of Antibiotic Production (VINING, HUTCHINSON, C. R., DECKER,H., MADDURI, K.,
L., STUTTARD,C., Eds.), pp. 65-102. Newton, OTTEN,S. L., TANG,L. (1994), Genetic control
MA: Butterworth-Heinemann. of polyketide biosynthesis in the genus Strepto-
5 References 101
myces, Anton. Leeuwenhoek Int. J. Gen. Micro- subtilis, Proc. Natl. Acad. Sci. USA 91, 5793-
biol. 64, 165-176. 5797.
IRR, J. D. (1972), Control of nucleotide metabolism LACALLE,R. A, TERCERO,J. A., JIMENEZ, A.
and ribosomal ribonucleic acid synthesis during (1992), Cloning of the complete biosynthetic
nutrient starvation of Escherichia coli, J. Bacteri- gene cluster for an aminonucleoside antibiotic,
01. 110,554-561. puromycin, and its regulated expression in het-
ISHIZUKA, H., HORINOUCHI, S., KIESER, H. M., erologous hosts, EMBO J. 11, 785-792.
HOPWOOD, D. A., BEPPU,T. (1992), A putative LANCE,R., HENGGE-ARONIS, R. (1991a), Identifi-
two-component regulatory system involved in cation of a central regulator of stationary-phase
secondary metabolism in Streptomyces spp., J. gene expression in Escherichia coli, Mol. Micro-
Bacteriol. 174, 7585-7594. biol. 5, 49-59.
JONES, G. H. (1985), Regulation of phenoxazinone LANGE,R., HENGGE-ARONIS, R. (1991b), Growth
synthase expression in Streptomyces antibioticus, phase-regulated expression of bolA and mor-
J. Bacteriol. 163, 1215-1221. phology of stationary-phase Escherichia coli
KELEMEN,G. H., ZALACAIN, M., CULEBRAS, E., cells are controlled by the novel sigma factor us,
SENO,E. T., CUNDLIFFE, E. (1994), Transcrip- J. Bacteriol. 173, 4474-4481.
tional attenuation control of the tylosin-resist- LANCE,R., HENGGE-ARONIS, R. (1994), The cel-
ance gene tlrA in Streptomyces fradiae, Mol. Mi- lular concentration of the us subunit of RNA
crobiol. 14, 833-842. polymerase in Escherichia coli is controlled at
KELLY,K. S., OCHI,K., JONES, G. H. (1991), Pleio- the levels of transcription, translation, and pro-
tropic effects of a relC mutation in Streptomyces tein stability, Genes Dev. 8, 1600-1612.
antibioticus, J. Bacteriol. 173, 2297-2300. LAVILLE, J., VOISARD,C., KEEL, C., MAURHO-
KHOKHLOV,A. S. (1982), Low molecular weight FER, M., DBFAGO,G., HAAS,D. (1992), Global
microbial bioregulators of secondary metabol- control in Pseudomonas fluorescens mediating
ism, in: Overproduction of Microbial Products antibiotic synthesis and suppression of black
(KRUMPHANZL, V., SIKYTA,B., VANEK,Z., root rot of tobacco, Proc. Natl. Acad. Sci. USA
Eds.), pp. 97-109. London: Academic Press. 89, 1562-1566.
KHOKHLOV,A. S., TOVAROVA, I. I., BORISOVA, L. LAWLOR,E. J., BAYLIS,H. A., CHATER,K. F.
N., PLINER, S. A., SCHEVCHENKO, L. A., KOR- (1987), Pleiotropic morphological and antibiotic
NITSKAYA, E. Y., IVKINA, N. S., RAPOPORT, I. deficiencies result from mutations in a gene en-
A. (1967), A-factor responsible for the produc- coding a tRNA-like product in Streptomyces
tion of streptomycin by a mutant strain of Acti- coelicolor A3(2), Genes Dev. 1, 1305-1310.
nomyces streptomycini, Dokl. Akad. Nauk SSSR LESKIW, B. K., LAWLOR, E. J., FERNANDEZ-ABA-
177,232-235. LOS, J. M., CHATER, K. F. (1991a), l T A codons
KIESER, H. M., KIESER, T., HOPWOOD,D. A. in some genes prevent their expression in a class
(1992), A combined genetic and physical map of of developmental, antibiotic-negative, Strepto-
the Streptomyces coelicolor A3(2) chromosome, myces mutants, Proc. Natl. Acad. Sci. USA 88,
J. Bacteriol. 174, 5496-5507. 2461-2465.
KILLHAM, K., FIRESTONE, M. K. (1984a), Salt LESKIW, B. K., BIBB,M. J., CHATER, K. F. (1991b),
stress control of intracellular solutes in strepto- The use of a rare codon specifically during de-
mycetes indigenous to saline soils, Appl. Envi- velopment, Mol. Microbiol. 5, 2861-2867.
ron. Microbiol. 47, 310-306. LESKIW, B. K., MAH,R., LAWLOR, E. J., CHATER,
KILLHAM,K., FIRESTONE, M. K. (1984b), Proline K. F. (1993), Accumulation of bldA-specified
transport increases growth efficiency in salt- transfer RNA is temporally regulated in Strepto-
stressed Streptomyces griseus, Appl. Environ. myces coelicolor A3(2), J. Bacteriol. 175, 1995-
Microbiol. 48, 239-241. 2005.
KINASHI,H., SHIMAJI, M., SAKAI,A. (1987), Giant LILLEY,G., CLARK,A. E., LAWRENCE, G. C.
linear plasmids in Streptomyces which code for (1981), Control of the production of cephamycin
antibiotic biosynthesis genes, Nature 328, 454- C and thienamycin by Streptomyces cattleya
456. NRRL 8057, J. Chem. Tech. Biotechnol. 31,127-
KOLTER, R., SIEGELE, D. A., TORMO,A. (1993), 134.
The stationary phase of the bacterial life cycle, LONEITO, M. A., BROWN,K. L., RUDD, K. E.,
Annu. Rev. Microbiol. 47, 855-874. BUTTNER,M. J. (1994), analysis of the Strepto-
KONG, L., DUBNAU,D. (1994), Regulation of com- myces coelicolor 9 gene reveals the existence of
petence-specific gene expression by Mec-me- a subfamily of eubacterial RNA polymerase u
diated protein-protein interaction in Bacillus factors involved in the regulation of extracyto-
plasmic functions, Proc. Natl. Acad. Sci. USA 91, scriptionally during carbon starvation, J. Bacteri-
7573-7577. 01. 175,2143-2149.
LOEWEN,P. c., VON OSSOWSKI, I., SWITALA, J., MCCUE,L. A., KWAK,J., BABCOCK,M. J., KEN-
MULVEY,M. R. (1993), KatF (a")synthesis in DRICK, K. E. (1992), Molecular analysis of spo-
Escherichia coli is subject to posttranscriptional rulation in Streptomyces griseus, Gene 115, 173-
regulation, J. Bacteriol. 175, 2150-2153. 179.
MACNEIL,D. J., OCCI,J. L., GEWAIN, K. M., MAC- MEIGHEN,E. A. (1991), Molecular biology of bac-
NEIL,T., GIBBONS,P. H., RUBY,C. L., DANIS, terial bioluminescence, Microbiol. Rev. 55, 123-
S. J. (1992), Complex organisation of the Strep- 142.
tomyces avermitilis genes encoding the avermec- MENDEZ,C., BRARA,A. F., MANZANAL,M. B.,
tin polyketide synthase, Gene 115, 119-125. HARDISSON, C. (1985), Role of substrate mycel-
MAGNUSON, R., SOLOMON, J., GROSSMAN, A. D. ium in colony development in Streptomyces,
(1994), Biochemical and genetic characterization Can. J. Microbiol. 31, 44-50.
of a competence pheromone from B. subtilis, MIYAKE,K., HORINOUCHI, S., YOSHIDA,M., CHI-
Cell 77,207-216. BA, N., MORI, K., NOGAWA,N., MORIKAWA,
MALPARTIDA, F., HOPWOOD,D. A. (1984), Molec- N., BEPPU,T. (1989), Detection and properties
ular cloning of the whole biosynthetic pathway of A-factor-binding protein from Streptomyces
of a Streptomyces antibiotic and its expression in griseus, J. Bacteriol. 171, 4298-4302.
a heterologous host, Nature 309,462464. MIYAKE,K., KUZUYAMA, T., HORINOUCHI, S.,
MALPARTIDA,F., NIEMI, J., NAVARRETE,R., BEPPU,T. (1990), The A-factor-binding protein
HOPWOOD,D. A. (1990), Cloning and expres- of Streptomyces griseus negatively controls strep-
sion in a heterologous host of the complete set tomycin production and sporulation, J. Bacteriol.
of genes for biosynthesis of the Streptomyces 172,3003-3008.
coelicolor antibiotic undecylprodigiosin, Gene
93, 91-99. MSADEK, T., KUNST,F., RAPOPORT,G. (1994),
MecB of Bacillus subtilis, a member of the ClpC
MARAHIEL,M. A., NAKANO,M. M., ZUBER,P. ATPase family, is a pleiotropic regulator con-
(1993), Regulation of peptide antibiotic produc- trolling competence gene expression and growth
tion in Bacillus, Mol. Microbiol. 7, 631-636. at high temperature, Proc. Natl. Acad. Sci. USA
MARTIN, J. F., GIL, J. A. (1984), Cloning and ex- 91,5788-5792.
pression of antibiotic production genes, Bio/
Technology 2,63-72. MULVEY,M. R., LOEWEN,P. C. (1989), Nucleotide
sequence of katF of Escherichia coli suggests
MARTfN, J. F., LIRAS,P. (1989), Organization and
KafF protein is a novel u transcription factor,
expression of genes involved in the biosynthesis
Nucleic Acids Res. 17, 9979-9991.
of antibiotics and other secondary metabolites,
Annu. Rev. Microbiol. 43,173-206. NAKANO,M. M., ZUBER,P. (1993), Mutational
MARTIN, J. F., MCDANIEL,L. E. (1975), Kinetics analysis of the regulatory region of the srfA ope-
of biosynthesis of polyene macrolide antibiotics ron in Bacillus subtilis, J. Bacteriol. 175, 3188-
in batch cultures: cell maturation time, Biotech- 3191.
nol. Bioeng. 17, 925-938. NAKANO, M. M., XIA,L., ZUBER,P. (1991), Tran-
MARTINEZ-COSTA, 0.H., ARIAS,P., ROMERO,N. scription initiation region of the srfA operon,
M., PARRO,V., MELLADO,R. P., MALPARTIDA, which is controlled by the comP-comA signal
F. (1996), A relAlspoT homologous gene from transduction system in Bacillus subtilis, J. Bacte-
Streptomyces coelicolor A3(2) controls antibiotic riol. 173,5487-5493.
biosynthetic genes, J. Biol. Chem. 271, 10627- NAKANO,M. M., CORBELL,N., BESSON,J., Zu-
10634. BER, P. (1992), Isolation and characterization of
MATSUMOTO, A., HONG,S . 4 , ISHIZUKA, H., Ho- sfp: a gene that functions in the production of
RINOUCHI, S., BEPPU,T. (1994), Phosphoryla- the lipopeptide biosurfactant, surfactin, in Bacil-
tion of the AfsR protein involved in secondary lus subtilis, Mol. Gen. Genet. 232, 313-321.
metabolism in Streptomyces species by a euka- NARVA,K. E., FEITELSON, J. S. (1990), Nucleotide
ryotic-type protein kinase, Gene 146, 47-56. sequence and transcriptional analysis of the
MCCANN, M. P., YIDWELL,J. P., MATIN, A. redD locus of Streptomyces coelicolor A3(2), J.
(1991), The putative u factor KatF has a central Bacteriol. 172, 326-333.
role in development of starvation-mediated gen- NEAL,R. J., CHATER,K. F. (1987), Nucleotide se-
eral resistance in Escherichia coli, J. Bacteriol. quence analysis reveals similarities between pro-
173,4188-4194. teins determining methylenomycin A resistance
MCCANN,M. P., FRALEY,C. D., MATIN,A. (1993), in Streptomyces and tetracycline resistance in eu-
The putative ufactor KatF is regulated posttran- bacteria, Gene 58, 229-241.
5 References 103
NEIDHARDT, F. C., INGRAHAM, J. L., Low, K. B., protein from Streptomyces virginiae, J. Biol.
MAGASANIK, B., SCHAECHTER, M., UMBAR- Chem. 270,12319-12326.
GER,H. E. (1987), Escherichia coli and Salmon- OKAMOTO, S., NIHIRA,T., KATAOKA, H., SUZUKI,
ella typhimurium: Cellular and Molecular Biolo- A., YAMADA, Y. (1992), Purification and molec-
gy. Washington, DC: American Society for Mi- ular cloning of a butyrolactone autoregulator re-
crobiology. ceptor from Streptomyces virginiae, J. Biol.
NGUYEN,L. H., JENSEN,D. B., THOMPSON, N. E., Chem. 267, 1093-1098.
GENTRY,D. L., BURGESS, R. R. (1993), In vitro OLUKOSHI, E. R., PACKTER, N. M. (1994), Impor-
functional characterization of overproduced tance of stored triacylglycerols in Streptomyces -
Escherichia coli katFlrpoS gene product, Bio- possible carbon source for antibiotics, Micro-
chemistry 32, 11 112-11 117. biology 140, 931-943.
NISHINO,T., MURAO,S. (1981), Possible involve- ONAKA,H., ANDO,N., NIHIRA,T., YAMADA,Y.,
ment of plasmid in nucleotide pyrophosphokin- BEPPU, T., HORINOUCHI, H. (1995), Cloning
ase production and the relationship between this and characterization of the A-factor receptor
productivity and cellular accumulation of guano- gene from Streptomyces griseus, J. Bacteriol. 177,
sine tetraphosphate (ppGpp) in Streptomycetes, 6083-6092.
Agric. Biol. Chem. 45, 199-208. POSTMA,P. W., LENGELER, J. W., JACOBSON, G.
NOVOA,M. A., DIAz-GUERRA,SANMILLAN,J. R. (1993), Phosphoenol pyruvate: carbohydrate
L., MORENO,F. (1986), Cloning and mapping of phosphotransferase systems of bacteria, Micro-
the genetic determinants for microcin (27 pro- biol. Rev. 57, 543-594.
duction and immunity, J. Bacteriol. 168, 1384- PU~TIKHUNT, C., OKAMOTO,S., NAKAMURA, T.,
1391. NIHIRA,T., YAMADA, Y. (1993), Distribution in
OCHI,K. (1986), Occurrence of the stringent re- the genus Streptomyces of a homolog to nusG, a
sponse in Streptomyces sp. and its significance gene encoding a transcriptional anti-terminator,
for the inhibition of morphological and physio- FEMS Microbiol. Lett. 110, 243-248.
logical differentiation, J. Gen. Microbiol. 132, RAIBAUD, A., ZALACAIN, M., HOLT, T. G., TI-
2621-2631. ZARD, R., THOMPSON, C. J. (1991), Nucleotide
OCHI,K. (1987). A re1 mutation abolishes the en- sequence analysis reveals linked N-acetyl hy-
zyme induction needed for actinomycin synthe- drolase, thioesterase, transport, and regulatory
sis by Streptomyces antibioticus, Agric. Biol. genes encoded by the bialaphos biosynthetic
Chem. 51,829-835. gene cluster of Streptomyces hygroscopicus, J.
OCHI,K. (1988), Nucleotide pools and stringent re- Bacteriol. 173, 4454-4463.
sponse in regulation of Streptomyces differentia- RETZLAFF, L., MAYER,G., BEYER,S., AHLERT,J.,
tion, in: Biology of Actinomycetes '88 (OKAMI, VERSECK,S., DISTLER,J., PIEPERSBERG, W.
Y., BEPPU,T., OGAWARA, H., Eds.), pp. 330- (1993), Streptomycin production in streptomy-
337. Tokyo: Scientific Societies Press. cetes: a progress report, in: Industrial Microrgan-
OCHI,K. (1990a), Streptomyces relC mutants with isms: Basic and Applied Molecular Genetics
an altered ribosomal protein ST-L11 and genetic (BALTZ,R. H., HEGEMAN, G. D., SKATRUD, P.
analysis of a Streptomyces griseus relC mutant, J. L., Eds.), pp. 183-194. Washington, DC: Ameri-
Bacteriol. 172, 4008-4016. can Society for Microbiology.
OCHI,K. (1990b), A relaxed (rel) mutant of Strep- ROGGIANO, M., DUBNAU,D. (1993), ComA, a
tomyces coelicolor A3(2) with a missing riboso- phosphorylated response regulator protein of
mal protein lacks the ability to accumulate Bacillus subtilis, binds to the promoter region of
ppGpp, A-factor and prodigiosin, J. Gen. Micro- srfA, J. Bacteriol. 175, 3182-3187.
biol. 136, 2405-2412. ROMEO,T., PREISS,J. (1989), Genetic regulation of
OCONNOR,K. A., ZUSMAN, D. R. (1988), Reex- glycogen biosynthesis in Escherichia coli: in vitro
amination of the role of autolysis in the develop- effects of cyclic AMP and guanosine 5'-diphos-
ment of Myxococcus xanthus, J. Bacteriol. 170, phate 3'-diphosphate and analysis of in vivo
41034112. transcripts, J. Bacteriol. 171, 2773-2782.
OHASHI,H., ZHENG,Y. H., NIHIRA, T., YAMADA, ROMEO,T., GONG,M., LIU,M. Y., BRUN-ZINKER-
Y. (1989), Distribution of virginiae butanolides NAGEL,A. M. (1993), Identification and molecu-
in antibiotic-producingactinomycetes, and iden- lar characterization of csrA, a pleiotropic gene
tification of the inducing factor from Strepto- from Escherichia coli that affects glycogen bio-
myces antibioticus as virginiae butanolide A, J. synthesis, gluconeogenesis, cell size, and surface
Antibiot. 42, 1191-1195. properties, J. Bacteriol. 175, 47444755.
OKAMOTO, S., NAKAMURA,K., NIHIRA, T., YA- RUDNER,D. Z., LE DEAUX,J. R., IRETON,K.,
MADA, Y. (1995), Virginiae butanolide binding GROSSMAN, A. D. (1991), The SPOOKlocus of
Bacillus subtilis is homologous to the oligopep- (199O), Signal transduction in bacteria, Nature
tide permease locus and is required for sporula- 344,395-400.
tion and competence, J. Bacteriol. 173, 1388- STRAUCH,E., TAKANO,E., BAYLIS,H. A., BIBB,
1398. M. J. (1991), The stringent response in Strepto-
SAIER,M. H. (1989), Protein phosphorylation and myces coelicolor A3(2), Mol. Microbiol. 5, 289-
allosteric control of inducer exclusion and cata- 298.
bolite repression by the bacterial phosphoenol- STROHL,W. R. (1992), Compilation and analysis of
pyruvate: sugar phosphotransferase system, Mi- DNA sequences associated with apparent Strep-
crobiol. Rev. 53, 109-120. tomyces promoters, Nucleic Acids Res. 20, 961-
SAIER,M. H. (1991), A multiplicity of potential 974.
carbon catabolite repression mechanisms in pro- STUTZMAN-ENGWALL, K. J., OTTEN,S., HUTCHIN-
karyotic and eukaryotic microorganisms, New SON,C. R. (1992), Regulation of secondary me-
Biol. 3, 1137-1147. tabolism in Streptomyces spp. and overproduc-
SARUBBI, E., RUDD, K. E., CASHEL,M. (1988), tion of daunorubicin in Streptomyces peucetius,
Basal ppGpp level adjustment shown by new J. Bacteriol. 174, 144-154.
spoT mutants affect steady state growth rates SWIFT,S., WINSON,M. K., CHAN,P. F., BAINTON,
and rrnA ribosomal promoter regulation in N. J., BIRDSALL, M., REEVES,P. J., REES,C. E.
Escherichia coli, Mol. Gen. Genet. 213, 214-222. D., CHHABRA, S. R., HILL,P. J., THROUP,J. P.,
SCHNIDER, U., KEEL,C., BLUMER,C., TROXLER, BYCROFI-,B. W., SALMOND,G. P. C., WIL-
J., DEFAGO,G., HASS,D. (1995). Amplification LIAMS,P., STEWART, G. S. A. B. (1993), A novel
of the housekeeping sigma factor in Pseudomon- strategy for the isolation of lux1 homologues:
as fluorescens CHAO enhances antibiotic pro- evidence for the widespread distribution of a
duction and improves biocontrol abilities, J. LuxR: Lux1 superfamily in enteric bacteria, Mol.
Bacteriol. 121, 416-421. Microbiol. 10, 511-520.
SCHONER, B., GEISTLICH, M., ROSTECK,P., RAO, TAKANO, E., BIBB,M. J. (1994), The stringent re-
R. N., SENO,E., REYNOLDS, P., Cox, K., BUR- sponse, ppGpp and antibiotic production in
GETT, S., HERSHBERGER, C. (1992), Sequence Streptomyces coelicolor A3(2). Actinomycetolog-
similarity between macrolide-resistance determi- ica 8, 1-10.
nants and ATP-binding transport proteins, Gene TAKANO,E., GRAMAJO,H. C., STRAUCH,E.,
115,93-96. ANDRES,N., WHITE, J., BIBB, M. J. (1992),
Transcriptional regulation of the redD transcrip-
SCHREIBER,G., METZGER,S., AIZENMAN, E., tional activator gene accounts for growth-phase
ROZA, S., CASHEL,M., GLASER,G. (1991), dependent production of the antibiotic undecyl-
Overexpression of the relA gene in Escherichia prodigiosin in Streptomyces coelicolor A3(2),
coli, J. Biol. Chem. 266, 3760-3767. Mol. Microbiol. 6, 2797-2804.
SENO,E. T., BALTZ,R. H. (1989), Structural organ- TAKAYANAGI, Y., TAMAKA, K., TAKAHASHI, H.
ization and regulation of antibiotic biosynthesis (1994), Structure of the 5 ' upstream region and
and resistance genes in actinomycetes, in: Regul- the regulation of the rpoS gene of Escherichia
ation of Secondary Metabolism in Actinomycetes coli, Mol. Gen. Genet. 243, 525-531.
(SHAPIRO,S., Ed.), pp. 2-48. Boca Raton, F L TANAKA, K., TAKAYANAGI, Y., FUJITA,N., ISHI-
CRC Press. HAMA, A., TAKAHASHI, H. (1993), Heterogene-
SIMUTH, J., HUDEC,J., CHAN,H. T., DANYI,O., ity of the principal sigma factor in Escherichia
ZELINKA,J. (1979), The synthesis of highly coli: The rpoS gene product, d8, is a second
phosphorylated nucleotides, RNA and protein principal sigma factor of RNA polymerase in
by Streptomyces aureofaciens,J. Antibiot. 32,53- stationary phase Escherichia coli, Proc. Natl.
58. Acad. Sci. USA 90,3511-3515.
STASTNA,J., MIKULIK, K. (1981), Role of highly TANG,L., GRIMM,A., ZHANG,Y.-X., HUTCHIN-
phosphorylated nucleotides and antibiotics in SON, C. R. (in press), Purification of the DNA-
the development of streptomycetes, in: Proc. 4th binding protein DnrI, a transcriptional factor of
Int. Symp. Actinomycete Biology, Cologne, pp. daunorubicin biosynthesis in Streptomyces peu-
481-486 (SCHAAL,K. P., PULVERER,G. Eds.). ceticus, Mol. Microbiol.
Stuttgart, New York: Gustav Fischer Verlag. THIARA, A. S., CUNDLIFFE, E. (1988), Cloning and
STEIN,D., COHEN,S. N. (1989), A cloned regulato- characterization of a DNA gyrase B gene from
ry gene of Streptomyces lividans can suppress the Streptomyces sphaeroides that confers resistance
pigment deficiency phenotype of different devel- to novobiocin, EMBO J. 7,2255-2259.
opmental mutants, J. Bacteriol. 171, 2258-2261. THIARA,A. S., CUNDLIFFE, E. (1989), Interplay of
STOCK,J. B., STOCK,A. M., MOTTONEN, J. M. novobiocin-resistant and -sensitive DNA gyrase
5 References 105
activities in self-protection of the novobiocin WATERS,B., VUJAKLIJA, D., GOLD, M. R., DAV-
producer, Streptomyces sphaeroides, Gene 81, IES,J. (1994), Protein tyrosine phosphorylation
65-72. in streptomycetes, FEMS Microbiol. Lett. 120,
THIARA, A. S., CUNDLIFFE, E. (1993), Expression 187-190.
and analysis of two gyrB genes from the novo- WEINRAUCH, Y., GUILLEN,N., DUBNAU,D. A.
biocin producer, Streptomyces sphaeroides, Mol. (1989), Sequence and transcription mapping of
Microbiol. 8, 495-506. Bacillus subtilis competence genes comB and
THOMPSON, C. J., SKINNER, R. H., THOMPSON, J., comA, one of which is related to a family of bac-
WARD,J. M., HOPWOOD,D. A., CUNDLIFFE, E. terial regulatory determinants, J. Bacteriol. 171,
(1982), Biochemical characterisation of resist- 5362-5375.
ance determinants cloned from antibiotic-pro- WEINRAUCH,Y., PENCHEV,R., DUBNAU,E.,
ducing streptomycetes, J. Bacteriol. 151, 678- SMITH,I., DUBNAU, D. (1990), A Bacillus subtil-
685. is regulatory gene product for genetic compe-
TSAO,S. W., RUDD,B. A. M., HE, X., CHANG,C., tence and sporulation resembles sensor protein
FLOSS,H. G. (1985), Identification of a red pig- members of the bacterial two-component signal-
ment from Streptomyces coelicolor A3(2) as a transduction systems, Genes Dev. 4, 860-872.
mixture of prodigiosin derivatives, J. Antibiot. WHITE,J., BIBB, M. J. (in press), The bldA-de-
38,128-130. pendence of undecylprodigiosin production in
VAN SINDEREN, D., GALLI,G., COSMINA,P., DE Streptomyces coelicolor A3(2) involves a path-
FERRA,F., WITHOFF,S., VENEMA,G., GRAND], way-specific regulatory cascade, J. Bacteriol.
G. (1993), Characterization of the srfA locus of WIETZORREK,A. (1996), Regulation of antibiotic
Bacillus subtilis: only the valine-activating do- production in Streptomyces coelicolor A3(2)-
main of srfA is involved in the establishment of analysis of the afsB mutant BH5. Ph D Thesis,
genetic competence, Mol. Microbiol. 8, 833- University of East Anglia, Norwich, UK.
841. WILLIAMS, P. (1994), Compromising bacterial com-
VICENTE,M., KUSHNER,S. R., GARRIDO,T., AL- munication skills, J. Pharmacol. 46, 1-10.
DEA,M. (1991), The role of the ‘gearbox’ in the XIAO,H., KALMAN, M., IKEHARA, K., ZEMEL,S.,
transcription of essential genes, Mol. Microbiol. GLASER,G., CASHEL,M. (1991), Residual gua-
5,2085-2091. nosine 3 ‘,5 ‘-bispyrophosphate synthetic activity
VOGTLI,M., CHANG,P.-C., COHEN,S. N. (1994), of relA null mutants can be eliminated by spoT
afsR2 a previously undetected gene encoding a null mutations, J. Biol. Chem. 266,5980-5990.
63-amino-acid protein that stimulates antibiotic YAMADA,Y., SUGAMURA, K., KONDO,K, YA-
production in Streptomyces lividans, Mol. Micro- NAGIMOTO, M., OKADA,H. (1987), The struc-
biol. 14, 643-653. ture of inducing factors for virginiamycin pro-
VOLZ, K. (1993), Structural conservation of the duction in Streptomyces virginiae,J. Antibiot. 40,
CheY superfamily, Biochemistry 32, 11741- 496-504.
11753. ZAMBRANO, M. M., SIEGELE,D. A,, ALMIR6N,
WASSERMAN, H. H., SHAW,C. K, SYKES,R. J. M., TORMO,A., KOLTER,R. (1993), Microbial
(1974), The biosynthesis of metacycloprodigiosin competition: Escherichia coli mutants that take
and undecylprodigiosin, Tetrahedron Lett. 33, over stationary phase cultures, Science 259,
2787-3091. 1757-1760.
Biotechnology
3 Screening of Novel
Receptor-Active Compounds of
Microbial Origin
HARUOTANAKA
SATOSHIOMURA
Tokyo, Japan
1 Introduction 109
2 Assay Methods for Screening of Receptor-Active Compounds 109
2.1 Assays Based on Physiological Activities in Animal Tissues and Cells 109
2.2 Assays Using Radio-Labeled Ligands 110
2.3 Functional Assays Using Recombinant Cells Transformed with a Receptor Gene 110
3 Receptor Antagonists 111
3.1 Antagonists of Low Molecular Weight Ligand Receptors 112
3.1.1 Muscarinic Acteylcholine Receptor Antagonists 112
3.1.2 Dopamine Receptor Antagonists 114
3.1.3 NMDA Receptor Antagonists 114
3.1.4 Leukotriene B4 Receptor Antagonists 114
3.1.5 PAF Receptor Antagonists 114
3.1.6 Fibrinogen Receptor Antagonists 115
3.1.7 Estrogen Receptor Antagonists 115
3.1.8 Androgen Receptor Antagonists 116
3.2 Antagonists of Peptide Ligand Receptors 116
3.2.1 Cholecystokinin Receptor Antagonists 116
3.2.2 Endothelin Receptor Antagonists 116
3.2.3 Substance P Receptor Antagonists 120
3.2.4 ANP Receptor Antagonists 120
3.2.5 Arginine - Vasopressin Receptor Antagonists 121
3.2.6 Oxytocin Receptor Antagonists 121
108 3 Screening of Novel Receptor-Active Compounds of Microbial Origin
3.2.7 Complement C5a Receptor Antagonists 121

3.2.8 Devazepide - A Non-Peptide Peptide Ligand Antagonist under Development for
Medical Use 121
4 Receptor Agonists 122
4.1 Motilides (Macrolides with Motilin Activity) 122
4.2 Other Agonists 123
5 Inhibitors of Virus Receptor Binding - gp12O-CD4 Binding Inhibitors 124
6 Current State and Future Perspectives 127
8 References 128
2 Assay Methods for Screening of Receptor-Active Compounds 109
1 Introduction screened, resulting in the discovery of new

subtances which act on receptors. They pro-
vide not only lead compounds for the devel-
Development of receptor agonists and an- opment of new drugs but also important tools
tagonists as drugs had already begun before for clarifying the receptor functions. For the
their receptors were characterized as sub- study of the individual functions of receptor
stances. These drugs account for a fairly high subtypes, which have been discovered in re-
percentage of the medications currently used. cent years, the identification of substances se-
Most of them are produced by chemical syn- lectively acting on receptor subtypes is
thesis. In recent years, many receptor-active needed now.
compounds of microbial origin have been dis- This paper reviews the general considera-
covered, stimulating efforts for developing tions of receptor-active compounds of micro-
new drugs from these lead compounds. These bial origin, the screening methods, and the
are now providing an important tool for clar- physiological and pharmacological activities
ifying the functions of receptors and the rela- of those discovered to date. In addition, fu-
tionship between receptor function and dis- ture perspectives for this class of compounds
ease. are also discussed.
Receptors are specific proteins, which re-
cognize exogenous signaling molecules or
physical stimuli, and induce cellular re-
sponses. Receptors located in the membrane
or the cytoplasm function as the first window
in the transmission of extracellular informa-
2 Assay Methods for
tion into the cell. Some receptors, e.g., virus Screening of
receptors and the LDL receptor, are involved
in the uptake of components cells and not di- Receptor-Active
rectly in signal transduction.
Early in the 1980s, the nicotinic acetylcho- Compounds
line receptor gene was cloned and its primary
structure was determined (NODAet al., 1983). Although animal experiments are the most
Since then, knowledge regarding the recep- reliable method of assessing pharmacological
tor-constituting proteins has been cultivated, actions, they require much cost and labor and
and cloning of many receptor genes has been are not suitable for the examination of nu-
performed, thus stimulating discussions about merous samples. A common alternative is
the relationship between receptor structure screening using radio-labeled ligands and re-
and function (PEROUTKA,1994). However, ceptor-containing cells or tissue homogen-
before their characterization as substances, ates. Various simple methods of screening
receptors had been recognized as a vague have been devised, some of which are pre-
concept and had contributed greatly to the sented below.
analysis of drug effects and to the develop-
ment of new drugs. A number of derivatives
and analogs of low molecular weight ligands 2.1 Assays Based on Physiological
have been synthesized. Through the study of Activities in Animal Tissues and
the effects and the pharmacological actions of
these substances on receptors, many agonists Cells
and antagonists have been developed and are
currently used as drugs. The Magnus test, using blood vessels, has
In recent years, the development of radio- been employed for the screening of numerous
labeled ligands has simplified receptor bind- samples, although the detection of ligands is
ing experiments using tissue homogenates or not very easy with this technique. FUJIIet al.
cells. With such techniques, many synthetic (1991) assessed the contraction inhibiting ac-
compounds and microbial cultures have been tivity with the following technique: First, an
isolated tracheal specimen was suspended in a commercially available. They can be used for
Magnus bath. Subsequently, bronchial con- experiments on receptor binding in cells, tis-
traction was induced by neurokinin A. Micro- sue homogenates and cell fractions (receptors
bial cultures were then added to assess the for steroid hormones etc. are located in the
contraction inhibitory activity. Using this cytoplasm). The amount of a radio-labeled li-
technique, active substances from 10000 sam- gand bound to the receptor is estimated from
ples were selected. This led to the isolation of the amount of the ligand bound to the recep-
actinomycin D. This compound inhibited the tor in the presence of excess cold ligand to
contraction induced by neurokinin A obtain the amount of specific receptor bind-
(IC5,,= 1.8 x 10-6M), but did not inhibit that ing, which can be used for selecting sub-
by substance P, acetylcholine, etc. (FUJI] et stances which specifically inhibit receptor
al., 1991). binding. The receptor binding inhibitors in-
Methods based on cellular responses to li- clude both agonists and antagonists. Agonists
gands allow an easier examination of many can be distinguished from antagonists by ex-
samples than those using tissues. One exam- amining the influence of receptor binding in-
ple for this is the use of platelets. Platelet ag- hibitors on the physiological actions of li-
gregation is known to be induced by collagen, gands, e.g., using the Magnus method. These
ADP, arachidonic, acid, thrombin, PAF etc. methods are estimated to have been em-
Their antagonists can be isolated by selecting ployed in screening of ligands from microor-
substances which inhibit platelet aggregation ganisms frequently. However, many publica-
(NAKAGAWA, 1992). With such a method, tions lack descriptions of the screening meth-
OKAMOTOet al. (1986a, b) discovered two ods. Following recent success in the cloning of
PAF receptor antagonists, i.e., FR-49175 pro- many receptor genes, it is now possible to ex-
duced by Penicillium terlikowskii and FR- amine the binding of radio-labeled ligands to
900452 produced by Streptomyces phaeofa- recombinant receptors by radioimmunoassay
ciens. LAUERet al. (1991) screened throm- or ELISA.
boxane A2 receptor antagonists taking inhibi-
tion of platelet aggregation as an indicator.
Another cell culture method is based on mac- 2.3 Functional Assays Using
rophage chemotaxis. TSUJIet al. (1992a), e.g., Recombinant Cells Transformed
identified the leukotriene B4 antagonist
WF11605 influencing chemotaxis of polymor- with a Receptor Gene
phonuclear leukocytes.
These methods examine the responses of A new type of assay for ligands using re-
tissues and cells and hence are advantageous combinant microorganisms has been re-
in that they allow a distinction of agonists ported. KING et al. (1990) constructed a re-
from antagonists. However, since all the var- combinant yeast with which ligands of the &-
ious responses of tissues or cells are shown in adrenergic receptors can be assessed by trans-
which many receptors other than the target fecting genes of the human &-adrenergic re-
receptor are included, selectivity may not be ceptor and a G protein a-subunit into the
very high with these methods. However, if the yeast. Saccharomyces cerevisiae possesses the
examiner is experienced and performs careful G protein but lacks the a-subunit which is
observation, the discovery of compounds with necessary for intracellular signal transduction.
novel physiological actions can be expected. Therefore, the human &-adrenergic receptor
gene (hPAR) and the mammalian a-subunit
gene (rat Gsa) were transfected into S. cerevi-
2.2 Assays Using Radio-Labeled siae, resulting in coupling of P and ysub-
Ligands units. Furthermore, a system for the identifi-
cation of &-adrenergic receptor agonists by
In recent years, a number of radio-labeled colorimetry was established by linking the &-
ligands (such as hormones, autacoids, cyto- receptor G protein to the P-galactosidase
kines and neurotransmitters) have become gene (Fig. 1).
3 Receptor Antagonists 11 1
Saccharomyces cerevisiae
Fig. 1. A new screening system for agonists and antagonists of G protein coupled receptors.
Growth
Fig. 2. A new screening system for Agonist +
agonists and antagonists of steroid
receptors. Saccharomyces cerevisiae Agonistt
Antagonist
-
ura3 was transformed with the
URA3 gene by homologous recom-
bination and the URA3 gene was
expressed under control of a ste-
roid-dependent promoter.
MCDONELL constructed another recombi- not be made with binding experiments, is pos-
nant yeast for the assessment of ligands for sible with such systems, they are expected to
steroid hormone receptors, utilizing the simi- provide simpler assay systems if automation
larities of the transcription factors between techniques are incorporated.
yeasts and mammals (ABBOTT, 1991). As ste-
roid hormone receptors are a family of tran-
scription factor, an in vivo transcription sys-
tem could be established using inducible ex-
pression vector system containing steroid hor-
mone receptor genes. The linking of this sys- 3 Receptor Antagonists
tem to the URA3 gene in yeast resulted in the
establishment of a system for measuring sub- Ever since antibiotics were discovered from
stances acting on the steroid hormone recep- microorganisms, microorganisms have been
tors in S. cerevisiae. As shown in Fig. 2, this regarded as a treasure-house of secondary
system uses yeast growth as an indicator. metabolites. They serve as an important
These approaches to assay systems using source of physiologically active substances, in
recombinant microorganisms can also be ap- addition to antibiotics. Plants and marine or-
plied to establish assay systems for many oth- ganisms have also been screened but the per-
er receptor ligands. Since the distinction be- centage of new substances discovered from
tween agonists and antagonists, which could microorganisms is much higher. Receptor-ac-
CI - HO
Ergotamine Muscarine Zealarenone
(Clavi=P Purpurea) (Amanita musearia) (Gibberellazeae)
a-Receptor antagosist Muscarinic acetylcholine Estrogen receptor

receptor agonist agonist
Fig. 3. Traditional receptor agonists and antagonists of microbial origin.
tive substances derived from microorganisms 3.1 Antagonists of Low Molecular

have a long history. It dates back to the dis-
covery of ergot alkaloids (leading to the clini- Weight Ligand Receptors
cal use of ergotamine and ergometrine), mus- (Tab. 1, Fig. 4)
carine (an acetylcholine receptor agonist) and
zealarenone (an estrogen receptor agonist) 3.1.1 Muscarinic Acetylcholine
(Fig. 3), although these substances were not
found by screening for receptor-active com- Receptor Antagonists
pounds. There were more systematic ap-
proaches to discover receptor-active com- TAKESAKO et al. (1988) screened microor-
pounds of microbial origin after the identifi- ganisms for muscarinic acetylcholine receptor
cation of the cholecystokinin antagonist as- antagonists in order to develop anticholiner-
perlicin in 1985. Receptor-active compounds gic agents useful as anti-ulcer agents. Taking
discovered from microorganisms by such ef- guinea pig brain homogenates as receptors
forts are described below. and [3H]quinuclidinyl benzilate as ligands,
Tab. 1. Antagonists of Microbial Origin of Low Molecular Weight Ligand Receptors
Ligand Antagonist Producer Reference

Acetylcholine IJ2702-I and -11 Soil isolate TAKESAKO et al. (1988),
UENOet al. (1988)
Dopamine Sch42029 Actinoplanes sp. HEDGEet al. (1991)
NMDA ES-242-1 to -8 Verticillium sp. TOKIet al. (1992a, b)
Leukotriene Ba WF11605 Fungus TSUJIet al. (1992a),
SHIGEMATSU et al. (1992)
Novobiocin Streptomyces sp. TSUJIet al. (1992b)
PAF FR-49175 Penicillium terlikowskii OKAMOTO et al. (1986a)
FR-900452 S. phaeofaciens OKAMOTO et al. (1986b)
Phomatins (A, B, A,, B,) Phoma sp. SUGANO et al. (1991)
Fibrinogen Tetrafibricin S. neyagawaensis KAMIYAMA et al. (1993a, b)
Estrogen R1128 A, B, C, D Streptomyces sp. HORIet al. (1993a, b, c)
Napiradiomycin A and B1 Streptomyces sp. HORIet al. (1993e)
Androgen 3-Chloro-4-(2-amino-3-chloro-Pseudomonas sp. HORIet al. (1993d)
pheny1)-pyrrole
WS9761 A and B Streptomyces sp. HORIet al. (1993f)
I I
IJ 2702-1 (R: CH f CHCHSH3) 42029 O C b OH FR-49175
IJ 2702-11 (R:CH&H&H&H3)
ES242-1 (R1= H, R2 = OCOCH3)
ES242-2 (R1 , R2 = OCOCb)
ES242-3 (R1= OH, & --OCOCH3)
ES242-4 (R1, R2 =OH)
ES-242-5 ( R1= H, R2 = OH)
S-cb
WF 11605 C b FR-W)0452
Phomatin A Phomatin B Phomalin81 Phomatin82
OH
OH
Tetrafibricin
OH OH OH OH OH
Jqp "W
R 0 OH OH 0 C b
Fig.antagonists
of 4. Structures
of HO OH OH
low molecular n
" &C OH
weight ligand
receptors of R1128A: R = CH&H&& WS9761 A: R = CH3
R1128B: R = CH&H&H&b WS9761 B: R = CH@H
microbial R1128C: R = CH&H&H(CH&
origin. R1128D: R = CH&H&H&H&b
they assayed inhibitory effects on the recep- 3.1.4 Leukotriene B4 Receptor

tor ligand binding of cultures of soil isolat-
ed actinomycetes, fungi and bacteria. The Antagonists
cultures showing inhibition were examined
for anticholinergic activity, using isolated gui- Leukotriene B, (LTB4) is an autacoid
nea pig ileum. In this way, two new com- which promotes aggregation, degranulation
pounds, IJ2702-I and IJ2702-I1 (IC50=0.3 and chemotaxis of polymorphonuclear leuko-
resp. 0.6 pg/mL), were isolated from the cul- cytes. LTB4 is thought to be involved in in-
ture of an actinomycete strain, I52702 (TAKE- flammatory reactions. TSUJIet al. (1992a) ex-
SAKO et al., 1988; UENOet al., 1988). amined microorganisms for substances that
inhibit the LTBCinduced chemotaxis of rat
polymorphonuclear leukocytes, leading to the
3.1.2 Dopamine Receptor discovery of WF11605 produced by a fungus.
WF11605 is a new compound with a triter-
Antagonists pene glucoside structure (SHIGEMATSU et al.,
1992). WF11605 was found to inhibit not only
HEDGEet al. (1991) searched for dopamine chemotaxis (IC50=1.7 x lo-' M) but also the
D1 receptor-active compounds of microbial binding of [3H]LTB4 to the membrane frac-
origin, using [3H]Sch23390 and rat striatum. tion of polymorphonuclear leukocytes
-'
In their study, Sch42029 (2,5-dihydroxyace- (IC5,, = 5.6 x 10 M) and the LTBCinduced
toanilide) was found to be a D1 receptor-spe- degranulation of polymorphonuclear leuko-
cific ligand. Many of the drugs for the treat- cytes (ICS0=3.0x lo-' M). These results in-
ment of Parkinson's disease, which is related dicate that WF11605 is an LTB4 receptor an-
to abnormal dopamine metabolism in the tagonist. Its LD,,, in mice was 1.0 g/kg or
brain, are D2 receptor-specific. Sch42029 is more (i.p.).
the first natural substance specific to the D1 During screening for LTB4 antagonists,
receptor (Ki =0.6 pM). TSUJIet al. (1992b) recently found that novo-
biocin, an antibiotic in clinical use, acts as an
antagonist of LTB4 receptors. This com-
3.1.3 NMDA Receptor pound inhibited the binding of [3H]LTB4 to
the membrane fraction of polymorphonuclear
Antagonists leukocytes (ICs0= 1.0 x 10 -'M). Since leuko-
trienes are known to be involved in the onset
TOKIet al. (1992a, b) carried out screening of ear edema in mice, the investigators exam-
work with [3H]TCP [l-(l-(2-thienyl)cyclohex- ined the effect of novobiocin on ear edema in
ylpiperidine] and a rat brain membrane frac- mice induced by arachidonic acid, and found
tion. They identified new compounds, ES- that local treatment with this antibiotic
242-1 through ES-242-8, which are produced (0.1 pg or more per ear) suppressed the for-
by Verticilliurnsp. and serve as NMDA recep- mation of edema. The compound was effec-
tor antagonists. These compounds inhibit the tive even when administered orally
binding of [3H]TCP to the synaptic mem- (ED50 =220 pg/kg).
brane (IC50:0.1 pM for ES-242-1) but not the
binding of [3H]kainic acid. Although MK801
and ketamine are also known as synthetic
compounds of this type, the ES-242 series are 3.1.5 PAF Receptor Antagonists
the first new compounds of microbial origin
(TOKIet al., 1992d). They have recently been Platelet activating factor (PAF) is produced
used in experiments to clarify the pharmaco- by various cells and tissues. Even very small
logical actions at the molecular level of amounts of PAF exert various biological ac-
NMDA receptors. tions. It is a new physiologically active sub-
stance involved in different diseases such as
bronchial asthma, inflammatory reactions,
renal disease, collagen disease, and anaphy- conformational changes within the molecule.
laxis. Therefore, PAF antagonists are ex- Fibrinogen binding to the receptors on the
pected not only to clarify the physiological ac- surface of platelets is a prerequisite for plate-
tions and pathophysiological roles of PAF but let aggregation. Thus, fibrinogen receptor an-
also to provide an effective therapeutic agent tagonism is a good target for a platelet aggre-
for the treatment of these diseases. gation inhibitor.
OKAMOTOet al. (1986a, b) examined mi- In the course of their screening program
croorganisms for PAF antagonists using inhi- for fibrinogen binding antagonists, KAMIYA-
bition of PAF-induced platelet aggregation as MA et al. (1993a, b) isolated a non-peptide an-
an indicator. They found FR-49175 and FR- tagonist, tetrafibricin, from the culture broth
900452; FR-49175 was identified as bisdes- of an actinomycete. Tetrafibricin strongly in-
thiobis(methylthio)gliotoxin, while FR- hibited the binding of fibrinogen to its recep-
900452 was a new compound. The ICSOof the tors with an ICSOof 46nM. It also inhibited
platelet aggregation inhibiting effect of FR- ADP-, collagen-, and thrombin-induced ag-
49175 was 8.5 pM. Intravenous injection gregation of human platelets with an ICSOof
(0.1 mglkg) to guinea pigs inhibited PAF-in- 5.6 pM, 11.0 pM, and 7.6 pM, respectively.
duced bronchial stenosis (OKAMOTOet al., Tetrafibricin is a novel non-peptide anta-
1986a) FR-900452 is a compound with a gonist of the fibrinogen receptor.
unique structure, including piperidine and
indolinone. It was able to inhibit PAF-
induced rabbit platelet aggregation 3.1.7 Estrogen Receptor
(ICS0=3.7 x lo-’ M), while its inhibitory ef-
fect on platelet aggregation induced by col- Antagonists
lagen, arachidonic acid or ADP was much
weaker. The compound markedly suppressed Non-steroidal estrogen receptor antago-
PAF-induced bronchial stenosis, hypotension nists, e.g., tamoxifen, have been used success-
and elevation in vascular permeability in gui- fully in the therapy of advanced breast can-
nea pigs when it was administered intrave- cer, especially estrogen receptor positive
nously, even in low doses below 10 pg/kg breast cancer. Although this therapy results
(OKAMOTO et al., 1986b). in remarkable improvements for breast can-
SUGANOet al. (1991) examined marine mi- cer patients, the development of tamoxifen
croorganisms for secondary metabolites and resistance frequently occurs and most patients
isolated the phomatins A, B, B,, and B2 from eventually relapse. One potential method to
Phoma sp., a species of Fungi Zmperfecfi liv- overcome the resistance is the use of estrogen
ing upon crabshells. These four compounds receptor antagonists with a new chemical
inhibited PAF-induced platelet aggregation, structure different from tamoxifen and re-
with an ICSOof 1 . 0 ~ l O - ~ M1 ,. 7 ~ 1 O - ~ M , lated compounds, containing the triphenyl
9.8 x M, and 1.6 x M, respectively. ethylene moiety.
Based on such considerations HORIet al.
screened microbial products for new non-ste-
3.1.6 Fibrinogen Receptor roidal estrogen receptor antagonists without
the triphenyl ethylene moiety. They found
Antagonists new non-steroidal estrogen receptor antago-
nists - R1128 A, B, C, and D - from the cul-
Platelet aggregation plays a key role in nor- ture broth of Streptomyces sp. No. 1128
mal hemostasis and thrombosis. Platelets first (HORIet al., 1993a, b, c). These compounds
adhere and spread onto the thrombogenic inhibited estrogen binding to its receptor. The
components of the vascular subendothelium ICSOvalues of R1128 A, B, C, and D for
at the sites of vascular lesions. When stimu- partially purified rat uterine cytosol re-
lated by an agonist, such as ADP, collagen or ceptor were 1.1 x M, 1.2 x lo-’ M,
thrombin, the fibrinogen receptors acquire 2.6 x l o p 7M, and 2.7 x M, respectively.
the ability to bind fibrinogen through some R1128B was a competitive inhibitor of es-
trogen receptor binding and inhibited the 3.2 Antagonists of Peptide Ligand
growth of estrogen-responsive human mam-
mary adenocarcinoma MCF-7 cells in soft Receptors
agar. This inhibition was reversed by addition (Tab. 2, Fig. 5)
of estradiol to the culture medium. R1128B
showed antitumor activities against MCF-7 3.2.1 Cholecystokinin Receptor
when xenografted to nude mice by implanta-
tion into the subrenal capsule of mice (SRC Antagonists
assay). The potency of R1128 B was about 8-
fold lower than that of tamoxifen both in v i m Cholecystokinin (CCK) is a digestive hor-
and in vivo (HORIet al., 1993~).A recent mone which promotes lipid degradation and
study by HORIet al. (1993d) revealed that na- absorption by stimulating gallbladder con-
piradiomycin A and B1, which have been traction, pancreatic juice secretion and small
known to possess antimicrobial activities, bowel motility. Its involvement in the central
are estrogen receptor antagonists regulation of appetite and pain has recently
(ICS0=4.2x M and 3.5 x M, re- been noted. Known CCK receptors include
spectively. CCK-A, primarily located in the periphery,
and CCK-B, primarily located centrally. After
the discovery of the CCK-A antagonists as-
3.1.8 Androgen Receptor perlicins (CHANGet al., 1985; GOETZ et al.,
1985, 1988; LIESCHet al., 1985, 1988), the
Antagonists CCK-B antagonist tetronothiodin (ICSO
against the CCK-B receptor = 3.6nM) was
Androgen plays an important role in the found (OHTSUKAet al., 1992, 1993a, b; WA-
prostatic growth including benign prostatic TANABE et al., 1993). Both compounds are
hyperplasia and prostate cancer. Androgen non-peptide antagonists. Using asperlicin as a
actions are thought to be mediated through lead compound, devazepide was synthesized
binding to its own receptor. Therefore, an- (GOETZet al., 1985; EVANS et al., 1986) and
drogen receptor antagonists can be used in is under development now as an oral agent
the treatment for androgen-responsive dis- for the treatment of pancreatitis etc., as men-
eases. tioned below (see Sect. 3.2.8). Recently, an-
During the course of search for non-steroi- thramycin (KUBOTAet al., 1989) and virginia-
dal androgen receptor binding inhibitors, mycin M1 (LAMet al., 1991) were found to be
HORIet al. (1993d) found that 3-chloro-4-(2- CCK-B antagonists. Anthramycin has a ben-
amino-3-chlorophenyl)-pyrrole (WB2838), a zodiazepin moiety like asperlicin, but it binds
known antifungal antibiotic, is a non-steroidal to the CCK-B receptor unlike asperlicin.
androgen receptor antagonist. More recently,
HORIet al. (1993f) discovered the novel an-
drogen receptor antagonists WS9761 A and 3.2.2 Endothelin Receptor
B. WS9761 A and B inhibited androgen re-
ceptor binding with ICs0 values of Antagonists
8.6 x M and 4.5 x M, respectively,
and showed weak inhibitory activity against Endothelin (ET) was discovered in 1988 as
estrogen receptor binding. a new peptide with potent activity to induce
vascular contraction. During the subsequent
five years, three isopeptides of ET (ET-1, -2,
and -3) were found and there are at least two
receptors (ETA and ETB) for ET. Studies of
the agonists and antagonists of ET have also
been carried out. Following the discovery of
cyclic peptide antagonists of microbial origin
(BE-18257A and B) (IHARAet al., 1991; Ko-
Tab. 2. Antagonists of Microbial Origin of Peptide Ligand Receptors
Ligand Subtype Antagonist Producer Reference
Cholecystokinin A Asperlicin A, B, C, Aspergillus alliaceus CHANGet al. (1985),

D, E" GOETZet al. (1985,
1988), LIESCHet al.
(1985, 1988)
B Anthramycin" S. spadicogriceus KUBOTAet al. (1989)
B Tetronothiodin a Streptomyces sp. OHTSUKA et al.
(l992,1993a, b),
WATANABE (1993)
Gastrinkhole- B Verginiamycin MI S. olivaceus LAMet al. (1991)
cystokinin
L-156586, L-156587,
L-156588, L-156906
Endothelin A BE-18257 A and B S. misakiensis IHARAet al. (1991),
KOJIRIet al. (1991),
NAKAJIMA et al.
(1991)
A WS-7338 C and D Streptomyces sp. MIYATAet al.
(1992a, b, c)
A WS-OOO9 A and B" Streptomyces sp. MIYATAet al.
(1992d, e)
A and B Cochinmicin I, 11, 111 Microbispora sp. LAMet al. (1992),
ZINKet al. (1992)
A Asterric acid" Aspergillus sp. OHASHIet al. (1992)
Substance P NK-1 and 2 WS-9326 A S. violaceusniger HAYASHIet al.
(1992)
NK-1 Actinomycin D Streptomyces sp. FUJIIet al. (1991)
NK-1 Fiscalin A, B, C" Neosartorya fischeri WONGet al. (1993b)
NK-1 Anthrotainin" Gliocladium catenulatum WONGet al. (1993a)
NK-1 and 2 WIN 64821" Aspergillus sp. SEDLOCKet al.
(1994)
ANP Anantin S. coerulescens WEBERet al. (1991),
WYSSet al. (1991)
HS-142-1" Aureobasidium pullulans MORISHITA et al.
(1991a, b)
Arginine-vaso- Hapalindolinone A Fischerella sp. SCHWARTZ et al.
pressin and B" (1987)
Oxytocin L-156373 S. silvensis PETTIBONEet al.
(1989)
C5a L-156602 = Streptomyces sp. HENSENSet al.
PD 124966 (1991), HURLEYet
al. (1986)
" Nonpeptide antagonist.
JIRIet al., 1991; NAKAJIMAet al., 1991) and ric acid (OHASHIet al., 1992) have been iden-
WS7338C and D (MIYATAet al., 1992a,b, c), tified. The cyclic peptide BE-18257B exhi-
non-peptide antagonists such as WS009A and bited IC,, values of 1.4 and 0.8 mM against
B (MIYATA et al., 1992e, d), cochinmicins [ '251]ET-1 binding to aortic smooth muscle
(LAMet al., 1992; ZINKet al., 1992) and aster- tissue and to ventricle membranes from pig,
HO
HN
O=d\ I
,c*O
H, NH
0
Telronothiodin BE-18%7A (R : H)
BE-182578 (R: C w
Cochinmidn X *
OH III H
CI s
R SCH&HNHCOCH3
111 CI R WSOOSA ( R H ) 'COOH
HO WS OOSB (R: OH)
-
GIG+
g1+6
Anantin
HS142-1
(cap),
~0OH
G I Cn4 -,30
m=5-15
Glc; Dglucose, Cap; capronic add

$y H
Hapalindolinone A (R: Cl)
Hapalindolinone B (R: H)
Asperlidn
L-156373
Fiscalin A Fiscalin B Fiscalin C
Anthrotainin
WIN 64821
Fig. 5. Structures of antagonists of peptide ligand receptors of microbial origin.
which are ETA-rich tissues. It did not inhibit tor antagonist (IHARAet al., 1991). On the
['251]ET-1binding to ETB-rich tissues. In iso- other hand, the peptolide cochinmicin 1 is a
lated rabbit iliac arteries, BE-18257B antag- nonselective antagonist for ETA and ETB
onized ET-1-induced vasoconstriction. Thus, sites (LAMet al., 1992). Studies using these
itwas found that BE-18257B is an ETArecep- antagonists are expected to clarify the physi-
ological and pathophysiological significance 3.2.4 ANP Receptor Antagonists

of ET isopeptides and their receptors and to
provide new therapeutic agents. Atrial natriuretic peptide (ANP) is a pep-
tide hormone involved in the regulation of
body water, electrolytes and blood pressure.
3.2.3 Substance P Receptor It has potent diuretic and vasoconstrictive ac-
tions. ANP and two other peptides with dif-
Antagonists ferent amino acid sequences (RNP and CNP)
constitute a natriuretic peptide family. Two
Neuropeptides such as substance P and types of ANP receptors are known, one carry-
neurokinin A, which markedly induce airway ing a guanylate cyclase domain inside the
constriction and promote mucosal secretion, cells and another without any guanylate cy-
have been investigated because of their rela- clase domain. The receptors with a guanylate
tionship to respiratory diseases such as asth- cyclase domain may be involved in the ANP-
ma. HAYASHIet al. (1992) searched for inhi- induced elevation of the intracellular cGMP
bitors of [3H] substance P binding to guinea level, while the receptor without this domain
pig lung membrane fractions and discovered is thought to be involved in the ANP metab-
WS9326A, a cyclic depsipeptide produced by olism. However, the exact physiological roles
an actinomycete. WS9326A exhibits an ICSO of ANP and its pathophysiological signifi-
value of 3.6 x M in the above assay and cance have not yet been fully clarified. For
acts as a tachykinin antagonist in various this reason, the development of ANP recep-
functional assays. Its tetrahydro derivative, tor antagonists is needed.
FK224, was more potent than WS9326A WEBER et al. (1991) examined microbial
(ICSO=1.0 x lo-' M) and antagonizes both metabolites for those substances which inhibit
the neurokinin 1receptor (involved in airway the binding of [ '251]-labeled rat ANP, i.e.,
edema) and the neurokinin 2 receptor (in- [ '251]rANP to the bovine adrenocortical
volved in airway constriction) (HASHIMOTO membrane. They discovered the peptide an-
et al., 1992). This compound is under devel- tagonist anantin produced by an actinomy-
opment for clinical use. In addition, non-pep- cete. This is a cyclic peptide composed of 17
tide inhibitors, termed fiscalins, with moder- amino acids (WYSSet al., 1991). The com-
ate neurokinin 1 binding activity have also pound was suggested to be an ANP receptor
been reported (WONG et al., 1993b). There antagonist because it inhibited both the bind-
has also been a report of a tetracyclic coming of [ lZI]rANP to receptors (ICSO=1.0 p,M)
pound, anthrotainin, with neurokinin 1 activi- and the ANP-induced intracellular cGMP ac-
ty (WONGet al., 1993a), but this compound cumulation in bovine aortic smooth muscle
was found to be a noncompetitive substance cells, and it did not show any agonist effect
P antagonist. (WEBERet al., 1991).
More recently, WIN64821, a non-peptide MORISHITAet al. (1991a, b) examined mi-
secondary metabolite produced by Aspergil- crobial cultures for substances inhibiting the
lus sp. was found to inhibit radiolabeled sub- binding of ['"IIrANP to the rabbit renal cor-
stance P binding in a variety of tissues with tical membrane. They isolated HS-142-1 from
Ki values ranging from 0.24 p,M in human the culture broth of a fungus belonging to the
astrocytoma U-373 MG cells to 7.89 FM genus Aureobasidiurn. HS-142-1 is a new po-
in submaxillary membranes. Additionally, lysaccharide composed of a linear pl,6-glu-
WIN64821 was found to inhibit [ '2SI]-neuro- a x e chain conjugated to caproic acid. In the
kinin A binding to the neurokinin 2 receptor target tissues and cells of ANP, i.e., in the bo-
in human tissue at a concentration equivalent vine adrenocortical membrane fraction (Mo-
to its neurokinin 1 activity (0.26 p,M). RISHITA et al., 1992; ODAet al., 1992), bovine
WIN64821 was shown to be a functional an- vascular smooth muscle cells (IMURAet al.,
tagonist of neurokinin 1 and neurokinin 2 re- 1992), rat renal glomerulus (SANO et al.,
ceptors (OLEYNEKet al., 1994). 1992a, b, c), LLC-Pkl cells (TANAKAet al.,
1992), and PC12 cells (TOKIet al., 1992c), this
compound specifically antagonized the bind- stance from an actinomycete (L-156373)

ing of [ 1251]rANPto ANP receptors contain- which inhibited the binding of [3H]oxytocin
ing guanylate cyclase (ICSOfor rabbit renal to the rat uterine membrane fraction. The Ki
cortical membranes = 0.3 p,g/mL), and inhi- value of this compound was 150 p,M. Its affin-
bited the ANP-induced elevation in cGMP ity for oxytocin receptors was more than 20
level. Although anesthetized rats treated in- times higher than that for the arginine-vaso-
travenously with HS-142-1 alone showed no pressin receptors (AVP-V, and AVP-V2). Its
reaction, the diuretic response of the animals derivative L-365209, produced by dehydroxy-
to exogenous or endogenous ANP was not lation of the N-hydroxyleucine unit and oxi-
seen after pretreatment with HS-142-1 (SANO dation of the piperazic acid residues of the L-
et al., 1992b, c). HS-142-1 thus seems to be a 156373, was 20 times as potent as L-156373
new non-peptide ANP antagonists useful for and had a K i of 7.3 p,M. This derivative was
the analysis of the physiological and patho- highly selective and antagonized the oxytocin
physiological role of ANP. In the future, this action to the rat uterus (IDS0=460 p,g/kg).
compound often will be used in cardiovascu-
lar studies.
3.2.7 Complement C5a Receptor
Antagonists
3.2.5 Arginine-Vasopressin
Receptor Antagonists C5a is thought to be involved in the aggra-
vation of various inflammatory allergic dis-
Arginine-vasopressin (AVP) is a peptide eases. HENSENSet al. (1991) isolated a sub-
composed of 9 amino acids. It is a hormone stance from actinomycete metabolites (L-
secreted from the posterior lobe of the pitui- 156602) that inhibited the binding of human
tary gland and possesses antidiuretic and hy- polymorphonuclear leukocytes. This com-
pertensive properties. The renal AVP-V2 re- pound was considered to be identical to PD-
ceptor is involved in the antidiuretic, the 124966 which had been discovered as an anti-
AVP-Al receptor of the cardiac smooth mus- tumor antibiotic. The structure of this com-
cle in the hypertensive action. pound shown in Fig. 5, was proposed by HEN-
SCHWARTZ et al. (1987) isolated substances SENS et al. (1991) and has never been deter-
from a microorganism of the genus Fischerel- mined.
la that inhibited the binding of [3H]AVP to
the renal tissue containing V2 receptors. They
called the substances hapalindolinone A and 3.2.8 Devazepide- A Non-Peptide
B. These compounds are non-peptide antag- Peptide Ligand Antagonist under
onists which carry cyclopropane and indoli-
none skeletons. They inhibit not only the Development for Medical Use
binding of [3H]AVP to renal tissue =
37.5 p,M) but also the AVP-induced activa- In 1985, CHANGet al. (1985) isolated the
tion of adenylate cyclase (ICSO= 44.6 p,M). CCK receptor antagonist asperlicin from a
microorganism (Aspergillus alliaceus), which
is the first non-peptide antagonist of peptide
3.2.6 Oxytocin Receptor ligand receptors. Asperlicin does not act on
the CCK-B receptors primarily located in the
Antagonists center of the body, but selectively acts on the
CCK-A receptors mainly located in the pe-
Oxytocin is a peptide composed of 9 amino riphery. Its discovery confirmed the presence
acids. It is a hormone secreted from the pos- of CCK receptor subtypes and made it possi-
terior lobe of the pituitary gland and induces ble to distinguish receptor A from receptor B.
uterine contraction and milk secretion. CCK is known to be involved in diseases such
PEITIBONE et al. (1989) isolated a sub- as pancreatitis. Asperlicin was initially ex-
Diare-pm
Fig. 6. Structures and activities of the CCK-A antagonist asperlicin and its analog derazepide.
pected to be useful as a therapeutic agent,

but, because of low solubility, it is ineffective
4 Receptor Agonists
when administered orally.
In 1983, KUBOTAet al. (1983, 1985) found 4.1 Motilides (Macrolides with
in an experiment with peripheral tissue and Motilin Activity)
brain that diazepam, %ananti-anxiety drug
which can be administered orally, antagon- ITOH et al. (1985) found that the side effect
ized CCK. In in v i m binding experiments, of erythromycin causing diarrhea etc., is simi-
however, it did not antagonize CCK (GOETZ lar to the effect of motilin which is a hormone
et al., 1985). Since asperlicin has a benzodia- promoting gastrointestinal motility. Subse-
zepine skeleton like diazepam, efforts have quently, OMURA and coworkers (OMURAet
been started to synthesize highly soluble, al., 1987; TSUZUKI et al., 1989; SUNAZUKA et
orally applicable derivatives. EVANSet al. al., 1989) synthesized a number of erythromy-
(1986) synthesized a number of derivatives cin derivatives which exert only motilin activi-
and analogs containing benzodiazepine and ty and no antibacterial activity. EM536, one
indole because the asperlicin molecule con- of these derivatives, showed 2890 times high-
tains benzodiazepine and L-tryptophan. De- er motilin activity than erythromycin A. Qua-
vazepide which contains D-tryptophan, carry- ternary ammonium derivatives of the deso-
ing a 2-indolyl bond to benzodiazepine as samine moiety of 6,9-hemiketal erythromycin
shown in Fig. 6, is more than loo0 times more A possess the most potent gastrointestinal
potent than asperlicin. It is highly soluble in motor stimulating activity. However, these
water while retaining selective activity. Thus ionized derivatives have low permeability.
devazepide was selected as an excellent can- Consequently, EM523 (18 times as potent as
didate for development (EVANSet al., 1986). erythromycin A) and EM574 (248 times as
At present, devazepide is under development potent) were selected from the tertiary am-
for treatment of acute pancreatitis, biliary monium derivatives as candidates for medical
colic, abdominal pain, and anorexia (EVANS, use (Fig. 7, Tab. 3). At present, these two der-
1989). ivatives (EM523 and EM574) are under de-
4 Receptor Agonists 123
HO
0 OH
Erythromycin A (EMA) EM201 R = C &

EM523 R=CHfi&
EM574 R = CH(CH&
X-
HO HO
0 OH 0 OH
EM485 R=CH3 EM502 R=CH3

EM491 R=CH&H3 EM506 R = CH&H=C&
EM511 R = CH&H=C& EM507 R = CH&ICH
EM536 R = CH&&H
Fig. 7. Structures of erythromycin A and its derivatives.
velopment to be used as gastrointestinal mo- 4.2 Other Agonists

tility regulators when administered either by
injection or orally. The macrolides exerting The traditional receptor agonists muscarine
motilin activity were called “motilides” by and zearalenone (see Fig. 3) are well known
TSUZUKI et al. (1989). KONDO et al. (1988) to be isolated from microorganisms. Musca-
demonstrated that motilides are agonists of rine is an alkaloid from the red variety of
the motilin receptor. Because motilin is a Amanita muscaria, a poisonous mushroom
peptide hormone composed of 22 amino (KUEHL et al., 1955; KOGL et al., 1957). It is
acids, motilin itself is ineffective when admin- an acetylcholine receptor agonist and used as
istered orally while the abovementioned mo- a important biochemical reagent.
tilide EM574 (a non-peptide agonist) is effec- Zearalenone is an estrogen receptor ago-
tive. At present, morphine, an enkephalin nist isolated from the mycelia of the fungus
agonist, is the only non-peptide agonist of Gibberella zeae (Fusarium graminearum)
peptide ligand receptors in clinical use, motil- (STOBet al., 1962; URRYet al., 1966).
ide EM374 is expected to be used as the sec- No agonists of peptide ligand receptors of
ond non-peptide agonist. microbial origin, except erythromycin A (see
Tab. 3. Antimicrobial Activities and Gastrointestinal Motor Stimulating Activities of Erythromycin A and
its Derivatives
Compound Antimicrobial Activity (MIC, pg/mL) Gastrointestinal Motor Stimulating

Activity (relative activity)
SA BS BC EC KP
EMA 0.2 0.1 0.1 12.5 6.25 1

EM201 50 25 25 >loo >lo0 10
EM523 >loo >loo >loo >loo >loo 18
EM511 loo >loo >loo >loo >loo 256
EM536 100 100 loo >loo >loo 2890
EM506 100 100 loo >loo >loo 115
SA Staphylococcus aureus ATCC6358P BS Bacillus subtilis ATCC6633; BC Bacillus cereus IF03001; EC

Escherichia coli NIHJ; KP Klebsiella pneumoniae ATCC10031.
Sect. 4.1), have been reported to date. Their In the screening program for new inhibitors
discovery is desired for the development of of gp120-CD4 binding from microorganisms,
new orally available drugs to replace peptide OMURAet al. (1993) discovered the novel in-
hormones and cytokines - in the same way as hibitors isochromophilone I and I1 (Fig. 8)
motilides are being developed as orally avail- from the culture borth of Penicillium sp. FO-
able gastrointestinal motor stimulating drugs. 2338, and chloropeptin I and I1 (Fig. 9) from
Sfrepfomyces sp. WK-3419 (OMURAet al.,
unpublished data). Chloropeptin 11, however,
was identified with complestatin (KANEKOet
al., 1989).
The inhibitory activities against gp120-
5 Inhibitors of Virus CD4 binding were determined by enzyme-
linked immunosorbent assay (ELISA) using
Receptor Binding - recombinant soluble CD4 and recombinant
gp120 as described by GILBERTet al. (1991).
gp120-CD4 Binding Isochromophilone I and I1 inhibited gp120-
CD4 binding with ICs0 values of 6.6 p M and
Inhibitors 3.9 pM, respectively. The IC,, values for
chloropeptin I and I1 were 2.0 pM and
The entry of viruses needs their specific 3.3 pM, respectively.
binding to a receptor of the susceptible cell. Anti-HIV activity was assayed as follows.
Human immunodeficiency virus (HIV) entry Peripheral human lymphocytes were isolated
begins with the highly specific binding of the by density gradient centrifugation. After stim-
HIV gp120 envelope glycoprotein with a CD4 ulation by a mitogen, the cells were infected
molecule on the surface of most susceptible with a standardized preparation of HIV-1.
cells (MCDOUGALet al., 1986; SADROSKIet Subsequently, the infected cells were cultured
al., 1986; LIFSONet al., 1986). Blocking of in the presence of the agent for 4 days. The
HIV entry is one of the most important tar- amount of viral core protein p24 synthesized
gets for HIV therapy (JOHNSTON and HOTH, and released by the infected cells was deter-
1993). mined by the capture-ELISA technique on
5 Inhibitors of Virus Receptor Binding - gp120-CD4 Binding Inhibitors 125
CH3
lsochromophiloneI
Fig. 8. Structures of the gp120-CD4 binding in-

hibitors isochromophilone I and 11. lsochromophiloneII
Tab. 4. Inhibition of HIV Replication on the Viral days 2, 3, and 4. By comparing with a stand-
Core Protein Level (for the assay method, see ard preparation, the amount of protein (p24)
text) produced by the virus infected cells was calcu-
lated. As shown in Tab. 4 isochromophilone
Sample Viral Core Protein p24
Synthesized (ng/mL) I1 and chloropeptin I significantly inhibited
HIV replication at 25 p M and 7.5 pM, respec-
Day2 Day3 Day4 tively. The inhibition of HIV replication by
isochromophilone I1 and chloropeptin I is
None 0 97.3 129.6 considered to be due to blocking of HIV en-
Isochromophilone I1 0 0 13.5 try into the cells. Isochromophilone I and I1
Chloropeptin I 0 0 7.3 are the first novel non-peptide compounds to
inhibit gp120-CD4 binding. Isochromophi-
lones and chloropeptins are expected to pro-
vide the lead compounds for development of
HIV therapy.
Chloropeptin I
Chloropeptin II (complestatin)
Fig. 9. Structures of the gp120-CD4 binding inhibitors chloropeptin I and 11.

6 Current State and domain of its receptor revealed that this com-
plex is composed of one hormone and two re-
Future Perspectives ceptor molecules. The hormone has four heli-
cal structures with abnormal topology, and
the receptor bound to it has two different
As described above, substances acting on binding domains. The two receptor molecules
receptors (i.e., agonists and antagonists) have bind through the same amino acid residues in
been synthesized before the exact nature of each domain to two structurally distinct sites
receptors was clarified. These substances of the hormone. At their C-terminal domains
have contributed greatly not only to treat- distant from the binding sites the two recep-
ment of diseases but also to advances in stud- tor molecules are in contact to each other.
ies in the field of cellular biology and pharma- This contact may play a crucial role in intra-
cology. cellular signal transduction.
Following recent commercialization of ra- Since the three-dimensional features of the
dio-labeled peptide ligands, screening of re- mode of binding between growth hormone
ceptor-active compounds in various speci- and its receptor has been clarified, molecular
mens such as microbial cultures has been per- designing of new agonists and antagonists us-
formed in experiments involving the binding ing computer graphics technology will ad-
of these ligands to tissue or cells. In this way, vance in the future. However, because crys-
new substances affecting receptors of peptide tallization of the ligand-receptor complex
ligands have been discovered. At present, usually is not easy, search for new receptor-
derivatives of these substances (i.e., devaze- active compounds and subsequent chemical
pide, FK-224, and motilide) are under devel- modification of the thus discovered com-
opment for clinical use. This class of drugs pounds will, for the time being, continue to
will further increase in the future. play a principal role in the development of
It is speculated that orally administered new receptor-active compounds.
non-peptide agonists and antagonists will
bring about an epochal reform of drug thera-
py. To date, however, no low molecular
weight compound acting on the receptors of
macromolecular peptide ligands (e.g., ligands 7 Concluding Remarks
with 100 or more amino acid residues) has
been reported although many natural peptide This chapter provides a general review of
ligands or their analogs have been clinically receptor-active compounds. Studies of sub-
used by injection. stances acting on receptors of peptide ligands
If the three-dimensional structure of li- still have only a very short history. We expect
gands and their receptors is identified and the more simple assay techniques to be devel-
mode of the ligand-receptor binding is clarif- oped in the future, facilitating the discovery
ied in detail, the development of new drugs of many receptor-active compounds. These
by computerized information processing will compounds will help to clarify the function of
be possible. Although the primary structure cells and elucidate the physiological and pa-
of many receptors has been clarified to date, thophysiological functions of receptors.
the three-dimensional structure of a receptor As studies on receptor-active compounds
is not known until the crystalline structure of of microbial origin have been advancing,
the growth hormone receptor complex (see some known substances have been high-
below) has been determined. lighted because of their additional action on
The first analysis of the crystalline structure receptors. Considerably small differences in
of a macromolecular peptide ligand receptor the chemical structures of ligands or receptors
complex was reported in 1992 (DEVOSet al., often reflect quite different actions and it is
1992). The analysis of the three-dimensional quite likely that the same compound can have
structure of the complex formed between hu- two or more target molecules. This indicates
man growth hormone and the extracellular that it is not easy to discover a highly selective
compound and that it is even possible to mod- Fermentation, isolation and biological proper-
ify a compound so that its minor action re- ties, J. Antibiot. 38, 1633-1637.
places the major action - as seen in the case GOETZ,M. A., MONAGHAN, R. L., CHANG,R. S.
of motilide. L., ONDEYKA,J., CHEN, T. B., LOTTI, V. J.
(1988), Novel cholecystokinin antagonists from
Aspergillus alliaceus. I. Fermentation, isolation,
and biological properties. J. Antibiot. 41, 875-
Acknowledgements 877.
HASHIMOTO, M., HAYASHI,K., MURAI,M., FUJII,
The authors are indebted to Dr. H. KLEIN- T., NISHIKAWA, M., KIYOTO,S., OKUHARA, M.,
KAUF for providing the opportunity of this KOHSAKA,M., IMANAKA H. (1992), WS9326A,
presentation, and also to Drs. S. TAKAMATSU a novel tachykinin antagonist isolated from
and J. INOKOSHIand Mr. K. MATSUZAKIfor Streptomyces violaceusniger No. 9326.11. Biolog-
their useful help in preparation of this manu- ical characterization and pharmacological char-
script. acterization of WS9326A and tetrahydro-
WS9326A (FK224), J. Antibiot. 45,1064-1070.
HAYASHI,K., HASHIMOTO, M., SHIGEMATSU, N.,
NISHIKAWA, M., EZAKI,M., YAMASHITA,M.,
KIYOTO, S., OKUHARA,M., KOHSAKA,M.,
IMANAKA, H. (1992), WS9326A, a novel tachy-
8 References kinin antagonist isolated from Streptomyces vio-
laceusniger No. 9326. I. Taxonomy, fermenta-
ABBOIT, A. (1991), Receptor screens detect bioac- tion, isolation, physico-chemical properties and
tivity, Biotechnology 9, 694. biological activities, J. Antibiot. 45, 1055-1062.
CHANG,R. S. L., LOTTI,V. J., MONAGHAM, R. L., HEDGE, V. R., PATEL, M. G., HORAN,A. C.,
BIRNBAUM, J., STAPLAY,E. O., GOETZ,M. A., SCHWARTZ, J. L., HART,R., PUOR,M. S., GUL-
ALBERS-SCHONBERG, G., PATCHEIT, A. A., LO,V. P., IYENGAR, S. (1991), Sch42029, a natu-
LIESCH,J. M., HENSENS,0. D., SPRINGER, J. P. rally produced dopamine rceptor ligand taxono-
(1985), A potent nonpeptide cholecystokinin an- my, fermentation, isolation and structure, J. Ind.
tagonist selective for peripheral tissues isolated Microbiol. 8, 187-192.
from Aspergillus alliaceus, Science 230,177-179. HENSENS,0. D., BORRIS,R. P., KOUPAL,L. R.,
DEVOS, A. M., ULTSCH,M., KOSSIAKOFF, A. A. CALDWELL,C. G., CURRIE,S. A., HAIDRI,A.
(1992), Human growth hormone and extracellu- A., HOMIK,C. F., HONEYCUTT, S. S., LINDEN-
lar domain of its receptor: crystal structure of MAYER,S. M., SCHWARTZ,C. D., WEISSBER-
the complex, Science 255, 306-312. GER, B. A., WOODRUFF, H. B., ZINK,D. L., ZI-
EVANS,B. E. (1989), Recent developments in cho- TANO, L., FIELDHOUSE, J. M., ROLLINS,T.,
lecystokinin antagonist research, Drugs of the SPRINGER,M. S., SPRINGER,J. P. (1991), L-
Future, 14,971-979. 156,602, A C5a antagonist with a novel cyclic
EVANS, B. E., BOCK,M. G., RIITLE, K. E., DIPAR- hexadepsipeptide structure from Streptomyces
DO,R.M., WHIITER,W. L., VEBER,D. F., AN- sp. MA6348. Fermentation, isolation and struc-
DERSON, P. s., FREIDINGER, R. M. (1986), De- ture determination, J. Antibiot. 44,249-254.
sign of potent, orally effective, nonpeptidal an- HORI,Y., ABE, Y., EZAKI,M., GOTO,T., OKUHA-
tagonists of the peptide hormone cholecystokin- RA, M., KOHSAKA,M. (1993a), R1128 sub-
in, Proc. Natl. Acad. Sci. USA 83,4918-4922. stances, novel non-steroidal estrogen-receptor
FUJII, T., MURAI,M., MORIMOTO,H., NISHIKA- antagonists produced by a Streptomyces. I. Tax-
WA,M., KIYOTO,S. (1991), Effects of actinomy- onomy, fermentation, isolation and biological
cin D on airway constriction induced by tachy- properties, J. Antibiot. 46, 1055-1062.
kinin and capsaicin in guinea-pigs, Eur. J. Phar- HORI,Y., TAKASE,S., SHIGEMATSU, N., GOTO,T.,
macol. 194,183-188. OKUHARA, M., KOHSAKA,M. (1993b), R1128
GILBERT,M., KIRIHARA, J., MILLS,J. (1991), En- substances, novel non-steroidal estrogen-recep-
zyme-linked immunoassay for human immuno- tor antagonists produced by a Streptomyces. 11.
deficiency virus type I envelope glycoprotein Physico-chemical properties and structure deter-
120, J. Clin. Microbiol. 29,142-147. mination. J. Antibiot. 46, 1063-1068.
GOETZ, M. A,, LOPEZ, M., MONAGHAN,R. L., HORI, Y., ABE, Y., NISHIMURA,M., GOTO, T.,
CHANG,R. S. L., LOTTI, V. J., CHEN, T. B. OKUHARA,M.,KOHSAKA,M. (1993c), R1128
(1985), Asperlicin, a novel non-peptidal chole- substances, novel non-steroidal estrogen-recep-
cystokinin antagonist from Aspergillus alliaceus. tor antagonists produced by a Streptomyces. 111.
8 References 129
Pharmacological properties and antitumor activ- (1989), Complestatin, a potent anti-complement

ities, J. Antibiot. 46, 1069-1075. substance produced by Streptomyces lavendulae.
HORI, Y., ABE, Y., NAKAJIMA, H., TAKASE, S., I. Fermentation, isolation and biological charac-
FUJITA, T., GOTO, T., OKUHARA, M., KOHSA- terization, J. Antibiot. 42,236-241.
KA, M. (1993d), WB2838 [3-chloro-4-(2-amino- KING,K., DOHLMAN,H. G., THORNER, J., CARON,
3-~hlorophenyl)-pyrrole]:Non-steriodal andro- M. G., LEFKOWITZ,R. J. (1990), Control of
gen-receptor antagonist produced by a Peudo- yeast mating signal transduction by a mammal-
monas, J. Antibiot. 46, 1327-1333. ian a-adrenergic receptor and Gs a subunit,
HORI,Y., ABE, Y., SHIGEMATSU, N. S., GOTO, T. Science 250,121-123.
OKUHARA,M., KOHSAKA,M. (1993e), Napyra- KOGL, F., Cox, H. C., SALEMINK,C. A. (1957),
diomycins A and B1: Non-steroidal androgen- Synthese eines Gemischs von Muscarin und
receptor antagonists produced by a Strepto- seinen stereoisomeren Formen, Experienta 13,
myces, J. Antibiot. 46, 1890-1894. 137-138.
HORI, Y., ABE, Y., NAKAJIMA,H., SHIGEMATSU, KOJIRI,K., IHARA,M., NAKAJIMA, S., KAWAMU-
N., TAKASE,S., GOTO, T., OKUHARA, M., KOH- RA, K., FUNAISHI, K., YANO, M., SUDA, H.
SAKA,M. (1993f), WS9761A and B: New non- (1991), Endothelial-binding inhibitors, BE-
steroidal androgen-receptor antagonists pro- 18257A and BE-18257B. 1. Taxonomy, fermen-
duced by a Streptomyces, J. Antibiot. 46, 1901- tation, isolation and characterization, J. Antibiot.
1903. 44, 1342-1347.
HURLEY,T. R., BUNGE,R. H., WILLMER,N. E., KONDO, Y., TORII, K., OMURA,S., ITOH, Z.
HOKANSON,G. C., FRENCH, J. C. (1986), (1988), Erythromycin and its derivatives with
PD124,895 and PD124,966, two new antitumor motilin-like biological activities inhibit the spe-
antibiotics, J. Antibiot. 39, 1651-1656. cific binding of lZsII-motilinto duodenal muscle,
IHARA,M., FUKURODA, T., SAEKI,T., NISHIKIBE, Biochem. Biophys. Res. Commun. 150,877-882.
M., KOJIRI,K., SUDA,H., YANO,M. (1991), An KUBOTA,K., SUNAGANE, N., SUGAYA,K., MAT-
endothelin receptor (ETA) antagonist isolated SUOKA,Y., URUNO,T. (1983), Competitive an-
from Streptomyces misakiensis, Biochem. Bio- tagonists of cholecystokinin and some benzodia-
phys. Res. Commun. 178,132-137. zepines at cholecystokinin receptors of smooth
IMURA,R., SANO, T., GOTO, J., YAMADA,K., muscle, Jpn. J. Pharmacol. 33, 87.
MATSUDA,Y. (1992), Inhibition by HS-142-1, a KUBOTA,K., SUGAYA,K., SUNGANE,N., MATSU-
novel nonpeptide atrial natriuretic peptide an- DA, I., URANO,T. (1985), Cholecystokinin an-
tagonist of microbial origin, of atrial natriuretic tagonism by benzodiazepines in the contractile
peptide-induced relaxation of isolated rabbit response of the isolated guinea-pig gallbladder,
aorta through the blockade of guanylyl cyclase- Eur. J. Pharmacol. 110,225-231.
linked receptors, Mol. Pharmacol. 42,982-990. KUBOTA,K., SUGAYA,K., KOIZUMI,Y., TODA,M.
ITOH, Z., SUZUKI,T., NAKAYA,M., INOUE,M., (1989), Cholecystokinin antagonism by anthra-
ARAI,H., WAKABAYASHI, K. (1985), Structure- mycin, a benzodiazepine antibiotic, in the cen-
activity relation among macrolide antibiotics in tral nervous system in mice, Brain Res. 485, 62-
intitiation of interdigestive migrating contrac- 66.
tions in the canine gastrointestinal tract, Am. J. KUEHL, F. A., JR., LEBEL, N., RICHTER,J. W.
Physiol. 248, G320-525. (1955), Isolation and characterization studies on
JOHNSTON,M. I., HOTH,G. F. (1993), Present sta- muscarine, J. Am. Chem. SOC.77,6663-6665.
tus and future prospects for HIV therapies, LAM, Y. K. T., BOGEN,D., CHANG,R. S., FAUST,
Science 260,1286-1293. K. A., HENSENS,0. D., ZINK, D. L.,
KAMIYAMA, T., UMINO,T., FUJISAKI, N., FUJIMO- SCHWARTZ,C. D., ZITANO,L., GARRITY,G.
RI, K., SATOH,T., YAMASHITA, Y., OHSHIMA, M., GAGLIARDI,M. M., CURRIE,S. A., WOOD-
S., WATANABE, J., YOKOSE,K. (1993a), Tetrafi- RUFF, H. B. (1991), Novel and potent gastrin
bricin, a novel fibrinogen receptor antagonist. I. and brain cholecystokinin antagonists from
Taxonomy, fermentation, isolation, characteriza- Streptomyces olivaceus. Taxonomy, fermenta-
tion and biological activities, J. Antibiot. 46, tion, isolation, chemical conversion, and physi-
1039-1046. co-chemical and biological properties, J. Anti-
KAMIYAMA, T., ITEZONO,Y., UMINO,T., SATOH, biot. 44, 613-625.
T., NAKAYAMA, N., YOKOSE,K. (1993b), Tetra- LAM, Y. K. T., WILLIAMS,D. L. JR., SIGMUND, J.
fibricin, a novel fibrinogen receptor antagonist. M., SANCHEZ,M., GENILLOUD,O., KONG, Y.
11. Structural elucidation, J. Antibiot. 46, 1047- L., STEVENS-MILES,S., HUANG,L., GARRITY,
1054. G. M. (1992), Cochinmicins, novel and potent
KANEKO, I., KAMOSHITA,K., TAKAHASHI, S. cyclodepsipeptide endothelin antagonists from a
Microbispora sp. I. Production, isolation and MIYATA,S., HASHIMOYO, M., FUJIE,K., SHOUHO,
characterization. J. Antibiot. 45, 1709-1716. M., SOGABE,K., KIYOTO, S., OKUHARA, M.,
LAUER,U., ANKE,T., HANSSKE,F. (1991), Anti- KOHSAKA,M. (1992e), WS009A and B, new en-
biotics from basidiomycetes XXXVIII. 2-Meth- dothelin receptor antagonists isolated from
oxy-5-methyl-1,4-benzoquinone, a thromboxane Streptomyces sp. No. 89009. 11. Biological char-
A2 receptor antagonist from Lentinus adhaerens, acterization and pharmacological characteriza-
J. Antibiot. 44,5945. tion of WS009A and B, J. Antibiot. 45, 1041-
LIESCH,J. M., HENSENS,0. D., SPRINGER,J. P., 1046.
CHANG,R. S. L., LOTH, V. J. (1985), Asperlicin, MORI, Y., ABE, Y., SHIGEMATSU,N., GOTO, T.,
a novel non-peptidal cholecystokinin antagonist OKUHARA,M., KOHSAKA,M. (1993c), Neoira-
from Aspergillus alliaceus. Structure elucidation, diomycins A and B1: non-steroidal estrogen-re-
J. Antibiot. 38,1638-1641. ceptor antagonists produced by a Streptomyces,
LIESCH, J. M., HENSENS,0. D., ZINK, D. L., J. Antibiot. 46, 1890-1893.
GOETZ,M. A. (1988), Novel cholecystokinin an- MORISHITA,Y., SANO,T., ANDO,K., SAITOH,Y.,
tagonists from Aspergillus alliaceus. 11. Structure KASE,H., YAMADA,K., MATSUDA,Y. (1991a),
determination of asperlicins B, C, D, and E, J. Microbial polysaccharide, HS-142-1, competi-
Antibiot. 41, 878-881. tively and selectively inhibits ANP binding to its
LIFSON,J. D., FEINBERG,M. B., REYES,G. R., RA- guanylyl cyclase-containing receptor, Biochem.
BIN, L., BANAPOUR,B., CHAKRABART,S., Biophys. Res. Commun. 176,949-957.
Moss, B., WONG-STALL,F., STEIMER,K. S., MORISHITA,Y., TAKAHASHI, M., SANO,T., KA-
ENGELMAN,E. G. (1986), Induction of CD4-de- WAMOTO,I., ANDO,K., SANO,H., SAITOH,Y.,
pendent cell fusion by the HTLV-IIIILAV en- KASE, H., MATSUDA,Y. (1991b), Isolation and
velope glycoprotein, Nature 323,725-728. purification of HS-142-1, a novel nonpeptide an-
MCDOUGAL,J. S., KENNEDY,M. S., SLIGH,J. M., tagonist for the atrial natriuretic peptide recep-
CORT, S. P., MAWLE,C. A., NICHOLSON,J. K. tor form Aureobasidium sp., Agric. Biol. Chem.
A. (1986), Binding of HTLV-III/LAV to T4+ 55,3017-3025.
cells by a complex of the llOK viral protein and MORISHITA,Y., SANO,T., KASE,H., YAMADA,K.,
the T4 molecule, Science 231, 382-385. INAGAMI,T., MATSUDA,Y. (1992), HS-142-1, a
MIYATA,S., HASHIMOTO,M.,MASUI,Y., EZAKI, novel nonpeptide atrial natriuretic peptide
M., TAKASE,S., NISHIKAWA,M., KIYOTO,S., (ANP) antagonist, blocks ANP-induced renal
OKUHARA,M., KOHSAKA,M. (1992a), WS- responses through a specific interaction with
7338, new endothelin receptor antagonists iso- guanylyl cyclase-linked receptors, Eur. J. Phar-
lated from Streptomyces sp. No. 7338. I. Taxono- macol. 225,203-207.
my, fermentation, isolation, physico-chemical NAKAGAWA, A. (1992), Vasoactive substances, in:
properties and biological activities, J. Antibiot. The Search for Bioactive Compounds from Mi-
45,7682. croorganisms (OMURA,S., Ed.), pp. 198-212.
MIYATA,S., HASHIMOTO,M., FUJIE,K., NISHIKA- Berlin-Heidelberg-New York: Springer-Verlag.
WA,M., KIYOTO,S., OKUHARA, M., KOHSAKA, NAKAJIMA,S., NIIYAMA,K., IHARA,M., KOJIRI,
M. (1992b), WS-7338, new endothelin receptor K., SUDA,H. (1991), Endothelin-binding inhibi-
antagonists isolated from Streptomyces sp. No. tors, BE-18257A and BE-18257B. 11. Structure
7338.11. Biological characterization and pharma- determination, J. Antibiot. 44, 1348-1356.
cological characterization of WS-7338B, J. Anti-
biot. 45,83-87. NODA,M., TAKAHASHI,H., TANABE,T., TOYOSA-
TO, M., KIKYOTANI, S., FURUTANI,Y., HIROSE,
MIYATA,S., FUKAMI, N., NEYA,M., TAKASE,S.,
KIYOTO,S. (1992c), WS-7338, new endothelin T., TAKASHIMA, H., INAYAMA, S., MIYATA,T.,
receptor antagonists isolated from Streptomyces NUMA,S. (1983), Structural homology of Torpe-
sp. No. 7338. 111. Structures of WS-7338A, B, C, do californica acetylcholine receptor subunits,
and D and total synthesis of WS-7338B, J. Anti- Nature 302,528-532.
biot. 45,788-791. ODA, S., SANO,T., MORISHITA, Y., MATSUDA,Y.
MIYATA,S., OHHATA,H., MURAI,H., MASUI,Y., (1992), Pharmacological profile of HS-142-1, a
EZAKI,M.,TAKASE,S., NISHIKAWA, M., KIYO- novel nonpeptide atrial natriuretic peptide
TO, S., KOHSAKA,M. (1992d), WS009A and B, (ANP) antagonist of microbial origin. 11. Resto-
new endothelin receptor antagonists isolated ration by HS-142-1 of ANP-induced inhibition
from Streptomyces sp. No. 89009. I. Taxonomy, of aldosterone production in adrenal glomerulo-
fermentation, isolation, physico-chemical prop- sa cells, J. Pharmacol. Exp. Ther. 263,241-245.
erties and biological activities, J. Antibiot. 45,
I
OHASHI,H., AKIYAMA,H., NISHIKORI,K., Mo-
1029-1040. CHIZUKI,J. (1992), Asterric acid, a new endo-
8 References 131
thelin binding inhibitor, J. Antibiot. 45, 1684- turally unique, potent, and selective oxytocin an-
1685. tagonist derived from Streptomyces silvensis, En-
OHTSUKA,T., KUDOH,T., SHIMMA,N., KOTAKI, docrinology 125,217-222.
H., NAKAYAMA,N., ITEZONO,Y., FUJISAWA, SADROSKI, J., GOH, W. C., ROSEN,C., CAMPBELL,
N., WATANABE,J., YOKOSE, K., SETO, H. K., HASELTIME,W. A. (1986), Role of HTLV-
(1992), Tetronothiodin, a novel cholecystokinin III/LAV envelope in syncytium formation and
type-B receptor antagonist produced by Strepto- cytopathicity, Nature 322, 47W74.
myces sp., J. Antibiot. 45, 140-143. SANO,T., IMURA,R., MORISHITA,Y., MATSUDA,
OHTSUKA,T., KOTAKI,H., NAKAYAMA, N., ITE- Y., YAMADA,K. (1992a), HS-142-1, a novel po-
ZONO,Y., SHIMMA,N., KUDO,T., KUWAHARA, lysaccharide of microbial origin, specifically re-
T., ARISAWA, M., YOKOSE, K., SETO, H. cognizes guanylyl cyclase-linked ANP receptor
(1993a), Tetronothiodin, a novel cholecystokinin in rat glomeruli, Life Sci. 51, 1445-1451.
type-B receptor antagonist produced by Strepto- SANO, T., MORISHITA,Y., MATSUDA, Y., YA-
myces sp. NR0489. 11. Isolation, characterization MADA, K. (1992b), Pharmacological profile of
and biological activities. J. Antibiot. 46, 11-17. HS-142-1, a novel nonpeptide atrial natriuretic
OHTSUKA,T., NAKAYAMA,N., ITEZONO, Y., peptide (ANP) antagonist of microbial origin. I.
SHIMMA,N., KUWAHARA,T., YOKOSE, K., Selective inhibition of the actions of natriuretic
SETO,H. (1993b), Tetronothiodin a novel chole- peptides in anesthetized rats, J. Pharmacol. Exp.
cystokinin type-B receptor antagonist produced Ther. 260,825-831.
by Streptomyces sp. NR0489. 111. Structural elu-
cidation, J. Antibiot. 46, 18-24. SANO,T., MORISHITA,Y., YAMADA,K., MATSU-
DA, Y. (1992c), Effects of HS-142-1, a novel
OKAMOTO,M., YOSHIDA,K., UCHIDA,I., NISHI- non-peptide ANP antagonist, on diuresis and
KAWA, M., KHOSAKA,M., AOKI, H. (1986a), natriuresis induced by acute volume expansion
Studies of platelet activating factor (PAF) antag- in anesthetized rats, Biochem. Biophys. Res.
onists from microbial products. I. Bisdethiobis- Commun. 182,824-829.
(methy1thio)gliotoxin and its derivatives, Chem.
Pharm. Bull. 34,340-344. SCHWARTZ,R. E., HIRSCH,C. F., SPRINGER, J. P.,
OKAMOTO,M., YOSHIDA, K., NISHIKAWA,M., PETTIBONE,D. J., ZINK,D. L. (1987), Unusual
ANDO,T., IWAMI,M., KHOSAKA,M., AOKI,H. cyclopropane-containing hapalindolinones from
(1986b), FR-900452, a specific antagonist of a cultured Cyanobacterium, J. Org. Chem. 52,
platelet activating factor (PAF) produced by 3704-3706.
Streptomyces phaeofaciens. I. Taxonomy, fer- SEDLOCK,D. M., BARROW,C. J., BROWNELL,J.
mentation, isolation and physico-chemical and E., HONG, A., GILLUM,A. M., HOUCK,D. R.
biological characteristics, J. Antibiot. 39, 198- (1994), WIN64821, a novel neurokinin antago-
204. nist produced by an Aspergillus sp. I. Fermenta-
OLEYNEK,J. J., SEDLOCK,D. M., BARROW,C. J., tion and isolation, J. Antibiot. 47, 391-398.
APPELL, K. C., CASIANO,F., HAYCOCK,D., SHIGEMATSU,N., TUJI, E., KAYAKIRI,N., TA-
WARD, S. J., KAPLITA, P., GILLUM,A. M. KASE, s., TANAKA,H., TADA, T. (1992),
(1994), .WIN64821, a novel neurokinin antago- WF11605, an antagonist of leukotriene B4 pro-
nist produced by an Aspergillus sp. 11. Biological duced by a fungus. 11. Structure determination,
activity, J. Antibiot. 47, 399-410. J. Antibiot. 45, 704-708.
OMURA,S., TSUZUKI,K., SUNAZUKA, T., MARUI, STOB, M., BALDWIN,R. S., TUITE,J., ANDREWS,
S., TOYODA,H., INATOMI,N., ITOH, Z. (1987), F. N., GILLETTE, K. G. (1962), Isolation of an
Macrolides with gastrointestinal motor stimulat- anabolic, uterotrophic compound from corn in-
ing activity, J. Med. Chem. 30, 1941-1943. fected with Gibberella zeae, Nature 1%,1318.
OMURA,S., TANAKA,H., MATSUZAKI, K., IKEDA, SUGANO,M., SATO, A., IIJIMA,Y., OSHIMA,T.,
H., MASUMA,R. (1993), Isochromophilones I FURUYA,K., KUWANO,H., HATA,T., HANZA-
and 11, novel inhibitors against gp120-CD4 bind- WA,H. (1991), Phomatin A: a novel PAF antag-
ing from Penicillium sp., J. Antibiot. 46, 1908- onist from a marine fungus Phoma sp., J. Am.
1911. Chem. SOC.113,5463-5464.
PEROUTKA,S. J. (1994), Handbook of Receptors SUNAZUKA, T., TSUZUKI,K., MARUI,S., TOYODA,
and Channels. Boca Raton, FL: CRC Press. H., OMURA,S., INATOMI,N., ITOH, Z. (1989),
PETTIBONE, D. J., CLINESCHMIDT, B. V., ANDER- Motilides, macrolides with gastrointestinal mo-
SON,P. S., FREIDINGER, R. M., LUNDELL,G. F., tor stimulating activity. 11. Quaternary N-substi-
KOUPAL, L. R., SCHWARTS,C. D., WILLIAM- tuted derivatives of 8,9-anhydroerythromycin A
SON, J. M., GOETZ, M. A., HENSENS,0. D., 6,9-hemiacetal and 9,9-dihydroerythromycin A
LIESCH,J. M., SPRINGER,J. P. (1989), A struc- 6,9-epoxide, Chem. Pharm. Bull. 37,2701-2709.
TANAKA, T., ICHIMURA,M., NAKAJO,S., SNAJ- Motilides, macrolides with gastrointestinal mo-
DAR, R. M., MORISHITA,Y., SANO, T., YA- tor stimulating activity. I. 0-Substituted and ter-
MADA,K., INAGAMI, T., MATSUDA,Y. (1992), tiary N-substituted derivatives of 8,9-anhydro-
HS-142-1, a novel non-peptide antagonist for erythromycin A 6,9-hemiacetal, Chem. Pharm.
atrial natriuretic peptide receptor, selectively in- Bull. 37,2687-2700.
hibits particulate guanylyl cyclase and lower cyc- UENO,M., TAKESAKO, K., IKAI,K., KATAYAMA,
lic GMP in LLC-PKI cells, Biosci. Biotech. Bio- K., SHIMANAKA, K., NAKAMURA, T. (1988),
chem. 56,1041-1045. Isolation of novel muscarine antagonists, IJ 2702
TASKESAKO, K., KURODA,H., UENO,M., SAITO, and PF6766, Nippon Nogei Kagaku Kaishi 62,
H., YAMAMOTO,J., NAKAMURA,T. (1988) 339.
Screening of muscarine antagonists by radiore- URRY,W. H., WEHRMEISTER, H. L., HODGE, E.
ceptor assay, Nippon Nogei Kagaku Kaishi 62, B., HIDY,P. H. (1966), The structure of zearale-
338. none, Tetrahedron Lett. 27,3109-3114.
TOKI,S., ANDO,K., YOSHIDA,M., KAWAMOTO, I., WATANABE,J., FUJISAKI,N., FUJIMORI, K., AN-
SANO,H., MATSUDA,Y. (1992a), ES-242-1, a ZAI, Y., OSHIMA,S., SANO,T., OHTSUKA,T.,
novel compound from Verticilliumsp., binds to a WATANABE,K., OKUDA,T. (1993), Tetrono-
site on N-methyl-D-aspartate receptor that is thiodin, a novel cholecystokinin type-B receptor
coupled to the channel domain, J. Antibiot. 45, antagonist produced by Streptomyces sp.
88-93. NR0489. I. Taxonomy, yield improvement and
TOKI, S., ANDO, K., KAWAMOTO,I., SANO,H., fermentation, J. Antibiot. 46,l-10.
YOSHIDA,M., MATSUDA,Y. (1992b), ES-242-2, WEBER,W., FISCHLI,W., HOCHULI,E., KUPFER,
-3, -4, -5, -6, -7, and -8, novel bioxanthracenes
E., WEIBEL,E. K. (1991), Anantin - a peptide
produced by Verticilliumsp., which act on the N- antagonist of the atrial natriuretic factor (ANF).
methyl-D-aspartate receptor, J. Antibiot. 45,
I. Producing organism, fermentation, isolation
1047-1054.
and biological activity, J. Antibiot. 44, 164-171.
TOKI,S., MORISHITA, Y., SANO,T., MATSUDA,Y.
(1992c), HS-142-1, a novel non-peptide ANP an- WONG,S.-M., KULLNIG, R., DEDINAS, J., APPELL,
tagonist, blocks the cyclic GMP production eli- K. C., KYDD,G. C., GILLUM,A. M., COOPER,
cited by natriuretic peptides in PC12 and R., MOORE,R.(1993a), Anthrotainin, an inhibi-
NG 108-5 cells, Neurosci. Lett. 135, 117-120. tor of substance P binding produced by Gliocla-
TOKI,S., TSUKUDA, E., NOZAWA,M., NONAKA, dium catenulatum, J. Antibiot. 46,214-221.
H., YOSHIDA,M., MATSUDA,Y. (1992d), The WONG,S.-M., MUSZA,L. L., KYDD,G. C., KULL-
ES-242s, novel N-methyl-D-aspartate antagonists NIG,R., GILLUM,A. M., COOPER,R. (1993b),
of microbial origin, interact with both the neuro- Fiscalins: New substance-P inhibitors produced
transmitter recognition site and the ion channel by the fungus Neosartorya fischeri. Taxonomy,
domain, J. Biol. Chem. 267,14884-14892. fermentation, structures, and biological activi-
TSUJI, E., TSURUMI,Y., MIYATA,S., FUJIE, K., ties, J. Antibiot. 46, 545-553.
KAWAKAMI, A., OKAMOTO, M., OKUHARA, M. WYSS, D. F., LAHM, H. W., MANNEBERG, M.,
(1992a), WF11605, an antagonist of leukotriene LABHARDT,A. M. (1991), Anantin - a peptide
B4 produced by a fungus. I. Producing strain, antagonist of the atrial natriuretic factor (ANF).
fermentation, isolation and biological activity, J. 11. Determination of the primary sequence by
Antibiot. 45,698-703. NMR on the basis of proton assignments, J. An-
TSUJI,E., SHIGEMATSU, N., HATANAKA, H., YA- tibiot. 44, 172-180.
MASHITA,M., OKAMOTO,M., OKUHARA,M. ZINK,D., HENSENS,0. D., LAM,Y. K. T., REAM-
(1992b), Novobiocin, an antagonist of leuko- ER,R., LIESCH,J. M. (1992), Cochinmicins, nov-
triene B4, J. Antibiot. 45, 1958-1960. el and potent cyclodepsipeptide endothelin an-
TSUZUKI, K., SUNAZUKA, T., MARUI,S., TOYODA, tagonists from a Microbispora sp. 11. Structure
H., OMURA,S., INATOMI,N., ITOH, Z. (1989), determination, J. Antibiot. 45, 1717-1722.
Biotechnology
4 Microbial Lipids
COLINRATLEDGE
Hull, United Kingdom
1 Introduction 135
1.1 Lipid Nomenclature and Major Lipid Types 138
2 Accumulation of Lipid 139
2.1 Patterns of Accumulation 139
2.2 Efficiency of Accumulation 142
2.3 Biochemistry of Accumulation 143
3 Triacylglycerols and Fatty Acids 146
3.1 Bacteria 147
3.1.1 Polyunsaturated Fatty Acids in Bacteria 147
3.2 Yeasts 148
3.2.1 Production of a Cocoa Butter Equivalent Yeast Fat 148
3.2.1.1 Direct Feeding of Stearic Acid 150
3.2.1.2 Inhibition of Stearoyl Desaturase 151
3.2.1.3 Mutation 152
3.2.1.4 Metabolic Manipulation 154
3.2.1.5 Conclusions 154
3.3 Molds 155
3.3.1 y-Linolenic Acid (GLA, 183 w-6) 159
3.3.2 Dihomo-y-Linolenic Acid (DHGLA, 20:3 w-6) 160
3.3.3 Arachidonic Acid (ARA, 20:4 w-6) 161
3.3.4 Eicosapentaenoic Acid (EPA, 20: 5 w-3) 162
3.3.5 Docosahexaenoic Acid (DHA, 22 :6 w-3) 162
3.3.6 Eicosatrienoic Acid (ETA, 20:3 w-9, “Mead Acid”) 163
3.3.7 Conclusions 163
3.4 Algae 164
3.4.1 y-Linolenic Acid (GLA, 18:3 w-6) 167
3.4.2 Arachidonic Acid (ARA, 20:4 w-6) 167
3.4.3 Eicosapentaenoic Acid (EPA, 20: 50-3) 168
3.4.4 Docosahexaenoic Acid (DHA, 22:6 0-3) 169
3.4.5 Conclusions 170
134 4 Microbial Lipids
4 Sterols, Carotenoids, and Polyprenes 170

4.1 Sterols 170
4.2 Carotenoids 172
4.3 Polyprenoids 175
5 Wax Esters and Polyesters 176
5.1 Wax Esters 176
5.2 Polyesters - Poly-P-Hydroxyalkanoates 177
6 Other Lipids 180
6.1 Biosurfactants 180
6.2 Ether (Archaebacterial) Lipids 181
6.3 Phospholipids and Sphingolipids 183
6.4 Prostanoid-Type Lipids 184
7 Conclusions 185
8 References 186
1 Introduction 135
1 Introduction low- (or zero-) erucic acid (20:l) oil which is

then a permitted oil for food manufacture.
Overall production of plant and animal oils
Since the publication of the 1st Edition of is increasing at about 3% per annum; produc-
“Biotechnology” and the earlier chapter on tion in 1992193 was about 85-106t and is ex-
the biotechnology of lipids in 1986, a consid- pected to reach 105*106t by the year 2000
erable number of developments have taken (MIELKE,1992). Pricing of these materials re-
place in this field. Some microbial lipid prod- mains highly competitive as most products us-
ucts have now been produced commercially ing oils can switch between the various types
and prospects for other developments appear according to the price of the day. The average
to be not too far away. In some cases, as e.g. price index for the major commodity oils is
with the bacterial lipid poly-p-hydroxybuty- about US$ 500-550 per t though groundnut
rate, no counterpart exists from plant or ani- oil, e.g., is always significantly higher than the
mal sources and consequently the economics average at $800-850 per t. The highest priced
of producing this product lie outside the nor- commodity oil, excluding the speciality mate-
mal oils and fats domain. With most other mi- rials, is always olive oil at $ 1,500-2,000 per t.
crobial lipids, these are the equivalent in com- Its price depends on its quality which includes
position to plant-derived oils and consequent- minor, but very important, flavor compo-
ly must compete against these in any potential nents. Animal fats (tallow and lard) have
market place. Only the highest valued oils steadily declined in consumption over the
have any chance of being produced by bio- past decade and are likely to fall even further
technological means as it is impossible for mi- to about 20% of the total market by 2001
croorganisms to produce oils and fats as (SHUKLA,1994). Their prices are therefore
cheaply as the main commodity oils are pro- usually at or below the average index level.
duced from plant and animal sources. Howev- The trends in world oil and fats supplies
er, there is always the possibility of producing are under constant surveillance and are fre-
a microbial oil as an adjunct to some waste quently reviewed in various publications: the
treatment process in a way similar to that of- extensive reviews by SHUKLA(1994) and
ten used to produce microbial proteins (SCP MIELKE(1992) can be recommended though
- single cell protein) for animal feed from for current information journals such as Lipid
some unwanted substrate. Microbial oils - Technology (P. T. Barnes & Associates), Oils
which could then be referred to as single cell and Fats International (Chase Webb, St Ives
oils (SCO) - would have the double advan- PLC), INFORM (American Oil Chemists’
tage over SCP in that they could probably sell Society, Illinois) provide invaluable and con-
for a higher price than SCP and, moreover, tinuously up-dated information in most areas.
could be used for a technical purpose should There are, in addition, a number of special-
the nature of the substrate prevent the prod- ized trade reviews that provide weekly prices
uct being returned into the food chain. of the traded oils.
The major commercial plant oils continue The fatty acid composition of the major
to be dominated by soybean oil (current 1993 commercial oils is given in Tab. 1. The nom-
production is about 18*106t); palm oil, enclature of lipids is given in Sect. 1.1. It will
though, continues to be the fastest growing be appreciated that, unlike say animal feed
market with 14-106 t now being produced protein, the composition of the fats varies
compared to 6.106 t in 1983. If the present considerably from species to species. In all
rate of expansion in palm oil production con- cases, however, the oil or fat is composed al-
tinues in Malaysia and Indonesia (BASIRON most entirely ( > 98%) of triacylglycerols
and IBRAHIM,1994; LEONARD,1994), then (formerly known as triglycerides) - Sect. 1.1.
palm oil will overtake soybean oil production For edible purposes, the oil or fat is retained
by the end of this decade. Rapeseed oil (now in this form though the individual fatty acyl
9 -lo6 t in 1993) is also expanding mainly due groups on the glycerol can be modified -
to increased cultivation in Europe and Cana- usually by chemical means - without affecting
da. The variety now under cultivation is the the triacylglycerol structure per se. Some en-
Tab. 1. Fatty Acid Composition of Fats and Oils of Animal and Plant Origin
FatslOils Relative Proportion of Fatty Acyl Groups [“h(w/w)]
4:O-10:0 12:O 14:O 16:O 16:l 18:O 18:l 18:2 18:3 20:O 20:l Others
Animal Fats
Butterfat 10 3 11 27 2 12 29 2 - - - 15:0+17:0, 3%
Beef tallow - - 3 24 4 19 43 3 1 - - 15:0+17:0, 2%;
14:1+17:1,2%
Lard - - 2 26 3 14 44 10 - - -
Plant Oils
Coconut oil 15 47 18 9 - 3 6 2 - - - -
Palm kernel oil 8 48 16 8 - 3 15 2 - - - -
Cocoa butter - - - 26 - 35 35 3 - 1 - -
Olive oil - - - 13 1 3 71 10 1 1 - -
Rapeseed oil - - - 4 - 2 62 22 10 - - -
Groundnut oil” - - - 11 - 2 48 32 - 1 2 22:0+24:0,5%
Sunflower oil - - - 7 - 5 19 68 1 - - -
Soybean oil - - - 11 - 4 24 54 7 - - -
Corn oil - - - 11 - 2 28 58 1 - - -
Cotton seed oil - - 1 22 1 3 19 54 1 - - -
“Exotic” Plant Oils
Borage seed oil - - - 11 - 4 16 39 22b - 4.5 22:1, 2.5%
24:0, 1.5%
Evening primrose seed oil - - - 8 - 2 9 70 9b - -
Blackcurrant seed oil - - - 6 - 1 10 48 17b - - (Y-18~3,13%
a Also known as peanut oil.
yLinolenic acid, 18:3 (6, 9,12).
I Introduction 137
zymatic reformulation of triacylglycerols oc- tential source of the widest types of lipids
curs on an industrial scale using stereospecific then it is possible to identify a number of po-
lipases to transesterify palm oil fractions into tentially attractive products. For the purposes
the much more expensive cocoa butter-like of this article, I have therefore used the broad
triacylglycerols (OWUSU-ANSAH, 1993). definition of a lipid as any material that is de-
With some technical applications of oils, it rived from a (micro)organism, is directly sol-
is the fatty acid that is required: consequently uble in organic solvents, and is essentially a
saponification (hydrolysis) of the triacylglyc- water-insoluble material. However, as there is
erol is carried out and the fatty acid used still considerable interest in the manner in
either as such, e.g., with soap manufacture, or which microorganisms synthesize large quan-
is modified to an appropriate derivative tities of lipids, much of the review will be tak-
which is then used in a multitude of products: en up with the more conventional types of
from detergents to adhesives. oils and fats that they produce. The entire
The aim of all biotechnological processes is subject of microbial lipids, encompassing all
to produce products that are either cheaper aspects and not just biotechnology, has been
than can be obtained from other sources, in- the above subject of a two-volume mono-
cluding possible chemical synthesis, or are not graph by RATLEDGEand WILKINSON (1988a,
available by any other means. Within the field 1989). The industrial applications of microbial
of lipids, the opportunities to produce triacyl- lipids have also been the subject of a mono-
glycerol lipids are limited to the highest val- graph edited by KYLE and RATLEDGE
ued materials. The highest priced bulk (com- (1992). Details concerning the degradation of
modity) oil is cocoa butter whose price has fats, oils, and fatty acids, including the action
varied between $ 8,000 to $ 3,000 per t over of lipases and phospholipases, which are not
the past decade. At the higher price level, the covered here have been recently reviewed
prospects of producing a cocoa butter equi- elsewhere by the author (RATLEDGE,1993).
valent oil by yeast technology have looked fa- It will be appreciated, of course, that al-
vorable. This topic is specifically reviewed lat- though microorganisms remain a potential
er (see Sect. 3.2.1). source of oils and fats, there is considerable
Other very high valued oils are those in the effort being put into the production of oils
health care market and which have had var- and fats from conventional plant sources.
ious claims made on their behalf for the ame- Such efforts include the modification of pea-
lioration of various diseases and conditions. nut oils (groundnut oil) to produce changes in
Of current interest are oils containing the po- the fatty acid composition so that the more
lyunsaturated fatty acids: y-linolenic acid, desirable oils can be produced more cheaply.
18:3 (0-6); arachidonic acid, 20:4 (w-6); eico- The application of genetic engineering is now
sapentaenoic acid, 20: 5 (w-3); and docosa- gathering pace as a means of producing “tail-
hexaenoic acid, 22 :6 (w-3). Oils containing or-made” oils and fats in plants and is likely
such fatty acids are found in a number of mi- to supersede the traditional plant breeding
croorganisms and are reviewed in Sects. 3.3 approach as a means of creating what is
and 3.4. wanted more quickly and with greater cer-
The very highest priced lipids though are tainty. This review, however, will not include
probably the prostanoid compounds encom- any detailed review of the current develop-
passing the prostaglandins, leukotrienes, and ments in plant genetic engineering as applied
thromboxanes. These are mainly used for to the commodity oils and fats. Readers
treatment of uncommon disorders or for ex- should though be aware that such advances
perimental purposes. Consequently, the are now likely to be a major influence in the
amounts required per annum are probably at availability of “improved” oils for everyday
the kilogram stage rather than the ton (or use and will undoubtedly ensure that these
kiloton) stage with other lipid products. Pros- materials remain highly competitively priced
pects for producing such materials are briefly for many years to come. The recent reviews
mentioned in Sect. 6.4. by HARWOOD(1994a, b), MURPHY(1994a)
Thus, if we view microorganisms as a po- and RATTRAY(1994) and the monographs
Tab. 2. Developments in Genetic Engineering (by in vitro Mutagenesis and

Gene Transfer) Being Applied for the Modification of Plant Oils (adapted
from RAITRAY,1994)
Plant Target Fatty Objective Application

Acid
Soybean and 16:O increase margarine

rapeseed 16:O decrease edible oil
18:O increase margarine; cocoa butter
substitute(?)
18:l increase improved edible oil
~~-18:3 decrease improved stability and
odor
Rapeseed 22: 1 increase erucic acid for
oleochemicals
Sunflower
Linseed
18:l
18:2+ 18:3
18:3
increase
decrease
increase
1olive oil
substitute
oleochemicals
Groundnut 18:l increase improved edible oil
edited by RATTRAY(1991) and by MURPHY acids may be saturated or unsaturated with

(1994b) will be found particularly useful in one or more double bonds which are usually
this respect. Tab. 2 summarizes some of the in the cis (or 2) form. The structure of a fatty
current developments that are now taking acid is represented by a simple notation sys-
place in this area. tem -X: Y , where X i s the total number of C
The opportunities for microorganisms to atoms and Y is the number of double bonds.
produce oils and fats of commercial value for Thus, 18:0 is octadecanoic acid, that is stearic
the bulk markets remain doubtful but where acid; 16:1 is hexadecenoic acid, that is palmit-
the product cannot be obtained from else- oleic acid, with one double bond, and 18:3
where then this provides a much better op- would represent octadecatrienoic acid, a C18
portunity for a microbial oil than attempting acid with three double bonds. The position of
to replicate what is already available from the double bond(s) is indicated by designat-
plant sources. Some opportunities neverthe- ing the number of the C atom, starting from
less do exist but they have to be identified the COOH terminus, from which the double
with some care. Hopefully, some of the fol- bond starts: oleic acid is thus 18: 1 (9) signify-
lowing material may indicate to the astute ing the bond is from the 9th (to the 10th) C
reader where such opportunities may lie. atom. If it is necessary to specify the isomer,
this is added as “c” (for “cis”=Z) or “t” for
“trans” = E). In this review, the cisltrans sys-
1.1 Lipid Nomenclature and Major tem is used. Thus, cis, cis-linoleic acid is 18:2
(c 9, c 12). Most naturally-occurring unsatu-
Lipid Types rated fatty acids are in the cis configuration
and, unless it is stated otherwise, this configu-
Fatty acids are long chain aliphatic acids ration may be assumed.
(alkanoic acids) varying in chain length from, With polyunsaturated fatty acids (PUFA),
normally, CI2 to CZ2though both longer and the double bonds are normally methylene-in-
shorter chain-length acids are known. In most terrupted: -CH= CH -CH2- CH= CH- .
cells (microbial, plant, and animal), the pre- Thus, once the position of one bond is speci-
dominant chain lengths are 16 and 18. Fatty fied all the others are also indicated. In num-
bering PUFAs, a “reverse” system is used With respect to the monoacylglycerols,

where the position of only the last double there are obviously three possible isomers
bond is given by the number of carbon atoms and similarly for the diacylglycerols.
it is from the CH3 terminus. To denote the Where different acyl groups are attached to
“counting back” system is in operation the the glycerol moiety, these can then be individ-
notation is given as 0-x or n-x; thus the two ually given. For example, 1-stearoyl-Zoleoyl-
main isomers of linolenic acid (18:3) are giv- 3-palmitoyl-sn-glycerol is the major triacyl-
en as 18:3 (0-3) and 18:3 (w-6) or as 18:3 glycerol of cocoa butter with stearic, oleic,
(n-3) and 18:3 (n-6). In some systems, the and palmitic acids on the three OH posi-
minus sign may be omitted giving, for exam- tions.
ple, 18:3 (w3) or 18:3 (n3) and 18:3 (&) and Phospholipids possess two fatty acyl groups
18:3 (n6). Respectively, these two isomers at the sn-1 and sn-2 positions of glycerol with
are: a phospho group at sn-3 which is also linked
in 17 16 IS 14 13 12 I1 10 9
CH3-CH2-CH=CH-CH2-CH=CH-CH2-CH=CH
I
HOOC-H~C-H~C-H~C-H~C-HZC-H~C-H~C
2 3 4 S 6 1
CH3-CH2-CH2- CH2- CH2-CH=CH-CHz-CH=CH

w w-1 w-2 0-3 0 4 w-5 w-6
n n-1 n-2 n-3 n-4 n-5 n-6
HOOC-H2C-HzC-H2C-H2C-HC=HC-H2C
The w-x system will be used in this review. to a polar head group: choline, serine, etha-
When fatty acids are esterified to glycerol, nolamine, and inositol are the common ones.
they give a series of esters: mono-, di-, and For the nomenclature and naming of phos-
triacylglycerols (1, 11, 111). This is the pre- pholipids and other microbial lipids, the mul-
ferred nomenclature to the older mono-, di-, tiauthored treatise Microbial Lipids, edited
and triglycerides. by RATLEDGEand WILKINSON (1988a, 1989),
CHp-0-CO-R CH2-O-COR CHp-0-COR
I I I
CH-OH CH -0-COR CH -0-COR
I I I
CHP-OH CHZ-OH CHZ-0-COR
I I1 I11
where R is a long alkyl chain and RCO- is, may be helpful though there are numerous
therefore, the fatty acyl group. text books on lipids that provide similar infor-
mation.
As various isomeric forms are possible, the
position of attached acyl group must be speci-
fied in most cases. For this, the stereospecific
numbering (sn-) system is used so that the
two prochiral positions of glycerol (IV)can 2 Accumulation of Lipid
be distinguished as sn-1 and sn-3.
’ CH20H 2.1 Patterns of Accumulation
2 LHOH Not all microorganisms can be considered
I as abundant sources of oils and fats, though,
CH20H like all living cells, microorganisms always
IV contain lipids for the essential functioning of
membranes and membranous structures.

Those microorganisms that do produce a high
content of lipid may be termed “oleaginous”
in parallel with the designation given to oil-
bearing plant seeds. Of the some 600 different
yeast species, only 25 or so are able to accu-
mulate more than 20% lipid; of the 60,OOO
fungal species fewer than 50 accumulate more
than 25% lipid (RATLEDGE,1989a).
The lipid which accumulates in oleaginous
microorganisms is mainly triacylglycerol (see
Sect. 1.1). If lipids other than this type are re-
quired then considerations other than those
expressed here might have to be taken into
account to optimize their production. With
few exceptions, oleaginous microorganisms
are eukaryotes and thus representative spe-
cies include algae, yeasts, and molds. Bacteria
do not usually accumulate significant Fig. 2. Electron micrograph of Cryptococcus curva-
amounts of triacylglycerol but many do accu- tus ( = Candida curvata =Apiotrichum curvatum)
mulate waxes and polyesters (see Sects. 5.1 strain D grown for 2 days on nitrogen-limiting me-
and 5.2) which are now of commercial inter- dium (viz. Fig. 1) showing presence of multiple lip-
est. id droplets. Total lipid content approx. 40%, mark-
The process of lipid accumulation in yeasts er bar: 1 km (from HOLDSWORTH et al., 1988).
and molds growing in batch culture was eluci-
dated in the 1930s and 1940s (see WOOD-
BINE, 1959, for a review of the early litera- culture to become exhausted within about
ture, and RATLEDGE,1982, for an updated 2 4 4 8 h. Exhaustion of nutrients other than
review of these aspects). A typical growth nitrogen can also lead to the onset of lipid ac-
pattern is shown in Fig. 1. This pattern is also cumulation (see GRANGERet al., 1993, for a
found with the accumulation of polyester ma- recent reference) but, in practice, cell proli-
terial in bacteria (see Sect. 5 ) . feration is most easily effected by using a lim-
The key to lipid accumulation lies in allow- iting amount of N (usually NH4+ or urea) in
ing the amount of nitrogen supplied to the the medium. The excess carbon which is
available to the culture after N exhaustion
15 continues to be assimilated by the cells and,
by virtue of the oleaginous organism possess-
60 ing the requisite enzymes (see below), is con-
verted directly into lipid.
The essential mechanism which operates is
that the organism is unable to synthesize es-
sential cell materials - protein, nucleic acids,
etc. - because of nutrient deprivation and
’3 0
0 20 40 60 80
thus cannot continue to produce new cells.
Because of the continued uptake of carbon
and its conversion to lipid, the cells can then
Culture time ( h l be seen to become engorged with lipid drop-
Fig. 1. Typical lipid accumulation pattern for a lets (Fig. 2). It is important to appreciate,
yeast (Rhodotorula glutinis = R. gracilis) growing however, that the specific rate of lipid biosyn-
on a high C:N ratio medium in batch culture. Bio- thesis does not increase; the cells fatten be-
mass M, % lipid content 0, NH: in medium 0 cause other processes slow down or cease al-
(from YOONet al., 1982). together and, as lipid biosynthesis is not
linked to growth, this may continue unabated. The exact ratio of C to N chosen for the
The process of lipid accumulation (Fig. 1) can medium was originally considered to be of lit-
be seen as a two-phase batch system: the first tle consequence provided N was the limiting
phase consists of balanced growth with all nu- nutrient and sufficient carbon remained to
trients being available; the subsequent “fat- ensure good lipid accumulation. However,
tening” or “lipogenic” stage occurs after the YKEMAet al. (1986) showed that a range of
exhaustion of a key nutrient other than car- lipid yields in an oleaginous yeast, Apiotri-
bon and, of course 02.The role of O2 during chum curvatum (originally Candida curvata
lipid formation was discussed briefly in the but now Cryptococcus curvatus; see BARNETT
1st Edition of “Biotechnology” (RATLEDGE, et al., 1990) were traversed in continuous cul-
1986). ture by varying the C:N ratio of the growth
Accumulation of lipid has also been medium. There was a hyperbolic relationship
achieved in single stage continuous culture between the C:N ratio and the maximum
(RATLEDGEet al., 1984) and a typical accu- growth (dilution) rate that the organism could
mulation profile dependent upon the dilution attain: the lowest growth rate was at the high-
rate (growth rate) is shown in Fig. 3. As with est C:N ratio of 50:l and this, in turn, con-
batch cultivation, the medium has to be for- trolled the amount of lipid produced and the
mulated with a high carbon-to-nitrogen ratio, efficiency of yield (g lipid per g glucose used)
usually about 50:l. The culture must be with which it was produced. Although the
grown at a rate which is about 25-30% of the highest lipid contents of the cell (50% w/w)
maximum. Under this condition, the concen- were obtained with a C:N ratio of 50:l or
tration of nitrogen in the medium is virtually over, the optimum ratio for maximum pro-
nil and the organism then has sufficient resi- ductivity (g L-’ h-’ lipid) was at a ratio of
dence time within the chemostat to assimilate 25: 1with glucose (YKEMAet al., 1986) and at
the excess carbon and convert it into lipid. 30-35 :1 when whey permeates were used
The rate of lipid production (i.e., g L h-’) -’ with same yeast (YKEMAet al., 1988). Similar
is usually faster in continuous cultures than in results for describing the optimum C:N ratio
batch ones (EVANSand RATLEDGE,1983; for lipid accumulation have been developed
FLOETENMEYER et al., 1985). by GRANGERet al. (1993) using Rhodotorula
glutinis.
Interestingly, YKEMA et al. (1986) com-
mented that Apiotrichum curvatum simulta-
7 neously accumulated about 20% carbohy-
drate in the cells along with the 50% lipid.
6 Such a phenomenon of carbohydrate forma-
tion had been conjectured by BOULTONand
-- 5 RATLEDGE(1983a) to be a likely event to ac-
‘1
count for an observed delay in lipid synthesis
ZL 60 -3 after glucose assimilation had been initiated.
VI
v) .. This carbohydrate was also recognized inde-
g 3
-
m LO E pendently by HOLDSWORTH et al. (1988) in
2
c
C
W
the same yeast and was considered to be gly-
L
cogen. As YKEMAet al. (1986) pointed out, if
20
1 the biosynthesis of the polysaccharide which,
a
P
2
like lipid, is a reserve storage material, could
0 0 be prevented then this would enhance the to-
0 0025 005 0075 01 tal amount of lipid producible with a cell.
Ollution rate (h-’) Although most studies on microbial lipid
Fig. 3. Typical lipid accumulation pattern for a accumulation have been conducted using
yeast (Rhodotorula glutinis) growing on nitrogen batch cultivation and, for accuracy, in contin-
limiting medium in continuous culture. Biomass W, uous culture, other growth systems have also
% lipid content 0 (from YOONand RHEE,1983). been explored. In particular, fed-batch cul-
ture has proved effective in increasing both Conversions of glucose and other carbohy-
the cell density and lipid contents of oleagi- drates including lactose and starch, to lipid up
nous yeasts: YAMAUCHIet al. (1983) used to 22% (w/w) have been recorded with a vari-
ethanol as substrate with Lipomyces starkeyi ety of yeasts (RATLEDGE,1982; YKEMAet
and achieved a biomass density of 150 g L-’ al., 1988; DAVIESand HOLDSWORTH, 1992;
with a lipid content of 54%. Similarly, PAN HASSANet al., 1993) which compares favora-
and RHEE(1986) achieved 185 g (dry wt.) of bly with the theoretical maximum of about
Rhodotorula glutinis per liter with a lipid con- 31-33% (RATLEDGE,1988). Somewhat lower
tent of 43% using glucose as the fed-batch yields appear to pertain with molds (WOOD-
substrate. In this latter case, Oz-enriched air BINE, 1959; WEETE, 1980). The reason for
(40% Oz + 60% air) had to be used to sus- this difference is not obvious though it may
tain the cells. At the density recorded, the be due to a somewhat slower growth rate of
packed cell volume was 75% of the total vol- molds than yeasts. It should be said, however,
ume of the fermentation medium. Without that there has not been the same amount of
using additional OZrit seems likely that cell detailed work carried out with molds as with
densities of up to 1OOgL-’ could be yeasts. Claims that microorganisms have
achieved with most oleaginous yeasts (see, achieved higher conversions of glucose or
e.g., YKEMAet al., 1988) though filamentous other sugars to lipid should be treated with
molds may pose other problems. Economic caution: either there will be found to be addi-
considerations, however, would probably be tional carbon within the medium and not tak-
against the use of OZ-enriched air for any en into the mass balance or, as may occasion-
commercial process. Interestingly, it is sug- ally happen, the “lipid” has been improperly
gested that higher rates of lipid formation extracted and may contain non-lipid material.
may occur with fed-batch techniques than However, if experimental data are calculated
with batch- or continuous-culture approaches so that the yield of lipid or fatty acids can be
(YKEMAet al., 1988). based on the fraction of glucose being used
At the end of the lipid accumulation phase solely for lipid biosynthesis, then values close
(see Fig. l), it is essential that the cells are to the theoretical value have been attained in
promptly harvested and processed. If glucose, practice (GRANGER et al., 1993).
or other substrate, has become exhausted on When ethanol is used as substrate, the the-
the end of the fermentation, then the organ- oretical yield of lipid is 54% (w/w) (RAT-
ism will begin to utilize the lipid as the role of LEDGE, 1988). Though only a 21% conversion
the accumulated material is to act as a reserve of ethanol to lipid was recorded by YAMAU-
store of carbon, energy, and possibly even CHI et al. (1983) in the fed-batch culture of
water. HOLDSWORTH and RATLEDGE(1988) Lipomyces starkeyi, higher conversions were
showed with a number of oleaginous yeasts recorded by EROSHINand KRYLOVA (1983)
that after carbon exhaustion following lipid also using a fed-batch system for the cultiva-
accumulation, the lipid began to be utilized tion of yeasts on ethanol: conversions of 26%,
within 1.5 h thus indicating the dynamic state 27%, and 31% were obtained using, respec-
of storage lipids in these organisms. tively, Zygolipomyces lactosus (Lipomyces te-
trasporus) and two strains of Cryptococcus al-
bidus var. aerius. The reason for these very
2.2 Efficiency of Accumulation high values lies in the efficiency by which
ethanol can be converted to acetyl- CoA, the
The efficacy of conversion of substrate to starting substrate for lipid biosynthesis (see
lipid has been examined in some detail in below). With glucose, the maximum yield of
both batch and continuous culture. In gener- acetyl-CoA can only be 2 mol per mol utilized
al, the latter technique offers the better whereas with ethanol the yield is 1 mol per
means of attaining maximum conversions as mol. On a weight-to-weight basis, therefore,
the cells are operating under steady state con- ethanol (MW 46) is almost twice as efficient
ditions and carbon is not used with different as glucose (MW 180) in providing Cz units.
efficiencies at each stage of the growth cycle. GRANGERet al. (1993) have recorded direct
conversions of ethanol to lipid (that is exclud- mulation occurs in this group of organisms. In
ing ethanol being converted to non-lipid bio- non-oleaginous organisms, one of the key en-
mass) of 42% (wlw) with Rhodotorula glutin- zymes, ATP-citrate lyase, does not occur and
is. consequently the formation of acetyl-CoA oc-
While most oleaginous microorganisms ac- curs by another route (see SHERIDAN et al.,
cumulate lipid equally well from a number of 1989) and does not lead to a lipogenic state
different carbon sources (see, e.g., EVANS being created.
and RATLEDGE,1983; YooN et al., 1982; In oleaginous microorganisms, the follow-
DAVIES and HOLDSWORTH,1992; HAM- ing sequence of events is considered to hap-
MOND et al., 1990), and do so without regard pen to cause lipid accumulation:
to the source of nitrogen used in the medium, (1) When the culture has consumed all avail-
a few yeasts are known which only accumu- able N from the medium, a nitrogen-scaveng-
late lipid when an organic source of nitrogen, ing process is initiated. This takes the form,
such as urea, glutamate, or aspartate, is used but there are likely to be other examples, of
(WITTERet al., 1974; EVANSand RATLEDGE, deaminating AMP via the enzyme AMP-
1984a, b). These yeasts appear to be mainly deaminase (see Reaction 1) which becomes
confined to strains of Trichosporon (Endo- activated at the point of N exhaustion
mycopsis) pullulans and Rhodosporidium to- (EVANSand RATLEDGE,1985~).
ruloides. A biochemical explanation for this
has been advanced based on nitrogen catabo-
AMP + IMP + NH3 (1)
lite inhibition of phosphofructokinase (see As a result of the activity of AMP deaminase,
Fig. 4) which is a key enzyme controlling the some NH3 is provided for the cell to help
rate of flux of glucose (as carbon substrate) to maintain protein and nucleic acid synthesis
acetyl-CoA (EVANSand RATLEDGE,1984~). but, simultaneously, the concentration of
The effect, though, is not explicable in terms AMP drops rapidly (BOULTONand RAT-
of a greater pH decrease with inorganic NH4+ LEDGE,1983a) and this is then the first major
salts than with glutamate or asparagine, as trigger in the lipogenic cascade mechanism.
has been suggested elsewhere (MORETON, (2) AMP is required as an activator of the
1988a). The same effect was produced in enzyme isocitrate dehydrogenase (Reaction
Rhodotorula gracilis using ammonium tar- 2) operating in the mitochondrion (EVANSet
trate as was produced with NH4Cl where the al., 1983). As a consequence of the absence of
former salt caused little downward pH drift AMP the reaction is unable to proceed as
but, without changing the biomass yield, in- part of the tricarboxylic acid cycle.
creased the lipid content of the cells from
18% to over 50% (EVANSand RATLEDGE,
+
Isocitrate NADP -,2-Oxoglutarate +
NADPH + CO2
+
1984a, c). (2)

(3) With the cessation or slowing down of
isocitrate dehydrogenase, isocitrate cannot be
2.3 Biochemistry of Accumulation further metabolized and both isocitrate and
citrate begin to accumulate. The equilibrium
The pathway of synthesis of any microbial lies in favor of citrate so as the assimilation of
product, or indeed plant or animal product, glucose continues unabated by the N-limited
needs to be understood so that it then be- cells, (BOTHAMand RATLEDGE,1979), ci-
comes possible to manipulate the cell to en- trate becomes a major product of its metabo-
hance the formation of that product. The for- lism (BOULTONand RATLEDGE,1983a).
mation of lipids is not different from any oth- (4) Citrate now exits from the mitochondrion
er product in this general concept. The path- in a malate-mediated citrate translocase reac-
way of triacylglycerol formation from glucose tion (see Fig. 4). Citrate is then cleaved by
is given in Fig. 4. The pathway accounts for ATP-citrate lyase (Reaction 3), an enzyme
lipid formation in oleaginous microorganisms which appears to be uniquely associated with
and, with the accompanying regulatory con- the lipogenic process. The dependency of iso-
trol mechanisms, can explain how lipid accu- citrate dehydrogenase (Reaction 2) upon
Fig. 4. Pathway of biosynthesis of triacylglycerols from glucose in oleaginous microorganisms. Numbers in

parentheses indicate approximate stoichiometry for conversion of glucose to fatty acyl-CoA as elucidated
in oleaginous yeasts (see also RATLEDGE,1988 RATLEDGEand EVANS,1989).
G3P Glycerol 3-phosphate (this is derived from 0.5 mol glucose during glycolysis)
AMP is; though, another metabolic feature, Therefore, if glucose is not used for the
possibly peculiar to the oleaginous organisms synthesis of any other product the yield of lip-
(EVANSand RATLEDGE,1985b, 1986). id is approximately 32 g per 100 g glucose.
Citrate + ATP + CoA Acetyl-
The role of ATP-citrate lyase in lipid accu-
+ + + mulation appears to be central. The enzyme
+
CoA Oxaloacetate ADP Pi (3) itself has been purified and partially charac-
(5) The acetyl-CoA from Reaction 3 serves terized from Rhodotorula gracilis (SHASHIet
as the primer for fatty acid synthesis. Howev- al., 1990). It has a M,of approx. 520 kDa and
er, in addition to a supply of C2 units, the cell comprises four identical subunits each about
must also provide NADPH as reductant for 120-130 kDa in size. Other properties of the
fatty acid synthesis. This is provided from the enzyme to establish its involvement in lipid
subsequent metabolism of oxaloacetate, first accumulation have been presented by Bo-
to malate via malate dehydrogenase, and then THAM and RATLEDGE(1979), BOULTONand
to pyruvate via malic enzyme (Reaction 4): RATLEDGE(l981,1983b), and by EVANS and
+
Malate NADP+ Pyruvate C 0 2
+ + + RATLEDGE(1985a, c). The enzyme appears
similar in overall size and structure to mam-
NADPH (4) malian ATP-citrate lyase (HOUSTON and
Some NADPH may also be supplied by NIMMO,1984, 1985). Microorganisms lacking
metabolism of glucose via the pentose phos- ATP-citrate lyase generally do not accumu-
phate pathway. late lipid above 10-15%. However, it is im-
The malic acid for the latter reaction is pre- portant to state that the corollary, that micro-
sumed to be by it leaving the mitochondrion organisms with ATP-citrate lyase will accu-
in exchange for pyruvate in a series of cou- mulate lipid, does not necessarily follow be-
pled transport reactions across the mito- cause other enzymes that function at key reg-
chondrial membrane (see Fig. 4). ulatory points such as AMP deaminase
The overall flux of glucose to fatty acyl- (Reaction l), the AMP-dependent isocitrate
CoA and then into triacylglycerol is given in dehydrogenase (Reaction 2) and malic en-
Fig. 4. The overall stoichiometry is approxi- zyme (Reaction 3) may be lacking or be un-
mately: der alternative control mechanisms.
15 Glucose + Triacylglycerol + 36C02 Malic enzyme (ME) itself is absent in some
I Enzymes:
PC: Pyruvate carboxylase (Pyruvate + C 0 2+ ATP + Oxaloacetate + ADP + Pi
PD: F'yruvate dehydrogenase (Pyruvate + CoA + NAD + -+ Acetyl-CoA+ C 0 2+ NADH)
cs: Citrate synthase (Acetyl-CoA + Oxaloacetate + Citrate + CoA)
ACL: ATP:citrate lyase
MDH: Malate dehydrogenase
ME: Malic enzyme
A CC: Acetyl-CoA carboxylase
FAS: Fatty acid synthase
PIMEX: Pyruvate/malate exchange (triple-linked system)
ICD H: Isocitrate dehydrogenase (NAD - requiring, AMP-dependent; see text Reaction 5)
+
CT: Citrate translocase

TC: Tricarboxylic acid cycle reactions
..
7. unknown (unspecified) reactions
Overall conversion (assuming the 9 mol of NADH required for the MDH reaction will be either from
glycolysis or the PD reaction):
+
4.5 Glucose + CoA + 9 NAD + + 7 NADPH + 17 ATP C18-fattyacyl-CoA 9 C 0 2+9NADH
+
+7 NADP++ 17 ADP+17 P,)

The production of 1 mol triacylglycerol, therefore, requires approx. 5 mol glucose as an additional 0.5 mol
glucose is needed to provide glycerol 3-phosphate (G3P) for the triacylglycerol and a further mol of glu-
cose must be oxidized to provide additional energy (ATP) and reducing power (NADPH).
oleaginous yeasts, notably Lipomyces spp. fungi; CARMAN and HENRY(1989) which co-
(BOULTON,1982), and here the supply of vers phospholipid biosynthesis in yeast; PIE-
NADPH is presumably by reactions of the RINGER (1989) covering the biosynthesis of
pentose phosphate cycle. ME activity has also non-terpenoid lipids from fatty acids; and
been found in oleaginous fungi and has also COOLBEARand THRELFALL (1989) review-
been implicated as the provider of NADPH ing the biosynthesis of terpenoid lipids.
for the desaturation of fatty acids (KENDRICK
and RATLEDGE,1992c) in Mucor circinello-
ides. Here, a second form of ME is associated
with the membranes of endoplasmic reticu-
lum - the microsomal fraction of the cell - 3 Triacylglycerols and
that then serves to provide NADPH directly
within the membranes. The transfer of reduc-
Fatty Acids
ing equivalents from the NADPH to the 02-
dependent desaturation reaction (Reaction 5 ) Fatty acids do not occur as such in living
may not be linked via cytochrome bs which is cells because of their inherent toxicity. Conse-
normally associated with such reactions in quently, fatty acids are esterified, usually as
other tissues (see Reaction 5). triacylglyerols (see Sect. 1.1, Structure 111), or
occasionally as wax esters (see Sect. 5 ) . Tri-
- CH, - CH,- acylglycerols (TAG) are produced as the ma-
n . n e p - m & desaturase
Fatty acid jor storage product of most oleaginous yeasts
Malate NADP+-
- CH = CH - and molds. Their proportion of the total lipid
2 H,O is usually over 80% (see RATLEDGE,1986)
and can be over 90%. Clearly, the proportion
(5)
of TAG will depend upon a number of factors
So far there have been no other reports of which would include the total amount of lipid
the occurrence of malic enzyme in microso- in the oleaginous cell (the more accumulated
ma1 membranes in other systems. lipid, the greater the amount of TAG) and
In non-oleaginous microorganisms, where also the method used for lipid extraction.
ATP-citrate lyase does not occur, acetyl-CoA More severe methods of extraction will re-
for fatty acid biosynthesis is generated from move both free and bound lipids whereas
the intramitochondrial pool by transfer of the gentler methods will extract only the freely
acetyl group via carnitine acetyl transferase soluble lipid which will be predominantly
(CAT). Oleaginous microorganisms also pos- TAGS. Thus, comparisons of lipid analysis,
sess this enzyme activity (RATLEDGEand carried out by different researchers using dif-
GILBERT,1985; HOLDSWORTH et al., 1988) ferent methods with different organisms,
though it is not immediately obvious why two must be treated with some caution. however,
routes to acetyl-CoA generation are neces- from a biotechnological viewpoint, the TAG
sary though it does emphasize that without content of the cell is usually of paramount im-
ATP-citrate lyase cells cannot presumably portance as this is the form in which an oil
provide sufficient acetyl-CoA under N-limit- will eventually be offered for sale. Should the
ing growth conditions to keep lipid biosynthe- TAG content of a cell be low, but the fatty
sis fully primed. Non-oleaginous microorgan- acids of interest, then total extraction of all
isms, therefore, tend to accumulate carbohy- fatty acyl lipids will have to be carried out
drate reserves when placed under the same and the fatty acids then recovered after hy-
growth conditions. drolysis. In these cases, the fatty acyl profile
The mechanism of fatty acid biosynthesis of the total lipid becomes the major determi-
itself from acetyl-CoA is well documented in nant of the potential usefulness of lipid. The
most standard text books. Recent reviews fatty acids can be esterified (usually methyl or
that may be consulted on this topic include ethyl derivatives) and then offered for sale.
those by SCHWEIZER (1989) which covers fat- Lipid extraction is fraught with a number
ty acid biosynthesis in bacteria, yeasts, and of difficulties and pitfalls for the unwary. The
article by RATLEDGE and WILKINSON (DHA). EPA has now been identified in the
(1988b) discussed these problems at some lipids (presumably phospholipids) of a num-
length and suggested methods that could be ber of species of marine bacteria: Alteromon-
used to minimize post-harvest changes in the as, Shewanella, Flexibacter, and Vibrio (RIN-
lipid composition of microbial cells. The pres- GO et al., 1992; AKIMOTOet al., 1990, 1991;
ence of partial acylglycerols (mono- and di- YAZAWAet al., 1988, 1992; HENDERSONet
acylglycerols) dong with free fatty acids in al., 1993), though it was first recognized in the
the extracts is indicative of faulty extraction lipids of Flexibacter polymorphus some years
procedures that have failed to subdue the la- ago by JOHNSand PERRY(1977). YAZAWA
tent activity of lipases and phospholipases of et al. (1992) carried out an extensive survey of
the cells. These enzymes function successfully some 24,000 bacteria isolated from various
in the very solvent systems that are used for fish and marine mammals. One isolate, which
lipid extraction and require rapid inactivation approximated taxonomically to Shewanella
(usually by heating) to ensure that the lipid putrefaciens, was found that produced EPA at
remains unchanged. Fractionation of ex- up to 40% of the total fatty acids; the lipid
tracted lipids that show the present of signifi- content of the bacterium was between 10 to
cant amounts of these hydrolysis products, 15% and the organism grew readily in the la-
and especially free fatty acids, are therefore boratory achieving 15 g (dry wt.) L - ' in 12-
seldom worth reporting and, as there are so 18 h. The content of EPA in dry cells was ap-
few good examples with oleaginous yeasts prox. 2%. Total hydrolysis of the extracted
and molds (see RATLEDGE,1986), the erudite lipids was needed to release the EPA from
reader is therefore referred to the more ex- the phospholipid fraction. EPA was found al-
tensive reviews of RATTRAY(1988) and LO- most exclusively at the sn-2 position of phos-
SEL (1988) who detail the lipid analyses of a phatidylethanolamine and phosphatidylglyce-
large number of yeasts and molds. This pres- rol. It was the only PUFA that was recog-
ent chapter, however, is concerned solely with nized in the total lipids; other unsaturated fat-
the biotechnological potential of microorgan- ty acids were 16:l (12%), 18:l (oleic+cis-
isms and consequently it is only those species vaccenic) (2%) and 17:l (6%). HENDERSON
that are prolific in the production of desirable et al. (1993), using a marine Vibrio isolate,
oils or fatty acids that will be considered showed that EPA formation was higher (at
here. 9% of the total fatty acids) when the cells
were grown at 5°C than 20"C, whereas YAZA-
WA et al. (1992) had only used 10-15°C for
3.1 Bacteria the growth of their bacterium. The gene(s) re-
sponsible for EPA production have been
Although bacteria do not produce triacyl- transferred into E. coli with the resultant pro-
glycerols and are usually considered as duction of EPA in this bacterium (WATA-
sources of novel lipids (see Sect. 5), they nev- NABE and YAZAWA, 1992).
ertheless are of current interest in that some DHA was first recognized in several ma-
species are known that produce polyunsatu- rine bacteria by DELONG and YAYANOS
rated fatty acids (PUFA) of dietary or even (1986). Like EPA, DHA was exclusively asso-
pharmaceutical importance. ciated with the phospholipids in all cases.
However, DHA was predominant in bacteria
grown at 2°C and at very high pressures
3.1.1 Polyunsaturated Fatty Acids (- lo5 Pa) making them unlikely candidates
for biotechnological exploitation. More re-
in Bacteria cently, YANOet al. (1994) have reported sim-
ilar findings with a further five deep-sea bac-
Several bacterial species have been recently terial isolates. Again there was a strong de-
isolated from marine sources that synthesize pendency on high pressures and low tempera-
either eicosapentaenoic acid (20: 5 0-3) tures for DHA formation though, for the first
(EPA) and docosahexaenoic acid (22:6 0-3) time, both DHA and EPA, together with
traces of 20:4 (0-3) and 20:3 (w3), were ly, the opportunities for exploiting yeast oils
noted. as a commercial possibility are limited and
The commercial values of EPA and DHA other more expensive targets have had to be
as individual fatty acids are unclear as they identified. One major target that has been
are both readily obtainable as a mixture in identified is to produce a yeast oil as a cocoa
fish oils which are relatively cheap. It is clear, butter equivalent (CBE). The fatty acid com-
however, that the bacterium described above position of cocoa butter is given in Tab. 1.
by YAZAWAet al. (1992) - Shewunellu putre-
fuciens - could represent a valuable source of
EPA but the commercial potential for EPA 3.2.1 Production of a Cocoa
and DHA would both be considerably en-
hanced if it should be shown that single fatty Butter Equivalent Yeast Fat
acids were required for medical purposes.
WATANABE et al. (1994) have recently shown CBE fats are used extensively in the con-
that isolated DHA (and by inference EPA fectionery business: in the manufacture of
would act similarly) can be readily incorpo- cooking chocolate and chocolate-type materi-
rated into bacterial phospholipids, including als and it can also be included, to an agreed
those of E. coli and also Rhodopseudomonus percentage - usually 5% - of cocoa butter it-
cupsulutu which is currently used for the pro- self, in chocolate manufacture in several
duction of larvae of marine fish. In this way countries including UK, Ireland, and Den-
DHA could be delivered to the developing mark (KERNON,1992). CBE materials have
larvae and potentiate their growth rate. DHA to have similar physical characteristics to co-
though at present must be obtained from fish coa butter itself and currently they are manu-
oils though mold and algae sources of it are factured from palm oil by fractional crystalli-
known (see Sects. 3.3.5 and 3.4.4). If recombi- zation though they can also be produced by
nant DNA technology could be used to in- enzymic transesterification of palm oil and
crease the levels of DHA and EPA in more stearic acid or its esters (WILLNERet al.,
easily cultivatable bacteria (see WATANABE 1993; OWUSU-ANSAH, 1993).
and YAZAWA,1992), or even yeasts, this The requirements for a satisfactory CBE is
could open up new and exciting horizons for that there should be approximately equal
these PUFAs. At the moment, however, EPA amounts of stearic acid (18:0), oleic acid
production by bacteria is someway off and (18: l), and palmitic acid (16:O) attached to
that of DHA would appear almost impossi- the glycerol moiety with the unsaturated acid
ble. Other sources of EPA, however, include being in the sn-2 position. This is known as
both molds and algae (see Sects. 3.3 and the POSt-TAG (1-palmitoyl-2-oleoyl-3-stearo-
3.4). yl-sn-glycerol). The price of cocoa butter it-
self has varied from $8600 per t in the mid
1980s (MORETON, 1988b;SMITet al., 1992) to
3.2 Yeasts about $3OOO per t in 1994. As CBEs com-
mand a price of about 80% of cocoa butter, it
The number of oleaginous yeasts (Tab. 3) is can be appreciated that at the higher price
relatively small in comparison to the total level, a yeast CBE would be an attractive eco-
number of species - 590 (see BARNETTet al., nomical proposition. Not surprisingly, a num-
1990). Besides producing triacylglycerols, ber of developments took place in the mid
these yeasts also usually produce between 2- 1980s using appropriate yeast technology with
10% of their total lipid as phospholipids with this objective in mind. Simultaneously, the at-
smaller amounts of sterol and sterol esters tractive price of cocoa butter and of CBEs
(RATLEDGE, 1986). The fatty acid profile of stimulated the current interest in lipase-cata-
the yeasts lipids (Tab. 3) is typically that of lyzed transformation of oils and fats. Howev-
several commercial plant oils (see Tab. 1) er, with the present low price of commercial
and, as such, would not command a high cocoa butter and CBEs, the yeast route has
price: say, maximally $600 per t. Consequent- been deemed to be non-profitable and even
Tab. 3. Lipid Contents and Fatty Acid Profiles of Oleaginous Yeastsa
Yeast Speciesb Maximum Major Fatty Acyl Residues Others
Lipid [Relative % (w/w)]
Content
["A (w/w)] 16:O 16:l 18:O 18:l 18:2 18:3
~ ~ ~~~
Candida diddensiae 37 19 3 5 45 17 5 18:4 (1%)

Candida sp. 107 42 44 5 8 31 9 1
Cryptococcus albidus 65 12 1 3 73 12
var. aerius
C. albidus 65 16 trace 3 56 -
var. albidus
C. curvatus D' 58 32 - 15 44 8
C. curvatus R' 51 31 - 12 51 6
C. laurentii 32 25 1 8 49 17
Endomyces (Endomycopsis) 28 17 19 1 36 25
magnusiid
Galactomyces geotrichume 50 no record'
Williopsis saturnusf 28 16 16 - 45 16 5
Waltomyces lipoferg 64 37 4 7 48 3 -
Lipomyces starkeyi 63 34 6 5 51 3
L. tetrasporus 67 31 4 15 43 6 1
(Zygolipomyces lactosus)
Rhodosporidium toruloides 66 18 3 3 66 - - 23:O (3%)
24:O (6%)
Rhodotorula glutinis 72 37 1 3 47 8 -
R. graminis 36 30 2 12 36 15 4
R. mucilaginosa 28 no record
Trichosporon .beigeliih 45 12 - 22 50 12 -
T.fermentans' 20 17 1 4 42 34 trace
T. pullulans' 65 15 - 2 57 24 1
Yarrowia lipolyticak 36 11 6 1 28 51 1
a A lower lipd content of cells for inclusion in the list has been taken as 20%. Data taken from RATLEDGE
and EVANS(1989) and RATLEDGE(1989b) except for Cryptococcus albidus var. aerius where original
data of ZVYAGINSTEVA et al. (1975) - see RATLEDGEand EVANS (1989) - is used. Not included in the
table and of uncertain oleaginicity are:
Candida (Pichia) guilliermondii (22% lipid), C. methylica (20%), C. stellatoidea (C. albicans) (20%), C.
tropicalis (23%), Cryptococcus (Filobasidiella) neoformis (22%), Hansenula (Pichia)ciferri (22%), Lipo-
myces spp. (two unspecified species: 59 and 67% lipid), Schwanniomyces occidentalis (23%), and Trig-
onopsis variabilis (an unusual yeast which only accumulates lipid - up to 40% - if grown with methio-
nine in the medium).
Nomenclature given accordingly to BARNETTet al. (1990).
Although Candida curvata strains D and R are often referred to as Apiotrichum curvatum, this name is
not recognized as such. BARNETTet al. (1990) consider C. curvata to be a synonym of Cryptococcus
curvatus which is now the recommended name.
Endomyces magnusii. This name is of dubious status and may be the imperfect stage of Trichosporon
pullulans (q.v.)according to KREGER-VAN RIJ (1984). It is not listed by BARNETTet al. (1990).
' Formerly Geotrichum candidum. Although this organism was quoted as containing up to 50% lipid in
the 1930s (see HESSE,1949; WOODBINE,1959) there have been no recent reports of such levels being
repeated. The original oleaginous strains may, therefore, be lost.
Formerly Hansenula saturnus.
g Formerly Lipomyces lipofer.
h Formerly Trichosporon cutaneum.
' Formerly Trichosporon fermentans.
1 Trichosporon pullulans possibly includes Endomycopsis vernalis (see KREGER-VAN RIJ, 1984) though
yeast correspond to this latter description appears to be no longer available.
Yarrowia lipolytica is variously referred to as Candida lipolytica and Saccharomycopsis lipolytica
amongst other names. Only a few strains may be oleaginous.
the lipase route must also be under severe fi-

nancial pressure. Nevertheless, the approach 0
used to achieving a yeast CBE is illustrative 2
of what can be achieved given limited re-
sources and manpower in trying to produce a
marketable biotechnological bulk product.
Four different approaches have been tried
to produce a satisfactory yeast CBE. The
main difficulty to be overcome has been how
to increase the inherently low content of
stearic acid (maximally from 10 to 12%) in
oleaginous yeasts (see Tab. 3) up to the re-
quired 30% stearate or more.
3.2.1.1 Direct Feeding of Stearic

Acid
The simplest way to increase the stearic
acid content of a yeast oil is to feed stearic
acid or its ester to the yeast. The fatty acid or
its ester is then taken up by the yeast and, al-
though some may be degraded, the bulk
seems to be directly esterified into the storage
triacylglycerols in the cell. Typical results of
such efforts are shown in Tab. 4. Principal
commercial concerns that attempted this
route were Fuji Oil Co. Ltd. (1979, 1981) and
CPC International Inc. (1979, 1982a, b). Best
results were obtained by feeding both palmit-
ic acid and stearic acid, as their methyl esters,
to Torufopsis ATC C 20507 (Fuji Oil Co.
Ltd., 1979, 1981). However, even this oil re-
quired fractionation to produce a satisfactory
CBE and this then significantly added to the
costs.
The obvious problem with this approach is
the cost of the stearic acid or the stearate -
palmitate mixture. Stearic acid itself is usually
produced by chemical hydrogenation of oleic
acid, which in turn is usually derived most
cheaply from animal fats. Unfortunately, this
origin of stearic acid then negates any claim
that may be made for the yeast CBE being
wholly derived from non-animal sources and
would make it unacceptable for vegetarians
and some religious groups. This situation has
now, however, changed with the advent of
high-oleic acid sunflower oil (see Tab. 2). For
the first time, it is therefore possible to pro- 0"0"?!
duce stearic acid from oleic acid at a relative- 22222
DS-9 DS-12 DS-15
4
180 -b 181 (9) 182(9,12) -b 183(9,12,15)
im4 4
stearic oleic acid linoleic acid a -1inolenic acid
acid
DS-6
DS-6
18:2 (6,9) 183 (6,9,12) 184 (6,9,12,15)
octadecadienoic Y-linolenic acid octadecatetraenoic
202 (8,ll) 203 (8,11,14) 204 (8,11,14,17)
1
eicosadienoic dihomo-y-linolenic eicosatetraenoic
acid acid (DHGLA) acid
DS-5
1
I
DS-17 DS-5
203 (5,8,11) 204 (5,8,11,14) + 205 (5,8,11,14,17)
eicosatrienoic arachidonic eicosapentaenoic
acid (ETA) acid acid
(“Mead Acid”) (AM) (El”
EL
225 (7,10,13,16,19)
Fig. 5. Formation of polyunsatu- docosapentaenoic
rated fatty acids in fungi. DS-n: fat- acid (DPA)
ty acyl desaturase acting at nth C
atom of fatty acid; EL: elongase us-
ing acetyl-CoA. The DS-17 is only
found in certain filamentous fungi;
1-
226 (4,7,10,13,16,19)
it is not found in animals. It could docosahexaeneoic acid
though conceivably also convert (DHA)
DHGLA (dihomo- ylinolenic acid) 0-9 0-6 0- 3
to eicosatetraenoic acid. series series series
ly low cost and still being classed as a plant fatty acyl group being detached to a phos-
oil. pholipid (KENDRICKand RATLEDGE1992c,
In spite of the development of a cheap, d)
plant source of stearic acid, this approach to a
yeast CBE appears to have been abandoned
though it did establish that yeast oils could be 3.2.1.2 Inhibition of Stearoyl
produced with an unprecedented high con- Desaturase
tent of saturated fatty acids.
The remaining three approaches to pro- The naturally occurring sterculic acid, cis-
duce a yeast CBE have all sought to limit the 9,10-methyleneoctadecenoic acid (V), which
conversion of stearic acid to oleic acid within is found in the seed oil of sterculia and kapok
the yeast cell. This reaction (see Reaction 5, plants is an effective inhibitor of the A 9 desa-
Sect. 2.3), functions at the level of the coen- turase.
zyme A derivatives of the fatty acids and re- CH3-(CH2)7-C=C-(CH2)7COOH
quires molecular O2 as well as a supply of re- \ /
ducing equivalents (NADPH). It is catalyzed CH2
by A 9-stearoyl desaturase. The subsequent Sterculic acid
desaturations (see Fig. 5) are carried out with V
Tab. 5. Effect of A9- and Al2-Cyclopropene Fatty Acids on the Fatty Acyl
Composition of Rhodosporidium toruloides IF0 0559 (from MORETON,
1988b)
Relative Fatty Acyl Composition

["h (w/w)]
14:O 16:O 16:l 18:O 18:l 18:2 18:3
Control (no additions) 1.7 27.8 0.3 7.3 40.0 17.5 4.6
Sterculia oil"0.3 mL L-' 0.7 14.9 - 48.3 18.5 9.1 3.7
A12-Cyclopropeneb0.4 mL L-' 1.0 36.0 0.3 8.3 44.7 1.8 -
Sterculia oil" + 0.3 mL L-' - 19.5 - 46.9 21.8 4.7 1.5
A12-Cyclopropeneb0.3 mL L-'
" Contains 50% (w/w)of the A9-cyclopropene c18:1

Al2-Cyclopropene c18:1
MORETON(1985) successfully demonstrated biotechnology process designed to produce

that as little as 100 mg sterculic acid per L of an edible oil was very uncertain. Clearly, reg-
growth medium was effective inhibiting the ulatory authorities would be extremely cau-
reaction in a number of yeasts: Cundidu sp. tious in allowing a yeast CBE to be used in
107, Rhodosporidium toruloides, and Tricho- foods that had some possibility of containing
sporon cutuneum. Stearic acid then accumu- any residual inhibitor. This approach, which
lated up to 48% of the total fatty acids (see had been pioneered by Cadbury-Schweppes
Tab.'5). However, the effect of the inhibitor plc, the large UK-based chocolate manufac-
was extremely specific and did not affect the turer, was abandoned in 1986.
subsequent conversion of oleic acid to linoleic
acid (18:2) via the action of the A 12 desatur-
ase enzyme. Consequently, the proportion of 3.2.1.3 Mutation
18:2 in the yeast oil was unaffected by ster-
culic acid, and this was detrimental to the Mutation and genetic manipulation of bac-
properties required of the CBE lipid. MORE- teria are now commonplace; haploid yeasts
TON (1988b) subsequently showed that when and molds are also similarly mutatable
the cis-A 12 analog of sterculic acid which had though, of course, many yeasts are diploid or
to be chemically synthesized, was added to aneuploid and would thus be not amenable to
the yeasts this now had the desired effect of simple mutational strategies. Nevertheless
decreasing the linoleic acid content (see and without ever apparently assessing wheth-
Tab. 5 ) . In the presence of both sterculic acid er their chosen yeast was haploid, aneuploid,
and cis-12,13-methyleneoctadecenoic acid, diploid, or polypoid SMITet al. at the Free
the oil of R. toruloides (the best of the yeasts University of Amsterdam embarked upon an
examined) was now almost exactly as re- ambitious project to delete the stearoyl desat-
quired: the three principal fatty acids, palmit- urase from the yeast Cryptococcus curvutus
ic, oleic, and stearic, were present at a ratio of (YKEMAet al., 1989, 1990; VERWOERTet al.,
1:1:2 (MORETONand CLODE,1985). 1989; SMITet al., 1992). (This yeast was origi-
Although this approach of MORETON nally isolated by HAMMONDet al. from dairy
(1985, 1988b) was scientifically very success- wastes as a yeast that would readily grow on
ful in meeting its objectives, the costs of the lactose, the principal carbohydrate of whey,
cyclopropene inhibitors were to prove with whey being judged to be a potential
beyond what the economics of the process cheap and plentiful substrate (MOON et al.,
could accommodate. Further, the acceptabili- 1978). Initially, the yeast was named as Cun-
ty of using known metabolic inhibitors in a didu curvucu and then reclassified as Apiotri-
chum curvatum which is now recognized as maintain the sharp melting point transition of
Cryptococcus curvatus (BARNETT et al., a CBE. To decrease the content of 18:2 and
1990).) 18:3 in the yeast oil, a further mutation pro-
In the work of SMITet al., C. curvata was gram would have been required but this was
treated with chemical mutagens and a num- never carried out.
ber of auxotrophic mutants isolated that re- From the initial oleate-auxotroph, Ufa 33, a
quired oleic acid for growth. Such mutants number of partial revertants and hybrids were
would thus be unable to produce unsaturated subsequently produced (YKEMAet al., 1990
fatty acyl groups for incorporation into their VERWOERTet al., 1989) that no longer re-
phospholipid membrane structures and would quired oleic acid to be added to the growth
be unable to maintain membrane fluidity and medium: in other words the A9-desaturase
growth. It was presumed, but never shown by was now partially functional allowing a small
direct enzyme assay, that these mutants amount of oleic acid to be synthesized inside
would have been affected in the gene coding the cell. The fatty acid profile of these yeasts
for the A 9 desaturase. One of these mutants, (Tab. 6) all contained substantial amounts of
Ufa 33, now contained 50% stearic acid (see stearic acid and, in some cases, notably that of
Tab. 6). However, as the mutant required the the hybrid F33.10, with excellent similarities
addition of oleic acid to the medium, it was to cocoa butter itself.
obvious from the fatty acid analysis that the Similar mutational programs have been re-
412- and Al5-desaturases (forming linoleic ported by BEAVANet al. (1992), working for
and then linolenic acids, respectively) were Diversified Research Laboratories as a sub-
unaffected in this mutant. Although small sidiary company of G. Weston Ltd., Canada,
amounts of the 18:2 and 18:3 fatty acids can and by HASSAN et al. (1993, 1994a) working
be tolerated in CBE preparations their in the group of GERARDGOMA,Toulouse,
amounts need to be very low in order to France. BEAVANet al. (1992) isolated 6,725
Tab. 6. Cocoa Butter Fatty Acids and the Fatty Acyl Compsition of Triacyl-
glycerols from Cryptococcus curvatus wild type (WT) strain, two unsaturated
fatty acid auxotrophic mutants (Ufa 33 and Ufa M3), a revertant mutant
(R22.72), and a hybrid (F33.10), both derived from Ufa 33 compared to the
best results obtained with the wild-type strain grown with limited O2 supply do
diminish desaturase activity
Major Fatty Acyl Groups

[Relative % (w/w)]
16:O 18:O 18:l 18:2 18:3 24:O
Cocoa butter 23-30 32-37 30-37 2-4 - trace

Yeast
WT 17 12 55 8 2 1
Ufa 33" 20 50 6 11 4 4
Ufa M3b 26 37 22 8 4 2
R22.72' 16 43 27 7 1 2
F33.10 24 31 30 6 - 4
WT-NZ' 18 24 48 3 1 2
a Grown with 0.2 g L - ' oleic acid; from YKEMAet al. (1990).
Grown with 0.2 g L-' oleic acid; from HASSAN et al. (1993, 1994a).
' Grown without oleic acid; from YKEMAet al. (1990).
Grown without oleic acid; from VERWOERT et al. (1989).
' Wild type grown on whey lactose in a 500 L bubble column fermenter with a
restricted O2 supply; from DAVIESet al. (1990).
yeasts (which would include many identical position (see Tab. 6). Significantly and inter-
organisms) of which one organism, DRL- estingly, this approach by DAVIES et al.
D221, was taken as the best with respect to (1990) also served to decrease the contents of
the rate of lactose utilization, lipid produc- the polyunsaturated fatty acids (18:2 and
tion, yield, and composition. The organism 18:3) which, of course, also require O2 in
was tentatively identified as Trichosporon cu- their formation. Thus by a single and obvious
tuneum which was originally suggested as an metabolic manipulation, the goal of a high-
identification of C. curvutu. A mutant of this stearate yeast oil was almost perfectly
yeast, DRL-JF34, contained 19% 16:0, 34% achieved.
18:0, and 23% 18:1. The work of HASSANet
al. (1993,1994a) used the same yeast as SMIT
et al.; their results are also shown in Tab. 6. 3.2.1.5 Conclusions
Although all three groups engaged on this
work produced greatly increased proportions The pursuit of yeast oil as a CBE has been
of stearic acid in the yeast oils, without dimin- now ceased after a decade of intensive activi-
ution of the lipid content of the cells and ty. DAVIES(1984) was the first to appreciate
without significant decrease in the overall the potential of this as a commercial target
growth performance of the yeasts, none of the and to carry out a sustained program to this
mutants have yet been used in commercial end. Of key significance was the appreciation
trials to produce CBE oils. Large-scale trials that a cheap substrate would be essential for
with mutants are, however, essential as the success as, from the previous sections (Sect.
characteristics of these organisms are often 2.2), approximately 5 t of glucose or equival-
not fully revealed until they are grown in ent carbohydrate are needed to produce 1 t of
large fermenters. Thus, the use of mutants to oil. DAVIES,by working in New Zealand,
produce yeast oil CBEs remains a potential, quickly realized that the waste whey gener-
rather than a proven, approach to this goal. ated from the extensive dairy industry of that
country could represent such a source of fer-
mentable carbohydrate. The earlier discovery
3.2.1.4 Metabolic Manipulation of the yeast C. curvutu (MOONet al., 1978)
that could be readily grown on lactose, which
The final approach to increasing the stearic is the major carbohydrate of whey, quickly in-
acid content of yeast oils has been to control dicated that it was probably the most likely
the amount of O2 entering the fermentation. one to use in this process. DAVIESet al. (see
As mentioned above (see Reaction 5, Sect. DAVIES, 1988, 1992a, b, c; DAVIES and
2.3), O2 is an essential co-reactant in the stea- HOLDSWORTH, 1992) then developed a pro-
royl desaturase reaction. Accordingly, DAV- cess up to a pilot-plant level of 200 m3 using a
IES et al. (1990) carried out a series of fer- bubble column fermenter with Cundidu cur-
mentation runs with C. curvufu in which the vatu and casein (milk) whey as its substrate.
oxygen uptake rate was progressively de- From those extensive trials, DAVIESwas able
creased by the simple expedient of restricting to calculate the probable economics of a yeast
the air supply (Tab. 6). Best results were ob- CBE process. This was based on the use of
tained with a 500 L fermenter in which the O2 six, similar sized reactors which would oper-
supply could be effectively regulated. (The ef- ate continuously and with a cell recycling
fect of O2 deprivation is extremely difficult to mode of operation to allow yeast densities of
demonstrate with conventional 5 L laboratory 50 g dry cells per L to be achieved. Recovery
fermenters as the amount of air that needs to of the oil from the yeast was also examined
be supplied to produce the effect is so small and the economics of this process taken into
that the usual air control systems are unsuita- account in the final calculations. A summary
ble; C. RATLEDGEand D. GRANTHAM, un- of DAVIES’costings is given in Tab. 7. The
published work.) The highest levels of stearic assumption is made that the yeast CBE prod-
acid so obtained by DAVIESet al. (1990) were uct would sell for approximately 80% of the
not far from the required cocoa butter com- price of cocoa butter and that no significant
Tab. 7. Yeast CBE Process: Estimated Operating Budget (from DAVIES
and
HOLDWORTH, 1992)
Basis of process
Available whey 200,000 m3/a-'
Lactose content 39% (WIV)
Cost of whey (in New Zealand) $ 0.5 per m3
CBE yield 0.16 kg kg-' lactose
CBE production 1,250 t a - '
Value of CBE $ 2,400 t -'
Total sales value of CBE $3,000,000
costs
Direct manufacturing costs
(untilities, substrate costs, downstream
processing, wages, etc.) $l,OoO,000
Manufacturing overheads
(laboratory costs, site charges, effluent
disposal, maintenance, insurance,
service overheads, etc.) $46O,OOo
Finance and sales
(distribution, research, and development) $300,000
Plant depreciation over 10 years
on capital of $ 5.4 M $540,000
Interest at 12% $650,000
Total costs = $ 2.95 M
Total sales = $ 3.0 M
Profit = $ 50,000
costs would be involved in toxicological trials the year to enable continual operation of the
before the CBE could be offered for sale. plant. Costs of labor and utilities are obvious-
DAVIES(1992a) and DAVIESand HOLDS- ly cheaper in some countries than in others
WORTH (1992) assumed that a fermentation and, consequently, it would not be surprising
plant would have to be built specifically for if this yeast CBE process or a similar one was
the process. Clearly, if a fermentation plant not adopted somewhere in the world over the
already existed and could be used without next decade. Only the continuing low price of
major modification, then this would signifi- cocoa butter will prevent it from becoming an
cantly decrease the indirect costs and the operating reality.
whole process might then show a significant It has recently been reported by Roux et
annual profit. al. (1994) that some species of Mucor, espe-
Although DAVIES'cost analysis will ob- cially M . circinelloides, will simultaneously
viously change from country to country and produce a CBE-SCO as well as producing a
will be heavily influenced by the cost of sub- valuable polyunsaturated fatty acid (y-lino-
strate, the overall conclusion is that this yeast lenic acid). This is described later in Sect.
CBE process is a process that is waiting for its 3.3.1.
day to come rather than being another uneco-
nomic biotechnology pipe dream. Clearly, a
number of other substrates can be used be- 3.3 Molds
sides whey (see BEDNARSKIet al., 1986;
GLATZet al., 1985, VEGAet al., 1988; FALL A list of oleaginous species of mold is given
et al., 1984; GUERZONIet al., 1985; DOSTA- below.
LEK, 1986; HASSANet al., 1994b) though it is
essential that these be available throughout Lipid contents of oleaginous molds grown in
the vegetative mycelial state are given in par- chea pulchella (27); Myrothecium sp.
entheses (taken from RATLEDGE, 1986 (30); Sclerotium bataticola (46)
1989a, 1993; LOSEL, 1988 and additional ref- Hymenomycetes
erences as indicated). Lepista (Tricholoma)nuda (48)
Ustilaginomycetes
Zygomycetes Ustilaginales
Entomophthorales Sphacelotheca reiliana (41); Tilletia con-
Conidiobolus nanodes (26); Entomoph- troversa (35); Tolyposporium ehrenbergii
thora conica (38); E. coronata (43); E. (41); Ustilago zeae (59)
obscura (34); E. thaxteriana (32); E. viru- Uredmiomycetes
lenta (26); Glomus caledonius (72) Cronartium fusiforme (28); Puccinia
Mucorales coronata (37)
Absidia corymbifera (27); A. spinosa
(28); Blakesleea trispora (37); Cunningh- For inclusion in the list, the lower cut-off of a
amella echinulata (45); C. japonica (60); lipid content of 25% has been used. There are
C. elegans (44, 56); C. homothallica (38); numerous molds that could have been listed if
Mortierella alpina (33) (TOTANIet al., the limit had been dropped to 20% and, in-
1992); M. elongata (34) (BAJPAIet al., deed, the commercial value of mold lipids, as
1992); M. isabellina (63-86); M. pusilla will be explained below, could still be high
(59); M. vinacea (66); Mucor albo-ater even with molds having lipid contents of less
(42); M. circinelloides (65); M. hiemalis than 20%. Commercialization of mold oils,
(42) (KENNEDYet al., 1993); M. miehei therefore, unlike yeast oils, does not depend
(25); M. mucedo (51); M. plumbeus (63); so much on the amount of oil that a mold pro-
M. pusillus (26); M. ramanniana (56); M. duces but on the quality of that oil. The qual-
spinosus (47); M. vinacea (25) (HANSSON ity of the oil is determined by its fatty acid
und DASTALEK, 1988); Phycomyces. profile (Tab. 8): some fatty acids, particularly
blakesleeanus (33); Rhizopus arrhizus the polyunsaturated ones (PUFA) that have
(32-57); R. delemar (32-45); R. oryzae nutritional and some medical importance, are
(57); Zygorhynchus moelleri (40) therefore select targets for current develop-
Peronosporales ments in this field.
Pythium irregulare (42); P. ultimum (48) The number of oleaginous species of mold
Ascomycotina is greater than the number of oleaginous
Ascomycetes yeasts (cf. Tab. 3 and the list given above) but
Aspergillusfisheri (53); A.flavus (35); A. this is hardly surprising considering that there
minutus (35); A. nidulans (51, 25); A. are some 100 times more molds (60,000 ap-
ochraceus (48); A. oryzae (37-57); A. ter- prox.) than yeasts. Although most of the 590
reus (57); A. ustus (28); Chaetomiumglo- species of yeast have probably been assessed
bosum (54); Fusarium bulbigenum (50); for oleaginicity, it is likely that most molds
F. equiseti (48); F. graminearum (31); F. have not. Therefore, one may expect the list
lini (35); F. lycopersicum (35); F. oxyspo- of oleaginous molds will be considerably in-
rum (29,34); Fusarium sp. NII (39); Gib- creased as further work continues to be car-
berella fujikuroi (F. moniliforme) (48); ried out. Extensive reviews of fungal lipids
Humicola lanuginosa (75); Penicillium have been prepared by LOSEL (1988) and
gladioli (32); P. javanicum (39); P. lila- WEETE(1980).
cinum (51,56); P. soppii (40); P. spinulo- As a generalization, molds produce higher
sum (64); Stilbella thermophila (38) levels of the polyunsaturated acids 18:2 and
Clavicipitacae 18: 3, than yeasts. Although there are various
Claviceps purpurea (31-60) isomeric possibilities, the fatty acids in yeasts,
Tulasuellales molds, and algae appear to be the same as
Pellicularia practicola (39) those found in plants. With linolenic acid
Hyphomycetes (18:3), two isomers occur in both plants and
Cladosporium herbarum (49); Malbran- microorganisms. The more common isomer is
Tab. 8. Fatty Acid Analyses of Lipid from Selected Molds (data taken from RATLEDGE(1986,1989a) and
LOSEL(1988) and additional references indicated)
Organism Major Fatty Acyl Groups Others

[Relative % (w/w)]of
14:O 16:O 18:O 18:l 18:2 18:3
Entomophthorales
Conidiobolus nanodes" 1 23 15 25 1 4' 20:l (13%)
22:l (8%)
20:4 (4%)
Entomophthora coronata 31 9 2 14 2 1' 12:o (40%)
E. obscura 8 37 7 4 trace trace 1 2 0 (41%)
Mucorales
Absidia corymbifera 1 24 7 46 8 10' -
Cunninghamella japonica trace 16 14 48 14 8' -
Mortierella alphab - 19 8 28 9 8' 20:3 (7%)
20:4 (21%)
M.elongata' - 7 2 18 12 25 ' 20:4 (16%)
205 (150/)
M. isabellina 1 29 3 55 3 3' -
Rhizopus arrhizus 19 18 6 22 10 12' -
Mucor alpina-peyron" 10 15 7 30 9 If 20:o (8%)
20:3 (6%)
20:4 (5%)
Peronosporales
Pythium ultimumd 7 15 2 20 16 1* 20:l (4%)
20:4 (15%)
205 (12%)
P. irregulare' 8 17 2 14 18 - 20:1 (5%)
20:4 (11'3'0)
20:5 (14%)
Ascomycetes
Aspergillus terreus 2 23 trace 14 40 21 -
Fusarium oxysporum trace 17 8 20 46 5 -
Pellicularia practicola trace 8 2 11 72 2 -
Pennicillium spinulosum - 15 7 42 31 1 -
Hyphomycetes
Cladosporium herbarum trace 31 12 35 18 1 -
Ustilaginales
Tolyposporium ehrenbergii 1 7 5 81 2 - -
Calvicipitacae
Claviceps purpurea trace 23 2 19 8 - 12-HO-
18:l (42%)
a KENRICKand RATLEDGE (1992a)

YAMADAet al. (1992)
BAJPAIet al. (1992)
GANDHIand WEETE(1991)
O'BRIENet al. (1993)
' y-linolenic acid, 18:3 (0-6)
Tab. 9. Fatty Acid Profiles of a Commercial Fungal Oil Product Compared with Evening Primrose Oil,
Borage Oil, and Blackcurrant Oils Containing y-Linolenic Acid
Mucor Evening Borage Blackcurrant

circinelloides" Primrose Seed Oil Seed Oil
Seed Oil
Oil content 20 16 30 30
["h (w/w)]
Fatty acid
16:O 22-25 6-10 9-13 6
16:1 0.5-1.5 - - -
18:O 5-8 1.5-3.5 3-5 1
18:l 3841 6-12 15-1 7 10
18:2 10-12 65-75 3741 48
y-18:3 15-18 8-12 19-25 17
(~-18:3 0.2 0.2 0.5 13
20: 1 - 0 4.5 -
22:l - - 2.5 -
24:O - - 1.5 -
a Production organism used by J & E Sturge Ltd., Selby, Yorkshire, UK, from 1985-1990.
a-linolenic acid, 18:3 (c 9, c 12, c 15), or 18:3 of this acid. Epoxy and dihydroxy fatty acids
(03), - see Sect. 1.1- which is found in most are also found in relative abundance in the
plants seed oils, yeasts, and the majority of lipids extracted from the spores of some ba-
molds. The more unusual isomer is y-linole- sidiomycetes. Some acetylenic acids may also
nicacid-18:3 ( c 6 , ~ 9 , ~ 1 2 ) o r 1 8 : 3 ( ~ 6 )occur
- (LOSEL, 1988). The occurrence of hy-
which occurs in the seed oils of Oenothera droxy fatty acids in fungi has been recently
(evening primrose), Ribes (blackcurrant, red- reviewed by VAN DYKet al. (1994).
currant, etc.) and the borage family (Boragi- Although molds contain an exceptional
naceae) (see Tab. 9). It also occurs through- diversity of fatty acids, current commercial in-
out the lower fungi, also known as phycomy- terest centers principally upon the formation
cetes (see Tab. 8). Both isomers occur simul- of particular PUFAs that have dietary or
taneously in some algae, though not in molds medical applications. The role of such
nor with Oenothera and borage plants. In PUFAs, in both healthy and dysfunctional pa-
Ribes, however, both isomers occur in equal tients, has been the subject of considerable
amounts. Longer chain polyunsaturated fatty research and investigation. For a more than
acids, up to 22:6, have been detected in the adequate statement of the current status of
lipids of many species of the phycomycetes this work, the reader is referred to a recent
and these are discussed separately below. Congress whose proceedings are given in a
Unusual fatty acids such as hydroxy fatty volume edited by SINCLAIRand GIBSON
acids or branched fatty acids are found, re- (1992). A discussion, or even prkcis, of the
spectively, in a few species of Claviceps and various conditions and aliments that seeming-
Conidiobolus (TYRRELLand WEATHER- ly benefit by a dietary intake of PUFA is
STONE, 1976). The high content of ricinoleic beyond the scope of this review but consulta-
acid, 12-hydroxyoleic acid, in Claviceps spp. tion of this symposium volume should give
(see Tab. 8) occurs only in the sclerotial tissue adequate information on most topics. The fol-
of the fungus and is absent from the vegeta- lowing sections review the current work being
tive mycelium (see LOSEL, 1988). It has, carried out to produce PUFAs from fungal
therefore, no potential as an alternative sources. The formation of PUFAs is set out in
source of castor oil which is the major source Fig. 5.
3.3.1 y-Linolenic Acid Co. Ltd. of Japan (see RATLEDGE,1989b)

which used Mortierella isabellina. However, a
(GLA, 18:3 w-6) recent report from NAKAJIMA and I z u (1992)
of that company makes no mention of any
Lower fungi, that is the “phycomycetes”, previous large-scale industrial process for
invariably produce the y-isomer of 18:3 in- GLA production and, indeed, highlights only
stead of the more common a-isomer. SHAW strains of Mucor circinelloides as potential
(1966) was the first to point out this distinc- candidate organisms. The largest scale of op-
tion amongst fungi, though the occurrence of eration was only 30 L but, optimistically, the
GLA in fungal lipids was first reported by authors consider that scale-up to 200 m3 was
BERNHARD and ALBRECHT(1948) who had now a feasible proposition, but no reports of
examined the lipid from Phycomyces blakes- work at this level were given. One strain of
leeanus. the mold was found that could produce 25%
Interest in oils containing this acid has a GLA in the total fatty acid with, however,
long history and the oil from the seeds of only a 6% oil content of cells. The highest oil
evening primrose (Oenothera biennis) have producer (30% of the cells) only produced
been used for many centuries, being de- 10% GLA in the fatty acid. This reverse cor-
scribed as the “King’s Cure-All”, as a remedy relation between oil and GLA content has
for a number of disorders. The efficacy of been previously noted by RATLEDGE(1989b)
evening primrose oil has been attributed to its and now been documented in some detail by
content of GLA (about &lo% of the total KENNEDYet al. (1993).
fatty acids). GLA itself has been reported as KENNEDY et al. (1993) showed that even in
suppressing acute and chronic inflammations, a single organism (they used Mortierella ra-
decreasing blood cholesterol concentrations, manniana, Mucor hiemalis, and Mucor circi-
and improving atopic eczema (HORROBIN, nelloides) the GLA content of the oil could
1992), however, SCHAFERand KRAGBALLE range from 5%-32% with oil contents from
(1991) found no clear evidence in support of 43%-4% with Mucor hiemalis with a narrow-
GLA being an efficacious treatment for atop- er range for Mortierella rammaniana. Maxi-
ic dermatitis. Alternative plant sources to mum productivity of GLA (that is g L-’ h-’)
evening primrose have been identified more was calculated for Mucor hiemalis with a
recently and include borage (Borago offici- GLA content of the oil at &lo%. For Mucor
nalis) and Ribes spp. (see Tab. 7). circinelloides, maximum productivity was with
In view of the known occurrence of GLA a GLA content of 14-16% which is close to
in fungi, steps to develop a biotechnological the commercial process outlined above and
route to GLA were first initiated in the au- shown in Tab. 9.
thor’s laboratory in 1976. A strain of Mucor High proportions of GLA of 20-26% in the
circinelloides (M. javanicus) was identified total fatty acids, together with oil contents of
from a large screening program as being a 15-25%, appear to be attainable by a number
suitable production organism. The first sales of species of Mucor and Mortierella (HANS-
of the GLA-rich oil were in 1985, the process SON and DOSTALEK,1988; DAVIES,1992a;
being run by J & E Sturge Ltd. (now Har- KENNEDYet al., 1993; R o u x et al., 1994). It
mann & Reimer) of Selby, Yorkshire, UK, at is, therefore, a matter of simple screening to
the 220 m3 level. The process ran until 1990 identify a likely candidate for large-scale pro-
when production ceased following transfer of duction. However, because of the high costs
the company to its present owners who are of fermentation processing, other parameters
part of the Bayer industrial group. have to be taken into account besides GLA
The specifications of the fungal oil were content: these include (1) the density to which
equal or better than evening primrose oil in the cells can be grown - values of up to
almost every respect including its much high- 2 5 g L - ’ are not uncommon (see KENNEDY
er content of GLA (see Tab. 9). A similar et al., 1993; NAKAJIMA and Izu, 1992) but are
process to the Sturge one, was considered to still far from optimal (about 40-80 g L-’
have been developed by Idemitsu Petroleum should be attainable); (2) the rate of growth -
full growth and lipid accumulation should be 1992) and in Mortierella spp. of the subgenus
attained within 72-96 h; and (3) the ability to Mortierella (AMANOet al., 1992). The highest
extract the oil from the cells. The oil should amounts occurred with M. alpina strain 1S-4
obviously be free from any deleterious mate- which had been grown in the presence of ses-
rial, including free fatty acids and partial acyl- ame oil (SHIMIZUet al., 1989a; YAMADAet
glycerols, but, as toxicological trials may be al., 1992). The concept behind adding the ses-
called for prior to the release of the oil for ame oil to the fungal culture had been to see
human consumption, it is an obvious advan- if exogenous oils could be taken up by the
tage if the organism being used already has an cells and then desaturated to particular fatty
established record of safe usage in foods. For acids. With sesame oil there was an apparent
these many reasons, Mucor circinelloides inhibition in the formation of arachidonic
seems to have been an excellent choice for a acid (20: 4), which being produced directly
GLA-production organism. from DHGLA (see Fig. 5), then led to the ac-
Roux et al. (1994) have recently reported cumulation of DHGLA. The inhibitor was
that some Mucor spp. when grown on acetic identified as a minor component of sesame
acid also produce a high content of stearic oil, sesamin, which acted specifically against
acid (18:O) besides GLA (up to 38 mg per g the A5 desaturase (SHIMIZUet al., 1991).
dry biomass was attained) and have suggested DHGLA was produced up to 23% of the to-
that, by appropriate fractionation, it should tal fatty acids and at a yield of 2.2 g L-'.
be possible to produce both a GLA-rich oil NAKAJIMA and Izu (1992) have similarly
and a cocoa butter equivalent fat. Although shown that a number of anisole derivatives
some strains of Mucor produced a high con- when presented to Conidiobolus nanodes also
tent of stearic acid when grown in glucose led to the accumulation of DHGLA. Like se-
(27% of the total fatty acids with M.flavus), same seedoil with M. alpina, these com-
the highest combined yields of stearic acid pounds had only a minor effect on cell growth
and GLA were with M . circinelloides grown and lipid accumulation. Maximum effect was
in a pH-stat, fed-batch culture using acetic produced with tert-butylhydroxyanisole
acid as sole carbon source. Stearic acid was (BHA) giving 18% DHGLA in the total lipid
up to 19% of the neutral lipid with GLA at fraction which was about 35% of the cells.
8%. As both BHA and sesamin appeared to act
as specific inhibitors of the A5 desaturase, the
next logical step was to delete this enzyme by
3.3.2 Dihomo- y-Linolenic Acid mutational techniques. JAREONKITMONGKOL
et al. (1992a, c) reported the results of such a
(DHGLA, 20:3 0-6) study using M. alpina and succeeded in isolat-
ing a mutant that produced 3.2gL-'
DHGLA is produced biosynthetically from DHGLA, that is 123 mg per g cells, and ac-
GLA by chain elongation (see Fig. 5). It is the counting for 23% of the total fatty acids. In
precursor of the Group 1 of prostaglandins comparison, the wild type produced less than
and thromboxanes and is often a minor com- a quarter of this amount. Other mutants of
ponent of lipids from fungi and algae but is the same organism have been reported that
also found in animals. There is probably no accumulated increased amounts of other
large market for DHGLA. Nevertheless, var- PUFAs (see Sect. 3.3.6, eicosatrienoic acid).
ious attempts have been made to develop a The approach of deleting various desatur-
process for its production as, undoubtedly, ases at the genetic level that are involved in
small amounts of oils containing high conversion of the PUFAs is obviously a very
amounts of DHGLA would command a very powerful one. Simple mutational techniques
high price if only for exploratory trials and for coupled with extensive screening for the cor-
experimental laboratory work. rect mutants can clearly pay handsome com-
Small amounts of DHGLA, up to 5% of mercial rewards.
the total fatty acids have been found in oils
from Conidiobolus spp. (NAKAJIMA and IZU,
3.3.3 Arachidonic Acid 18:l. As the identity of the ARA was not
confirmed by capillary GC or by CG-MS,
(ARA, 20:4 0-6) only by argentation TLC and by packed col-
umn GC, there must be grave doubts about
Arachidonic acid (ARA) and eicosapenta- the authenticity of these claims. Biochemical-
enoic acid (EPA, see below) are interme- ly, it would be an unprecedented reaction that
diates in the formation of several key prosta- could convert a fatty acid of the w-3 series,
glandins and leukotrienes which exert pro- which are invariably produced by these fungi
found physiological control over various (LOSEL,1988), into the 0-6 series (see Fig. 5)
bodily functions and are the subject of much as this would involve a saturation reaction of
nutritional and medical research (see SIN- a double bond at the 0 3 position which has
CLAIR and GIBSON,1992). never yet been recorded in any aerobically-
ARA has a much more restricted distribu- growing microorganism. It is, therefore, more
tion than GLA and it is clear that many than likely that the 20:4 PUFA reported by
molds do not synthesize fatty acids beyond RADWANand SOLIMAN(1988) as arachi-
CI8in length (SHAW,1966). However, phyco- donic acid was not the w-6 isomer (i.e., ARA)
mycetes molds of the subdivision Mastigomy- but was the 0-3 isomer, that is 20:4 (8,11, 14,
cotina synthesize fatty acids up to Cz2and for- 17) which would have behaved in both the
mation of ARA has been recorded, for exam- GC and TLC analyses as ARA. Nevertheless,
ple, in several Pythium spp. (GANDHIand it is still quite exceptional for these higher
WEETE, 1991), Saprolegnia parasitica (GEL- fungi to be recorded as producing any fatty
LERMAN and SCHLENK,1979), in several acid beyond CI8 in length (LOSEL,1988) and
Conidiobolus and Entomophthora spp. (TYR- one can only conclude that it was the cultiva-
RELL, 1967, 1968,1971; KENDRICK and RAT- tion of these fungi on shorter chain fatty acids
LEDGE, 1992a, b, C; NAKAJIMAand IZU, that led to this most unusual result, a result,
1992) and in a number of species of Mortierel- though, which has yet to be confirmed in an-
la subgenus Mortierella (Totani and OBA, other laboratory.
1987, 1988; YAMADAet al., 1987a; SHIMIZU The highest ARA contents have been re-
et al., 1988a; SHINMEN et al., 1989 AMANOet corded in Mortierella alpina with up to 79%
al., 1992). The subject has been recently re- ARA in the total fatty acids which repre-
viewed by RADWAN(1991) and BAJPAIand sented 26% of the cell dry weight (TOTANI
BAJPAI(1992), the latter recorded the ARA and OBA, 1987). Further work with this or-
contents in 27 species of phycomycetes fungi ganism has been developed up to the 300L
as well as in 42 species of marine algae (see scale (TOTANIet al., 1992) with some slight
also below). The more general review of Lo- diminution of yield. Significantly, exception-
SEL (1988) records a number of fungi that ally long fermentation times up to 16 d were
contain ARA, though without regard to any needed to produce the greatest yields. Such
biotechnological potential. lengthy times would invariably increase the
All researchers and reviewers to date have costs of any large-scale process.
been generally agreed that the family of w-6 Both YAMADAet al. (1992) and BAJPAIet
PUFA, which includes ARA as well as GLA, al. (1991~)have reported some possibly inter-
are confined to the “lower fungi” or phyco- esting developments in which the mold of
mycetes. RADWANand SOLIMAN(1988), choice is first grown for up to 10 d then al-
however, reported that they had found ARA lowed to stand at a lower temperature with-
in the lipids of a number of ascomycetes (or out further aeration. Under these conditions,
higher) fungi: Aspergillus versicolor, A. niger, the content of ARA (and EPA) increased up
A. oryzae, A. ustus, A. fumigatus, Paecilo- to 70% of the total fatty acids. However, mass
myces lilacinus, Penicillium sp., Fusarium balances were not carried out with these
oxysporum, and another Fusarium sp. In all stored cells so it is not immediately apparent
cases, the fungi had been cultivated on single, where the extra ARA or lipid might have ori-
shorter chain fatty acids, either saturated or ginated; although the lipid content of M. alpi-
monounsaturated, i.e., 14:0, 16:0, 18:0, or nu increased from 1415% to 3345% during
the aging of the cells for 6 d (BAJPAIet al., tion (28°C) (SHIMIZUet al., 1988b). Highest
1991c) it is not clear from where this extra lip-yields were attained with Mortierella alpina
id could have arisen. One simple explanation and M . hygrophila which produced 29 and
is that biomass other than lipid was self-uti- 41 mg EPA per g cells. Subsequent work
lized. Thus the total amount of lipid may not showed that exogenously added a-linolenic
have changed but only appeared to increase acid (18:3 w-3) was converted by M. alpina
as the remainder of the biomass was con- into EPA eventually pushing up the yield to
sumed. 67 mg per g dry cells (YAMADAet al., 1992;
SHIMIZUet al., 1989b). This was an unusual
finding as most fungi will not modify exoge-
3.3.4 Eicosapentaenoic Acid nously added fatty acids (RATLEDGE,
1989b).
(EPA, 20 :5 w-3) Other groups have not been as successful
as the Japanese in finding productive fungi
EPA exerts a number of physiological ef- for EPA. BAJPAI et al., (1992) found the
fects when fed to experimental animals in- highest yields with Mortierella elongata were
cluding lowering of blood triacylglycerol con- 15 mg per g dry cells and when a-linolenic
centration. Consequently, this would decrease acid was added this increased to only 36 mg
the potential for a coronary heart attack and, per g. In Pythium ultimum the maximum con-
therefore, the consumption of EPA is encour- tent was 34 mg EPA per g dry weight which
aged by many advocates (SIMOPOULOS, was attained only after careful selection of the
1989). EPA, like ARA is the precursor of strain and its culture conditions (GANDHIand
prostaglandins and leukotrienes. Currently, WEETE, 1991). O’BRIEN et al. (1993) re-
the major source of EPA is fish oil where it ported a maximum yield of 25 mg EPA per g
occurs, usually in low concentrations along dry cells using Pythium irregulare and have
with docosahexaenoic acid (DHA) (see be- recently described a pilot-plant process using
low). a colloid mill for the recovery of EPA at 96%
In molds, EPA frequently occurs along yield from this fungus (O’BRIENand SENSKE,
with ARA. The conversion of ARA to EPA 1994).
occurs directly in fungi (see Fig. 5) but not in
animals (GELLERMANand SCHLENK,1979);
the necessary 417 desaturase for this conver- 3.3.5 Docosahexaenoic Acid
sion is thus apparently unique to fungi. Small
amounts (usually < 10% of the fatty acids) of (DHA, 22:6 0-3)
EPA have been recorded in a number of low-
er fungi (LOSEL, 1988) as well as in the ma- DHA is abundant in the phospholipids of
rine fungi Thraustochytrium and Schizochy- retina and brain tissues and is usually re-
trium spp. (ELLENBOGEN et al., 1969) though garded as an essential fatty acid for humans
these latter fungi also contain higher amounts and other animals (THOMASand HOLUB,
of DHA (see below). 1994). It occurs in the oils of many fish where,
SHIMIZUet al. (1988a) were the first re- along with EPA, it may account for over 50%
search group to search specifically for the oc- of the total fatty acids (ACKMAN,1994; NI-
currence of EPA at high levels in fungi CHOLS et al., 1994). However, fish do not syn-
though its presence had long been known as thesize either EPA or DHA but acquire them
minor component amongst many of arachi- by ingestion of planktonic algae. DHA is con-
donic acid-containing fungi (SHAW, 1966). sidered important in the development of
SHIMIZUet al. (1988a, b) having earlier brain tissue of babies but conclusive evidence
screened fungi for their ARA contents (YA- for the nutritional role of DHA is lacking as it
MADA et al., 1987a, b) observed that many of is always administered during feeding trials
these fungi showed enhanced contents of along with EPA as fish oil is always used in
EPA if they were grown at a lower tempera- such studies. A number of different nutrition-
ture (12°C) than that used for ARA forma- al roles for DHA have been proposed (THO-
MAS and HOLUB,1993; see also YADWADet first described by MEADand SLATON(1956).
al., 1991). Its nutritional status still remains unclear
Although algae are clearly a potential though it may be converted to the 12-hydroxy
source of DHA (see below), there are several derivative which can affect blood platelet ag-
fungi that have been considered as possible gregration (LAGARDEet al., 1985). It is pre-
candidate organisms for its production sumably produced because animal tissues are
though it ,must be said that prospects for a unable to desaturate fatty acids between the
biotechnological route to DHA via fungi existing double bond, in this case at the A9
seems remote at this stage. position, and the terminal CH3 group.
The presence of DHA has been recorded There has only been one report of the for-
in small amounts in the lipids of Conidiobolus mation of ETA in fungi: JAREONKITMONG-
and Entomophthora spp. (TYRRELL,1967, KOL et al. (1992b), following their work on
1968, 1971) but more abundantly in the lipids the deletion of various fatty acid desaturases
of the marine fungi Thraustochytrium and in Mortierella alpina (see Sects. 3.3.2 and
Schizochytrium (ELLENBOGEN et al., 1969), 3.3.3) which led to increased production of
in which the high content of DHA and EPA DHGLA and ARA, found another mutant
were considered to have a role in maintaining that was no longer able to convert oleic acid
membrane fluidity in the organism whilst at (18:l w-9) to linoleic acid (18:2 w-6). This
low temperatures and in saline conditions. mutant accumulated several fatty acids of the
BAJPAIet al. (1991a, b) and KENDRICKand 0-9 series: 18:2 (6,9), 20:2 (8,ll) and also
RATLEDGE(1992b) both independently re- ETA. Under optimal conditions, in a fed-
examined the marine fungi as potential batch submerged cultivation for 10 d at 2 0 T ,
sources of DHA. All these fungi grew slowly ETA reached 56 mg per g dry biomass =
and generally to low growth yields giving only O.SgL-’. Growth of the mutant itself
low contents of lipid. None contained the key reached about 15 g L-’ indicating that a
enzyme for oleaginicity, ATP-citrate lyase small scale process might be possibly develop-
(see Sect. 2.3) and KENDRICKand RAT- able to produce this acid if sufficient commer-
LEDGE, (1992b) concluded that these fungi cial demand for it was forthcoming.
would be extremely difficult to exploit direct-
ly for the production of DHA. Maximum
amounts of DHA were produced by T. au- 3.3.7 Conclusions
reum ATCC 34304 at up to 50% of its total
fatty acids but with a lipid content of less than Although molds contain a large number of
15% (BAJPAIet al., 1991a, b) and usually not fatty acids, their biotechnological potential
more than 10% (KENDRICK and RATLEDGE, lies in their ability to produce a few selected
1992b). About two-thirds of the lipid was polyunsaturated fatty acids (PUFA) in some
neutral lipid (triacylglycerols) and this still quantity. Already we have seen the commer-
had a high content (30%) of DHA. However, cial production of ylinolenic acid (GLA)
the greatest difficulty with the exploitation of during the 1980s and it is likely that the pro-
this fungus was its low growth yield: KEN- duction of arachidonic acid (ARA) may not
DRICK and RATLEDGE(1992b) obtained only be far off. A number of industrial companies
4 g biomass per L over 72 h under conditions are known to be developing processes for this
which yielded up to 12 g L-’ of other fungi. PUFA and at least one company (Martek
Corp. Inc, Maryland, USA) now offers a tri-
acylglycerol oil containing 48% ARA for sale
3.3.6 Eicosatrienoic Acid at $2,OOO per kg. Free ARA itself at 80%
(ETA, 20:3 0-9, “Mead Acid”) purity is offered at $20 per g (i.e., $20,000
per kg). Demand may increase for this parti-
This comparatively rare PUFA occurs in cular fatty acid if it can be unequivocally de-
small amounts in lipids of animal tissues and monstrated that it is beneficial when added to
arises by direct elongation and desaturation milk destined for neonatal babies. The ab-
of oleic acid (18:l w-9) (see Fig. 5). It was sence of ARA, DHA, and GLA in cow’s milk
(but not mother’s milk) has suggested that As with all biotechnological products, what
these acids may fulfil important nutritional is produced will be dictated by market forces.
roles in the development of the early brain of Increased demand for a commodity invaria-
children. Current sources of ARA are from bly pushes up the price: should demand begin
animals and as long as there is continuing and to increase for any of the PUFAs then rapid
developing concern over the presence of un- exploitation of molds could then be antici-
detected viruses and prions in animal prod- pated. The high quality of the oils ensures
ucts, the greater will be the driving force to that molds are realistic alternative sources to
identify alternative and safe sources of these either plant or animal products.
acids. PUFAs derived from molds by fermen-
tation technology are, of course, free from
such infectious agents and can be produced to 3.4 Algae
higher levels of quality control than plant oils.
Furthermore, they do not contain herbicide The term “algae“ covers 14 distinct biologi-
or pesticide residues that would occur in oils cal groups and includes both the macro- and
derived from plant crops that have been microalgae. The macroalgae are the seaweeds
treated with these chemical agents as part of and related families; the microalgae are the
the usual agricultural regimen of routine crop equivalent of eukaryotic microorganisms but
spraying. Experience with GLA from Mucor also include the cyanobacteria, formerly
circinelloides has indicated it to be a very high termed the blue-green algae, which are part
quality oil free from all deleterious sub- of the prokaryotic eubacteria. It is the micro-
stances. Once a high market demand for an algae that are the subject of most research for
ARA-rich oil has been developed, the price the production of designated lipids and will
should fall dramatically from the level quoted therefore be covered here.
above. Production costs for a finished mold Microalgae have long been used as sources
oil should lie in the region of $25-35 per kg. of protein for use in animal and human foods.
This price would include refinement (removal Their potential as sources of biomass and fine
of non-TAG lipids) and decolorization or chemicals has been the subject of several
deodorization if needed by passage through recent major monographs (BECKER, 1993;
charcoal which are standard procedures for CRESSWELLet al., 1989; BOROWITZKAand
producing all high quality oils. BOROWITZKA, 1988; STADLERet al., 1988).
Demand for PUFAs other than GLA and Prospects for the production of oils and fatty
ARA from molds is less certain although acids by algae biotechnology have been re-
DHGLA and ETA (“Mead acid”) could be viewed, in general by KYLE (1991), YONG-
produced if needed but these tend to be re- MANITCHAI and WARD (1989), VOLKMANN
garded as “rare” PUFAs which are probably (1989), and BOROWITZKA (1988), and, with
only required in small amounts (say 10 kg an- respect to the production of specific fatty
nually) for experimental purposes. With EPA acids by KYLE (1992), SETO et al. (1992),
and DHA, mold sources are not as good as COHENand HEIMER(1992), BOSWELLet al.
algae or bacteria for these acids: EPA rarely (1992), and KYLEet al. (1992).
exceeds 20% of the total fatty acids in a mold The main problem in the biotechnological
oil and is often much less though the recent exploitation of algae is their cultivation. Most
description of a pilot-plant extraction process algae will only grow photosynthetically and,
for the recovery of EPA from Pythiurn irregu- therefore, require illumination and, although
lure (O’BRIENand SENSKE,1994) may indi- their carbon source, being C02, is regarded as
cate possible future developments in this free, growth is usually limited by the C 0 2
area. Although DHA can occur at up to 50% content of the air. When algae are grown out-
of the total fatty acids in a few molds, these doors in ponds or lagoons, a warm ambient
species are slow-growing and may be difficult temperature is needed besides high illumina-
to develop on a large scale though their ex- tion which limits their geographical develop-
ploitation appears under active considera- ment but many algae are susceptible to con-
tion. tamination by other algae or even bacteria or
may be attacked by predatory protozoa. Con- which is acceptable in appearance: a clear,

sequently, to maintain a pure monoculture of pale yellow oil with no taste, or aftertaste, is
an alga, high-cost illuminated fermenters may usually needed. This means that algae lipids
be needed which become impractical because may have to be fractionated or alternatively
of cost for large-scale growth. Although nu- the constituent fatty acids removed from the
merous devices, including clear plastic tubular total lipid by hydrolysis, either chemically or
fermenters, have been suggested for photo- enzymatically, and then re-esterified to etha-
synthetic algae growth (see, e.g., LEE, 1986), nol or glycerol. Ethyl esters of PUFAs are ac-
no satisfactory system has yet been develop- ceptable alternatives to the natural TAGS.
ed. Those commercial algae units that do ex- Such additional processing though increases
ist, mainly for the production of carotenoids the costs of the final product quite considera-
(see Sect. 4.2), use robust algae that have a bly.
particular nutritional advantage that prevents A list of oleaginous microalgae is given be-
contaminants or predators affecting the cul- low. Maximum reported lipid contents as %
ture. For example, Dunaliella salina grows in biomass dry weight are given in parentheses
hypersaline ponds or lakes and little else can (from ROESSLER, 1990 and RATLEDGE,
survive in such environments. However, an 1989a).
alternative to autotrophic growth (sunlight
and C02) of algae may be possible in some Prokaryota (Cyanobacteria, blue-green algae)
cases. Thus KYLE(1991, 1992) has described Anabaena cylindrica (9); Calothrix castelli
the heterotrophic growth (darkness and glu- (10); Nostoc sp. (8); Oscillatoria ssp. (18); Spi-
cose) of several algae that continue to pro- rulina maxima (2); S. platensis (17).
duce high amounts of lipid in the absence of
light. These examples are described below Eukaryota
(see Sects. 3.4.3 and 3.4.4). Amphiprora pyalina (30); Ankistrodesmus
Current attention on microalgae as sources sp. (40); Biddulphia aurita (40); Botryococcus
of lipids is mainly because of their high con- braunii (53-70); Chlorella minutissima (23);
tents of PUFAs. However, there is also inter- C. pyrenoidosa (36, 72); C. vulgaris (40);
est in algae as potential sources of essential Chlamydomonas applanta (33); Chrysochro-
fatty acids for marine animals, particularly by mulina ssp. (3348); Crypthecodinium cohnii
developing fish larvae, mollusks, and crusta- (25); Cyclotella cryptica (37); Dunaliella bar-
cea (VOLKMANN, 1989). In these cases, the dawil (D. salina) (47); Euglena gracilis (14-
actual microalgae itself is of importance as 20); Isochyrysis galbana (22); Monalanthus
the whole cells, not the isolated lipid, be- salina (72); Nannochloris sp. (48,55); N. ocu-
comes the feed. lata (42); Navicula acceptata (38); N. pelliculo-
With PUFA production, the choice of algae sa (22-45); Neochloris oleoabundans (35-54);
is less critical as the lipid itself will be ex- Nitzschia palea (40); Ochromonas danica (39-
tracted and used as the source of fatty acids. 71); Oocystis polymorpha (35); Ourococcus
This, though, highlights a second major prob- sp. (50); Peridinium cinctum (36); Phaeodac-
lem with algal lipids. Unlike oleaginous yeasts tylum tricornutum (14); Porphyridium cruen-
and molds, algae produce a large number of tum (14, 22); Prymnesium parvum (22-38);
lipids many of which have functional roles in Radiosphaera negevensis (43); Scenedesmus
connection with the photosynthetic process. acutus (26); S. dimorphus (1640); S. obliquus
Only a relatively small proportion (1040%) (49); Scotiella sp. (16-35).
of the total lipid may be triacylglycerol (see
RATLEDGE,1986) with the remainder being The fatty acid profiles of selected species of
phospholipids, other polar lipids, and a varie- microalgae are given in Tab. 10. The cyano-
ty of glycolipids. In the context of algae being bacteria (blue-green algae) tend to have low
used for animal feeding, the type of lipid does lipid contents and do not produce fatty acids
not matter as long as it is accessible and diges- longer than CIS. Although some species con-
tible. With oils for human consumption, the tain y-linolenic acid (see below), the amount
commercial emphasis is on producing an oil is too low to warrant recovery though the
Tab. 10. Fatty Acid Profiles of Selected Microalgae
Major Fatty Acyl Residues in Lipids [Relative YO] Others Reference

14:O 16:O 16:l 18:l 18:2 18:3 18:3 20:3 20:4 20:5 22:6
(0-7) (0-9) (0-6) (0-6) (0-3) (0-6) (0-6) (0-3) (0-3)
Prokaryota 8 6 3 2 4 9 12 - - - - - HUDSONand
KARIS(1974)
Spirulina maxima
S. platensis 1 26 5 23 10 21 - - - - - TORNABENE
et al. (1985)
S. platensis (SRS-lh) - 41 5 3 25 24 - - - - - COHENet al. (1992
Eukaryota
Chlorella minutissima 12 13 21 1 2 - - - 3 4 5 - SETOet al. (1984)
Chlorella vulgaris - 1 6 2 58 9 - 14 - - - - SHIFRIN (1984)
Chlorella NKG042401 <1 22 3 8 28 11 14 - - - - HIRANOet al.
(1990)
Chlorella CHLOR-1 <1 35 <1 44 7 <1 9 - - - - GUCKERTand
COOKSEY(1990)
Crypthecodium cohrii 47 19 1 5 - - - - - - 9 12:0,l6% HENDERSON
et al. (1988)
Zsochrysis galbana 12 10 11 3 2 - - - c1 25 11 18:4, 11% MOLINAGRIMA
et al. (1993)
Nannochloropsis 4 1 5 22 3 1 - - 1 4 3 8 - HODGSON
oculata (849/1) et al. (1991)
Nannochloropsis sp. 5 1 4 21 4 3 - - - 7 3 8 - SETOet al. (1992)
Phaeodactylum tricornutum - 1 0 21 1 4 1 - - 1 3 3 4 YONGM ANITCH AI
and WARD(1992)
Porphyridium cruentum (SRP-7) - 30 5 <1 5 1 - <1 16 - - COHENet al. (1992
Unspecified
Martek isolate MK 8908 2229 1 2 6 3 - 1 - - 5 - BOSWELL
et al. (1992)
presence of this PUFA in dry biomass may found that above pH 11 the triacylglycerol
have a marginal nutritional benefit if algae (TAG) fraction of the total lipid increased to
are used as a dietary supplement. However, over 20% whereas below this pH it was less
the cyanobacteria are not readily digested than 3%. GLA was only about 10% of the to-
and may possess some toxicity (TORNABENE tal fatty acids in the TAG fraction but was
et al., 1985). Spirulina spp. have though been over 40% in the glycolipid fraction which was
used as a source of supplementary dietary between 50 and 60% of the total lipids.
protein in both Mexico and Chad where Thus no readily recognizable, useful source
blooms of the algae occur on Lake Texcoco of GLA has been found in any microalgae
and Lake Chad (RATLEDGE,1989a). These whether prokaryotic or eukaryotic. The ple-
species, therefore, may be presumed to have a thora of lipid types (see, e.g., GUCKERTand
safe health record. Most attention on microal- COOKSEY,1990) means that direct produc-
gae has, though, focussed on the eukaryotic tion of a GLA-TAG oil from algae is an un-
species as sources of PUFA. economic proposition.
3.4.1 y-Linolenic Acid 3.4.2 Arachidonic Acid

(GLA, 18:3 w-6) (ARA, 20 :4 0-6)
Spirulina platensis and S. maxima have Surveys of the lipids of macro- and microal-
been known to contain GLA for some time gae (WOOD, 1988; YONGMANITCHAI and
(NICHOLSand WOOD, 1968) and have occa- WARD,1989; ROESSLER,1990) indicate that
sionally been considered as potential sources the red alga (Rhodophyceae) Porphyridium
of this PUFA. HIRANOet al. (1990), who ap- cruentum is superior to all other species for
pear to be the last research group to look at the formation of ARA. Small amounts of
this source of GLA in any seriousness, were ARA do, though, occur in many of the ma-
only able to achieve a maximum content of rine microalgae but only in P. cruentum does
GLA in dry biomass of S. platensis of 12 mg its content exceed 30% of the total fatty acids.
per g and this was after heterotrophic cultiva- Very high amounts of ARA in the alga were
tion for 7 d at 30°C. The total fatty acid con- originally reported by AHERNet al. (1983) at
tent of the cells was less than 5%. These val- up to 60% of the total fatty acids. COHENand
ues are not substantially different from those HEIMER(1992) have confirmed the potential
recorded by previous workers (see RAT- of this alga for large-scale production of
LEDGE,1989a) using these and other cyano- ARA which can be grown satisfactorily in
bacteria. HIRANOet al. (1990) also screened a large outdoor ponds (VONSHAK et al., 1985).
large number (>300) of marine eukaryotic However, under such conditions, the alga
algae for GLA formation and found that the produces mainly a reserve polysaccharide at
highest production was with a Chlorella sp. 40% of the biomass; ARA is only 1.5% of the
(NKG 042401) that contained about 10% to- cell dry weight. According to how this alga is
tal fatty acids with a 10% content of GLA. cultivated, it may also contain equal amounts
In an attempt to improve GLA production of EPA (qv.) to ARA (COHENet al., 1988)
in S. platensis, COHENet al. (1992) isolated a and may, therefore, be considered as a poten-
number of cell lines that were resistant to the tial source of either or both PUFAs. A meth-
herbicide known as SAN 9785 which is a sub- od for fractionating the glycolipid fraction of
stituted pyridazinone that selectively inhibits fatty acids, which is the major lipid fraction,
fatty acid desaturation. Slight increases in to- has been described and has yielded ARA at
tal fatty acid contents ( 4 4 % dry wt.) were 80% purity and EPA at 97% purity (COHEN
obtained in the resistant cells with GLA in- and COHEN,1991).
creasing from 21.5%-23.5%. GUCKERTand Improvements to this alga are still being
COOKSEY(1990) used an alkaline stress culti- sought by the selection of herbicide-resistant
vation regimen with a Chlorella isolate and cell lines so that they will have higher propor-
Tab. 11.Cultivation of Porphyridiurn curenturn in Open (Outdoor) Ponds for the Production of ARA and
EPA (from COHENand HEIMER,1992)
Cell ARA EPA Output Rates

Concentration ph dry wt.=] [% dry wt."] [g m-' d-'1
Biomass ARA EPA
Winterb
Low 0.76 2.2 3.6 0.03 0.08
Medium 0.75 2.3 5.0 0.04 0.12
High 0.73 2.2 2.1 0.02 0.05
Summer'
Low 0.71 1.3 19.7 0.14 0.25
Medium 1.2 1.2 24.0 0.28 0.28
High 1.3 1.o 13.0 0.17 0.17
a Ash-free dry wt.

Maximum daily temperature 1618°C.
Maximum daily temperature 28-31 "C.
tions of desirable PUFAs (COHENand HEI- this and other alga as a source of EPA has
MER,1992). The best results so far achieved continued (YONGMANITCHAI and WARD,
in outdoor cultivation (see Tab. l l ) , which is 1991).
the only route available for large-scale pro- Tab. 12 summarizes the principal photosyn-
duction, still indicate that there is little chance thetically-grown algal species that have been
of this being an economic route to ARA, or considered of some potential in the produc-
EPA, production. Contents of ARA at less tion of EPA. Although there are other algae
than 2% of the cell dry weight compare unfa- with higher contents of EPA in their total fat-
vorably with yields attained with molds which ty acids (see YONGMANITCHAI and WARD,
can achieve 5% and possibly 6% ARA con- 1989, 1991), the lipid content of these algae
tents (4.v.). Moreover, molds produce TAG may not be very high and furthermore, they
oils requiring the minimum of downstream are mostly macroscopic algae which are not
processing and can be easily grown to pro- easily cultivatable under controlled condi-
duce zero EPA. Thus fractionation and sepa- tions.
ration of the w-6 and 0-3 fatty acids would be A separate approach to PUFA production
unnecessary using mold technology. by algae has been developed by the Martek
Corp. Inc. (Maryland, USA) in which se-
lected algae are grown heterotrophically us-
3.4.3 Eicosapentaenoic Acid ing glucose as carbon and energy source.
(EPA, 20:5 w-3) Thus, photobioreactors are unnecessary and
the organism behaves, and performs, as a eu-
SETO et al. (1984) reported that the high karyotic yeast or mold (KYLE,1992). Of sev-
contents of EPA in Chlorellu minutissima eral thousand algae that were screened for
made this alga an attractive source of this acid PUFA production, one - MK8909 - was se-
for both health foods and for the pharmaceu- lected for EPA production. (Another isolate
tical industry. Maximum contents of EPA of was subsequently exploited for DHA produc-
the total fatty acids reached 45% but were tion, see below.) This isolate has been de-
only 2.7% of the dry biomass. Cell yields, scribed as a apochlorotic diatom, that is not
moreover, were very low at 3OOmgL-l possessing chlorophyll and, therefore, incapa-
(compared to molds which reach cell densities ble of photosynthesis. The alga may be re-
of over 50 g L -I). Nevertheless, interest in lated to Nuviculu suprophillu which KYLE et
Tab. U.Photosynthetically-Grown Algae as Potential Sources of EPA
Alga Lipid Content EPA in Reference

of Biomassa Fatty Acids
[To 1 [%I
Phaeodacty lum 15 35 YONGMANITCHAI and
tricornutum WARD(1992)
Nannochloropsis 14 45 SETOet al.
oculata (1992)
Isochrysis galbana 22 26 MOLINAGRIMAet al.
(1992, 1993, 1994)
Porphyridium 5.6b 41 COHENet al.
cruentum (1992)
Chlorella 15 45 SETOet al. (1984)
minutissima
a The lipids are usually composed of glycolipids, phospho- and other polar lipids with triacylglycerol as a
minority component. The total fatty acid content of the lipids may be less than 50%.
Total fatty acids.
al. (1988) had earlier described as having along with EPA and also 18:4 (YONGMANIT-
some potential for PUFA production. CHAI and WARD, 1989). Y~NGMANITCHAI
The content of EPA in MK8909 was only and WARD (1989) highlighted three species
5% of the total fatty acids but the oil content that could warrant further attention for DHA
was 50% (w/w) of the cells. The organism production: Crypthecodinium cohnii, Go-
grew readily and yielded 50 g dry cells per L nyaulax catenella, and Gymnodinium nelsonii
in 3 d. As the alga produced an easily extract- as all contained over 30% DHA in their total
able oil, the EPA was considered of potential fatty acids. HENDERSON et al. (1988) have ex-
commercial value as there was a complete ab- amined the first algae in some details growing
sence of any other PUFA: 18 :2 was present at the cells non-photosynthetically. The har-
only 3% and 18: 1 was at 26%. Thus enrich- vested biomass had a total lipid content of
ment of EPA from the oil is a relatively easy 25% of which the triacyglycerol fraction con-
proposition and such oils are now offerred for stituted over 50% though this fraction only
sale by Martek. contained 9% DHA. The major DHA-con-
The overall problem with EPA production taining component was phosphatidylcholine
from any microbial source has already been (PC) having 57% DHA in its total fatty acids
discussed (Sect. 3.3.4) in that EPA is usually and PC itself constituted 18% of the lipid.
easily obtainable from fish oils though as a The complete fatty acid profile is given in
mixture with DHA. Unless, and until, some- Tab. 9.
one demonstrates the clear need for EPA (or Of the other photosynthetic algae, only the
DHA) as a single PUFA, there seems little species of Zsochrysis have been reported as
prospect that EPA will be required for any- containing more than a small amount of
thing more than experimental work. DHA. This marine alga has been recently ex-
amined by MOLINAGRIMAet al. (1992,1993,
1994) for its potential to produce DHA and
3.4.4 Docosahexaenoic Acid EPA. Following screening of a number of iso-
lates, one was selected for further work (Lo-
(DHA, 22:6 w-6) PEZ ALONSOet al., 1992). This strain, when
grown in chemostat culture with illumination,
DHA has a very limited distribution in produced up to 2% of the cell dry weight as
most algae though it occurs throughout the DHA (MOLINAGRIMAet al., 1993). EPA
family of dinoflagellated algae (Dinophyceae) though reached up to 6% of the cells (see
above and Tab. 12). As a source of DHA, the them heterotrophically: treating them as
alga is, therefore, no better than many fish yeasts or lower fungi and growing them with-
oils; DHA would be difficult to purify as a out illumination in conventional bioreactors.
single PUFA and its presence in all lipid frac- Martek Corp. Inc. in the USA have pio-
tions (MOLINAGRIMAet al., 1994) at about neered this approach and since 1992 have on
10% of the total fatty would not be regarded sale oils containing EPA and DHA as single
as particularly encouraging. PUFA not containing any other PUFA or
On the other hand, the concept developed even diunsaturated fatty acids. It is under-
by the Martek Corp. Inc. (see Sect. 3.4.3) of stood, however, that demand for these oils
using algae growing heterotrophically has has been very modest principally because
succeeded in identifying an unknown (i.e., un- there is, as yet, no clear dietary or clinical in-
declared) phytoplankton species (MK 8805) dication that one acid or the other would be
that produced DHA as the sole PUFA at useful in the treatment of any disorder or nu-
50% of its total fatty acids (KYLEet al., 1992). tritional imbalance. In view of the current
The only other unsaturated fatty acids in the high costs of these oils, not unnaturally the
algae oil were 18:l (at 11%) and 16:l (at public have preferred to buy various fish oils
2%). The only drawback with this alga was its and PUFA concentrates derived from fish oils
low lipid content at between 10-15%. Nev- that contain both EPA and DHA together.
ertheless, it could be grown to high cell densi- The possible inclusion of DHA in infant food
ties (about 25 g L-') in 84 h in a conventional formulations (see Sect. 3.3.7), however, is
(non-illuminable) fermenter. The oil and likely to stimulate interest to find an alterna-
DHA-enriched fractions of it with up to 80% tive source to fish oil for this PUFA. The al-
purity are now offered for sale by Martek. gae and fungi already mentioned will be the
This oil is, therefore, the only one which is principal candidates.
commercially available that does not contain
EPA.
3.4.5 Conclusions
4 Sterols, Carotenoids,
Algae are undoubtedly a rich source of
PUFA. They are the sources of PUFA in fish,
and Polyprenes
which do not carry out de novo synthesis of
these acids, but rely on the ingestion of algae 4.1 Sterols
for them. Nevertheless, algae are not excep-
tional sources of PUFA. ARA is possibly Although steroids, including their precur-
producible from Porphyridium cruentum be- sor squalene, have been reported occasionally
ing grown in outdoor ponds but the econom- from the prokaryotic microorganisms, these
ics do not look as attractive as obtaining are only of academic interest as the quantities
ARA from fungi. EPA and DHA may occur involved are extremely small. Sterols, howev-
together in algae in which case the lipid er, have been produced commercially from
would be no better than fish oil as a source of eukaryotic microorganisms though only as a
either acid. Some algae though only produce by-product by the extraction of either spent
EPA and, therefore, might be useful sources fungal mycelium recovered from a fermenta-
of this acid if it were not for the fact that EPA tion process or from spent brewer's yeast.
is distributed throughout all the many lipid Sterols for commercial use are usually ob-
classes that occur in algae. This plethora of tained by extraction of plant materials which
lipid classes necessitates complete hydrolysis is not only a cheaper route than from micro-
of the total lipid and recovery of the EPA as organisms but also provides the correct sub-
the free acid. stituents on the molecule for easy transforma-
The more realistic commercial approach to tion into the commercially lucrative steroid
PUFA production by algae has been to grow hormone market. Most microbial processes
for steroid formation now appear to be de- the yeasts, especially in Saccharomyces, Kluy-
funct but some extraction of brewer’s yeast veromyces, Metschnikowia, Pichia, and Toru-
for the production of ergosterol (VI) does ap- laspora (RATTRAY, 1988). While the major
pear to continue in a few locations though the steroid is ergosterol, over 14 other sterols
exact number and scale of the various opera- have also been identified (WEETE,1980), the
tions is not generally revealed by industrial- principal ones of which are lanosterol (VII),
ists. zymosterol (VIII), and ergosta-5,7,22,24(28)-
Ergosterol (VI) is the commonest microbial tetraen-3/3-01 (IX).
sterol and it occurs both in the free form and Although the usual range of extractable
as its fatty acyl ester in algae, yeasts, and sterols (free sterols plus sterol esters) in
molds. Highest levels of sterol are found in yeasts is generally in the range of 0.034% of
Ergosterol Lanosterol
vn
Ergosta-5,7,22,24(28)tetraen-3@-01
Zymosterol
VIII IX
7-Dehydrocholesterol
Cholecalciferol
X XI
the cell dry weight (RATTRAY,1988), DULA- sum. The highest level of sterol reported
NEY et al., (1954) found several strains of S. would appear to be that found by OSMANet
cerevisiae which produced ergosterol up to al. (1969) for Aspergillus fumigatus which
10% of the biomass. When the yeasts were contained 5% of its dry biomass as ergosterol.
grown in submerged culture yields of sterol of SHIMIZUet al. (1992) have recently reported
up to 4 g L - ’ were obtained. A process for the occurrence of a novel sterol, 24,25- me-
the production of ergosterol was patented by thylenecholest-5-en-3p-01(XIV) in Mortierel-
DULANEY(1957). Other workers have been la alpina being grown for arachidonic acid
also able to achieve similar results: e.g., EL- production (Sect. 3.3.3). This sterol contains a
REFAI and EL-KADY (1968a, b; 1969) re- cyclopropane ring in its side chain. Similar
corded up to 23% sterol contents in Saccha- sterols occur in sponges but this is the first ex-
romyces fermentati (now Torulaspora del- ample of a sterol with a cyclopropane in this
brueckii) if grown in the presence of potas- particular position in any organism.
sium persulphate, hydroquinone, or indigo-
carmine.
Increased accumulation of sterols in S. cere- 4.2 Carotenoids
visiae occurs at low specific growth rates un-
der nitrogen-limited, aerobic growth condi- Carotenoids occur through the whole of the
tions (NOVOTNYet al., 1988) and by exploit- microbial world and, of course, throughout
ing such information BEHALOVAand VORI- nature. Their occurrence and structure has
SEK (1988) increased the total sterol content been comprehensively reviewed by GOOD-
of this yeast to 7.2% of the dry biomass. The WIN (1980, 1983). The review of LOSEL
total lipid content of the cells, including the (1988) on fungal lipids also includes coverage
sterols, reached 31% making the yeast a can- of the carotenoids of fungi. A useful mono-
didate for inclusion in the list of oleaginous graph on the various pigments of microorgan-
yeasts (see Tab. 3). However, such lipid con- isms has been prepared (MARGALITH,1992)
tents in S. cerevisiae are quite exceptional in- and contains a short but informative chapter
dicating that these authors may be using an on carotenoids, their chemistry, biochemistry,
atypical strain. and functions. The biosynthesis of carote-
Ergosterol is not of major economic signifi- noids has been described by HARRISON
cance though it has some commercial value. (1986), BRAMLEY(1985), and BRAMLEYand
MARGALITH (1989) has outlined the various MACKENZIE(1988).
attempts to produce it using yeasts. Its main Although all algae contain carotenoids,
application is that of an analog for cholecalci- these are part of the chloroplast photosyn-
ferol, vitamin D3, (X), which arises from 7- thetic apparatus and, therefore, do not usual-
dehydrocholesterol (XI) by the action of ul- ly form a major constituent of the biomass. In
traviolet light; likewise ergosterol (VI) is con- the marine brown macroalgae (the Phaeophy-
verted to ergocalciferol, vitamin D2 (XII). ta), the total annual biosynthesis of carote-
However, the latter is only 10% as nutrition- noids through the oceans of the world has
ally effective in chickens as is vitamin D3.The been calculated as 1.2-10’ t (JENSEN,1966) -
prospects of being able to produce mutants of but, of course, none of this is harvested for
S. cerevisiae which would accumulate 7-dehy- the carotenoids. The major carotenoids, in or-
drocholesterol rather than ergosterol are, der of abundance, in these seaweeds are fu-
therefore, extremely attractive (PARKSet al., coxanthin (XV), violaxanthin (XVI), and p
1984). carotene (XVII).
In molds, a large range of sterols has been Amounts of carotene in microalgae do not
recognized (WEETE, 1980; LOSEL, 1988), usually exceed a few mg per g dry weight
though once more ergosterol is the major (GOODWIN,1980). However, Spirulina pla-
constituent in many, though not every, spe- tensis, which is used as a source of food by
cies. Cholesterol (XIII) has been noted as the inhabitants around Lake Chad in Africa and,
major sterol in a number of Mucorales and which will be grown on brackish water, con-
also, unexpectedly, in Penicillium funiculo- tains about 0.4% of its dry weight as pcaro-
HO &&
Ergocalciferol
XI1
HO
Cholesterol XI11
24,25-methylenecholest-5en-3~-ol
XTV
Fucoxanthin xv
Violaxanthin XVI
p-Carotene XVII
tene and other xanthophylls (CLEMENT, RON,1980). PCarotene may be up to 10% of

1975). The halophilic Dunaliella salina or the dry weight of D . bardawil (MARGALITH,
sometimes referred to as D . barduwil, which 1992) and with such yields has been grown
can grow in waters of high salinity and pro- commercially in Australia by Western Bio-
duce glycerol in some abundance (DUBINSKYtechnology Ltd. at Hutt Lagoon in Western
et al., 1978), can produce up to 400 mg p- Australia, in Israel (Nature Beta Technolo-
carotene m-' d-' (BEN-AMOTZand Av- gies), and in the USA (Microbio Resources).
All systems use outdoor lagoons: that in Aus- viewpoint, the most important carotenoid is
tralia, e.g., has a coverage of 50 ha (10x5 ha pcarotene (XVII) which, besides being an
ponds) and was scheduled to add a further important foodstuffs colorant, has provitamin
25 ha pond in 1994. (The author is grateful to A activity (vitamin A: XVIII). It has also
Dr. M. A. BOROWITZKA for this informa- been suggested that pcarotene may also act
tion.) Full descriptions of this process have through its role as an antioxidant as a tumor-
been provided by BOROWITZKA and BORO- suppressing agent and be useful in chemo-
WITZKA (1989, 1990) and BOROWITZKA prevention of cancer (MARGALITH,1992).
(1992). In Israel, Dunaliella is grown in a sec- However it is the p-cis isomer which appears
tion of the Dead Sea. to be effective rather than the all-trans, chem-
Algal pcarotene is comprised of 60% 9-cis ically-produced pcarotene. Demand may
isomer (BEN-AMOTZ and AVRON, 1983; consequently shift towards the more expen-
BEN-AMOTZet al., 1988). This appears to be sive natural p-carotene if these claims for its
assimilated by experimental animals more therapeutic effectiveness are confirmed.
rapidly than the all-trans isomer (BEN- pCarotene is also the predominant carote-
AMOTZ et al., 1989). From a commercial noid in many fungi and is a minor constituent
Vitamin A
XVIII
Torulene
Astaxanthin (3R. 3 R isomer)

xx
Botryococcene
XXI
in most others, though not all fungi do con- ceed 100t a-' by the end of this century
tain carotenoids. The most abundant produc- (JOHNSONand AN, 1991). The yeast carote-
tion of p-carotene has been achieved with the noid has, however, the opposite chirality at
phycomycete fungi, Blakesleea trispora and 3R and 3R' (i.e., the hydroxyl group on the
Phycomyces blukesleeanus (MURILLOet al., cyclohexene end groups) to the lobster asta-
1978; NINETand RENAUT,1979; CERDA-OL- xanthin (see GOODWIN,1980; 1983) but this
MEDO,1989) where yields of carotene in mu- does not affect its acceptability. Synthetic as-
tant strains of up to 2% of the biomass have taxanthin which is the all-trans isomer, is pro-
been recorded. Commercial processes for the duced by Hoffman-La Roche Ltd. but awaits
production of &carotene using B. trispora approval from the FDA for use in the USA
have been developed and at least one, in the (JOHNSONand AN, 1991). Consequently the
Ukraine, is still in operation. SHLOMAIet al. emphasis is now on P. rhodozymu as a poten-
(1991) have recently re-examined the p-caro- tial worldwide source.
tene produced by P. blakesleeanus and, con- Initial yields of astaxanthin by P. rhodozy-
trary to the original supposition that the all- mu were less than 500 Fg g-' (JOHNSONet
trans isomer would be found (BRAMLEYand al., 1977; 1980). Whilst some increase is possi-
MACKENZIE,1988), 15% of the total p-caro- ble by careful selection of the growth medium
tene was the 9-cis isomer. The remaining 85% and conditions (HAARD,1988; NELISand DE
was though the all-trans form. Nevertheless, LEENHEER,1989), high yields can only be
this result was considered of considerable in- achieved after mutagenesis (AN et al., 1989;
terest as further research should be able to in- MEYER et al., 1993). Even these yields
crease the proportion of 9-cis-p-carotene. though can be enhanced by selecting individ-
Such formation of the highly desirable isomer ual cells by a cell sorter using the fluorescence
for cancer prevention (see above) could now of astaxanthin as the indicator (AN et al.,
re-stimulate interest in the exploitation of 1991). Yields of astaxanthin of 2.5 mg per g
fungi for &carotene production. Cultivation dry cells have been achieved by using a mu-
of heterotrophic fungi appears to be a better tant and carefully selected growth conditions
commercial proposition than having to grow (MEYERet al., 1993). However, it is consid-
algae autotrophically where the costs of land, ered that commercial production probably
harvesting and the necessity for high light in- needs to reach 4 to 5 mg g-' in order to be
tensities and high ambient temperatures pose economic.
many severe limitations (MARGALITH, 1992). An alternative microbial source of astaxan-
Torulene (XIX), which has only half the thin is the freshwater green alga, Huemufo-
provitamin A capability as p-carotene, is the coccus pluviulis (BOROWITZKA,1992). Out-
major carotenoid pigment of the red Rhodo- door cultivation of this alga so far has proved
torula and Rhodosporidium yeasts, of which unreliable and commercialization seems un-
several species are oleaginous (see Tab. 3). likely at the moment even though astaxanthin
Other carotenoids also occur in these species may reach up to 20 mg per g cell dry weight.
(GOODWIN,1980). However, the amounts are
far less than needed for any commercial inter-
est and these carotenoids remain of academic 4.3 Polyprenoids
interest.
The pink yeast, Phaffiu rhodozyma pro- The only polyprenoid of possible biotech-
duces astaxanthin (XX) which is the carote- nological significance is botryococcene (XXI)
noid giving the characteristic pink color to which occurs as the principal lipid component
salmon, crabs, lobsters, and other crustaceans of the alga Botryococcus bruunii. This alga
(and indirectly flamingos and other birds liv- has been reported as containing up to 85% of
ing off these life forms). It is now produced its biomass as a mixture of polyprene hydro-
commercially by Gist-Brocades, Netherlands, carbons when it has been isolated from coal
for use in feed formulations for poultry but deposits, which appears to be its natural eco-
mainly for pen-reared salmonids. The de- logical niche. The alga can be grown in the
mand for astaxanthin for fish feed may ex- laboratory, though yields of botryococcene
and related C,, hydrocarbons are then only dual purpose: sewage cleanup and hydrocar-
from 24 to 45% of the biomass (HILLENet al., bon production. As the hydrocarbon would
1982; YAMAGUCHI et al., 1987; CASADEVALL be used as a fuel and not as a food supple-
et al., 1985). The hydrocarbons can be ther- ment, there is obviously no restriction on
mally cracked to give gasoline and other fuels what the alga may be grown on as none of it
of commercial value (HILLENet al., 1982) and would be returned into the food chain.
thus have been suggested as possible sources
of energy. However, it seems unlikely that
this would represent a commercially exploita-
ble source of fuel hydrocarbons particularly
in view of the slow rate of hydrocarbon pro-
duction at about 0.15 g L-' d-' (CASADE-
5 Wax Esters and
VALL et al., 1985) though a slight increase oc-
curs when the cells are immobilized (BAIL-
Polyesters
LIEZ et al., 1985; 1986) and used in an air lift,
illuminated bioreactor (BAILLIEZet al., 5.1 Wax Esters
1988).
The impracticality of scaling-up such a sys- Wax esters of the type RCOOR' where R
tem would, though, preclude any serious bio- and R' are long alkyl chains occur in bacte-
technological application but SAWAYAMA et ria, algae, and yeast. Their route of biosynthe-
al. (1992) have recently reported that B. bruu- sis is given in Fig. 6. The fatty acid and alco-
nii can be usefully cultivated on secondarily hol moieties may be saturated or unsaturated.
treated sewage from domestic wastewaters so The diunsaturated wax ester is desirable as a
that it not only removes N and P but still pro- substitute for jojoba oil. With algae, it is the
duces about 50% of its biomass weight as the protozoan, Euglena grucilis, that has been the
hydrocarbon. Thus, the only hope for com- most studied (see KAWABATA and KANEYA-
mercial take-up would be to use the alga for a NA, 1989) for wax ester production but the
Feedstock
- lntracellular activities
-I
Hydrocarbon Hydrocarbon
I
Oxidation
Fatty alcohol Fatty'alcohol Oxidationw Unsaturated

fatty alcohol
Fatty a& ___j

Oxidation
lt Reduction
Fatty acid Oxidation ~ Unsaturated

fatty acyl CoA
Diunsaturated
monoester
(Fatty acyl-CoAester)
Degradation
it
Acetyl-CoA
Biosynthesis
acetic acid, etc.
Fig. 6. Pathways to a diunsaturated wax ester (jojoba oil type substitute) from various substrates.
amounts are less than 10% of the cell biomass process has occurred. Indeed, prospects for a
(about 100 kg per lo6 cells). With yeasts, wax microbial route to wax ester production
esters are not usual components of the lipid would now seem to have receded even fur-
fraction but SEKULA(1992) has reported for- ther with the recent description of how oleoyl
mation of them in several yeasts when grown oleate can be chemically synthesized using
in fatty alcohols. The amounts formed were oleic acid, oleoyl alcohol, and a zeolite cata-
not disclosed but appeared, from TLC evi- lyst (SANCHEZet al., 1992).
dence, to be equal in amount to the triacyl-
glycerol fraction. Evidence was presented
that the fatty acid moiety of the ester may be 5.2 Polyesters -
oxidized to a e l keto fatty acid which could
then be incorporated into the wax ester (see Poly-P-Hydroxyalkanoates
Fig. 6).
The greatest amounts of monoester appear The major microbial polyesters of commer-
to occur in bacteria of the genera Acineto- cial interest are the poly-p-hydroxy alka-
bacter, Micrococcus, Nocardia, Mycobacteria, noates (PHA) of which poly-p-hydroxybuty-
and Corynebacterzurn. In the latter three gen- rate (PHB, R=CH3 in XXII) is of major im-
era, the fatty acids involved may be long- portance. PHB and PHA are found principal-
chained ( > &) and/or methyl branched; the ly in bacteria though related molecules have
waxes may also be associated with virulence been found in small amounts in the mem-
and thus biotechnological exploitation is un- branes of eukaryotic cells (ANDERSONand
likely. Acinetobacter and Micrococcus, which DAWES,1990). The subject has been the topic
are now probably synonymous at least as far of a number of monographs and international
as the wax-producing strains are concerned, symposia (DAWES,1990 DOI, 1990; SCHLE-
have been studied by various groups includ- GEL and STEINBUCHEL, 1992; VERT et al.,
ing that of Cetus Corp., California, USA and 1992) and more are to follow, see below. A
numerous patents (see RATLEDGE,1986) tak- typical electron micrograph of a PHB-con-
en out. The aim of this work had been to taining cell is given in Fig 7.
achieve production of a jojoba oil-like materi- Present interest in PHB/PHA arises from
al which is essentially a 20: 1-20: 1 fatty acid/ its use as a biodegradable plastic. PHB itself
fatty alcohol ester. Yields, however, have reis considered too brittle to be conveniently
mained low (< 1g L-') and no take-up of the molded into appropriate shapes and so is con-
r 1
x = 10 000 to 20 000
R = -CH3 for poly-Ehydroxybutyrate (PHB)

R = - C2H5 for poly-P-hydroxyvalerate (PHV)
R = - CnHh-1 for poly-P-hydroxyalkanoates(PHA) up to n = 9
Poly-p-hydroxybutyrate and alkanoates

XXII
cose. PHB and the co-polymer, PHB/V, are

produced by Zeneca plc, UK, (formerly ICI
plc) and uses Alcaligenes eutrophus as pro-
duction organism. Yields are up to 80% of
the total cell biomass. For commercial pur-
poses the molecular weight of the polymer
should be as high as possible and certainly in
excess of lo6 Da. The process has been de-
scribed, with perhaps understandable per-
functoriness, by BYROM(1990, 1992). The
commercial product is sold under the trade
name of Biopol@. A rival industrial process
operated by Chemie Linz GmbH, Austria,
has been described by HRABAK(1992) but it
is uncertain whether this operates other than
as a demonstration unit.
Fig. 7. Electron micrograph of poly-p-hydroxybuty- Accumulation of PHB in bacteria is fa-
rate granules in Alcaligenes eutrophus; marker bar: vored by much the same environmental fac-
1 pm (photograph kindly supplied by Dr. A. J. AN- tors that are needed for triacylglycerol accu-
DERSON, University of Hull, UK). mulation in yeasts and molds: that is a deple-
tion of N (or other nutrient) from the culture
with the provision of excess carbon to ensure
continued formation of the polymer (see Sect.
sequently produced as a heteropolymer along 2.1). As bacteria do not readily synthesize
with Phydroxyvalerate (V, R = -C2H5 in triacylglycerols, PHB and PHA may be re-
XXII) as the other monomeric unit to P-hy- garded as the bacterial equivalent to triacyl-
droxybutyrate. Whilst Phydroxybutyrate is glycerols serving the same metabolic func-
synthesized from acetate-acetate condensa- tions as a (chemically) reduced storage com-
tion, P-hydroxyvalerate is produced from ace- pound; that is, it is accumulated under condi-
tate-propionate condensation. This requires tions of nutrient deficiency and utilized dur-
propionic acid to be presented to the bacteri- ing periods of carbon starvation. The pathway
al cultures as a cosubstrate along with glu- of biosynthesis of PHB (Fig. 8) is less com-
Acetyl-CoA & Acetoacetyl-CoA --D-3-Hydroxybutyryl-CoA C

(3-ketobutyryl-CoA)
COA-SH
3-Ketovaleryl-CoA PHB/V
C
Fig. 8. Biosynthesis of poly-/3-hydroxybutyrate (PHB) and the copolymer of poly-(p-hydroxybutyrate/@

hydroxyvalerate) (PHBN). PHB, R = CH3 in Fig. 10 PHV, R = C2H5in Fig. 10. Enzymes: A: 3-ketothio-
lase; B: acetoacetyl-(3-ketoacyl-)CoAreductase (NADPH-dependent);C: PHB synthase (or polymerase);
l x : one mol; 2 x : two mol.
plex than that of triacylglycerols. Indeed, the tion would be an oil-producing one as such a
biosynthesis requires only three additional plant would already have the necessary bio-
enzymes to those already present for fatty chemical machinery present to produce ace-
acid biosynthesis: 3-ketothiolase, acetoacetyl- tyl-CoA in some abundance. Thus, if fatty
CoA reductase, and PHB (PHA) synthase or acid biosynthesis could be prevented, say by
polymerase. All three enzymes have been deletion of acetyl-CoA carboxylase or even
studied in some detail (see ANDERSONand impairment of the fatty acid synthase com-
DAWES,1990) and the genes for each of them plex, then, with the three PHB genes being
have been identified, sequenced (STEINBU- successfully introduced and fully expressed in
CHEL et al., 1992), and now cloned and ex- the plant, carbon should now flow into PHB
pressed in plants (POIRIERet al., 1992). (The production.
PHA polymerase is probably a different en- Such scenarios are now being developed
zyme from that of PHB polymerase and the for oilseed rape by Zeneca Seeds in the UK
corresponding gene awaits to be described.) (SMITHet al., 1994) and by the Carnegie In-
This latter work has considerable commer- stitution in conjunction with Procter & Gam-
cial potential. Whilst the economics of PHB/ ble in the USA using Arabidopsis as as a
V production by large-scale bacterial fermen- model plant system (POIRIERet al., 1992).
tation are regarded as only just favorable for Alternative plants to oilseed rape could clear-
commercial exploitation, the economics of ly include sunflower, which is clearly amena-
production would be considerably enhanced ble to genetic modification (see Tab. 2), and
if the same production could be achieved in possibly even the palm oil tree. Yields, how-
plants. Just as plants produce triacylglycerol ever, at the moment are very far from any
oils and fats at a tenth, or even less, of the practical value, and considerable technical
cost of producing the same materials biotech- work, if not scientific innovation, will be nec-
nologically, so the costs of producing PHB essary to achieve commercially viable yields.
would be considerably decreased by switching The future demand for PHB and related
production into plants. For the work to be molecules is currently seen to depend on their
successful, yields of PHB will have to equal value as a biodegradable plastic. Its “environ-
those currently achieved by oilseed crops for mentally friendly” nature has been repeatedly
the production of triacylglycerol oils, i.e., up stressed. However, progress to producing oth-
to at least 30% of the harvested crop. Present er non-PHB, biodegradable plastics using
results indicate that PHB formation is only chemical synthesis is now proceeding apace
about 100 pg in the hybrid plants (POIRIERet (VERTet al., 1992). Should large-scale chemi-
al., 1992) which places the work as being still cal synthesis of an alternative, but still “envi-
in the preliminary experimental stages. ronmentally friendly” polymer be achieved in
For successful production of PHB in plants, the near future, this would have serious con-
the most suitable plant would appear to be sequences for the economic viability of PHB/
one which has the essential machinery for ac- V. Conceivably though, if production of PHB
cumulation of a (plant) product already in could be achieved in plants to the same yield
place. Moreover, not only will the genes for that they now produce oils, then this could re-
PHB biosynthesis have to be inserted but the main an alternative route to production. Long
genes for producing the existing product will term prospects for the production of PHB by
have to be deleted so that the flux of carbon microbial fermentation would appear uncer-
can then be diverted wholly into PHB: tain. However, just as microorganisms have
been dismissed as alternative producers of
COz - Sugar /‘PHB

Old product (oil or fat) oils and fats that are already commercially
available, the way forward for microbial
PHBs may be to identify higher valued prod-
ucts for niche markets.
As PHB is synthesized by direct condensa- The range of polyalkanoates produced by
tion of acetyl-CoA units (Fig. S), it is fairly microorganisms extends far beyond the sim-
obvious that the ideal plant for PHB produc- ple PHB and PHV polymers. Other potential-
ly useful polymers could include the poly-p- may be a minor component of apparently
hydroxyalkanoates where the side chain (R in only academic curiosity but then in the hands
XXII) could be an alkyl chain up to C9. These of an appropriate industrial company is scaled
are known as the medium chain length PHAs. up into being a significant product. In the fol-
Their occurrence has been described mainly lowing some examples of the range of micro-
in Pseudomonas oleovorans (WITHOLTet al., bial lipids are given; in some cases commer-
1990 ANDERSON and DAWES,1990). Forma- cial exploitation has occurred, in others the
tion is considerably enhanced by growing the lipids remain academic curiosities.
bacteria on long chain alkanes, fatty alcohols,
and fatty acids. However, growing the bacte-
rium on a long chain fatty acid, such as oleic 6.1 Biosurfactants
acid (18:l ) , only induces formation of the CI2
monomer (R=C9 in XXII) though, by using Most microorganisms produce a range of
it, PHAs are produced with an unsaturated surfactants that seemingly allow them to be-
side chain (WITHOLTet al., 1990) but still no come attached to water-insoluble substrates
longer than C,. The properties of these PHAs or to surfaces of leaves and other parts of
have been described (DE KONING,1995) and plants or soils. Production of surfactants may
have indicated some potential for producing, be enhanced by growing the organism on hy-
after chemical modification, a rubber latex drocarbons or vegetable oils and, for a while,
material that still retained its biodegradabili-this was considered to be a prerequisite for
ty. growth to take place. It is now not certain if
Considerably further work with the PHAs this is the case as some surfactants are pro-
may be anticipated over the next few years. duced in some abundance even when water-
Although there is currently only one indus- soluble substrates are used.
trial producer of a PHA (see above), a recent A wide diversity of chemical types is
(1994) conference held in Montreal, Canada, known ranging from glycolipids which usually
attracted over 300 delegates. The proceedings involve one or more fatty acyl residues at-
of this meeting were published in the Cana- tached glycosidically to a mono- or di-saccha-
dian Journal of Microbiology, probably dur- ride. Examples include the sophorolipid from
ing 1995. Candidu bombicolu (XXIII) and the rham-
nolipids from Pseudomonus aeruginosa
(XXIV). In some cases, macromolecular sur-
factants are formed with a M , of up to lo6 Da
in which the lipid component may not be the
major one. An extensive monograph on the
6 Other Lipids entire subject has been recently published
(KOSARIC,1993). Readers are therefore re-
The range of microbial lipids is, of course, ferred to this book for full details of the range
extensive (see RATLEDGEand WILKINSON, and potential of these molecules.
1988a, 1989). The biotechnological exploita-
tion of any particular lipid will depend upon
the perception of the purpose to which that
lipid can be put. In some instances the forma-
tion of a particular lipid may have been
known for many years but how it may be of
commercial benefit is not so obvious. An ex-
ample of this would be the formation of the
sophorolipids by several Candida spp. which, Ladone form
although excellent surfactants, have not R=HorCH3
proved sufficiently superior to chemically-
Sphorolipid
produced materials to warrant full scale pro-
duction (see below). In other cases, the lipid M(m
6 Other Lipids 181
ness of these organisms from conventional

bacteria, contain a plethora of novel lipid
types not seen elsewhere in either prokaryotic
or eukaryotic organisms. The organisms
which are regarded as the most ancient of all
life forms, are amongst the most resistant of
all organisms to the extremes of environment.
They therefore include the thermoacido-
philes, the extreme halophiles, and the me-
R = H or - CH-CHz - COOH thanogens.
I The lipids of these bacteria are character-
(CH2)6 ized by being ether, rather than ester, deriva-
I
CH3
tives of glycerol. The alkyl groups, however,
are not derived from fatty acids but are
formed by the mevalonate pathway which
Rhamnolipid leads to the formation of the isopentenyl lip-
ids. (The isopentenyl-derived lipids include
XXIV carotenoids, sterols, and other terpenoid lip-
ids which are of ubiquitous distribution but
the archaebacterial ether lipids are confined
to the Archaea.) Some typical examples of
these lipids are given in Fig. 9. Although
these lipids also use glycerol as the polyol
unit to which the acyl groups are attached,
they are attached to the sn-2 and sn-3 posi-
tions (c.f. Sect. 1.1) rather than the sn-1 and
sn-2 positions of the conventional diacylglyc-
COOH erols and phospholipids. The ether lipids may
be classed as di- or tetraethers depending on
the number of ether linkages that are present
Spiculisporic acid
in a particular molecule.
Excellent reviews of these highly unusual
XXV
lipids have been written by DE ROSA and
GAMBACORTA(1988), SMITH(1988), KATES
Of potential commercial relevance in this (1992, 1993), and by GAMBACORTAet al.,
field but not included in KOSARIC'Smono- (1995). A comprehensive monograph has also
graph, is spiculisporic acid (XXV) which is recently been published (KATES et al., 1993).
produced at up to 110 g L-' by Penicillium The much greater chemical stability of these
spiculisporum and whose possible uses have lipids than the acylated glycerol ester lipids of
included acting as a surfactant or as a synthet- Eubacteria and the Eukaryota (the other two
ic intermediate for such materials (ISHIGAMI, domains of the microbial world) has sug-
1993). Other fatty acid derivatives may also gested the means whereby the archaebacteria
have a similar potential: the review by ISHI- are able to withstand temperatures of over
GAMI (1993) may, therefore, prove helpful in 110°C, salinities approaching that of saturated
delineation of these different molecules. salt solution, and pH values of 1 or even less.
The biotechnological applications of the lip-
ids are not immediately apparent though they
6.2 Ether (Archaebacterial) Lipids may present interesting ideas to chemists for
synthesis of novel compounds. The biotech-
The newly-created bacterial domain of Ar- nological applications of the Archaea them-
chaea, which has arisen from the appreciation selves, however, are the subject of much cur-
of the chemical and biochemical distinctive- rent speculative research; this area has been
'rd/JvJd
VH-0
I
~CH,-O
a
CH,OH
I
C
CH,OH
Fig. 9. Isoprenoid lipids of Archaea.

a 2,3-di-O-phytanyl-sn-glycerol (archaeol) found in several genera; b 2,3-di-O-biphytanyl-sn-glycerol, a
macrocyclic diether found in the thermophilic methanogen, Methanococcus, jannadschii; c Glycerol-dial-
kyl-glycerol tetraether (caldarchaeol) found in Sulfobolus and other genera; d Tetracyclized glycerol dial-
kylnonitol tetraether (cyclized nonitolcaldarchaeol) found in thermoacidophiles and some methanogens.
6 Other Lipids 183
recently reviewed by VENTOSAand NIETO Very few novel biotechnological applications

(1995) and LEUSCHNERand ANTRANIKANhave been identified as the amounts are
(1995) amongst others (see AGUILAR,1995). usually less than 5% of the dry microbial bio-
mass and growth of a microorganism specifi-
cally to produce a phospholipid would be
6.3 Phospholipids and clearly uneconomic. An attempt though was
made in the early 1980s to extract the phos-
Sphingolipids pholipid from bacteria being grown on me-
thanol as a source of single cell protein
Phospholipids, specially sn-1,2-diacyl glyc- (SMITH,1981). Some yeast “lecithin” (which
erol-3-phosphorylated compounds, are found is the unfractionated phospholipid fraction of
in all living cells save for the Archaea where which phosphatidylcholine is the predomi-
alternative phospholipids occur (see Fig. 14). nant type) is produced for the health food
The range of phospholipids in microorgan- market by extraction of spent brewer’s yeast.
isms is extensive and has been reviewed by In most cases, large-scale commercial de-
the present author (RATLEDGE,1987,1989b). mand for phospholipids is satisfied by using
R - CH - CH - CH20-Y
I I
OH NH
I
X
I. Sphingosinebases
X=H; Y=H;R=
Cm(CH2)12CH=CH- Sphingosine
CH3(CH2)14- Dihydrosphingosine
CH~(CH~)I~CH(OH)- Phytosphingosine
CH~(CH~)SCH=CH(CH&!CH(OH)- Dehydrophytosphingosine
11. Ceramides IIa. Ceramide phosphates
R=asinI; Y=H; X = R=asinI;X=asinII; Y =
CHNH2)n- 0
I
CH3(CH2)n-ICH(OH)CO- - 0 - P - OH
I
(n=10 to 24) 0
111. Sphingomyelins IV. Cerebrosides
R = a s i n I ; X=asinII; Y = R=asinI; X=asinII; Y =
0 - sugar (glucosyl or galactosyl)

Fig. 10. Structures and nomenclature II - phosphoinositol
of sphingolipids found in yeasts and
- CH2CH2k(CH3)3 - phosphoinositol-rnannose
I
fungi. 0-
plant-derived materials, which are described

loosely as lecithin, and are recovered as by- i
products following the refining of plant seed
oils (see SZUHAJ,1989, for their applications
in the food industry). However, phospholip- OH
ids also have important applications in the
pharmaceutical industries and can be used for OH
a number of functions (HANINand ANSELL, Stearoylphytosphingone
1987). Such functions include the use of high-
ly purified phosphatidylcholine as a lung sur- XXVI
factant for neonatal children (BANGHAM,
1992) and its uses to improve biocompatibility
of various medical devices including contact amides in which the base is either phytosphin-
lenses, implants, and disposable devices used gosine or dehydrosphingosine. KULMACZand
in human health care. It is now sold by a SCHROEPFER(1978) observed that addition
number of commercial companies. The mate- of pentadecanoic acid to the growth
rial may be produced by large-scale chroma- medium of Pichia ciferri increased the pro-
tographic purification but is more likely to be duction of the extracellular materials. This
synthesized from glycero-sn-3-phosphocho- technique is now used to produce a number
line, which can be produced from plant leci- of ceramides with this yeast. Patent applica-
thin, being reacted with ethyl esters of highly tions have been filed by Gist-Brocades (Ne-
purified individual fatty acids in the presence therlands) with respect to the production of
of appropriate phospholipases (see RAT- N-stearoylphytosphingosine(XXVI) and oth-
LEDGE,1994). Other uses for phospholipids er related compounds. The ceramides, after
could include preparation of artificial lipo- purification, are used in cosmetic industry for
somes which may be used as a means of de- the controlled release of dermatologically ac-
livery of chemotherapeutic agents to selected tive (or beneficial) compounds into the epi-
tissues of a patient (HANINand ANSELL, dermal, dermal, and subcutaneous tissues of
1987). If such liposomes require the presence the skin. The yeast ceramides are regarded as
of specific acyl groups or polar head groups identical to the ceramides that occur naturally
then it may be possible to identify these in in the human skin; this includes the correct
certain microorganisms: however, for reasons chirality at the three chiral centers in the mol-
already given it is unlikely that such lipids ecule (see XXVI). This ensures optimal per-
could be produced cost-effectively as the sole formance in various skin-care formulations. (I
microbial product from a biotechnological am grateful to Dr. H. STREEKSTRA of Gist-
process. Brocades for supplying the above informa-
Sphingolipids (see Fig. 10) are usually only tion.)
minor components in microbial lipids. They
occur in bacteria (WILKINSON,1988), in
yeasts (RATTRAY, 1988; RATLEDGE and 6.4 Prostanoid-Type Lipids
EVANS,1989), and in some fungi ( L ~ S E L ,
1988). All types of sphingolipid shown in Fig. 5, Prostaglandins, and the related leuco-
have been recognized in some microorgan- trienes and thromboxanes, exert unique phys-
ism. Their contents in cells are usually less iological control over many key metabolic se-
than 0.5% of the dry biomass. The four sphin- quences in animals, including humans (SIN-
gosine bases (see Fig. 5) do not usually occur CLAIR and GIBSON,1992). Such materials are
free but usually as acylated derivatives, or used clinically for induction of childbirth and
ceramides. are also required by a large number of medi-
The major microorganism of interest in this cal research groups. Prospects for producing
area is Hansenula ciferri (now known as Pich- such materials microbiologically may be re-
ia ciferri). This yeast appears to be unique in mote but some indication has been given that
producing both intra- and extracellular cer- the yeast Dipodascopsis uninucleata may be
7 Conclusions 185
e C ) H
Fig. 11. ARA metabolites pro-

duced by Dipodascopsis uninuclea- - -
ta; a a-pentanor PGF2, y-lactone;
b 3-HETE 3-hydroxyarachidonic Ho OH
acid (from VAN DYKet al., 1994). a b
able to convert exogenously supplied arachi- offset the cost of waste disposal via a biotech-
donic acid to a compound or compounds (see nological route. With respect to a microbial
Fig. 11) showing prostaglandin-like activities oil - or single cell oil (SCO) - this too could
when administered to experimental animals be similarly produced. There are two obvious
(KOCK et al., 1991; VAN DYK et al., 1991; additional advantages with this alternative
KOCK and RATLEDGE, 1993). Current ongo- strategy to producing SCP. Firstly, the oil
ing research should be able to determine would, if carefully selected, be worth consid-
within the next two or three years if such erably more than just chicken-feed SCP. This
prospects of producing prostanoid lipids in is nicely illustrated by the attempt to produce
this way are realistic. A recent review (VAN a cocoa butter equivalent (CBE) in New Zea-
DYKet al., 1994) has indicated that the well- land using deproteinized whey as substrate
characterized oxygenase systems that occur in (see Sect. 3.2.1.5). Even here though, given
plants and animals for the formation of hy- the comparatively high price of a CBE, there
droxy PUFAs, which become the immediate was still insufficient profit margin to proceed
precursors of the prostaglandins, may be against rival plant products selling for about
found in some lower fungi (Saprolegniales $2,000 per t.
and Lagenidiales) as well as species of Dipo- The second potential advantage of SCO
dascopsis. over SCP is that the SCO need not be used
for food or feed. SCP must be put back with
the food chain by being fed to animals which,
in turn, will be consumed, in whole or in part,
by ourselves. This then requires that the sub-
7 Conclusions strate, or feedstock, is of an acceptable food-
grade quality or poses no long-term toxico-
Microbial lipids seemingly offer an almost logical problems. SCO can, of course, when
bewildering array of possibilities for biotech- produced from food-acceptable substrates
nological exploitation. However, careful ex- also be used in food materials (see, e.g., the
amination reveals that in many cases the oil yeast SCO-CBE product already cited), but
or fat that is produced by a microorganism is an SCO could also be used for some technical
not essentially different from that found in a purpose. In this case, the requirement for the
plant oil, or occasionally, an animal fat. In substrate to be of food-grade quality no long-
these circumstances, the cost of the microbial er applies. In the extreme case, SCO could be
route of production is likely to be many times produced from any fermentable substrate if it
that of the existing route. The only way that were subsequently put to some non-food use
such processes could, therefore, be economi- such as incorporation into paints, lubricants,
cally viable would be if the microbial oil was detergents, plasticizers, etc. However, the
being produced as an adjunct to some waste cost-effectiveness of this scenario would have
disposal process. Such concepts were devel- to rely almost entirely on the biological proc-
oped widely in the 1960s and 70s for the pro- ess being used in some form of waste removal
duction of single cell proteins (SCP) which, or environmental cleanup process as the sell-
although just selling, literally, as chicken feed, ing price of the non-food SCO would proba-
nevertheless produced a positive income to bly be quite low. The advantage would be
that the SCO-producing microorganism this gives the microbial product a significant
achieves environmental cleanup and simulta- edge over the chemical product that rarely
neously produces a saleable end product has the right chirality. Thus where stereospe-
more valuable than SCP. cificity is an important attribute of a product,
As it appears at the present time, the future a biotechnological route will usually be found
of microbial oils probably lies outside these to be superior to a chemical process. This is
areas and the best chance for producing a also seen in the formation of ceramides by
commercially viable product is probably with yeast technology (Sect. 6.3).
high valued materials that are difficult to pro- The range of microbial lipids is enormous.
duce from plant or animal sources. This ap- Which are of potential commercial value and
proach began in the 1980s with the use of fun- which are of academic interest is hard to as-
gi to produce polyunsaturated fatty acids sess. Insight of a particular field may bring an
(PUFA) and especially y-linolenic acid appreciation to one person that a microbial
(GLA) that was available from only one or lipid is of value but this may not be apparent
two plant sources at exceedingly high prices. to most others. Therefore, the perusal of the
However, as with any high-priced material range of types of lipid molecules that are
which appears to generate considerable profit available from microbial sources could bring
for one or two producers, other producers rich rewards to the shrewd reader. However,
quickly began rival processes of the GLA- let me conclude by saying I have touched on
producing plant crops so that in a few short only some of the microbial lipids; there are
years the price of GLA-rich oils was halved many others that have not been included di-
and fell even lower. Profit margins which rectly here but may just have been referred to
were sufficient to maintain commercial inter- obliquely or en passant. Nevertheless, I hope
est in producing SCO-GLA in the mid to late that I have given sufficient references for the
1980s then disappeared under this downward reader to pursue some of these more esoteric
commercial pressure. opportunities for themselves: fortune will al-
The scene in PUFAs was now moved on ways favor the prepared mind.
beyond GLA with other PUFAs, especially
ARA and DHA, becoming the principal tar-
gets for microbial production. Whether these
other PUFAs will follow the example of GLA 8 References
and become more abundant, and thus less ex-
pensive, from other non-microbial sources ACKMAN, R. (1994), Animal and marine lipids, in:
will be a risk that the biotechnologist must, Technological Advances in Improved and Alter-
therefore, try to assess. In the longer term, native Sources of Lipids (KAMEL,B. s., KAKU-
though, it is probably these selected, very DA, Y., Eds.), pp. 292-328. London: Blackie.
high-priced fatty acids, or derivatives from AGUILAR, A. (Ed.) (1995), Special Review Topic:
them, that will become the next SCOs to en- Biotechnology of Extremophile Microorgan-
isms, World J. Microbiol. Biotechnol. 11, issue
ter the marketplace. NO. 1, pp. 7-131.
Certain lipids, besides those based on the AHERN,T. J., KATOH,S., SADA,E. (1983), Arachi-
triacylglycerols, are already being produced donic acid production by the red alga Porphyrid-
biotechnologically where there is a market ium cruentum, Physiol. Plant. 16,36-643.
opportunity at the correct (i.e., profitable) AKIMOTO, M., ISHII,T., YAMAGAKI, K., OHTAGU-
price level. Carotenoids, such as pcarotene CHI, K., KOIDE, K., YAZAWA,K. (1990), Pro-
and astaxanthin, are produced commercially duction of eicosapentaenoic acid by a bacterium
(see Sect. 4.2) but need to be marketed very isolated from mackerel intestines, J. Am. Oil
shrewdly to convince the purchaser that these Chem. SOC. 67,911-915.
AKIMOTO, M., ISHII,T., YAMAGAKI, K., OHTAGU-
microbial products are superior to the chemi- CHI, K., KOIDE, K., YAZAWA,K. (1991), Metal
cally-produced, and thus cheaper, products. salts requisite for the production of eicosapen-
As microorganisms tend to produce only one taenoic acid by a marine bacterium isolated from
chiral isomer - and usually this is equivalent mackerel intestines, J. Am. Oil Chem. SOC. 68,
to the existing plant or animal material - then 504-508.
8 References 187
AMANO,N., SHINMEN, Y., AKIMOTO,K., KAWA- BANGHAM, A. (1992), “Surface tension” in lungs,
SHIMA, H., AMACHI, T., SHIMIZU,S., YAMADA, Nature (London) 359, 110.
H. (1992), Chemotaxonomic significance of fatty BARNETT,J. A., PAYNE,R. W., YARROW,D.
acid composition in the genus Mortierella (Zygo- (1990) Yeasts; Characteristics and Identification,
mycetes, Mortierellaceae), Mycotaxon 44, 257- 2nd Edn. Cambridge: University Press.
265. BASIRON,Y., IBRAHIM, A. (1994), Palm oil: old
AN, G.-H., SCHUMAN, D. B., JOHNSON,E. A. myths new facts, INFORM 5, 977-980.
(1989), Isolation of Phafla rhodozyma mutants BEAVAN,M., KLIGERMAN, A., DRONIUK,R.,
with increased astaxanthin content, Appl. Envi- DROUIN, C., GOLDENBERG, B., EFFIO,A., Yu,
ron. Microbiol. 55, 116-123. P., GIULIANY, B., FEIN,J. (1992), Production of
AN, G. H., BIELICH,J., AUERBACH, R., JOHNSON, microbial cocoa butter equivalents, in: Industrial
E. A. (1991), Isolation and characterization of Applications of Single Cell Oils (KYLE, D. J.,
carotenoid hyperproducing mutants of yeast by RATLEDGE, C., Eds.), pp. 156-184. Champaign,
flow cytometry and cell sorting, Biotechnology 9, I L American Oil Chemists’ Society.
70-73. BECKER,E. W. (1993), Microalgae: Biotechnology
ANDERSON, A. J., DAWES,E. A. (1990), Occur- and Microbiology. Cambridge: University Press.
rence, metabolism, metabolic role, and industrial BEDNARSKI, W., LEMAN,J., TOMASIK, J. (1986),
uses of bacterial polyhydroxyalkanoates, Micro- Utilization of beet molasses and whey for fat
biol. Rev. 54, 450-472. biosynthesis by a yeast, Agric. Wastes 18, 19-26.
BAILLIEZ,C., LARGEAN,C., CASADEVALL, E. BEHALOVA, B., VORISEK,J. (1988), Increased ste-
(1985), Growth and hydrocarbon production of rol formation in Saccharomyces cerevisiae. Anal-
Botryococcus braunii immobilized in calcium al- ysis of cell components and ultrastructure of vac-
ginate gel, Appl. Microbiol. Biotechnol. 23, 99- uoles, Folia Microbiol. 33, 292-297.
105. BEN-AMOTZ,A., AVRON,M. (1980), Glycerol,
beta-carotene and dry algal meal production by
BAILLIEZ, C., LARGEAN,C.,BERKALOFF, C., CASA- commercial cultivation of Dunaliella, in Algae
DEVALL,E. (1986), Immobilization of Botryo- Biomass (SHELEF,G., SOEDER,C. J., Eds.), pp.
coccus braunii in alginate: influence on chloro- 603-610. Amsterdam: Elsevier.
phyll content, photosynthetic activity and degen-
eration during batch cultures, Appl. Microbiol. BEN-AMOTZ,A., AVRON,M. (1983), On factors
which determine massive p-carotene accumula-
Biotechnol. 23,361-366.
tion in the halotolerant alga Dunaliella bardawil,
BAILLIEZ,C., LARGEAN,C., CASADEVALL, E., Plant Physiol. 72, 593-597.
YANG,L. W., BERKALOFF, C. (1988), Photosyn- BEN-AMOTZ,A., LERS, A., AVRON,M. (1988),
thesis, growth and hydrocarbon production of Steroisomers of pcarotene and phytoene in the
Botryococcus braunii immobilized by entrap- alga Dunaliella bardawil, Plant Physiol. 86,
ment and adsorption in polyurethane foams, 1286-1291.
Appl. Microbiol. Biotechnol. 29, 141-147. BEN-AMOTZ, A,, MOKADY, S., EDELSTEIN, S., Av-
BAJPAI,P., BAJPAI,P. K. (1992), Arachidonic acid RON, M. (1989), Bioavailability of natural isomer
production by microorganisms, Biotechnol. mixture as compared with all-trans-p-carotene in
Appl. Biochem. 15, 1-10. rats and chicks, J. Nutr. 119, 1013-1019.
BAJPAI,P. K., BAJPAI,P., WARD,0. P. (1991a), BERNHARD, K., ALBRECHT, H. (1948), Die Lipide
Optimization of production of docosahexaenoic aus Phycomyces blakesleeanus, Helv. Chim. Acta
acid (DHA) by Thraustochytrium aureum 31,977-988.
ATCC 34304, J. Am. Oil Chem. SOC. 68, 509- BOROWITZKA, M. A. (1988), Fats, oils and hydro-
514. carbon, in: Micro-Algal Biotechnology (BORO-
BAJPAI,P., BAJPAI,P. K., WARD,0. P. (1991b), WITZKA, M. A., BOROWITZKA, L. J., Eds.), pp.
Production of docasohexaenoic acid by Thraus- 257-287. Cambridge: University Press.
tochytrium aureum, Appl. Microbiol. Biotechnol. BOROWITZKA, M. A. (1992), Comparing caroteno-
35, 706-710. genesis in Dunalliella and Haematococcus: impli-
BAJPAI,P., BAJPAI,P. K., WARD,0. P. (1991c), cations for commercial production strategies, in:
Effects of ageing Mortierella mycelium on pro- Profiles on Biotechnology (VILLA, T . G.,
duction of arachidonic and eicosapentaenoic ABALDE,J., Eds.), pp. 301-310. Servicio de Pub-
acids, J. Am. Oil Chem. SOC.68,775-780. lications, Universidade de Santiago, Spain.
BAJPAI,P. K., BAJPAI,P., WARD,0. P. (1992), Op- BOROWITZKA, M. A., BOROWITZKA, L. J. (Eds.)
timization of culture conditions for production (1988), Micro-Algal Biotechnology, Cambridge:
of eicosapentaenoic acid by Mortierella elongata University Press.
NRRL 5513, J. Ind. Microbiol. 9, 11-18. BOROWITZKA, L. J., BOROWITZKA, M. A. (1989),
PCarotene (provitamin A) production with al- CERDA-OLMEDO, E. (1989), Production of carote-

gae, in: Biotechnology of Vitamins, Pigments and noids by fungi, in: Biotechnology of Vitamins,
Growth Factors (VANDAMME, E. J., Ed.), pp. Pigments and Growth Factors (VANDAMME, E.
15-26. London: Elsevier Applied Science. J., Ed.), pp. 27-42. London: Elsevier Applied
BOROWITZKA, L. J., BOROWITZKA, M. A. (1990), Science.
Commercial production of p-carotene by Dunal- CIFERRI,O., TIBONI,0. (1985), The biochemistry
iella salina in open ponds, Bull. Marine Sci. 41, and industrial potential of Spirulina, Ann. Rev.
244-252. Microbiol. 39, 503-526.
BOSWELL,K. D. V., GLADUE,R., PRIMA,B., CLEMENT,G. (1975), Producing Spirulina with
KYLE,D. J. (1992), SCO production by fermen- C02, in: Single Cell Protein, Vol. 2 (TANNEN-
tative microalgae, in: Industrial Applications of BAUM,S. R., WANG,D. I. C., Eds.), pp. 79-89.
Single Cell Oils (KYLE,D. J., RATLEDGE,C., Cambridge, MA: M.I.T. Press.
Eds.), pp. 274-286. Champaign, IL: American COHEN,Z., COHEN,S. (1991), Preparation of eico-
Oil Chemists’ Society. sapentaenoic acid (EPA) concentrate from Por-
BOTHAM, P. A., RATLEDGE, C. (1979), A biochem- phyridium cruentum, J. Am. Oil Chem. SOC. 68,
ical explanation for lipid accumulation in Candi- 16-19.
da 107 and other oleaginous micro-organisms,J. COHEN,Z., HEIMER,Y. M. (1992), Production of
Gen. Microbiol. 114, 351-375. polyunsaturated fatty acid (EPA, ARA and
BOULTON, C. A. (1982), The biochemistry of lipid GLA) by the microalgae Porphyrium and Spiru-
accumulation in oleaginous yeasts, Ph. D. The- lina, in: Industrial Applications of Single Cell
sis, University of Hull, UK. Oils (KYLE, D. J., RATLEDGE,C., Eds.), pp.
BOULTON,C. A., RATLEDGE,C. (1981), Correla- 243-273, Champaign, I L American Oil Chemis-
tion of lipid concentration in yeasts with posses- ts’ Society.
sion of ATP:citrate lyase, J. Gen. Microbiol. 127, COHEN,Z., VONSHAK, A., RICHMOND, A. (1988),
169-176. Effect of environmental conditions on fatty acid
BOULTON,C. A., RATLEDGE,C. (1983a). Use of composition of the red alga Porphyridium cruen-
transition studies in continuous cultures of Lipo- rum: correlation to growth rate, J. Phycol. 24,
myces starkeyi, an oleaginous yeast, to investi- 328-332.
gate the physiology of lipid accumulation, J. COHEN,Z., DIDI,S., HEIMER,Y. M. (1992), Over-
Gen. Microbiol. 129, 2871-2876. production of y-linolenic acid and eicosapenta-
BOULTON,C. A., RATLEDGE,C. (1983b), Partial enoic acids by algae, Plant Physiol. 98, 569-572.
purification and some properties of ATP:citrate COOLBEAR, T., THRELFALL, D. R. (1989), Biosyn-
lyase from the oleaginous yeast Lipomyces star- thesis of terpenoid lipids, in: Microbial Lipids,
keyi, J. Gen. Microbiol. 129,2863-2869. Vol. 2 (RATLEDGE, C., WILKINSON, S. G., Eds.),
BRAMLEY, P. M. (1985), The in vitro biosynthesis pp. 115-254, London: Academic Press.
of carotenoids, Adv. Lipid Res. 21,243-279. CPC International INC. (1979), Process for the mi-
BRAMLEY, P. M., MACKENZIE, A. (1988), Regula- crobiological production of oil, Eur. Patent
tion of carotenoid biosynthesis, Curr. Top. Cell. 0005 277.
Regul. 29,291-343.
CPC International INC. (1982a), Multistage pro-
BYROM,D. (1990), Industrial production of copo- cess for the preparation of fats and oils, Brit.
lymer from Alcaligenes eutrophus, in: Novel Patent 2091285A.
Biodegradable Microbial Polymers (DAWES,E.
A., Ed.), pp. 113-117 (NATO AS1 Series No CPC International INC. (1982b), Preparation of
186). Dordrecht: Kluwer. fats and oils, Brit. Patent 2091286A.
BYROM,D. (1992), Production of poly-p-hydroxy- CRESSWELL, R. C., REES, T. A., SHAH,N. (Eds.)
butyrate: poly-p-valerate copolymers, FEMS Mi- (1989), Algal and Cyanobacterial Technology.
crobiol. Rev. 103, 247-250. London: Longmans.
CARMAN, G. M., HENRY,S. A. (1989), Phospholip- DAVIES,R. (1984), Oil from whey, Food Technol.
id biosynthesis in yeast, Annu. Rev. Biochem. 58, N. 2. (June), 33-37.
635-669. DAVIES,R. J. (1988), Yeast oil from cheese whey -
CASADEVALL, E., DIF, D., LARGEAU,C., GUDIN, process development, in: Single Cell Oil (MORE-
C., CHANMON, D., DESANTI,0. (1985), Studies TON, R. S., Ed.), pp. 99-145. Harlow: Long-
on batch and continuous cultures of Botryococ- man.
cus braunii, hydrocarbon production in relation DAVIES,R. J. (1992a), Advances in lipid processing
to physiological state, cell ultrastructure and in New Zealand, Lipid Technol. 4, 6-13.
phosphate nutrition, Biotechnol. Bioeng. 27, DAVIES,R. J. (1992b), Scale up of yeast oil tech-
286-295. nology, in: Industrial Applications of Single Cell
8 References 189
Oils (KYLE,D. J., RATLEDGE,C., Eds.), pp. S., BELSKY,M. (1969), Polyunsaturated fatty
196-218. Champaign, I L American Oil Chem- acids of aquatic fungi: possible phytogenetic sig-
ists’ Society. nificance, Comp. Biochem. Physiol. 29, 805-811.
DAVIES,R. J., HOLDSWORTH, J. E. (1992), Synthe- EROSHIN, V. K., KRYLOVA, N. I. (1983), Efficiency
sis of lipids in yeasts: biochemistry, physiology, of lipid synthesis by yeasts, Biotechnol. Bioeng.
production, Adv. Appl. Lipid Res. 1, 119-159. 25,1693-1700.
DAVIES,R. J., HOLDSWORTH, J. E., READER,S. L. EVANS,C. T. (1983), The control of metabolism in
(1990), The effect of low oxygen uptake rate on lipid-accumulating yeasts, Ph. D. Thesis, Univer-
the fatty acid profile of the oleaginous yeast, sity of Hull, UK.
Apiotrichum curvatum, Appl. Microbiol. Bio- EVANS,C. T., RATLEDGE,C. (1983), A compari-
technol. 33, 569-573 (erratum: ibid. 1991, 34, son of the oleaginous yeast, Candida curvata,
832-833). grown on different carbon sources in continuous
DAWES,E. A. (Ed.) (1990), Novel Biodegradable and batch culture, Lipids 18,623-629.
Microbial Polymers, (NATO AS1 Series No. EVANS,C. T., RATLEDGE,C. (1984a), Effect of ni-
186). Dordrecht: Kluwer. trogen source on lipid accumulation in oleagi-
DE KONING,G. (1995), Physical properties of bac- nous yeasts, J. Gen. Microbiol. 130,1693-1704.
terial poly((R)-3-hydroxyalkanoates), Can. J. EVANS,C. T., RATLEDGE,C. (1984b), Influence of
Microbiol. 41 (Suppl. l), 303-309. nitrogen metabolism on lipid accumulation by
DE LONG,E. F., YAYANOS,A. A. (1986), Bio- Rhodosporidium toruloides CBS 14, J. Gen. Mi-
chemical function and ecological significance of crobiol. 130, 1705-1710.
novel bacterial lipids in deep-sea prokaryotes, EVANS,C. T., RATLEDGE,C. (1984c), Phospho-
Appl. Environ. Microbiol. 51, 730-737. fructokinase and the regulation of the flux of
DE ROSA,M., GAMBACORTA, A. (1988), The lip- carbon from glucose to lipid in the oleaginous
ids of Archaebacteria, Prog. Lipid Res. 27, 153- yeast Rhodosporidium toruloides, J. Gen. Micro-
175. biol. 130,3251-3264.
DELONG,E. F., YAYANOS, A. A. (1986), Biochem- EVANS,C. T., RATLEDGE, C. (1985a), Partial puri-
ical function and ecological significance of novel fication and properties of pyruvate kinase and its
bacterial lipids in deep-sea prokaryotes, Appl. regulatory role during lipid accumulation by the
Environ. Microbiol, 51, 730-737. oleaginous yeast Rhodosporidium toruloides
Dot, Y. (1990), Microbial Polyesters. Weinheim: CBS 14, Can. J. Microbiol. 31, 479-484.
VCH. EVANS,C. T., RATLEDGE,C. (1985b), The role of
DOSTALEK,M. (1986), Production of lipid from the mitochondria1 NAD :isocitrate dehydro-
+
starch by a nitrogen-controlled mixed culture of genase in lipid accumulation by the oleaginous

Saccharomyces fibuliger and Rhodosporidium yeast Rhodosporidium toruloides CBS 14, Can.
toruloides, Appl. Microbiol. Biotechnol. 24, 19- J. Microbiol. 31, 845-850.
23. EVANS,C. T., RATLEDGE,C. (1985c), Possible
DUBINSKY, Z., BERNER,T., AARONSON, S. (1978), regulatory roles of ATP:citrate lyase, malic en-
Potential of large-scale algal culture for biomass zyme and AMP deaminase in lipid accumulation
and lipid production in arid lands, Biotechnol. by Rhodosporidium toruloides CBS 14, Can. J.
Bioeng. Symp. 8, 51-68. Microbiol. 31, 1000-1005.
DULANEY,E. L. (1957), Preparation of ergosterol EVANS,C. T., RATLEDGE,C. (1986), The physio-
containing yeasts, US.Patent 2817624. logical significance of citric acid in the control of
DULANEY,E. L., STAPLEY,E. O., SIMPF, K. metabolism in lipid-accumulating yeasts, Bio-
(1954), Studies on ergosterol production by technol. Gen. Eng. Rev. 3,349-375.
yeasts, Appl. Microbiol. 2,371-379. EVANS,C. T., SCRAGG,A. H., RATLEDGE,C.
EL-REFAI,A. E. M., EL-KADY,I. A. (1968a), Ste- (1983), A comparative study of citrate efflux
rol production of yeast strains, Z. Allg. Mikro- from mitochondria of oleaginous and non-olea-
bwl. 8,355-360. ginous yeasts, Eur. J. Biochem. 130, 195-204.
EL-REFAI, A. E. M., EL-KADY,I. A. (1968b), Ste- FALL,R., PHELPS,P., SPINDLER,D. (1984), Bio-
rol biosynthesis in Saccharomyces fermentati, Z. conversion of xylan to triglycerides by oil-rich
A&. Mikrobiol. 8, 361-366. yeasts, Appl. Environ. Microbiol. 47, 1130-
EL-REFAI,A. E. M., EL-KADY,I. A. (1%9), Utili- 1134.
zation of some industrial by-products for the mi- FLOETENMEYER, M. D., GLATZ, B. A., HAM-
crobiological production of sterols from Saccha- MOND,E. G. (1985), Continuous culture fermen-
romyces fermentati, J. Bot. U.A.R. 12, 55-66. tation of whey permeate to produce microbial
ELLENBOGEN, B. A., AARONSON, S., GOLDSTEIN, oil, J. Dairy Sci. 68, 633-637.
Fuji Oil CO. LTD. (1979), Method for producing HANSSON, L., DOSTALEK, M. (1989), Effect of cul-
cacao butter substitute, Brit. Patent 1555OOO. ture conditions on mycelial growth and produc-
Fuji Oil CO. LTD. (1981), Method for producing tion of ylinolenic acid by the fungus Mucor ra-
cacao butter substitute, U.S. Patent 4268527. manniana, Appl. Microbiol. Biotechnol. 28,240-
GAMBACORTA, A,, GLIOZZI,A., DE ROSA, A. 246.
(1995), Archaeal lipids and their biotechnologi- HARRISON, D. M. (1986), The biosynthesis of caro-
cal applications, World J. Microbial Biotechnol. tenoids, Nut. Prod. Rep. 3, 205-215.
U,115-131. HARWOOD, J. L. (1994a), Environmental effects on
GANDHI, S. R., WEETE,J. D. (1991), Production of plant lipid metabolism, INFORM, 5, 835-839.
the polyunsaturated fatty acids arachidonic acid HARWOOD,J. L. (1994b), Environmental factors
and eicosapentaenoic acid by the fungus Phy- which can alter lipid metabolism, Prog. Lipid
tium ultimum, J. Gen. Microbiol. 137, 1825- Res. 33, 193-202.
1830. HASSAN,M., BLANC,P. J., GRANGER, L. M., PAR-
GELLERMAN, J. L., SCHLENK, H. (1979), Methyl- EILLEUX,A., GOMA,G. (1993), Lipid produc-
directed desaturation of arachidonic to eicosa- tion by an unsaturated fatty acid auxotroph of
pentaenoic acid in the fungus Saprolegnia para- the oleaginous yeast Apiotrichum curvatum
sitica, Biochim. Biophys. Acta 573,23-30. grown in single-stage continuous culture, Appl.
Microbiol. Biotechnol. 40,483-488.
GLATZ, B. A., FLOETENMEYER, M. D., HAM-
MOND,E. G. (1985), Fermentation of bananas HASSAN,M., BLANC, P. J., PAREILLEUX, A.,
and other food wastes to produce microbial lip- GOMA,G. (1994a), Selection of fatty acid auxo-
id, J. Food Prot. 48,574-577. trophs from the oleaginous yeast Cryptococcus
curvatus and production of cocoa butter equiva-
GOODWIN, T. W. (1980), The Biochemistry of the lents in batch culture, Biotechnol. Lett. 16, 819-
Carotenoids, 2nd Edn., Vol. 1. London: Chap- 824.
man & Hall. HASSAN,M., BLANC, P. J., PAREILLEUX, A.,
GOODWIN,T. W. (1983), The Biochemistry of the GOMA,G. (1994b), Production of single cell oil
Carotenoids, 2nd Edn., Vol. 2. London: Chap- from prickly pear juice fermentation by Crypto-
man & Hall. coccus curvatus grown in batch culture, World J.
GRANGER,L. M., PERLOT,P., GOMA,G., PAREIL- Microbiol. Biotechnol. 10, 534-537.
LEUX,A. (1993), Effect of various nutrient limi- HENDERSON, R. J., LEFTLEY,J. W., SARGENT,J.
tations on fatty acid production by Rhodotorula R. (1988), Lipid composition and biosynthesis in
glutinis, Appl. Microbiol. Biotechnol. 38, 784- the marine dinoflagellate Crypthecodium cohnei,
789. Phytochemistry 27, 1679-1683.
GUCKERT, J. B., COOKSEY, K. E. (1990), Triglycer- HENDERSON, R. J., MILLAR,R. M., SARGENT,J.
ide accumulation and fatty acid profile changes R., JOSTENSEN, J. P. (1993), Trans-neonoenoic
in Chlorella (Chlorophyta) during high pH-in- and polyunsaturated fatty acids in phospholipids
duced cell cycle inhibition, J. Phycol. 26, 72-79. of a Vibrio species in relation to growth condi-
GUERZONI,M. E., LAMBERTINI, P., LERCKER,G., tions, Lipids, 28, 389-396.
MARCHEITI,R. (1989, Technological potential HESSE,A. (1949) Industrial biosyntheses, part I.
of some starch degrading yeasts, StarcWStarke fats, Adv. Entymol. 9, 653-704.
37,52-57. HILLEN, L. W., POLLARD,G., WAKE, L. V.,
HAARD,N. F. (1988), Astaxanthin formation by WHITE,N. (1982), Hydrocarbon of the oils of
the yeast Phafia rhodozyma on molasses, Bio- Bdtryococcus braunii to transport fuels, Biotech-
technol. Lett. 10, 609-614. nol. Bioeng. 24, 193-205.
HAMMOND, E. G., GLATZ,B. A. (1990), Biotech- HIRANO,M., MORI,H., MIURA,Y., MATSUNAGA,
nology applied to fats and oils, in: Food Technol- N., NAKAMURA, N., MATSUNAGA, T. (1990), ‘y-
ogy, Vol. 2 (KING,R. D., CHEETHAM,P. S. J., Linolenic acid production by microalgae, Appl.
Eds.), pp. 173-217. London: Elsevier Applied Biochem. Biotechnol. 24/25, 183-191.
Science. HODGSON, P. A., HENDERSON, K. J., SARGENT, J.
HANIN,I., ANSELL,G. B. (Eds.) (1987), Lecithin: R.,LEFTLEY,J. W. (1991), Patterns of variation
Technological, Biological and Therapeutic As- in the lipid class and fatty acid composition of
pects. New York: Plenum Press. Nannochloropsis oculata (Eustigmatophyceae)
HANSSON, L., DOSTALEK, M. (1988), Effect of cul- during batch culture, I. The growth cycle, J.
ture conditions on mycelial growth and produc- Appl. PhyCOl. 3, 169-181.
tion of ylinolenic acid by the fungus Mortierella HOLDSWORTH, J. E., RATLEDGE,C. (1988), Lipid
rarnanniana, Appl. Microbiol. Biotechnol. 28, turnover in oleaginous yeasts, J. Gen. Microbiol.
240-246. 134,339-346.
8 References 191
HOLDSWORTH, J. E., VEENHUIS, M., RATLEDGE, Phaffia rhodozyma as an astaxanthin source in

C. (1988), Enzyme activities in oleaginous yeasts salmonid diets, Aquaculture 20, 123-134.
accumulating and utilizing exogenous or endoge- KATES,M. (1992), Archaebacterial lipids: struc-
nous lipids, J. Gen. Microbiol. 134, 2907-2915. ture, biosynthesis and function, Biochem. SOC.
HORROBIN, D. F. (1992), Clinical applications of Symp. 58,51-72.
n-6 essential fatty acids: atopic eczema and in- KATES, M. (1993), Membrane lipids of extreme ha-
flammation, diabetic neuropathy and retinopa- lophytes: biosynthesis, function and evolutionary
thy, breast pain and viral infections, in: Essential significance, Experientia 49, 1027-1036.
Fatty Acids and Eicosanoids, 3rd Int. Congr. KATES,M., KUSHNER,D. J., MATHESON,A. T.
(SINCLAIR, A., GIBSON,R., Eds.), pp. 367-372. (Eds.) (1993), The Biochemistry of Archaea (Ar-
Champaign: IL: American Oil Chemists’ Socie- chaebacteria). Amsterdam: Elsevier Science
tY. Publishers B.V.
HOUSTON,B., NIMMO,H. G. (1984), Purification KAWABATA, A., KANEYAMA, M. (1989), The ef-
and some kinetic properties of rat liver ATP:cit- fect of growth temperature on wax ester content
rate lyase, Biochem. J. 224, 437443. and composition of Euglena gracilis, J. Gen. Mi-
HOUSTON,B., NIMMO, H. G. (1985), Effects of crobiol. 135, 1461-1467.
phosphorylation on the kinetic properties of rat KENDRICK, A., RATLEDGE,C. (1992a), Lipids of
liver ATP:citrate lyase, Biochim. Biophys. Acta selected molds grown for the production of n-3
884,233-239. and n-6 polyunsaturated fatty acids, Lipids 27,
HRABAK, 0. (1992), Industrial production of poly- 15-20.
bhydroxybutyrate, FEMS Microb. Rev. 103, KENDRICK, A., RATLEDGE,C. (1992b), Lipid for-
251-256. mation in the oleaginous mold Entomophthora
HUDSON,B. J. F., KARIS,I. G. (1974), The lipids of exitalis grown in continuous culture: effects of
the alga Spirulina, J. Sci. Food Agric. 25, 759- growth rate, temperature and dissolved oxygen
763. tension on polyunsaturated fatty acids, Appl.
Microbiol. Biotechnol. 37, 18-22.
ISHIGAMI, Y. (1993), Biosurfactants face increasing
interest, INFORM 4, 1156-1165. KENDRICK, A., RATLEDGE, C. (1992c), Desatura-
tion of polyunsaturated fatty acids in Mucor cir-
JAREONKITMONGKOL, S., KAWASHIMA, H., SHI- cinelloides and the involvement of a novel mem-
RASAKA, N., SHIMIZU, S., YAMADA, H. (1992a), brane-bound malic enzyme, Eur. J. Biochem.
Production of dihomo-y-linolenic acid by a A5- 209,667-673.
desaturase-defective mutant of Mortierella alpi- KENDRICK, A., RATLEDGE,C. (1992d), Phospho-
nu 1s-4, Appl. Environ. Microbiol. 58, 2196-
lipid fatty acyl distribution of three fungi indi-
2200. cates positional specificity for n-6 vs. n-3 fatty
JAREONKITMONGKOL, S., KAWASHIMA, H., SHIMI- acids, Lipids 27, 505-508.
zu, S., YAMADA,H. (1992b), Production of KENNEDY,M. J., READER,S. L., DAVIES,R. J.
5,8,11-cis-eicosatrienoic acid by a A12-desatur- (1993), Fatty acid production characteristics of
ase-defective mutant of Mortierella alpina 1s-4, fungi with particular emphasis on y-linolenic
J. Amer. Oil Chem. SOC. 69, 939-944. acid, Biotechnol. Bioeng. 42, 625-634.
JAREONKITMONGKOL, S., SHIMIZU,S., YAMADA, KERNON,J. (1992), Vegetable fats in chocolate -
H. (1992c), Fatty acid desaturation-defective EC and US legislation issues, Lipid Technol. 4,
mutants of an arachidonic acid-producing fun- 113-114.
gus, Mortierella alpina 1s-4, J. Gen. Microbiol. KOCK,J. L. F., RATLEDGE,C. (1993), Changes in
138,997-1002. lipid composition and arachidonic turnover dur-
JENSEN,A. (1966), Report No. 31, Norwegian Insti- ing the life cycle of the yeast Dipodascopsis uni-
tute of Seaweed Research. nucleata, J. Gen. Microbiol. 139, 459464.
JOHNS,R. B., PERRY,G. J. (1977), Lipids of the KOCK,J. L. F., COETZEE,D. J., VAN DYK,M. S.,
marine bacterium Flexibacter polymorphus, TRUSCOTT,M., CLOETE,F. C., VAN WYK,V.,
Arch. Microbiol. 114, 267-271. AUGUSTYN, 0. P. H. (1991), Evidence for phar-
JOHNSON,E. A., AN, G. H. (1991), Astaxanthin macologically active prostaglandin in yeasts,
from microbial sources, CRC Crit. Rev. Biotech- South Afr.J. Sci. 87, 73-76.
nol. 11,297-321. KORITALA,S. (1989), Microbiologial synthesis of
JOHNSON,E. A., CONKLIN, D. E., LEWIS,M. J. wax esters by Euglena gracilis, J. Am. Oil Chem.
(1977), The yeast Phaffia rhodozyma as a dieta- SOC. 66,133-134.
ry pigment source for salmonids and crusta- KOSARIC,N. (Ed.) (1993), Biosurfactants: Produc-
ceans, J. Fish. Res. Board Can. 34,2417-2421. tion, Properties, Applications. New York Marcel
JOHNSON, E. A., VILLA,T. G., LEWIS,M. J. (1980), Dekker.
KREGER-VAN RIJ, N. J. W. (Ed.) (1984), The MARGALITH, P. Z. (1992), Pigment Microbiology.

Yeasts. A Taxonomic Study, 3rd Edn. Amster- London: Chapman & Hall.
dam: Elsevier Science Publishers. MEAD,J. F., SLATON,W. H. (1956), Metabolism of
KULMACZ, R. J., SCHROEPFER, G. J. (1978), Sphin- essential fatty acids. 111. Isolation of 5,8,11-eico-
golipid base metabolism, J. Am. Chem. Soc. 100, satrieonic acid from fat-deficient rats, J. Biol.
3963-3964. Chem. 219,705-709.
KYLE, D. J. (1991), Speciality oils from microalgae: MEYER,P. S., DU PREEZ, J. C., KILIAN,S. G.
new perspectives, in: Biotechnology of Plant Fats (1993), Selection and evaluation of astaxanthin-
and Oils (RATTRAY, J. Ed.), pp. 130-143. Cham- overproducing mutants of Phaffia rhodozyma,
paign, IL: American Oil Chemists’ Society. World J. Microbiol. Biotechnol. 9, 514-520.
KYLE,D. J. (1992), Production and use of lipids MIELKE,S. (1992), Trends in supply, consumption
from microalgae, Lipid Technol. 4, 59-64. and prices, in: Oils and Fats in the Nineties
KYLE,D. J., RATLEDGE,C. (1992), Industrial Ap- (SHUKLA,V. K. S., GUNSTONE, F. D., Eds.), pp.
plications of Single Cell Oils, Champaign, IL: 10-22. Lystrup, Denmark: International Food
American Oil Chemists’ Society. Science Centre AIS.
KYLE,D. J., BEHRENS,P., BINGHAM, S., ARNETT, MILLER,M., KOCK, J. L. F., BOTES,P. J. (1989a),
K., LIEBERMAN, D. (1988), Microalgae as a The significance of long-chain fatty acid compo-
source of EPA-containing oils, in: Biotechnology sitions and other phenotypic characteristics in
for the Fats and Oils Industry (APPLEWHITE, T. the yeast Pichia Hansen emend. Kurtzman, Syst.
H., Ed.), pp. 117-121. Champaign, IL: American Appl. Microbiol. 12, 70-79.
Oil Chemists’ Society. MOLINAGRIMA,E., SANCHEZ PEREZ,J. A., GAR-
KYLE,D. J., SICOTTE,V. J., REEB,S. E. (1992), CIA SANCHEZ, J. L. (1992), EPA from Isochrysis
Bioproduction of docosahexaenoic acid (DHA) galbana. Growth conditions and productivity,
by microalgae, in: Industrial Applications of Sin- Process Biochem. 21, 299-305.
gle Cell Oils (KYLE,D. J., RATLEDGE, C., Eds.), MOLINAGRIMA,E., SANCHEZ PEREZ,J. A., GAR-
pp. 287-300. Champaign, I L American Oil CIA CAMACHO,F., GARCIASANCHEZ,J. L.,
Chemists’ Society. LOPEZALONSO,D. (1993), n-3 PUFA produc-
LAGARDE,M., BURTIN,M., RIGAND,M., SPRE- tivity in chemostat cultures of microalgae, Appl.
CHER, H., DECHAVANNE,M., RENAND,S. Microbiol. Biotechnol. 38, 599-605.
(1985), Prostaglandin E2 - like activity of 2 0 MOLINAGRIMA,E., SANCHEZ PEREZ,J. A., GAR-
3n-9 platelet lipoxygenase end-product, FEBS CIA CAMACHO, F., FERNANDEZ SEVILLA, J. M.,
Lett. 181, 53-56. ACIEN FERNANDEZ, F. G. (1994), Effect of
LEE, Y. K. (1986), Enclosed bioreactors for the growth rate on the eicosapentaenoic acid and
mass cultivation of photosynthetic microorgan- docasahexaenoic acid content of Isochrysis gal-
isms: the future trend, Trends Biotechnol. 4,186- bana in chemostat culture, Appl. Microbiol. Bio-
189. technol. 41,23-27.
LEONARD, E. C. (1994), Salim group - major factor MOON, N. J., HAMMOND, E. G., GLATZ,B. A.
in oil palm trade, INFORM 5, 987-992. (1978), Conversion of cheese whey and whey
LEUSCHNER, C., ANTRANIKAN, G. (1995), Heat- permeate to oil and single-cell protein, J. Dairy
stable enzymes from extreme thermophilic and Sci. 61, 1537-1547.
hyperthermophilic microorganisms, World J. Mi- MORETON,R. S. (1985), Modification of fatty acid
crobiol. Biotechnol. l l , 95-114. composition of lipid accumulating yeasts with cy-
LOPEZ ALONSO,D., MOLINAGRIMA,E., SAN- clopropene fatty acid desaturase inhibitors,
CHEZ PEREZ,J. A., GARCIASANCHEZ, J. L., Appl. Microbiol. Biotechnol. 33,4145.
GARCIACAMACHO, F. (1992), Fatty acid varia- MORETON,R. S. (1988a), Single Cell Oil. Harlow:
tion among different isolates of a single strain of Longman.
Isochrysis galbana, Phytochemistry 31, 3901- MORETON,R. S. (1988b), The physiology of lipid
3904. accumulating yeasts, in: Single Cell Oil (MORE-
LOSEL, D. M. (1988), Fungal lipids, in: Microbial TON,R. S., Ed.), pp. 1-32. Harlow: Longman.
Lipids, Vol. 2 (RATLEDGE,C., WILKINSON, S. MORETON,R. S., CLODE,D. M. (1985), Microbial
G. Eds.), pp. 699-806. London: Academic desaturase enzyme inhibitors and their use in a
Press. method of producing lipids. Brit. Patent
MARGALITH, P. (1989), Vitamin D: the biotechnol- 8407 195.
ogy of ergosterol, in: Biotechnology of Vitamins, MURILLO,F. J., CALDERON, I. L., LOPEZ-DIAZ,I.,
Pigments and Growth Factors (VANDAMME, E. CERDA-OLMENDO, E. (1978), Carotene-super-
J., Ed.), pp. 81-104. London: Elsevier Applied producing strains of Phycomyces, Appl. Environ.
Science. Microbiol, 36. 477-479.
8 References 193
MURPHY,D. J. (1994a), Transgenic plants - a fu- the filamentous fungus Pythium irregulare, Appl.
ture source of novel edible and industrial oils, Microbiol. Biotechnol. 40,211-214.
Lipid Technol. 6, 84-91. O’LEARY,W. M., WILKINSON, S. G. (1988), Gram-
MURPHY,D. J. (Ed.) (1994b), Designer Crop Oils. positive bacteria, in: Microbial Lipids, Vol. 1
Weinheim: VCH. (RATLEDGE,C., WILKINSON, S. G., Eds.), pp.
NAKAHARA, T., YOKOCKI, T., KAMISAKA, Y., Su- 117-201. London: Academic Press.
ZUKI, 0.(1992), Gamma-linolenic acid from OSMAN,H. G., MOSTAFA,M. A., EL-REFAI,A. H.
genus Mortierella, in: Industrial Applications of (1969), Production of lipids and sterols by As-
Single Cell Oils (KYLE,D. J., RATLEDGE,C., pergillus funigatus. Some culture conditions fa-
Eds.), pp. 61-97. Champaign, I L American Oil vouring the formation of lipids and sterols, J.
Chemists’ Society. Chem. U.A.R. 12, 185-197.
NAKAJIMA, T., Izu, S. (1992), Microbial produc- OWUSU-ANSAH, Y. J. (1993), Enzymes in lipid
tion and purification of w-6 polyunsaturated fat- technology and cocoa butter substitutes, in:
ty acids, in: Essential Fatty Acids and Eicosa- Technological Advances in Improved and Alter-
noids (SINCLAIR, A., GIBSON,R., Eds.), pp. 57- native Sources of Lipids (KAMEL,B. s., KAKU-
64. Champaign, I L American Oil Chemists’ So- DA, Y., Eds), pp. 360-389. Glasgow: Blackie
ciety. Academic & Professional.
NEIDLEMAN, S., HUNTER-CEVERA, J. (1992), Wax PAN,J. G., RHEE,J. S. (1986), Biomass yields and
ester production by Acinetobacter sp. HOl-N, in: energetic yields of oleaginous yeasts in batch
Industrial Applications of Single Cell Oils culture, Biotechnol. Bioeng. 18, 112-114.
(KYLE,D. J., RATLEDGE,C., Eds.), pp. 16-28. PARKS,L. W., RODRIQUEZ, R. J., MCCAMMON, M.
Champaign, IL: American Oil Chemists’ Socie- T. (1984), Sterols of yeast: a model for biotech-
ty. nology in the production of fats and oils, in: Bio-
NELIS,H. J., DE LEENHEER, A. P. (1989), Micro- technology for the Oils and Fats Industry (RAT-
bial production of carotenoids other than @a- LEDGE c.,DAWSON, P., RAITRAY,J., Eds.), pp.
rotene, in: Biotechnology of Vitamins, Pigments 177-187. Champaign, I L American Oil Chem-
and Growth Factors (VANDAMME, E. J., Ed.), ists’ Society.
pp. 43-80. London, New York: Elsevier Applied PIERINGER, R. A. (1989), Biosynthesis of non-ter-
Science. penoid lipids, in: Microbial Lipids, Vol. 2 (RAT-
NICHOLS, B. W., WOOD,B. J. B. (1968), The occur- LEDGE, c., WILKINSON, S. G., Eds.), pp. 51-114.
rence of biosynthesis of ylinolenic acid in a London: Academic Press.
blue-green alga, Spirulina platensis, Lipids 3,46- POIRIER, Y., DENNIS,D., KLOMPARENS, K., NAW-
50. RATH,C., SOMERVILLE, C. (1992), Perspectives
NICHOLS,P. D., NICHOLS,D. S., BAKES,M. J. on the production of polyalkanoates in plants,
(1994), Marine oil products in Australia, FEMS Microbiol. Rev. 103, 237-246.
INFORM 5,254-261. POIRIER,Y., NAWRATH,C., SCHECHTMAN, L. A.,
NINET,L., RENAUT,J. (1979), Carotenoids, in: Mi- SATKOWSKI, M. M., NODA,I., SOMERVILLE, C.
crobial Technology, 2nd Edn., Vol. l (PEPPLER, (1994), PHB production in plants: some recent
H. J., PERLMAN, D., Eds.), pp. 529-544. San advances, Abst. Intern. Symp. on Bacterial PHA,
Diego: Academic Press. Montreal, Canada, p. 44.
NOGUCHI,Y., KAME,M., IWAMOTO,H. (1982), RADWAN, S. S. (1991), Sources of C2,,-polyunsatu-
Studies on lipid production by yeasts. Fatty acid rated fatty acids for biotechnological use, Appl.
composition of lipid from glucose and fatty acid Microbiol. Biotechnol. 35, 421430.
esters by Rhodotorula sp. and Candida sp., Yu- RADWAN, S. S., SOLIMAN, A. H. (1988), Arachi-
kagawa 31, 431437. donic acid from fungi utilizing fatty acids with
NOVOTNY,C., BEHALOVA,B., STRUZINSKY, R., shorter chains as sole sources of carbon and en-
NOVAK, M., ZAJICEK,J. (1988), Sterol composi- ergy, J. Gen. Microbiol. 134, 387-393.
tion of a AS, 7-sterol-rich strain of Saccharo- RATLEDGE,C. (1982), Microbial oils and fats: an
myces cerevisiae during batch growth, Folia Mi- assessment of their commercial potential, Prog.
crobiol. 33, 377-385. Ind. Microbiol. 16, 119-206.
O’BRIEN,D. J., SENSKE, G. E. (1994), Recovery of RATLEDGE,C. (1986), Lipids, in: Biotechnology,
eicosapentaenoic acid from fungal mycelia by 1st Edn., Vol. 4 (REHM,H. J., REED,G., Eds.),
solvent extraction, J. Am. Oil Chem. SOC. 71, pp. 185-213. Weinheim: VCH.
947-950. RATLEDGE,C. (1987), Microorganisms as sources
O’BRIEN,D. J., KURANTZ,M. J., KWOCZAK,R. of phospholipids, in: Lecithin: Technological,
(1993), Production of eicosapentaenoic acid by Biological and Therapeutic Aspects (HANIN,I.,
ANSELL,G. B., Eds.), pp. 17-35. New York Ple- Production of eicosapentaenoic acid by fresh-
num Press. water Vibrio, Lipids 27, 564-566.
RATLEDGE,C. (1988), Biochemistry, stoichiome- ROESSLER, P. G. (1990), Environmental control of
try, substrates and economics, in: Single Cell Oil glycerolipid metabolism in microalgae: commer-
(MORETON,R. S., Ed.), pp. 33-70. Harlow: cia1 implication and future research directions, J.
Longman. Phycol. 26,393-399. .
RATLEDGE,C. (1989a), Biotechnology of oils and Roux, M. P., KOCK, J. L. F., BETHA, A., DU
fats, in: Microbial Lipids (RATLEDGE,C., WIL- PREEZ,J. C., WELLS,G. V., BOTES,P. J. (1994),
KINSON, S. G., Eds.), pp. 567-668. London: Aca- Mucor - a source of cocoa butter and gamma-
demic Press. linolenic acid, World J. Microbiol. Biotechnol.
RATLEDGE, C. (1989b), Microbiological sources of 10,417-422.
phospholipids, in: Lecithins: Sources, Manufac- SANCHEZ, N., MARTINEZ, M., ARACIL,J., CORMA,
ture and Uses (SZUHAJ,B. F., Ed.), pp. 72-96. A. (1992), Synthesis of oleyl oleate as a jojoba
Champaign, I L American Oil Chemists’ Soci- oil analog, J. Am. Oil Chem. SOC. 69, 1150-
ety. 1153.
RATLEDGE,C. (1994), Biodegradation of oils, fats SAWAYAMA, S., MINOWA, T., DOTE,Y., YOKOYA-
and fatty acids, in: Biochemistry of Microbial MA, S. (1992), Growth of the hydrocarbon-rich
Degradation (RATLEDGE,C., Ed.), pp. 89-141. microalga Botryococcus braunii in secondarily
Dordrecht: Kluwer Academic Publisher. treated sewage, Appl. Microbiol. Biotechnol. 38,
RATLEDGE,C., EVANS,C. T. (1989), Lipids and 135-138.
their metabolism, in: The Yeasts, 2nd Edn., Vol. SCHAFER,L., KRAGBALLE, K. (1991), Supplemen-
3 (ROSE,A. H., HARRISON, J. S., Eds.), pp. 367- tation with evening primrose oil in atopic derma-
455. London: Academic Press. titis: effect of fatty acids in neutrophils and epi-
RATLEDGE,C., GILBERT,S. C. (1985), Carnitine dermis, Lipids 26,557-560.
acetyltransferase activity in oleaginous yeasts, SCHLEGEL, H. G., STEINBUCHEL, A. (Eds.) (1992),
FEMS Microbiol. Lett. 21,273-215. Int. Symp. Bact. Polyalkanoates ISBP ’92, FEMS
RATLEDGE, C., WILKINSON, S. G. (1988a), Micro- Microbiol. Rev., Vol. 103 (special issue).
bial Lipids, Vol. 1. London: Academic Press. SCHWEIZER, E. (1989), Biosynthesis of fatty acids
RATLEDGE,C., WILKINSON, S. G. (1989), Micro- and related compounds, in: Microbial Lipids,
bial Lipids, Vol. 2. London: Academic Press. Vol. 2 (RATLEDGE, C., WILKINSON, S. G., Eds.),
RATLEDGE,C., WILKINSON,S. G. (1988b), An pp. 3-50. London: Academic Press.
overview of microbial lipids, in: Microbial Lip- SEKULA,B. (1992), Wax ester production by yest,
ids, Vol. 1 (RATLEDGE,C., WILKINSON, s. G., in: Biotechnology of Plant Fats and Oils (RAT-
Eds.), pp. 3-22. London: Academic Press. TRAY,J. B. M., Ed.), pp. 162-176. Champaign,
RATLEDGE,C., BOULTON,C. A., EVANS,C. T. I L American Oil Chemists’ Society.
(1984), Continuous culture studies of lipid pro- SETO,A., WANG,H. L., HESSELTINE, C. W. (1984),
duction by oleaginous microorganisms, in: Con- Culture conditions affect eicosapentaenoic acid
tinuous Culture 8: Biotechnology, Medicine and content of Chlorella minutissima, J. Am. Oil
the Environment (DEAN,A. C. R., ELLWOOD, Chem. SOC.61,892-894.
D. C., EVANS,C. G. T., Eds.), pp. 272-291. Chi- SETO,A., KUMASAKA, K., HOSAKA,M., KOJIMA,
Chester: Ellis Honvood. E., KASHIWAKURA, M., KATO,T. (1992), Pro-
RATTRAY,J. B. M. (1988), Yeasts, in: Microbial duction of eicosapentaenoic acid by a marine mi-
Lipids, Vol. 1 (RATLEDGE,C., WILKINSON, S. croalgae and its commercial utilization for aqua-
G., Eds.), pp. 555-697. London: Academic culture, in: Industrial Applications of Single Cell
Press. Oils (KYLE, D. J., RATLEDGE,C., Eds.), pp.
RATI-RAY, J. B. M. (1991), Plant biotechnology and 219-234. Champaign, I L American Oil Chem-
the oils and fats industry, in: Biotechnology of ists’ Society.
Plant Fats and Oils (RATTRAY,J. B. M., Ed.), SHASHI, K., BACHHAWAT,A. K., JOSEPH, R.
pp. 1-35. Champaign, I L American Oil Chem- (1990), ATP:Citrate lyase of Rhodotorula gracil-
ists’ Society. is: purification and properties, Biochim. Bio-
RATTRAY,J. B. M. (1994), Biotechnological ad- phys. Acta 1033, 23-30.
vances in improved and alternative sources of SHAW,R. (1966), The polyunsaturated fatty acids
lipids, in: Technological Advances in Improved of microorganisms, Adv. Lipid Res. 4,107-174.
and Alternative Sources of Lipids (KAMEL,B. S., SHERIDAN, R., RATLEDGE,C., CHALK, P. A.
KAKUDA, Y., Eds.), pp. 50-92. London: (1989), Pathways to acetyl-CoA formation in
Blackie. Candida albicans, FEMS Microbiol. Lett. 69,
RINGO,E., JOSTENSEN, J. P., OLSEN,R. E. (1992), 165-170.
8 References 195
SHIFRIN,N. S. (1984), Oils from microalgae, in: tants, in: Industrial Applications of Single Cell
Biotechnology for the Oils and Fats Industry Oils (KYLE, D. J., RATLEDGE,C., Eds.), pp.
(RATLEDGE,C., DAWSON,P., RATTRAY,J., 185-195. Champaign, I L American Oil Chem-
Eds.), pp. 145-163. Champaign, I L American ists’ Society.
Oil Chemists’ Society. SMITH,E., WHITE,K. A., HOLT, D., FENTEM,P.
SHIMIZU, S., KAWASHIMA,H., SHINMEN, Y., AKI- A., BRIGHT,S. W. J. (1994), Expression of poly-
MOTO, K., YAMADA, H. (1988a), Production of hydroxybutyrate in oilseed rape, Abstr. Znt.
eicosapentaenoic acid by Mortierella fungi, J. Symp. Bacterial PHA, Montreal, Canada, p. 44.
Am. Oil Chem. SOC.65,1455-1459. SMITH,P. F. (1988), Arachaebacteria and other
SHIMIZU, S.,SHINMEN, Y., KAWASHIMA, H., AKI- specialized bacteria, in: Microbial Lipids, Vol. 1
MOTO,K., YAMADA, H. (1988b),Fungal mycelia (RATLEDGE,C., WILKINSON, S. G., Eds.), pp.
as a novel source of eicosapentaenoic acid. Acti- 489-553. London: Academic Press.
vation of enzyme(s) involved in eicosapentaeno- SMITH,S. R. L. (1981),Some aspects of ICI’s Single
ic acid production at low temperature, Biochem. Cell Protein Process, in: Microbial Growth on C,
Biophys. Res. Commun. 150,335-341. Compounds (DALTON,H., Ed.), pp. 343-348.
SHIMIZU,S., AKIMOTO,K., KAWASHIMA, H., London: Heyden and Sons.
SHINMEN, Y., YAMADA,H. (1989a), Production STADLER, T., MOLLION, J., VERDUS,M. C., KARA-
of dihomo-y-linolenic acid by Mortierella alpina MANOS, Y., MORVAN, H., CHRITIAEN, D. (Eds.)
1s-4,J. Am. Oil Chem. SOC. 66,237-241. (1988), Algal Biotechnology. Amsterdam: Else-
SHIMIZU,S., KAWASHIMA, H., AKIMOTO,K., vier.
SHINMEN, Y., YAMADA,H. (1989b), Microbial STEINSUCHEL, A., HUSTEDE,E., LIEBERGESELL,
conversion of an oil containing a-linolenic acid M., PIEPER,U., TIMM,A., VALENTIN, H. (1992),
to an oil containing eicosapentaenoic acid, J. Molecular basis for biosynthesis acids in bacte-
Am. Oil Chem. SOC.66,342-437. ria, FEMS Microbiol. Rev. 103,217-230.
SHIMIZU,S., AKIMOTO,K., SHINMEN, Y., KAWA- SZUHAJ,B. F. (Ed.) (1989), Lecithins: Sources,
SHIMA, H., SUGANO, M., YAMADA, H. (1991), Manufacture and Uses. Champaign, IL: Ameri-
Sesamin is a potent and specific inhibitor of 85- can Oil Chemists’ Society.
desaturase in polyunsaturated fatty acid biosyn- THOMAS,L. M., HOLUB,B. J. (1994), Nutritional
thesis, Lipids 26, 512-516. aspects of fats and oils, in: Technological Aware-
SHIMIZU,S., KAWASHIMA, H., WADA, M., YA- ness in Improved and Alternative Sources of Lip-
MADA,H. (1992), Occurrence of a novel sterol, ids (KAMEL,s., KAKUDA,Y., Eds.), pp. 16-49.
24,25-methylenecholest-5-en-3~-ol, in Mortierel- London: Blackie.
la alpina 1s-4,Lipids 21, 481433. TORNABENE, T. G., BOURNE,T. F., RAZIUDDIN,
SHINMEN, Y., SHIMIZU, S., AKIMOTO,K., KAWA- S., BEN-AMOTZ,A. (1985), Lipid and lipopoly-
SHIMA, H., YAMADA, H. (1989), Production of saccharide constituents of cyanobacterium Spi-
arachidonic acid by Mortierella fungi, Appl. Mi- rulina platensis, Mar. Ecol. Prog. Ser. 22, 121-
crobiol. Technol. 31,ll-16. 125.
SHLOMAI,P., BEN-AMOTZ,A,, MARGALITH,P. TOTANI,N., OBA,K. (1987),The filamentous fun-
(1991),Production of carotene stereoisomers by gus Mortierella alpina, high in arachidonic acid,
Phycomyces blakesleeanus, Appl. Microbiol. Lipids 22, 1060-1062.
Biotechnol. 34,458-462. TOTANI,N., OBA,K. (1988), A simple method for
SHUKLA,V. K. S. (1994), Present and future out- production of arachidonic acid by Mortierella al-
look of the world fats and oil.supply, in: Tech- pina, Appl. Microbiol. Biotechnol. 28, 135-137.
nological Awareness in Improved and Alternative TOTANI,N., SOMEYA,K., OBA, K. (1992), Indus-
Sources of Lipids (KAMEL,B. S., KAKUDA,Y., trial production of arachidonic acid by Mortierel-
Eds.), pp. 1-15. London: Blackie. la, in: Industrial Applications of Single Cell Oils
SIMOPOULOS, A. P. (1989),Summary of the NATO (KYLE,D. J., RATLEDGE,C., Eds.), pp. 52-60.
Advanced Research workshop on dietary 03 Champaign, I L American Oil Chemists’ Soci-
and 06 fatty acids: biological effects and nutri- ety.
tional essentiality,J. Nutr. 119,521-528. TYRRELL,D. (1967),The fatty acid composition of
SINCLAIR, A., GIBSON,R. (Eds.) (1992), Essential 17 Entomophthora isolates, Can. J. Microbiol.
Fatty Acids and Eicosanoids, 3rd Int. Congr. 13,755-760.
Champaign, I L American Oil Chemists’ Soci- TYRRELL, D. (1968),The fatty acid composition of
ety. some Entomophthoraceae 11, Lipids 3,368-372.
SMIT, H., YKEMA,A., VERBREE,E. C., VER- TYRRELL,D. (1971),The fatty acid composition of
WOERT,I. I. G. S., KATER,M. M. (1992),Pro- some Entomophthoraceae 111, Can. J. Microbiol.
duction of cocoa butter equivalents by yeast mu- 17,1115-1118.
TYRRELL, D., WEATHERSTONE, J. (1976), The fat- Eds.), pp. 98-117. Champaign, I L American Oil
ty acid composition of some Entomophthora- Chemists’ Society.
ceae IV. The Occurrence of branched-chain fatty WILKINSON, S. G. (1988), Gram-negative bacteria,
acids in Conidiobolus species, Can. J. Microbiol. in: Microbial Lipids, Vol. 1 (RATLEDGE,C.,
22, 1058-1060. WILKINSON, S. G., Eds.), pp. 299-488. London:
VANDYK,M. S., KOCK,J. L. F., COETZEE,D. J., Academic Press.
AUGUSTYN, 0. P. H., NIGAM, S. (1991), Isola- WILLNER,T., SITZMANN, W., WEBER,K. (1993),
tion of a novel arachidonic acid metabolite 3-hy- Dry fractionation for cocoa butter replacers, in:
droxy-5,8,11,14-eicosatetraenoicacid (3-HETE) Oil and Fats in the Nineties, pp. 162-175. Lys-
from the yeast Dipodascopsis uninucleata trup: International Food Science Centre AIS.
UOFS-Y128, FEBS Lett. 283, 195-198. WITHOLT,B., HUISMAN, G. W., PREUSTING, H.
VAN DYK, M. S., KOCK, J. L. F., BOTHA, A. (1990), Bacterial poly(3-hydroxyalkanoates) in:
(1994), Hydroxy long chain fatty acids in fungi, Novel Biodegradable Microbial Polymers
World J. Microbiol. Biotechnol. 10, 495-504. (NATO AS1 Series No. 186), (DAWES,E. A.,
VEGA, E. Z., GLATZ,B. A., HAMMOND, E. G. Ed.), pp. 161-173. Dordrecht: Kluwer.
(1988), Optimization of banana juice fermenta- WI-ITER,B., DEBUCH,H., STEINER,M. (1974), Die
tion for the production of microbial oil, Appl. Lipide von Endomycopsis vernalis bei verschie-
Environ. Microbiol. 54, 748-752. dener Stickstoff-Ernahrung, Arch. Microbiol.
VENTOSA,A., NIETO,J. J. (1995), Biotechnological 101,321-335.
applications and potentialities of halophilic mi- WOOD, B. J. B. (1988), Lipids of algae and proto-
croorganisms, World J. Microbiol. Biotechnol. zoa, in: Microbial Lipids, Vol. 1 (RATLEDGE, C.,
11,85-94. WILKINSON, S. G., Eds.), pp. 807-867. London:
VERT, M., FEIJEN,J., ALBERTOSON, A., Scorn, Academic Press.
G., CHIELLINI, E. (Eds.) (1992), Biodegradable WOODBINE, M. (1959), Microbial fat: microorgan-
Polymers and Plastics. Cambridge: Royal Soci- isms as potential producers, Prog. Ind. Micro-
ety of Chemistry. biol. 1, 179-245.
VERWOERT, I. I. G. S., YKEMA,A., VALKENBURG, YADWAD,V. B., WARD,0. P., NORONHA,L. C.
J. A. C., VERBREE,E. C., NIJKAMP,H. J. J., (1991), Application of lipase to concentrate the
SMIT,H. (1989), Modification of the fatty acid docosahexaenoic acid (DHA) fraction of fish oil,
composition in lipids of the oleaginous yeast Biotechnol. Bioeng. 38,956-959.
Apiotrichum curvatum by intraspecific sphero- YAMADA,H., SHIMIZU, S., SHINMEN, Y. (1987a),
plast formation, Appl. Microbiol. Biotechnol. 32, Production of arachidonic acid by Mortierella
327-333. elongata IS-5, Agric. Biol. Chem. 51, 785-790.
VOLKMANN, J. K. (1989), Fatty acids of microalgae YAMADA,H. S., SHIMIZU, S., SHINMEN, Y., KA-
used as feedstocks in aquaculture, in: Fats for the WASHIMA, H., AKIMOTO, K. (1987b). Produc-
Future (CAMBIE,R. C., Ed.), pp. 263-283. Chi- tion of arachidonic acid and eicosapentaenoic
acid by microorganisms in: Proc. World Con$
Chester: Ellis Horwood.
Biotechnol. for the Oils and Fats Industry (AP-
VONSHAK, A., COHEN,Z., RICHMOND, A. (1985), PLEWHITE, T. H., Ed.), pp. 173-177. Champaign,
The feasibility of mass cultivation of Porphyrid- I L American Oil Chemists’ Society.
ium, Biomass 8,13-25. YAMADA,H., SHIMIZU, S., SHINMEN, Y., AKIMO-
WATANABE, K., YAZAWA,K. (1992), Cloning of TO, K., KAWASHIMA, H., JAREONKITMONGKOL,
DNA coding for EPA - biosynthesis from a ma- S. (1992), Production of dihomo-y-linolenic acid,
rine bacterium into E. coli, J. Jpn. Oil Chem. arachidonic acid and eicosapentaenoic acid by
SOC. 42, 42-49. filamentous fungi, in: Industrial Applications of
WATANABE, K., ISHIKAWA, C., INOVE,H., CEN- Single Cell Oils (KYLE,D. J., RATLEDGE,C.,
HUA, D., YAZAWA,K., KONDO,K. (1994), In- Eds.), pp. 118-138. Champaign, I L American
corporation of exogenous docosahexaenoic acid Oil Chemists’ Society.
into various bacterial phospholipids, J. Am. Oil YAMAGUCHI, K., NAKANO,H., MURAKAMI, M.,
Chem. SOC. 71,325-330. KONOSU,S., NAKAYAMA, O., KANDA,M., NA-
WEETEJ. D. (1980), Lipid Biochemistry of Fungi KANUIRA, A., IWAMOTO, H. (1987), Lipid com-
and other Organisms. New York: Plenum position of a green alga, Botryococcus braunii,
Press. Agric. Biol. Chem. 51, 493-498.
WEETE,J. D., GANDHI, S. R. (1992), Enhancement YAMAUCHI, H., MORI,H., KOBAYASHI, T., SHIMI-
of C,, polyunsaturated fatty acid production in zu, S. (1983), Mass production of lipids by Lipo-
Pythium ultimum, in: Industrial Applications of myces starkeyi in microcomputer aided fed-batch
Single Cell Oils (KYLE, D. J., RATLEDGE,C., culture, Ferment. Technol. 61,275-280.
8 References 197
YANO, Y., NAKAYAMA, A., SAITO,H., ISHIHARA, SMIT,H. (1990), Lipid production of revertants
K. (1994), Production of docosahexaenoic acid of Ufa mutants from the oleaginous yeast Apio-
by marine bacteria isolated from deep sea fish, trichum curvatum, Appl. Microbiol. Biotechnol.
Lipids 29,527-528. 33, 176-182.
YAZAWA,K., ARAKI,K., OKAZAKI,N., WATA- YONGMANITCHAI, W., WARD, 0. P. (1989), Ome-
NABE, K., ISHIKAWA, C., INOVE, A., NUMAO, ga-3 fatty acids: alternative sources of produc-
N., KONDO,K. (1988), Production of eicosapen- tion, Process Biochem. 24, 117-125.
taenoic acid by marine bacteria, J. Biochem. 103, YONGMANITCHAI, W., WARD, 0. P. (1991),
5-1 1. Screening of algae for potential alternative
YAZAWA, K., WATANABE, K., ISHIKAWA,C., sources of eicosapentaenoic acid, Phytochemis-
KONDO,K., KIMURA,S. (1992), Production of try 30,2963-2967.
eicosapentaenoic acid from marine bacteria, in: YONGMANITCHAI,W., WARD, 0. P. (1992),
Industrial Applications of Single Cell Oils Growth and eicosapentaenoic acid production
(KYLE, D. J., RATLEDGE,C., Eds.), pp. 29-51. by Phaeodactylum tricornutum in batch and con-
Champaign, I L American Oil Chemists’ Soci- tinuous culture systems, Am. Oil Chem. SOC.69,
ety. 584-590.
YKEMA,A., VERBREE,E. C., VAN VERSEVELD, H.
W., SMIT,H. (1986), Mathematical modelling of YONGMANITCHAI, W., WARD, 0. P. (1993), Posi-
lipid production by oleaginous yeasts in contin- tional distribution of fatty acids, and molecular
uous culture, Antonie van Leeuwenhoek 52,491- species of polar lipids, in the diatom Phaeodacty-
506. lum tricornutum, J. Gen. Microbiol. 139, 465-
YKEMA,A., VERBREE,A. C., KATER, M. M., 472.
SMIT,H. (1988), Optimization of lipid produc- YOON,S. H., RHEE,J. S. (1983), Quantitative Phys-
tion in the oleaginous yeast Apiotrichum curva- iology of Rhodotorula glutinis for microbial lipid
tum in whey permeate, Appl. Microbiol. Biotech- production, Process Biochem. 18, 2-4.
nol. 29,211-218. YOON, S. H., RHIM,J. W., CHOI, S. Y., RYU, D. D.
YKEMA,A., VERBREE,E. C., NIJKAMP,H. J. J., W., RHEE,J. S. (1982), Effect of carbon and ni-
SMITH,H. (1989), Isolation and characterization trogen sources of lipid production of Rhodotoru-
of fatty acid auxotrophs from the oleaginous la gracilis, J. Ferment. Technol. 60,243-246.
yeast Apiotrichum curvatum, Appl. Microbiol. ZVGAGINTSEVA, I. S., PITRYUK,I. A., BAB’EVA,I.
Biotechnol32,76-84. P., RUBAN,E. L, (1975), Fatty acid composition
YKEMA,A., VERBREE, E. C., VERWOERT,I. I. G. of lipids of soil and epiphytic yeasts, Mikrobio-
S., VAN DER LINDEN,K. H., NIJKAMP, H. J. J, loiya USSR (Int. Edn.) 44, 625-631.
Biotechnology
5 Microbial Siderophores
GUNTHERWINKELMANN
HARTMUTDRECHSEL
Tiibingen, Federal Republic of Germany
1 Introduction 200
2 General Aspects 200
3 Bacterial Siderophores 201
3.1 Catecholate Siderophores 201
3.2 Hydroxamate Siderophores 209
3.3 Peptide Siderophores 212
3.4 Mycobactins and Related Siderophores 217
3.5 Citrate Hydroxamate Siderophores 220
3.6 Carboxylate Siderophores 223
3.7 Keto Hydroxy Bidentates 225
4 Fungal Siderophores 225
4.1 Ferrichromes 225
4.2 Coprogens 227
4.3 Rhodotorulic Acid 228
4.4 Fusarinines (Fusigens) 228
4.5 Rhizoferrins 229
5 Miscellaneous Compounds 230
6 Transport Mechanisms 233
7 Conclusion and Perspectives 235
8 References 236
200 5 Microbial Siderophores
1 Introduction may be predicted that a variety of fermenta-

tion processes will be optimized and precisely
controlled by the addition or withdrawal of
Iron nutrition in microbes and higher or- iron and siderophores. Other examples will
ganisms has become a most fascinating topic be briefly addressed at the end of this re-
in microbiology and biotechnology. Extensive view.
studies in a variety of microorganisms have
shown that low-molecular weight, ferric-spe-
cific ligands, named siderophores, are essen-
tial for growth and survival in natural envi-
ronments. The biosynthesis of siderophores 2 General Aspects
and their cognate membrane transport sys-
tems in bacteria are regulated at the tran- The majority of aerobic and facultative
scriptional level by internal iron sensors anaerobic microorganisms respond to a de-
named ferric uptake regulation proteins (Fur) creasing iron content in their environment by
which respond to the external iron concentra- the expression of siderophores and their cog-
tion and function as iron-loaded repressors. nate uptake systems. Because of the highly
The literature on siderophores has increased specific ferric iron binding, the ligands have
considerably during the past five years and attracted the attention of chemists who devel-
several comprehensive books offer detailed oped strategies for the synthesis of a number
descriptions of structures and functions of the of natural siderophores (LEE and MILLER,
various siderophores (WINKELMANN et al., 1983; BERGERON and MCMANIS,1991) and
1987; WINKELMANN, 1991a; BARTON and biomimetic analogs (SHANZERand LIBMAN,
HEMMING, 1993; WINKELMANN and WINGE, 1991). The isolation of siderophores from bio-
1994). Attention has been paid primarily to logical fluids is based on various extraction
the particular functions of siderophores, their and detection methods (NEILANDSand NA-
occurrence and distribution among bacteria KAMURA, 1991). However, yields are very
and fungi. Thus, plenty of information on si- low sometimes because of the fact that pro-
derophores already exists. From a biotechno- duction and excretion of siderophores are ne-
logical point of view iron is an extremely im- gatively regulated by iron. Therefore, mu-
portant element for fermentation processes in tants defective in regulation of siderophore
order to enhance growth and production biosynthesis or in the expression of outer
yields. Thus, concentration, presence or ab- membrane receptors have been used for the
sence of iron and other ions or medium con- production of siderophores (YOUNG and
stituents may greatly affect synthesis and ex- GIBSON,1979; WINKELMANN et al., 1994).
cretion of fermentation products. The biosynthesis and excretion of many side-
The present review is intended to update rophores can be detected in culture filtrates
the existing structural diversity of sidero- by the appearance of color due to metal-to-
phores and to specifically address some bio- ligand charge transfer bands. Thus, ferric hy-
technological aspects of siderophores which droxamate complexes show absorption maxi-
had already been the aim of the previous re- ma at 420-440 nm, resulting in a yellow-
view in Vol. 4 of the first edition of this series brown color, while ferric catecholate com-
published about ten years ago (WINKEL- plexes are red-blue in color due to an absorp-
MA“, 1986). Since then, a number of novel tion maximum at 495-520 nm. Ferric carbox-
siderophore structures have been described ylate siderophores show only a weak absorp-
which are not only variations of known struction at 335 nm (DRECHSEL et al., 1992) which
tures but even represent novel types or results in a pale yellow color.
classes of siderophores. Overall, the topic of Survival of microorganisms in an aerobic
siderophores seems to be a never ending sto- atmosphere where the concentration of sol-
ry that continues to inspire the community of uble ferric ions in solution at neutral pH is
microbiologists. The biotechnological value smaller than lO-”M has been a challenge
of these results still is not fully appreciated. It for about 3.5 billion years and enabled the
evolution of siderophores and their cognate cally stable chromic complexes have been
membrane transport systems. Moreover, side- used to study the metabolism of siderophores
rophore-mediated iron acquisition has been in various microorganisms. The physicochem-
correlated with the ability of various microor- ical properties of natural and synthetic sidero-
ganisms to establish and maintain infection of phores have been compiled in a recent review
a host, although examples of virulence en- by CRUMBLISS (1991). The reader is also re-
hancement by siderophores are still scarce ferred to a comprehensive description de-
(BAGG and NEILANDS,1987; HESEMANN, voted solely to the solution and structural
1987; ACTISet al., 1988; ENARD et al., 1988; chemistry of siderophores (MATZANKEet al.,
STOJILJKOVIC and HANTKE,1992). 1989).
The history of siderophores started with
the discovery of “growth factors” for myco-
bacteria and fungi (reviewed in: WINKEL-
MANN, 1991a). NEILANDS (1952) was the first
to isolate a siderophore in crystalline form 3 Bacterial Siderophores
and thus opened an era of intensive search for
new iron-binding compounds, which were
previously named sideramines or sideroch- 3.1 Catecholate Siderophores
romes and are now collectively designated as
siderophores. Siderophores are designed for Enterobactin (Fig. l), also referred to as
the solubilization, transport, and storage of enterochelin, is the prototype catecholate si-
iron in microorganisms. It is now generally derophore of enteric bacteria. It forms highly
agreed that siderophores including their bio- stable octahedral complexes with ferric iron
synthesis and the expression of the corre- (&= The three catecholamide-binding
sponding membrane-located transport sys- subunits of enterobactin are attached to a tri-
tems are carefully regulated by iron. Iron- L-serine lactone ligand backbone. While the
binding compounds that are not regulated ligand is uncharged at neutral pH, the ferric
seem to occur but do not fall into the category form is a trianionic complex. Convenient
of natural siderophores. Moreover, the term methods for the isolation of enterobactin (en-
siderophore should be reserved for the metal- terochelin) have been described by YOUNG
free ligand, although in some cases the names
had been given to the metal complex earlier
(ferrichrome, ferrioxamine, coprogen). The
corresponding iron-free compounds then
need the prefix desferri- or deferri-; sidero-
phores which represent the iron-free form
(enterobactin, aerobactin, staphyloferrin, rhi-
zoferrin) need the prefix ferri- or ferric when hi OH
iron is bound to the ligand.
The ligands involved in iron(II1) binding
are either phenolates or catecholates, hydrox- OH
amates, oxazolines, a-hydroxy carboxylates
(e.g., citrate derivatives), or keto hydroxy bi-
dentates. Ferric siderophores are octahedral
com lexes in which the coordinated metal ion
P
is d , high spin, and rapidly exchangeable.
After reduction, the ferrous ion shows little 0
affinity for the ligand, and this is obviously HO
the mechanism by which iron is removed
from siderophores within the cell or at the HO
cell surface of microorganisms. Non-reducible
aluminum and gallium complexes and kineti- Fig. 1. Enterobactin.
and GIBSON(1979), NEILANDSand NAKA- lier by this group (JALAL and VAN DER
MURA (1991), and BERNER et al. (1991). HELM,1989) and crystal structures are to be
HPLC separation and isolation from culture expected in the near future. Five additional
fluids have been described which also allow gene products are required to complete the
the simultaneous detection of degradation transport into the cells, of which only FepB
products of linear 2,3-dihydroxybenzoyl ser- has been sequenced so far (ELKINSand EAR-
ine derivatives (WINKELMANN et al., 1994). HARDT, 1989). The actual mechanism of ente-
The production procedure of YOUNG and robactin transport is still unsettled. Due to
GIBSON (1979) employs a mutant strain the very low redox potential ( -790 mV vs.
CfepA) of Escherichia coli which is unable to NHE at pH 7.4) a direct reduction model
transport the ferric enterobactin complex into could be excluded and an esterolytic degrada-
the cell. Whereas wild-type strains produce tion prior to reduction has been proposed.
enterobactin only during severe iron deficien- Alternative mechanisms involving protonated
cy ( < 0.2 p,M) and terminate production after molecular species of the ferric enterobactin
iron transport into the cell, fepA mutants con- complex during membrane transport have
tinue to produce high amounts of enterobac- been discussed (CASS et al., 1989; MAT-
tin in a medium not necessarily iron-defi- ZANKE, 1991). An additional internal redox
cient. reaction has also been suggested as 55Fe-ente-
Enterobactin can be estimated from aque- robactin is taken up while 67Ga-enterobactin
ous solutions (4 mL) after acidification with is not. Recent HPLC data from the authors'
conc. H2S04and extracted with an equal vol- laboratory have shown that the esterase
ume of ethyl acetate. If required, the extract seems to attack ferric enterobactin after en-
may be washed with an equal volume of sodi- trance via FepA, producing several linear di-
um phosphate buffer (0.1 M, pH 7) to remove hydroxybenzoyl serine products (WINKEL-
any of the hydrolysis products of enterobactin MANN et al., 1994) which then can be easily
which might be present in varying amounts reduced (-350mV vs. NHE). These results
(WINKELMANN et al., 1994), and the concen- are in favor of a combined esterase-reductase
tration is determined at 316 nm. The concen- mechanism and thus obviate any protonation
tration of ferri-enterobactin is determined or internal redox reaction of the ferric entero-
using the molar extinction coefficient bactin complex.
( ~ ~ ~ ~ ~ , , , =M-'cm-').
5600
The transport of ferric enterobactin into E.
coli requires the expression of the FepA out- Chrysobactin
er membrane protein (81 kDa), which has
been cloned, sequenced (LUNDRIGANand Chrysobactin (Fig. 2), a simple catechol-
KADNER,1986), and also crystallized (JALAL type siderophore containing only one 2,3-di-
and VAN DER HELM, 1989). Enterobactin is hydroxybenzoyl group connected to a dipep-
the most powerful natural iron sequestering
agent (reviewed by CRUMBLISS, 1991). Physi-
cochemical properties, analogs as well as me- OH
chanistic aspects of transport have been dis-
cussed in detail (MATZANKE, 1991; SHANZER
and LIBMAN,1991). The FepA receptor re- OH
cognizes the ferric enterobactin complex pres-
ent in a delta-cis configuration as shown by
comparison of natural enterobactin and syn- I
thetic lambda-cis-enantio-enterobactin(NEI-
LANDS et al., 1981). Purification of FepA by
FPLC has been reported from the plasmid-
harboring E. coli strain UT5600/pBB2 (ZHOU
I
I
CH20H
et al., 1995). Successful crystallization of the
purified FepA protein had been reported ear- Fig. 2. Chrysobactin.
tide was isolated from Erwinia chrysanthemi

which is a phytopathogenic species in Saint-
paulia plants (PERSMARK et al., 1989; PERS-
MARK and NEILANDS, 1992). Structure eluci-
dation revealed that chrysobactin is N-[N2-
(2,3-dihydroxybenzoyl)-~-lysyl]-~-serine.
While for similar natural monocatechol si-
derophores, like 2,3-dihydroxybenzoyl serine
and 2,3-dihydroxybenzoyl lysine, an L-confi-
guration was determined, the lysyl residue in Fig. 4. Azotochelin: R = COOH; myxochelin A:
chrysobactin has a D-configuration. However, R = CH20H.
as shown by transport experiments, the confi-
guration of the amino acids in chrysobactin
seems to be of minor importance compared to azotochelin can be extracted from the culture
that of the catechol-iron center (PERSMARK filtrate using ethyl acetate. Aminochelin is
et al., 1992). 4-N-2,3-dihydroxybenzoyl-l ,Cdiaminobutane,
and azotochelin is N,N'-bis-(2,3-dihydroxy-
benzoy1)-lysine. It was anticipated that these
Aminochelin, Azotochelin, Myxochelin A, compounds may possibly represent sidero-
and Protochelin phore precursors or degradation products ori-
ginating from a more complex siderophore
Azotobacter vinelandii produces several which indeed has been found recently and
catecholate siderophores under iron-limiting was named protochelin (Fig. 5) (TARAZet al.,
conditions (FUKASAWA et al., 1972; PAGE 1990). Although recently detected in Azoto-
and TIGERSTROM, 1988; CORBINand BULEN, bacter, (CORNISHand PAGE,1995), protoche-
1969; DEMANGEet al., 1988; BUDZIKIEWICZlin has been previously shown to occur in a
et al., 1992): 2,3-dihydroxybenzoic acid bacterium (DSM 5746) which is able to grow
(DHBA), aminochelin (Fig. 3), and azotoche- on methanol as the sole carbon source. The
lin (Fig. 4). These three compounds seem to structure of protochelin is a direct combina-
be biogenetically related. Aminochelin and tion of aminochelin and azotochelin. As
shown by PAGEand HUYER(1984) DHBA is
produced constitutively by Azotobacter,
OH whereas azotochelin and azotobactin are pro-
duced only under iron limitation.
Myxochelin A (Fig. 4) is a new catecholate
OH siderophore isolated from Angiococcus disci-
formis (Myxobacteria). The structure is N,N'-
bis-(2,3-dihydroxybenzoyl)-lysinol(KUNZEet
al., 1989) which is related to aminochelin
Fig. 3. Aminochelin. (N,N '-bis-(2,3-dihydroxybenzoyl)-lysine)pro-
mo" OH
OH
Fig. 5. Protochelin. 0
duced by Azotobacter vinelandii (PAGEand amino group and the Phydroxyl group of
TIGERSTROM, 1988). Myxochelin A exerts threonine are linked to an oxazoline ring. The
weak antibiotic activity on various gram-posi- crystal structure of the ligand was published
tive bacteria, e.g., Bacillus brevis, B. cereus, B. by ENG-WILMOT and VAN DER HELM(1980).
subtilis, B. megaterium, Micrococcus luteus, Exposure to acid leads to the formation of the
Staphylococcus aureus, Arthrobacter simplex, oxazoline ion which opens to yield the ester
Brevibacterium linens, Corynebacteriurn fas- and subsequently under neutral conditions
cians, and Nocardia corallina (KUNZEet al., undergoes the N-O-acyl shift to give the am-
1989). ide. This open form was named agrobactin A
(Fig. 6b). Whereas enterobactin adopts a del-
ta-cis configuration about the iron center, the
Agobactin configuration of agrobactin is lambda-cis.
Agrobactin and several other polyamine cate-
Agrobactin (Fig. 6a) is the characteristic si- cholamide chelators, such as parabactin and
derophore of Agrobacterium tumefaciens vibriobactin (see below), have been chemical-
(ONGet al., 1979). In low-iron media A . tu- ly synthesized (reviewed by BERGERON and
mefaciens B6 was found to produce a neutral, MCMANIS,1991). Recently, the structure of
ethyl acetate-extractable substance. The UV serratiochelin from Serratia marcescens has
spectrum revealed absorption maxima at 316 been elucidated, which is a derivative of agro-
and 252nm. Agrobactin was shown by its bactin (Fig. 6a) lacking the tetramethylenedi-
blue fluorescence in the ultraviolet range and hydroxybenzoylamide (EHLERTet al., 1994).
positive Arnow reaction to belong to the ca-
techol-type siderophores. Acid hydrolysis
yielded DHBA, spermidine, and L-threonine. Parabactin
The carboxyl group of DHBA as well as the
In 1975, TAITisolated a siderophore from
low-iron cultures of Micrococcus denitrificans
(now Paracoccus denitrificans) which he
named compound 111. A reinvestigation (PE-
a TERSON and NEILANDS, 1979) Of the StrUC-
ture (Fig. 6a) revealed an analogous structure
to agrobactin lacking the OH group of the
central catechol in position 3. This compound
was named parabactin. Again two forms are
conceivable: one possessing a closed oxazol-
ine ring (parabactin) and an open form (para-
bactin A).
Inspection of the molecular models of agro-
bactin and parabactin revealed that the tertia-
ry N-atom of the oxazoline ring and the o-hy-
droxyphenyl function are involved in the six-
coordinate ferric complex. Because of the op-
tically active substituent L-threonine, a parti-
cular coordination isomer of parabactins and
agrobactins can be expected. The CD spectra
revealed a lambda-cis configuration as in fer-
richrome. In addition, steric constraints ruled
out the possible presence of geometrical iso-
mers of the trans variety. The binding
strength of parabactins and agrobactins is
Fig. 6a. Agrobactin: R=OH, parabactin: R=H; b comparable to that of enterobactin which is
agrobactin A. approximately lo5*at pH 7.4.
The study of biological activities of agro- a

bactins and parabactins in Agrobucterium tu-
mefuciens and Purucoccus denitrificuns
showed that only the closed but not the open
oxazoline forms counteract growth retarda-
tion caused by EDTA. Using both labeled
metal and labeled ligand complexes, BERGE-
RON et al. (1985) presented evidence that pa-
rabactin operates by the “iron taxi” mecha-
nism. Iron was delivered to the cell, but the
ligand was not taken up in higher amounts.
The same authors also presented evidence
that enunrio-parabactin possessing a delta
coordination isomer was unable to transfer
higher amounts of labeled metal to Purucoc-
mO”
cus denitrificans. Therefore, the parabactin
receptor stereoselectively recognizes the nat- b
ural lambda coordination isomer.
Vibriobactin
From low-iron cultures of Vibrio cholerue

the siderophore vibriobactin (Fig. 7a) was iso-
lated by GRIFFITHet al. (1984). V. cholerue
(serovar 01 and 0139) is known to have Fig. 7a. Vibriobactin: R = OH; vulnibactin: R =H, b
caused several severe pandemic diarrheal dis- fluvibactin.
eases, and the latest one (the 7th) is still exist-
ing in Asia, Australia, and South America
(WACHSMUTH et al., 1994). Vibriobactin, like
agrobactin, contains three 2,3-dihydroxyben- ciency. Vibriobactin is not the only sidero-
zoyl residues and two residues of L-threonine phore utilized by V. cholerue as mutants de-
per molecule both of which form an oxazoline fective in vibriobactin synthesis can alterna-
ring. The polyamine backbone proved to tively use ferric citrate (SIGELet al., 1985) or
be N-(3-aminopropyl)-l,3-diaminopropane obtain iron from heme (STOEBNERand
(norspermidine), in contrast to spermidine PAYNE,1988). As shown below, additional
which is found in agrobactin and parabactin. carboxylate transport systems for vibrioferrin
Cells of V. cholerue Lou15 were used for may exist. The outer membrane receptor for
the original isolation of vibriobactin (GRIF- vibriobactin has been identified (STOEBNER
FITH et al., 1984). The yield of the prepara- et al., 1992) and the gene (viuA) encoding the
tion varied between 10-25 mg per 6 L batch. outer membrane receptor (74 kDa) has re-
Thin layer chromatography (TLC) on silica cently been cloned and shown to be negative-
gel plates using chloroform-methanol 4: 1 ly regulated by the fur gene at the transcrip-
gave one spot detected by its blue fluores- tional level (BUTTERTONet al., 1992). The
cence under UV light or by spraying with fer- role of fur in the iron-regulated expression of
ric chloride or iodine. The UV absorption the outer membrane receptor for vibriobactin
spectrum is qualitatively the same as that of (ViuA) and the outer membrane virulence
agrobactin. Biological activity testing re- protein (IrgA) has been demonstrated recent-
vealed that the growth of the producing v. ly (LITWINand CALDERWOOD, 1994). The
cholerue strain Loul5, which is only weakly fact that further proteins were observed
pathogenic, was stimulated by vibriobactin which were negatively regulated by iron but
and also by agrobactin with nearly equal effi- independent of Fur suggested that the gene
regulation in V. cholerue by Fur and iron is ture elucidation of the amonabactins have
much more complex than previously as- shown that four structural varieties of amona-
sumed. bactins exist (TELFORDet al., 1994). Amona-
bactins contain 2 mol DHBA, glycine and ly-
sine and 1mol of either tryptophan or phe-
Vulnibactin nylalanine. Two of them are glycine-deleted
forms (Fig. 8). A survey on various Aeromon-
The group of YAMAMOTOhas also re- as strains revealed that although most strains
ported on the structures of further polyam- produced amonabactins, some produced an
ine-containing catecholate siderophores from enterobactin-like or a so far uncharacterized
the human pathogen Vibrio vulnificus (OKU- siderophore (ZYWNO et al., 1992). In strains
JO et al., 1994a). The principal compound iso- producing amonabactins a biosynthetic gene
lated contained two oxazoline-bound salicyl (umoA) was identified in a Suu3A gene libra-
moieties and one outer DHBA residue and ry of Aeromonus hydrophilu 495A2 chromo-
was named vulnibactin (Fig. 7a). somal DNA (BARGHOUTHI et al., 1991). This
gene seems to correspond to the first 2,3-
DHBA biosynthetic enzyme (isochorismate
Fluvibactin synthetase) in E. coli (MASSADet al., 1994).
Fluvibactin (Fig. 7b) from Vibrio fluvialis

(YAMAMOTO et al., 1993) resembles agrobac- Anguibactin
tin with respect to the three DHBA residues
and the central oxazoline ring. However, flu- While the characteristic moiety of the side-
vibactin, like vibriobactin and vulnibactin, rophores described above are oxazoline rings,
contains norspermidine as a backbone (iden- the following structures contain a thiazoline
tical propane spacer between the two nitro- ring. The first member of this series is angui-
gens). This is consistent with the finding that bactin (Fig. 9a). Isolated from the fish patho-
Vibrio species contain norspermidine in its gen Vibrio unguillurum (ACTISet al., 1986), it
free form as a major polyamine (YAMAMOTO was one of the first siderophores regarded as
et al., 1991). an important plasmid-encoded virulence fac-
tor (CROSA, 1989). Structure elucidation
based on single-crystal X-ray diffraction stud-
Amonabactins ies revealed that anguibactin consists of 2,3-
dihydroxybenzoic acid linked to cysteine by a
Several amonabactins have been isolated thiazoline ring which in turn is linked by a hy-
from the fish pathogen Aeromonus hydrophi- droxamic acid bound to eN-hydroxy hista-
la (BARGHOUTHI et al., 1989). Recent struc- mine (JALALet al., 1989).
OH
/++=/OH
OH
Fig. 8. Amonabactins.
Phe can be substituted
by Trp, Gly-deleted
forms occur.
R' ferri-ferrithiocin is produced by Streptomyces

antibioticus (TU 1998; Fig. 9b) (NAEGELIand
ZAHNER, 1980) which resembles the structure
OH of aeruginic acid published previously (Fig.
9c). Desferri-ferrithiocin contains a hydroxy
pyridine residue linked to a thiazoline ring.
Although a function in iron transport has
never been shown in the producing strain, it
has been suggested to be a potent, orally
available iron chelator.
PyocheIin
HO
Pyochelin (Fig. 10a) is an iron-chelating
compound isolated from low-iron cultures of
Fig. 9. Anguibactin: R'=OH, R2=A, R3=H, Pseudomonas aeruginosa strain PAO-1. Pyo-
X = CH; desferri-ferrithiocin: R'= H, R2= CH3, chelin has also been found in clinical isolates.
R3= COOH, X =N; aeruginic acid: R' = R2= H, The structure, 2-(2-o-hydroxyphenyl-2-thia-
R3= COOH, X = CH. zolin -4-yl) -3-methylthiazolidine -4 -carboxylic
acid, is presumed to be biosynthesized from
Desferrithiocin salicylcysteinyl cysteine by cyclization and hy-
dration of the thiazolidine ring. Pyochelin was
Besides the indicator antibiotics p- and 7- purified by preparative TLC on silica gel G
rubromycins the iron-binding compound des- (Cox et al., 1981) and detected both by fluo-
OH
Fig. 1Oa. Pyochelin; b yersiniabactin.

rescence and a red reaction using an iron

spray. It forms a stable, red complex with fer-
ric iron. Although the ligand groups involved
in iron chelation have not been identified, it
may be assumed that iron is bound to the
phenyl O H group, the carboxyl group, and
one nitrogen in a 2:l ligand-to-iron ratio. The
simpler compound 2-o-hydroxyphenyl-2’-
thiazoline-4-carboxylic acid, named aeruginic
II Hzo
acid, is known to occur as a fermentation

product of P. aeruginosu. Ferrithiocin isolated
from Streptomyces untibioticus is related to
aeruginic acid, but contains hydroxy pyridine
instead of phenyl and a methyl group in the 0
vicinity of the carboxyl group resulting in an Fig. 11. Proferrorosamine A.
additional chiral center.
Yersiniabactin Ferrorosamine A
From cultures of Yersinia enterocolitica, an Erwinia rhupontici has been shown to pro-
iron-complexing and iron-transporting com- duce proferrorosamine A, ~-(2-pyridyl)-l-
pound named yersiniabactin was isolated pyrroline-5-carboxylic acid, a Fe(I1)-complex-
(HAAG et al., 1993; CHAMBERSand SOKOL, ing agent (Fig. 11) which after complexing
1994). The structure of the siderophore was gives a reddish pigment named ferrorosamine
determined by a variety of spectroscopic A (FEISTNERet al., 1983). While proferroros-
methods, including 2D NMR experiments on amine A in neutral solution exists in the bi-
the metal-free ligand as well as its gallium cyclic form, the pyrroline ring opens under
complex (DRECHSELet al., 1995b). The novel acidic conditions to give the a-amino acid re-
siderophore contains a benzene and a thiazol- sidue which no longer complexes ferrous ions.
idine ring as well as two thiazoline rings (Fig. The proferrorosamine producing Pseudo-
lob) and thus resembles the thiazoline-con- monas sp. strain G H (LMG 11358), has later
taining siderophores pyochelin, anguibactin, been reclassified as E. rhuptontici (VANDE
and desferri-ferrithiocine. The molecule con- WOESTYNEet al., 1991; DE Vos et al.,
tains five chiral centers, four of them inherent 1993).
in the connectivity and substituents of the
thiazoline rings. The compound forms stable
complexes with trivalent cations such as ferric Siderochelin A
iron and gallium. Iron binding constants have
not yet been determined. From pH depend- Siderochelin A (Fig. 12) isolated from fer-
ent UV and CD measurements it is clear, mentation broths of Nocardia sp. SC 11340
however, that the yersiniabactin-iron(II1) was identified as truns-3,4-dihydro-4-hydroxy-
complexes are stable at least between pH 4-
11. The complexes with Ga3+ and Fe3+ are
monomeric, 1:1 complexes. Yersiniabactin
constitutes the principal siderophore of Yer-
siniu enterocolitica sought for a long time. The
same siderophore has been isolated as alumi-
nium complex and has been termed yersinio-
phore (CHAMBERS et al., 1996).
Fig. 12. Siderochelin A.

5 - (3- hydroxy -2-pyridinyl) -4-methyl-2H -pyr- Ferrioxamines

role-Zcarboxamide (LIUet al., 1981).
Streptomyces pilosus produces a group of
closely related desferrioxamines (Al, AZ, B,
3.2 Hydroxamate Siderophores C, D1, Dz, E, F, G, and H) among which des-
ferrioxamine B or E generally predominate
Generally, hydroxamate siderophores in (BICKELet al., 1960 KELLER-SCHIERLEIN et
bacteria can be divided into two main groups: al., 1964). Although commonly isolated in
(1) exclusively containing hydroxamate their ferric forms as ferrioxamines, the stuc-
groups as iron-binding bidentates, (2) con- tures shown in Fig. 13 represent the iron-free
taining additional iron-binding ligands, e.g., forms. Desferrioxamine B is produced indus-
a-hydroxycarboxylate residues, like citrate or trially by large-scale fermentation of S.pilo-
catecholate groups as found in the pyover- sus strain A 21748 as the methane sulfonate
dins. Members of the first group comprise the salt (DesferaP). Desferale still is the only
ferrioxamines originally isolated from the drug approved for treatment of iron overload
gram-positive streptomycetes and are now diseases although several other useful iron
known to occur also in the gram-negative chelators are currently being developed
genus Pseudomonas and in Enterobacteria, (DIONISet al., 1991).
such as P. stutzeri (MEYERand ABDALLAH, Desferrioxamine B is a linear trihydrox-
1980) and Erwinia, Enterobacter, Hafnia amate. Acid hydrolysis yields acetic acid, suc-
(BERNERand WINKELMANN, 1990; REISS- cinic acid, and 1-amino-5-hydroxylaminopen-
BRODT et al., 1990). tane in a ratio of 1:2:3. After addition offer-
R'
I
NH
N-C N-C
\c=o
N-C
I I
Hb Hb I, HO 0
NH
I
\
N-C N-C N-C
I I I I
HO 0 HO 0
Fig. 13. Ferrioxamines (iron-free).

a Ferrioxamine B: R' =H, R2= CH3, n =5; ferrioxamine D1:R' = COCH3, R2 = CH3, n =5; ferrioxamine
'
G1: R = H, R2= CH2CH2COOH,n =5; ferrioxamine G2:R' =H, R2 = CH2CH2COOH,n =4; b ferriox-
amine E; ferrimycin A,: R 1=A,
R2 = CH3.
ric iron, a 1:1complex is formed with a stabil- ing number and size of the consituent di-
ity constant of Kf=1030.6 showing a red- amines, cyclic structure, and C-terminal acetyl
brown color due to the metal-to-ligand charge residues (FEISTNER,1995).
transfer band at h,,,=428 nm ( ~ = 2 8 0 0 Ferrioxamine B is the most intensely stud-
M cm -I). Since ferrioxamine B possesses a ied ferrioxamine. Today it is used in the me-
free amino group (pK=9.74), it migrates as a sylate form (DesferaP) for the treatment of
cation during electrophoresis in a weak acid iron storage diseases in man. The biological
buffer. Ferrioxamine D, (N-acetyl-ferriox- activity of ferrioxamine B has been studied
amine B) was isolated as a minor constituent using growth promotion tests with iron-auxo-
of the ferrioxamine fraction and can be pre- trophic bacteria, such as Microbacterium lacti-
pared synthetically by acetylation of ferriox- cum ATCC 8181, Arthrobacter flavescens JG9
amine B with acetic acid anhydride. The li- ATCC 29091, and A. terregens as reviewed by
gand of ferrioxamine E is identical with the WINKELMANN(1991a). Transport experi-
antibiotic “nocardamine” isolated from No- ments with iron-labeled ferrioxamines in
cardia. The crystal structure of ferrioxamine Streptomyces pilosus (ATCC 19797) de-
E was determined (VANDER HELMand POL- scribed by MULLER and RAYMOND(1984)
ING, 1976), and the coordination about the and MULLERet al. (1984) showed that fer-
iron center is cis. The ligand is a cyclic struc- rioxamine B has the highest ratio of iron up-
ture consisting of 3mol succinic acid and take ( K , =0.2 pM). Chromic and gallium
3 mol l-amino-5-hydroxylamino pentane. complexes of ferrioxamine B were trans-
Formal hydrolysis of one peptide bond in fer- ported at similar rates indicating that the in-
rioxamine E leads to ferrioxamine G which in tact complexes were taken up. Differences in
turn can be converted to ferrioxamine E by cis and trans isomers were not observed. In
treatment with dicyclohexyl carbodiimide. contrast, ferrioxamine transport studies using
Some of the isolated ferrioxamines contained in vivo Mossbauer spectroscopy and radioac-
a mixture of l-amino-5-hydroxylamino pen- tive labeling in Enterobacteria, such as Er-
tane and l-amino-4-hydroxylamino butane winia herbicola (now called Pantoea herbico-
(molar ratio 2:l) and were designated as fer- la), revealed that the ligand is not accumu-
rioxamine A (related to ferrioxamine B), and lated inside the cells (MATZANKEet al.,
ferrioxamine A*, (related to ferrioxamine D1) 1991).
and ferrioxamine DZ(related to ferrioxamine
E). Recent investigations in the authors’ labo-
ratory have shown that ferrioxamines are not
confined to the genus Streptomyces but are Bisucaberin and Alcaligin
also characteristic siderophores of several en-,
terobacterial genera, e.g., Erwinia, Pantoea, Bisucaberin (Fig. 14a), a dihydroxamate si-
Enterobacter, Ewingella, and Hafnia (BER- derophore isolated from low-iron cultures of
NER et al., 1988; BERNERand WINKELMANN,the marine bacterium Alteromonas haloplanc-
1990; REISSBRODTet al., 1990). Earlier re- tis is a cyclic dimer of succinyl-(N-hydroxy)-
ports have shown that Pseudomonas stutzeri cadaverine. It thus resembles ferrioxamine E
is able to produce desferri-ferrioxamine E, which is a cyclic molecule composed of succi-
previously named nocardamine (MEYERand nyl diaminopentane residues.
ABDALLAH,1980). Although the name pro- From the heterotrophic marine bacterium
ferrioxamine has been suggested for the li- Alcaligenes denitrificans a similar cyclic dihy-
gand instead of desferrioxamine, deferriox- droxamate siderophore named alcaligin (Fig.
amine, or deferoxamine (FEISTNERet al., 14b) was isolated (NISHIOet al., 1988). Alcal-
1993), the new terminology of pFOs has not igin is a cyclic molecule consisting of two
been accepted in the biological literature so N1-hydroxy-3(0H)-putrescine residues linked
far but seems to be of great value in distin- by two succinic acid residues. Alcaligin has
guishing the vast number of structurally dif- recently been found in the obligate pathogens
fering ligand derivatives by a simple subscript Bordetella pertussis and B. bronchiseptica
(e.g., desferrioxamine B = PFOsss~c)indicat- (MOOREet al., 1995; BRICKMAN et al., 1996).
Ferrimycins and Albomycins
The ferrimycins (Fig. 13f) represent ferri-

oxamine-type siderophores possessing anti-
biotically active residues. Ferrimycins (A,,
A2, B) and the compound A-22765 are ferri-
oxamine-type antibiotics isolated from Strep-
tomyces griseoflavus and other species. The
ferrimycins inhibit growth of gram-positive
bacteria, such as Staphylococcus aureus and
Bacillus subtilis. A revised structure of ferri-
mycin A was published by KELLER-SCHIER-
LEIN et al. (1984). Agar diffusion tests re-
vealed that the antibiotic activity of the ferri-
mycins is antagonized by the presence of fer-
rioxamines. This has been explained by the
existence of a transport antagonism in which
the ferrioxamines and ferrimycins compete
for the same target in the transport system
Fig. 14a. Bisucaberin; b alcaligin, R =OH; putre- (ZAHNERet al., 1977). It remains an open
bactin, R =H. question whether the antibiotically active
group is split off from the transport moiety or
whether ferrimycin acts as a whole after en-
A related cyclic dihydroxamate named putre- tering the cell.
bactin (Fig. 14c), containing putrescine in- Albomycins (al, &, E) (Fig. 15) have also
stead of hydroxy-putrescine, was isolated been isolated from Streptomyces strains, al-
from Shewanella putrefaciens (ALISONBUT- though the triornithine peptide is a character-
LER, personal communication). istic feature of the fungal siderophores, e.g.,
ferrichromes. The originally published albo-
mycin structure has been revised (BENZet al.,
Fig. 15. Albomycins.

Albomycin al: X = 0; albomycin &: X =NCONH2; albomycin E: X =NH.
1982) and was shown to consist of a linear

peptidonucleoside (Oml-Om2-Om3-Ser-nu-
cleoside). Albomycins are highly active
against gram-negative and gram-positive bac-
teria. Ferrimycins and albomycins use the
iron transport systems of ferrioxamine and
ferrichrome, respectively. Albomycin is taken
up in E. coli via the FhuA receptor protein
and the tonB gene product, as shown with
F u A and tonB mutants (HARTMANN et al., 3 0 L o\ A0
NH
I
OYNH.f
1979). Moreover, using 3H- and 35S-labeled
albomycin it was confirmed that, contrary to
the iron atom, the ligand is not accumulated
inside the cell. From the small portion of la-
beled sulfur detected within the cells it was
concluded that the nucleoside residue was OH
split off to exert its antibiotic activity. Al- Fig. 16. Alterobactin.
though inhibition of protein biosynthesis was
demonstrated, it was also suggested that the
inhibition of oxidative phosphorylation is the bacterium, Alteromonus luteovioluceu (REID
initial inhibitory effect of albomycins. Growth et al., 1993). It has been shown to possess an
of E. coli is inhibited at very low concentra- exceptionally high formation constant
tions (lo-' M). Because of the high frequen- (K=1049-1053). Alterobactin A is a cyclic si-
cy of resistant mutants, ferrimycins and albo- derophore containing a lactone ester bond,
mycins have not found any application in the while alterobactin B is the corresponding
therapy of bacterial infections so far. open-chain form. So far, only two sidero-
phores from Alteromonus have been isolated
- alterobactin and bisucaberin from Alfero-
3.3 Peptide Siderophores monus huloplunctis (TAKAHASHIet al.,
1987).
Alterobactin
Alterobacin, a peptide siderophore con- Ferrocins
taining the sequence cyc1.-(DHBA-Ser-Gly-
Arg-/?-OH-Asp-POH-Asp) (Fig. 16), was Ferrocins (Fig. 17) represent a new group
isolated from a marine Pseudomonds-like of iron-containing peptide antibiotics pro-
Fig. 17. Ferrocins.

Ferrocin A: R'=H,
R2=R3=R4=H,ferrocin B:
R'=OH, R2=R3=R4=H,ferrocin
C and D: R'=H, RZ,R3, R4=2-H,
1* CH3.
duced by Pseudomonas fluorescens YK-310 thine residues and the peptidic nature of the
(KATAYAMA et al., 1993). These siderophore ferrocins they resemble the albomycins, al-
antibiotics showed antibacterial activity though the latter contain a nucleoside resi-
against E. coli and P. aeruginosa and strong due.
therapeutic effects on P. aeruginosa. The
structure elucidation revealed that the ferro- Pseudobactins and Pyoverdms
cins are cyclic decapeptides containing three
hydroxamate residues .(TSUBOTANI et al., Strains of the Pseudomonas fluorescens
1993). Four different ferrocins (A, B, C, D) group (P. fluorescens, P. putida, P. aerugino-
have been isolated. Ferricrocin A is a cyclic sa, P. syringae, and related species) produce
peptide containing a fatty acid residue and a yellow-green fluorescent siderophores. Typ-
lactone ring between the C-terminal Gly and ical structures are the linear pseudobactin
the N-terminal Ser: (Z)-3-Decenoicacid-cyc1.- (Fig. 18) (TEINTZEet al., 1981) and pyover-
Ser-X-Gly-Val-Ser-X-Ala-Gly-X-Gly. X may din Pfl2, containing a cyclic substructure (Fig.
represent N-Acetyl-N-OH-Orn or N-propio- 19) (BUDZIKIEWICZ, 1993). Pseudobactins
nyl-N-OH-Om. Because of the three orni- and pyoverdins contain a common structural
Fig. 18. Pseudobactin. 0
Fig. 19. Pyoverdin Pfl12.

element, the chromophore which is a deriva- KIEWICZ et al. (1992) and LINGET et al.
tive of 2,3-diamino-6,7-dihydroxyquinoline, (1992) have generally confirmed the composi-
tion of ferribactin. Both compounds, ferribac-
possessing an additional 1-carboxyl-pyrimidi-
no ring (S configuration) and an amino grouptin and pyoverdin, isolated from the same
acylated with different dicarboxylic acid resi-
strain have been shown to differ in the chro-
dues: succinic acid, succinamide, 2-0x0-glu-mophore part but are identical in their pep-
taric acid, glutamic acid, malic acid (amide).
tide sequence (BUDZIKIEWICZ et al., 1992).
The chromophore is connected via the C-1 Thus, the composition of ferribactin isolated
carboxyl group predominantly (for excep- from Pseudomonus uptutu was similar to
tions, see BUDZIKIEWICZ, pyoverdin Pap (P. uptufa)but lacked the fluo-
1993) to the N-ter-
minus of a peptide chain of varying chain rescent chromophore and contained Glu and
length (6-12 amino acids), half of which pos-
Tyr/Dbu instead. Ferribactin is a much weak-
sess D configuration. In addition to the cate-
er iron-binding compound than the pyover-
dins as the catechol group provided by the
cholate bidentate of the quinoline residue two
further amino acids within the peptide chainchromophore is absent. However, the phenol
function as iron bidentates among which N5- group of Tyr and the carboxylate group of
acyl-N5-hydroxy ornithine is present in all Glu in ferribactin may possibly substitute the
compounds studied so far. The N5-acyl group chromophore ligand binding sites.
of ornithine may be formyl, acetyl, or p-hy- Because of the diversity of the pyoverdin
droxybutyryl or may even be connected to itsstructures produced by the various Pseudo-
own carboxyl group (cyclic OH-ornithine). rnonus species a pronounced specificity dur-
The second iron-binding amino acid is gener-ing uptake of ferric pyoverdins can be ex-
ally threo-p-hydroxy aspartic acid. However,pected and has indeed been observed by
some of the pyoverdins lack p-hydroxy aspar-HOHNADELand MEYER(1988) and MEYER
tic acid and possess two NS-acyl-N5-hydroxy (1992).
ornithine residues instead. In addition to the The structural gene for the ferric pseudo-
cyclic NS-OH-Orn the peptide chain may pos- bactin receptor (PupA) in Pseudomonus puti-
du WCS358 has been cloned and sequenced
sess larger internal cyclic substructures where
&-aminogroups (Lys) may be connected via (MARUGGet al., 1989; BITTER et al., 1991).
amide bonds to distant C-terminal carboxyl An inducible ferric pseudobactin receptor
groups or where Ser and Thr form additional (PupB) of P. putidu WCS358 has also been
tetrahydropyrimidine rings. Although inter- identified and characterized (KOSTERet al.,
nal rings have been detected in a variety of1993).
pyoverdins, pseudobactin is a linear molecule Different types of pyoverdin-defective
as confirmed by X-ray analysis (TEINTZEet (pvd) mutants have been isolated and charac-
al., 1981). A list of the currently known pyo-
terized (VISCAet al., 1992). The pvd-l mutant
verdin structures is shown in Fig. 20. Afteris an ~-N’-hydroxy ornithine auxotroph un-
chelation with ferric iron the complex adopts
able to oxygenate L-ornithine and requiring
a red-brown color. At pH 7 the Fe complexes ~-N’-hydroxy ornithine for pyoverdin pro-
show UV-VIS absorption maxima at 400 (log duction. Other types of mutants appear to be
&=4.2), 320 and 280nm and additional blocked in further steps of the biosynthetic
charge transfer bands at 470 and 550 nm. For-
pathway leading to pyoverdin, the acylation
mation constants have been reported to be inof ~-N~-hydroxy ornithine (pvd-2)and chro-
the range of 1024-1026. mophore biosynthesis (pvd-3).The oxygenase
gene pvd-A (previously named pvd-2) from
P. ueruginosu has been cloned (VISCAet al.,
Fembactin 1994). From a gene bank using the broad-host
range cosmid pLAFR3 mobilized in a pvdA-
During the isolation of iron-binding pig- defective mutant a trans-complementing cos-
ments from P. fluorescens a non-fluorescent mid pPV4 was obtained which restored pyo-
precursor named ferribactin was isolated verdine synthesis and oxygenase activity in
(MAURERet al., 1968). The studies of BUDZI- the pvdA mutant.
Bacteria Structures of pyoverdins and related compounds

Pseudomoms putidu ATCC 12633 Chr-L-Asp-L-Lys-D-OHsp-(D,L)Ser-L-Thr-D-Ala-D-Glu-(D,L)Ser-L-cOHOrn
Pseudomonas putidu CFBP 246 1 Chr-L-Asp-(D,L)-Lys-D-OHAsp-D-Ser-L-Thr-D-AIa-(D,L)-Lys-L-Thr-L-cOHOrn
Pseudomonas chlororaphis ATCC 9446 Chr-D-Ser-L-Lys--Gly-(D,L)-FoOHOrn-c~-Lys-~,L)-FoOHOrn-L-Ser]
PseudomonasjluorescensATCC I3525 Chr-D-Ser-L-Lys-Gly-(D,L)-FoOHOrn-c~-Lys-(D,L~FoOHOrn-L-Ser]
PseudomonasjluorescensSB 83 Chr-D-Ala-L-Lys-L-Thr-D-Scr-L-AcOHOrn-L-cOHOrn
PseudomonasjluorescenrATCC I 7400 Chr-D-Ala-D-Lys-Gly-Gly-L-OHAsp-D-GlnCTHPMD-L-Ser-D-Ala-L-cOHOrn
(Demange et al., 1990a)
PseudomonasjluorescensCCM 2798 Chr-SerCTHPMD-Gly-L-Ser-D-OHAsp-L-Ala-Gly-D-Ala~ly-L~OHOrn
(Demange et al., 1990b)
Pseudomonasaeruginosa ATCC I 5692 Chr-D-Ser-L-Arg-DSer-L-FoOHOrn-c~-Lys-L-FoOHOrn-L-~r-L-Thr]
(Briskot et al., 1989)
Pseudomoms rolnasii NCPPB 2 192 Chr-D-Ser-L-Lys-L-Ser-DSer-L-Thr-D-Ser-L-AcOHOrn-L-Thr-D-Ser-D-cOHOrn
(Demange et at., 1990b)
Azorobacter vinelandii CCM 289 Chr A-L-Asp-DSer-L-Hse-Gly-D-OHAspL-Ser-P

(azotobactin D) (Demange et al., 1988)
PseudomonasjluorescensATCC I 3525 L-Glu-D-TyrCTHPMD-DSer-L-Lys-Gly~D,L)-FoOHOrn~[L-Lys-(D,L)-F~HOrn-L-Ser]
(desferribactin)(Linget et al., 1992)
Azomonas macrocyrogenes ATCC I2334 Chr-L-Hse-D-AcOHOrn-D-Ser-L-AcOHOrn-D-Hse-L-CTHPMD
(azovcrdin) (Bcrnardini et al., 1996)
For additional structures see:Abdallah 1991 and Budzikiewicz 1993.
Chr = Chromophore, ChrA = chromophoreof azotobactin, OHOm and OHAsp = N'-hydroxyornithine and Bhydroxyaspartic acid,
c = cyclic, cOHOrn = cyclo-N*-hydroxyornithine,Fo and Ac = formyl and acetyl present at the N'-nitrogen atom of hydroxyornithines,
SelCTHPMD, TyrCTHPMD and GlnCTHPMD represent the amino acids derived from tetrahydropyrimidine resulting from ring
formation of L-2,4-diaminobutyric acid with Ser, Gln or Tyr:
R = succinic acid (amide)

malic acid (amide)
2-ketoglutaric acid (amide)
glutamic acid
Chromophore
OH
-NH
k3-
SerCTHPMD
0
Fig. 20. Structures of pseudobactins and pyoverdins.

TyrCTHPMD GlnCTHPMD
Azotobactin nithine acylated with 3-hydroxy butyric acid

(BUDZIKIEWICZ et al., 1992).
The peptide siderophore azotobactin (Fig. Using 55Fe-labeled compounds it was
21) seems to be the most prominent sidero- shown that azotochelin and azotobactin do
phore of the Azotobacteriaceae showing a function as siderophores in A. vinelandii
striking similarity to the pyoverdins. The (KNOSPet al., 1984). However, siderophore-
main structural difference between the azoto- mediated uptake of iron was not observed un-
bactins and pyoverdins is that azotobactins til substantial non-specific binding of iron to
form an extra imidazolone ring connecting the cells was eliminated. Transport of azoto-
the amino group at C-5 to the chromophore bactin and azotochelin was also decreased by
N-4 by an urea unit, whereas the pyoverdins high concentrations of cations, such as Na +,
possess an amidically bound dicarboxylic acid K + , Li +,or Mg2+, due to some interference
at that position (CORBINet al., 1970). As this with the siderophore transport system.
is a minor structural difference, some authors VISWANATHA and his group (MENHARDT
have included the azotobactins in the pyover- et al., 1991) have recently discussed the heter-
din group (MENHARDT et al., 1991). ogeneity of the pyoverdins and azotobactins
observed during growth under iron limitation.
From their pulse labeling experiments it was
OH inferred that an amino acid deficiency or a
lack of stringency may be responsible for the
observed heterogeneity of the produced azo-
OH tobactins and pyoverdins.
0
Ornibactins
i
Fig. 21. Chromophore of azotobactin. Ornibactins (Fig. 22) represent modified te-
trapeptide siderophores containing N-termi-
nal3-hydroxy-acyl residues of different chain
FUKASAWA et al. (1972) described the first lengths (C4, C6, (3).The peptide contains
structure of the yellow-green fluorescent pep- two t-Orn (N5-OH, N5-acyl), one Asp (p-
tide, azotobactin 0, produced by Azotobacter OH), one L-Ser and a C-terminal lP-diamino
vinelandii (strain 0) which contained only two butane residue (STEPHANet al., 1993a, b).
bidentate chelating groups. Two further struc- Thus the ornibactins resemble the pyoverdins
tures have been elucidated by now: azotobac- but they possess an acyl residue instead of a
tin D (DEMANGE et al., 1988) and azotobac- chromophore and are much smaller peptides.
tin from A. vinelandii DSM 87 (ATCC They have been isolated from Pseudomonas
12837),the latter containing an N-hydroxy or- cepacia-like strains. The “cepacia group”
0
I
Fig. 22. Ornibactins.

Ornibactin C4
(R =methyl); ornibac-
tin C6 (R=propyl);
ornibactin C8
(R = pentyl).
(rRNA homology group 11) has been recently in ethanol was added to convert the mycobac-
transferred to the genus Burkholderia (YA- tins completely to the ferric form. An equal
BUUCHI et al., 1992). Ornibactin production volume of chloroform was added, followed by
has been found in clinical isolates as well as in water until two layers were formed. The chlo-
nitrogen-fixing bacterial isolates from the rhi- roform layer was washed three times with wa-
zosphere of rice plants. Moreover, ornibactin- ter to remove excess iron; it was then dried
mediated iron uptake was observed in all with MgS04, evaporated to dryness, and the
Burkholderia strains but was absent in Pseu- residue was dissolved in methanol. As
domonas aeruginosa, P. fluorescens, and P. pointed out by SNOW(1970), the characteris-
stutzeri (MEYERet al., 1995). tic UV absorption spectra of the iron-free
mycobactins arise from two parts of the
molecule, the 2-(o-hydroxyphenyl)-oxazoline
3.4 Mycobactins and Related structure and the acyl hydroxamic acid. De-
pending on the substitution of the benzene
Siderophores ring the following A,, values in methanol are
given: (no methyl group) 243, 249, and
Mycobacteria produce a series of lipid-sol- 304 nm with inflection at 258 nm; (with me-
uble, iron-binding compounds termed myco- thyl group at position 6) 250 and 311 nm with
bactins (Fig. 23a) which have been described a shoulder at 254nm and an inflection at
in previous reviews by SNOW (1970) and 265 nm.
RATLEDGE(1987). Among the different my- Growth of Mycobacterium paratuberculosis
cobactins (A, F, H, M, N, P, R, S, T), myco- (formerly named M. johnei) was stimulated in
bactin P, isolated from Mycobacterium phlei, the presence of 5-15 ng mycobactin P (FRAN-
and mycobactin T, isolated from M. tubercu- CIS et al., 1953). However, other mycobactins
losis, were analyzed in more detail (SNOW, also gave significant effects so that there
1965 ). After splitting the ester link, two prod- seems to be little specificity. M. paratubercu-
ucts known as mycobactic acid and cobactin losis and some Mycobacteria related to Myco-
were obtained. Regarding the P type mycobacterium avium have lost the ability to syn-
bactin, chemical analysis of the mycobactic thesize mycobactins and are, therefore, stimu-
acid unit yields (1) an aromatic acid, either lated by the addition of exogenous mycobac-
salicylic acid or 6-methyl salicylic acid, the lat- tins. Nocobactins have been isolated from
ter undergoing further degradation to m-cre- Nocardia species as compounds equivalent in
sol and carbon dioxide; (2) a P-hydroxy ami- function to the mycobactins and similar in
no acid, either serine or threonine; (3) a mix- structure (RATLEDGEand PATEL,1976).
ture of homologous long-chain fatty acids, Mycobactins are extremely lipophilic com-
usually with a double bond adjacent to the pounds and are generally not excreted into
carboxyl group; (4) N6-hydroxy lysine. The the medium. Therefore, RATLEDGE(1987)
cobactin unit yields a hydroxy acid, either 3- suggested that mycobactins function as an
hydroxy butyric acid or 3-hydroxy-2-methyl iron shuttle within the lipid boundary of the
pentanoic acid, and a second molecule of N6- cell surface. Although salicylic acid is ex-
hydroxyl lysine. creted into the medium, its contribution to
The mycobactins are associated with the iron solubilization seems to be insignificant.
mycobacterial cell walls and have to be ex- Instead, water-soluble exochelins are thought
tracted by organic solvents. In order to avoid to function as iron scavengers and to trans-
extraction of undesired fatty material, the port iron to the mycobactins (STEPHENSON
most convenient method is to suspend the and RATLEDGE,1979). The mechanism pro-
cells in cold ethanol. HALL and RATLEDGE posed for iron uptake into Mycobacterium
(1982) have reported a simple method for the smegmatis involves the mediation of myco-
production and isolation of mycobactins: My- bactin in shuttling iron across the boundary
cobacteria were grown on solid media and layers of the cell (RATLEGEand MARSHALL,
scraped from the agar after growth. The bac- 1972). The loading of iron into mycobactin is
teria were then held for 24 h in ethanol. FeC13 presumed to occur via exochelins which are
a
HO 3
a 3
\
N 0 m q N
0 , OH
OH 0 0
Fig. 23a-c. General structure of a mycobactins (for side chains and nomenclature see SNOW,1970);
b exochelin MS: c exochelin MN.
able to solubilize iron from the environment containing three N'-hydroxy-ornithines (Fig.
(MACHAM et al., 1975; STEPHENSONand Z3b) (SHARMANet al., 1995a). Another novel
RATLEDGE, 1979). Structure elucidation of water-soluble iron-binding linear hexapeptide
exochelins from M.smegmatis has shown the named exochelin MN has recently been iso-
presence of a modified linear pentapeptide lated from M. neoaururn (SHARMANet al.,
1995b) containing P-hydroxyhistidine, 2 @ has previously been reported to produce a

alanines, N"-methyl-'N-hydroxyornithine, or- novel citrate-based siderophore, acinetoferrin
nithine, and cyclic 'N-hydroxyornithine (Fig. (OKUJOet al., 1994b).
23c). In addition, exochelins showing struc-
tural similarities with mycobactins have been
isolated from M. tuberculosis (GOBINet al., Frankobactin
1995, LANEet al., 1995; RATLEDGEand Ew-
ING, 1996). These exochelins contain dicarb- The siderophore of Frankia (strain 52065),
oxylic acid methyl esters instead of the long named francobactin, has recently been iso-
chain fatty acids occurring in the mycobac- lated and shown to be a novel hydroxamate-
tins. containing peptide siderophore (Fig. 25). The
following components have been tentatively
identified (MW 731): a phenyl oxazoline ring,
Acinetobactin serine, glycine, ornithine, and P-HO-aspartate
(BOYERand ARONSON,1994; ARONSON and
From low-iron cultures of Acinefobacter BOYER,1994). The actinomycete Frankia is a
baumannii ATCC19606 another compound gram-positive, aerobic, filamentous, sporulat-
with both catecholate and hydroxamate func- ing, and nitrogen-fixing bacterium that exists
tional groups was isolated containing 2,3-di- as a free-living saprophyte or as a symbiont in
hydroxy benzoic acid and threonine com- the roots of the actinorhizal plants including
bined to an oxazoline ring and o-N-hydroxy alder trees (e.g., Afnus glufinosa), Ceanofhus
histamine (YAMAMOTOet al., 1994a). The americanus, Myrica gale, and several species
structure was elucidated by chemical degrada- of Casuarina. Unlike the Rhizobium strains,
tion, fast atom bombardment mass spectro- Frankia strains are also able to fix nitrogen in
metry, and 'H and 13C NMR spectroscopy. aerobic culture which makes Frankia an inter-
Acinetobactin (Fig. 24) was identified in 4 of esting organism to study siderophore produc-
tion under symbiotic and non-symbiotic con-
ditions.
0
7
NH
Maduraferrin
An Acfinomadura madurae strain (Actino-

mycetes) produces the novel siderophore
Fig. 24. Acinetobactin. maduraferrin (KELLER-SCHIERLEIN et al.,
1988). Maduraferrin (Fig. 26) is the iron com-
plex of an oligopeptide siderophore com-
12 clinical isolates of A . baumannii. Acineto- posed of salicylic acid, p-alanine, glycine, L-
bactin is structurally related to anguibactin, a serine, N5-hydroxy-N2-methyl-~-ornithine,
siderophore isolated earlier from Vibrio an- and ~-hexahydropyridazine-3-carboxylic acid.
guiffarumwhich has a thiazoline ring instead Although most Acfinomadura strains pro-
of an oxazoline ring (JALALet al., 1989). An- duced various ferrioxamines, some strains,
other Acinefobacter species, A . huemolyticus, like strain DSM 43067, were found to pro-
Serine
Glycine
+
+ ornithine
HYhXYaSpertate
Fig. 25. Frankobactin. 0
R'
i -C n -R'
OH 0
+
N H y ( C H d m-N
Fig. 26. Maduraferrin.
duce desferri-maduraferrin. Because of the

low growth in chemically defined media, mad-
uraferrin production was performed in a com-
plex medium (soy, yeast extract) with the ad- A H
dition of AlC13 which mimics iron deficiency.
Transport of iron mediated by maduraferrin
was demonstrated in the producing strains
but was absent in E. coli or Staphylococcus H
aureus, suggesting a special uptake route in
Actinomadura strains. B
3.5 Citrate Hydroxamate

Siderophores
C
Schizokinen
Schizokinen (Fig. 27a), originally described Fig. 27. Citrate hydroxamates.

as a factor that reduced the division lag of Ba- Schizokinen: R'=R2=H, R3=R4=CH3,n=2; rhi-
zobactin 1021: R' = R2=H, R3= A, R4= CH3,
cillus megaterium, B. cereus, and B. subtilis, n=2; acinetoferrin: R'=R2=H, R3=R4=B,n=2;
was isolated from a culture of B. megaterium arthrobactin: R' = R2= H, R3= R4= CH3, n =4;
ATCC 19213 (BYERS et al., 1967). The aerobactin: R ' =R2= COOH, R3=R4= CH3, n =4;
growth factor activity was confirmed with a nannochelin A: R ' =R2= COOCH3, R3= R4= C,
schizokinen-auxotroph of B. megaterium. The n =4; nannochelin B: R' = COOCH3, R2= COOH,
structure of schizokinen was elucidated by R3= R4=C, n=4; nannochelin c
MULLISet al. (1971). Hydrolysis of schizokin- R1=R2=COOH,R3=R4=C,n=4.
en yielded l-amin0-3-(N-hydroxyamino)pro-
pane, and oxidation with periodate yielded
acetate as the acyl moiety of the hydroxamic (Amax =390 nm) was pH independent between
acid bonds. The central backbone is com- pH 5 and 9.5, but shifted to red below pH 5.
posed of a citric acid residue. Titration and The cultures were grown in a chemically
CPK models confirmed that both the central defined low-iron medium at 35 "C with strong
hydroxyl and carboxyl group can bind to the aeration for one week (3% inoculum). After
metal ion to give an octahedral six-oxygen removing the cells, the supernatant was con-
coordination sphere. At neutral pH the ferric centrated to 1/10 of the original volume, acid-
complex is an anion indicating the presence ified with conc. HCI to pH 2, extracted with
of ionized citrate hydroxyl. The absorption chloroform-phenol (1 :l), and transferred
with ether to the aqueous phase. This was en. The results indicated an initial tempera-
then concentrated by evaporation and ap- ture-independent binding of ferric schizokin-
plied to a column of the acetate form of Dow- en, followed by a temperature-dependent (ac-
ex AG-2-XlO. The column was washed with tive) transport of the chelate into the cell, and
distilled water and schizokinen was eluted an enzyme-catalyzed separation of iron from
with 0.2M ammonium chloride and further the chelate (ARCENEAUX et al., 1973). More-
purified on Bio-Gel P-2. TLC on silica gel over, a mutant of B. megaterium S K ll unable
with methanol (Rf=0.60) using tetrazolium to synthesize schizokinen showed recognition
chloride or FeC13 in 0.05 N HCl as sprays was capacity for the chemical structure of schizo-
performed to confirm purity. Schizokinen was kinen. Ferrioxamine B (Desferalm) was taken
also found in Anabaena strains (LAMMERS up at lower rates and aerobactin did not sup-
and SANDERS-LOHR, 1982) and may be a si- port uptake of iron (HAYDONet al., 1973). A
derophore in other cyanobacteria. recent investigation has shown that B. subtilis
A derivative of schizokinen, schizokinen A utilizes three types of hydroxamate sidero-
(Fig. 28), has recently been isolated from al- phores: ferrichromes, ferrioxamines, and schi-
calophilic Bacillus strains containing a cyclic zokinen (SCHNEIDERand HANTKE,1993).
side chain due to the condensation of the carb- Moreover, the transport system showed sig-
oxyl at the quarternary C atom with one of nificant homology to the binding protein-de-
the amide NH groups (STEPHANet al., un- pendent components of E. coli. As a peri-
published results). Schizokinen A has been plasm is missing in gram-positive bacteria, the
synthesized by LEE and MILLER(1983). A FhuD-corresponding protein is anchored as a
similar compound was isolated earlier from lipoprotein in the cytoplasmic membrane.
B. megaterium. In the original publication,
however, this product was incorrectly de-
scribed to have a six-membered ring (MULLIS Rhizobactin 1021
et al., 1971). It is still unclear whether or not
cyclization to schizokinen A is an artifact aris- Rhizobactin 1021 (Fig. 27b) is a novel
ing during the isolation procedure of schizo- asymmetric citrate-based dihydroxamate iso-
kinen or if it is a real natural product. The oc- lated from Rhizobium meliloti strain 1021.
currence of imido forms in citrate containing While acinetoferrin contains two identical hy-
siderophores has also been reported for sta- droxamic acid acyl groups as (E)-Zoctenoic
phyloferrin A (KONETSCHNY-RAPP et al., acids, rhizobactin 1021 contains two different
1990) and rhizoferrin (DRECHSELet al., acyl residues, (E)-2-octenoic acid and (E)-2-
1992). Rhizoferrin, e.g., tends to give both, decenoic acid (PERSMARK et al., 1993). Ferric
imidorhizoferrin and bis-imidorhizoferrin rhizobactin 1021 predominantly exists in solu-
(DRECHSELet al., 1992). tion in the lambda configuration in an appar-
BYERSand co-workers were the first to in- ent equilibrium of a monomeric and a dimeric
vestigate the transport of iron in B. megater- species.
ium, using double-labeled 59Fe-3H-schizokin-
Acinetoferrin
Acinetoferrin (Fig. 27c) has been isolated

from Acinetobacter haemolyticus and shown
to belong to the citrate-based dihydroxamates
(OKUJOet al., 1994b). Structure elucidation
revealed that acinetoferrin is a structural ana-
log of schizokinen as it contains a citric acid
\
c-0 backbone linked amidically to two l-amino-
I 3-(N-hydroxy propane) residues which in turn
are condensed by hydroxamic acid bonds to
Fig. 28. Schizokinen A. two (E)-2-octenoic acid moieties. Fast atom
bombardment mass spectroscopy revealed a 1969). Aerobactin consists of a conjugate of

quasi molecular ion peak (MH +) at mlz =585 N6-acetyl-N6-hydroxy lysine and citric acid.
corresponding to a molecular mass of 584 After complexing with ferric iron the internal
amu. Maximal production of acinetoferrin oc- citrate carboxyl and the a-carboxyls of the
curred when A . huemolyticus ATCC 17906 amino acid residues remain free resulting in a
was grown in a chemically defined medium ferric complex with three negative charges at
containing approximately 0.1 pM iron. Ab- neutral pH. The synthesis of aerobactin was
sorption spectra of acinetoferrin showed an described by MAURERand MILLER(1982).
absorption maximum at 210 nm (2.9.104 Coordination chemistry, stability constants, li-
M-’ cm-’) with a shoulder at 250 nm, indica- gand protonation, and metal ligand equilibria
tive of unsaturated bonds. After addition of were determined by HARRISet al. (1979).
iron a red-brown color develops that is char- The ferric aerobactin complex shows an over-
all formation constant of log Kf =22.93 and an
acteristic for a ligand-to-metal charge transfer
band (h,,,=486 nm; ~ =9 .8 -1 0 *M-’ cm-’1. absorption of A,,,=398 nm (e=2170 M-’
cm-’). Based on the pM value (stability un-
der specified concentration and pH) aerobac-
Arthrobactin tin can effectively compete with transferrin.
The comparatively low redox potential
Arthrobactin (Fig. 27d), previously known (-0.336 V) allows reduction via physiological
as the “terregens factor”, is produced by reductants and reuse of the ligand, which
strains of Arthrobacter and supported the seems to be an advantage for the producing
growth of the iron-auxotrophs Arthrobacter organism. The CD spectrum revealed that
terregens and A . jlavescens JG9 ATCC 29091. ferric aerobactin predominantly exists as the
The structure of arthrobactin was elucidated lambda optical isomer in aqueous solution.
by LINKEet al. (1972). It contained two l-ami- Aerobactin can be isolated from strains of
no-5-(N-acetyl-N-hydroxy)aminopentane re- Enterobacter aerogenes using chemically de-
sidues linked symmetrically to citric acid by fined media without added iron as described
peptide bonds. Arthrobactin was isolated by GIBSON and MAGRATH(1969). After
from low-iron cultures of Arthrobacter pm- inoculation and incubation for 20 h at 37°C
cens (ATCC 13346) in an asparagine-salts with stirring and aeration, the cells are sepa-
medium. After inoculation with a spore sus- rated and the supernatant is passed through a
pension the culture was incubated for 24 h at Dowex-1 (Cl -) column. All the material giv-
27°C. Then FeC13 was added, the culture fil- ing a red color with FeC13 is held on the co-
trate was saturated with ammonium sulfate, lumn. The column is then eluted with an am-
and extracted with benzyl alcohol. Diethyl- monium chloride gradient (0.4-1 M). 2,3-Di-
ether was added to transfer the ferric arthro- hydroxy benzoate derivatives are eluted first,
bactin complex into the aqueous phase. The followed by aerobactin. The fraction contain-
crude material was purified by countercurrent ing aerobactin is then evaporated to dryness,
extraction and by chromatography and gel fil- dissolved in water, and further purified on a
tration. Arthrobactin and schizokinen were Dowex 50W-X8 ( H + ) (200-400 mesh) col-
synthesized chemically (LEE and MILLER, umn with water. FeC13-positive fractions are
1983). collected, freeze-dried, and dried over P205
for several days.
Aerobactin is also produced by Enterobact-
Aerobactin er cloacae and Shigella flexneri (PAYNE,
1980), Erwinia carotovora (ISHIMARUand
Examination of supernatants from cultures LOPER,1992), and by Escherichia coli strains
of Aerobacter aerogenes 62-1 (now called En- containing the plasmid ColV (BRAUN,1981).
terobacter aerogenes) and related strains It was shown that aerobactin uptake requires
showed the presence of a hydroxamic acid for the expression of an outer membrane recep-
which the trivial name, aerobactin (Fig. 27e), tor (Iut A, 74 kDa) which serves as a common
was suggested (GIBSON and MAGRATH, binding site for aerobactin and the bacterio-
cin cloacin (VAN TIEL-MENKVELD et al.,

1982; BINDEREIF et al., 1982). The genes cod-
ing for aerobactin synthesis were analyzed by
GROSSet al. (1985) and DELORENZOet al.
(1986). Aerobactin production also seems to
be widespread in other bacterial genera. Be-
sides in enterobacteria, it has been detected OH NH
in cultures of a halophilic pseudomonad
(BUYERet al., 1991) and in a culture of a
gram-positive DSM 4640 strain (WINKEL- u
COOH o
MANN,unpublished data). Fig. 29. Rhizobactin DM4.
Nannochelins
amine bridge between a (non-amidically
Citrate hydroxamate siderophores named linked) alanine and lysine residue, the latter
nannochelins have been isolated from low- being N6-acylated by malic acid. Thus, rhizo-
iron cultures of Nannocystis excedens (Myxo- bactin DM4 is regarded as a complexone type
bacteria) (KUNZEet al., 1992) (Fig. 27f). Nan- or carboxylate type siderophore, while rhizo-
nochelins resemble aerobactin, but contain bactin 1021 is a citrate-based dihydroxamate.
two N-E-cinnamoyl residues instead. While It was anticipated that the iron complex of
nannochelin C contains two non-esterified ly- rhizobactin DM4 is hexacoordinated involv-
syl carboxyl groups, nannochelin B has one ing the two ethylene diamine nitrogens, the
and nannochelin A two methyl-esterified carb- two amino acid carboxylates, and the a-hy-
oxyl groups. droxy acid function of malic acid (NEILANDS,
1993). There is structural similarity of rhizo-
bactin DM4 and the mugineic acids isolated
as phytosiderophores from grasses having a
diamine backbone with (a-hydroxy)carboxy-
3.6 Carboxylate Siderophores late side functions. Mugineic acids may be
also regarded as complexone-type sidero-
Because of their high stability constants phores (NOMOTOet al., 1987).
and their pronounced colored ferric com-
plexes, catecholate and hydroxamate sidero-
phores were the first microbial iron chelates Staphyloferrin A
detected. However, with novel chemical and
biological assay methods being available, e.g., Staphyloferrin A (Fig. 30) isolated from
the chrome azurole S test (CAS test), addi- staphylococci is a carboxylate type sidero-
tional carboxylate-type siderophores have phore consisting of two citric acid residues
been detected (SCHWYNand NEILANDS, linked amidically to D-ornithine (KONETSCH-
1987). NY-RAPPet al., 1990; MEIWESet al., 1990).
Staphylococcus hyicus DSM 20459 was used
to produce staphyloferrin A in larger
Rhizobactin DM4 amounts. In addition, three dehydration
products were obtained, two mono forms and
Rhizobactin DM4 (Fig. 29) isolated from one diimido form, as a result of intramolecu-
Rhizobium meliloti DM4 (SMITHet al., 1985) lar condensation of the central carboxyl
should not be confused with rhizobactin 1021 groups with the amide NH groups. Interest-
isolated from R. meliloti 1021 (PERSMARK et ingly feeding with D- and L-ornithine led to
al:, 1993), although both have been isolated incorporation and synthesis of D-ornithine-
from isolates of the same species. Rhizobactin containing staphyloferrin A. Transport ex-
DM4 contains a characteristic ethylene di- periments with 55Fe-labeled compounds
COOH taric acid residue can easily form a five-mem-

a I
bered ring by nucleophilic addition of the am-
ide nitrogen to the carbonyl carbon. NMR ex-
periments showed NOE contacts confirming
the existence of a cyclization product as
shown in the structural formula of staphylo-
ferrin B. Whether this shift of equilibrium
from the open to the cyclized form is caused
by sample preparation or measurement con-
ditions is still unclear.
Vibrioferrin
A structurally novel siderophore was iso-

lated by successive column and TLC from Vi-
brio parahaemolyticus (YAMAMOTOet al.,
1992). This organism is known to occur in ma-
rine and estuarine environments and causes
diarrhea associated with seafood consump-
tion. Vibrioferrin was negative in the chemi-
cal assays of ARNOWand CSAKY(for descrip-
C tion, see NEILANDSand NAKAMURA, 1991)
excluding the presence of catecholates and
hydroxamates. The constituents of the sidero-
phore found after hydrolysis were L-alanine,
ethanolamine, citric acid, and 2-oxoglutaric
acid. Fast atom bombardment mass spectro-
metry revealed a mass of 446 amu. A com-
plete structure has recently been reported
(YAMAMOTOet al., 1994b). 2-oxoglutaric
Fig. 3Oa. Staphyloferrin A, b staphyloferrin B; c vi- acid is amidically linked to L-alanine and, like
brioferrin. in staphyloferrin B, exists in equilibrium with
a five-membered hydroxylactam (Fig. 3012).
showed that ferric staphyloferrin A acts as a

siderophore in staphylococci. Cepabactin
Cepabactin (Fig. 31) is a natural bacterial

Staphyloferrin B pyridinone siderophore (l-hydroxy-5-meth-
oxy-6-methyl-2(1H)-pyridinone) isolated
Staphyloferrin B has also been isolated from Pseudomonas cepacia ATCC 25416
from Staphylococci, however, the structural (MEYERet al., 1989). Cepabactin can be re-
components differ completely from staphylo-
ferrin A (DRECHSELet al., 1993a) in that cit-
ric acid is linked by a 2,3-diamino propionic
acid residue at one side and by a diamino
ethane and a 2-0x0 glutaric acid residue at the
other side (Fig. 30b). The molecular mass was 0
determined by pneumatically assisted electro
OH
spray from aqueous solution and indicates hy-
dration of the 2-0x0 function. The 2-oxoglu- Fig. 31. Cepabactin.
garded as a cyclic hydroxamate or as a keto fungal siderophore (see Sect. 4.3, uptake ki-
hydroxy bidentate. netics and characterization of the genes in-
volved in Morganella morganii have recently
been described (CARRANO et al., 1996 KUHN
3.7 Keto Hydroxy Bidentates et al., 1996).
There seems to be a wide range of com-

pounds in nature possessing keto hydroxy bi-
dentate ligands, most of which have not been
shown to be involved in iron transport. How-
ever, some selected keto hydroxy bidentate li-
4 Fungal Siderophores
gands proved to be very effective in iron
transport, as demonstrated recently (THIE- 4.1 Ferrichromes
KEN and WINKELMANN, 1993) using bacteria
of the group Proteeae (Proteus, Providencia, Ferrichrome (Fig. 32), first isolated in 1952
Morganella) all of which are able to produce from Ustilago sphaerogena (NEILANDS,
a-keto acids as a result of oxidative deamina- 1952), was found later in other Ustilago spe-
tion of amino acids (DRECHSEL et al., 1993b). cies (DEML,1985) as U. maydis (BUDDEand
Deamination to 2-0x0 acids is a characteristic LEONG, 1989), in Neovossiu indica (Tille-
trait of the Proteeae and is only of minor im- tiales) (DEML et al., 1984), Lipomycetaceae
portance in other enterobacterial genera. Be- (VANder WALTet al., 1990) and the derma-
cause of the high production rates in Proteeae tophyte Trichophyton mentugrophytes (MOR
and their activity in iron nutrition, a-keto et al., 1992). Besides ferrichrome, Ustilago
acids have been assigned a siderophore func- species also produce ferrichrome A, the struc-
tion in Proteeae and in other genera (DRECH- ture of which is closely related to ferrichrome.
SEL et al., 1993b; REISSBRODT et al., 1994). Crystal structures of ferrichrome A and fer-
Interestingly, strains of Proteus mirubilis were richrome were determined by ZALKIN et al.
found to be unable to transport iron via the (1966) and VAN DER HELM et al. (1980)
carboxylate type siderophore rhizoferrin but showing a cyclic, modified Om-0rn-0rn-Gly-
still were able to transport ferric complexes of Gly-Gly hexapeptide. Ferrichrome may be re-
keto hydroxy bidentate ligands (THIEKEN garded as a prototype siderophore from
and WINKELMANN, 1993), suggesting that the which a variety of other “ferrichromes” are
transport systems for ferric rhizoferrin and derived, e.g., ferricrocin, ferrichrysin, ferri-
ferric complexes of keto hydroxy bidentate li- chrome c , ferrichrome A, ferrirubin, ferri-
gands are different. Although rhizoferrin is a rhodin, malonichrome. An updated detailed
R’ 0
( C’
O0
Fig. 32. Ferrichromes.

Ferrichrome: R’ =R2=H, R3=CH3; ferrichome A: R’=R2=H, R 3 = A ferricrocin: R’=H,
R2= CH20H, R3=CH3; ferrichrysin: R’ = R2= CH20H, R3= CH3; ferrirubin: R’ = R2= CH20H, R3=B;
ferrirhodin: R’ = R2= CH20H, R3= C; asperchromes: R’ = R2= CH,OH, R3=mixture of C and CH3.
description is given by VAN DER HELMand moved on a Sephadex LH 20 column and a

WINKELMANN (1994). With the exception of final separation on a silica gel-silica gur (1:l)
tetraglycyl ferrichrome, a heptapeptide column with dichloro methane-methanol
(DEMLet al., 1984), all other members of the (2:l) as an eluting solvent (DEML et al.,
ferrichrome family possess a cyclic hexapep- 1984). HPLC separation using acetonitrile-
tide backbone. Ferrichrome type sidero- water mixtures (KONETSCHNY-RAPP et al.,
phores contain a characteristic tripeptide se- 1988) or TLC on silica gel plates using dichlo-
quence of NS-acyl-NS-hydroxy-~-ornithinero methane-methanol-water (65:25:4) as a
and a variable tripeptide sequence of short solvent system are recommended for identifi-
amino acids (Gly, Ser, Ala). Variations in the cation and isolation. A detailed survey on the
latter tripeptide sequence and in the nature of methods of isolation and identification of fun-
the N-acyl substituents of the ornithine resi- gal siderophores is given by JALALand VAN
dues account for the number of distinct fer- DER HELM(1991).
richrome-type siderophores mentioned The biological properties of siderophores
above. X-ray diffraction studies and CD can be studied by growth promotion tests or
measurements in solution have revealed that by transport experiments with radio-labeled
the configuration of all ferrichromes is lamb- siderophores as discussed in previous reviews
da-cis (VAN DER HELMet al., 1980; LEONG (WINKELMANN, 1991a, 1993). Originally fer-
and RAYMOND, 1974). The stability of desfer- richrome and ferrichrome A were character-
ri-ferrichrome with ferric ions is in the range ized by their growth-stimulating effect on Pi-
of K ~ 1029-1031.
= lobolus and Arfhrobacter species. A. flavesc-
VAN DER HELMand co-workers have iso- ens (now reclassified as Aureobacterium flav-
lated the so-called “asperchromes” as minor escens) JG9 (ATCC 25091) is a siderophore-
constituents in low-iron culture filtrates of auxotroph and requires trace amounts of fer-
Aspergillus ochraceus, containing a common richrome or other hydroxamate siderophores
cyclic Om-0rn-0rn-Ser-Ser-Gly hexapeptide but does not respond to catechols (RABSCH
sequence as in ferrichrysin, ferrirhodin, and and WINKELMANN, 1991; SHANZERet al.,
ferrirubin but different ornithyl-N’-acyl 1988). EMERY(1971) was the first to show
groups (JALALet al., 1984). Interestingly, a that ferrichrome indeed functions as a sidero-
ferrichrome-type siderophore with an incom- phore in Ustilago sphaerogena. Ferrichrome is
plete, open-chain peptide was detected and actively taken up. The iron-free ligand is ex-
named des(diserylglycy1)ferrirhodin (JALAL creted again and can be used for another
et al., 1985). The latter has recently also been round of transport. Contrary to that, ferri-
found in the edible mushroom Agaricus bi- chrome A does not enter the cells but donates
sporus (ENG-WILMOT et al., 1992). iron to the external face of the cell membrane
Ferrichrome was originally isolated from where it is removed by a reduction process
Ustilago sphaerogena grown in iron-contain- (ECKERet al., 1982a). A mutant of Neurospo-
ing medium. A key observation was made ra crassa, unable to synthesize siderophores
when the production of ferrichrome could be in the absence of ornithine, has been used to
enhanced by growing Ustilago in an iron-defi- study uptake of ferrichrome-type sidero-
cient medium. The composition of the me- phores and coprogen (WINKELMANN and
dium for the production of ferrichrome and ZAHNER,1973). Uptake of ferrichromes in N.
ferrichrome A is a modified Grimm-Allen crassa and Penicillium parvum has been
medium. To isolate ferrichromes, iron is ad- shown to be highly stereospecific (WINKEL-
ded under stirring to the cell-free supernatant MA”, 1979; WINKELMANN and BRAUN,
of a low-iron culture after a week of growth 1981). Thus enanfio-ferrichromes (NAEGELI
with strong aeration. The ferrichromes can and KELLER-SCHIERLEIN, 1978) were not re-
easily be adsorbed on a column of XAD-2 recognized by the fungal siderophore transport
sin and, after washing with distilled water to systems (HUSCHKAet al., 1985, 1986).
remove salts and other polar constituents, the
ferrichromes can be desorbed with acetone-
water (1:l) or methanol. Lipids can be re-
4.2 Coprogens ia sp. (for review, see JALALand VAN DER

HELM,1991).
Coprogen (Fig. 33) has been isolated from Analysis of coprogen transport in wild type
Penicillium species, from Neurospora crassa, and mutant strains of Neurospora crassa using
Curvularia, and a variety of other fungi (JAL- single-labeled 55Fe-coprogen, 14C-coprogen,
AL and VAN DER HELM,1991). A method for or double-labeled 55Fe-'4C-coprogen (WIN-
coprogen production and isolation from N. KELMANN and ZAHNER, 1973; HUSCHKAet
crassa was described by WONGet al. (1983). al., 1985) indicated that the intact coprogen
The structure of coprogen was elucidated by molecule can enter the cell. Furthermore,
KELLER-SCHIERLEIN and DIEKMANN (1970) Mossbauer spectroscopic studies (MAT-
and revealed a linear trihydroxamate sidero- ZANKE and WINKELMANN, 1981) confirmed
phore composed of 3 mol of N5-acyl-N5-hy- that after uptake a major portion of intracel-
droxy-L-ornithine, 3 mol of anhydro meva- lular coprogen remains in the iron complex
lonic acid and 1 mol of acetic acid. Desacetyl form and iron is removed by reduction only
coprogen, named coprogen B, was isolated slowly. Moreover, an additional ferrichrome-
from Fusarium dimerum (DIEKMANN, 1970). type siderophore, ferricrocin, was found to be
A further coprogen, triornicin, was found in synthesized only as an intracellular sidero-
low-iron culture filtrates of Epicoccum pur- phore functioning as an iron storage com-
purescens (FREDERICKet al., 1981). Curvu- pound from which iron is gradually removed
laria lunata produces neocoprogen I and neo- by reduction. Uptake and competition experi-
coprogen 11, in which the two outer anhy- ments have shown that N. crassa possesses
dromevalonic acid residues are substituted by two different siderophore recognition sites,
one or two acetyl groups, respectively. The one for ferrichrome-type siderophores one
CD spectrum of coprogen showed a delta confor coprogen, both of which donate their side-
figuration in solution (WONG et al., 1983), rophores to a common transport system lo-
and the crystal structure of neocoprogen I re- cated in the cytoplasmic membrane (HUSCH-
vealed a delta-trans configuration (HOSSAIN KA et al., 1985; CHUNG et al., 1986). Details
et al., 1987). Further coprogen derivatives, of the actual transport process are still unre-
such as hydroxy coprogen, N"-dimethyl co- solved. However, there is evidence that the
progen as well as their corresponding neo- transport of siderophores in fungi is function-
derivatives have been isolated from Alternar- ally connected to the membrane potential
Fig. 33. Coprogens. A HO"(^

a Coprogen: R' =H,
R2=COCH3, R3= R4 = A.
Coprogen B: R ' = R 2 = H ,
R3=R4 =A.
Neocoprogen I: R' =H,
R2= COCH3, R3= CH,,
R4= A.
Neocoprogen 11: R'= H,
R~ =COCH,,
R3= R4= CH3.
Nu-dimethyl coprogen:
R' = R = ~ CH,,
R3=R4=A.
b Desferricoprogen:
R3=R4 =A.
generated by the plasma membrane ATPase

(HUSCHKA et al., 1983).
4.3 Rhodotorulic Acid C

I
0 0 a
A compound with the properties of a sec-
ondary hydroxamic acid strongly binding
iron(II1) was isolated from supernatants of
iron-deficient cultures of Rhodotorula pilima-
nae (ATKINand NEILANDS,1968). This com-
pound, named rhodotorulic acid (RA) (Fig.
34a), was characterized as the diketo pipera-
zine of N-acetyl-L-N-hydroxy ornithine and
has subsequently been found in species of
Leucosporidium, Rhodosporidium, Sporidio-
bolus, and Sporobolomyces (ATKIN et al.,
1970). The diketopiperazine moiety is also b
present in coprogen as dimerum acid (Fig.
34b). Thus, rhodotorulic acid (RA) is struc-
turally related to the coprogen family. The
iron complex of rhodotorulic acid at neutral
pH has been found to be dimeric with the for-
mula Fe2(RA)3,where both iron atoms have NH
the delta-cis configuration (CARRANO and I
R
RAYMOND,1978). Below pH 3 this complex
dissociates to give the monomer, FeRA+, in Fig. 35. Fursarinines.
a Fusarinine A. n =2, fusarinine B: n = 3; b fusarin-
which each iron is bound to two hydroxamate ine C (Fusigen), R = H ; triacetyl fusarinine C (tria-
ligands. Ferric RA has an absorption maxi- cetyl fusigen), R = COCH3.
mum at 425 nm ( ~ = 2 7 0 0M-' cm-') at pH 7
and becomes red at pH 2 with an absorption
maximum at 480 nm ( ~ = 1 7 5 0M-' cm-').
Ferric RA functions as an iron transport
agent in Rhodotorula pilimanae and probably 4.4 Fusarinines (Fusigens)
in other RA-producing heterobasidiomyce-
tous yeasts. RA mediates iron transport to Fusarinines (Fig. 35) had originally been
the cell but does not actually transport iron isolated from Fusarium cubense (DIEKMANN,
into the cell (CARRANOand RAYMOND, 1967) and F. roseum (SAYER and EMERY,
1978). Citrate was found to be as effective in 1968) and were later shown to occur in other
iron transport as RA in this fungus. genera, e.g., Giberella, Aspergillus, Penicil-
lium, and a variety of other fungi (DIEK-
MANN,1968; JALALet al., 1986, 1987). Re-
cently, fusarinine C was described as an endo-
genous siderophore in Agaricus bisporus
(ENG-WILMOT et al., 1992). Alkaline degrad-
ation yielded 3 mol of the monohydroxamic
acid cis-fusarinine (Fig. 35a) in which N5-hy-
&
droxy-L-ornithine is N-acylated by cis-5-hy-
droxy-3-methyl-2-pentenoic acid. The three
A: OH fusarinine molecules are esterified head-to-
Fig. 34. Rhodotorulic acid: R = CH3; dimerum acid tail to build the cyclic triester fusigen (Fig.
R=A. 35b). The free amino groups show a pK of 7.1
and the ferric iron complex is stable up to pH an enzyme which hydrolyzes the ornithine es-
1. In addition to the cyclic triester, the linear ter bonds of fusigen (fusarinine C) but not the
trimer (fusarinine B), the linear dimer (fusar- ferric chelate of fusigen. Penicillium sp. was
inine A), and the monomer (fusarinine) can found to hydrolyze triacetyl fusigen and was
be found in the culture filtrate. Another fully active on the ferric trihydroxamate che-
member of the fusigen family, N,N',N"-tria- late. The iron-free form of triacetyl fusigen
cetyl fusarinine C (TAFC) or triacetyl fusigen was also reported to act as an antibiotic on a
(Fig. 35c), has been isolated from Aspergillus variety of bacteria grown on minimal media
fumigatus strains (DIEKMANNand KREZ- (ANKE, 1977). Bacillus brevis, Clostridium
DORN, 1975) and from Penicillium sp. pasteurianum, Pseudomonas fluorescens, and
(MOOREand EMERY,1976). The ferric tri- Streptomyces viridochromogenes were found
acetyl fusarinine C is a relatively flat molecule to be sensitive even on complex media.
with a total thickness of about 45 pm as deter-
mined from the crystal structure (HOSSAINet
al., 1980). Two different coordination isomers 4.5 Rhizoferrins
of the triacetyl fusigen molecule have been
observed. Crystals from ethanolbenzene ad- Rhizoferrin (Fig. 36) is a novel carboxylate-
opted a lambda-cis absolute configuration. type siderophore first isolated from low-iron
However, in solution and in the morphologi- cultures of Rhizopus microsporus var. rhizo-
cally different crystals from chloroform, the podiformis (DRECHSELet al., 1991, 1992;
molecule predominantly existed as a delta-cis WINKELMANN, 1992). Further studies have
isomer, as determined by circular dichroism. shown that rhizoferrin seems to be the char-
Fusigen can be isolated from low-iron cul- acteristic siderophore in the order of Mucor-
tures of Fusarium cubense in an asparagine- ales (Zygomycetes) as a variety of other
salts medium (DIEKMANNand ZAHNER, strains from different genera (Mucor, Phyco-
1967). The medium is inoculated with a sus- myces, Chaetostylum, Absidia, Cokeromyces,
pension of spores and incubated at 27°C on a Cunninghamella, Mycotypha, and Mortierel-
rotary shaker or in a fermenter with aeration. la) produce rhizoferrin as their only sidero-
After 3-5 d FeC& solution is added, and the phore (THIEKENand WINKELMANN, 1992).
siderophores are adsorbed on Servachrome Hydroxamates could not be found. However,
XAD-2, washed with three volumes of dis- recent results from the authors' laboratory
tilled water, and desorbed with one volume of point to the existence of hydroxamate sidero-
methanol. The crude siderophore solution is phores in strains of Basidiobolus and Conid-
evaporated to dryness and dissolved in 0.01 M iobolus (THIEKENand WINKELMANN, un-
ammonium acetate buffer (pH 5 ) and passed published results) which are regarded as a
through a CM-Sephadex column equilibrated separate genus of Zygomycetes (ZYCHAand
and eluted with the same buffer. Bound fusig- SIEPMANN, 1969).
en is then eluted with 0.1 M ammonium ace- The stability constant of rhizoferrin with
tate buffer, and again adsorbed on XAD-2, ferric iron has been determined to be
washed, and desorbed as described before. A Kf= loB (ALBRECHT-GARY, personal com-
further purification can be achieved on Se- munication). The existence of rhizoferrins as
phadex LH-20 in methanol. a novel carboxylate siderophore in Mucorales
Triacetyl fusigen can be isolated from Peni-
cillium and Aspergillus strains using the same
medium as described for the production of fu-
sigen (DIEKMANNand KREZDORN,1975).
Transport experiments with 55Fe-labeled fu-
sigen in Aspergillus fumigatus revealed high
uptake rates, suggesting rapid utilization of
the chelated iron by the producing strain
(WIEBE and WINKELMANN, 1975). EMERY 'COOH HOOC'
(1976) reported Fusarium roseum to contain Fig. 36. Rhizoferrin.
is especially interesting, since the Zygomy- reduced. Uptake of iron rhizoferrin and its
cetes are the only fungal class where ferritins metal analogs (Cr, Rh, Ga) was studied in
as iron storage proteins have been found yet. Absidiu spinosu (Mucorales) and in Morgu-
Ferritins have not been detected so far in As- nellu morgunii (Proteeae) (CARRANOet al.,
comycetes and Basidiomycetes (MATZANKE, 1996). While uptake kinetics in the fungus
1994). Therefore, rhizoferrin iron transport could be explained by an active transport sys-
and ferritin iron storage might coexist without tem via a shuttle mechanism, growth promo-
metabolic disturbance. tion assays with M. morgunii could not easily
The siderophore activity has been investi- be explained and were dependent on a num-
gated using ”Fe-labeled rhizoferrin (DRECH- ber of factors including the nature of the che-
SEL et al., 1991; CARRANOet al., 1996). Fur- lating agents used to induce iron deficiency.
thermore, a comparison with ferrioxamines Two genes, rumA and rumB (rhizoferrin up-
revealed that iron uptake rates mediated by take into Morgunellu), encoding an outer
rhizoferrin and ferrioxamines in R. microspo- membrane protein and a periplasmic protein,
rus were similarly effective, suggesting that respectively, have been identified in M. mor-
fungi of the Mucorales are also able to use gunii (KUHNet al., 1996).
ferrioxamines as an iron source (DRECHSEL
et al., 1991). This has important consequences
for their role as pathogens, since some mem-
bers of this group have been repeatedly iso-
lated during fatal cases of mucormycosis in dia- 5 Miscellaneous
lysis patients (BOELAERTet al., 1993, 1994).
Thus, a substantial number of patients treated Compounds
with desferrioxamine (DesferaP) for either
aluminium or iron overload have developed Several miscellaneous compounds have
mucormycosis caused by Rhizopus (BOE- been isolated from fungi containing iron-
LAERT et al., 1991). binding catecholate, phenolate, or keto hy-
A simple growth promotion assay using droxy bidentate ligands. Most of these com-
bacteria of the Proteus group has been devel- pounds originate from the shikimate-choris-
oped to detect bioactive siderophores con- mate pathway and represent condensation
taining keto hydroxy bidentate ligands includ- products of tyrosine. They have been charac-
ing rhizoferrin and simple keto and hydroxy terized earlier as pigments from fungi, mainly
acids (THIEKENand WINKELMANN, 1993). of the Boletales, e.g., Boletus, Suillus, Paxil-
This bioassay not only allows the detection of lus, Hydnellum, Polyporus, and Xerocomus
further fungal carboxylate siderophores but (GILL and STEGLICH,1987). The blueing of
also discriminates between carboxylate and members of the Boletaceae, and the red and
hydroxamate producing fungi. Several deriva- yellow stains of several Aguricus and Corti-
tives of rhizoferrin have been prepared by nurius species have long been used as taxon-
directed fermentation (DRECHSEL et al., omic characteristics by fungal taxonomists. In
1995a). Thus the addition of analogous diami- many cases quinones react by the combined
no acids or diamines to the culture medium action of alkali, oxygen, or ferric choride.
yielded rhizoferrin analogs with either shorter Thus compounds like the terphenyl qui-
(norrhizoferrin) or longer diamino residues nones (polysporic acid, atromentin, leucomel-
(homorhizoferrin). Even substituted amines one, and variegatin) (Fig. 37) possess both
were incorporated leading to 2-ketorhizofer- keto hydroxy and phenolic iron binding func-
rin and oxarhizoferrin. Variation of the citryl tionalities. Internal cyclization leads to cyclo-
residues of rhizoferrin by addition of tricarb- leucomelone, cyclovariegatin, and thelephoric
allylic acid to the culture medium yielded the acid (Fig. 38) having only phenolic groups.
corresponding mono- and didesoxy rhizofer- Another series of compounds are the pulvinic
rin derivatives. As the latter alterations affect acids (vulpinic acid, atromentic acid, xero-
the iron-liganding properties of the rhizofer- comic acid, and variegatic acid) (Fig. 39)
rins, the chelate stability with iron should be which, depending on the kind and number of
5 Miscellaneous Compounds 231
b
COZH
I
OH
I
Fig. 37s. Polysporic acid; b atromen-

tin; c leucomelone; d variegatin.
functional groups, possess good iron-binding cin (HL) (Fig. 40) and terricolin (FeL,), re-
properties. Because of the quinoid and poly- spectively (JEGOROVet al., 1993). Members
phenolic structure these compounds can be of the entomopathogenic genus Tolypocla-
regarded as iron-binding pigments, but their dium are known for their ability to produce
involvement in iron transport has still to be cyclosporin A and a variety of pigments. Al-
proven. though the iron complex (terricolin) has been
A novel pyridinone-type iron-binding com- characterized by X-ray structure determina-
pound has recently been isolated from strains tion, iron transport properties have not been
of the fungus Tolypocladium named tolypo- reported. The absolute configuration of terri-
P a
0
6H
b
OH
C
Fig. 38a. Cycloleucomelone: R =H,cyclovariegatin:
R = OH, b thelephoric acid.
Colin has been shown to be lambda-cis both in

solution and in the solid state.
Pulcherriminic acid (Fig. 41a) and its iron
complex pulcherrimin have been isolated
from Candida pulcherrima (KLUYVER et al.,
1953). Pulcherrimin, obtained by extraction
of the cells with methanolic KOH as an amor- d
phous red powder represents a polymer of di-
hydroxy-pyrazine dioxide which is insoluble
in water and may be regarded as an iron stor-
age compound rather than a siderophore.
Several related compounds, like the anti-
biotic mycelianamide (Fig. 41b) from Penicil-
lium griseofulvum (OXFORD and RAISTRICK,
OH
\
1948) or aspergillic acid (Fig. 41c), hydroxy

aspergillic acid, muta-aspergillic acid, and Fig. 39a. Vulpinic acid b atromentic acid; c xero-
neo-aspergillic acid from culture fluids of As- comic acid; d variegatic acid.
pergillus flavus and A. oryzae (DUTCHER,
1958) were isolated. The latter compounds OH
are water-soluble and showed growth promo- I
tion activity with A. flavescens JG. However,
siderophore activity in the producing strains
has not been demonstrated.
Fig. 40. Tolypocin.

6 Transport Mechanisms 233
(1) Diffusion of siderophores across the

membranes along a diffusion gradient.
(2) Active transport of the entire iron chelate
into the cell and intracellular reductive
removal of iron with degradation or ex-
pulsion of the free ligand.
(3) Transport of siderophores only to the
OH membrane surface, with or without re-
duction of iron, and donation of iron to
the cell interior without penetration of
the complex or ligand into the cell (taxi
model). Indeed, all three modes of side-
rophore uptake have been shown to oc-
cur in certain microorganisms under cer-
tain experimental conditions (WINKEL-
MANN, 1991a).
Siderophore iron transport in fungi has re-

cently been reviewed with emphasis on the
conditions and precautions of kinetic meas-
urements (WINKELMANN, 1993). Generally,
there are several approaches to unravel the
Fig. 41a. Pulcherrimic acid b mycelianamide; c as- mechanisms of fungal iron transport: radioac-
pergillic acid. tive labeling, photoaffinity labeling, and spec-
troscopy. Using 55Fe-and 14C-labeledsidero-
phores four types of transport have been de-
fined (VAN DER HELMand WINKELMANN,
6 Transport Mechanisms 1994): (1) a shuttle mechanism (e.g., ferri-
chromes in Neurospora); (2) a taxi cab mech-
To determine whether or not an iron-bind- anism (e.g., Fe-rhodotorulate in Rhodotoru-
ing compound can be regarded as a sidero- la); (3) a hydrolytic mechanism (e.g., fusarin-
phore, its physiological function and genetic ines in Mycelia sterilia); and (4) a reductive
regulation must be studied. An increasing mechanism (siderophores and ferric ions in
number of microbial iron transport systems Saccharomyces and other non-siderophore
are currently being identified and character- containing yeasts). Using a photoaffinity label
ized at the molecular level. Siderophore (p-azidobenzoyl coprogen) 50% irreversible
transport in enterobacteria and pseudomon- inhibition of coprogen and ferrichrome up-
ads has been reviewed by BRAWN and take was measured in Neurospora crassa
HANTKE(1991) and MARWGGand WEIS- (BAILEYet al., 1986) - an argument in favor
BEEK (1991). Fungal iron transport, although of a common transport system for these side-
mainly characterized in terms of structure- rophores. These results correspond to earlier
function relationship (VAN DER HELMand transport kinetics showing a common trans-
WINKELMANN, 1994) is coming to the same port system but different siderophore recep-
molecular level of understanding that has tors (HUSCHKA et al., 1985). Although a vari-
been achieved in bacterial systems (MEI and ety of siderophores may be recognized by
LEONG, 1994; LESWISSE and LABBE, 1994). fungal siderophore receptors, recognition is
Siderophore-mediated iron transport has highly stereoselective as determined by com-
been studied in bacteria and fungi using ra- parative studies with ferrichrome and enan-
dio-labeled siderophores in wild type strains tio-ferrichrome (WINKELMANN, 1979; WIN-
and mutants. In principle, three mechanisms KELMANN and BRAUN, 1981). These results
of uptake into microorganisms are conceiv- confirm that fungal siderophore transport is
able: not a diffusion-controlled process but re-
quires specific interactions with components and the antibiotic albomycin (HANTKEand
of the transport system, although proteins of BRAUN,1975). Although ferrichrome is first
fungal siderophore transport systems have bound to the FhuA receptor, it has to be
not been identified so far. processed further by additional gene products
There are several reports on the applica- (FhuB, C, D) to permit completion of sidero-
tion of EPR and Mossbauer spectroscopic phore transport into the cell (KOSTER,1991).
measurements to fungal siderophore iron up- This mechanism of nutrient translocation is
take (MATZANKEand WINKELMANN, 1981; known as a periplasmic binding-dependent
ECKERet al., 1982; MATZANKEet al., 1988) transport which earlier was described for oth-
supporting the view that iron undergoes sub- er transport processes (AMES,1986).
sequent reductive removal and that under Different ferric hydroxamate uptake (Fhu)
certain conditions siderophores can also serve outer membrane receptors of E. coli have
as iron storage compounds (see also the re- been shown to mediate the transport of var-
view of MATZANKE,1994). Interestingly, si- ious fungal siderophores, e.g., ferrichromes,
derophores are also found in conidiospores coprogen, and rhodotorulic acid (HANTKE,
and seem to have an important function dur- 1983). Surprisingly, the fungal siderophores
ing sporulation and germination (HOROWITZ coprogen and Fe rhodotorulate enter the cells
et al., 1976; MATZANKE et al., 1987; WINKEL- of E. coli by the same outer membrane recep-
MANN,1991b). Up to now only few reports on tor protein (FhuE) as the bacterial ferriox-
the transport of carboxylate type sidero- amines, whereas members of enterobacterial
phores, like rhizoferrin, in fungi are available Erwina herbicola (Pantoea herbicola) use a
(DRECHSEL,et al., 1991; THIEKENand WIN- separate outer membrane receptor FoxA to
KELMANN, 1992). The mechanism of uptake transport ferrioxamines (BERNERand WIN-
in fungi of the Mucorales (Zygomycetes) is KELMANN, 1990).
still under investigation. While FhuD is located in the periplasmic
The mechanism of hydroxamate iron trans- space, FhuB and FhuC are integrated in the
port in enterobacteria was examined in early cytoplasmic membrane. All fhu genes are or-
studies with E. coli K-12 and Salmonella ty- ganized in an operon; they have been se-
phimurium LT-2 mutants defective in entero- quenced, and the deduced molecular weights
bactin synthesis using 55Fe-and 3H-labeled of the polypeptides are in accordance with
ferrichrome and their chromic analogs those found by SDS-PAGE. Although the hy-
(LEONGand NEILANDS,1976). These investi- droxamate siderophores enter the cells by dif-
gations already showed that ferrichrome is ferent outer membrane receptors, e.g., fer-
taken up as an intact chelate in both E. coli richromes (FhuA), coprogens (FhuE) and Fe
and Salmonella. Later studies revealed that rhodotorulate (FhuE), aerobactin (Iut), and
the ferrichrome ligand is modified by acyla- ferrioxamines (FoxA), their transport across
tion of the hydroxamate N-OH groups after the cytoplasmic membrane is accomplished
iron delivery (HARTMANN and BRAUN,1980) by the same transport proteins (FhuD, B, C).
thereby reducing the stability of ferrichrome In addition, a TonB protein is essential for an
with iron within the cell. Our current knowl- “energized state” of the outer membrane re-
edge of ferrichrome uptake in E. coli has in- ceptors as shown by tonB mutations which
creased considerably by a detailed genetic prevent correct interaction of the TonB pro-
analysis of the various membrane compo- tein with outer membrane receptor proteins
nents involved (BRAUNand HANTKE,1991). (BRAUNet al., 1991). Thus TonB is regarded
In gram-negative bacteria, the transport of as a coupling device between the outer and
iron bound to hydroxamate siderophores gen- inner membrane. The interaction of TonB
erally depends on the expression of outer with the receptor proteins requires the pres-
membrane receptor proteins. Ferrichrome, ence of a common binding domain which is
e.g., is taken up via the FhuA receptor pro- characterized on the genetic level by a con-
tein (78kDa) localized in the outer mem- sensus sequence, the so-called “TonB box”,
brane of gram-negative bacteria. FhuA also present in all TonB-dependent receptor pro-
binds the phages T1, T5, @80, to colicin M, teins (BRAUNand HANTKE,1991).
7 Conclusion and Perspectives 235
ERNSTet al. (1978) were the first to ob- ments of receptor and binding proteins in-
serve a constitutive production of sidero- volved. There is a structure-function relation-
phores in a S. typhimurium mutant and ship between iron-containing siderophores
coined the term fur (ferric uptake regulation) and the surface of transport proteins. Confor-
for the gene responsible for the regulation of mation of siderophores seems to be the most
siderophore biosynthesis and expression of important factor. In gram-negative bacteria
the cognate receptor proteins. The fur gene two membranes and a periplasmic space have
later was also found in E. coli and was subse- to be penetrated before iron can be utilized
quently cloned and sequenced (HANTKE, by the cell metabolism. Although molecular
1984). The corresponding protein was charac- genetics has become the most powerful tool
terized by WEE et al. (1988). It turned out in analyzing the transport of siderophores,
that all flu genes and other siderophore re- many open questions remain, e.g.: Which part
ceptor genes (enterobactin, aerobactin, ferri- of the siderophore molecule is recognized and
oxamines) are regulated by iron via the f i r what kind of interaction is to be expected?
gene product. This also applies to the biosyn- Should the outer membrane receptor protein
thetic genes of siderophores. be a channel protein through which the side-
Uptake of ferrichrome in fungi is stereo- rophore can pass with specific interaction, as
specific, as proven with synthetic enantio-fer- has been inferred from recent results (KILL-
richrome which has the opposite chirality MANN et al., 1993; RUTZet al., 1993)? HOW is
(WINKELMANN and BRAUN,1981). However, the outer membrane receptor gated or ener-
uptake of enantio-ferrichrome (delta configu- gized by the TonB protein? What kind of in-
ration) in E. coli was still 50% as compared to teractions prevail at the periplasmic binding
uptake of the natural ferrichrome (lambda proteins (FhuD) and, finally, how does the
configuration), indicating either an incom- translocation through the inner cytoplasmic
plete recognition or additional routes for iron membrane proceed?
uptake. Enantioselectivity seems to be more In fungi similar structure-function relation-
pronouced with the cloned Yersiniu ferri- ships have been documented and allocated to
chrome receptor protein (FcuA) accepting putative cytoplasmic membrane proteins.
ferrichrome but excluding enantio-ferri- However, since siderophore transport mu-
chrome completely (BAUMLERand HANTKE, tants are not available, membrane proteins
1992). A recent analysis of the siderophore involved in siderophore binding and trans-
specificity of different Fhu receptors in E. coli port have never been identified. However,
has revealed that certain ferrichromes like the principal mechanisms of interaction with
ferrichrysin and ferrirubin may not only enter specific membrane proteins also seem to be
via FhuA but may also use the FhuE receptor valid in fungal siderophore transport. The ba-
(KILLMANN and BRAUN,1992). sic question as to whether the entire chelate
molecule or only the iron atom enters the cell
has been solved with double-labeled sidero-
phores in various fungi. According to our
present knowledge only some siderophores,
7 Conclusion and mainly ferrichromes, can enter the fungal cell
and remain intact. Several other siderophores
Perspectives need to interact with membrane receptors to
allow exchange of iron. In every case, howev-
The knowledge of siderophores and their er, specific interaction is observed. Thus the
cognate iron transport systems in microorgan- essential features of structure-function rela-
isms is crucial for an understanding of basic tionship among bacterial and fungal sidero-
events, such as growth, metabolic activity, phore systems seem to be identical.
host invasion, and virulence. In all cases iron Future applications of siderophores are
nutrition is a prerequisite. The diversity of si- manifold. The lag phase of slowly growing mi-
derophores is remarkable and can only be croorganisms, e.g., might be shortened to a
discussed in relation to the structural require- certain extent by simply adding very small
amounts of genus-specific siderophores.

Moreover, addition of siderophores to biolog-
8 References
ical fluids or food may overcome iron restric- ABDALLAH, M. A. (1991), Pyoverdins and pseudo-
tion due to the iron-withholding properties of bactins, in: Handbook of Microbial Iron Chelates
iron-binding proteins (SOUKKAet al., 1992) (WINKELMANN, G., Ed.), pp. 139-153, Boca Ra-
and may thus allow outgrowth and detection ton, FL: CRC Press.
of otherwise dormant microorganisms. The ACTIS,L. A., FISH,W., CROSA,J. H., KELLERMAN,
use of genetically engineered organisms that K., ELLENBERGER, S. R., HAUSER,F. M., SAN-
possess improved iron transport systems rec- DERS-LOHR, J. (1986), Characterization of an-
ognizing specific siderophores may be an al- guibactin, a novel siderophore from Vibrio an-
ternative approach to enhance growth and guillarum 775 (pJMl), J. Bacteriol. 167, 57-65.
ACTIS,L. A., TOMALSKY, M. E., FARREL,D. H.,
metabolite production. Also the use of non- CROSA,J. H. (1988), Genetic and molecular
utilizable or antibiotic-linked siderophores characterization of essential components of the
(NIKAIDOand ROSENBERG, 1990) designed Vibrio anguillarum plasmid-mediatediron trans-
for reducing the bioavailablity of iron by con- port system, J. Biol. Chem. 263,2853-2860.
taminating bacteria might be an interesting AMES,G . F. L. (1986), Bacterial periplasmic trans-
future biotechnological application. The best port systems: structure, mechanism, and evolu-
examples of siderophore application can be tio, Ann. Rev. Biochem. 55, 397-425.
found in nature where biocontrol and host ANKE,H. (1977), Metabolic products of microor-
defense mechanisms based on nutritional iron ganisms: desferritriacetylfusigen, an antibiotic
from Aspergillus deflectus, J. Antibiot. 30, 125-
deficiency are active in a finely tuned manner 128.
without negative effects on the surrounding ARCENEAUX, J. E. L., DAVIS,W. B., DOWNER, D.
ecosystems. The biotechnological applications N., HAYDON,A. H., BYERS,B. R. (1973), Fate
of siderophores to medicine, agriculture, the of labeled hydroxamates during iron transport
environment, and the food industry are nu- from hydroxamate-iron chelates, J. Bacteriol.
merous. For example, the affinity of sidero- 115,919-927.
phores for toxic metals, such as chromium ARONSON, D. B., BOYER,G . L. (1994), Growth
and plutonium, suggests that siderophores and siderophore formation in six iron-limited
may be useful agents for remediation of pol- strains of Frankia, Soil Biol. Biochem. 265, 561-
567.
luted environments. Likewise, the develop- ATKIN,C. L., NEILANDS, J. B. (1968), Rhodotorul-
ment of orally effective siderophores suitable ic acid, a diketopiperazine dihydroxamic acid
for iron chelation therapy through directed with growth factor activity: isolation and charac-
fermentation is a virtually unexplored area of terization, Biochemistry 7, 3734-3739.
investigation. A possible role of bacterial si- ATKIN,C. L., NEILANDS, J. B., PHAFF,H. J. (1970),
derophores in reducing inflammation (COFF- Rhodotorulic acid from species of Leucosporid-
MANN et al., 1990) as well as protecting cells ium, Rhodosporidium, Rhodotorula, Sporidio-
from free radical damage has also been sug- bolus, and Sporobolornyces and a new alanine
gested. containing ferrichrome from Cryptococcus meli-
biosum, J. Bacteriol. 103, 722-733.
However, addressing of possible applica- BAGG, A., NEILANDS, J. B. (1987), Molecular
tions of siderophores in more detail is beyond mechanism of regulation of siderophore-me-
the scope of this chapter. The authors are diated iron assimilation, Microbiol. Rev. 51,509-
convinced that the future will bring exciting 518.
advances in this area. The present review BAIIEY, C. T., KIME-HUNT, E. M., CARRANO, C.
aims at understanding the wealth of the struc- J., HUSCHKA, H., WINKELMANN, G . (1986), A
tural diversity of siderophores, their produc- photoaffinity label for the siderophore-mediated
tion, and mode of action and will hopefully iron transport system in Neurospora crassa, Bio-
provide a stimulus for future research in mi- chim. Biophys. Acta 883, 299-305.
BARGHOUTHI, S., YOUNG,R., ARCENEAUX, J. E.
crobial physiology, ecology, and fields of ap- L., BYERS,B. R. (1989), Physiological control of
plied biotechnology. arnonabactin biosynthesis in Aeromonas hydro-
phila, BioMetals 2, 155-160.
BARGHOUTHI, S., PAYNE,S. M., ARCENEAUX, J.
E. L., BYERS,B. R. (1991), Cloning, mutagene-
8 References 237
sis, and nucleotide sequence of a siderophore erichia coli is part of the Fe”-aerobactin trans-
biosynthetic gene (amoA) from Aeromonas hy- port system, J. Bacteriol. 150, 1472-1475.
drophila, J. Bacteriol. 173,5121-5128. BITTER, W., MARUGG,J. D., DE WEGER,L. A.,
BARTON,L. L., HEMMING, B. C. (Eds.) (1993), TOMMASSEN, J., WEISBEEK,P. J. (1991), The
Iron Chelation in Plants and Soil Microorgan- ferric-pseudobactin receptor PupA of Pseudo-
isms. San Diego, CA: Academic Press. monas putida WCS 358: homology to TonB-de-
BAUMLER,A. J., HANTKE,K. (1992), Ferrioxam- pendent Escherichia coli receptors and specifici-
ine uptake in Yersinia enterocolitica: characteri- ty of the protein, Mol. Microbiol. 5, 647-655.
zation of the receptor protein FoxA, Mol. Mi- BOELAERT,J. R., FENVES,A. Z., COBURN,J. W.
crobiol. 6, 1309-1321. (1991), Deferoxamine therapy and mucormyco-
BENZ,G., SCHRODER, T., KURZ,J., WUNSCHE,C. sis in dialysis patients: report of an international
K. W., STEFFENS,G., PFITZNER,J., SCHMIDT, registry, Am. J. Kidney Dis. 18, 660-667.
D. (1982), Konstitution der Desferriform der Al- BOELAERT,J. R., DELOCHT,M., VAN CUTSEM,J.,
bomycine a,, &, E, Angew. Chemie 94, 552-553. KERBELS,V., CANTINIEAUX, B., VERDONCK,
BERGERON, R. J., MCMANIS,J. S. (1991), Synthe- A., VAN LANDUYT,SCHNEIDER, Y.-J. (1993),
sis of catecholamide and hydroxamate sidero- Mucormycosis during desferoxamine therapy is
phores, in: Handbook of Microbial Iron Chelates a siderophore-mediated infection. In vitro and in
(WINKELMANN, G., Ed.), Boca Raton, F1: CRC vivo animal studies, J. Clin. Invest. 91, 1979-
Press. 1986.
BERGERON,R. J., DIONIS,J. B., ELLIOT,G. T., BOELAERT,J. R., VANCUTSEM,J., DELOCHT,M.,
KLINE,S. J. (1985), Mechanism and stereospe- SCHNEIDER, Y.-J., CRICHTON, R. R. (1994), De-
cificity of the parabactin-mediated iron transport feroxamine augments growth and pathogenicity
system in Paracoccus denitrificans, J. Biol. of Rhizopus, while hydroxypyridinone chelators
Chem. 260,7936-7944. have no effect, Kidney Int. 45, 667-671.
BERNER,I., WINKELMANN, G. (1990), Ferrioxam- BOYER,G. L., ARONSON,D. B. (1994), Iron up-
ine transport mutants and the identification of take and siderophore formation in the actinorhi-
the ferrioxamine receptor protein (FoxA) in Er- zal symbiont Frankia, in: Biochemistry of the
winia herbicola (Enterobacter agglomerans), Metal Micronutrients in the Rhizosphere (MAN-
BioMetals 2, 197-202. THEY,J. A., CROWLEY,D. E., LUSTER,D. G.,
BERNER,I., KONETSCHNY-RAPP, S., JUNG, G., Eds.), pp. 41-54. Boca Raton, FL: CRC Press.
WINKELMANN, G. (1988), Characterization of BRAUN,V. (1981), Escherichia coli cells, containing
ferrioxamine E as the principal siderophore of the plasmid COW produce the iron ionophore
Envinia herbicola (Enterobacter agglomerans), aerobactin, FEMS Microbiol. Lett. 11, 225-228.
BioMetals 1, 51-56. BRAUN,V., HANTKE,K. (1991), Genetics of bacte-
BERNER,I., GREINER, M., METZGER,J., JUNG,G., rial iron transport, in: Handbook of Microbial
WINKELMANN, G. (1991), Identification of ente- Iron Chelates (WINKELMANN, G., Ed.), pp. 107-
robactin and linear dihydroxybenzoylserine 138. Boca Raton, FL: CRC Press.
compounds by HPLC and ion spray mass spec-
trometry (LC/MS and MSIMS), BioMetals 4, BRAUN,V., GUNTER, C., HANTKE,K. (1991),
113-118. Transport of iron across the outer membrane,
BioMetals 4, 14-22.
BERNHARDINI, J., LINGET-MORICE, C., HOH, F.,
COLLINSON,S. K., KYSLIK,P., PAGE, W. J., BRICKMAN, T. J., HANSEL,J.-G., MILLER,M. J.,
DELL,A., ABDALLAH, M. A. (in press), Bacteri- ARMSTRONG, S. K. (1996), Purification spectro-
al siderophores: Structure elucidation and ‘H, scopic analysis, and biological activity of the ma-
13C, I5N two-dimensional NMR assignments of crocyclic dihydroxamate siderophore alcaligin
azotoverdin and related siderophores synthe- produced by Bordetella pertussis and Bordetella
sized by Azomonas macrocytogenes ATCC bronchiseptica, BioMetals 9, 191-203.
12334, BioMetals 9, 107-120. BRISKOT, G., TARAZ, K., BUDZIKIEWICZ, H.
BICKEL, H., BOSSHARDT,R., GAUMANN,E., (1989), Pyoverdin-type siderophores from Pseu-
REUSSER,P., VISCHER,E., VOSER,w., WE=- domonas aeruginosa, Liebigs Ann. Chem. 1989,
STEIN, A., ZAHNER,H. (1960), Stoffwechselpro- 375-384.
dukte von Actinomyceten. Uber die Isolierung BUDDE,A. D., LEONG,S. A. (1989), Characteriza-
und Charakterisierung der Ferrioxamine A-F, tion of siderophores from Ustilago maydis, My-
neuer Wuchsstoffe der Sideramin-Gruppe, Helv. copathologia 108, 125-133.
Chim. Acta 43,2118-2128. BUDZIKIEWICZ, H. (1993), Secondary metabolites
BINDEREIF,A., BRAUN,V., HANTKE,K. (1982), from fluorescent pseudomonads, FEMS Micro-
The cloacin receptor of ColV-bearing Esch- biol. Rev. 104, 209-228.
BUDZIKIEWICZ, H., SCHAFFNER, E. M., TARAZ,K. Azotobacter vinelandii, J. Chem. SOC., Chem.
(1992), A novel azotobactin from Azotobacter Commun. D 186-187.
vinelandii, Nat. Prod. Lett. 1, 9-14. CORNISH, A. S., PAGE,W. J. (1995), Production of
BUTTERTON, J. R., STOEBNER, J. A., PAYNE,S. the tricatecholate siderophore protochelin by
M., CALDERWOOD, S. B. (1992), Cloning, se- Azotobacter vinelandii, BioMetals 8, 332-338.
quencing, and transcriptional regulation of vizA, Cox, C. D., RINEHART,K. L., MOORE, M. L.,
the gene encoding the ferric vibriobactin recep- COOK,J. C. (1981), Pyochelin: novel structure of
tor of Vibrio cholerae, J. Bacteriol. 174, 3729- an iron-chelating growth promotor for Pseudo-
3738. monas aeruginosa, Proc. Nut. Acad. Sci. USA 78,
BUYER,J. S., DELORENZO,V., NEILANDS, J. B. 42564260.
(1991), Production of the siderophore aerobac- CROSA,J. H. (1989), Genetics and molecular biolo-
tin by a halophilic Pseudomonad, Appl. Environ. gy of siderophore-mediated iron transport in
Microbiol. 57, 2246-2250. bacteria, Microbiol. Rev. 53, 517-530.
BYERS,B. R., POWELL,M. V., LANKFORD, C. E. CRUMBLISS, A. L. (1991), Aqueous solution equili-
(1967), Iron-chelating hydroxamic acid (schizo- brium and kinetic studies of iron siderophore
kinen) active in initiation of cell division in Ba- and model siderophore complexes, in: Hand-
cillus megaterium, Bacteriol. 93, 286-294. book of Microbial Iron Chelates (WINKELMANN,
CARRANO, C. J., RAYMOND, K. N. (1978), Coordi- G., Ed.), pp. 177-233. Boca Raton, FL: CRC
nation chemistry of microbial iron transport Press.
compounds: Rhodotorulic acid and iron uptake DELORENZO, V., BINDEREIF, A., PAW, B. H., NEI-
in Rhodotorula pilimanae, J. Bacteriol. 136, 69- LANDS, J. B. (1986), Aerobactin biosynthesis
74. and transport genes of plasmid ColV-K30 in Es-
CARRANO,C., THIEKEN,A., WINKELMANN, G. cherichia coli K-12, J. Bacteriol. 165, 570-578.
(1996), Specificity and mechanism of rhizoferrin
mediated metal ion uptake, BioMetals 9, 185- DEMANGE,P., BATEMAN, A., DELL,A., ABDAL-
189. LAH,M. A. (1988). Structure of azotobactin D, a
siderophore of Azotobacter vinelandii strain D
CASS,M. E., GARRETT,T. M., RAYMOND, K. N. (CCM 289), Biochemistry 27,2745-2752.
(1989), The salicylate mode of bonding in pro-
tonated ferric enterobactin analogs, J. Am. DEMANGE, P., BATEMAN,A., MACLEOD,J. K.,
Chem. SOC. 111, 1677-1682. DELL,A., ABDALLAH, M. A. (1990a), Bacterial
CHAMBERS, C. E., MCINTYRE, D. D., MOUCK,M., siderophores: Unusual 3,4,5,6-Tetrahydropyrim-
SOKOL,P. (1996), Physical and structural charac- idine-based amino acids in pyoverdins from
terization of yersiniophore, a siderophore pro- Pseudomonas fluorescens, Tetrahedron Lett. 31,
duced by clinical isolates of Yersinia enterocoliti- 7611-7614.
ca, BioMetals 9, 157-168. DEMANGE, P., BATEMAN, A., MERZ,C., DELL,A,,
CHAMBERS, C. E., SOKOL,P. A. (1994), Compari- PIBMONT,Y., ABDALLAH, M. (1990b), Bacterial
son of siderophore production and utilization in siderophores: Structures of pyoverdines Pt, side-
pathogenic and environmental isolates of Yersin- rophores of Pseudomonas tolaassii NCPPB 2192,
ia enterocolitica, J. Clin. Microbiol. 32, 32-39. and pyoverdins Pf, siderophores of Pseudomon-
CHUNG,T. D. Y., MATZANKE, B. F. M., WINKEL- as fluorescens CCM 2789. Identification of an
MANN, G., RAYMOND, K. N. (1986), The inhibi- unusual natural amino acid, Biochemistry 29,
tory effect of the partially resolved coordination 11041-11 051.
isomers of chromic desferricoprogen on copro- DEML,G. (1985), Studies in heterobasidiomycetes,
gen uptake in Neurospora crassa, J. Bacteriol. Part 34. A survey on siderophore formation in
165,283-287. low-iron cultured smuts from the floral parts of
COFFMAN, T., Cox, C. D., EDECKER, B. L., BRITI- Polygonaceae, Syst. Appl. Microbiol. 6, 23-24.
GAN, B. E. (1990), Possible role of bacterial si- DEML,G., VOGES,K., JUNG,G., WINKELMANN,
derophores in inflammation, J. Clin. Invest. 86, G. (1984), Tetraglycylferrichrome- the first hep-
1030-1 037. tapeptide ferrichrome, FEBS Lett. 173, 53-57.
CORBIN,J. L., BULEN,W. A. (1969), The isolation DEVOS, P., VANDEWOESTYNE,M., VANCAN-
and identification of 2,3-dihydroxybenzoic acid NEYT, M., VERSTRAETE, w., KERSTERS,K.
and 2-N, 6-N-di (2,3-dihydroxybenzoyI)-~-lysine (1993), Identification of proferrorosamine pro-
formed by iron-deficient Azotobacter vinelandii, ducing Pseudomonas sp. strain GH (LMG
Biochemistry 8,757-762. 11358) as Erwinia rhapontici, System. Appl. Mi-
CORBIN,J. L., KARLE,I. L., KARLE,J. (1970), crobiol. 16, 252-255.
Crystal structure of the chromophore from the DIEKMANN, H. (1967), Fusigen - ein neues Sider-
fluorescent peptide produced by iron-deficient amin aus Pilzen, Arch. Microbiol. 58, 1-5.
8 References 239
DIEKMANN, H. (1968), Stoffwechselprodukte von ECKER,D. J., LANCASTER,J. R., EMERY,T.
Mikroorganismen. 68. Die Isolierung von trans- (1982), Siderophore iron transport followed by
5-Hydroxy-3-methylpenten-(2)-saure,Arch. Mi- electron paramagnetic resonance spectroscopy,
crobiol. 62, 322-327. J. Biol. Chem. 257,8623-8626.
DIEKMANN, H. (1970), Stoffwechselprodukte von EHLERT, G., TARAZ, K., BUDZIKIEWICZ, H.
Mikroorganismen, 81. Mitteilung. Vorkommen (1994), Serratiochelin, a new catechoiate sidero-
und Strukturen von Coprogen B und Dimerum- phore from Serratia marcescens, 2. Naturforsch.
saure, Arch. Microbiol. 73,65-76. 4 9 ~ 11-17.
,
DIEKMANN, H., KREZDORN, E. (1975), Stoffwech- ELKINS,M. F., EARHARDT, C. F. (1989), Nucleo-
selprodukte von Mikroorganismen. 150. Ferri- tide sequence and regulation of the Escherichia
crocin, Triacetylfusigen und andere Sideramine coli gene for ferrienterobactin transport protein
aus Pilzen der Gattung Aspergillus, Gruppe fu- FepB, J. Bacteriol. 171, 5443-5451.
migatus, Arch. Microbiol. 106, 191-194. EMERY,T. (1971), Role of ferrichrome as a ferric
DIEKMANN, H., ZAHNER,H. (1967), Konstitution ionophore in Ustilago sphaerogena, Biochemis-
von Fusigen und dessen Abbau zu delta-2-Anhy- try 10, 1483-1488.
dromevalonsaurelacton, Eur. J. Biochem. 3,213- EMERY, T. (1976), Fungal ornithine esterases: rela-
218. tionship to transport, Biochemistry 15, 2723-
DIONIS,J. B., JENNY,H., PETER, H. H (1991), 2728.
Therapeutically useful iron chelators, in: Hand- ENARD,C., DIOLEZ,A., EXPERT,D. (1988), Sys-
book of Microbial Iron Chelates (WINKELMANN, temic virulence of Erwinia chrysanthemi on
G., Ed.), pp. 339-356. Boca Raton, F L CRC Saintpaulia chrysanthemi 3937 requires a func-
Press. tional iron assimilation system, J. Bacteriol. 170,
DRECHSEL,H., METZGER,J., FREUND,S., JUNG, 2419-2426.
G., BOELAERT,J., WINKELMANN, G. (1991), ENG-WILMOT, L. D., VAN DER HELM,D. (1980),
Rhizoferrin - a novel siderophore from the fun- Molecular and crystal structure of the linear tri-
gus Rhizopus microsporus var. rhizopodiformis, catechol siderophore, agrobactin, J. Am. Chem.
BioMetals 4, 238-243. SOC.102,7719-7725.
DRECHSEL,H., JUNG, G., WINKELMANN, G. ENG-WILMOT,L. D., ADJIMANI, J.P., VAN DER
(1992), Stereochemical characterization of rhizo- HELM, D. (1992), Siderophore mediated iron-
ferrin and identification of its dehydration prod- (111) transport in the mycelia of the cultivated
ucts, BioMetals 5, 141-148. fungus, Agaricus bisporus, J. Inorg. Biochem. 48,
DRECHSEL,H., FREUND,S., NICHOLSON, G., 183-1 95.
HAAG, H., JUNG,O., ZAHNER,H., JUNG,G. ERNST,J. F., BENNET,R. L., ROTHFIELD,L. I.
(1993a). Purification and chemical characteriza- (1978), Constitutive expression of the iron-ente-
tion of staphyloferrin B, a hydrophilic sidero- rochelin and ferrichrome uptake systems in a
phore from staphylococci, BioMetals 6, 185- mutant strain of Salmonella typhimurium, J.
192. Bacteriol. 135, 928-934.
DRECHSEL,H., THIEKEN,A., REISSBRODT,R., FEISTNER, G. F. (1995), Suggestion for a new, sem-
JUNG,G., WINKELMANN, G. (1993b), a-Keto irational nomenclature for the free chelators of
acids are novel siderophores in the genera Pro- ferrioxamines, BioMetals 8, 193-196.
teus, Providencia, and Morganella and are pro- FEISTNER, G., KORTH,H., KO, H., PULVERER, G.,
duced by amino acid deaminases, J. Bacteriol. BUDZIKIEWICZ, H. (1983), Ferrorosamine A
175,2727-2733. from Erwinia rhapontici, Curr. Microbiol. 8,
DRECHSEL,H., TSCHIERSKE, M., THIEKEN,A., 239-243.
ZAHNER, H., JUNG, G., WINKELMANN, G. FEISTNER,G. J., STAHL,D. C., GABRIK,A. H.
(1995), The carboxylate-type siderophore rhizo- (1993), Proferrioxamine siderophores of Erwinia
ferrin and its analogs produced by directed fer- amylovora. A capillary liquid chromatographic/
mentation, J. Ind. Microbiol. 14, 105-112. electrospray tandem mass spectrometric study,
DRECHSEL, H., STEPHAN, H., LOTZ,R., HAAG,H., Org. Mass Spectrom. 28, 163-175.
ZAHNER,H., HANTKE,K., JUNG,G. (1995), FRANCIS, J., MACTURK, H. M., MADINAVEITIA, J.,
Structure elucidation of Yersiniabactin, a sidero- SNOW,G. A. (1953), Mycobactin, a growth fac-
phore from highly virulent Yersinia strains, Lie- tor for Mycobacterium johnei. I. Isolation from
bigs Ann. Chem., 1727-1733. Mycobacterium phlei, Biochem. J. 55, 596-607.
DUTCHER,J. D. (1958), Aspergillic acid: An anti- FREDERICK, C. B., BENTLEY,M. D., SHIVE,W.
biotic substance produced by Aspergillus flavus. (1981), The structure of triomicin, a new sidero-
111. The structure of hydroxyaspergillic acid, J. phore, Biochemistry 20, 2436-2438.
Biol, Chem. 232,785. FUKASAWA, K., GOTO,M., SASAKI,K., HIRATA,
Y., SATO,S. (1972), Structure of the yellow- albomycin by Escherichia coli K-12, Eur. J. Bio-
green fluorescent peptide produced by iron-defi- chem. 99,517-524.
cient Azotobacter vinelandii strain 0, Tetrahe- HAYDON, A. H., DAVIS,W. B., ARCENEAUX, J. E.
dron 28,5359-5365. L., BYERS,B. R. (1973), Hydroxamate recogni-
GIBSON,F., MAGRATH, D. I. (1969), The isolation tion during iron transport from hydroxamate-
and characterization of a hydroxamic acid (aero- iron chelates, J. Bacteriol. 115, 912-918.
bactin) formed by Aerobacter aerogenes 62-1, HESEMANN, J. (1987), Chromosomal-encoded side-
Biochim. Biophys. Acta 192, 175-184. rophores are required for mouse virulence of en-
GILL,M., STEGLICH, W. (1987), Pigments of fungi teropathogenic Yersinia species, FEMS Micro-
(Macromycetes), in: Progr. Chem. of Organic biol. Lett. 18, 229-233.
Natural Products, Vol. 51 (HERZ,W., GRISE- HOHNADEL, D., MEYER,J.-M. (1988), Specificity
BACH,H., KIRBY,G. W., TAMM,C., Eds.). Hei- of pyoverdin-mediated iron uptake among fluo-
delberg: Springer-Verlag. rescent Pseudomonas strains, J. Bacteriol. 170,
GOBIN, J., MOORE,C. H., REEVE,J. R. WONG,D. 48654873.
K., GIBSON,B. W., HORWITZ,M. A. (1995), HOROWITZ, N. H., CHARLANG, G., HORN,G.,
Iron acquisition by Mycobacterium tuberculosis: WILLIAMS, N. P. (1976), Isolation and identifica-
isolation and characterization of a family of iron- tion of the conidial germination factor of Neu-
binding exochelins, Proc. Natl. Acad. Sci. USA rospora crassa, J. Bacteriol. 127, 135-140.
92,5189-5193. HOSSAIN, M. B., ENG-WILMOT, D. L., LOGHRY, R.
GRIFFITH, G. L., SIGEL,S. P., PAYNE,S. M., NEI- A., VAN DER HELM, D. (1980), Circular di-
LANDS, J. B. (1984), Vibriobactin, a siderophore chroism, crystal structure, and absolute configu-
from Vibrio cholerae, J. Biol. Chem. 259, 383- ration of the siderophore ferric N, N‘, N”-tri-
385. acetylfusarinine, J. Am. Chem. SOC. 102, 5766-
GROSS,R., ENGELBRECHT, BRAUN,V. (1985), 5773.
Identification of the genes and their polypeptide HOSSAIN, M. B., JALAL,M. A. F., BENSON, B. A.,
products responsible for aerobactin synthesis by BARNES,c. L., VAN DER HELM,D. (1987),
ColV plasmids, Mol. Gen. Genet. 201, 204-212. Structure and conformation of two coprogen-
HAAG,H., HANTKE,K., DRECHSEL, H., STROJIL- type siderophores: Neocoprogen I and Neoco-
KOVIC, I., JUNG,G., ZAHNER, H. (1993), Purifi- progen 11, J. Am. Chem. SOC. 109,49484954.
cation of yersiniabactin: a siderophore and pos- HUSCHKA,H., MULLER,G., WINKELMANN, G.
sible virulence factor of Yersiniaenterocolitica,J. (1983), The membrane potential is the driving
Gen. Microbiol. 139,2159-2165. force for siderophore iron transport in fungi,
HALL,R. M., RATLEDGE, C. (1982), Mycobactins FEMS Microbiol. Lett. 20, 125-129.
as chemotaxonomic characters for some rapidly HUSCHKA, H., NAEGELI, H., LEUENBERGER-RYF,
growing mycobacteria, J. Gen. Microbiol. 130, H., KELLER-SCHIERLEIN, W., WINKELMANN,
1883-1892. G. (1985), Evidence for a commQn siderophore
HANTKE,K. (1983), Identification of an iron up- transport system but different siderophore re-
take system specific for coprogen and rhodoto- ceptors in Neurospora crassa, J. Bacteriol. 162,
rulic acid in Escherichia coli K12, Mol. Gen. Ge- 715-721.
net. 191, 301-306. HUSCKA, H., JALAL,M. A. F., VAN DER HELM,D.,
HANTKE, K. (1984), Cloning of the repressor pro- WINKELMANN, G. (1986), Molecular recognition
tein gene of iron regulated systems in E. coli of siderophores in fungi: Role of iron-surround-
K12, Mol. Gen. Genet. 197, 337-341. ing N-acyl residues and peptide backbone during
HANTKE, K., BRAUN,V. (1975), A function com- membrane transport in Neurospora crassa, J.
mon to iron-enterochelin transport and action of Bacteriol. 167, 1020-1024.
colicins B, I, V in Escherichia coli, FEBS Lett. ISHIMARU, C. A., LOPER,J. E. (1992), High-affini-
59,277-281. ty iron uptake systems present in Erwinia caro-
HARRIS,W. R., CARRANO, C. J., RAYMOND, K. N. tovora subsp. carotovora include the hydroxa-
(1979), Coordination chemistry of microbial iron mate siderophore aerobactin, J. Bacteriol. 174,
transport compounds. 16. Isolation, characteriza- 2993-3003.
tion, and formation constants of ferric aerobac- JALAL,M. A. F., VAN DER HELM,D. (1989), Puri-
tin, J. Am. Chem. SOC. 101,2722-2727. fication and crystallization of ferric enterobactin
HARTMANN, A., BRAUN, V. (1980), Iron transport receptor protein, FepA, from outer membranes
in E. coli: uptake and modification of ferri- of Escherichia coli UT5600IpBB2, FEBS Lett.
chrome, J. Bacteriol. 143, 246-255. 243,366-370.
HARTMANN, A., FIEDLER,H.-P., BRAUN,V. JALAL,M. A. F., VAN DER HELM,D. (1991), Isola-
(1979), Uptake and conversion of the antibiotic tion and spectroscopic identification of fungal
8 References 241
siderophores, in: Handbook of Microbial Iron KELLER-SCHIERLEIN, W., HAGMANN, L., ZAH-
Chelates (WINKELMANN, G., Ed.), pp. 235-269. NER, H., HUHN, W. (1988), Stoffwechselpro-
Boca Raton, F L CRC Press. dukte von Mikroorganismen. 167. Maduraferrin,
JALAL,M. A. F., MOCHARLA, R., BARNES,C. L., ein neuartiger Siderophor aus Actinomadura
HOSSAIN, M. B., POWELL, D. R., ENG-WILMOT, madurae, Helv. Chim. Acta 71, 1528-1540.
D. L., GRAYSON, S. L., BENSON,B. A., VAN KILLMANN, H., BRAUN,V. (1992), An aspartate
DER HELM, D. (1984), Extracellular sidero- deletion mutation defines a binding site of the
phores from Aspergillus ochraceus, J. Bacteriol. multifunctional FhuA outer membrane receptor
158,683-688. of Escherichia coli, J. Bacteriol. 174, 3479-3486.
JALAL,M. A. F., GALLES,J. L., VAN DER HELM, KILLMANN, H., BENZ,R., BRAUN,V. (1993), Con-
D. (1985), Structure of des(diserylg1ycyl)ferri- version of the FhuA transport protein into a dif-
rhodin, DDF, a novel siderophore from Asper- fusion channel through the outer membrane of
gillus ochraceus, J. Org. Chem. 50,5642-5645. Escherichia coli, EMBO J. 12,3007-3016.
JALAL,M. A. F., LOVE,S . K., VAN DER HELM,D. KLUYVER, A. J. J., VAN DER WALT,J. P., VAN
(1986,) Siderophore mediated iron(II1) uptake TRIET,A. J. (1953), Pulcherrimin, the pigment
in Gliocladium virens. 1. Properties of cis-fusar- of Candida pulcherrima, Proc. Natl. Acad. Sci.
inine, trans-fusarinine, dimerum acid, and their U.S.A. 39, 583.
ferric complexes, J. Inorg. Biochem. 28, 417- KNOSP,O., TIGERSTROM, M., PAGE,W. J. (1984),
430. Siderophore-mediated uptake of iron in Azoto-
JALAL,M. A. F., LOVE,S. K., VAN DER HELM,D. bacter vinelandii, J. Bacteriol. 159, 341-347.
(1987), Siderophore mediated iron(II1) uptake KONETSCHNY-RAPP, S., HUSCHKA, H., WINKEL-
in Gliocladium virens. 2. Role of ferric mono- MA”, G., JUNG,G. (1988), High-performance
and dihydroxamates as iron transport agents, J. liquid chromatography of siderophores from
Inorg. Biochem. 29,259-267. fungi, BioMetals 1,9-17.
KONETSCHNY-RAPP, S., JUNG,G., MEIWES,J.,
JALAL,M. A. F., HOSSAIN, M. B., VAN DER HELM,
D., SANDERS-LOHR, J., ACTIS,L. A., CROSA,J. ZAHNER,H. (1990), Staphyloferrin A a struc-
H. (1989), Structure of anguibactin, a unique turally new siderophore from staphylococci, Eur.
J. Biochem. 191, 65-74.
plasmid-related bacterial siderophore from the
fish pathogen Vibrio anguillarum, J. Am. Chem. KOSTER,M., VAN DE VOSSENBERG, J., LEONG,J.,
SOC. 111,292-296. WEISBEEK, P. J. (1993), Identification and char-
acterization of the pupB gene encoding an indu-
JEGOROV, A., MATHA,V., HUSAK,M., KRATOCH- cible ferric-pseudobactin receptor of Pseudo-
VIL,B., STUCHLIK, J., SEDMERA, P., HAVLICEK, monas putida WCS358, Mol. Microbiol. 8, 591-
V. (1993), Iron uptake system of some members 601.
of the genus Tolypocladiurn: Crystal structure of KOSTER,W. (1991), Iron(II1) hydroxamate trans-
the ligand and its iron(II1) complex, J. Chem. port across the cytoplasmic membrane of
SOC. Dalton Trans. 1993, 1287-1293.
Escherichia coli, BioMetals 4, 23-32.
KATAYAMA, N., NOZAKI,Y., OKONOGI, K., HA- KUNZE, B., BEDORF,N., KOHL,W., HOFLE,G.,
RADA, S, ONO, H. (1993), Ferrocins, new iron- REICHENBACH, H. (1989), Myxochelin A, a new
containing peptide antibiotics produced by bac- iron-chelating compound from Angiococcus dis-
teria. Taxonomy, fermentation and biological ac- ciformis (Myxobacteriales), J. Antibiot. 42, 14-
tivity, J. Antibiot. 46, 65-70. 17.
KELLER-SCHIERLEIN, W., DIEKMANN, H. (1970), KUNZE, B., TROWITZSCH, W., HOFLE, G., REI-
Zur Konstitution des Coprogens, Helv. Chim. CHENBACH, H. (1992), Nannochelins A, B and
Acta 53, 2035-2044. C, new iron-chelating compounds from Nanno-
KELLER-SCHIERLEIN, W., PRELOG,V., ZAHNER, cystis excedens (Myxobacteria), J. Antibiot. 45,
H. (1964), Siderochrome, in: Forrschritte der 147-1 50.
Chemie organischer Naturstoffe (ZECHMEISTER, KUHN,S., BRAUN,V., KOSTER,W. (1996), Ferric
L., Ed.), pp. 279-322. Heidelberg: Springer-Ver- rhizoferrin uptake into Morganella morganii:
lag. characterization of genes involved in the uptake
KELLER-SCHIERLEIN, W., HUBER,P., KAWAGU- of a polycarboxylate siderophore, J. Bacteriol.
CHI, H. (1984), Chemistry of danomycin, an 178,496-504.
iron-containing antibiotic, in: Natural Products LAMMERS, P. J., SANDERS-LOHR, J. (1982), Active
and Drug Development (KROGSGAARD-LAR- transport of ferric schizokinen in Anabaena sp.,
SEN, P., BROGGERCHRISTENSEN, s., KOFOD, J. Bacteriol. 151, 28&294.
H., Eds.), pp. 213-227. Copenhagen: Munks- LANE,S. J., MARSHALL, P. S., UFTON, R. J., RAT-
gaard. LEDGE, c., EWING, M. (1995), Novel extracellu-
lar mycobactins, the carboxymycobactins from take of Fe3+ in Pseudomonas putida WCS358, J.
Mycobacterium avium, Tetrahedron Lett. 36, Bacteriol. 171, 2819-2826.
41294132. MASSAD,G., ARCENEAUX, J. E. L., BYERS,B. R.
LEE, B. H., MILLER,M. J. (1983), Natural ferric (1994), Diversity of siderophore genes encoding
ionophores: Total synthesis of schizokinen, schi- biosynthesis of 2,3-dihydroxybenzoic acid in
zokinen A, and arthrobactin, J. Org. Chem. 48, Aeromonas spp., BioMetals 7, 227-236.
24-31. MATZANKE, B. F. (1991), Structures, coordination
LEONG,J., NEILANDS,J. B. (1976), Mechanisms of chemistry and functions of microbial iron che-
siderophore iron transport in enteric bacteria, J. lates, in: Handbook of Microbial Iron Chelates
Bacteriol. 126, 823-830. (WINKELMANN, G., Ed.) pp. 15-64. Boca Raton,
LEONG,J., RAYMOND, K. N. (1974), Coordination F L CRC Press.
isomers of biological iron transport compounds. MATZANKE, B. F. (1994), Iron storage in fungi, in:
11. The optical isomers of chromic desferrifer- Metal Ions in Fungi (WINKELMANN, G., WINCE,
richrome and desferriferrichrysin, J. Am. Chem. D. R., Eds.), pp. 179-214. New York: Marcel
SOC.96,66284630. Dekker.
LESUISSE, E., LABBE,P. (1994), Reductive iron as- MATZANKE, B. F., WINKELMANN, G. (1981), Side-
similation in Saccharomyces cerevisiae, in: Metal rophore transport followed by Mossbauer spec-
Ions in Fungi (WINKELMANN, G., WINCE,D. R., troscopy, FEBS Lett. 130, 50-53.
Eds.), pp. 149-178. New York: Marcel Dekker. MATZANKE,B. F., BILL, E., TRAUTWEIN, A. X.,
LINGET,C., STYLIANOU, D. G., DELL,A., WOLFF, WINKELMANN, G. (1987), Role of siderophores
R. E., PIEMONT,Y., ABDALLAH,M. (1992), in iron storage in spores of Neurospora crassa
Bacterial siderophores: the structure of a desfer- and Aspergillus ochraceus, J. Bacteriol. 169,
ribactin produced by a Pseudomonas fluorescens 5873-5876.
ATCC 13525, Tetrahedron Lett. 33,3851-3854. MATZANKE,B. F., BILL,E., TRAUTWEIN, A. X.,
LINKE, W. D., CRUEGER,A., DIEKMANN, H. WINKELMANN, G. (1988), Ferricrocin functions
(1972), Stoffwechselprodukte von Mikroorganis- as the main intracellular iron-storage compound
men. 106. Zur Konstitution des Terregens-Fak- in mycelia of Neurospora crassa, BioMetals 1,
tors, Arch. Microbiol. 85, 44-50. 18-25.
LITWIN, C. M., CALDERWOOD, S. B. (1994), Analy- MATZANKE, B. F., MULLER,G., RAYMOND, K. N.
sis of the complexity of gene regulation by Fur in (1989), Siderophore mediated iron transport, in:
Vibrio cholerae, J. Bacteriol. 176, 240-248. Iron Carriers and Iron Proteins (LOEHR, T.,
LIU, W., FISHER,S., WELLS,J. S., RICCA,C. S., Ed.), pp. 1-121. Weinheim: VCH.
PRINCIPE, P. A., TREJO,W. H., BONNER,D. P., MATZANKE, B. F., BERNER,I., BILL,E., TRAUT-
GOUGOUTOS, J. Z., POEPLITZ,B., SYKES,R. B. WEIN,A. X., WINKELMANN, G. (1991), Trans-
(1981), Siderochelin, a new ferrous-ion chelating port and utilization of ferrioxamine-E-bound
agent produced by Nocardia, J. Antibiot. 34, iron in E. herbicola (Pantoea herbicola), BioMe-
791-799. tals 4,181-185.
LUNDRIGAN, M. D., KADNER,R. J. (1986), Nu- MAURER,P. J., MILLER,M. J. (1982), Microbial
cleotide sequence of the gene for the ferriente- iron chelators: Total synthesis of aerobactin and
rochelin receptor FepA in Escherichia coli, J. its constituent amino acid N6-acetyl-N6-hydroxy-
Biol. Chem. 261, 10797-10801. lysine, J. Am. Chem. SOC. 104,3096-3101.
MACHAM,L. P., RATLEDGE,C., NOCTON,J. C. MAURER,B., MULLER,A., KELLER-SCHIERLEIN,
(1975), Extracellular iron acquisition by myco- W., ZAHNER,H. (1968), Ferribactin, ein Sidero-
bacteria: Role of the exochelins and evidence phor aus Pseudomonas fluorescens Migula,
against the participation of mycobactin, Infect. Arch. Microbiol. 60, 326-339.
Immun. U ,1242-1251. MEI, B.,LEONG,S. A. (1994), Molecular biology of
MARUGG,J. D., WEISBEEK, P. J. (1991), Molecular iron transport, in: Metal Ions in Fungi (WINKEL-
genetics of siderophore biosynthesis in fluores- MA", G., WINGE,D. R., Eds.), pp. 117-148.
cent pseudomonads, in: Handbook of Microbial New York: Marcel Dekker.
Iron Chelates (WINKELMANN, G., Ed.), pp. 155- MEIWES,J., FIEDLER,H., HAAG,H., ZAHNER,H.,
175. Boca Raton, F L CRC Press. KONETSCHNY-RAPP, S., JUNG,G. (1990), Isola-
MARUGG,J. D., DE WEGER,L. A., NIELANDER, tion and characterization of staphyloferrin A, a
H. B., OORTHUIZEN, M., RECOURT,K., LUG- compound with siderophore activity from Sta-
TENBERG, B., VAN DER HOFSTAD,G. A. J. M., phylococcus hyicus DSM 20459, FEMS Micro-
WEISBEEK,P. J. (1989), Cloning and characteri- biol. Lett. 67, 201-206.
zation of a gene encoding an outer membrane MENHARDT, N., THARIATH, A., VISWANATHA, T.
protein required for siderophore-mediated up- (1991), Characterization of the pyoverdins of
8 References 243
Azotobacter vinelandii ATCC 12837 with regard transport and siderophore systems in Rhizobia,
to heterogeneity, BioMetals 4, 223-232. in: Iron Chelation in Plants and Soil Microorgan-
MEYER,J. M. (1992), Exogenous siderophore-me- isms (BARTON,L. L., HEMMING,B. C., Eds.),
diated iron uptake in Pseudomonas aeruginosa: pp. 179-195. San Diego, CA: Academic Press.
possible involvement of porin OprF in iron NEILANDS,J. B., NAKAMURA,K. (1991), Detec-
translocation, J. Gen. Microbiol. 138, 951-958. tion, determination, isolation, characterization
MEYER,J. M., ABDALLAH,M. A. (1980), The side- and regulation of microbial iron chelates, in:
rochromes of non-fluorescent pseudomonads: Handbook of Microbial Iron Chelates (WINKEL-
production of nocardamine by Pseudomonas MANN,G., Ed.), pp. 1-14. Boca Raton, FL: CRC
stutzeri, J. Gen. Microbiol. 118, 125-129. Press.
MEYER,J. M., HOHNADEL,D., HALLB,F. (1989), NEILANDS,J. B., ERICSON,T. J., RASTETTER,W.
Cepabactin from Pseudomonas cepacia, a new H. (1981), Stereospecificity of the ferric entero-
type of siderophore, J. Gen. Microbiol. 135, bactin receptor of Escherichia coli K-12, J. Biol.
1479-1 487. Chem. 256,3831-3832.
MEYER, J.-M., TRAN VAN, V., STINTZI, A., NIKAIDO, H., ROSENBERG,E. Y. (1990). Cir and
BERGE,O., WINKELMANN, G. (1995), Ornibac- Fiu proteins in the outer membrane of Escher-
tin production and transport properties in strains ichia coli catalyze transport of monomeric cate-
of Burkholderia vietnamiensis and Burkholderia chols: Study with p-lactam antibiotics containing
cepacia (formerly Pseudomonas cepacia), Bio- catechol and analogous groups, J. Bacteriol. 172,
Metals 8, 309-317. 1361-1367.
MOORE,R. E., EMERY, T. (1976), N'-Acetylfusar- NISHIO,T., TANAKA,N., HIRATAKE,JuN., KAT-
inines: isolation, characterization, and proper- SUBE, Y., ISHIDA, Y., ODA, J. (1988), Isolation
ties, Biochemistry 15, 2719-2722. and structure of the novel dihydroxamate side-
MOORE,C. H., FOSTER, L.-A., GERBIGJR., D. G., rophore alcalignin, J. Am. Chem. SOC. 110,8733-
DYER,D. W., GIBSON,B. (1995), Identification 8734.
of alcaligin as the siderophore produced by Bor- NOMOTO, K., SUGIURA,Y., TAKAGI, S. (1987),
detella pertussis and B. bronchiseptica, J. Bacteri- Mugineic acids, studies on phytosiderophores,
01. 7, 1116-1118. in: Iron Transport in Microbes, Plants and Ani-
MOR, H., KASHMAN,Y., WINKELMANN, G., BA- mals (WINKELMANN, G., VAN DER HELM, D.,
RASH, I. (1992), Characterization of sidero- NEILANDS,J. B., Eds.), pp. 401-425, Weinheim:
phores produced by different species of the der- VCH.
matophytic fungi Microsporum and Trichophy- OKUJO,N., SAITO,M., YAMAMOTO,S., YOSHIDA,
ton, BioMetals 5, 213-216. T., MIYOSHI,S., SHINODA,S. (1994a), Structure
MULLER,G., RAYMOND,K. N. (1984), Specificity of vulnibactin, a new polyamine-containing side-
and mechanism of ferrioxamine-mediated iron rophore from Vibrio vulnificus, BioMetals 7,
transport in Streptomyces pilosus, J. Bacteriol. 109-116.
160,304-312. OKUJO,N., SAKAKIBARA, Y., YOSHIDA,T., YA-
MULLER,G., MATZANKE,B. F., RAYMOND,K. N. MAMOTO, S. (1994b), Structure of acinetofferin,
(1984), Iron transport in Streptomyces pilosus a new citrate-based dihydroxamate siderophore
mediated by ferrichrome siderophores, rhodoto- from Acinetobacter haemolyticus, BioMetals 7 ,
rulic acid, and enantio-rhodotorulic acid, J. Bac- 170-176.
teriol. 160, 313-318. ONG,S. A., PETERSON,T., NEILANDS, J. B. (1979),
MULLIS,K. B., POLLACK,J. R., NEILANDS,J. B. Agrobactin, a siderophore from Agrobacterium
(1971), Structure of schizokinen, an iron trans- tumefaciens, J. Biol. Chem. 254, 1860-1865.
port compound from Bacillus megaterium, Bio- OXFORD, A. E., RAISTRICK,H. (1948), Studies on
chemistry 10, 4894-4898. the biochemistry of microorganisms. 76. Myce-
NAEGELI, H., KELLER-SCHIERLEIN, W. (1978), lianamide, a metabolic product of Penicillium
Stoffwechselprodukte von Mikroorganismen. griseofulvum. Part I. Preparation, properties and
Eine neue Synthese des Ferrichroms; enantio- breakdown products, Biochem. J. 42,323.
Ferrichrom, Helv. Chim. Acta 61, 2088-2095. PAGE, W. J., HUYER,M. (1984), Derepression of
NAEGELI,H., ZAHNER,H. (1980), Stoffwechsel- the Azotobacter vinelandii siderophore system
produkte von Mikroorganismen. Ferrithiocin, using iron-containing minerals to limit iron re-
Helv. Chim. Acta 63, 1400-1406. pletion, J. Bacteriol. 158, 496-502.
NEILANDS,J. B. (1952), A crystalline organo-iron PAGE,W. J., TIGERSTROM, M. (1988), Aminoche-
pigment from a rust fungus (Ustilago sphaero- lin, a catecholamine siderophore produced by
gena), J. Am. Chem. SOC. 74,4846-4847. Azotobacter vinelandii, J. Gen. Microbiol. 134,
NEILANDS, J. B. (1993), Overview of bacterial iron 453-460.
PAYNE,S. M. (1980), Synthesis and utilization of Pasteurella haemolytica, J. Basic Microbiol. 34,
siderophores by Shigella flexneri, J. Bacteriol. 61-63.
143, 1420-1424. RUTZ, J. M., LIU, J., LYONS,J., GORANSON, J.,
PERSMARK, M., NEILANDS,J. B. (1992), Iron(II1) ARMSTRONG, S. K., MCINTOSH, M. A., FEIX,J.
complexes of chrysobactin, a siderophore of Er- B., KLEBBA, P. E. (1993), Formation of a gated
winia chrysanthemi, BioMetals 5, 29-36. channel by a ligand-specific transport protein in
PERSMARK,M., EXPERT, D., NEILANDS,J. B. the bacterial outer membrane, Science 258,471-
(1989), Isolation, characterization, and synthesis 475.
of chrysobactin, a compound with siderophore SAYER,J. M., EMERY,T. (1968), Structures of the
activity from Erwinia chrysanthemi, J. Biol. naturally occurring hydroxamic acids, fusarin-
Chem. 264,3187-3193. ines A and B, Biochemistry 7, 184190.
PERSMARK, M., EXPERT,D., NEILANDS,J. B. SCHNEIDER, R., HANTKE,K. (1993), Iron-hydroxa-
(1992), Ferric iron uptake in Erwinia chrysan- mate uptake systems in Bacillus subtilis: identifi-
themi mediated by chrysobactin and related ca- cation of a lipoprotein as part of a binding pro-
techol-type compounds, J. Bacteriol. 174, 4783- tein-dependent transport system, Mol. Micro-
4789. biol. 8, 111-121.
PERSMARK,M., PITTMAN,P., BUYER,J. S., SCHWYN,B., NEILANDS,J. B. (1987), Universal
SCHWYN,B., GILL, P. R., NEILANDS,J. B. chemical assay for the detection and determina-
(1993), Isolation and structure of rhizobactin tion of siderophores, Anal. Biochem. 160, 47-
1021, a siderophore from the alfalfa symbiont 56.
Rhizobium meliloti 1021, J. Am. Chem. SOC.115, SHANZER, A., LIBMAN, J. (1991), Biomimetic side-
3950-3956. rophores, in: Handbook of Microbial Iron Che-
PETERSON,T., NEILANDS,J. B. (1979), Revised lates (WINKELMANN, G., Ed.), pp. 309-338.
structure of a catecholamide spermidine sidero- Boca Raton, FL: CRC Press.
phore from Paracoccus denitrificans, Tetrahe- SHANZER,A., LIBMAN, J., LAZAR,R., TOR, Y.,
dron Lett. 50,48054808. EMERY,T. (1988), Synthetic ferrichrome ana-
RABSCH,W., WINKELMANN, G. (1991), The specif- logues with growth promotion activity for Ar-
icity of bacterial siderophore receptors probed throbacter flavescens, Biochem. Biophys. Res.
by bioassays, BioMetals 4, 244-250. Commun. 157,389-394.
RATLEDGE,C. (1987), Iron metabolism in myco-
SHARMAN, G., WILLIAMS, D. H., EWING,D. F.,
bacteria, in: Iron Transport in Microbes, Plants
RATLEDGE,C. (1995a), Isolation, purification
and Animals (WINKELMANN, G., VAN DER and structure of exochelin MS, the extracellular
HELM,D., NEILANDS, J. B., Eds.), pp. 207-233, siderophore from Mycobacterium smegmatis,
Weinheim: VCH.
Biochem. J. 305, 187-196.
RATLEDGE, C., EWING,M. (1996), The occurrence
of carboxymycobactin, the siderophore of pa- SHARMAN, G., WILLIAMS, D. H., EWING,D. F.,
thogenic mycobacteria, as a second extracellular RATLEDGE,C. (1995b), Determination of the
siderophore in Mycobacterium smegmatis, Mi- structure of exochelin MN the extracellular side-
crobiology 142, 2207-2212. rophore from Mycobacterium neoaurum, Chem.
RATLEDGE,C., MARSHALL, B. J. (1972), Iron Biol. 2, 553-561.
transport in Mycobacterium smegmatis: the role SIGEL, S. P., STOEBNER,J. A., PAYNE,S. M.
of mycobactin, Biochim. Biophys. Acta 279, 58- (1985), Iron-vibriobactin transport system is not
74. required for virulence of Vibrio cholerae, Infect.
RATLEDGE, C., PATEL,P. V. (1976), The isolation, Immun. 47,360-362.
properties and taxonomic relevance of lipid-sol- SMITH, M. J., SHOOLERY, J. N., SCHWYN,B.,
uble iron-binding compounds (the nocobactins) HOLDEN,I., NEILANDS,J. B. (1985), Rhizobac-
from Nocardia, J. Gen. Microbiol. 93, 141-152. tin, a structurally novel siderophore from Rhizo-
REID,R. T., LIVE,D. H., FAULKNER, D. J., BUT- bium meliloti, J. Am. Chem. SOC. 107, 1739-
LER,A. (1993), A siderophore from a marine 1743.
bacterium with an exceptional ferric ion affinity SNOW,G. A. (1965), Isolation and structure of my-
constant, Nature 366,455-458. cobactin T, a growth factor from Mycobacterium
REISSBRODT, R., RABSCH,W., CHAPEAUROUGE, tuberculosis, Biochem. J. 97, 166-175.
A., JUNG,G., WINKELMANN, G. (1990), Isola- SNOW,G. A. (1970), Mycobactins: iron chelating
tion and identification of ferrioxamine G and E growth factors from Mycobacteria, Microbiol.
in Hafnia alvei, BioMetals 3, 54-60. Rev. 34, 99-125.
REISSBRODT, R., ERLER,W., WINKELMANN, G. SOUKKA, T., TENOVUO, J., LENANDER-LUMIKARI,
(1994), Iron supply of Pasteurella multocida and M. (1992), Fungicidal effect of human lactoferrin
8 References 245
against Candida albicans, FEMS Microbiol. Lett. TSUBOTANI, S., KATAYAMA, N., FUNABASHI, Y.,
90,223-228. ONO, H., HARADA,S. (1993), Ferrocins, new
STEPHAN,H., FREUND,S., BECK,W., JUNG,G., iron-containing peptide antibiotics produced by
WINKELMANN, G. (1993a), Ornibactins - a new bacteria. Isolation, characterization and struc-
family of siderophores from Pseudomonas, ture elucidation, J. Antibiot. 46, 287-293.
BioMetals 6, 93-100. VAN DER HELM,D., POLING, M. (1976), The crys-
STEPHAN,H., FREUND,S., MEYER,J.-M., WIN- tal structure of ferrioxamine E, J. Am. Chem.
KELMANN, G., JUNG,G. (1993b), Structure elu- SOC. 98, 82-86.
cidation of the gallium-ornibactin complex by VAN DER HELM,D., BAKER, J. R., ENG-WILMOT,
2D-NMR spectroscopy, Liebigs Ann. Chem. D. L., HOSSAIN,M. B., LOGHRY,R. A. (1980),
1993,4348. Crystal structure of ferrichrome and comparison
STEPHENSON, M. C., RATLEDGE,C. (1979), Iron with the structure of ferrichrome A, J. Am.
transport in Mycobacterium smegmatis: Uptake Chem. SOC. 102, 42244231.
of iron from ferriexochelin, J. Gen. Microbiol. VAN DER HELM,D., WINKELMANN, G. (1994), Hy-
110,193-202. droxamates and polycarboxylates as iron trans-
STOEBNER, J. A., PAYNE,S. M. (1988), Iron-regul- port agents (siderophores) in fungi, in: Metal
ated hemolysin production and utilization of Ions in Fungi (WINKELMANN, G., WINGE,D. R.,
heme and hemoglobin by Vibrio cholerae, Infect. Eds.), pp. 39-98. New York: Marcel Dekker.
Immun. 56,2891-2895. VAN DER WALT, J. P., BOTHA,A., EICKER,A.
STOEBNER,J. A., BUTTERTON, J. R., CALDER- (1990), Ferrichrome production by Lipomyceta-
WOOD,S. B., PAYNE, S. M. (1992), Identification ceae, Syst. Appl. Microbiol. 13, 131-135.
of the vibriobactin receptor of Vibrio cholerae, J. VAN TIEL-MENKVELD,G.J., MENTJOX-VER-
Bacteriol. 174, 3270-3274. VUURT,M., OUDEGA,B., DEGRAAF, F. K.
STOJILJKOVIC, I., HANTKE,K. (1992), Hemin up- (1982), Siderophore production by Enterobacter
take system of Yersinia enterocolitica:similarities cloacae and a common receptor protein for the
with other TonB-dependent systems in gram-ne- uptake of aerobactin and cloacin DF13, J. Bacte-
gative bacteria, EMBO J. 11, 4359-4367. riol. 150, 490-497.
TAKAHASHI, A., NAKAMURA, H., KAMEYAMA, T., VANDEWOESTYNE,M., BRUYNEEL, B., MER-
KURASAWA, S., NAGANAWA, H., OKAMI,Y., GEAY,M., VERSTRAETE, W. (1991), The Fez+
TAKEUCHI, S, UMEZAWA, H. (1987), Bisucaber- chelator proferrorosamine A is essential for the
in, a new siderophore, sensitizing tumor cells to siderophore-mediated uptake of iron by Pseudo-
macrophage-mediated cytolysis. 11. Physico- monas roseus fluorescens, Appl. Environ. Micro-
chemical properties and structure determination, biol. 57, 949-954.
J. Antibiot. 40,1671-1676. VISCA,P., SERINO,L., ORSI, N. (1992), Isolation
TARAZ,K., EHLERT,G., GEISEN,K., BUDZIKIE- and characterization of Pseudomonas aeruginosa
WICZ,H., KORTH, H., PULVERER,G. (1990), mutants blocked in the biosynthesis of pyover-
Protochelin, ein Catecholat-Siderophor aus ei- din, J. Bacteriol. 174, 5727-5731.
nem Bakterium (DSM 5746), 2. Naturforsch. VISCA,P., CIERVO,A., ORSI,N. (1994), Cloning
45b,1327-1332. and nucleotide sequence of the pvdA gene en-
TEINTZE,M., HOSSAIN,M. B., BARNES,C. L., coding the pyoverdin biosynthetic enzyme L-or-
LEONG,J., VAN DER HELM,D. (1981), Structure nithine N5-oxygenasein Pseudomonas aerugino-
of ferric pseudobactin, a siderophore from a sa, J. Bacteriol. 176, 1128-1140.
plant-deleterious Pseudomonas, Biochemistry WACHSMUTH, I. K., BLAKE,P. A., OLSVIK,0.
20,6446-6457. (Eds.) (1994), Vibrio cholerae and Cholera -
TELFORD,J. R., LEARY,J. A., TUNSTAD, L. M. G., Molecular to global perspectives. Washington,
BYERS,B. R., RAYMOND, K. N. (1994), Amona- DC: Am. SOC.Microbiol.
bactin: Characterization of a new series of side- WEE, S., NEILANDS, J. B., BITTNER,M. L., HEM-
rophores from Aeromonas hydrophila, J. Am. MING,B. c.,HAYMORE, B. L., SEETHARAM,R.
Chem. SOC. 116,44994500. (1988), Expression, isolation and properties of
THIEKEN, A., WINKELMANN, G. (1992), Rhizofer- Fur (ferric uptake regulation) protein of Esche-
rin: A complexone type siderophore of the Mu- richia coli K12, BioMetals 1, 62-68.
corales and Entomophthorales, FEMS Micro- WIEBE,C., WINKELMANN, G. (1975), Kinetic stud-
biol. Lett. 94, 37-42. ies on the specificity of chelate iron uptake in
THIEKEN, A., WINKELMANN, G. (1993), A novel Aspergillus, J. Bacteriol. 123, 837-842.
bioassay for the detection of siderophores con- WINKELMANN, G. (1979), Evidence for stereospe-
taining keto-hydroxy bidentate ligands, FEMS cific uptake of iron 14 chelates in fungi, FEES
Microbiol. Lett. 111, 281-286. Lett. 97, 43-46.
WINKELMANN, G. (1986), Iron complex products gen. nov. and transfer of seven species of the
(siderophores), in: Biotechnology, 1st Edn., Vol. genus Pseudomonas homology group I1 to the
4, (REHM,H.-J., REED,G., Eds), pp. 215-243. new genus, with the type species Burkholderia
Weinheim: VCH. cepacia (PALLERONI and HOLMS1981) comb.
WINKELMANN, G. (Ed.) (1991a), Handbook of Mi- nov., Microbiol. Immunol. 36, 1251-1275.
crobial Iron Chelates. Boca Raton, F L CRC YAMAMOTO, S., CHOWDHURY, M. A. R., KURO-
Press. DA, M., et al. (1991), Further study on polyam-
WINKELMANN, G. (1991b), Importance of sidero- ine compositions in Vibrionaceae, Can. J. Micro-
phores in fungal growth, sporulation and spore biol. 37, 148-153.
germination, in: Frontiers in Mycology (HAWK- YAMAMOTO, S., FUJITA,Y., OKUJO,N., MINAMI,
SWORTH,D. L., Ed.), pp. 49-65. Wallingford C., MATSUURA,S., SHINODA, S. (1992), Isola-
CAB International. tion and partial characterization of a compound
WINKELMANN, G. (1991c), Specificity of iron trans- with siderophore activity from V. parahaemolyti-
port in bacteria and fungi, in: Handbook of Mi- cus, FEMS Microbiol. Lett. 94, 181-186.
crobial Iron Chelates (WINKELMANN, G., Ed.), YAMAMOTO, S., OKUIO,N., FUJITA,Y., SAITO,M.,
pp. 65-105. Boca Raton, F L CRC Press. YOSHIDA,T., SHINODA, S. (1993), Structures of
WINKELMANN, G. (1992), Structures and functions two polyamine-containing catecholate sidero-
of fungal siderophores containing hydroxamate phores from Vibrio fluvialis, J. Biochem. 113,
and complexone type iron binding ligands, My- 538-544.
col. Res. %, 529-534. YAMAMOTO,S., OKUJO, N., SAKAKIBARA, Y.
WINKELMANN, G. (1993), Kinetics, energetics, and (1994a), Isolation and structure elucidation of
mechanisms of siderophore iron transport in acinetobactin, a novel siderophore from Acine-
fungi, in: Iron Chelation in Plants and Soil Mi- tobacter baumannii, Arch. Microbiol. 249-254.
croorganisms (BARTON,L. L., HEMMING, B. C., YAMAMOTO, S., OKUJO,N., YOSHIDA,T., MAT-
Eds.), pp. 219-239. Boca Raton, F L CRC SUURA,S., SHINODA,S. (1994b), Structure and
Press. iron transport activity of Vibrioferrin, a new si-
WINKELMANN, G., BRAUN,V. (1981), Stereoselec- derophore from Vibrio parahaemolyticus, J. Bio-
tive recognition of ferrichrome by fungi and bac- chem. 115, 868-874.
teria, FEMS Microbiol. Lett. 11, 237-241. YOUNG,I. G., GIBSON,F. (1979), Isolation of ente-
WINKELMANN, G., WINCE, D. R. (Eds.) (1994), robactin from Escherichia coli, Methods Enzy-
Metal Ions in Fungi. New York: Marcel Dek- mol. 56,394-398.
ker.
WINKELMANN, G., ZAHNER,H. (1973), Stoffwech- ZAHNER,H., DIDDENS,H., KELLER-SCHIERLEIN,
selprodukte von Mikroorganismen. 115. Eisen- W., NAEGELI,H. (1977), Some experiments with
aufnahme bei Neurospora crassa. Zur Spezifitat semisynthetic sideromycin, J. Antibiot. 30
des Eisentransportes, Arch. Microbiol. 88, 49- (Suppl.), s-201-206.
60. ZALKIN, A., FORRESTER, J. D., TEMPLETON, D. H.
WINKELMANN, G., VAN DER HELM, D., NEI- (1966), Ferrichrome A tetrahydrate. Determina-
LANDS,J. B. (Eds.) (1987), Iron Transport in Mi- tion of crystal and molecular structure, J. Am.
crobes, Plants and Animals. Weinheim: VCH. Chem. SOC. 88, 1810-1814.
WINKELMANN, G., CANSIER,A., BECK,W., JUNG, ZHOU, x. H., VAN DER HELM,D., VENKATRA-
G. (1994), HPLC separation of enterobactin and MANI,L. (1995), Binding characterization of the
linear 2,3-dihydroxybenzoylserinederivatives: a iron transport receptor from the outer mem-
study on mutants of Escherichia coli defective in brane of Escherichia coli (FepA): differentiation
regulation (fur), esterase (fes) and transport between FepA und FecA, BioMetals 8, 129-
(FepA),BioMetals 7 , 149-154. 136.
WONG,G. B., KAPPEL,M. J., RAYMOND, K. N., ZYCHA,H., SIEPMANN, R. (Eds.) (1969), Muco-
MATZANKE,B., WINKELMANN, G. (1983), rales - Eine Beschreibung aller Gattungen und
Coordination chemistry of microbial iron trans- Arten dieser Pilzgruppe. Mit einem Beitrag zur
port compounds. 24. Characterization of copro- Gattung Mortierella von G. Linnemann, Lehre:
gen and ferricrocin, two ferric hydroxamate side- Verlag von Cramer.
rophore, J. Am. Chem. SOC.105,810-815. ZYWNO,S. R., ARCENEAUX, J. E. L., ALTWEG,M.,
YABUUCHI, E., KOSAKO,Y., OYAIZU,H., YANO, BYERS,B. R. (1992), Siderophore production
I., HOTTA,H., HASHIMOTO, Y., EAZAKI,T., and DNA hybridization groups of Aeromonas
ARAKAWA, M. (1992), Proposal of Burkholderia spp., J. Clin. Microbiol. 30, 619-622.
Biotechnology
6 Advances in the Molecular

Genetics of fl=LactamAntibiotic
Biosynthesis
PAULL. SKATRUD
Indianapolis, IN, USA
SCHWECKE
TORSTEN
Cambridge, UK
HENKVAN LIEMPT
Bonn, Germany
MATTHEWB. TOBIN
Indianapolis, IN, USA
1 Introduction 249
2 Organisms which Produce p-Lactam Antibiotics 250
3 Biosynthesis of @Lactam Antibiotics 251
4 Development of Genetic Transformation Systems 253
5 Enzymes Involved in Penicillin Biosynthesis 253
5.1 &(L-a-Aminoadipy1-L-Cysteinyl-D-ValineSynthetase (ACVS) 254
5.2 Isopenicillin N Synthase (IPNS) 256
5.3 Acyl-Coenzyme A:Isopenicillin N Acyltransferase (AT) 257
6 Cloning of the Genes Involved in the Biosynthesis of Penicillin G, Cephalosporin C, and
Cephamycin C 259
7 Clustering of /I-Lactam Biosynthetic Genes 259
248 6 Advances in the Molecular Genetics of P-Lactam Antibiotic Biosynthesis
8 Compartmentalization of P-Lactam Biosynthetic Enzymes 261

9 Evolution of the &Lactam Biosynthetic Pathway: The Horizontal Transfer Hypothe-
sis 262
10 p-Lactam Antibiotics and Recombinant DNA Technology: Practical Applications 264
11 Regulation of &Lactam Biosynthetic Gene Expression 265
12 Summary 267
13 References 267
I Introduction 249
1 Introduction inactive penicilloyl enzyme resistant to hydro-

lysis or aminolysis.
Why the inhibition of cell wall biosynthesis
should ultimately lead to cell death is a com-
120 years ago scientists observed growth in- plex question (reviewed by SHOCKMAN et al.,
hibition when certain microbes were cultured 1982). Peptidoglycan assembly is performed
together and they recognized the potential by several specific enzymes forming glyco-
therapeutic value of this antagonism (PA- sidic, amide, and peptide bonds, and as such
STEUR and JOUBERT,1877). Half a century requires some kind of coordinate regulation
later a similar observation in the British labo- of these enzymes. Early studies conducted by
ratory of SIR ALEXANDER FLEMINGpiqued LEDERBERG(1957) assumed that inhibition
interest in this phenomenon (FLEMING, of cell wall biosynthesis led directly to a
1929). Over a decade later, these discoveries shortage of building blocks, so that as grow-
and other observations led SIR EDWARD ing cells increased in size, the limited or
ABRAHAMand his colleagues to the first weakened cell wall was unable to contain the
amazingly successful clinical use of penicillin contents and lysed. However, more recent ob-
in 1941 (ABRAHAM et al., 1941). In the ensu- servations indicate that not all bacterial spe-
ing years, the clinical use of plactam antibiot- cies undergo lysis after treatment with cell
ics (i.e., penicillins, cephalosporins, and ceph- wall inhibitors such as penicillins. Indeed, the
amycins) became the most successful foray of historically accepted dogma concerning that
mankind in his on-going battle against infec- bacteriolysis is the direct result of penicillin
tious diseases. Indeed, it has been suggested treatment of susceptible bacteria has recently
that in the United States the use of antibiotics been challenged. GIESBRECHTet al. (1993)
has added approximately ten years to life ex- were able to distinguish between the onset of
pectancy (MCDERMOTTand ROGERS,1982). killing and the onset of bacteriolysis after
Due to their intrinsically high therapeutic in- treatment of S. aureus with p-lactam antibiot-
dex and clinical success, plactam antibiotics ics, suggesting that the penicillin-induced bac-
have been and continue to be the most in- teriolysis of sensitive bacteria is not the cause
tensely used class of antibacterial compound. of death, but rather its consequence.
The mode of action of p-lactam antibiotics The unparalleled clinical success of p-lac-
on bacteria has been extensively studied since tam antibiotics stimulated intense interest and
the identification of penicillin (reviewed by research into these compounds as well as the
WAXMANand STROMINGER,1982). WISE organisms that produce them. For decades,
and PARK(1965) and TIPPERand STROMIN- industrial research was focused on improving
GER (1965) demonstrated that penicillin inhi- overall productivity and on understanding the
bits the transpeptidation reaction required for biosynthesis of penicillin G in Penicillium
peptidoglycan crosslinking, and hypothesized chrysogenum and of cephalosporin C in
that penicillin acts as a structural analog of Cephalosporium acremonium (syn. Acremon-
the acyl-D-ala-D-ala terminus of the stem pep- ium chrysogenum). Strain improvement stud-
tide (Park nucleotide) found in peptidogly- ies utilized natural selection and extensive
can. Other studies have demonstrated that mutagenesis programs coupled with biochem-
penicillin also binds at the same active-site ical selections. These strain improvement
serine in two bacterial D-alanine carboxypep- studies resulted in strains vastly improved in
tidases (YOCUMet al., 1979). Regeneration of their capacity to produce p-lactam antibiotics.
these enzymes requires cleavage of the acyl- By 1985, the central reaction pathways for the
enzyme complex by hydrolysis (of carboxy- biosynthesis of p-lactams had been described
peptidases) or transaminolysis (of transpep- (O'SULLIVANand SYKES,1983). Most of the
tidases). The scissile bond of the p-lactam intermediates had been isolated and the order
ring replaces the natural D-ala-D-ala substrate of reactions could be assigned (QUEENER
at the enzyme active site. Acylation of the ca- and NEUSS,1982) (see Fig. 1). The stage was
talytically active amino acid would occur with set for the complete elucidation of biosyn-
the opening of the plactam ring, forming an thetic pathway mechanisms, including the cat-
alytic entities and the corresponding genes. in understanding the molecular biology of p
Specifically the production of modified P-lac- lactam biosynthesis as well as the genetic ma-
tam antibiotics seemed feasible and this nipulation and characterization of organisms
spurred research in both industrial groups producing these compounds. The reader is re-
and academic centers. Initial recombinant ferred to several of these reviews for more ex-
DNA studies on plactam producing fungi tensive coverage of topics discussed in this
were done in the industrial setting (QUEENER chapter (MILLERand INGOLIA,1989; INGOL-
et al., 1985; PENALVA et al., 1985) and as col- IA and QUEENER,1989; SKATRUD,1991,
laborative studies between industry (Eli Lilly 1992; MARTfN et al., 1991; AHARONOWITZ et
and Company, U.S.) and academia (Oxford al., 1992; VICHITSOONTHONKUL et al., 1994;
University, U.K.) (SAMSONet al., 1985). MART~Nand GUTIERREZ,1995). On the fol-
Since that time both communities have con- lowing pages, the developments in these lines
tributed heavily to the expansion of our of research are highlighted and examples of
knowledge in this arena. In the years since the use of these advances in basic knowledge
1983, two lines of research have come togeth- for the improvement of antibiotic production
er to prepare the way for a rationally de- are given.
signed exploitation of plactam antibiotic pro-
ducing organisms. The first major step was
the identification and cloning of the plactam
biosynthetic genes which engendered a means
to influence the catalytic properties of the 2 Organisms which
corresponding enzymes. And second, the de-
velopment of transformation protocols for Produce P-Lactam
achieving stable expression of foreign DNA
in P-lactam producing strains led to an en- Antibiotics
compassing array of methods for drug design
and production. A variety of prokaryotic and eukaryotic mi-
Several reviews have been written in the croorganisms are able to synthesize bicyclic
past few years which address progress made and monocyclic p-lactams (Tab. 1). Of these
Tab. 1. Representative P-Lactam-Producing Organisms (modified from TURNER,

1992)
@-Lactam Structure Fungi Bacteria
Actinomycetes Eubacteria
Penam Aspergillus nidulans

Penicillium chrysogenum
Epidermophyton
Trichophyton
Polypaecilum
Malbranchea
Microsporum
Cephem Cephalosporium acremonium Streptomyces Flavobacterium
Paecilomyces Nocardia Lysobacter
Scopulariopsis Xanthomonas
Diheterospora
Spiroidium
Clavam Streptomyces Erwinia
Serratia
Carbapenem Streptomy ces
3 Biosynthesis of p-Lactam Antibiotics 251
compounds, the bicyclic p-lactams possess the ase (IPNS) enzymes, respectively. In Nocar-
most potent antimicrobial activity. Thus far, dia lactamdurans and S. clavuligerus, a-ami-
only P. chrysogenum and C. acremonium noadipic acid is formed from L-lysine by ly-
have been used extensively at the industrial sine-E-aminotransferase, encoded by the lat
scale for production of bicyclic p-lactam com- gene. In the fungi, a-aminoadipate is re-
pounds. Despite their relatively potent anti- cruited from L-lysine biosynthesis. Subse-
microbial activity, only a minority of these quent to the formation of isopenicillin N
compounds are directly used in the clinical (IPN), the penicillin and cephalosporin/ceph-
setting. However, chemical modification of amycin biosynthetic pathways diverge. In P.
bicyclic p-lactam fermentation products has chrysogenum and A. nidulans, hydrophobic
produced a wide array of clinically useful penicillins (e.g., penicillin G) are formed by a
semi-synthetic p-lactams. For the purposes of transacylation reaction catalyzed by acyl-
this chapter we will focus on the biosynthesis, coenzyme A:isopenicillin N acyltransferase
molecular genetics, and enzymology of bicy- (AT), encoded by the penDE gene. In cepha-
clic p-lactams. losporin/cephamycin producing organisms
(e.g., C. acremonium, N. lactamdurans, and
several strains of Streptomycetes), IPN is con-
verted to penicillin N by an epimerase, en-
coded by the cefD gene. The thiazolidine ring
3 Biosynthesis of of the penicillin is then ring-expanded to form
deacetoxycephalosporin C (DAOC), contain-
P-Lactam Antibiotics ing the dihydrothiazine ring characteristic of
the cephalosporins. DAOC is subsequently
A clear understanding of p-lactam biosyn- hydroxylated at the C-3 position, forming
thesis (QUEENERand NEUSS,1982) enabled deacetylcephalosporin C (DAC). In the fil-
investigators to efficiently dissect and modify amentous fungus C. acremonium, the cefEF
this pathway through the use of molecular gene product (DAOC/DAC synthase, expan-
biology. All p-lactams contain the four-mem- dase/hydroxylase, REX/H) catalyzes both of
bered azetidinone ring system in which anti- these reactions. In the prokaryotes S. clavuli-
microbial activity resides (Fig. 1). Bicyclic @- gerus and N. lactamdurans, these two reac-
lactam antibiotics fall into two natural classes: tions are catalyzed by separate enzymes,
penicillins with the second ring system being DAOC synthase (REX, expandase) and
a five-membered thiazoladine ring and cepha- DAC synthase (hydroxylase), encoded by two
losporins/cephamycins in which the second separate genes which bear striking homology
ring system is a six-membered dihydrothia- to one another (cefE and cefc respectively).
zine ring. The biosynthetic pathways leading At this stage, the biosynthetic pathways of ce-
to the production of penicillin G in P. chryso- phalosporins and cephamycins diverge. In C.
genum, cephalosporin C in C. acremonium, acremonium, DAC is converted to cephalos-
and cephamycin C in Streptomyces clavulige- porin C, the end product of cephalosporin
rus are illustrated in Fig. 1 as a branched biosynthesis, by an acetyltransferase reaction
pathway. Designations for the genes encoding at the C-3 position. The enzyme catalyzing
enzymes of this pathway were established by this reaction is encoded by the cefG gene. The
INGOLIAand QUEENER(1989) and SKA- formation of cephamycin C in S. clavuligerus
TRUD (1991); pcb refers to genes common to involves further modification of DAC. A 3'-
all three end products, pen defines genes hydroxymethylceph-3-em-0-carbamoyltrans-
found only in penicillin biosynthesis, cef ref- ferase, encoded by the cmcH gene, catalyzes
ers to genes found only in cephalosporin C the carbamoylation at the C-3 position of
biosynthesis, and cmc denotes genes specific DAC to form 0-carbamoyldeacetylcephalos-
for cephamycin C biosynthesis. The so-called porin C. The last two steps in cephamycin C
early genes (pcbAB and pcbC) encode the 6 biosynthesis involve 7 '-a-hydroxylase and
(L-a-aminoadipoy1)-L-cysteinyl-D-valinesyn- methyltransferase reactions, encoded by the
thetase (ACVS) and isopenicillin N synth- cmcl and cmcJ genes, respectively.
252 6 Advances in the Molecular Genetics of p-Lactam Antibiotic Biosynthesis
L-alpha-aminoadipic acid + L-cysieine + L-valine
LLDACV
I ACV SYNTHETASE W A B )
1
4-membered CqH
0
I S O F ' E N I C I L ~ ~ ~ f l T H E T A(IPNS)
SE
(a)
6 A P A ACYLCoA
ACYL'IRANSlTRASE p h c n y a c e ty l C o A
CqH
I
PENICILLIN N
0
F'ENICILLING 0 DAOCS (EXF'ANDASE)
W F in C.
uacl?roXY-
CQWALCSFCfUNC
OAW
( a E F in C. acremonium)
DE4cliTYL- ( U F in S.
C
-
OAO
3-HYDROXYMETHYLCEPH-3-EM
CEPHALOSPORIN C SYNIRETASE
II
(3-3
CEPHALOSPORWC
C ~ H
CqH
Fig. 1. Biosynthesis of p-lactam antibiotics.

4 Development of Genetic cin B phosphotransferase (HPT) gene signifi-

cantly increased the rate of transformation
Transformation Systems for C. acremonium (SKATRUD et al., 1987b).
Transformation occurs by integration rather
than replication - in contrast to S. cerevisiae
The application of recombinant DNA tech- where both modes of transformation are
nology was based on the development of ge- readily available. In C. acremonium and P.
netic transformation systems for p-lactam chrysogenum, most transformation events oc-
producing organisms and cloning of biosyn- curred by ectopic integration into genomic
thetic genes. The ability to transform indus- DNA; only a small percentage of the transfor-
trially relevant organisms (specifically the fil- mants harbored homologous integration
amentous fungi P. chrysogenum and C. acre- events. Despite these limitations, transforma-
monium) would allow investigators to manip- tion systems such as these opened the door
ulate the biosynthetic pathway in a precise for the application of recombinant DNA
manner and provide an avenue for practical technology to these organisms.
applications (e.g., gene dosage studies for Transformation systems for prokaryotes
rate-limiting steps, gene disruptions to alter which produce p-lactam antibiotics have also
end products, or the addition of novel genes been developed. Recently, an efficient genetic
to modify the end product). Experimental transformation system was developed for the
conditions utilized in the development of the prokaryote Nocardia lactamdurans, a cepha-
first fungal transformation system (i.e., Sac- mycin C-producing organism, using an endo-
charomyces cerevisiae, HINNENet al., 1978) genous plasmid found in a species of Amyco-
were relied upon heavily in establishing con- latopsis (KUMARet al., 1994). This develop-
ditions for transformation of plactam pro- ment made possible the genetic manipulation
ducing fungi. The first fungal plactam proof Nocardia species and the related genus
ducing organism to be transformed was As- Amycolatopsis, some of which are used on an
pergillus nidulans (YELTONet al., 1984), how- industrial scale for the production of antibiot-
ever, this organism is not used on the indus- ics (e.g., efrotomycin, vancomycin, and rifa-
trial scale for production of p-lactam antibiot- mycin).
ics. C. acremonium was transformed the fol-
lowing year (QUEENER et al., 1985; PENALVA
et al., 1985). While the transformation of A .
nidulans relied on auxotrophic complementa-
tion, transformation of C. acremonium made
use of a dominant selectable marker, hygro- 5 Enzymes Involved in
mycin B phosphotransferase. Mutagenesis of
a highly developed production strain fre- Penicillin Biosynthesis
quently resulted in a decline in overall pro-
ductivity, thus the use of a dominant selecta- While the basic biosynthetic routes reading
ble marker which permitted transformation of to the different p-lactam antibiotics (see
a broad range of C. acremonium strains with- Fig. 1) have been known for several decades,
out mutagenesis was a significant advance. the enzymes catalyzing the reactions have
Several laboratories also developed transfor- only recently been isolated and studied in de-
mation systems for P. chrysogenum shortly tail. Knowledge about primary and tertiary
thereafter (BERIand TURNER,1987; CANTO- structures and catalytic properties of these
RAL et al., 1987; SANCHEZ et al., 1987; SKA- rather unique enzymes provided a rationale
TRUD et al., 1987a). Typical for filamentous for planning new biosynthetic routes and for
fungi, the transformation frequencies for the design of novel biocatalysts. Here, we will
these organisms are low (i.e., 1-10 transfor- focus on the three enzymes leading to the for-
mants per Fg of transforming DNA). Subse- mation of penicillin G. As stated above, iso-
quently, utilization of a homologous promot- penicillin N is a precursor for both penicillin
er (pcbC promoter) attached to the hygromy- G and cephalosporins, thus the first two en-
254 6 Advances in the Molecular Genetics of b-Lactam Antibiotic Biosynthesis
zymes are common to all organisms produc- WITZ et al., 1992). So far, the multienzymes
ing penicillin and cephalosporin-type com- from A. niduluns (VANLIEMFTet al., 1989) S.
pounds. cluvuligerus (BALDWIN et al., 1990; JENSENet
al., 1990; SCHWECKE et al., 1992a; ZHANGet
al., 1992), and C. ucremonium (BALDWIN et
5.1 S(L-a-Aminoadipy1)-L- al., 1990 ZHANGand DEMAIN,1990a) have
Cysteinyl-D-Valine Synthetase been purified to homogeneity and enzymati-
cally characterized (ZHANG,1991; ZHANG
(ACVS) and DEMAIN,1992). The S. cluvuligerus
ACVS was a monomeric protein (JENSENet
8- (L - a-Aminoadipyl) - L- cysteiny1- D - valine al., 1990; SCHWECKEet al., 1992b; ZHANG
(ACV), the first common intermediate in the and DEMAIN,1990b) whereas the C. ucre-
formation of all penicillins and cephalospo- monium ACV synthetase appeared to be a
rins by eukaryotic and prokaryotic microor- homodimer with weak physical interactions
ganisms, was initially thought to be synthe- between the two subunits (ZHANGand DE-
sized by two different enzymes in analogy to MAIN,1990a, b, 1992; ZHANGet al., 1992;
the structurally similar tripeptide glutathione. ZHANG,1991). In analogy to many known
This type of process would predict that a dipeptide synthetases, ACVS activates the
peptide existed as a free intermediate. How- three substrate amino acids at the expense of
ever, BANKOet al. (1986, 1987) showed that the a-p-phosphate bond of ATP (SCHWECKE
an extract of C. ucremonium catalyzed the et al., 1992a; VAN LIEMPTet al., 1989). For-
synthesis of ACV at a substantially higher mation of aminoacyl adenylates with all three
rate from the individual amino acids than amino acids was demonstrated by measuring
from G(L-a-Aminoadipy1)-L-cysteine (AC) formation of labeled ATP from [32P]-PPiin
and L-valine, and also faster than AC was the presence of the single amino acids. One
formed from first component amino acids. mole of ATP was consumed per peptide bond
The authors concluded that ACV synthesis in formed (KALLOWet al., 1994). Substrate spe-
C. ucremonium involved the action of a single cificity studies have demonstrated that ACVS
multifunctional enzyme. JENSENet al. (1988) accepts a broad range of amino acid analogs
reported on a cell-free extract of S. cluvulige- (SHIAUet al., 1995; ZHANGand DEMAIN,
rus capable of synthesizing ACV from the ap- 1992). The a-aminoadipate dependent ATP/
propriate component amino acids. The actual PPi exchange reaction catalyzed by the S. clu-
demonstration of catalysis by a multienzyme vuligerus enzyme was inhibited by Tris buffer
was achieved by VAN LIEMPTet al. (1989) (SCHWECKEet al., 1992a) as well as tripep-
with the isolation of a large synthetase from tide formation (ZHANGet al., 1992). ZHANG
A. niduluns which catalyzed the formation of and DEMAIN(1992) compared several P-lac-
ACV from the L-amino acid precursors. The tam biosynthetic enzymes in crude cell-free
corresponding gene was identified and found extracts of C. ucremonium and S. cluvuligerus
to direct the biosynthesis of a protein of and found that ACVS possessed only 1-10%
422kDa (MACCABEet al., 1991). To date of the specific activity of isopenicillin syn-
several other genes encoding ACV synthetase thase, isopenicillin epimerase, and deacetoxy-
from prokaryotes as well as from eukaryotes cephalosporin C-synthetase. After being acti-
have been cloned and characterized. The de- vated as aminoacyl adenylates, the substrate
duced sizes of the encoded polypeptides were amino acids were covalently bound to the cys-
between 404 and 424 kDa (AHARONOWITZ et teamine moieties of 4g-phosphopantetheine
al., 1993). A common feature found in all (ROLANDet al., 1975) which were attached to
these enzymes was the presence of three re- each activation domain of the enzyme by a
peated modules, each comprised of more posttranslational modification (BALDWINet
than 500 amino acids. These modules exhi- al., 1991; SCHLUMBOHM et al., 1991). In the
bited extensive sequence homology with non- final product formed by ACVS, valine oc-
ribosomal peptide synthetases such as firefly curred in the D-configuration, however, only
luciferase and acyl-CoA ligases (AHARONO- the L-configuration of covalently bound val-
ine could be released from the enzyme indi- ing blocks in the final product, as has been
cating that epimerization took place upon or convincingly shown for other peptide synthet-
after peptide bond formation (SHIAUet a]., ases (HORIet al., 1991; KRAUSEet al., 1985;
1995; SCHWECKE and VON DOHREN,unpub- KRAUSEand MARAHIEL, 1988; KUROTSOet
lished data). The putative thioesterase func- al., 1991; PIEPERet al., 1995; SAITOet al.,
tion located in the C-terminal region of 1995; STACHELHAUS et al., 1995; TURGAY et
ACVS has been proposed to specifically real., 1992; VAN SINDREN et al., 1993; WEBER
lease the LLD-tripeptide (KLEINKAUF and et al., 1994).
VON DOHREN,1996). Heterologous expression of individual
The order of the partial reactions involved modules in E. coli has so far been hampered
in the biosynthesis of the complete tripeptide by the fact that the fragments expressed were
is still not completely resolved. A particulate insoluble within the bacterium (SCHWECKE et
preparation of lysed protoplasts from a Ce- al., 1992b; V. UHLMANN,Masters Thesis,
phalosporium sp. incorporated DL-[14C]-val- Technical University Berlin, cited in KLEIN-
ine into penicillin when incubated with the di- KAUF and VON DOHREN,1996). After rena-
peptide AC and ATP (LODER and ABRA- turation of protein from unfolded inclusion
HAM, 1971b). BANKO and colleagues ob- bodies containing the N-terminal module of
tained ACV from AC and valine (BANKOet the S. clavuligerus ACVS, SCHWECKEet al.
al., 1986,1987). However, SHIAUet al. (1995) (1992b) demonstrated adenylation of all three
recently reported on the formation of cyste- constituent amino acids with valine showing
inyl-valine (CV) as an intermediate in ACV highest activity in the ATP/[32]PPi exchange
biosynthesis implicating a different mecha- reaction followed by cysteine and a-amino-
nism, which implies the initial formation of adipate. This somewhat surprising result
the second peptide bond. Nevertheless, the showed that this approach to determine ki-
yield was low and no biosynthesis of ACV oc- netic constants is of limited use in intact
curred from a-aminoadipate and the dipep- ACVS. However, upon incubation with all
tide CV (SHIAUet al., 1995). constituent 14C-labeled amino acids only a-
Such results do not resolve the issue of aminoadipate gave rise to radioactivity coval-
whether the order of peptide bond formation ently bound to the refolded fragment, sup-
is ACV or VCA. A critical requirement for porting the idea of a modular arrangement of
analyzing and manipulating ACVS is, there- activation units on the multienzyme that re-
fore, the ability to characterize each individu- flects the order of amino acids in the product.
al module or domain thereof. TAVANLAR No investigation into the nature of the coval-
and co-workers used limited proteolysis to ent bond was conducted. Therefore, it cannot
isolate and analyze ACVS fragments from C. be ruled out that the observed enzyme-bound
acremonium (TAVANLAR, SCHWECKE,VAN radioactivity was at least partially due to an
LIEMPT, and VON DOHREN, unpublished unidentified contaminant known to be pres-
data, cited in KLEINKAUF and VON DOHREN, ent in the 14C-a-aminoadipate(SCHWECKE et
1996). Three major bands with a Mr of ap- al., 1992a).
proximately 116 kDa were observed and pu- While cysteine and valine are available in
rified after digestion of a homogenous ACVS all cells, a-aminoadipate has to be synthe-
sample using either subtilisin or proteinase K. sized exclusively for p-lactam formation from
The digestion products were investigated by lysine in actinomycetes (KERN et al., 1980
sequence analysis and biochemical assays. KIRKPATRICK et al., 1973; MADDURIet al.,
The activation of cysteine as aminoacyl ade- 1989). Cephalosporin and cephamycin pro-
nylate could undoubtedly be attributed to the ducing strains possess an enzyme, L-lysine &-
middle fragment by means of ATP/PP, ex- aminotransferase (LAT), which mediates the
change assay. Surprisingly, the N-terminal formation of a-aminoadipate by removal of
fragment activated valine to a higher degree the &-aminogroup from lysine (MADDURI et
than a-aminoadipate. These results were in al., 1989, 1991; VININGet al., 1990). The re-
contrast to the assumption that the order of sulting 1-piperidine-6-carboxylic acid was
modules are colinear with the order of build- postulated to be oxidized by a yet unknown
dehydrogenase in analogy to a pipecolic acid two histidine residues, rendering the other-
pathway found in Pseudomonas sp. (CAL- wise highly reactive iron less damaging to the
VERT and RODWELL,1966). The gene encod- cell. The IPNS enzyme catalyzed both cycliza-
ing LAT has been mapped to the pcblcef gene tions found in the structure of isopenicillin N.
cluster in plactam antibiotic producing Since the initial purification of IPNS, the en-
strains of Strepromyces (COQUEet al., 1991b; zyme has been isolated from many different
MADDURIet al., 1991) and is absent in non- sources, including fungi, gram-positive and
producers (MADDURIet al., 1989). MALM- gram-negative bacteria. In each case, IPNS
BERG and colleagues found 4-fold elevated was characterized as a protein with a molecu-
levels of LAT and 2-5 times increased cepha- lar mass very close to that of IPNS from C.
mycin production after introduction of an ad- acremonium.
ditional copy of the gene into the chromo- The IPNS reaction mechanism has been ad-
some of S. clavuligerus (MALMBERGet al., dressed in elaborate studies by BALDWIN and
1993, 1995). Fungal p-lactam producing coworkers (BALDWIN and ABRAHAM, 1988;
strains make a-aminoadipate as an interme- BALDWIN and BRADLEY,1990). By analyzing
diate in the synthesis of lysine and do not kinetic isotope effects, they were able to show
seem to require such an enzyme. that the reaction proceeds in a two-step
mechanism, the intermediate being a mono-
cyclic plactam in which the sulphur is bound
5.2 Isopenicillin N Synthase to iron dioxygen of the enzyme. Based on
cell-free catalytic systems using purified re-
(IPNS) combinant IPNS, HUFFMANand colleagues
demonstrated that IPNS could recognize and
A key step in the biosynthesis of p-lactam utilize substrates similar to ACV. Such stud-
antibiotics is the conversion of the tripeptide ies led to the formation of many novel plac-
ACV into the bicyclic fused plactam-thiazol- tam compounds, some of which possessed an-
idine ring structure. This particular step has tibiotic activity (HUFFMANet al., 1992).
been the subject of many studies over several The three-dimensional structure of IPNS
decades. The structure of the plactam ring has recently been elucidated by BALDWIN
was correctly established through the work of and coworkers (ROACH et al., 1995). In a
ABRAHAM and colleagues over 50 years ago. complex with manganese, the recombinant
The structure of the direct precursor (ACV) IPNS protein from A. nidulans was crystal-
was determined in 1971 (LODERand ABRA- lized, and the structure was determined at a
HAM,1971a). The enzymatic conversion of resolution of 2.5 A. Structural analysis of
the tripeptide ACV into a p-lactam com- crystalized IPNS revealed the presence of 10
pound was finally established in 1979 ( O W L - helices and 16 P-strands, eight of which fold
LIVAN et al., 1979). ABRAHAMet al. (1981) into a jelly-roll motif. The active site was bu-
noted that IPNS required Fez+ and molecu- ried within the p-barrel. A glutamine residue
lar oxygen for activity. The catalytic activity which is conserved in all IPNS primary struc-
of IPNS was greatly stimulated by the pres- tures available to date interacted with the me-
ence of ascorbate and thiol compounds such tal ion (manganese in the case of the crystal-
as D m and by catalase (ABRAHAMet al., line form of IPNS) as well as two histidines
1981). The first purification of IPNS was re- and one aspartate residue. Furthermore, two
ported from C. acremonium in 1984 (PANGet water molecules were coordinated to the
al., 1984). Analysis of purified IPNS enzyme manganese. Based on these findings, a reac-
predicted a protein of 38 kDa, which required tion mechanism was proposed in which ACV
ferrous iron and ascorbate and used dioxygen and dioxygen bind to the coordination sites
as a co-substrate during catalysis. The nature occupied by the water molecules and gluta-
of the predicted iron-binding site within the mate.
active site cleft was addressed by CHENand IPNS is now the most well understood en-
colleagues (KRIAUCIUNAS et al., 1991). They zyme in the p-lactam biosynthetic pathway.
suggested that the metal ion was bound to The elucidation of the reaction mechanism
aided by knowledge of the three dimensional cursing fermentations with phenylacetic acid
crystal structure combined with the known or phenoxyacetic acid, respectively.
amino acid sequences of IPNS molecules The composition of the active form of AT
from many different sources may provide an has been the subject of some controversy. Ini-
avenue for rational design of IPNS proteins tial purifications of the A T enzyme suggested
with new desired activities. that it was a monomeric protein of approxi-
mately 30 kDa (ALVAREZet al., 1987; ALON-
so et al., 1988). However, subsequent protein
5.3 Acyl-Coenzyme A: purifications (e.g., WHITEMANet al., 1990)
Isopenicillin N Acyltransferase have indicated that A T is a heterodimer com-
posed of two dissimilar subunits of 11 kDa (a
(AT) subunit) and 29 kDa ( p subunit).
The eventual cloning of the gene encoding
The final step in the biosynthesis of hydro- A T (penDE) from both P. chrysogenum
phobic penicillins in P. chrysogenum and A . (BARREDOet al., 1989c; TOBINet al., 1990)
niduluns involves the removal of the L-a-ami- and A. niduluns (MONTENEGROet al., 1990;
noadipoyl side chain from isopenicillin N and TOBINet al., 1990), demonstrated that the
its exchange with one of many coenzyme A- penDE gene is composed of four exons en-
derived monosubstituted acetic acids (e.g., coding 357 amino acids, translated as a
phenylacetyl, forming benzylpenicillin) 40 kDa proenzyme, and posttranslationally
(BRUNNER et al., 1968; FAWCE-IT et al., 1975; processed to generate the two subunits
ALVAREZet al., 1987, 1993; WHITEMANet (Fig. 2). Based on NH2-terminal amino acid
al., 1990). Side chain exchange either occurs sequences, proenzyme cleavage occurs be-
directly or as a two-step process, forming 6- tween Glylo2 and Cyslo3 (BARREDOet al.,
aminopenicillanic acid (6-APA) as an inter- 1989~; WHITEMAN et al., 1990). However, the
mediate (QUEENERand NEUSS, 1982) (see presence of the -11 kDa subunit in active
Fig. 3). A T has been questioned by some investiga-
A single multifunctional enzyme, acyl- tors (e.g., BARREDOet al., 1989~).
coenzyme A:isopenicillin N acyltransferase Recent electrospray mass spectrometric
(AT), catalyzes all of the possible steps in- analysis of both recombinant and native P.
volved in this transacylation, including 6- chrysogenum AT has verified the location of
APA formation via IPN side chain removal the proenzyme cleavage site between Gly'O'
(IPN amidohydrolase, IAH), 6-APA acyla- and CySlo3(APLINet al., 1993a, b). These re-
tion (acyl-CoA:6-APA acyltransferase, ports also indicated that both the 11 kDa (a
AAT), IPN transacylation (acyl-CoA:IPN subunit) and the 29 kDa ( p subunit) proteins
acyltransferase, IAT), and acyl-CoA hydroly- are present in purified AT forming an a,@
sis (ALVAREZet al., 1993). The kinetic pa- heterodimer, in agreement with the report of
rameters reported for each of these reactions WHITEMANet al. (1990). To further investi-
suggests that removal of the a-aminoadipoyl gate the requirement for both of these sub-
side chain from IPN is the rate-limiting step units, TOBINet al. (1993) separated the re-
in the acyltransferase reaction (ALVAREZet gions of penDE encoding each subunit. Inde-
al., 1993). pendent production of either subunit in E.
The ability of the A T enzyme to accept a coli did not yield active AT, and post-expres-
variety of acyl-CoA derivatives (LUENGOet sion complementation by the mixing of cell
al., 1986; ALONSOet al., 1988), together with sonicates containing separately produced sub-
the capacity of P. chrysogenum to form a units did not regenerate activity. However,
number of acyl-CoA thioesters, has allowed reconstitution of A T activity was accom-
the in vivo production of more than 100 dif- plished by coproduction of the two subunits
ferent penicillins in P. chrysogenum fermen- produced from separate plasmids in the same
tations containing the corresponding precur- cell. Further, in vitro refolding of separately
sor acid (COLE, 1966). The production of produced 11 kDa and 29 kDa subunits only
penicillin G or V, e.g., is accomplished by pre- resulted in active A T when mixed together
pcbAB penDE
+ 1
I I '"
I 1 I1
II
I1
transcription
-1.4-kb mRNA
AAA
I
translation
AT 40-kDa proenzyme I co;
cleavage
Fig. 2. Transcription,transla-
I 11-kDa a-subunitI I
- 1
29-kDa Esubunit I
I tion, and posttranslational
Mature AT (a, heterodimer?) modification of AT in P. chry-
sogenum.
prior to refolding (TOBIN,1994). These re- stitution of this residue with Cys did not de-
sults indicate that the 11 kDa and the 29 kDa stroy AT activity, however, Ala at this posi-
proteins interact at an intermediate step in tion did not result in either proenzyme cleav-
the folding pathway to form AT enzyme. age or activity. Further, a thioesterase-like
Studies of the posttranslational cleavage of domain (G-X-S309-X-G) has been identified
the proenzyme have been hampered by the in AT (ALVAREZet al., 1993). TOBINet al.
inability to repeatably isolate the -40 kDa (1994) observed that Ser3@+Cys retained ac-
AT proenzyme from P. chrysogenum. Using a tivity, however, Ser3@+Ala was not active.
heterologous penDE expression system for E. AT enzyme containing either of these muta-
coli APLINet al. (1993a) observed the soluble tions was posttranslationally cleaved, indicat-
production of heterodimeric AT, suggesting ing that this residue is involved in the AT en-
that other specific P. chrysogenum factors zyme mechanism itself. Together with the
were not required for proenzyme cleavage. sensitivity of AT activity to chloromercuri-
Using insolubly produced, resolubilized, and benzoate and iodoacetamide and the require-
purified 40 kDa recombinant protein, TOBIN ment for reducing agents (e.g., DTT) for A T
(1994) further demonstrated cleavage of the activity (ALVAREZet al., 1993), these results
AT proenzyme by in vitro refolding. Cleavage imply that a thioesterase activity may be in-
occurred between Gly'O2 and Cys103, as ob- volved in the AT enzyme mechanism. Al-
served for native and solubly produced re- though there are five conserved Cys residues
combinant AT, suggesting that this event is in P. chrysogenum and A. nidulans AT, the
autolytic. requirement for Cys lo3 tempts speculation
Mutational analysis and recombinant ex- that this residue may bind the CoA-derived
pression of the penDE gene has identified acyl group, subsequently transferred to 6-
three amino acid residues essential for A T ac- APA or IPN by a thioesterase activity. The
tivity. Substitution of several residues in the apparent absolute requirement for proen-
proenzyme cleavage site region demonstrated zyme cleavage to produce active AT addition-
that Cyslo3 was absolutely required for AT ally suggests that proenzyme cleavage may be
proenzyme cleavage and AT activity. The an enzyme activation mechanism.
presence of Ser or Ala at this position yielded Substrate specificity studies have demon-
uncleaved and inactive recombinant AT (To- strated that, in addition to IPN and 6-APA,
BIN et al., 1995). Ser227,residing in an S-Q-N AT will hydrolyze and acylate penicillin N
motif in AT and other plactam biosynthetic forming 6-APA and penicillin G, respectively
enzymes was also examined for a potential (ALVAREZet al., 1987; TOBIN,1994). Fur-
role in AT activity (TOBINet al., 1994). Sub- ther, it has recently been shown that the ceph-
7 Clustering of PLactam Biosynthetic Genes 259
em nucleus 7-aminodeacetoxycephalospo-
rank acid (7-ADCA) is poorly acylated by re-
7 Clustering of P-Lactam
combinant AT, forming deacetoxycephalo-
sporin G (TOBIN,1994). Transacylation of
Biosynthetic Genes
cephalosporins containing aliphatic side
chains (e.g., deacetoxycephalosporin C, de- The isolation of antibiotic biosynthetic
acetylcephalosporin C, and cephalosporin C) genes from Streptomyces species revealed that
has not been observed. However, the accept- these genes tend to be organized in clusters,
ance of 7-ADCA and penicillin N (the latter and that they are often associated with anti-
containing the D-configured a-aminoadipoyl biotic resistance genes (MALARTIDAand
side chain) as a substrate would suggest that HOPWOOD,1984; SENO and BALTZ, 1989).
AT could be used to produce hydrophobic This clustering was also to be true for p-lac-
cephalosporins for industrial application to tam biosynthetic genes when BAILEYet al.
semi-synthetic cephalosporin production, (1984) observed the production of clavulanic
pending mutagenesis, and screening for the acid by introducing a cosmid clone containing
optimization of these activities. a segment of S. clavuligerus DNA into a non-
producing mutant. BURNHAMet al. (1987)
provided evidence that the cefE, cefF, and
cmc genes were clustered in S. clavuligerus.
KOVACEVIC et al. (1989) later isolated a cos-
6 Cloning of the Genes mid clone from S. clavuligerus containing
both the pcbC and cefE genes involved in
Involved in the cephalosporin Ckephamycin C biosynthesis.
This clone was subsequently used to isolate
Biosynthesis of Penicillin the cefD (KOVACEVICet al., 1990), cefF
(KOVACEVICand MILLER, 1991), lat, and
G, Cephalosporin C, and pcbAB genes (TOBINet al., 1991).
Both DiEz et al. (1989) and BARREDOet
Cephamycin C al. (1989b) demonstrated that the pcbC and
penDE genes were adjacent to one another in
A combination of approaches has been the filamentous fungus P. chrysogenum. A
used to identify, clone, and characterize genes study employing genetic complementation
involved in plactam biosynthesis including and heterologous cloning techniques later in-
genetic complementation, gene disruption, dicated that all of the genes involved in the
genetic linkage, and reverse genetics (MIL- biosynthesis of penicillins resided in close
LER and INGOLIA, 1989; SKATRUD, 1991). proximity to one another in this organism
Over the past decade, a substantial array of (SMITHet al., 1990b). In addition to demon-
plactam biosynthetic genes have been cloned strations of linkage in s. clavuligerus and P.
and sequenced from several different organ- chrysogenum, clustering of plactam biosyn-
isms (Tab. 2). A striking similarity exists be- thetic genes has been demonstrated in Asper-
tween corresponding genes and enzymes iso- gillus nidulans (SMITHet al., 1990a) as well as
lated from various organisms, despite their in the bacteria Flavobacterium and N. lactam-
phylogenetic diversity. This characteristic has durans (COQUEet al., 1991a, b). In addition,
been exploited to permit the isolation of a p-lactamase gene has been identified within
genes from several different organisms by the Flavobacterium (Lysobacter) and N. lac-
cross-hybridization techniques. tamdurans clusters (KIMURAet al., 1990;
COQUEet al., 1993). Fig. 3 illustrates the ar-
rangement of p-lactam biosynthetic genes in
various representative organisms. Recent re-
sults have demonstrated that the clavulanic
acid biosynthetic gene cluster in S. clavulige-
rus is adjacent to the gene cluster for p-lac-
260 6 Advances in the Molecular Genetics of @-LactamAntibiotic Biosynthesis
Tab. 2. DNA Sequences Reported for @-LactarnBiosynthetic Genes
Gene (Enzyme Endocded) Organism Reference
lat S. clavuligerus TOBINet al. (1991)

(lysine-E-arninotransferase) N. lactamdurans COQUEet al. (1991a)
pcbAB P. chrysogenum SMITHet al. (199Oc), D ~ E et
Z al. (1990)
(ACV synthetase) A. nidulans MACCABEet al. (1991)
C. acremonium GUTIBRREZ et al. (1991)
HOSKINSet al. (1990)
S. clavuligerus (partial) TOBINet al. (1991)
pcbC C. acremonium SAMSON et al. (1985)
(IPN synthase) P. chrysogenum CARR et al. (1986), BARREDOet al.
(1989a)
A. nidulans RAMdN et a]. (1987), WEIGEL et al.
(1988)
S. lipmanii WEIGELet al. (1988)
S. clavuligerus LESKIWet al. (1988)
S. jumonjinensi SHIFFMAN et al. (1988)
S. griseus GARCIA-DOMINGUEZ et al. (1991)
Flavobacterium sp. 12154 SHIFFMAN et al. (1990)
N. lactamdurans COQUEet al. (1991b)
penDE P. chrysogenum BARREDO et al. (1989~)
(IPN acyltransferase) A . nidulans MONTENEGROet al. (1990), TOBINet al.
(1990)
cefD S. clavuligerus KOVACEVIC et al. (1990)
(IPN epirnerase) N. lactamdurans COQUEet al. (1993)
cefEF C. acremonium SAMSON et al. (1987)
(REZH)
cefE S. clavuligerus KOVACEVICet al. (1989)
(expandase) N. lactamdurans COQUEet al. (1993)
cefF S. clavuligerus KOVACEVIC and MILLER(1991)
(hydroxylase)
cefG C. acremonium GUTII~RREZ
et al. (1992)
(acyltransferase)
cmcH N. lactamdurans COQUEet al. (1995a)
(carbarnoyltransferase)
cmcl N. lactamdurans COQUEet al. (1995b)
(7 ‘-a-hydroxylase)
cmcl N. lactamdurans COQUEet al. (1995b)
(rnethyltransferase)
tam antibiotic biosynthesis (WARD and and QUEENER (1989) reported that the early
HODGSON, 1993). genes were located on chromosome VI and
An anomaly regarding linkage of biosyn- the late genes on chromosome I1 in an indus-
thetic genes occurs in the filamentous fungus trial strain of C. acremonium (strain 394-4).
C. acremonium. The early genes (pcbAB, However, a later report (FIERRO et al., 1994)
pcbC) are linked (HOSKINS et al., 1990), how- demonstrated that in C. acremonium ATCC
ever, they reside on a different chromosome 28901 the early and late genes resided in
from the late (cefEF, cefG) genes. SKATRUD chromosomes VII and I, respectively. In both
--
8 Compartmentalization of PLactam Biosynthetic Enzymes 261
I P. chrysogenumand A. nidulans
I
pcbAS pcbC penD€
C. acremonium
I 4 *I I 4- I
pcbAB pcbC cef€f c e f ~
S. clavuligerus
w - a c - w -1
ORF OAF ORF ORFpCbC pcbAS /at cefD cef€ cefF
10 8, 9 2-7 1
N. lactamdurans
- 1 4+-----q
bla ORF J I H pcbc pcbAB /at cefDcefE CmcTORFpbp
10 cmc 12
I Lysobacter (Flavobacterium)
Fig. 3. Clustering of p-lac- 1- I
tam biosynthetic genes in bla cefD ceff cefE pcbC pcbAS
various organisms.
studies the early and late genes were in sepa- tion, KOVACEVIC et al. (1989) reported that
rate clusters. It is conceivable that these ap- the cefD and cefE genes are cotranslated.
parently conflicting reports are due to differ- Thus, portions of the biosynthetic pathway
ences between the strains. The location of the constitute operons. In the fungi, however, the
cefD gene in C. acremonium has not been de- early genes are transcribed in opposite direc-
termined. tions. While clustering of biosynthetic genes
In A. niduluns ATCC 28901 the penicillin in fungi is common separate genes have their
biosynthetic pathway is located on chromo- own promoter, so relative orientation is not
some VI (3.0-Mb) (MONTENEGROet al., important in a regulatory sense (TURNER,
1992). In P. chrysogenum AS-P-78 and P2 the 1992).
cluster is located on chromosome I (FIERRO
et al., 1994). In P. norarum, the organism orig-
inally used for penicillin production, the bio-
synthetic genes reside on chromosome I1
(FIERROet al., 1994). 8 Compartmentalization
The orientation of transcription of the p-
lactam biosynthetic genes relative to one an- of P-Lactam Biosynthetic
other is not entirely conserved. In the proka-
ryotes (Flavobacrerium,N. lactamdurans, and Enzymes
S. clavuligerus) the early genes tend to be
transcribed in the same direction. PETRICH The role of subcellular compartmentaliza-
and JENSEN(1994) reported that the sequen- tion in the biosynthesis of penicillin in P.
tial arrangement of the lat, pcbAB, and pcbC chrysogenum has been elucidated in some de-
genes affords coordinate regulation at the lev- tail. Early studies indicated that ACVS activi-
el of transcription, involving transcription ty was associated with a particular fraction of
units of various lengths. In addition to a short the crude cell homogenate (FAWCETT and
transcript for pcbC alone, a polycistronic mes- ABRAHAM,1976). IPNS is apparently a sol-
sage encoding LAT, ACVS, and IPNS has uble enzyme, although its activity in cell-free
been suggested by these investigators. Wheth- extracts appears to be stimulated by Triton X-
er or not the lut gene may be transcribed by 100 or sonication (SAWADAet al., 1980).
itself is not clear (S. JENSEN,University of More recently, KURYLOWICZet al. (1987)
Alberta, personal communication). In addi- suggested that penicillin G is synthesized in
the Golgi system. Pursuant to this finding, In light of the compartmentalization of the
MULLERet al. (1991) determined that the last penicillin biosynthetic pathway it seems likely
step in penicillin biosynthesis (acyl transfer) is that the alleviation of substrate limitations,
accomplished in organelles of 200-800 nm in particularly with regard to the formation of
diameter. A subsequent report by this group ACV, would be an effective route to strain
determined that a short signal peptide (Ala- improvement.
Arg-Leu) located at the COOH-terminus of
the AT enzyme is required for its localization
in organelles (“microbodies”) and for benzyl-
penicillin biosynthesis (MULLERet al., 1992).
This signal sequence is similar to that found
at the COOH-terminus of several proteins
9 Evolution of the
known to be transported to peroxisomes P-Lactam Biosynthetic
(GOULDet al., 1989). In addition, evidence
for the compartmentalization of the amino Pathway: The Horizontal
acid precursors required for ACV formation
has been obtained (AFFENZELLER and KUBI- Transfer Hypothesis
CEK,1991; LEDENFELDet al., 1993) and con-
firms the suspicions of earlier studies regard- The improbable identity demonstrated be-
ing the limitation of ACVS activity in vivo by tween particular genes and enzymes involved
one of its substrates (HONLINGER et al., 1988; in p-lactam biosynthesis from a wide diversity
HONLINGERand KUBICEK,1989). LEDEN- of microorganisms has generated a great deal
FELD et al. (1993) ultimately concluded that of speculation concerning the horizontal
the a-aminoadipate, L-cysteine, and L-valine transfer of the p-lactam biosynthetic pathway
precursors for ACV formation are derived from a prokaryote to a eukaryote. The first
from the vacuole, and ACVS is located either theory - as put forward by T. D. INGOLIAin
within or bound to the vacuolar membrane. A CARRet al. (1986) and later expanded (WEI-
current model (as suggested by LEDENFELD GEL et al., 1988; SKATRUD,1991) - was de-
et al., 1993) regarding the compartmentaliza- signed to explain the identity of the pcbC
tion of events involved in penicillin biosyn- genes of P. chrysogenum and C. acremonium
thesis in P. chrysogenum is illustrated in and the presence of IPNS activity in gram-po-
Fig. 4. sitive bacteria of the Sfreptomyces genus. As
9
PSC
Extracellular
A
PSC Vacuole
Pen
. I Microbodies
Microbodies Fig. 4. Compartmentalizationof
a-AAA the penicillin biosynthetic path-
IPNS ACVS
PSC-CoA way in P. chrysogenum (ad-
IF” 6ACV
-1
apted from LENDENFELD et al.,
1993). Open circles represent
possible transport steps, filled
a-AAA circles represent enzymatic
steps; a-AAA: a-aminoadipic
Intracellular
4
acid; PSC: precursor side chain;
P6!C Val Cys a-AAA PSC-CoA: CoA-activated PSC,
Pen: penicillin; 6-OPC 6-0x0-
piperidine-2-carboxylic acid;
enzymes are indicated in bold
letters.
9 Evolution of the p-Lactam Biosynthetic Pathway: The Horizontal Transfer Hypothesis 263
originally stated, the so-called “horizontal and Staphylococcus aureus from unknown
transfer hypothesis” argued for the transfer of sources have been proposed (DOWSON et al.,
the pcbC genes from Streptomyces to a euka- 1989; MATSUHASHI et al., 1986). Thus, in a
ryotic organism. Since then, the DNA se- mechanistic sense, a horizontal transfer of the
quences of several more pcbC and other /?- p-lactam biosynthetic genes from a proka-
lactam biosynthetic genes have been eluci- ryote to a eukaryote appears plausible.
dated, and the theory remains consistent with While a great deal of attention has been
the new information. Supportive evidence paid to the acquisition of the biosynthetic
from other genes, particularly the pcbAB, pathway by various organisms, little notice
penDE, cefD, cefE, cefF, and cefEF genes, has has been given to the explanation of how a
lent credence to the theory of horizontal number of genes recognizing similar sub-
transfer (SKATRUD, 1991). The close linkage strates appeared and how genetic linkage
of the genes in a variety of organisms makes a arose. There are two general scenarios: first,
single transfer event plausible. However, the it is conceivable that several of the genes in
DNA sequence of the pcbC gene from a the pathway arose via convergent evolution
gram-negative bacterium, Flavobacterium, of substrate specificity to perform the various
has produced an extension/modification of reactions. The second possibility is that some
the suggested transfer theory. Based on this of the genes may have arisen by divergent
DNA sequence AHARONOWITZ et al. (1992) evolution, requiring (1) a gene duplication,
suggested that more than one horizontal and (2) divergent evolution of function.
transfer event may have occurred during the There does not appear to be a great deal of
evolution of this biosynthetic pathway. DNA or amino acid homology between most
It is virtually impossible to accurately pre- genedenzymes for different steps in the path-
dict or suggest how this horizontal transfer way, with the exception of a conserved motif
might have occurred. However, there are sev- (S ...Q ...N) identified by KOVACEVICand
eral known mechanisms of DNA transfer, in- MILLER(1991) and TOBINet al. (1994). This
cluding conjugative plasmids, DNA transfor- motif is surrounded by a block of similar ami-
mation, protoplast fusion, and bacteriophage no acids in all the enzymes examined. Howev-
transduction. In one study plasmids in plac- er, the case of the expandase and hydroxylase
tam producing streptomyces were detected functions provides an elegant testimonial for
recently (KINASHI et al., 1995) The plasmids the second possibility. In C. acremonium, one
ranged in size from 12 kb to 450 kb - an ade- gene (cefEF) encodes both activities, howev-
quate size to contain the entire P-lactam bio- er, in S. clavuligerus, these two activities are
synthetic pathway in at least some cases. encoded by separate genes (cefE and cefF). In
However, hybridization analysis did not pro- the latter organism, each enzyme retains a re-
vide evidence for the genes on these plasmids sidual level of the other activity, however, the
(KINASHI et al., 1995). The ability to conju- DNA and amino acid sequences are strikingly
gate E. coli to both Saccharomyces cerevisiae similar (KOVACEVIC and MILLER, 1991). The
(HEINEMANN and SPRAGUE, 1989) and Schi- level of identity indicates that the separate
zosaccharomyces pombe (SIKORSKI et al., cefE and cefF genes arose by gene duplication
1990) in the laboratory is a direct demonstra- in S. clavuligerus (KOVACEVIC and MILLER,
tion of DNA transfer between a prokaryote 1991). An extended account of the horizontal
and a eukaryote. BORK and DOOLITTLE transfer hypothesis has been given in SKA-
(1992) have proposed that the fibronectin TRUD (1991).
type I11 domain of animal proteins has been
acquired by bacteria through horizontal
transfer from mammalian cells. In addition,
bacterial acquisition of antibiotic resistance
by plasmids, transposable elements, and gene
transfers is well known (DAVIES,1994) and
recent examples of horizontal DNA transfers
of PBP genes to Streptococcus pneumoniae
264 6 Advances in the Molecular Genetics of fi-Lactam Antibiotic Biosynthesis
10 P-Lactam Antibiotics These investigators returned multiple copies

of this gene via genetic transformation to a
and Recombinant DNA different strain of C. acremonium. They were
able to show a direct correlation between
Technology: Practical cefG copy number, message level, and cepha-
losporin C titer suggesting a different rate-
Applications limitation in their strain. It is interesting to
note that in the study by SKATRUDet al.
(1989) the fragment containing the cefEF
The use of gene dosage of genes involved gene was large enough to include the cefG
in penicillin biosynthesis for industrial strain gene as well. Thus cefG encoded activity may
improvement of P. chrysogenum was unsuc- have been augmented as well, depending on
cessful (SKATRUD et al., 1987a). Shortly the site of integration within the transforming
thereafter, evidence emerged which explained DNA, and may have exerted some influence
such results. Analysis of industrial strains of on the results in that study.
P. chrysogenum revealed that the cluster of Several steps in the biosynthesis of cepha-
genes responsible for penicillin production losporin C are oxygen dependent. DEMO-
was reiterated several times (BARREDOet al., DENA et al. (1993) reasoned that increasing
1989b; SMITH et al., 1989; TOBIN, unpub- the intracellular availability of oxygen might
lished data). Thus gene dosage was already positively affect overall antibiotic productivi-
utilized by these strains as a means of increas- ty. To accomplish this goal they expressed a
ing overall productivity. To further increase bacterial hemoglobin gene in C. acremonium
yields in such strains, it may be more fruitful (strain C10). Fermentation results demon-
to study and alter regulation of gene expres- strated increased cephalosporin C production
sion within the penicillin biosynthetic clus- in some transformants carrying the bacterial
ter. hemoglobin gene. It should be noted, howev-
In contrast, gene dosage studies in C. acre- er, that the strain transformed was not a high-
monium did meet with success. In the first ex- ly developed strain in terms of cephalosporin
ample, characterization of an industrial strain C production. Also, fermentation analyses
of C. acremonium suggested that the expand- were performed in shake flasks (baffled and
ase/hydroxylase step was rate-limiting in the unbaffled) in which oxygen limitation is an in-
biosynthesis of cephalosporin C due to the ac- herent problem. It is not clear that the same
cumulation of isopenicillin N (see Fig. 1). To type of effect would be observed in large-
overcome this rate limitation, this strain was scale fermentations.
transformed with a plasmid containing the Another exciting area for application of re-
cefEF gene encoding expandase/hydroxylase. combinant DNA technology to plactam pro-
Isolate LU4-79-6 contained one extra copy of ducing organisms was the production of inter-
cefEF ectopically integrated into its genome. mediate compounds useful in the manufac-
In fermentations utilizing strain LU4-79-6, ture of semi-synthetic cephalosporins (SKA-
isopenicillin N no longer accumulated, sug- TRUD, 1992). Medically important cephalo-
gesting that this rate-limitation was over- sporins such as cephalexin, cephradine, and
come. Further analysis revealed that this cefadroxil are made by the addition of differ-
strain possessed twice the amount of expand- ent side-chains to the 7-amino group of 7-ami-
ase/hydroxylase activity as compared to the nodeacetoxycephalosporin C (7-ADCA)
recipient and produced approximately 15% (BUNNELLet al., 1986). This procedure nor-
more cephalosporin C when scaled up to in- mally includes several chemical modification
dustrial level (SKATRUDet al., 1989). In an- steps with production of organic solvent
other example, the cefG gene which is direct- waste products. A biosynthetic/enzymatic
ly adjacent to cefEF was cloned from C. acre- process to produce 7-ADCA was proposed by
monium (MATHISONet al., 1993). This gene CANTWELLet al. (1990) which would elimi-
encodes the enzyme which carries out the last nate three chemical modifications and reduce
step in cephalosporin C production (Fig. 1). pollution. The proposed process involved in
11 Regulation of P-Lactam Biosynthetic Gene Expression 265
vivo ring expansion of penicillin V in P. chry- pansion of adipoyld-APA with expandase or

sogenum via a genetically engineered cefE expandase/hydroxylase produced by recombi-
gene to form cephalosporin V and subse- nant E. coli cells. In contrast, BALDWINet al.
quent enzymatic deacylation resulting in the (1987), working with cell-free extracts from C.
desired end product, 7-ADCA. The initial ucremonium, were able to detect ring expan-
step in establishing feasibility of such a pro- sion of this substrate.
cess was accomplished by these investigators.
The cefE gene from S.clavuligerus was mod-
ified by fusing it in frame with the pcbC pro-
moter of P. chrysogenum. The modified gene
was transformed into an industrial strain of P. 11 Regulation of
chrysogenum. Transformants were recovered
which maintained their capacity to produce P-Lactam Biosynt hetic
penicillin and possessed expandase activity
(CANTWELLet al., 1990). In order for this Gene Expression
work to achieve maximal impact, the amount
of cefE gene product produced in P. chryso- Investigations of fungal strains used in the
genum transformants would have to be in- production of plactam antibiotics have re-
creased and the substrate specificity of cefE vealed two general strategies to enhance pro-
would have to be modified to efficiently act ductivity. In general terms, increasing the
on penicillin V. amount of biosynthetic enzymes present in an
CRAWFORD et al. (1995) took a somewhat organism should increase productivity as long
different approach to achieve the production as primary metabolites or other factors uti-
of 7-ADCA in P. chrysogenum. Their process lized in biosynthesis are not depleted. Two
involved the expression of the S. clavuligerus routes for increasing the concentration of bio-
cefE or the C. acremonium cefEF gene in ge- synthetic enzymes are readily apparent. An
netic transformants of P. chrysogenum, and increase in the rate of transcription (i.e., up-
subsequent fermentation in the presence of regulation) or an increase in the number of
adipic acid. Transformants produced adipoyl- genes (i.e., gene dosage) or a combination of
6-APA (presumably by acyltransferase-cata- both, should achieve enhanced production of
lyzed transacylation of IPN and adipoyl- plactams. As detailed earlier, genes involved
CoA), which is subsequently ring-expanded, in the biosynthesis of p-lactam antibiotics are
producing high titers of adipoyl-7-aminode- closely linked in several organisms. Such
acetoxycephalosporanic acid (ad-7-ADCA) close physical proximity of the biosynthetic
(cefE-catalyzed), or adipoyl-7-aminodeacetyl- genes entails obvious benefits regarding the
cephalosporanic acid (ad-7-ACA) (cefEF-cat- inheritance of these genes and may simplify
alyzed). The adipic acid side chain from ad- the possibility of coordinate transcriptional
7-ADCA is removed by the Proteus SY77-1 regulation.
cephalosporin acylase, producing 7-ADCA in In P. chrysogenum, physical linkage of the
high yields. Current work is directed at at- penicillin biosynthetic genes proved to be an
taching this acylase to a resin to enable the advantage in terms of improving overall peni-
removal of the adipic acid from the acylase cillin productivity by gene dosage. Two
reaction. One apparent drawback to this ap- groups independently demonstrated that the
proach may involve isolation of ad-7-ADCA penicillin biosynthetic cluster in P. chrysoge-
or ad-7-ACA from the fermentation broth, num was amplified several-fold in strains
however, these products may prove to be ex- highly developed for penicillin production
tractable by an organic solvent. It is interest- (SMITHet al., 1990a; D ~ E Z et al., 1990). The
ing to note that the observed in vivo ring ex- three genes formally recognized as a part of
pansion of adipoyld-APA could not be dupli- the biosynthetic pathway are contained with-
cated in vitro with cell-free extracts of S. clu- in an approximately 15-kb region. The total
vuligerus (MAEDA et al., 1995). YEH et al. amplified region was about 35 kb in size. In a
(1994) similarly reported no in vitro ring ex- more recent study, FIERRO et al. (1995) found
a 106.5-kb region repeated several-fold in penicillin biosynthesis is sensitive to environ-

tandem on chromosome I in a different strain mental factors such as nitrogen source, car-
of P. chrysogenum which was highly devel- bon source, and growth stage in P. chrysoge-
oped for the production of penicillin. An- num suggesting the presence of several regu-
other strain they investigated contained a latory factors. ESPESOand PENALVA(1992)
- 58-kb repeat containing the penicillin bio- described a temporal pattern of expression of
synthetic pathway. These investigators also penicillin biosynthetic genes in A. nidulans
noted a hexanucleotide repeat (TTTACA) which was dependent upon carbon source.
which bounded the amplified units in all An enhancement of that knowledge came
strains examined, regardless of copy number when FENGet al. (1994) verified that expres-
or size of amplified unit. FIERROet al. (1995) sion of the penicillin biosynthetic genes in P.
suggested that amplification occurred by mu- chrysogenum was regulated by these factors
tation-induced site-specific recombination at at the level of transcription. Using a reporter
the conserved hexanucleotide sequence. gene analysis, their study revealed that pro-
A salient point regarding the amplified unit moters found within the intergenic region be-
of genomic DNA in P. chrysogenum is that tween pcbAB and pcbC were sensitive to ni-
the length of the pcbAB, pcbC, and penDE trogen repression, glucose repression, and
genes is only a fraction of the amplified regrowth stage. The gene (NRE) encoding the
gion. It is possible that other genes involved major nitrogen regulatory protein has recent-
in penicillin biosynthesis and the regulation ly been cloned from P. chrysogenwn (HAAS
thereof are encoded by this region. Indeed, a et al., 1995). This gene resembles other major
recent report suggests that the pcbAB and nitrogen regulatory genes (e.g., 60% identical
pcbC genes are subject to truns-acting regula- to ureA from A. niduluns and 30% identical
tion (CHU et al., 1995). A transcriptional reg- to nit2 from Neurosporu crussu). HAASand
ulatory protein which binds to the upstream MARZLUF(1995) carried these studies further
regions of both pcbAB (-32 to -200 and by demonstrating that the NRE encoded pro-
-188 to -344) andpcbC( -10 to -131) was tein binds specifically to the intergenic pro-
identified by electromobility shift assay. Bind- moter regions of nitrate assimilation and pen-
ing of this regulatory protein to these regions icillin biosynthetic gene clusters of P. chryso-
results in a repression of transcription of both genum. All three of the above mentioned ma-
genes. A similar situation has been observed jor nitrogen regulatory proteins are members
in A . niduluns (PEREZ-ESTEBAN et al., 1993). of the GATA protein family - as such they
Whether these elements effect penDE tran- are DNA binding proteins characterized by
scription, either directly or indirectly by re- the presence of a zinc finger motif (Cys-X2-
pression of pcbAB and pcbC transcription, Cys-X17-Cys-X2-Cys) and they recognize the
has not been reported. FENGet al. (1995) re- sequence GATA which is present in the in-
ported several nuclear proteins which bind to tergenic region of pcbABlpcbC. ESPESO et al.
the intergenic region between pcbAB and (1993) and TILBURN et al. (1995) demon-
pcbC. One particularly abundant nuclear pro- strated that the pcbC gene was regulated by
tein, nuclear factor A (NF-A), recognizes and PACC, a regulatory protein which mediates
binds to a specific 8 base pair sequence pH response of several other A. niduluns
(GCCAAGCC) found in the intergenic re- genes as well. THEN BERGH et al. (1996)
gion. BRAKHAGE and VANDENBRULLEidentified a major cis-acting DNA element,
(1995) utilized a reporter gene system in A . located between pcbAB and pcbC in A . nidu-
niduluns to identify recessive truns-acting mu- luns which regulates expression of both genes.
tations residing in two complementation In this organism 872 base pairs separate these
groups (prgAl and prgB1) which regulate p two divergently transcribed genes. Deletion
lactam biosynthesis. Mutations in these com- of nucleotides -353 to -432 upstream of
plementation groups resulted in a depression pcbAB lead to an altered expression of both
of penicillin production. genes. Further, analysis revealed the se-
Although not clearly understood at the mo- quence CCAAT (designated PENR1) found
lecular level, it has been known for years that within this region, was bound specifically by a
I3 References 267
protein or protein complex. Evidence was provement studies). Currently, strain im-
also provided which suggested that the corre- provement may be approached not only from
sponding genes in C. aremonium and P. chry- the standpoint of enhanced productivity but
sogenum contained a similar regulatory cir- also biosynthetic pathway engineering result-
cuit. ing in production of valuable new end prod-
The regulation of genes involved in cepha- ucts useful in the manufacture of semisynthet-
losporin C production in C. acremonium or ic p-lactam compounds (SKATRUD,1992).
cephamycin C production in the streptomy- In the same time frame, the clinical effec-
cetes has not been studied as extensively as tiveness of p-lactam antibiotics was once
those involved in penicillin production. It is again threatened by wide-spread emergence
interesting to note that the first two genes of of serious resistance problems (e.g., methicil-
the biosynthetic pathway (pcbAB and pcbC) lin-resistant Staphylococcus aureus [MRSA],
in C. acremonium are physically arranged in penicillin-resistant Streptococcus pneumo-
the same manner as in A. nidulans and P. niae). The primary mode of resistance in
chrysogenum (see Fig. 2). Despite a similar these organisms was not due to the elabora-
physical arrangement, gene amplification for tion of p-lactamases which has been a major
the cephalosporin C biosynthetic pathway has clinical problem in the past. Rather, the tar-
not been observed. This observation suggests geted penicillin-binding proteins (PBPS) were
that alteration in gene regulation produced either modified by mutation (PBP2x of S.
enhanced cephalosporin C production in in- pneumoniae) or replaced with a novel PBP
dustrial strains of C. acremonium. The rela- (PBP2a in MRSA) resulting in a target with
tive strength of the pcbAB and pcbC promot- low affinity for p-lactam antibiotics (SPRATT,
ers was recently studied with a reporter gene 1989; HARTMANand TOMASZ, 1984). Once
system (MENNEet al., 1994). Results of that again recombinant DNA technology may
study suggested that the pcbC promoter was have a dramatic impact on the use of p-lactam
approximately five times stronger than the antibiotics. However, in this case the applica-
pcbAB promoter. Such a contrast in the relation of this technology will also be on the or-
tive strength of promoters in the same biosyn- ganisms which are killed by these antibiotics,
thetic pathway might suggest differences in not on the producing organisms. One novel
protein stability, protein localization (i.e., approach to development of new antibiotics
compartmentalization vs. cytoplasmic), or targeted at the modified PBPs may involve
specific activity of the enzymes involved in structure-based drug design in which the
these steps. Regulation of the rate of tran- three-dimensional structure of a low-affinity
scription for the cephalosporin C biosynthetic PBP will be determined (Wu et al., 1992).
pathway can be positively influenced by the Based on that information, modifications to
addition of exogenous methionine to the me- existing compounds may lead to new more ef-
dium (VELASCOet al., 1994). fective plactam antibiotics and novel ap-
proaches for rapid diagnostic detection of
problematic organisms (UNALet al., 1992).
12 Summary
The last decade has produced numerous
significant advances in the p-lactam antibiotic
13 References
field. Most noteworthy perhaps has been the
ABRAHAM,E. P., CHAIN,E, FLETCHER,C. M.,
application of recombinant DNA technology GARDNER,A. D., HEATLEY,N. G., JENNINGS,
to. the organisms which produce these life- M. A., FLOREY,H. W. (1941), Further observa-
saving compounds. This application of tech- tions on penicillin, Lancet 241 (2), 177-188.
nology has opened new doors to many ave- ABRAHAM, E. P., HUDDLESTON, J. A., JAYATI-
nues of research which were in some in- LAKE, G . S., O'SULLIVAN,J., WHITE, R. L.
stances becoming less fruitful (e.g., strain im- (1981), in: Recent Advances in the Chemistry of
PLactam Antibiotics, R. SOC. Chem., Special BALDWIN J. E., ABRAHAM, E. P. (1988), The bio-
Publication No. 38, pp. 125-134. synthesis of penicillin and cephalosporins, Nat.
AFFENZELLER, K., KUBICEK,C. P. (1991), Evi- Prod. Rep. 5, 129-145.
dence for a compartmentation of penicillin bio- BALDWIN, J. E., BRADLEY, M. (1990), Isopenicillin
synthesis in a high- and a low-producing strain of N synthase mechanistic studies, Chem. Rev. 90,
Penicillium chrysogenum, J. Gen. Microbiol. 131, 1079-1088.
1653-1660. BALDWIN, J. E., ADDLINGTON, R. M., COATES,J.
AHARONOWITZ, Y., COHEN,G., MARTIN, J. F. B., CRABBE, M. J., CROUCH,N. P., KEEPING,J.
(1992), Penicillin and cephalosporin biosynthetic W., KNIGHT,B. C., SCHOFIELD, C. J., TING,H.-
genes: structure, organization, regulation and H., VALLEJO,C. A,, THORNILEY, M., ABRA-
evolution, Annu. Rev. Microbiol. 46, 461495. HAM,E. P. (1987), Purification and initial char-
AHARONOWITZ, Y., BERGMEYER, H., CANTORAL, acterization of an enzyme with deacetoxycepha-
J. M., COHEN,G., DEMAIN,A. L., FINK,U., losporin C synthase and hydroxylase activities,
KINGHORN,J., LANCE,M., MCCABE,A., PAL- Biochem. J. 245, 831-841.
ISA, H., PFEIFER,E., SCHWECKE,T., VAN BALDWIN, J. E., BIRD,J. W., FIELD,R. A., O’CAL-
LIEMPT, H., VON DOHREN,H., ZHANG,J. LAGHAN, N. M., SCHOFIELD, C. J. (1990), Isola-
(1993), S(L-cu-Aminoadipyl)-L-cysteinyl-D-val- tion and partial characterization of ACV syn-
ine synthetase, the multienzyme integrating the thetase from Cephalosporium acremonium and
four primary reactions in plactam biosynthesis, Streptomyces clavuligerus, J. Antibiot. 43, 1055-
as a model peptide synthetase, BiolTechnology 1057.
11,807-810. BALDWIN, J. E., BIRD,J. W., FIELD,R. A., O’CAL-
ALONSO,M. J., BERMEJO,F., REGLERO,A., FER- LAGHAN, N. M., SCHOFIELD, C. J., WILLIS,A.
NANDEZ-CAN6N, J. M., GONZALEZDE BUI- C. (1991), Isolation and partial characterization
TRAGO,G., LUENGO,J. M. (1988), Enzymatic of ACV synthetase from Cephalosporium acre-
synthesis of penicillins, J. Antibiot. 41, 1074- monium and Streptomyces clavuligerus:evidence
1084. for the presence of phosphopantothenate in
ALVAREZ,E., CANTORAL, J. M., BARREDO,J. L., ACV synthetase, J. Antibiot. 44,241-248.
D~Ez,B., MART~N, J. F. (1987), Purification to BANKO,G., WOLFE, S., DEMAIN,A. L. (1986),
homogeneity and characterization of acyl-coen- Cell-free synthesis of S(L-a-aminoadipy1)-L-cys-
zyme A: 6-aminopenicillanic acid acyltransferase teine, the first intermediate of penicillin and
of Penicillium chrysogenum, Antimicrob. Agents cephalosporin biosynthesis, Biochem. Biophys.
Chemother. 31, 1675-1682. Res. Commun. 131,528-535.
ALVAREZ,E., MEESSCHAERT, B., MONTENEGRO, BANKO,G., DEMAIN, A. L., WOLFE,S. (1987), 6
E., GUTIERREZ,S., D~Ez,B., BARREDO,J. L., (L-a-Aminoadipyl)-L-cysteinyl-D-vaIine synthet-
MART~N, J. F. (1993), The isopenicillin-N acylase (ACV synthetase): a multifunctional enzyme
transferase of Penicillium chrysogenum has iso- with broad substrate specificity for the synthesis
penicillin-N amidohydrolase, 6-aminopenicil- of penicillin and cephalosporin precursors, J.
lanic acid acyltransferase and penicillin amidase Am. Chem. SOC. 109,2858-2860.
activities, all of which are encoded by the single BARREDO,J. L., CANTORAL, J. M., ALVAREZ,E.,
penDE gene, Eur. J. Biochem. 215,323-332. D~Ez,B., MART~N, J. F. (1989a), Cloning, se-
APLIN,R. T., BALDWIN, J. E., COLE,S. C. J., Su- quence analysis and transcriptional study of the
THERLAND, J. D., TOBIN,M. B. (1993a), On the isopenicillin N synthase of Penicillium chrysoge-
production of a,P-heterodimeric acyl-coenzyme num AS-P-78, Mol. Gen. Genet. 216,91-98.
A:isopenicillin N acyltransferase of Penicillium BARREDO, J. L., DIEz, B., ALVAREZ,E., MART~N,
chrysogenum: studies using a recombinant J. F. (1989b), Large amplification of a 35-kb
source, FEBS Lett. 319, 166-170. DNA fragment carrying two penicillin biosyn-
APLIN,R. T., BALDWIN, J. E., ROACH,P. L., Ro- thetic genes in high penicillin producing strains
BINSON, c. V., SCHOFIELD, c. J. (1993b), Inves- of Penicillium chrysogenum, Curr. Genet. 16,
tigations into the post-translational modification 453459.
and mechanism of acyl-coenzymeAisopenicillin BARREDO, J. L., VANSOLINGEN, P., D~Ez,B., AL-
N acyltransferase using electrospray mass spec- VAREZ, E., CANTORAL, J. M., KAITEVILDER,
trometry, Biochem. J. 294, 357-363. A., SMAAL,E. B., GROENEN,M. A. M., VEEN-
BAILEY,C. R., BUTLER,M. J., NORMANSELL, I. STRA, A. E., MARTfN, J. F. (1989~),Cloning and
D., ROWLANDS,R. T., WINSTANLEY, D. J. characterization of the acyl-coenzyme A:6-ami-
(1984), Cloning a Streptomyces clavuligerus ge- nopenicillanic acid acyltransferase gene of Peni-
netic locus involved in clavulanic acid biosynthe- cillium chrysogenum, Gene 83,291-300.
sis, BiolTechnology 2, 808-811. BERI,R. K., TURNER, G. (1987), Transformation
13 References 269
of P. chrysogenum using the A . nidulans amdS tamdurans are clustered together in an organiza-
gene as a dominant selective marker, Curr. Ge- tion different from the same genes in Acremon-
net. 11, 639-641. ium chrysogenum and Penicillium chrysogenum,
BORK,P., DOOLITTLE, R. F. (1992), Proposed ac- Mol. Microbiol. 5, 1125-1133.
quisition of an animal protein domain by bacte- COQUE,J. J. R., LIRAS,P., MARTIN,J. F. (1993),
ria, Proc. Natl. Acad. Sci. USA 89, 8990-8994. Genes for a plactamase, a penicillin-binding
BRAKHAGE, A. A., VANDENBRULLE, J. (1995), protein and a transmembrane protein are clus-
Use of reporter genes to identify recessive trans- tered with the cephamycin biosynthetic genes in
acting mutations specifically involved in the reg- Nocardia lactamdurans, EMBO J. 12,631- 639.
ulation of Aspergillus nidulans penicillin biosyn- COQUE,J. J. R., LLARENA, F. J. P., ENGUITA, F. J.,
thesis genes, J. Bacteriol. 177(lo), 2781-2788. DE LA FUENTE, J. L., MART~N, J. F., LIRAS,P.
BRUNNER, R., ROHR, M., ZINNER,M. (1968), Zur (1995a), Characterization of the cmcH genes of
Biosynthese des Penicillins, Hoppe-Seyler’s 2. Nocardia lactamdurans and Streptomyces clavu-
Physiol. Chem. 349.95-103. ligerus encoding a functional 3 ’-hydroxymethyl-
BUNNEL,C. A., LUKE,W. D., PERRY,F. M., JR. cephem 0-carbamoyltransferase for cephamycin
(1986), in: PLactam Antibiotics for Clinical Use biosynthesis, Gene 162,21-27.
(QUEENER, S . F., WEBER,J. A., QUEENER,S. COQUE,J. J. R., ENGUITA, F. J., MART~N, J. F., LI-
W., Eds.), pp. 255-284. New York: Marcel Dek- RAS, P. (1995b), A two-protein component 7a-
ker. cephem-methoxylase encoded by two genes of
BURNHAM, M. K. R., HODGSON,J. E., NORMAN- the cephamycin C cluster converts cephalosporin
SELL, I. D. (1987), Isolation and expression of C to 7-methoxycephalosporin C, J. Bacteriol. 177
genes involved in the biosynthesis of p-lactams, (8), 2230-2235.
Eur. Patent Application No. 0233715. CRAWFORD, L., STEPAN,A. M., MCADA,P. C.,
CALVERT, A. F., RODWELL,V. W. (1966), Metab- RAMBOSEK, J. A., CONDER,M. J., VINCI,V. A.,
olism of pipecolic acid in a Pseudomonas sp., J. REEVES,C. D. (1995), Production of cephalo-
Biol. Chem. 241,409-414. sporin intermediates by feeding adipic acid to re-
CANTORAL, J. M., DIEZ, B., BARREDO,J. L., AL- combinant Penicillium chrysogenum strains ex-
VAREZ, E., MART~N, J. E. (1987), High-frequen- pressing ring expansion activity, BiolTechnology
cy transformation of P. chrysogenum, BiolTech- 13, 58-62.
nology 5,494-497. DAVIES,J. (1994), Inactivation of antibiotics and
CANTWELL, C. A., BECKMANN, R. J., DOTZLAF,J. the dissemination of resistance genes, Science
E., FISHER,D. L., SKATRUD, P. L., YEH,W.-K., 264,375-382.
QUEENER, S. W. (1990), Cloning and expression DEMODENA,J. A., GUTIERREZ, S., VELASCO,J.,
of a hybrid S. clavuligerus cefE gene in P. chry- FERNANDEZ, F. J., FACHINI, R. A., GALAZZO,
sogenum, Curr. Genet. 17,213-221. J. L., HUGHES,D. E., MART~N, J. F. (1993), The
CARR,L. G., SKATRUD, P. L., SCHEETZ,M. E., production of cephalosporin C by Acremonium
QUEENER, S. W., INGOLIA, T. D. (1986), Clon- chrysogenum is improved by the intracellular ex-
ing and expression of the isopenicillin N synthet- pression of a bacterial hemoglobin, BiolTechnol-
ase gene from Penicillium chrysogenum, Gene ogy 11, 926-929.
48,257-266. D~Ez,B., BARREDO,J. L., ALVAREZ, E., CANTO-
CHU, Y. W., RENNO,D., SAUNDERS, G. (1995), RAL, J. M., VAN SOLINGEN, P., GROENEN, M.
Detection of a protein which binds specifically to A. M., VEENSTRA, A. E., MARTIN,J. F. (1989),
the upstream region of the pcbAB gene in Peni- Two genes involved in penicillin biosynthesis are
cillium chrysogenum, Curr. Genet. 28(2), 184- linked in a 5.1 kb Sal I fragment in the genome
189. of Penicillium chrysogenum, Mol. Gen. Genet.
COLE,M. (1966), Microbial synthesis of penicillins: 218, 572-576.
Part 1, Process Biochem. 1, 334-338. D~Ez,B., GUTIERREZ,S., BARREDO,J. L., VAN
COQUE,J. J. R.,LIRAS,P., LAIZ,L., MARTIN, J. F. SOLINGEN,P., VAN DER VOORT, L. H. M.,
(1991a), A gene encoding lysine 6-aminotrans- MART~N, J. F. (1990), The cluster of penicillin
ferase, which forms the p-lactam precursor a- biosynthetic genes: Identification and characteri-
aminoadipic acid, is located in the cluster of zation of the pcbAB gene encoding the a-ami-
cephamycin biosynthetic genes in Nocardia lac- noadipyl-cysteinyl-valinesynthetase and linkage
tamdurans, J. Bacteriol. 113, 6258-6264. to the pcbC and penDE genes, J. Biol. Chem.
COQUE,J. J. R., MARTIN,J. F., CALZADA,J. G., 265, 16358-16365.
LIRAS, P. (1991b), The cephamycin biosynthetic DOWSON,C. G., HUTCHISON, A., BRANNIGAN, J.
genes pcbA B, encoding a large multidomain A., GEORGE,R. C., HANSMAN, D., LINARES,J.,
peptide synthetase, and pcbC of Nocardia lac- TOMASZ,A., MAYNARD SMITH,J., SPRATT,B.
G. (1989), Horizontal transfer of penicillin bind- induced death, in: 50 Years of Penicillin Applica-
ing protein genes in penicillin-resistant clinical tion: History and Trends (KLEINKAUF, H., VON
isolates of Streptococcus pneumoniae, Proc. Natl. DOHREN,H., Eds.), Technische Universitat Ber-
Acad. Sci. USA 86,8842-8846. lin, Germany (printed by Public Ltd., Czech Re-
ESPESO,E. A., PENALVA, M. A. (1992), Carbon ca- public), pp. 353-363.
tabolite repression can account for the temporal GOULD,S. J., KELLER,G.-A., HOSKEN,N., WIL-
pattern of expression of a penicillin biosynthetic KINSON, J., SUBRAMANI, S. (1989), A conserved
gene in Aspergillus nidulans, Mol. Microbiol. 6 tripeptide sorts proteins to peroxisomes, J. Cell
( l l ) , 1457-1465. Biol. 108, 1655-1664.
ESPESO,E. A., TILBURN, J., ARST,H. N., JR., PEN- GUTIBRREZ,S., D~Ez, B., MONTENEGRO, E.,
ALVA,M. A. (1993), pH regulation is a major MART~N, J. F. (1991), Characterization of the
determinant in expression of a fungal biosyn- Cephalosporium acremonium pcbAB gene en-
thetic gene, EMBO J. 12, 3947-3956. coding a-aminoadipyl-L-cysteinykbvahe syn-
FAWCETT,P. A., ABRAHAM, E. P. (1976), & ( ~ - a - thetase, a large multidomain synthetase: linkage
Aminoadipy1)cysteinylvaline synthetase, Meth- to the pcbC gene as a cluster of early cephalo-
ods Enzymol. 43,471-473. sporin biosynthetic genes and evidence of multi-
FAWCETT,P. A,, USHER,J. J., ABRAHAM, E. P. ple functional domain, J. Bacteriol. 173, 2354-
(1975), Behavior of tritium-labelled isopenicillin 2365.
N and 6-aminopenicillanic acid acyltransferase GUTIERREZ,S., VELASCO,J., FERNANDEZ, F. J.,
from Penicillium chrysogenum, Biochem. J. 151, MARTfN, J. F. (1992), The cefF gene of Cepha-
741-746. losporium acremonium is linked to the cefEF
FENG,B., FRIEDLIN, E., MARZLUF,G. A. (1994), gene and encodes a deacetylcephalosporin C
A reporter gene analysis of penicillin biosynthe- acetyltransferase closely related to homoserine
sis gene expression in Penicillium chrysogenum 0-acetyltransferase, J. Bacteriol. 174, 3056-
and its regulation by nitrogen and glucose cata- 3064.
bolite repression, App. Environ. Mirobiol. 60 HAAS,H., MARZLUF,G. A. (1995), NRE, the ma-
(12), 4 4 3 2 4 3 9 . jor nitrogen regulatory protein of Penicillium
FENG,B., FRIEDLIN, E., MARZLUF,G. A. (1995), chrysogenum, binds specifically to elements in
Nuclear DNA-binding proteins which recognize the intergenic promoter regions of nitrate assim-
the intergenic control region of penicillin biosyn- ilation and penicillin biosynthetic gene clusters,
thetic genes, Curr. Genet. 27, 351-358. Curr. Genet. 28, 177-183.
FIERRO,F., MONTENEGRO, E., MARCOS,A. T., HAAS,H., BAUER,B., REDL, B., STOFFLER,G.,
MART~N, J. F. (1994), Chromosomal location of MARZLUF,G. A. (1995), Molecular cloning and
the plactam biosynthetic genes in fungal over- anlysis of nre, the major nitrogen regulatory
producing strain, presented at 7th Int. Symp. Ge- gene of P. chrysogenum, Curr. Genet. 27, 150-
net. Ind. Microorg., Montreal, Quebec, Canada, 158.
June 26-July 1, 1994. HARTMAN, B. J., TOMASZ,A. (1984), Low-affinity
FIERRO,F., BARREDO,J. L., D~Ez,B., GUTIBR- penicillin-binding protein associated with P-lac-
REZ, S., FERNANDEZ, F. J., MART~N, J. F. tam resistance in Staphylococcus aureus, J. Bac-
(1995), The penicillin cluster is amplified in tan- teriol. 158, 513-516.
dem repeats linked by conserved hexanucleotide HEINEMANN, J. A., SPRAGUE,G. F. (1989), Bacte-
sequences, Proc. Natl. Acad. Sci. USA 92, 6200- rial conjugative plasmids mobilize DNA transfer
6204. between bacteria and yeast, Nature 340, 205-
FLEMING, A. (1929), On the antibacterial action of 209.
cultures of a Penicillium, with special reference HINNEN, A., HICKS,J. B., FINK,J. R. (1978), Trans-
to their use in the isolation of B. injluenzae, Br. formation of yeast, Proc. Natl. Acad. Sci. USA
J. Exp. Pathol. 10, 226-236. 75,1929-1933.
GARCIA-DOMINGUEZ, M., LIRAS,P., MARTIN,J. HONLINGER, C., KUBICEK,C. P. (1989), Regula-
F. (1991), Cloning and characterization of the tion of &(L-a-aminoadipy1)-L-cysteinyl-D-valine
isopenicillin N synthase gene of Streptomyces and isopenicillin N biosynthesis in Penicillium
griseus NRRL 3851 and studies of expression chrysogenum by the a-aminoadipate pool size,
and complementation of the cephamycin path- FEMS Microbiol. Lett. 65, 71-76.
way in Streptomyces clavuligerus, Antimicrob. HONLINGER, C., HAMPEL,W. A., ROHR,M., Ku-
Agents Chemother. 35,4452. BICEK, C. P. (1988), Differential effects of gener-
GIESBRECHT, P., FRANZ,M., KRUGER,D., LABIS- al amino acid control of lysine biosynthesis on
CHINSKI, H., WECKE,J. (1993), Bacteriolysis of penicillin formation in strains of Penicillium
Staphylococci is only a side-effect of penicillin- chrysogenum, J. Antibiot. 41, 255-257.
I3 References 271
HORI,K., YAMAMOTO, Y., TOKITA,K., SAITO,F., (cefEF)from Streptomyces clavuligerus: gene du-
KUROTSU, T., KANDA,M., OKAMURA, K., Fu- plication may have led to separate hydroxylase
RUYAMA, J., SAITO,Y. (1991), The nucleotide- and expandase activities in the actinomycetes, J.
sequence for a proline-activating domain of Bacteriol. 173, 398-400,
gramicidin-S-synthetase-2 gene from Bacillus KOVACEVIC, S., WEIGEL,B. J., TOBIN,M. B., IN-
brevis, J. Biochem. 110, 111-119. GOLIA, T. D., MILLER,J. R. (1989), Cloning,
HOSKINS, J., ~'CALLAGHAN, N., QUEENER, S. W., characterization, and expression in Escherichia
CANTWELL, C. A., WOOD, J. S., CHEN,V. J., coli of the Streptomyces clavuligerus gene encod-
SKATRUD, P. L. (1990), Gene disruption of the ing deacetoxycephalosporin C synthetase, J.
pcbAB gene encoding ACV synthetase in Ce- Bacteriol. 171, 754-760.
phalosporium acremonium, Curr. Genet. 18, KOVACEVIC, S., TOBIN,M. B., MILLER,J. R.
523-530. (1990), The p-lactam biosynthesis genes for iso-
HUFFMAN, G. W., GESELLSCHEN, P. D., TURNER, penicillin N epimerase and deacetoxycephalo-
J. R., ROTHENBERGER, R. B., OSBORNE, H. E., sporin C synthetase are expressed from a single
MILLER,F. D., CHAPMAN, J. L., QUEENER, S. transcript in Streptomyces clavuligerus, J. Bacte-
W. (1992), Substrate specificity of isopenicillin N riol. 172, 3952-3958.
synthetase, J. Med. Chem. 35(10), 1897-1914. KRAUSE,M., MARAHIEL, M. A. (1988), Organiza-
INGOLIA, T. D., QUEENER, S. W. (1989), PLactam tion of the biosynthesis genes for the peptide an-
biosynthetic genes, Med. Res. Rev. 9 (2), 245- tibiotic gramicidin S, J. Bacteriol. 170, 4669-
264. 4674.
JENSEN,S. E. (1985), Biosynthesis of cephalospo- KRAUSE,M., MARAHIEL,M. A., VON DOHREN,
rins, Crit. Rev. Biotechnol. 3, 277-301. H., KLEINKAUF, H. (1985), Molecular cloning of
JENSEN,S. E., WESTLAKE, D. W. S., WOLFE,S. an ornithine-activating fragment of the gramici-
(1988), Production of the penicillin precursor 6 din-S-synthetase-2-genefrom Bacillus brevis and
(L-a-aminoadipyl)-L-cysteinyl-D-valine(ACV) its expression in Escherichia coli, J. Bacteriol.
by cell-free extracts from Streptomyces clavulige- 162, 1120-1125.
rus, FEMS Microbiol. Lett. 49, 213-218. KRIAUCIUNAS, A., FROLIK,C. A., HASSELL, T. C.,
JENSEN,S. E., WONG,A., ROLLINS,M. J., WEST- SKATRUD, P. L., JOHNSON, M. G., HOLBROOK,
LAKE,D. W. S. (1990), Purification and partial N. L., CHEN,V. J. (1991), The functional role of
characterization of S(L-a-aminoadipy1)-L-cyste- cysteines in isopenicillin N synthase, J. Biol.
inyl-D-Valine synthetase from Streptomyces cla- Chem. 266, 11779-11 788.
vuligerus, J. Bacteriol. 172, 7269-7271. KUMAR,C. V., COQUE,J.-J. R., MART~N, J. F.
KALLOW,W., VON DOHREN,H., KENNEDY, J., (1994), Efficient transformation of the cephamy-
TURNER,G. (1994), Structure and functional cin C producer Nocardia lactamdurans and de-
study of ACV synthetase, 7th Int. Congr. Bacte- velopment of shuttle and promoter-probe clon-
rial, Appl. Microbiol. Mycol. Prague, Czech Re- ing vectors, Appl. Environm. Microbiol. 60(11),
public, p. 426. 4068-4093.
KERN,B. A., HENDLIN, D., INAMINE, E. (1980), L- KUROTSO, T., HORI,K., SAITO,Y. (1991), Charac-
Lysine-E-aminotransferaseinvolved in cephamy- terization and location of the L-proline activat-
cin biosynthesis, in Streptomyces lactamdurans, ing fragment from the multifunctional gramici-
Antimicrob. Agents Chemother. 17,679-685. d i n 4 synthetase-2, J. Biochem. 109, 763-769.
KIMURA, H., MIYANOSHITA, H., SUMINO,Y. KURYLOWICZ, W., KURZATKOWSKI, W., KURZAT-
(1990), DNA and applications thereof, Jpn. Pa- KOWSKI, J. (1987), Biosynthesis of benzylpenicil-
tent Application No. 2-3762. lin by Penicillium chrysogenum and its Golgi ap-
KINASHIH., DOI, M., NIMI,0. (1995), Isolation of paratus, Arch. Immunol. Ther. Exp. 35, 699-
large linear plasmids from p-lactam producing 724.
actinomycete strains, Biotechnol. Lett. 17(3), LEDERBERG, J. (1957), Mechanism of action of
243-246. penicillin, J. Bacteriol. 73, 144.
KIRKPATRICK,J. R., DOOLIN, L. E., GODFREY,L. LENDENFELD, T., GHALI,D., WOLSCHEK, M., Ku-
W. (1973), Lysine biosynthesis in Streptomyces BICEK-PRANZ, E. M., KUBICEK,C. P. (1993),
lipmannii: implications in antibiotic biosynthesis, Subcellular compartmentation of penicillin bio-
Antimicrob. Agents Chemother. 4, 542-550. synthesis in Penicillium chrysogenum: the amino
KLEINKAUF, H., VON DOHREN,H. (1996), A non- acid precursors are derived from the vacuole, J.
ribosomal system of peptide biosynthesis, Eur. J. Biol. Chem. 268,665471.
Biochem. 236,335-351. LESKIW,B. K., AHARONOWITZ, Y., MEVARECH,
KOVACEVIC,S., MILLER,J. R. (1991), Cloning and M., WOLFE,S., VINING,L. C., WESTLAKE, D.
sequencing of the Plactam hydroxylase gene W. S., JENSEN,S. E. (1988), Cloning and nucleo-
tide sequence of the IPNS gene from Strepto- cephalosporin antibiotic biosynthesis, in: Molec-
myces clavuligerus, Gene 62, 87-196. ular Industrial Mycology: Systems and Applica-
LODER,P. B., ABRAHAM, E. P. (1971a), Isolation tions for Filamentous Fungi (LEONG,S . A., BER-
and nature of intracellular peptides from a ceph- KA, R. M., Eds.), pp. 149-196. New York Mar-
alosporin C-producing Cephalosporium sp., Bio- cel Dekker.
chem. J. 123,471-476. MATHISON,L., SOLIDAY,C., STEPAN,T., AL-
LODER,P. B., ABRAHAM, E. P. (1971b), Biosyn- DRICH, T., RAMBOSEK, J. (1993), Cloning, char-
thesis of peptides containing a-aminoadipic acid acterization, and use in strain improvement of
and cysteine in extracts of a Cephalosporium sp., the C. acremonium gene cefG encoding acetyl
Biochem. J. 123,477-482. transferase, Curr. Genet. 23, 33-41.
LUENGO,J. M., IRISO,J. L., L6PEZ-NIET0, M. J. MATSUHASHI, M., SONG,M. D., ISHINO,F., WA-
(1986), Direct enzymatic synthesis of natural CHI, M., Dot, M., INOUE,M., UBUKATA,K.,
penicillins using phenylacetyl-CoA:6-APA phe- YAMASHITA, N., KONNO,M. (1986), Molecular
nylacetyl transferase of Penicillium chrysoge- cloning of the gene of a penicillin-binding pro-
num: minimal and maximal side chain require- tein supposed to cause high resistance to p-lac-
ments, J. Antibiot. 39, 1754-1759. tam antibiotics in S. aureus, J. Bacteriol. 167,
MACCABE,A. P., VAN LIEMPT,H., PALISSA,H., 975-980.
UNKLES,S. E., RIACH,M. B. R., PFEIFER,E., MCDERMOTI-, W., ROGERS,D. E. (1982), Social
VON DOHREN,H., KINGHORN,J. R. (1991), S ramifications of control of microbial disease,
(L-a-Aminoadipyl)-L-cysteinyl-D-valine synthet- Johns Hopkins Med. J. 151, 302-312.
ase from Aspergillus nidulans, J. Biol. Chem. MENNE,S., WALZ,M., KUCK, U. (1994), Expres-
266,12646-12654. sion studies with the bidirectional pcbAB-pcbC
MADDURI,K., STUTTARD,C., VINING,L. C. promoter region from Acremonium chrysoge-
(1989). Lysine catabolism in Streptomyces spp. is num using reporter gene fusions, Appl. Micro-
primarily through cadverine: p-lactam producers biol. Biotechnol. 42, 57-66.
also make a-aminoadipate, J. Bacteriol. 171, MILLER,J. R., INGOLIA, T. D. (1989), Cloning p-
299-302. lactam genes from Streptomyces spp. and fungi,
MADDURI,K., STUTTARD,C., VINING,L. C. in: Genetics and Molecular Biology of Industrial
(1991), Cloning and location of a gene governing Microorganisms (HERSHBERGER, C . L.,
lysine E-aminotransferase, an enzyme initiating QUEENER,S. W., HEGEMAN, G., Eds.), pp. 246-
p-lactam biosynthesis in Streptomyces sp., J. 255. Washington, D C American Society for Mi-
Bacteriol. 173, 985-988. crobiology.
MAEDA, K., LUENGO,J. M., FERRERO,O., MONTENEGRO, E., BARREDO, J. L., G U T I ~ R R E Z ,
WOLFE,S., LEBEDEV,M. Y., FANG,A., DE- S., DLEz, B., ALVAREZ,E., MARTIN,J. F.
MAIN, A. L. (1995), The substrate specificity of (1990), Cloning, characterization of the acyl-
deacetoxycephalosporin C synthase (“expand- CoA:6-aminopenicillianic acid acyltransferase
ase”) of Streptomyces clavuligerus is extremely gene of Aspergillus nidulans and linkage to the
narrow, Enzyme Microb. Technol. 17, 231-234. isopenicillin N synthase gene, Mol. Gen. Genet.
MALMBERG, L. H., Hu, W.-S., SHERMAN, D. H. 221,322-330.
(1993), Precursor flux control through targeted MONTENEGRO, E., FIERRO,F., FERNANDEZ, F. J.,
chromosomal insertion of the lysine &-amino- GUTIERREZ, s., MARTIN, J. F. (1992), Resolu-
transferase (LAT) gene in cephamycin C biosyn- tion of chromosomes 111 and VI of Aspergillus
thesis, J. Bacteriol. 175, 6916-6924. nidulans by pulse-field gel electrophoresis shows
MALMBERG, L. H., Hu, W. S., SHERMAN, D. H. that the penicillin biosynthetic pathway genes
(1995), Effects of enhanced lysine E-aminotrans- pcbAB, pcbC, andpenDE are clustered on chro-
ferase activity on cephamycin biosynthesis in mosome VI (3.0 Megabases), J. Bacteriol. 174,
Streptomyces clavuligerus, Appl. Microbiol. Bio- 7063-7067.
technol. 44,198-205. MULLER,W. H., VAN DER KRIFT,T. P., KROUW-
MALPARDITA, F., HOPWOOD, D. A. (1984), Molec- ER, A. J. J., WOSTEN, H. A. B., VAN DER
ular cloning of the whole biosynthetic pathway VOORT,L. H. M., SMAAL,E. B., VERKLEIJ, A.
of a Streptomyces antibiotic and its expression in J. (1991), Localization of the pathway of the
a heterologous host, Nature 309,462-464. penicillin biosynthesis in Penicillium chrysoge-
MARTIN, J. F., GUTI~RREZ, S. (1995), Genes for p- num, EMBO J. la, 489-495.
lactam antibiotic biosynthesis, Antonie van MULLER,W. H., BOVENBERG, R. A. L., GROOT-
Leeuwenhoek 67, 181-200. HUIS,M. H., KATTEVILDER, F., SMAAL,E. B.,
MARTIN, J. F., INGOLIA,T. D., QUEENER,s. W. VAN DER VOORT, L. H. M., VERKLEIJ, A. J.
(1991), Molecular genetics of penicillin and (1992), Involvement of microbodies in penicillin
I3 References 213
biosynthesis, Biochim. Biophys. Acta 1116, 210- ROACH,P. L., CLIFTON,I. J., FULOP, V., HARLOS,
213. K., BARTON,G. J., HAJDU,J., ANDERSSON,I.,
O'SULLIVAN, J., SYKES,R. B. (1986), PLactam an- SCHOFIELD,C. J., BALDWIN,J. E. (1995), Crys-
tibiotics, in: Biotechnology 1st. Edn., Vol. 4 tal structure of isopenicillin N synthase is the
(REHM, H.-J., REED, G., Eds.), pp. 247-273. first from a new structural family of enzymes,
Weinheim: VCH. Nature 375, 700-704.
O'SULLIVAN,J., BLEANEY,R. C., HUDDLESTON, ROLAND,I., FROYSHOV,O., LALAND,G. (1975),
J. A., ABRAHAM, E. P. (1979), Incorporation of On the presence of pantothentic acid in the
3H from &(~-(~-amino(4,5-3H)adipyl)-~-cystei- three complementary enzymes of bacitracin syn-
nyl-D-(4,4-3H)valine into isopenicillin N, Bio- thetase, FEBS Lett. 60,305-308.
chem. J. 184,421-426. SAITO,F., HORI, K., KANDA,M., KUROTSU,T.,
PANG,C. P., CHAKRAVARTI, B., ADLINGTON, R. SAITO,Y. (1995), Entire nucleotide sequence for
M., TING,H. H., WHITE,R. L., JAYATILAKE, G. Bacillus brevis Nagano Grs2 gene encoding
S., BALDWIN, J. E., ABRAHAM, E. P. (1984), Pu- gramicidin S synthetase 2: a multifunctional pep-
rification of isopenicillin N synthetase, Biochem. tide synthetase, J. Biochem. 116, 357-367.
J. 222, 789-795. SANCHEZ,F., LOZANO,M., RUBIO,V., PENALVA,
PASTEUR,L., JOUBERT,J. (1877), Charbon et sep- M. A. (1987), Transformation of P. chrysoge-
ticemie, C. R. Acad. Sci. Paris 85, 101-115. num, Gene 51,97-102.
PENALVA,M. A., TOURINO, A., PATINO,C., SAN- SAMSON,S . E., BELAGAJE,R., BLANKENSHIP, D.
CHEZ, F., FERNANDEZ SOUSA, J. M., RUBIO,V. T., CHAPMAN,J. L., PERRY,D., SKATRUD,P.
(1985), Studies of transformation of Cephalospo- L., VAN FRANK,R. M., ABRAHAM, E. P.,
rium acremonium, in: Molecular Genetics of Fil- BALDWIN, J. E., QUEENER,S. W., INGOLIA,T.
amentous Fungi (TIMBERKLAKE, W. E., Ed.), D. (1985), Isolation, sequence determination
pp. 59-68. New York: Alan Liss. and expression in E. coli of the isopenicillin N
P~REZ-ESTEBAN, B., OREJAS,M., G6MEZ-PAR- synthetase gene from Cephalosporium acremon-
DO, E., PENALVA, A. (1993), Molecular charac- ium, Nature (London) 318, 191-194.
terization of a fungal secondary metabolism pro- SAMSON,S. M., DOTZLAF,J. E., SLISZ, M. L.,
moter: transcription of the Aspergillus nidulans BECKER,G. W., VAN FRANK,R. M., VEAL,L.
isopenicillin N synthetase gene is modulated by E., YEH, W. K., MILLER,J. R., QUEENER,S. W.,
upstream negative elements, Mol. Microbiol. 9 INGOLIA,T. D. (1987), Cloning and expression
(4), 881-895. of the fungal expandaselhydroxylase gene in-
PETRICH,A. K., JENSEN,S. E. (1994), Transcrip- volved in cephalosporin biosynthesis, BiolTech-
tional regulation of the genes involved in the nology 5,1207-1214.
early steps of the cephamycin biosynthetic path- SAWADA,Y., BALDWIN, J. E., SINGH,P. D., SOLO-
way in Streptomyces clavuligerus, presented at MON,N. A., DEMAIN,A. L. (1980), Cell-free cy-
7th Int. Symp. Genet. Ind. Microorg. Montreal, clization of &(L-cY-aminoadipyl)-L-cysteinyl-D-
Canada, June 26-July 1, 1994. valine to isopenicillin N, Antimicrob. Agents
PIEPER,R., HAESE,A., SCHRIDER,W., ZOCHER, Chemother. 18,456-470.
R. (1995), Arrangement of catalytic sites in the SCHLUMBOHM, W., STEIN,T., ULLRICH,C., VA-
multifunctional enzyme enniatin synthetase, TER, J., KRAUSE, M., MARAHIEL,M. A.,
Eur. J. Biochem. 230, 119-126. KRUFI-,V., WI~MANN-LIEBOLD, B. (1991), An
QUEENER,S. W., NEUSS,N. (1982), The biosynthe- active serine is involved in covalent substrate
sis of p-lactam antibiotics, in: The Chemistry and amino acid binding at each reaction center of
Biology of p-Lactam Antibiotics (MORIN,R. B., gramicidin S synthetase, J. Biol. Chem. 266,
MORGAN,M., Eds.), pp. 1-81. London: Aca- 23 135-23 141.
demic Press. SCHWECKE, T., AHARONOWITZ, Y., PALISSA,H.,
QUEENER,S. W., INGOLIA, T. D., SKATRUD, P. L., VON DOHREN, H., KLEINKAUF,H., VAN
CHAPMAN,J. L., KASTER,K. R. (1985), A sys- LIEMFT,H. (1992a), Enzymatic characterization
tem for genetic transformation of Cephalospo- of the multifunctional enzyme G(L-a-aminoadi-
rium acremonium, in: Microbiology I985 py1)-L-cysteinyl-D-valine synthetase from Strep-
(LIEVE, L., Ed.), pp. 468-472. Washington, DC: tomyces clavuligerus, Eur. J. Biochem. 205, 687-
American Society for Microbiology. 694.
RAM6N, D., CARRAMOLINO, L., PATINO, c., SAN- SCHWECKE,T., TOBIN,M.,KOVACEVIC,S., MIL-
CHEZ,F., PENALVA,M. A. (1987), Cloning and LER, J. R., SKATRUD,P. L., JENSEN, S. E.
characterization of the isopenicillin N synthetase (1992b), The A module of ACV synthetase: ex-
gene mediating the formation of the plactam pression, renaturation and functional characteri-
ring in Aspergillus nidulans, Gene 57, 71-181. zation, in: Industrial Microorganisms: Basic and
274 6 Advances in the Molecular Genetics of p-.Lactam Antibiotic Biosynthesis
Applied Molecular Genetics (BALTZ, R. H., SKATRUD,P. L., QUEENER,S. W., CARR,L. G.,
HEGEMAN, G. E., SKATRUD, P. L., Eds.), p. 291. FISHER, D. L. (1987b), Efficient integrative
Washington D C American Society for Micro- transformation of C. acremonium, Curr. Genet.
biology. 12,337-348.
SENO,E. T., BALTZ,R. H. (1989), Structural organ- SKATRUD, P. L., TIETZ, A. J., INGOLIS,T. D.,
ization and regulation of antibiotic synthesis and CANTWELL, C. A., FISHER,D. L., CHAPMAN, J.
resistance genes in Actinomycetes, in: Regula- L., QUEENER, S. W. (1989), Use of recombinant
tion of Secondary Metabolism in Actinomycetes DNA to improve production of cephalosporin C
(SHAPIRO, S., Ed.), pp. 1-48. Boca Raton, FL: in C. acremonium, Bio/Technology 7, 411-485.
CRC Press. SMITH,D. J., BULL,J. H., EDWARDS, J., TURNER,
SHIAU,C. Y., BALDWIN, J. E., BYFORD,M. F., So- G. (1989), Amplification of the isopenicillin N
BEY,w. J., SCHOFIELD, C. J. (1995), &(L-cr-Am- synthetase gene in a strain of Penicillium chryso-
inoadipyl)-L-cysteinyl-D-vahe synthetase: the genum producing high levels of penicillin, Mol.
order of peptide bond formation and timing of Gen. Genet. 216,492-497.
the epimerisation reaction, FEBS Lett. 358, 97- SMITH,D. J., BURNHAM, M. K. R., BULL,J. H.,
100. HODGSON,J. E., WARD, J. M., BROWNE,P.,
SHIFFMAN, D., MEVARECH,M., JENSEN,S. E., BROWN,J., BARTON,B., EARL,A. J., TURNER,
COHEN,G., AHARONOWITZ, Y. (1988), Cloning G. (1990a), PLactam antibiotic biosynthetic
and comparative sequence analysis of the gene genes have been conserved in prokaryotes and
coding for isopenicillin N synthase in Strepto- eukaryotes, EMBO J. 9,741-747.
myces, Mol. Gen. Genet. 214, 562-569. SMITH,D. J., BURNHAM, M. K. R., EDWARDS, J.,
SHIFFMAN, D., COHEN,G., AHARONOWITZ, Y., EARL,A. J., TURNER,G. (1990b), Cloning and
PALISSA,H., VON DOHREN,H., KLEINKAUF, expression of the penicillin biosynthetic gene
H., MEVARECH, M. (1990), Nucleotide sequence cluster from Penicillium chrysogenum, BiolTech-
of the isopenicillin N synthase gene (pcbC) of nology 8,39-41.
the gram-negative Flavobacterium sp. SC 12,154, SMITH, D. J., EARL, A. J., TURNER,G. (1990c),
Nucleic Acids Res. 18, 660. The multifunctional peptide synthetase perform-
ing the first step in penicillin biosynthesis in Pen-
SHOCKMAN, G. D., DANEO-MOORE, L., McDow- icillium chrysogenum is a 421 037 Dalton protein
ELL,T. D., WONG,W. (1982), The relationship
similar to Bacillus brevis peptide antibiotic syn-
between inhibition of cell wall synthesis and bac-
thetases, EMBO J. 9, 2743-2750.
terial lethality, in: The Chemistry and Biology of
PLactam Antibiotics, Vol. 3 (MORIN,R. B., SPRATT,B. G. (1989), Resistance to p-lactam anti-
GORMAN, M., Eds.), pp. 303-338. London: Aca- biotics mediated by alterations of penicillin-
demic Press. binding proteins, in: Handbook of Experimental
Pharmacology, Vol. 91 (BRYAN,L. E., Ed.), pp.
SIKORSKI,R. S., MICHAUD,W., LEVIN,H. L., 77-100. Berlin: Springer-Verlag.
BOEKE,J. D., HIETER,P. (1990), Trans-king- STACHELHAUS, T., SCHNEIDER, A., MARAHIEL,
dom promiscuity, Nature 345,581-582. M. A. (1995), Rational design of peptide anti-
SKATRUD, P. L. (1991), Molecular biology of the biotics by targeted replacement of bacterial and
plactam producing fungi, in: More Gene Mani- fungal domains, Science 269, 69-72.
pulations in Fungi (BENNET,J. W., LASURE,L. THENBERGH,K., LITZKA,O., BRAKHAGE, A. A.
L., Eds.), pp. 364-395. New York: Academic (1996), Identification of a major cis-acting DNA
Press. element controlling the bidirectionally tran-
SKATRUD, P. L. (1992), Genetic engineering of p scribed penicillin biosynthesis genes acvA
lactam antibiotic biosynthetic pathways in fil- (pcbAB) and ipnA (pcbC) of Aspergillus nidu-
amentous fungi, Trends BioTechnol. 10(9), 324- lans, J. Bacteriol. 178(13), 3908-3916.
329. TILBURN, J., SARKAR, S., WIDDICK, D. A., ESPE-
SKATRUD, P. L., QUEENER, S. W. (1989), An elec- SA, E. A., OREJAS,M., MUNGROO, J., PENAL-
trophoretic molecular karyotype for an indus- VA,M. A., ARST,H. N., JR. (1995), The Asper-
trial strain of Cephalosporium acremonium, gillus PacC zinc finger transcription factor me-
Gene 78,331-338. diates regulation of both acidic- and alkaline-ex-
SKATRUD, P. L., FISHER,D. L., CHAPMAN, J. L., pressed genes by ambient pH, EMBO J. 14,779-
CANTWELL, C. A., QUEENER,S. W. (1987a), 790.
Strain improvement studies in P. chrysogenum TIPPER,D. J., STROMINGER, J. L. (1965), Mecha-
using the cloned P. chrysogenum isopenicillin N nism of action of penicillins: a proposal based on
synthetase gene and the amdS gene of Aspergil- their structural similarity to acyl-D-alanyl-D-ala-
lus nidulans, SIM News 37(4), 77. nine, Proc. Natl. Acad. Sci. USA 54, 1133-1141.
13 References 275
TOBIN,M. B. (1994), Genetic engineering of the valine synthetase from Aspergillus nidulans, J.
acyl-coenzyme A:isopenicillin N acyltransferase Biol. Chem. 264,3680-3684.
from Penicillium chrysogenum, Thesis, Univer- VAN SINDREN, D., GALLI,G., COSMINA,F., DE-
sity of Oxford. FERRA,S., WITTHOFF,S. VENEMA,G., GRAN-
TOBIN,M. B., FLEMING,M. D., SKATRUD, P. L., DI,G. (1993), Characterization of the srfA locus
MILLER,J. R. (1990), Molecular characteriza- of Bacillus subtilis: only the valine-activating do-
tion of the acyl-coenzyme A:isopenicillin N acyl- main of srfA is involved in the establishment of
transferase gene (penDE)from Penicillium chry- genetic competence, Mol. Microbiol. 8, 833-
sogenum and Aspergillus nidulans and activity of 841.
recombinant enzyme in Escherichia coli, J. Bac- VELASCO,J., GUTIERREZ,S., FERNANDEZ, F. J.,
teriol. 172, 5908-5914. MARCOS,A. T., ARENOS,C., MART~N, J. F.
TOBIN, M. B., KOVACEVIC, S., MADDURI, K., Hos- (1994), Exogenous methionine increases levels
KINS, J., SKATRUD, P. L., VINING,L. C., STUD- of mRNAs transcribed from pcbAB, pcbC, and
DARD,C., MILLER,J. R. (1991), Localization of cefEF genes, encoding enzymes of the cephalo-
the lysine E-aminotransferase (lat) and S(L-(Y- sporin biosynthetic pathway, in Acremonium
aminoadipyl)-L-cysteinyl-D-valine synthetase chrysogenum, J. Bacteriol. 176(4), 985-991.
(pcbAB) genes from Streptomyces clavuligerus VICHITSOONTHONKUL, T., CHU, Y. W., SODHI,H.
and production of lysine e-aminotransferase ac- S., SAUNDERS,G. (1994), p-Lactam antibiotics
tivity in E. coli, J. Bacteriol. 173, 6223-6229. produced by genetically engineered filamentous
TOBIN,M. B., BALDWIN,J. E., COLE, S. C. J., fungi, in: Recombinant Microbes for Industrial
MILLER,J. R., SKATRUD, P. L., SUTHERLAND, and Agricultural Applications (MUROOKA,Y.,
J. D. (1993), The requirement for subunit inter- IMANAKA, E., Eds.), pp. 119-135. New York:
action in the production of Penicillium chrysoge- Marcel Dekker.
num acyl-coenzyme A:isopenicillin N acyltrans- VINING,L. C., SHAPIRO,S., MADDURI,K., STUT-
ferase in Escherichia coli, Gene 132, 199-206. TARD,C. (1990), Biosynthesis and control of p-
TOBIN,M. B., COLE, S. C. J., KOVACEVIC,S., lactam antibiotics: the early steps in the “classi-
MILLER,J. R., BALDWIN, J. E., SUTHERLAND, J. cal” tripeptide pathway, Biotechnol. Adv. 8,159-
D. (1994), Acyl-coenzyme A:isopenicillin N 183.
acyltransferase from Penicillium chrysogenum: WARD,J. M., HODGSON,J. E. (1993), The biosyn-
effect of amino acid substitutions at Ser”’, thetic genes for clavulanic acid and cephamycin
Ser’” and Ser3won proenzyme cleavage and ac- production occur as a “super-cluster” in three
tivity, FEMS Microbiol. Lett. 121, 9-46. Streptomyces, FEMS Microbiol. Lett. 110, 239-
TOBIN,M. B., MILLER,J. R., BALDWIN, J. E., Su- 242.
THERLAND, J. D. (1995), Amino acid substitu- WAXMAN, D. J., STROMINGER, J. L. (1982), p-Lac-
tions in the cleavage site of acyl-coenzymeA:iso- tam antibiotics: biochemical modes of action, in:
penicillin N acyltransferase from Penicillium The Chemistry and Biology of p-Lactam Anti-
chrysogenum: effect on proenzyme cleavage and biotics, Vol. 3 (MORIN,R. B., GORMAN,M.,
activity, Gene 162, 29-35. Eds.), pp. 209-285. London: Academic Press.
TURGAY,K., KRAUSE,M., MARAHIEL,M. A. WEBER,G., SCH~RGENDORFER, K., SCHNEIDER-
(1992), Four homologous domains in the prima- SCHERZER, E., LEITNER,E. (1994), The peptide
ry structure of GrsB are related to domains in a synthetase catalyzing cyclosporine production in
superfamily of adenylate forming enzymes, Mol. Tolypocladium niveum is encoded by a giant
Microbiol. 6, 529-546. 45.8-kilobase open reading frame, Curr. Genet.
TURNER,G. (1992), Genes for the biosynthesis of 26, 120-125.
plactam compounds in microorganisms, in: Sec- WEIGEL,B. J., BURGETT,S. G., CHEN,V. J., SKA-
ondary Metabolites: Their Function and Evolu- TRUD,P. L., FROLIK,C. A., QUEENER,S. W.,
tion (CHADWICK, D. J., WHELAN,J., Eds.), pp. INGOLIA,T. D. (1988), Cloning and expression
113-128. Chichester: Wiley and Sons (Ciba in Escherichia coli of isopenicillin N synthetase
Foundation Symposium 171). genes from Streptomyces lipmanii and Aspergil-
UNAL,S., HOSKINS,J., FLOKOWITSCH, J. E., Wu, lus nidulans, J. Bacteriol. 170, 3817-3826.
C. Y. E., PRESTON,D. A., SKATRUD,P. L. WHITEMAN, P. A., ABRAHAM, E. P., BALDWIN, J.
(1992), Detection of methicillin-resistant staphy- E., FLEMING,M. D., SCHOFIELD, C. J., SUTHER-
lococci by using the polymerase chain reaction, LAND, J. D., WILLIS,A. C. (1990), Acyl coen-
J. Clin. Microbiol. 30,1685-1691. zyme A:6-aminopenicillanic acid acyltransferase
VAN LIEMPT,H., VON DOHREN,H., KLEINKAUF, from Penicillium chrysogenum and Aspergillus
H. (1989), G(L-a-Aminoadipyl)-L-cysteinyl-D- nidulans, FEBS Lett. 262, 342-344.
WISE,E. M., JR., PARK,J. T. (1965), Penicillin: its STROMINGER, J. L. (1979), Mechanism of peni-
basic site of action as an inhibitor of a peptide cillin action: penicillin and substrate bind coval-
cross-linking reaction in cell wall mucopeptide ently to the same active site serine in two bacte-
synthesis, Proc. Natl. Acad. Sci. USA 54, 75-81. rial D-alanine carboxypeptidases, Proc. Natl.
Wu, C. Y. E., HOSKINS,J., BLASZCZAK,L. C., Acad. Sci. USA 76,2730-2734.
PRESTON,D. A., SKATRUD,P. L. (1992), Con- ZHANG,J. (1991), ACV synthetase in cephalsporin
struction of a water-soluble form of penicillin- biosynthesis, Thesis, Massachussets Institute of
binding protein 2a from a methicillin-resistant Technology, Cambridge, MA.
Staphylococcus aureus isolate, A ntimicrob. ZHANG,J., DEMAIN, A. L. (1990a), Purification
Agents Chemother. 36,533-539. from Cephalosporium acremonium of the initial
YEH, W. K., DOTZLAF, J. E., HUFFMAN,G. W. enzyme unique to the biosynthesis of penicillins
(1994), Biochemical characterization and evolu- and cephalosporins, Biochem. Biophys. Res.
tionary implication of plactam expandaselhy- Commun. 137,528-535.
droxylase, expandase and hydroxylase, in: 50 ZHANG,J., DEMAIN, A. L. (1990b), Purification of
Years of Penicillin Application; History and ACV synthetase from Streptomyces clavuligerus,
Trends (KLEINKAUF,H., VON DOHREN, H., Biotechnol. Lett. 12, 649454.
Eds.), pp. 208-223. Prague: Public. ZHANG,J., DEMAIN, A. L. (1992), ACV synthet-
YELTON,M. M., HAMER,J. E., TIMBERLAKE, W. ase, Crit. Rev. Biotechnol. 12, 245-260.
E. (1984), Transformation of Aspergillus nidu- ZHANG,J., WOLFE,S., DEMAIN, A. L. (1992), Bio-
fans using a trpC plasmid, Proc. Natl. Acad. Sci. chemical studies of ACV synthetase activity
USA 81,1470-1474. from Streptomyces clavuligerus, Biochem. J. 383,
YOCUM,R. R., WAXMAN,D. J., RASMUSSEN, J. R., 691698.
Biotechnology
7 Peptide Antibiotics
HORSTKLEINKAUF
HANSVON DOHREN
Berlin, Germany
List of Abbreviations 278

1 Introduction 279
2 Structures 280
2.1 Ribosomal Origin 280
2.1.1 Structural Classifications 280
2.1.2 Similar Structures of Ribosomal and Nonribosomal Peptides - Monocyclic Struc-
tures 280
2.1.3 D-Amino Acids in Ribosomally Produced Peptides 282
2.1.4 Posttranslational Modifications of Side Chains 282
2.1.5 Multicyclic Peptides 283
2.1.6 Linear Peptides 284
2.1.7 Production of Ribosomally Formed Peptides 284
2.1.8 Similar Open Questions in Gene-Encoded and Enzymatically Formed Pep-
tides 284
2.2 Nonribosomal Origin 285
3 Biotechnology of Peptides 285
3.1 Biosynthesis 285
3.1.1 Biosynthetic Gene Clusters 287
3.1.2 Biosynthetic Modules 287
3.1.3 Biosynthetic Enzyme Systems 293
3.2 Production 293
3.2.1 Fermentation Procedures 293
3.2.2 Structural Alterations 295
3.2.2.1 Family Exploitations 295
3.2.2.2 Biosynthetic Manipulations 297
4 Future Prospects 298
4.1 Peptides of Ribosomal Origin 298
4.2 Peptides of Nonribosomal Origin 299
5 Compilation of Compounds 299
6 References 316
278 7 Peptide Antibiotics
aThr allo-threonine
List of Abbreviations Aze azetidine-2-carboxylic acid
Bmt (4R)-4-[(4E)-Zbutenyl]-4-
Note: If no D- or D-prefix is given, all ami- methyl-L-threonine
no acids are in the L-configuration. C
- cysteine linked as thio-
ether
l-C1-D-vinylGly 1-chloro-D-vinylglycine Cit citrulline
2a3h4buOA 2-amino-3-hydroxy-4- CPG cyclopropyl-glycine
butyl-octanoic acid cyclodihydroBmt cyclodehydro-Bmt
2a4m4HEA 2-amino-4-methyl-hex-4- (see Bmt)
enoic acid cyclopropylGly cyclopropyl-glycine
2a3h4mOA 2-amino-3-hydroxy-4- deoxyBmt deoxy-Bmt (see Bmt)
methyl-octanoic acid Dbu diaminobutyric acid
2a3h4,8m2NA 2-amino-3-hydroxy-4,8- fIle formyl-isoleucine
dimethyl-nonanoic acid Nal formyl-valine
2-Cl-~Ala 2-chloro-~-alanine fOrn f-ornithine
2-F-~Ala 2-fluoro-~-alanine Himv 2-hydroxy-4-methyl-valeric
2hAsn 2-hydrox y-asparagine acid
2hBu 2-hydroxy-butyric acid Hiv hydroxy-isovaleric acid
2hiCap 2-hydroxy-isocaproic acid HmP hydroxy-methyl-pentanoic
2hPhe 2-h ydroxy-phen ylalanine acid
2hVal 2-hydroxy-valine hPro 4-hydroxy-proline
2h3mVa 2-hydroxy-3-methyl-valeric HYP 4-h ydroxy-proline
acid Lac lactic acid
2,3h3mP 2,3-hydroxy-3-methyl- LYSA lysergic acid
pentanoic acid mFPhe meta-fluoro-phenylalanine
2,4h3mP 2,4-hydroxy-3-methyl- Mha 4-methyl-hydroxy-anthran-
pentanoic acid ilic acid
2,5h3mP 2,5-hydroxy-3-methyl- Mtz methyloxazole (formed by
pentanoic acid cyclization of threonyl side
3hAsn 3-hydroxy-asparagine chain)
3hCHA 3-hydroxy-cyclohexyl- N2MeAsp N,Zmethyl-aspartic acid
alanine Nle norleucine
3-h ydroxy-tetradecanoic Nva norvaline
acid oFPhe ortho-fluoro-pheny lalanine
3hLeu 3-hydroxy-leucine OMeSer O-methyl-serine
3S-Pro 3-thioproline oxz oxazole (formed by cycli-
3,4APro 3,4-dehydro-proline zation of seryl side chain)
4hPro 4-hydroxy-proline pGlu pyro-glutamic acid
4S-Pro 4-thioproline PhSer phenyl-serine
4,5hIle 4,5-hydroxy-isoleucine PPT phosphinothricine
6 'hTrp 6 '-hydroxy-tryptophan Q quinaldic acid
a-Aad a-aminoadipic acid Qaa quinaldinic acid
pAla palanine Qoa quinazol-4-one-3-acetic
AAla dehydro-alanine acid
AAbu dehydro-aminobutyric acid Qxa quinoxaline-2-carboxylic
Abu aminobutyric acid acid
Aeo 2-amino-S-oxo-9,lO-epoxi- Sar sarcosine
decanoic acid Ser* modified serine contained
aIle ah-isoleucine in tetrahydropyridine
allylGly allyl-glycine moiety of thiopeptides
AOC amino-octanoic acid Spro thioproline
I Introduction 279
T
- dehydrated threonyl side Main applications of peptides still are phar-
chain linked as thioether macological formulations applied in both hu-
tbuAla t-butyl-alanine man and animal health care. Besides antibac-
tbuGly t-butyl-glycine terial uses, increasing attention is paid to the
Thz thizole (formed by side introduction of antifungal compounds while
chain cyclization of immunomodulators have been firmly estab-
cysteine) lished. A considerable number of proteinase
vinylGly vinyl-glycine inhibitors and other enzyme inhibitors are
D-prefix for D-configuration, used without- continuously emerging and find various appli-
Me-prefix used for N-methyl- cations. In Sect. 5 a compilation of recently
Nme-prefix used for N-methyl- described peptides can be consulted for cur-
rent screening targets and their results.
As will be shown below, antibiotics evolved
in the ribosomal or enzymatic nonribosomal
system may have similar structures and prop-
1 Introduction erties. Similar genetic backgrounds for the
manufacturing of complex metabolites can be
followed from prokaryotes to lower and high-
Peptides as a chemically defined group of er eukaryotes, and boundaries between fields
metabolites have attracted considerable at- of microbial, plant or animal origin may dis-
tention for a variety of reasons. The first anti- appear. This becomes obvious in the screen-
biotics discovered - penicillins, tyrocidines, ing of marine organisms where in some in-
and gramicidins - are peptides, and new pep- stances metabolites isolated from sponges
tides emerge from screenings at a steady rate. could be traced to associated microorganisms.
However, in addition to the enzymatic path- Such possibly symbiotic relationships may in-
ways of biosynthesis which, e.g., produce po- deed trigger the production of useful metab-
lyketides or terpenoids, there is the additional olites. Likewise, experimental simulations of
route of ribosomal peptide synthesis. The microbial invasions lead to the production of
number of publications on peptide structures antimicrobials, and this principle has been ap-
reflects that increasing attention is paid to this plied successfully, e.g., in the detection of ri-
route. About a half of the 100 references on bosomally produced antibiotics of animal ori-
peptide antibiotics traced from December gin.
1995 until December 1996 deal with peptides The aim of these new approaches is to un-
of ribosomal origin. Those detected via non- derstand the ecological significance of such
antibiotic properties like hormone functions metabolites and to try to trace the concepts of
are not even part of this cluster. In recent the evolution of structures. These concepts
treatises on commercially important micro- will be beneficial in the design of selection
bial products peptides have an increasing screens in biosynthetic approaches utilizing
share (VININGand STUTTARD1995; STROHL, the natural combinatorial potential. In addi-
1996). In this volume, further chapters deal tion, boundaries between compound struc-
with highly significant fields: the established tures and their biosynthetic origins show the
plactams (SKATRUDet al., Chapter 6), the tendency of opening up due to the unifying
promising antibacterials called dalbaheptides concept of genetic structures. This is particu-
(LANCINIand CAVALLERI, Chapter 9), and larly evident from the similar genetic back-
the ribosomally formed lantibiotics (JACKet ground of various polyketide structures and
al., Chapter 8). The currently predominant their combination with peptide forming sys-
immunosuppressor cyclosporin is treated in tems, e.g., in the prominent immunomodula-
Chapter 12 by FLIRIet al. Other compounds tors FK506 and rapamycin (SCHWECKE et al.,
of peptidic origin are contained in the chapt- 1995).
ers on siderophores (WINKELMANN and
DRECHSEL,Chapter 5) and on antitumor
agents (GRAFEet al., Chapter 14).
2 Structures sponse proteins. Signaling functions have now

been reported in a variety of cases including
the involvement of peptide factors like the 45-
2.1 Ribosomal Origin residue ComS in competence in Bacillus sub-
tilis ( D ’ S o u z ~et al., 1994; HAMOENet al.,
Ribosomally encoded peptide antibiotics in 1995; MAGNUSONet al., 1994; SOLOMON et
microorganisms have been a topic throughout al., 1995).
the history of antibiotic research. The promis- In this chapter several aspects of mainly an-
ing field of lanthionine-containing peptides timicrobial peptides are presented. As the
for which the term “lantibiotics” has been field has been reviewed recently (LEHRERet
coined is treated in Chapter 8 of this volume al., 1993; BOMANet al., 1994; BOMAN,1995),
by JACK et al. These peptides are used in the we focus on biotechnological aspects.
food industry. As animal antibiotics of riboso-
mal origin they have led to very promising
fields of research. 2.1.1 Structural Classifications
Defenses against microbial infections - be-
sides the highly specific system of adaptive There have been several general ap-
immunity in vertebrates - were detected in a proaches to group or classify peptides of ri-
variety of targets. Thus compounds with an- bosomal origin. Two groups of linear peptides
timicrobial activity, mainly peptides, were with or without disulfide linkages is a com-
found in animals, plants, and bacteria. Ani- mon scheme, which is further refined into re-
mal antibiotics are located at surfaces like spective subgroups of dominating residues,
skin or mucosal surfaces, e.g., epithelial cells, antibacterial actions, or processed from pre-
and are components of body fluids like neu- cursors, or two, three, or four disulfides (Bo-
trophils, macrophages, and killer cells in MAN, 1995). Recently, HANCOCK et al. intro-
mammals and haemocytes in insects. These duced the group of cationic bactericidal pep-
antibacterials function as a primary defense tides (HANCOCK et al., 1995) which are de-
system which acts faster and simpler and fined as polypeptides with less than 100 ami-
saves energy as well as information compared no acid residues carrying a net charge of at
to the slow and highly specific memory-based least +2. It is obvious that there will be var-
system of adaptive immunity (BOMAN,1995). ious overlaps within such restricted systems as
Fundamental problems are the growth rates many compounds show several properties, in-
of bacteria compared to lymphocytes which cluding linear portions, disulfides and anti-
differ 50- to 100fold. bacterial properties. Their linear parts may be
Focusing on peptides, molecular genetics dominated by certain residues or they may in-
have opened up research to the complicated clude cationic regions. Schemes of this kind
network of innate immunity. Current work will be useful in some respects, but some com-
describes the structure of gene clusters, regul- pounds will be part of several subgroups. As
ation of the expression of peptide genes, their will be shown below, there is a similar situa-
posttranslational processing and modification, tion in the classification of nonribosomally
and their fate in the respective environment. generated peptides.
Obviously, these approaches are not limited
to antimicrobial peptides and apply as well to,
e.g., hormones or toxins. It has been pointed 2.1.2 Similar Structures of
out that peptides signaling defense reactions Ribosomal and Nonribosomal
in plants exhibit analogies to defense signal-
ing in animals (BERGEYet al., 1996). Thus Peptides - Monocyclic Structures
the release of the 18-residue peptide systemin
from a 200-amino acid precursor is induced Many cyclic and branched cyclic peptides
by mechanical wounding of tomato plants. of microbial origin containing nonprotein
Systemin induces a lipid-based signaling cas- constituents are well known antimicrobials.
cade leading to the expression of wound re- Cyclic peptides of ribosomal origin were char-
2 Structures 281
acterized from bovine neutrophils (bactene- three identical 198-base pair open reading
cins (I); ROMEOet al., 1988), and from the frames which encode identical 66-amino acid
skin of bullfrogs (brevinin 1 (11), brevinin 2 peptides, but differ in their presumably regul-
(111), Rana brevipode; esculentin (IV), Rana atory untranslated 5 '-regions. The amino ter-
esculenta; MORIKAWA et al., 1992; SIMMACO minal prepropeptide region is composed of a
et al., 1993; ranalexin (V), Rana catesbeiana; signal sequence separated by a prohormone
CLARKet al., 1994). processing site from the acidic region preced-
R4L-C +R+I+V+V
I \ 1
R+C+C+R+I
FLPVLAGIAAKVVPALF+ C+ K+ I +T
I1 \ 1
C+K+K
GIMDTLKNLAKTAGKGALQSLLNKAS +C A +V+T +
I11 \ 1
C+K+K
GIFSKLGRKKIKNLLISGLKNVGKEVGMDVVRTGIDIAG C + K-' I K + +
IV \ 1
C + E +G
FLGGLIKIVPAMI + C+ A +V+T
V \ 1
C+K+K
The cycloheptapeptide moiety appears to be ing the antibiotic peptide. Similar sequences
a common structural element of the Rana are known from other propeptides (BEVINS
peptides, carrying positively charged side and ZASLOFF,1990). These signal sequences
chains as found in polymyxin (VI), a nonri- resemble sequences found in amphibian
bosomal antibacterial from Bacillus poly- opioid peptides including dermorphins and
myxa. Cyclononapeptide structures as in bac- deltorphins (VIII and IX). These opioid re-
tenecins are found, e.g., in nonapeptidolac- ceptor-binding heptapeptides were first iso-
tones like lysobactidkatanosin (VII, Lyso- lated from the skin of South American frogs
bacter sp. and Cytophaga sp.). (Phyllomedusa sp.) and have been shown to
acyl+ Dbu +Thr +DDbu+ D bu+ Dbu DLeu+Thr+
VI
I 1
ThrcDbutDbu
DLeu +Leu+PhSer+3hLeu +Leu+ Arg+ Val
VII I 1
Sert3hAsn +Gly+aThr
Variable side chains point to additional func- contain D-amino acids (ERSPAMERet al.,
tional features, possibly even linked to the 1989).
processing of precursor structures. CLARKet
al. (1994) isolated the gene for ranalexin us- VIII Tyr-DAla-Phe-Gly-Tyr-Pro-SerNH2
. .
ing ' the 'cDNA approach and detected an

acidic amino acid-rich propeptide sequence at IX Tyr-DAla-Phe-AsP/G1u-val-val-
the amino terminal end. The ranalexin struc- GlyNH2
tutral gene has been found to be contained in
X pGlu-Asp-Tyr( HS03)-Thr/Leu/Met- A different route of conversion is found in

Gly-Trp-Met-Asp-PheNH2 lantibiotics where Ser or Thr residues are ini-
XI pGlu-Leu-Tyr-Glu-Asn-Lys-Pro- tially reduced to the dehydro derivatives
Arg-Arg-Pro-Tyr-Ile-Leu (SKAUGENet al., 1994). These may further
react with Cys side chains to thioethers, but
Both types of bioactive peptides are produced isolated dehydro residues are known, e.g., in
in granular glands of the skin, compounds all lantibiotics (JACK et al., Chapter 8, this
like dermorphin, caeruleins (X), and neuro- volume). These reactive side chains may be
tensin (XI) being noxious to predators. The involved in covalent binding of peptides to
common processing/transport signals indicate target proteins, as has been shown for the
their common secretory pathway. cyanobacterial cyclopeptide microcystin and
protein phosphatase 1 (MACKINTOSH et al.,
2.1.3 D-Amino Acids in 1995). A desdehydro derivative of nisin A
shows reduced antibacterial activity (ROLLE-
Ribosomally Produced Peptides MA et al., 1996).
Epimerizations of amino acid residues in

gene-encoded peptides have been verified for 2.1.4 Posttranslational
dermorphin (VIII) and dermenkephalin
(XII). Dermorphin binds exclusively to the p- Modifications of Side Chains
opioid receptor and is 1000 times more potent
Non-disulfide cyclic peptides include lanti-
than morphine upon intracerebroventricular
biotics (JACKet al., Chapter 8, this volume;
injection. The D-residue is of crucial impor-
JACKet al., 1995) with stable thioether links
tance for this high affinity. Isolation of the re-
forming multiple 3- to 6-membered cyclic
spective gene revealed the Ala codon as pre-
peptide structures. These unique antibacter-
cursor of D-Ala (RICHTERet al., 1987). The
ials, which are approved as food components,
predicted biosynthetic precursor contains 4
have been found exclusively in bacteria so far.
dermorphin sequences together with one for
Examples given are epidermin (XVI) and
dermenkephalin flanked by putative cleavage
subtilin (XVII). Several respective gene clus-
sites. In addition, D-amino acids have been
ters have been characterized and, besides the
found in the neuroexcitatory snail peptides
structural genes of the propeptides, they con-
from Achatia fulica achatin I (XIII) and fuli-
cin (XIV; FUJITA tain modifying enzymes and proteins involved
et al., 1995), and the 48 ami-
in export. Structural modifications to direct
no acid-containing Ca2+ antagonist W-
specificity and improved stability are under
AgaIVC from the venom of the Agelopsis
investigation (for details, see Chapter 8). In
aperra spider (XV). This venom has been the
contrast to disulfide bonds, enzymes opening
source of a cofactor-independent isomerase
thioether bonds have not yet been identi-
acting on peptides with the common sequence
fied.
Leu-Xua-Phe-Ala. This enzyme has been
Side chain modifications of cysteine and ser-
shown to isomerize Ser, Cys, OMeSer, and
ine residues to oxazoles and thiazoles, estab-
Ala residues in a reversible reaction (HECKet
lished for nonribosomal products for a long
al., 1995). At least in the case of dermorphins
time, have recently been detected in riboso-
the conversion to D-Ala was shown to pro-
mal peptides as well, including microcins
ceed at the prepropeptide stage (MOR et al.,
1991). (MORENOet al., 1995) and the rhizobial pep-
tide trifolitoxin (BREILet al., 1996; TRIPLETT,
XI1 Tyr-DMet-Phe-His-Leu-Met-AspNH2personal communication). A respective trans-
XI11 Gly-DPhe-Ala-Asp forming enzyme complex has been identified
by the analysis of the microcin B17 (XVIII)
VIV Phe-DAsn-Glu-Phe-ValNH2
XV EDN CIAEDIGK CTWGGTK CC RGRP CR CSMIGTN C E CTPRLIMEGL-D-Ser-SFA
I I I I
I
1 -
(XVI)
--
IASKFICTPGCAKTGSFNSYCC
(XVII) WKSETLCTPGCVTGALQTCFLQTLTCNCKISK
-
2 Structures 283
-r- V T
gene cluster and it contains three proteins (LI bosomal origin ..as now been verified by the
et al., 1996). Microcin B17 is the first known detection of the structural gene during the
gyrase inhibitor of peptidic nature. Its 69-resi- Bacillus subtilis genome sequencing pro-
due precursor peptide is modified at Gly-Cys gram.
and Gly-Ser segments, respectively, or at Gly-
Ser-Cys or Gly-Cys-Ser segments, with the (XXI) 4,5hIle-6’hTrp-Gly
formation of bicyclic thiazol-oxazole deriva-
tives. Replacement of oxazoles by thiazoles I I 1
4hPro SO Ile
leads to an inactive product (BAYERet al., I I I
1995). All respective examples reported so far Asn-Cy s-Gly
seem again to be restricted to bacterial
sources, including thiopeptides like thiostrep- (XXII) Qxa
ton (STROHLand FLOSS,1995), the antitumor Qaa
peptide bleomycin, bacitracin, and various Qoa-DSer-Ala-Cys-NMeVal
marine peptides like ulithiacylamide (XIX).
These cyclic peptides isolated from sponge I I
extracts are likely to originate from asso- NMeVal-Cys-Ala-D Aer -Qoa
ciated microorganisms, although this has not Qaa
been generally established (BEWLEYet al., Qxa
1996).
L Qxa
XVIII VGIGGGGGGGGGG-Oxz-Thz- Qaa
GGQGGG-Thz-GG-Thz-SNG-Thz- Qoa-DSer -Ala-Cys- NMeVal
Oxz-GGNGG-Thz-GG-Thz-GSHI
XIX Leu-Thz-Cys-Mtz I I
I I
I NMeVal-Cys-Ala-dSer-Qoa
YS
Mtz- Cys-Thz-Leu
Qaa
Qxa
(XX)
[Q]-Ile-Ala-AAla-Ala- [Ser*]-Thz-Thr-AAbu-DCys-3,4hIle-Thz-Thr-Thz-[Ser*]
-Thz-AAla-
I I I I AAla(NH2)
2.1.5 Multicyclic Peptides The principle of multiple rings in gene-en-

coded peptides is well established for numer-
ous structures containing up to 4 disulfide
Prominent examples of multicyclic peptides links (see compilation in Sect. 5 ) . Common
from microorganisms include thiostrepton cyclization patterns emerge and point to a
(XX), amanitins (XXI), and quinomycin similar conserved enzymology. Properties of
(XXII). For the latter in vivo transformation the peptides identified so far are mainly an-
of triostins (XXIII) to quinomycins was de- timicrobial, cytotoxic, or like trifoil peptides
monstrated (WILLIAMSON et al., 1982). More involved in tissue repair and cell proliferation
recently, the structural elucidation of the Ba- (CHINERY and COFFEY,1996). The rat intes-
cillus subtilis antibacterial subtilosin (XXIV) tinal trifoil factor rITF has even been shown
introduced a new level of complexity. Its ri- to form covalent dimers (CHINERYand COF-
(XXIV) GLGLWGNKGCATCSIGAACLVDGPIPDG*IAGAX
LS-S] IXI
FEY, 1996). Recent excitement has been pro- er, regioselective formation of several disul-
voked by the discovery of microbial multicyc- fide bonds still remains a challenge for chem-
lics with promising antiviral properties like ists and overall yields are in the range of a
aborycin (RP71955, XXV) (POTTERATet al., few percent (KELLENBERGER et al., 1995).
1994; FRECHET et al., 1994). Microbial peptides produced in their hosts do
not present a special problem. The produc-
- -I
(XXV) tion of animal or plant antibacterial peptides,
r, CLGIGSCNDFAGCGXAVVCFW
in bacterial species, however, poses the prob-
lem of a suicide situation. This could be over-
come by simultaneously introducing resist-
ance or export genes. So far, either the ex-
s-s pression of fusion constructs has been prac-
s-s ticed (PIERSet al., 1993) or insensitive yeast
cells have been employed (REICHHART et al.,
1992).
2.1.6 Linear Peptides
A number of peptides without any side
chain modifications or disulfide bonds with 2.1.8 Similar Open Questions in
diverse biological activities are known. For
antimicrobial peptides at least several groups
Gene-Encoded and Enzymatically
have been described of (1) amphipathic a-he- Formed Peptides
lical peptides (e.g., magainin) and (2) two a-
helices linked by a hinge region (e.g., cecro- The findings on multicyclic peptides illus-
pins). Three other groups show high contents trate well that the formation of constrained
of (1) glycine (e.g., diptericin), (2) proline and disulfide linkages to stabilize and functional-
arginine (e.g., apedaecin), or (3) tryptophan ize peptide chains is not restricted to higher
(e.g., indolicidin) (LEHRERet al., 1993; Bo- eukaryotes. The production of various multi-
MAN et al., 1994; BOMAN,1995). The trypto- cyclic peptides may thus be well achieved by
phan-rich indolicidin (XXVI) shows antibac- microbial cultures. However, a number of
terial and hemolytic activity. Replacement of questions might be relevant, brought up re-
tryptophan by phenylalanine abolishes only cently in connection with defensins (SELSTED
hemolytic activity (SUBBALAKSHMI et al., and OULE-I'TE, 1995):
1996) while antibacterial activities remain,
even when proline residues are substituted by How can the spontaneous lysis of granules
alanine. producing cytolytic peptides be prevented?
The same question is unresolved for the
(XXVI) ILPWKWPWWPWRRNHZ many microbial compounds with lytic prop-
erties; there might be natural antagonists or
A compilation of structures can be found in storage molecules.
Sect. 5. Are there extracellular roles for neutrophil
defensins, or are there intracellular roles
for enteric defensins? This question applies
2.1.7 Production of Ribosomally to most metabolites with, e.g., antimicrobial
Formed Peptides properties and it has not yet been answered
for any secondary metabolite convincingly.
Besides solid-phase synthesis production Do defensin isoforms cooperate to form
by fermentation is in an early phase. Howev- heteromer assemblies in target cell mem-
branes? The question of assembly is of con-
siderable importance since dimerization or
3 Biotechnology of
multimerization properties cannot be pre-
dicted yet. The human defensin HNP-3
Peptides
forms dimers, and dimer formation is es-
sential for the function of many microbial In the biotechnology of peptides fermenta-
peptides like the linear gramicidin pore for- tion of pure microbial cultures is used tradi-
mation or vancomycin binding to peptido- tionally. An increasing number of processes
glycan precursors. deals with cell culture techniques that are still
dominated by larger peptides and proteins as
products. Obviously, the manufacturing of
2.2 Nonribosomal Origin most bioactive peptides, especially antibiotics,
needs self-protective systems. Such systems of
In contrast to those of ribosomal origin, self-immunity are inherent in producer cells
more unusual structures and compounds are but may limit the levels of production. Prob-
found in nonribosomally formed peptides. lems arise when non-self products are to be
Systematic approaches to peptide structures generated, e.g., the expression of animal-de-
are difficult, since in reports which are basi- rived antibacterials in bacterial cultures. Solu-
cally oriented towards structural chemistry tions of such problems might be the use of in
the compounds may be found as peptide alka- vitro production systems or the introduction
loids (LEWIS,1996), as indole alkaloids (IHA- of resistance genes into the expression organ-
RA and FUKUMOTO, 1996), or they may be ism.
contained in source-oriented reviews (e.g.,
marine metabolites; FAULKNER,1996, and
references therein). Peptide structures linked 3.1 Biosynthesis
with polyketides and terpenoids may raise ad-
ditional problems of compound classification. The essential questions to be addressed re-
In a biosynthetic approach simply the types of fer to the biosynthetic origin of the com-
enzyme structures involved are used. Since all pound and to the availability of precursors. It
peptides originate from linear precursors fol- has to be decided whether a structural gene
lowed by various processing reactions, several or a protein template is involved. This deci-
approaches to systematics are possible (VON sion can be made upon inspection of the
DOHREN, 1990). However, in many cases structural features. A collection of modified
there are several structural features involved amino acids found in peptides of ribosomal
and convenient approaches should then per- origin is given in Tab. 1. A respective collec-
mit the search for a structural profile. tion of amino acids encountered in nonribos-
Tab. 1. Unusual Features of Ribosomally Derived Peptides
Modified Amino Acid Modification Process
Glu pyroGlu terminal modification

Ala D-Ala in-chain epimerization
Met D-Met
ASP D-As~
Ser D-Ala enzymatic dehydration
Thr D-Abu
CyslSer lanthionine enzymatic dehydration and
Cys/Thr methyl-lanthionine thioether linkage
SeriThr oxazole-derivatives enzymatic cyclization
CYS thiazole-derivatives
SYSTEM PRODUCT
Nucleic Acid Dependent Ribosomal System

1 2 3
- linear polypeptidesof any
sequence and Imgihfmm
TYPE 1 -
20 proteinogenicamino acids
activationas
choica of Program
Asssmbiy Line sminoacyl adenyla1.s
-
aminoacyl-tRNA-.ynUIetases
high specificitjthmugh
- proof reading mechanisms
-
high energy requiring
processingof peptides by
poomanslational processes
Complex Mechaniam
Up to 180 Compomntr Participating
Nucleic Acid Independent
&a -
Enzymatic Systems
%bP&
TYPE 2 PolyenzyrneSystem
Fixed Program - iinoar, branchedand cyclic
Assembly Line pptidea and depsipeptides.
-
e.g. Gramicidin S
Multionryme activetimas
All Intermediates Remain Enzyme-Bound

- amlmcyl adenyiates
proteinand nonprotein amino acids
Sequence Is Determined on the MuitlfunctionaiEnzyme-Chain (Dconfiguratlon. omithine.
-
4’-Phosphopantetheine is Cofactor hydroxy acids)
MultlenzymesIdentified so far only in Bacteria and Fungi
-
lowspecificity
--
Incorporationof anal-
modifkatbn, 9.9. Nmethylation
length of peptides up to
about 20 amino acids
Single Step Enzymes

- linear peptides, 9.9.
- OlUblhlolr, bacterialcoil wall

activationas phosphates or
- aminoacyl adenylates
synthesis of short peptides
separate enzymes - (generally less than 5 amino acids)
also non-protein amino acids are
components
Intarmediatesare soluble
only one component is activated
Fig. 1.Mechanisms of peptide biosynthesis. Scheme comparing ribosomal (type 1)and nonribosomal systems (type
2 - multistep systems, type 3 - single-step systems).
omally formed peptides is presented in Sect. tion (*Flu-Cys),Gly, with n=2-11 (RAUSER,
5. The principal differences of peptide form- 1990 STEFFENS,1990). Besides Gly also p
ing systems are shown in Fig. 1. Nonriboso- Ala or Ser may function as C-terminal resi-
ma1 peptides are formed either by single-step dues (GEKELERet al., 1989; KLAPHECKet al.,
enzymes with free intermediates or by multis- 1994). In maize seedlings two additional pep-
tep systems. Well-known examples of peptide families of the type (fllu-cys),, and
tides made by single-step enzymes are gluta- (+lu-Cys),Glu have been observed upon ex-
thione, pantetheine, or muropeptides. Thiol posure to Cd2+ (MEUWLYet al., 1995). Bio-
peptides termed “phytochelatins” which upon synthetic systems for these peptides are sup-
exposure to Cd2+, Cu2+, or Zn2+ are in- posed to resemble glutathione forming en-
duced in the fission yeast Schizosaccharo- zymes.
myces pombe (KONDOet al., 1984) or in cul- The modification of peptide structures by
tured plant cells (GRILLet al., 1989) are com- alteration of their genetic background is an
posed of Glu, Cys, and Gly of the composi- emerging approach (see Sect. 3.2.2). In gener-
al, the structural genes for peptides or peptide tems, or reductions and dehydration in the
synthetases are clustered together with those polyketide systems. A module is thus com-
for modifying enzymes, regulatory signals, prised of submodules. Boundaries of these
and - if secreted - for export proteins. submodules have been determined at the pro-
tein level by limited proteolysis studies and
sequence alignments. Therefore, the respec-
3.1.1 Biosynthetic Gene Clusters tive DNA modules have corresponding func-
tional protein modules, which contain do-
Clusters identified for peptide biosynthesis mains and subdomains of defined structures
are listed in Tab. 2. So far, only microbial and functions. All modules can be readily
clusters have been characterized in detail. Ge- identified by the presence of highly conserved
nome sequencing has revealed that there are core sequences (PFEIFERet al., 1995; STEIN
no nonribosomal peptide biosynthetic activi- and VATER,1996) (Fig. 2); a selection is giv-
ties in yeast, Haemophilus influenzae, Myco- en in Tab. 3.
plasma genitalium, Synechocystis, and Metha-
nococcus jannaschii. However, evidence for
multiple peptide forming clusters has been Arrangement of modules
obtained in Bacillus subtilis, Microcystis aeru-
ginosa, Amycolatopsis mediterranei, and Ba- Modules and their corresponding function-
cillus brevis. al domains show a linear arrangement corre-
Gene structures generally permit the iden- sponding to the sequence of catalytic events.
tification of the involved biosynthetic reac- As these sequential reactions are catalyzed by
tions. Structural genes for ribosomally en- protein surfaces this strict linearity is not ob-
coded peptides are a direct proof of their ori- vious. Domains which are not directly linked
gin (e.g., subtilosin). A respective nonriboso- need to interact frequently, e.g., in condensa-
ma1 system is readily identified by the pres- tion reactions (see below). Domains may be
ence of very large genes (up to 45 kb) with reused for repeated sequences as in the for-
modular structures. The most prominent mation of the cyclodepsipeptide enniatin
structural element is the amino acid activating (XXVII). The respective synthetase contains
module or adenylate domain. Domains like only two, not six activation modules.
this can be identified by a number of highly
conserved motifs (KLEINKAUFand VON NMeVal-Hiv-NMEVal
DBHREN, 1996). The signature sequence I I
SGTTGxPKG is most prominent and is also (XXVII) Hiv-NMeVal-Hiv
contained in acyl-CoA synthetases. Peptide
forming systems are recognized by the re-
peated occurrence of these modules directly Adenylate forming domains
related to the number of amino acids con-
tained in the respective peptide. In contrast to polyketide forming systems
peptide synthetases activate their building
blocks, i.e., amino acids, imino acids, and hy-
3.1.2 Biosynthetic Modules droxy acids, as adenylates on an integrated
activation domain. These activation domains
The term “module” has been introduced show structural similarities to the carboxyl-ac-
for a gene-encoded functional unit perform- tivating acyl-CoA synthetases of the polyke-
ing an elongation step in either polyketide or tide systems. The positioning of activating do-
peptide biosynthesis (DONADIOet al., 1991; mains corresponds in sequence to the amino
KLEINKAUF and VON DBHREN,1996). Dur- acid sequence of the product in either wholly
ing the process the particular residue entering or partially integrated interacting enzyme sys-
the elongation cycle may be structurally al- tems (Fig. 3). The reaction sequence includes
tered, which implies epimerization, methyla- binding of carboxyl substrate and MATPZ-
tion, or hydroxylation in peptide forming sys- (with M=Mg2+, Mn2+) and formation of the
Tab. 2. Peptide Biosynthetic Clusters Identified
Compound Type" Organism Reference
Ribosomally Formed Peptides:

Cytolysins P-?-M Enterococcus faecalis SAHLet al., 1995
Epidermin P-21-M Staphylococcus epidermidis SCHNELL et al., 1992
Lactocin P-37-M Lactobacillus sake SKAUGEN et al., 1994
Lactococcin P-27-M Lactococcus lactis RINCEet al., 1994
Microcin B17 P-43-M Escherichia coli LI et al., 1996
Nisin P-34-M Lactococcus lactis JACKet al., 1995
Pep-5 P-34-M Staphylococcus epidermidis K A L E ~etA al., 1989
Subtilin P-32-M Bacillus subtilis HANSEN,1993
Trifolitoxin P-11-M Rhizobium leguminosarum BREILet al.. 1996
Nonribosomally Formed Peptides:

A54145 R-P-13-L-10 Streptomyces fradiae BALTZ,1996
Actinomycin R-(L-5)2 Streptomyces chrysomallus SCHAUWECKER et al.,
1996
Anguibactin R-P-2-M Vibrio anguillarum CHENet al., 1996
Ardacin P-7-M Kibdelosporangium aridum PIECQet al., 1994
Bacilysin P-2-M Bacillius subtilis SAKAJOH et al., 1987;
HILTONet al., 1988
Bacitracin R-P-12-C-7 Bacillus licheniformis HERZOG-VELIKONJA
et al., 1994
Bialaphos P-3 Streptomyces viridochromogenes SCHWARTZ et al., 1996
CDA R-P-11-L-10 Streptomyces coelicolor CHONGet al., 1996
Clavulanic acid modified amino acid Streptomyces clavuligerus HODGSONet al., 1995
Cephalosporin P-3-M Acremonium chrysogenum MART~N and GUTIER-
REZ,1995
Lysobacter lactamgenus KIMURA et al., 1996
Cephamycin P-3-M Nocardia lactamdurans COQUEet al., 1995a,b
Coronatin acyl amino acid Pseudomonas syringae BENDERet al., 1996
Cyclosporin c-11 Tolypocladium niveum WEBERet al., 1994
Daptomycin R-P-13-L-10 Streptomyces roseosporus BALTZ,1996
Destruxin L-6 Metarhizium anisopliae BAILEYet al., 1996
Enniatin D-6 Fusarium scirpi HAESEet al., 1993
Enterochelin P-C-E-3 Escherichia coli REICHERTet al., 1992
Bacillus subtilis
Fengymycin R-P-lo-? Bacillus subtilis LIU et al., 1996
Ferrichrome C-6 Ustilago maydis MEI et al., 1993
Gramicidin S C-(P-5)* Bacillus brevis TURGAYand MARA-
HIEL, 1995
HC-toxin c-4 Helminthosporium carbonum PITKIN et al., 1996
Immunomycin modified polyketide Streptomyces sp. MOTAMEDI, 1996
Iturin C-8 Bacillus subtilis HUANGet al., 1993
Lysobactin P-11-L-9 Lysobacter sp. BERNHARD et al., 1996
Microcystin c-7 Microcystis aeruginosa MEISSNER et al., 1996
Nikkomycins modified peptides Streptomyces tendae BORMANN et al., 1996
Nosiheptide R-P-13-C-10-M Streptomyces actuosus STROHLand FLOSS,1995
Penicillin P-3-M Aspergillus nidulans SMITHet al., 1990; MAC-
Penicillium chrysogenum CABEet al., 1990
DIEZ et al., 1990
Tab. 2. Continued
Compound Type" Organism Reference

Phaseolotoxin P-4-M Pseudomonas syringae TURGAYand
MARAHIEL, 1995
Pristinamycin A R-C-6 Streptomyces pristinaespiralis BLANCet al., 1997
Pristinamycin M polyketide/peptide
Pyoverdin R-P-8-M Pseudomonas fluorescens STINTZIet al., 1996
Rapamycin modified polyketide Streptomyces hygroscopicus SCHMECKE et al., 1995;
MOLNARet al., 1996;
APARICIOet al., 1996
Saframycin P-4-M Myxococcus xanthus POSPIECHet al., 1996
SDZ214-103 L-11 Cylindrotrichum oligospermum BERNHARD et al., 1996
Surfactin L-8 Bacillus subtilis COSMINA et al., 1993
Syringomycin R-L-9 Pseudomonas syringae Mo et al., 1995
Syringostatin
Thiostrepton R-P-17-C-10-M Streptomyces laurentii STROHLand FLOSS,1995
Tolaasin R-P-18-L-5 Pseudomonas tolaasi RAINEYet al., 1993
Tyrocidine c-10 Bacillus brevis MITTENHUBER et al.,
1989
Yersiniabactin R-P-3-M Yersinia enterocolitica GUILVOUT et al., 1993
a The abbreviations used are: P peptide, C cyclopeptide, L lactone, D depsipeptide, E ester, R acyl, M modified. The
structuraltypes are defined by the number of amino-, imino- or hydroxy acids in the precursor chain. The ring sizes
of cyclic structures are indicated by the number following C, L, D, or E, defining the type of ring clo-
sure.
Fig. 2. Domain construction of peptide synthetases.Peptide synthetase sequencescan be identified by a spacing of

motifs, which have been assigned to adenylate formation (A1 and A2), aminoacylation of pantetheine (S),
N-methylation of aminoacylresidues (M), condensation (peptide bond formation) (C), epimerization (E), and
thioesterase (TE).The respective consensus sequences are given in Tab. 3.
acyladenylate which, as a highly reactive sitive assay for peptide synthetases and their
mixed anhydride, is stabilized against hydro- activation domains. In the presence of high
lysis by the two subdomains. The released concentrations of ATP the adenylate may
MPF- may reverse the activation reaction react to form AP4A, but it is not known if this
leading to the two substrates. This reaction reaction has any physiological significance.
employing [P-321-PPi usually is used as a sen-
Tab. 3. Consensus Sequences of Peptide Synthetase Motifs
Adenylate Domains
A: LTxxELxxxAxxLxR
B: AVxxAxAxWxIDxxYPxER
C. YSTGTTGxPKG
D: IIxxYGxT
E: GELxIxGxxVAR
F: RLYRTGDL
G: IEYLGRxDxQVKIRxxRIELGEIE
H: LxxYMVP
I: LTxxGKLxRKAL
Acyl Carrier Domains

J: LGGxSIxAI
Elongation (1) Epimerization" (2) Domains
K (Bl): YPSVxxQxRMYIL, (B2): LxPIQxWF
L (Bl): LIxRHExL, (B2): LxxxHD
M (Bl): DMHHIIxDGxSxxI (B2) HHxxVDxVSWxIL
N (Bl): LSKxGQxDIIxGTPxAGR (B2): VxxEGHGRE
0 (Bl): IxGMFVNTxLALR, (B2): TVGWFTxxxPxxL
P: PxxGxGy
Q:
N-Methyl-Transferasesb
M1: GxDFxxWTSMYDG;
M 2 LEIGTGTGMVLFNLxxxxGL;
M3: VxNSVAQYFP
M4: ExxEDExLxxPAW,
M5: HVExxPKxMxxxNELSxYRYxAV
N6: GxxVExSxARQxGxLD.
a Derived from Bacillus synthetases only.

Derived from fungal domains.
Aminoacylation or carrier domains carrier protein synthase structure from Esch-

erichia coli (LAMBALOTet al., 1996). Re-
Adenylates formed on acyl-CoA synthet- spective genes have been detected in several
ases are transferred to CoA, and these thioes- biosynthetic gene clusters, and for each clus-
ters are the stable transport form of aliphatic ter a specific transferase to activate the path-
carboxylic acids. Amino acids are transferred way has been proposed. The finding that apo-
to acyl carrier modules integrated within the enzymes form unusually tight complexes with
multienzyme structures. These carrier mod- the transferases in the absence of CoA sug-
ules have a structure similar to acyl carrier gests an additional control mechanism (LAM-
proteins involved in various polyketide form- BALOT and WALSH,1995; PFEIFER et al., un-
ing systems including fatty acid biosynthesis published data).
(STEINand VATER, 1996; STACHELHAUS et As in the case of the acyl transfer to CoA
al., 1996a). Their essential functional group, the adenylate is cleaved by the phosphopan-
4'-phosphopantetheine, has to be introduced tetheine thiol to release AMP and an enzyme-
by posttranslational modification from CoA. bound intermediate thioester is produced
The respective transferases have recently (PFEIFERet al., 1995).
been identified with the aid of the holo-acyl
Fig. 3. Domains in peptide forming multienzyme systems. ACVS: ACV synthetase, ES: enniatin synthet-
ase, Ent: enterobactin biosynthetic enzymes EntE and EntF, GS: gramicidin S synthetases 1 and 2, T Y
tyrocidine synthetases 1 and 2, Srf: surfactin synthetases 1, 2 and 3, CY: cyclosporin synthetase, H C HC-
toxin synthetase. Amino or hydroxy acid specificity of the domains is indicated. Note that two different
types of condensation domains are found, slightly deviating in their consensus sequences, and one very
closely related epimerization domain. The E-C domain in the HC-system catalyzes both epimerization and
condensation.
Condensation and epimerization domains (KLEINKAUFand VON DOHREN,1997)

(Fig. 4).
Adjacent to the carrier domains condensa- In several systems D-amino acid residues
tion or epimerization domains with similar are formed during a transfer reaction be-
structures are found. These are structurally tween two multienzymes. In these cases an
stable regions of about 50 kDa (DIECKMANN epimerization domain terminates the first en-
et al., unpublished results). Epimerization do- zyme while the second starts with a condensa-
mains are readily identified by two additional tion domain (DE CRECY-LAGARDet al.,
motifs (see Tab. 3). Epimerizations are cata- 1995). If the transfer occurs within an enzyme
lyzed at the aminoacyl or peptidyl stage re- system and is thus an intraenzymatic transfer,
sulting in a thioester-bound mixture of both a third type of domain with a size of about
isomers (KLEINKAUFand VON DOHREN, 100 kDa is found resembling a fusion of both
1997). In the following step the selection of types of domains. Such a domain was de-
the stereospecific intermediate is controlled. tected first in HC toxin synthetase, a multien-
The current model of peptide bond formation zyme forming the plant pathogenic cyclotetra-
assumes that aminoacyl and peptidyl interme- peptide HC toxin (SCOTT-CRAIG et al.,
diates of two adjacent modules are directed 1992).
towards peptide bond formation at the en-
closed condensation domain. In order to ex- N-methylation domains
plain the direction of bond formation amino-
acyl and peptidyl binding sites have been pos- Methylation of amino groups is catalyzed
tulated in analogy to the ribosomal system by a transferase domain at the aminoacyl-
4-Phosphopantetheine
/
ondensation domain 1
'2
Activation domain
domain
I Thiogsterasedomain
Exit site
Fig. 4. Schematic view of peptide biosynthesis using a tripeptide model system, ACV synthetase. The three
amino acids are activated at the activation domains 1-3. Each activation domain is composed of two sub-
domains. Adjacent to each activation domain is a carrier domain, which contains the cofactor 4 '-phospho-
pantetheine. This factor may swing into three positions: binding the activated amino acid (activation do-
main), transport of the activated amino acid to the condensation site (A-site), and transport of the starter
unit or peptidyl intermediate to the condensation P-site. In the terminating reaction epimerization occurs
and the peptide is stereospecifically released at the thioesterase domain.
thioester stage (BILLICHand ZOCHER,1987, change of a highly conserved serine residue

1990). This domain is inserted within the ade- contained in the motif GXSXG in the thioes-
nylate domain and has a size of about 50 kDa terase domain, G(L-a-aminoadipy1)-L-cystei-
(HAESEet al., 1993; WEBERet al., 1994). The nyl-L-valine is released as the main product
highly conserved core motifs (see Tab. 3) are instead of exclusively S(L-a-aminoadipy1)-L-
not found in other known methyl transferases cysteinyl-D-valine (KALLOW et al., 1996).
which all share the substrate S-adenosyl-me- This result indicates that the thioesterase ca-
thionine (BURMESTER et al., 1995). The reac- talyzes the stereospecific release of the prod-
tion has been investigated in some detail and uct. Similarily condensation domains adjacent
methods to detect methylation domains at the to epimerization domains are thought to cata-
protein or gene level have been developed lyze the stereospecific condensation of only
(ZOCHERet al., unpublished data). one of the peptidyl or aminoacyl interme-
diates.
Thioesterase domains
Enzymes still to be identified
Domains with similarities to thioesterases
have been detected in peptide synthetases A number of reactions known to be in-
and polyketide synthases as well. Their posi- volved in enzymatic peptide biosynthesis
tioning in the C-terminal region of terminat- have not been studied yet. It could not be es-
ing multienzymes indicates their functions in tablished so far, if these proceed within the
terminating a catalytic cycle. Such functions context of integrated multienzymes or if sepa-
have been demonstrated in several polyketide rate single-step enzymes interact with the re-
forming enzyme systems. Experimental work spective systems. Such reactions include the
on peptide synthetases is only available so far modification of amino acids (like hydroxyla-
on ACV synthetase producing the tripeptide tion or O-methylation), side chain modifica-
precursor of plactam antibiotics (for details, tions (like oxazole and thiazole ring forma-
see Chapter 6, this volume). Since a linear tion of serine, threonine and cysteine side
peptide is released the thioesterase function chains; cf. Sect. 2.1.4), and ether linkages or
seems to be obvious in this case. Upon direct linkages of aromatic side chains.
3.1.3 Biosynthetic Enzyme of Bacillus subtilis genome sequencing has re-

vealed a peptide biosynthetic cluster with a
Systems yet unidentified product (TOGNONIet al.,
1995). The conditions of expression of this
Multienzyme systems for most types of cluster are unknown. Tentatively, from corre-
peptides have been characterized (KLEIN- lating activation modules with known struc-
KAUF and VON DOHREN,1996) (Tab. 4). It tures, as has been discussed by COSMINAet
seems evident that systems of eukaryotic ori- al. (1993), the already identified peptide fen-
gin are fully integrated while prokaryotic sys- gycin could be the product.
tems contain up to three interacting multien-
zymes. The only system deviating from this
rule is ACV synthetase forming the p-lactam 3.2 Production
precursor tripeptide. This enzyme is recov-
ered in a fully integrated state from both pro- The best studied system for peptide pro-
karyotic and eukaryotic sources. Isopenicillin duction is the penicillin fermentation where
N synthase forming the bicyclic penam pre- relevant parameters have been investigated in
cursor from the tripeptide is located adjacent considerable detail. These include precursor
to the synthetase in the respective gene clus- concentrations and transport, intracellular
ters and has been implicated in horizontal compartmentation, levels and activity of bio-
gene transfer (AHARONOWITZ et al., 1992; synthetic enzymes, their regulation, expres-
BUADESand MOYA,1996). Transfer, howev- sion, and stability in relation to the gene copy
er, is thought to proceed from prokaryotic to number, their posttranslational modification
eukaryotic hosts. and intracellular localization, their possible
The integration of all sequential condensa- inhibition by metabolites or products, and fi-
tion functions does not rule out various internally their export. This still uncomplete list of
actions with other enzymes involved. In cy- factors illustrates the complexity of tasks to
closporin formation D-alanine and the com- be addressed. Many of the details are dis-
plex amino acid (4R)-4-[(E)-2-butenyl]-4-me-cussed in Chapter 6. For an overview of the
thyl-L-threonine (Bmt) have to be supplied as state of metabolic flow studies in Penicillium
direct precursors. Thus a respective alanine chrysogenum the reader is referred to the re-
racemase and an enzyme system forming a cent publications of NIELSEN et al. (JoR-
polyketide precursor which has to be trans- GENSEN et al., 1995a, b; NIELSENand JOR-
aminated are involved in the process as well GENSEN,1995). Comparable work on the fer-
(OFFENZELLERet al., 1996). Likewise, acyl- mentation of other peptides is scarce, but pre-
ated peptides can be assumed to have specific cursor-directed feeding work has been re-
polyketide synthases required for the respec- viewed recently (THIERICKE and ROHR,
tive acyl-CoA starter unit contained in or cor- 1993). Some of the results are discussed in
egulated with their respective peptide biosyn- Sect. 3.2.2.1.
thetic genes.
Biosynthetic systems can be identified by
the respective alignments identifying the var- 3.2.1 Fermentation Procedures
ious modules. For isolation of genes conve-
nient use can be made of the highly conserved It is beyond the scope of this chapter to dis-
core sequences to PCR-specific gene seg- cuss data on fermentation procedures which
ments (BORCHERTet al., 1992). Adjacent re- actually are similar in secondary metabolite
gions are explored then, and the correlation is production in general. Therefore, just a few
attempted by gene disruption and agreement remarks on recent developments are added:
of the peptide structure with the predicted The control of parameters in metabolite fer-
modular arrangement (BERNHARDet al., mentations has been improved by measuring
1996; MEISSNERet al., 1996). These proce- not only dissolved oxygen but also redox po-
dures are in complete analogy to polyketide tential, dissolved carbon dioxide (DAHOD,
systems (SCHWECKE et al., 1995). In the case 1993), and culture fluorescence (NIELSENet
Tab. 4. Peptide Synthetases Currently Studied
Peptide Organism Structural Multi-Enzymes Activation

Type" (size kDa)b Modules
Linear Peptides:
ACV Aspergillus nidulans P-3' 425
Acremonium chrysogenum 415b
Streptomyces clavuligerus 420
Ergopeptines Claviceps purpurea R-P-3d 140
370
Gramicidin Bacillus brevis P-15-M' 300
700
Cyclopeptides:
HC-toxin Helminthosporium carbonum P-4' 570b 4
Gramicidin S Bacillus brevis C-(P-5)zg 127b 1
512b 4
Tyrocidine Bacillus brevis C-lOh 123b 1
400 3
800 6
Cyclosporin Beauveria niveum c-lli 17Wb 11
Lactones:
Destruxin Metarhizium anisopliae L-6-MJ 800 6
Actinomycin Streptomyces chrysomallus R-(L-5)zk 49 1
280 2
480 3
SDZ215-104 Cylindrotrichum oligospermum L-111 1650 11
Surfactin Bacillus subtilis R-L-7" 402 3
401 3
144b 1
Branched Cyclopeptides:
Lysobactin Lysobacter sp. P-11-L-9h 450
900
Bacitracin Bacillus licheniformis P-12-C-7" 700
350
750
Cyclodepsipeptides:
Enniatin Fusarium scirpi C-(D-2)3" 347b 2
a The abbreviations used are: P peptide, C cyclopeptide, L lactone, D depsipeptide, R acyl, M modified.
The number of residues activated by each (mu1ti)enzyme is indicated. 3 x 2 stands for the trimerization
of 2 residues.
Size verified by gene sequencing.
' a-Aad-Cys-DVal
DLYSA-Ala-Phe-Pro
f-Val-Gly-Ala-DLeu-Ala-DVal-Val-DVal-T~-(DLeu-T~)3-ethanolamine
c(DPro-Ala-DAla-Aeo)
c(DPhe-Pro-Val-Om-Leu)2
c(DPhe-Pro-Phe-DPhe-Asn-Gln-Val-Orn-Leu)
' c(DAla-NMeLeu-NMeLeu-NMeVal-NMeBmt-Abu-Sar-NMeLeu-Va1-NMeLeu-Ala)
1 c(DHimv-Pro-Ile-NMeVal-NMeAla-PAla)
Mha-c(Thr-DVal-Pro-Sar-Val)
I c(DHiv-NMeLeu-Leu-NMeVal-NMeBmt-Thr-Sar-NMeLeu-Leu-NMeLeu-Ala)
~(3hC~~-Glu-Leu-DLeu-Val-Asp-DLeu-Leu)
DLeu-Leu-c(2hPhe-2hLeu-Leu-DArg-Ile-aThr-Gly-2hAsn-Ser)
O (Ile-Cys)-Leu-DGlu-Ile-c(Lys-DOrn-Ile-Phe-Asn-DAsp-His)
c(NMeVal-DHiV)3
al., 1994). Monitoring of the state of mycelia defined targets under conditions of enhanced
in fermentations permits more detailed inves- recombination could become a valuable tool.
tigations of morphology and vacuolation
(PAULand THOMAS,1996; VANHOUTTEet
al., 1995; NIELSENet al., 1995; NIELSENand 3.2.2.1 Family Exploitations
KRABBEN,1995). Oils have been used with
advantage as carbon sources in cephamycin The concept of peptide families is illus-
fermentations (PARKet al., 1994a, b). Solid- trated in Fig. 5 using the examples of grami-
state fermentation for the production of sur- cidin S, gramicidin, cyclosporin, and aureoba-
factin and iturin by Bacillus subtilis has been sidin. Analogs of gramicidin S have not been
exploited (OHNOet al., 1995, 1996). The sep- detected for a long time. Modern mass spec-
aration of products is a promising approach trometric techniques have considerably en-
to avoid feedback problems. Attempts with hanced analytical resolution (NOZAKI and
an aqueous two-phase system have been MURAMATSU, 1987; THIBAULT et al., 1992).
made in subtilin fermentation with Bacillus Both bacterial peptides show quite limited
subtilis (KUBOIet al., 1994). Alternatively, a structural variations. The only remarkable
microfiltration module was used in nisin pro- feature is that the intiating valine in linear
duction with Lactococcus lactis (TANIGUCHI gramicidin has been found to be replacable
et al., 1993). by isoleucine. Likewise, only tryptophan in
The discovery of the role of autoregulators position 11 of gramicidin has been exchanged
in metabolite production (CHATERand BIBB, by other aromatic amino acids. From large-
Chapter 2, this volume) has led to production scale fermentations of the fungal metabolites
processes now. Product yields are enhanced cyclosporin and aureobasidin 32 and 29 ana-
by the controlled addition of an autoregula- logs have been purified and characterized. In
tors at high cell concentrations (YANGet al., general, the number of tolerated substitutions
1996). is not more than two. This indicates a sieving
mechanism presumably exerted by the bio-
synthetic process. The combinatorial game as
played in chemical synthesis of analogs thus
3.2.2 Structural Alterations meets only limited success in the enzymatic
system. In fact, only a small amount of pre-
In order to achieve structural alterations dictable analogs is detected. 18 variable posi-
two approaches are available, either to screen tions should permit the formation of more
for analog producers or to manipulate the than lo4 cyclosporin analogs, but only 32
peptide composition by precursor feeding. have been detected (FLIRIet al., Chapter 12,
Genetic manipulations are less obvious be- this volume). The respective abortion prod-
cause the respective selection processes are ucts, however, have not been investigated yet.
still unclear. It seems obvious that peptides Likewise, 14 replacements of aureobasidins
encountered in various sources show a unique permit the synthesis of more than lo3 ana-
design with respect to structural variations. logs, 29 of which have been found (IKAIet al.,
These variations are evident from the obser- 1991a, b; AWAZUet al., 1995).
vation of peptide families. It is fascinating to Within a notably small range of variations
observe variations in certain residues while analogs have been detected by screening
others are invariant. Remarkable invariance methods. The scope of this approach is illus-
can be achieved solely by selection within the trated by LANCINIand CAVALLERI in Chap-
enzymatic steps of peptide synthesis and may ter 9 on dalbaheptides. Screening for cyclo-
reflect proof-reading mechanisms known sporin analogs gave rise to the peptidolac-
from the ribosomal machinery. It can be as- tone analog SDZ204-125 (Tab. 4), where D-ala-
sumed that structural selections are con- nine is replaced by D-hydroxyisovalerate, the
nected with biological targets which still need variable position 2 is dominated by threonine,
to be identified. Once biological selection and one N-methylation is missing. In the case
processes are understood selection against of the cyclodepsipeptide enniatin different
fOm
Leu Cit
Abu Lys
'DPhe +2Pro +Val +*Om +'Leu
t .1
"Leu +*'Om +"Val t2'Prot"DPhe
flle
N a l 4 l y +'Ma -+DLeu+Ma +DVal +Val +'DVal-+
Trp +DLeu -+"Trp +DLeu +Trp +DLeu +Trp +EA

Phe
TYr
DVaY
MeLeu
MeAOC*
DSer Leu Val MedeoxyBmt
8DAla+YeLeu+MeLeu+MeVal+1MeBmt
t .1 (3)
Ala+MeLeutVal+MeLeu+MeGly+2Abu
Abu Leu Nva Val Gly Ala
Leu lle Thr
Melle Val
Nva
2,5h3mP
2,4h3mP
2,3h3mP MeTyr
Dhiv Val Me2hPhe
'DHmp+Me2VaI+'Phe~Me4Phe
t .1
Meg2hVal+aLeu+Me7Val+6alletsPro
MeVal MeLeu Val

N2MeAsp Melle Met
Leu
MeTyr
mFPhe MemFPhe
oFPhe MeoFPhe
'DHmp+Me2Val+'Phe+Me*Phe
t .1
Ye82hVal+8Leu+Me7Valt6alle+6Pm
alle Nle Hyp
Nva Met Spro (4 b)
Fig. 5. Peptide analogs isolated from fermentation broths: (1) gramicidin S analogs, (2) linear gramicidins,
(3) cyclosporins, and (4) aureobasidins. Only two positions are variable in the Bacillus peptides (1) and (2).
Note that only one of the four tryptophans has been found to be exchanged. The fungal peptides show
conserved and less restricted positions. In all analogs found usually one of the variations is present, and in
rare cases two positions are exchanged compared to the main products. More exchanges apparently reduce
the rate of synthesis so that such products become extremely rare. Aureobasidin analogs (4b) were ob-
tained by directed biosynthesis.
4 Future Prospects 297
Fusaria have been shown to have altered pro- DOHREN,1982; LAWENand TRABER,1993).
files of branched-chain analogs. A study of Rates of synthesis of gramicidin S analogs
the respective peptide synthetases revealed have been studied with respect to multiple
altered substrate binding sites (PIEPERet al.,
1992). D-CHA
D-T~A'
D-Trp
D-OmTyr
3.2.2.2 Biosynthetic Manipulations D-mTyr
D-oTyr
The potential of biosynthetic manipulations D-Tyr
has to address either the precursor situation D-IPhe Sar Leu
D-BrPhe 4s-Pro Nva
or the enzymatic template. A convenient pro- D-CIPhe 3s-Pro Nle
cedure includes feeding of substrates or sub- D-oFPhe hPro alle Nle
strate analogs. Template alterations are still D-mFPhe Aze Nle Lys alle
in their infancy and they will become of spe- D-pFPhe 3,4APro lle Arg lle
cial importance when selection against new
targets not encountered in natural or ecologi- 'DPhe -'Pro -Val -*Om -'Leu (1)
cal systems becomes available. t 1
"Leu +*'Om $ ' V a l" P r o " D P h e
MeCys
Directed biosynthesis DLys MeaThr
DPhe MeSer
DVal+ Me2a4m4HEA
Work in directed biosynthesis has been suc- DCys MecyclodihydroBmt
cessful in various cases including tyrocidines, DcyclopropylGly+ Me2a3h4buOA
actinomycins, viridogriseins (where the natu- 1-CI-D-vinylGly MeNva+ Me2a3h4,8mNA
ral D-hydroxyproline is replaced by feeding of DtbuAla MetbuGly Me2a3h60EA
MetbuAla Me2a3h4m20A+
proline to the nonhydroxylated analog which 2-F-DAla+
2-CI-DAla+ MeallylGly Me2a3h4mOA
is a more effective antibiotic; OKUMURA,R-Ala+ MeCPG+ MeNle
1990), ergot peptides, enniatins, cyclosporins, DAbu+ Mealle+ Me3hCHA+
and aureobasidins (for a review, see THIER- Gly+ Melle+ MeCHA
ICKE and ROHR,1993). The remarkable ex-
MedihydroBmt
tension of products is illustrated with aureo- MeLeu
basidin analogs which are not available with- vinylGly MeAOC'
out feeding (Fig. 5). It is possible to replace DSer Leu Leu Val MedeoxyBmt
proline by hydroxyproline and thioproline 'DAla+MeLeu+MeLeu+Me%al+'MeBmt
(TAKESAKO et al., 1996). However, systems t 1 (2)
may be surprisingly restrictive. Attempts to AlatMeLeut~altMeLeui-MeGlytEAbu
replace 4-hydroxyproline in the echinocandin Abu Leu Nva Val Gly Ala
Leu Ile Thr
analog L-671329 by feeding a culture of the Melle Val
producer Zalerion arboricola were not suc- Leu Nva
cessful (ADEFARATI et al., 1991). It has been
established later that the hydroxy group is es- Gly+ Ile+ allylGly
sential for the antifungal activity. Such results Nva alle+ aThr
imply a target adaptation of biosynthetic sys- Nle CPG CYS
vinylGly allylGly Ile
tems. Val Abu PPT
An extension of in vivo feeding approaches CYS tbuAla
is the in vitro enzymatic synthesis of peptides Phe tbuGly
employing the isolated multienzyme systems aAla
(KLEINKAUFand VON DOHREN,1997). The Fig. 6. Peptides formed by in vitro synthesis with
potential of the method is illustrated in Fig. 6 supply of amino acids, ATP, and SAM if required
using gramicidin S, cyclosporin, and the pep- (1) gramicidin S, (2) cyclosporin, and (3) SDZ214-
tolide SDZ214-125 (KLEINKAUF and VON 125. For abbreviations see p. 278.
2hiCap+ hydration, the reprogramming approach is

D2h3mVa+ MetbuAla quite limited. In the peptide field a much
D2hVa+ MeallylGly Me-dihydroBmt
D2hBu+ MeCPG Me2a3h4mOA larger reservoir of structures seems to be
vinylGIy Mealle+ MeLeu’ available. Using a double recombination ap-
DLac MeAbu+ MeCHA proach STACHELHAUS et al. (1995, 1996b)
succeeded in replacing the last amino acid
‘D H i v ~ M e L e u ~ L e u ~ M e l l a l - r M e ’ B ~
r
AlatMeLe~t~LeutMeLeutMeGIy+~Thr
J-
module of surfactin synthetase by substituting
(3) the terminal leucine domain with bacterial or-
nithine and phenylalanine domains and with
AbU+ Leu 3hNva fungal cysteine and valine domains. This ap-
vinylGly Abu proach offers exciting opportunities of vary-
Nva Nva ing residues within a given structural type. Fi-
CYS Nle nally, even new peptide structures should be
available from designed enzyme systems.
+ molecular mass by FAB-MS
Fig. 6. Continued.
substitutions. It has been shown that rates are

4 Future Prospects
roughly additive and overall rates decrease
drastically when poor substrates are com- 4.1 Peptides of Ribosomal Origin
bined. In the evaluation of properties of cy-
closporin and SDZ213-125 analogs a large Impressive examples of antimicrobial prop-
number of cyclopeptides have been produced erties of peptides of ribosomal origin have
with these giant multienzymes in the mg been demonstrated. So far uses have been
range. A comparison revealed that the pep- limited to topical applications (a magainin
tolide synthetase had a much more restricted analog is in clinical trial for cutaneous appli-
substrate profile (LAWEN and TRABER, cation in skin and eye infections), and no data
1993). are available on the uptake in tissues. An ob-
In v i m investigation of enzyme properties vious limitation seems to be the degradation,
is essential for the efficient utilization of sub- especially by trypsin-like proteases. Promis-
strate feeding and the evaluation of structural ing fields of research include the use of pre-
alterations of enzymes. cursor peptides and sequence alterations to
improve proteolytic stability. Efforts for the
exploitation of peptide structures are justified
Genetic approaches by the ease of engineering these peptide
structures compared to enzymatic systems
Metabolic engineering could approach both and research has been directed towards multi-
substrate pools and composition as well as the drug resistant bacteria, sepsis, and even anti-
substrate selection properties of enzymes in- cancer drug development (BOMAN,1995).
volved. Should the latter approach attempt to A fascinating area of research aims at the
change the amino acid sequence of the con- in vivo production of peptide factors to coun-
sidered peptides the term “reprogramming” teract insect-borne diseases like malaria, try-
has been introduced. The idea of reprogram- panosomiasis, and filariasis. It has been pro-
ming is to exchange biosynthetic modules in posed to modify the insect immune system of
order to arrive at new structures. This ex- the respective vectors or symbionts in order
change has been demonstrated in several to eliminate the parasites (HAMet al., 1994).
cases for the polyketide forming systems. As
polyketide building blocks are restricted to
acetate, propionate, or butyrate residues
which may be modified by reductions and de-
5 Compilation of Compounds 299
4.2 Peptides of Nonribosomal retrieved best from the available data banks
(Chemical Abstracts; Kitasato Microbial
Origin Chemistry Database, Usako Corp., Tokyo;
Antibase, Database, Chemical Concepts, Hei-
Peptides of enzymatic origin continue to delberg). Data on ribosomal peptides are eas-
emerge from various screening programs. ily accessed from the standard gene banks.
New promising sources are, e.g., marine or- For retrieval of additional information the
ganisms like sponges as a rich source of pep- given names of compounds, organisms, au-
tidic structures, harboring various kinds of thors, and years of publication are sufficient.
microorganisms. The exploitation of such mi- Peptides of ribosomal origin have been
crobes that are often surface-associated has grouped in Tab. 5. This compilation illus-
just begun and is mainly hampered by the trates the historically grown pathways of the
lack of experience in cultivation. studies which originated mainly in the insect
The extensive understanding of biosynthet- field, then achieved spectacular results with
ic reactions permits various structural modifi- frog peptides, and now identify many antimi-
cations to be introduced into the genetic crobial peptides from mammalian sources.
background, thus facilitating the biotechno- For details, the reader is referred to recent re-
logical production of compounds. New at- views (HANCOCK et al., 1995; LEHRERet al.,
tempts will be made by the implementation of 1993; BOMANet al., 1994; BOMAN,1995). Mi-
strategies from ribosomal antibiotic induction crobial peptides of ribosomal origin are the
approaches to various microbial, plant, and subject of Chapter 8 on lantibiotics (JACKet
animal sources. A lot can be learned from the al., this volume) while the fields of neuropep-
interaction of organisms in ecosystems, and tides and hormones are not within the scope
genome sequencing will certainly enhance of the chapter.
this understanding considerably. The advanc- Peptides of nonribosomal origin are listed
ing analytical techniques will permit a more in Tab. 6. A number of observations can be
rapid access to even more complicated natu- made concerning the sources especially suited
ral products. The production of such com- for peptides. Thus the actinomycetes are still
pounds will remain a challenge for biotechnol- the leading group for new products, and here
%Y* extensive variations in the structural types of
compounds are found. The groups of dalba-
heptides and thiopeptides in particular have
benefitted from directed screening ap-
proaches. Bacilli have received less attention,
5 Compilation of but the potential of various cyclopeptides
from well-known strains and also from ma-
Compounds rine Bacilli with slightly deviating structures
has not yet been explored in depth. A surpris-
It is the purpose of this compilation covering group are the blue-green algae, mainly
ing most peptides and related structures pub- cyanobacteria, which show an impressive var-
lished during the last five years (1992-1996) iability of peptide structures with various
to illustrate trends in sources, screening tar- properties. This variability implies frequent
gets, variability of structures within groups of horizontal transfer of biosynthetic genes, a
organisms, and to address some problems of promising field to be exploited in order to
source identification and stability. This list make use of natural combinatorial ap-
updates and extends the table of peptides proaches. Fungi continue to provide interest-
presented in Vol. 4 of the First Edition of ing compounds, most peptides being rather
“Biotechnology (KLEINKAUFand VON small. Large fungal peptides are mainly as-
”
DOHREN,1986). A more extensive compila- sembled in the family of peptaibols.

tion of peptide structures including publica- Plant peptides of presumably nonribosomal
tions until 1989 can be found in VON DOH- origin often are not modified and could as
REN (1990). Data on specific compounds are well originate from ribosomal precursors by
Tab. 5. Compilation of Peptides of Ribosomal Origin
Peptide Source Sequencea Properties Reference

Microbial Sources
Leukocin A Leuconostoc gelidum KYYGNGVHCTKSGCSVNWGEAFSAGVHR- antibacterial HASTINGet al. (1991
Ual 187 LANGGNGFW (G -)
Nisin Lactococcus lactis ITSISLCTPGCKTGALMGCNMKTATCHC- antibacterial HURST(1981)
SIHVSK (G*)
Pep-5 Staphylococcus epidermidis TAGPAIRASVKQCQKTLKATRLFTVSCK- antibacterial KALETTAet al. (198
GKNGCK (G*)
Plantaricin Lactobacillus plantarum AY SLQMGATAIKQVKKLFKKW antibacterial NISSEN-MEYER et al.
A (1993)
Sillucin Rhizomucor pusillus ACLPNSCVSKGCCCGBSGYWCRQCGIKYTC antibacterial BRADLEYand SOM-
(G*) kuti (1979)
Subtilin Bacillus subtilis MSKFDDFDLDVVKVSKQDSKITPQWKSES- antibacterial BANERJEE and
LCTPGCVTGALQTCFLQTLTCNCKISK HANSEN(1988)
Plants
AFPl Rape (Brassica napus) QKLCERPSGTWSGVCGNNNACKNQCINLE- antifungal TERRASet al. (1992)
KARHGSCNWFPAHK
AFP2 Turnip (Brassica rapa) QKLCERPSGTXSGVCGNNNACKNQCIR antifungal TERRASet al. (1992)
Ac-AMP Amaranth (Amaranthus VGECVRGRCPSGMCCSQFGYCGKGP- antibacterial, BROEKAERT et al.
caudatus) KYCGR antifungal (1992)
Crambin Crambe plants (Crambe TTCCPSIVARSNFNVCRIPGTPEAICATYTG- TEETERet al. (1981)
abyssinica) CIIIPGATCPGDYAN
MBP-1 Maize (Zea mays) RSGRGECRRQCLRRHEGQPWET- antifungal DUVICKet al. (1992)
OECMRRCR
Mj-Amp1 Mirabilis lalapa QCIGNGGRCNENVGPPYCCSGFCLRQPGQY- antibacterial CAMMUEet al. (1992
(2 types) GYCKNR (G*); anti-
fungal
Rs-AFP1 Radish (Raphanus sativus) QKLCERPSGTWSGVCGNNNACKNQCINLE- antibacterial TERRASet al. (1990)
Insects
Abaecin Honey bee (Apis mellificu) YVPLPNVPQPGRRPFPTFPGQGPFNP- antibacterial CASTREELS
et al.
KIKQP QG Y (1990)
Andropin Fruit fly (Drosophilu VFIDILDKVENAIHNAAQVGIGFAKPFEK-
melunoguster) LINPK
Apedaecin Honey bee (Apis rnellificu, GNNRPVYIPQPRPPHPRI antibacterial CASTREELS et al.
IA lymph fluid) (G-) (1989)
Bactericidin Tobacco hornworm (Munducu WNPFKELERAGQRVRDAVISAAPAVATVG- antibacterial, DICKINSON et al.
B2 sextu, larvae hemolymph) QAAA IAR G cytotoxic (1988)
Bombolitin Bumblebee (Megubombus IKITTMLAKLGKVLAHVa antibacterial, ARGIOLASand
pennsylvunicus, venom) cytotoxic PISANO(1985)
Cecropin Silk moth (Bombyx mori) RWKIFKKIEKVGQNIRDGIVKAGPAVAVVG- antibacterial Qu et al. (1987)
QAATI
Crabrolin Hornet (Vespu crubro, venom) FLPLILRKIVTALa cytotoxic ARGIOLAS and
PISANO(1984)
Drosocin Fruit fly (Drosophilu GKPRPYSPRPTSIHPRPIRV antibacterial BULETet al. (1993)
melunoguster)
Insect Dragonfly (Aeschnu cyuneu, GFGCPLDQMQCHRHCQTITGRSG- antibacterial BULETet al. (1992)
defensin larvae) GYCSGPLKLTCTCYR (G+)
Lepidopte- Silkworm (Bombyx rnori) RWKLFKKIEKVGRNVRDGLIKAGPAIAVIG- TESHIMAet al. (1987
ran C QAKSL
Mastoparan Wasp (Vespulu lewisii, venom) INLKALAALAKKIL antibacterial BERNHEIMER and
(G+) RUDI(1986)
Melittin Bee (Apis rnellificu, venom) GIGAVLKVLTTGLPALISWIKRKRQQ antibacterial, TOSTESONand
cytostatic, TOSTESON (1984)
antifungal
Phormicin Blowfly (Phormiu terrunovu) ATCDLLSGTGINHSACAAHCKLRGNRG- antibacterial LAMBERTet al. (198
A (B) GYCNGKGVCVCRN (G+)
Royalisin Bee, royal jelly (Apis mellificu) VTCDLLSFKGQVNDSACAANCLGKAGGH- antibacterial FUJIWARA
et al. (199
CEKGVCICRKTSFKDLWDKYF (G+)
Sapecin Fleshfly (Surcophugu peregrinu) ATCDLLSGTGINHSACAAHCLLRGNRG- antibacterial HANZAWA et al.
GYCNGKAVCVCRN (1990)
Sarcotoxin Fleshfly (Surcophugu peregrinu) GWLKKIGKKIERVGQHTRDATIQGLGIAQ- antibacterial OKADAand NATOR
QAANVAATARa
Tab. 5. Continued
Polyphe- Horseshoe crab (Limulus RRWCFRVCYRGFCYRKCRa antibacterial LAMBERT

et al. (198
musin polyphemus)
Tachyplesin Horseshoe crab (Limulus KWCFRVCYRGICYRRCR antibacterial; NAKAMURA et al.
(3 types) polyphemus) antifungal (1988)
Toxin 2 Sahara scorpion (Androctonus VKDGYIVDDVNCTYFCGRNAYC- BONTEMSet al. (199
australis Hector) NEECTKLKGESGYCQWASPY GNA-
CYCKLPDHVRTKGPGRCH
Molluscs
CARP Mytilus edulis AMPMLRLa catch-relaxing HIRATAet al. (1987)
peptide
(muscle)
MIP Land snail (Achatina fulica) AAPKFVGRRGAPYFV contraction IKEDAet al. (1992)
(10 types) inhibitor
MIP Land snail (Helix pomatia) GAPAFV contraction IKEDAet al. (1992)
(12 types) inhibitor
Frogn oads
Adeno- Two-colored leaf frog GLWSKIKEVGKEAAKAAAKAAGKAAL- antibacterial, DALYet al. (1992)
egulin (Phyllomedusa bicolor) GAVSEAV antifungal
Bombinin Yellow-bellied toad (Bombina GIGALSAKGALKGLAKGLAEHFANa anibacterial CSORDASand MICH
variegata) (1970)
BLP-1 Asian toad (Bombina orientalis) GIGASELSAGKSALKGLAKGLAEHFANa antibacterial GIBSONet al. (1991)
(4 types)
Brevinin European frog (Rana esculenta) FLPLLAGLAANFLPKIFCKITRKC antibacterial, SIMMACOet al. (199
cytotoxic
Dermasep- Sauvage’s leaf frog ALWKTMLKKLGTMALHAGKAALKAAAD- antibacterial, MOR et al. (1991)
tin 1 (Phyllomedusu suuvugii) TISQGTQ antifungal
(6 types)
Esculentin European frog (Runa esculentu) GIFSKLGKKIKNLLISGLKNVG- antibacterial SIMMACO
et al. (199
KEVGMDVVRTGIDIAGCKIKGEC
Magainin I Xenopus luevis, skin GIGKFLHSAGKFGKAFVGEIMKS anibacterial, ZASLOFF(1987)
antifungal
PGLa Xenopus luevis, skin GMASKAGAI AGKI AKVALKAL antibacterial KUCHLERet al. (198
PGQ Xenopus luevis, stomach GVLSNVIGYLKKLGTGALNAVLK antibacterial; MOOREet al. (1991)
antifungal
Ranalexin Bullfrog ( R u m cutesbeiunu, FLGGLIKIVPAMICAVTKKC antibacterial CLARKet al. (1994)
skin)
XPF Xenopus luevis, skin GWASKIGQTLGKIAKVGLKELIQPK antibacterial SURESand CRIPPA
( 1984)
Snakes
Toxin 1 Waglers pit viper (Trimeresurus GGKPDLRPCHPPCHY IPRPKPR SCHMIDT
et al. (1992
wugleri, venom)
Mammals
Bactenecin bovine neutrophils RLCRIWIRVCR antibacterial ROMEOet al. (1988)
BacS bovine neutrophils RFRPPIRRPPIRPPFYPPFRPPIRPPIFP- antibacterial FRANKet al. (1990)
PIRPPFRPPLRFP (G-)
BNBD-2 bovine neutrophils VRNHVTCRINRGFCVPIRCPGRTRQIGTCFG- antibacterial SELSTEDet al. (1993
(13 types) PRIKCCRSW
Cecropin P1 Pig (Susscrofu, small intestine) SWLSKTAKKLENSAKKRISEGIAIAIQGGPR antibacterial LEE et al. (1989)
(G-)
Cryptdin Mouse (Mus musculus, LRDLVCYCRSRGCKGRERMNGTCRKGHL- SELSTEDet al. (192)
(5 types) intestine) LYTLCCR
Endozepine Pig (Susscrofu domesticu) KQATVGDINTERPDILDKGKAKW- antibacterial AGERBERTH et al.
DAWNGLKGTSKEDAMKAYINKVEELKK- (1993)
KYGI
Gastric Pig (Sus scrofu domesticu) ISDYSIAMDKIRQQDFVNWL- antibacterial AGERBERTH
et al.
Tab. 5. Continued

Hiastadin Macaca fascicularis DSHEERHHGRHGHHKYGRKFHEKHHSHR- antifungal Xu et al. (1990)
GYRSNYLYDN
HNP-1 Human neutrophils ACYCRIPACIAGERRYGTCIYQGRLWAFCC antibacterial, LEHRERet al. (1991)
(6 types) antifungal,
cytotoxic
Indolicidin Bovine neutrophils ILPWKWPWWPWRR antibacterial SELSTED et al. (1992)
Lacto- Bovine (N-terminal of lacto- FKCRRWQWRMKKLGAPSITCVRRAF antibacterial BELLAMY et al. (199
ferricin ferrin)
MCPI Rabbit (Oryctolagus cuniculus, WCACRRALCLPRERRAGFCRIR- antibacterial, SELSSSTEDet al.
(2 types; macrophage; neutrophils) GRIHPLCCRR antifungal, (1983)
NP-1 cytotoxic
(5 types)
Peptide Pig (Susscrofa dornestica) RADTQTYQPYNKDWIKEKIYVLLRRQAQ- antibacterial AGERBERTH et al.
3910 QAGK (1993)
RatNP-1 Rat ( R a m s norvegicus) VICYCRRTRCGFRERLSGACGYR- antifungal, EISENHAUER et al.
(4 types) GRIY RLCCR antibacterial (1989)
Sarcolipin Rabbit (Oryctolagus cuniculus, MERSTRELCLNF'IVVLITVILIWLLVRSY QY proteolipid-like, WAWRZYNOW et al.
skeletal muscle) ionophore? (1992)
Seminal- Bovine (Bos taurus, seminal SDEKASPDKHHRFSLSRYAKLANRLANPKL- antibacterial, REDDYand BHAR-
plasmin plasma) LETFLSKWIGDRGNRSV antifungal, GAVA (1979)
cytotoxic
TAP Bovine (Bos taurus, tracheal NPVSCVRNKGICVPIRCPGSMKQIGTCV- antibacterial, DIAMOND
et al. (199
mucosa) GRAVKCCRKK antifungal
a Only plain sequences are given, no disulfide links are indicated: a: C-terminal amidated; most accession numbers of the gene sequences can b
retrieved from HANCOCKet al. (1995).
Tab. 6. Compilation of Peptides of Nonribosomal Origin (Bacteria/Fungi/Plants/Animals)
Compound Organism Structural Properties Reference

Type"
Bacteria - Bacilli
FR901537 Bacillus sp. AA-M antitumor OOHATAet al. (1995)
Fusaricidin Bacillus polymyxa R-L-6 antibacterial KAJIMURAand KANEDA(1996)
N-4909 Bacillus sp. L-8 stimulates apolipoprotein HIRAMOTO et al. (1996)
E secretion
Halobacillin Bacillus sp. (marine) L-8(2) cytotoxic TRISCHMANN et al. (1994)
Isohalobacillin B Bacillus sp. L-8 ACAT inhibitor HASUMI et al. (1995)
Surfactins Bacillus subtilis natto L-S(2) OKAet al. (1993)
Surfactin-like Bacillus pumilus (marine) L-8(2 KALINOVSKAYA et al. (1995)
Bacillopeptin Bacillus subtilis C-8(2) antifungal (phytopathol.) KAJIMURA et al. (1995)
Cereulide Bacillus cereus C-D-12 K +-complex SUWANet al. (1995)
Homocereulide Bacillus cereus (marine) C-D-12 extremely cytotoxic WANGet al. (1995)
Bacitracins Bacillus sp. P-12-C-7- antibacterials IKAIet al. (1995)
M(2)
Actinomycetes
Monamidocin Streptomyces sp. P-2 fibrinogen receptor KAMIYAMA
et al. (1995)
antagonist
Matylstatin Actinomadura atra- R-P-2-M type IV collagenase OGITAet al. (1992)
mentaria inhibitor
PhevaIin Streptomyces sp. C-2-M calpain inhibitor ALVAREZet al. (1995)
Thaxtomin Streptomyces scabies C-2-M phytotoxin GELINet al. (1993)
Cutinostatin Actinomycete R-P-2-M cutinase inhibitor HIGASHIet al. (1996)
Amonobactins Aeromonas hydrophila R-P-3 siderophore TELFORD et al. (1994)
Napsamycin Streptomyces sp. R-P-3-M(U) antibacterial CHATTERJEEet al. (1994)
(Pseudomonas)
Nerfilin I Streptomyces halstedii R-P-3 neurite outgrowth inducer HIRAOet al. (1995)
Pepticinnamins Streptomyces sp. R-P-3-M farnesyl-protein OMURAet al. (1993)
Tab.6. Continued
Type"
~Y-MAPI Streptomyces sp. HIV-1 protease inhibitor STELLAet al. (1991)
Micromonospora sp.
Nocardia sp.
Mer-N5075A Streptomyces chromo- P-4-M(U) HIV-1 protease inhibitor KANETOet al. (1993)
fuscus
MTMTLA Streptomyces griseus P-4-M calpain inhibitor ALVAREZ et al. (1994)
GE20372 Streptomyces sp. P-4-M HIV-1 protease inhibitor STEFANELLIet al. (1995)
MR-387 Streptomyces neyagawa- P-4-M aPase inhibitor CHUNGet al. (1996)
ensis
Formobactin Nocardia sp. R-D-4 radical scavenger, neu- MURAKAMI
et al. (1996)
ronal protecting
Muredomycins Streptomyces flavidovirens R-P4-M(U) antibiotic ISONOet al. (1993)
Echinoserine Streptomyces tendae (R-P-4)2-M antibacterial (weak) BLUMet al. (1995)
BE-22179 Streptomyces gangtokensis (R-P-4)2-M topoisomerase I1 inhibitor OKADAet al. (1994)
Poststatin Streptomyces virido- P-5-(CO) pro-endopeptidase AOYAGIet al. (1991)
chromogenes inhibitor
Cyclot hialidine Streptomyces filipinensis P-5-M DNA gyrase inhibitor KAMIYAMA
et al. (1994)
WS 1279 Streptomyces willmorei R-P-5 bone marrow growth
stimulating
Sandramycin Nocardioides sp. (R-p-5)~ antitumor MATSONet al. (1993)
Rotihibin Streptomyces R-P-5-AA-M plant growth regulator FUKUCHI
et al. (1995)
graminofaciens
ws-7338 Streptomyces sp. c-5 endothelin receptor MIYATAet al. (1992)
antagonist
BE-18257 Streptomyces misakiensis c-5 enothelin binding inhibitor KOJIRIet al. (1991)
Aurantimycin Streptomyces aurantiacus C-6 antibacterial, cytotoxic GRAFEet al. (1995)
WF11899AIBIC Coleophoma emperri R-C-6(4) antifungal IWAMOTO et al. (1994)
Protactin Streptomyces cum- R-L-6 antibacterial. antitumor HANADAet al. (1992)
merosporus
Salinamide Streptomyces sp. (marine) R-L-6-M anti-inflammatory TRISCHMANN et al. (1994)
IClOl Streptomyces sp. R-L-6-M extracellular matrix UENOet al. (1993)
antagonist
RPI-856 Streptomyces sp. P-6(7)-M HIV protease inhibitor ASANOet al. (1994)
Bottromycin A2 Streptomyces bottropensis P-7-C-4-M KANEDA(1992)
Cochinmicins Microbispora sp. P-7-L-5 endothelin antagonist LAMet al. (1992)
Piperastatin A Streptomyces lavendofoliae R-P-7 sprine carboxypeptidase MURAKAMI et al. (1996)
inhibitor
WS9326A Streptomyces viola- R-L-7 tachykinin antagonist SHIGEMATSU
et al. (1993)
ceusniger
Kistamicins Microtetraspora parvosata P-7-M antiviral NARUSEet al. (1993)
MM 5526618 Amycolatopsis sp. R-P-7-M antibacterial Box et al. (1991)
(GLYC)
Galacardins Saccharothrix sp. R-P-7-M antibacterial TAKEUCHI
et al. (1992)
(GLYC)
Balhimycin Amycolatopsis sp. R-P-7-M antibacterial NADKARNI
et al. (1994)
(GLYC)
Chloropeptins Streptomyces sp. P-7-M gp120-CD4 binding MATSUZAKIet at. (1994)
inhibitor
WS9326A Streptomyces viola- R-C-7 tachykinin inhibitor HAYASHIet al. (1992)
ceusniger
A21459 Actinoplanes sp. C-8 antibacterial SELVAet al. (1996)
Himastatin Streptomyces hygro- (W2 antitumor LEET et al. (1996)
scopicus
Quinoxapeptin MA7095 (nocardioform) R-(P-5)2-M HIV-1,2 reverse tran- LINGHAINet al. (1996)
scriptase inhibitor
Cypemycin Streptomyces sp. P-11-M antibiotic MINAMIet al. (1994)
Promotiocin Streptomyces sp. P-13-C-10 tip A promoter inducing YUNet al. (1994)
[PYRI-M
Berninamycin Streptomyces bernensis P-14-C-12 LAU and RINEHART(1994)
[PYR]-M
Geninthiocin Streptomyces sp. P-15-C-13 tip A promoter inducing YUNet al. (1994)
[PYRI-M
GE37,468 Streptomyces sp. P-15-C-10-M antibacterial FERRARIet al. (1995)
Tab. 6. Continued

Type"
A10255BlGlJ Streptomyces gardneri P-1742-14 DEBONOet al. (1992)
[PYRI-M
Promoinducin Streptomyces sp. P-18-C-14 YUNand SETO (1995)
[PYRI-M
Cyanobacteria
Aeruginosin Microcystis aeruginosa R-P-3 trypsin inhibitor MURAKAMI
et al. (1995)
Aeruginosin 298A Microcystis aeruginosa P-4 thrombin and trypsin MURAKAMI
et al. (1994)
inhibitor
Muscoride A Nostoc muscorum P-4-M NAGATSUet al. (1995)
(TJ=P)
Cryptophycin Nostoc sp. L-4(4) BARROWet al. (1995)
Antillatoxin Lyngbya majuscula L-4-M(4) ichthyotoxic TAKIZAWA et al. (1995)
Linear peptides Microcystis sp. R-P-4(2,3) CHOIet al. (1993)
Microginin Microcystis aeruginosa P-5 angiotensin converting OKINOet al. (1993)
enzyme inhibitor
Microcolins Lyngbya majuscula R-P-5-M immunosuppressive KOEHNet al. (1992)
Nodularins Nodularia spumignea C-5(2,3) NAMIKOSHI et al. (1994)
Anabaenopeptins Anabaena flos-aquae P-6-C-5(4) HARADAet al. (1995)
Nostocyclamide Nostoc sp. C-6 antialgal, anticyano- TODOROVA et al. (1995)
bacterial
Westiellamide Westiellopsisprolifica C-6-M cytotoxic (moderate) PRINSEPet al. (1992)
Oscillamide Oscillatoria agardhii C-6-M(4) chymotrypsin inhibitor SANOand KAYA(1995)
Micropeptin 90 Microcystis aeruginosa R-L-6-M plasminltrypsin inhibitor ISHIDAet al. (1995)
DHB-microcystin-RR Oscillatoria agardhii C-7-M(2) SANOand KAYA(1995)
Nostocyclin Nostoc sp. R-P-7-L-5-M protein phosphatase-1 KAYAet al. (1996)
inhibitor
Microcystilide A Microcystis aeruginosa R-P-7-L-6-M cell differentiation TSUKAMOTO
et al. (1993)
Aeruginopeptins Microcystis aeruginosa R-P-8-L-6-M HARADA
Calophycin Calothrix fusca C-10-M fungicidal MOONet a
Puwainaphycins Anabaena sp. R-C-lO(2) cardioactive GREGSO
Laxaphycins Anabaena laxa c-11 antifungal FRANKM
Hormothamnin Hormothamnion entero- c-11 GERWIC
morphoides
Schizotrin Schizothrix sp. R-P-13-C- antimicrobial PERGAM
1W)
Microviridins Microcystis aeruginosa R-P-14- elastase inhibitor OKINOet
(C-4L-4)
Bacteria - Various
Epothilon Mucor hiemalis AA-M(PK) antifungal, cytotoxic GERTHet
Fluvibactin Vibrio fluvialis R3-A siderophore YAMAMO
WS75624 Saccharothrix sp. R-P-2-M endothelin converting TSURUM
enzyme inhibitor
CI-4 Pseudomonas sp. (marine) c-2 chitinase inhibitor IZUMIDA
e
TAN-057 Flexibacter sp. P-3-M antibacterial ( S . aureus KATAYA
methicillin resistant)
Methanofurans Methanobacterium R-P-3-M natural cofactor SULLION
thermoautotrop hicum
Cyclotetrapeptides Lactobacillus sp. c-4 metal binding, melanin KURANA
inhibitor; tyrosinase et al. (19
inhibitor
Rakicidins Micromonospora sp. L-4-M cytotoxic MCBRIEN
Glycopeptidolipid Mycobacterium sene- R-P-4-M LOPEZMA
antigen galense (GLYC)
Thiangole Polyangium sp. R-P-4-M HIV-1 inhibitor BOYCEet
FR-901,228 Chromobacterium C-R-P-4-M antitumor LI et al. (1
violaceum
Thiangazole Polyangium sp. R-PJ-M JANSENet
Aletrobactin A Alteromonas luteoviolacea R-P-6-L-5 siderophore DENGet a
Azoverdin Azomonas macrocytogenes R-P-7-M siderophore LINGETet
YM4714112 Flexibacter sp. P-7-L-5 elastase inhibitor ORITAet
Myxochromide A Myxococcus virescens R-L-6 TROWITZ
TAN-1511 Streptosporangium R-S-R ‘-P-7 induction of cytokines TAKIZAW
amethystogenes
WLIP Pseudomonas reactans R-P-9-L-6 white line inducing HANet al.
principle
Tab. 6. Continued
Type"
Vioprolides Cystobacter violaceus L-9-M(2) antifungal, cytotoxic SCHUMMER et al. (1996)
Syringomycin Pseudomonas syringae R-L-9 phytotoxin FUKUCHIet al. (1992)
Ferrocins Pseudomonas jluorescens R-L-10-M siderophore TSUBOTANI et al. (1993)
Pholipeptin Pseudomonas sp. R-P-1l-L- UI et al. (1995)
9(2)
Bu-2841 unidentified gram-negative R-P-12-L-11 antibacterial FUKAI
et al. (1995)
Fungi
Bassiatin Beauveria bassiana C-R-AA platelet aggregation KAGAMIZONO
et al. (1995)
inhibitor
Cathestatin Microascus longiirostris R-AA-A cysteine protease inhibitor Yu et al. (1996)
AM4299A Chromelosporium fulvum R-AA-M thiol protease inhibitor MORISHITA et al. (1994)
Lipoxamycin Aspergillus fumigatus R-AA palmitoyl-transferase MANDALA et al. (1994)
inhibitor
NK-374200 Talaromyces P-2 insecticidal
WIN 64821 Aspergillus sp. P-2-M POPPet al. (1994)
OPC-15161 Thielavia minor C-2-M superoxide anion KITAet al. (1994)
generation inhibitor
Gypsetin Nannizia gypsea C-2-M ACAT inhibitor SHINOHARA et al. (1994)
Tryprostatins Aspergillus fumigatus C-2-M cell cycle inhibitor CUIet al. (1996)
(TERP)
PEDT Aspergillus jlavus (C-2)~-M substance P inhibitor BARROW and SEDLOCK(1994)
Sch529OOll Gliocladium sp. (C-2)z-M oncogene inhibitor CHUet al. (1995)
Benzomalvins Penicillium sp. C-3-M substance P inhibitor SUNet al. (1994)
Ustiloxins Ustilaginoidea virens L-3-M(2,5) antimitotic, cancerostatic KOISOet al. (1994)
Citreoindole Penicillium citreo-viride P-3-C-2-M MATSUNAGA et al. (1991)
Fellutamide Penicillium fellutanum R-P-3-M nerve growth factor SHIGEMORI et al. (1991), TSUJIet al.
(marine) promoter (1994)
Stevastelins Penicillium sp. L-4(2) immunosuppressant MORINOet al. (1996)
Beauvericins Beauveria bassina C-D-4-M insecticidal GUPTAet al. (1995)
YF-044P-D Candida albicans R-P-5 aspartic proteinase SATOet al. (1994)
inhibitor
Plactin F-165 (Fungi Imperfecti) c-5 stimulates fibrinolytic INOUEet al. (1996)
activity
Leualacin Hapsidospora irregularis L-5-M-M(2) Ca*+-blocker HAMANOet al. (1992)
Malformin Aspergillus niger C-5-M phytotoxic KIMet al. (1993)
Pithomycolide Pithomyces chartarum C-D-5 MOUSSAand LE QUESNE(1996)
Aselacins Acremonium sp. R-P-5-L-4(3) endothelin receptor HOCHLOWSKI et al. (1994)
antagonist
Pneumocandins Zalerion arboricola R-C-6(3) antifungal HENSENSet al. (1992), MORRISet al.
(1994)
Deoxymulundocandin Aspergillus sydowii R-C-6(3) antifungal MUKHOPADHAY et al. (1992)
Destruxin A415 Aschersonia sp. L-6-M insecticidal KRASNOFFet al. (1996)
Desmethyldestruxin Metarhizium anisopliae L-6 suppresses hepatitis B CHENet al. (1995)
virus surface antigen
production
Bursaphelocides perfect fungus L-6-M nematicidal KAWAZUet al. (1993)
Enniatins D-F Fusarium sp, C-D-6 TOMODAet al. (1992)
Enniatin A2 Fusarium avenaceum C-D-6 KASTICet al. (1992)
WIN 66306 Aspergillus sp. C-7-M neurokinin antagonist BARROWet al. (1994)
(TERP)
PF1022 Mycelia sterilia C-D-8 antihelmintic SCHERKENBECK et al. (1995)
BZR-cotoxin IV Bipolyris zeicola L-8-M UEDA et al. (1995)
BZR-cotoxin 1/11 Bipolyris zeicola C-D-9-M phytotoxic? UEDA et al. (1992, 1994)
Leucinostatins, P-168 Paecilomyces sp. R-P-9-M antibiotic KUWATAet al. (1992), ISOGAIet al.
(1992)
Helioferins Mycogone rosea R-P-10-M antifungal GRAFEet al. (1995)
LF'237-F8 Tolypocladium geodes P-10-M antibiotic TSANTRIZOS et al. (1996)
Trichogin A Trichoderma longi- P-10-M AUVIN-GUETTEet al. (1992)
brachiatum
FR901459 Stachybotrys chartarum C-11-M immunosuppressive SAKAMOTO et al. (1993)
Petriellin A Petriella sordida L-13-M antifungal LEE et al. (1995)
Tab. 6. Continued
Type”
Hypelcins Hypocrea peltata R-P-19-M antibiotic MATSURAet al. (1993, 1994)
Trichosporin B Trichoderma polysporum R-P-19-M ion channel forming IIDAet al. (1993), NAGAOKAet al.
(1994)
Plants
p-Coumaroyl-tryptophan Coffea canephora R-AA MURATAet al. (1995)
Aurantiamide Piper aurantiacum R-P-2-M BANERJEEet al. (1993)
Clematis hepaulensis
Cyclopeptide alkaloids Zizyphus mucronata R-P3-M(E) BARBONIet al. (1994)
(roots)
Mucronine J Zizyphus mucronata R-P-CM(C) AUVINet al. (1996)
(root, bark)
Astin A Aster tataricus C-5(2) antitumor MORITAet al. (1993)
Astins D, E Aster tataricus c-5 MORITAet al. (1993)
Astin J Aster tataricus (roots) P-5 MORITAet al. (1995)
Asterinins A-C Aster tataricus (roots) P-5 CHENGet al. (1994)
Asterinins D, E Aster tataricus P-5-M CHENGet al. (1996)
Astericins D-F Aster tataricus (roots) P-5 CHENGet al. (1996)
Astins Aster tataricus C-S-(U) antitumor MORITAet al. (1996)
Astins Aster tataricus C-5(2) MORITAet al. (1994)
Pseudostellarins A-C Pseudostellaria hetero- c-5 tyrosinase inhibitor MORITAet al. (1994)
phylla
Mucronine J Zizyphus mucronata R-P-4-M AUVINet al. (1996)
(bark)
Stellarins B, C Stellaria yunnanensis C-6 ZHAOet al. (1995)
(roots)
Dichotomins Stellaria dichotoma C-6 MORITAet al. (1996)
Segetalin A/B-D Vaccaria segetalis C-6/C-5 MORITAet al. (1995)
Cyclogossine A Jatropha gossypifolia c-7 HORSTENet al. (1996)
(latex)
Stellarins F, G Stellaria yunnanensis C-8 ZHAOet al. (1995)
Pseudostellarin DIH Pseudostellaria hetero- C-718 MORITAet al. (1995)
phylla
Podacycline A Jatropha podagrica (latex) c-9 VAN DENBERG et a]. (1996)
Segetalin A Vaccaria segetalis (seeds) C-6 MORITAet al. (1994)
RA VII Rubia cordifolia C-6 eukaryotic ITOKAWAet al. (1993)
translation inhibitor
Stellarins D, E Stellaria yunnanensis c-7 ZHAOet al. (1995)
Segetalin E Vaccaria segetalis (seeds) c-7 MORITAet al. (1996)
Lyciumin A Lycium chinense P-8-C-5(CN) MORITAet al. (1996)
Cycloneoluripeptides Leonurus heterophylus c-9 MORITAet al. (1996)
Sponges
Geodiamolide G Cymbastela sp. L-5-M(7) cytotoxic COLEMANet al. (1995)
Cyclotheonamide Theonella sp. C-5(C0,3) senne protease inhibitor LEE et al. (1993)
Microsclerodermins Microscleroderma sp. C-6(2,3) antifungal BEWLEYet al. (1994)
Konbamide Theonella sp. P-6(U)-C-5 calmodulin antagonist SCHMIDTand WEINBRENNER(1996)
Orbiculamide Theonella sp. R-P-7-C-6- cytotoxic FUSETANIet al. (1991)
M(3)
Discobahamins Discodertnia sp. R-P-7-C-6- antifungal (Candida GUNASEKERA
et al. (1994)
M(3,4) albicans)
Pseudoaxinellin Pseudoaxinella massa c-7 KONGet al. (1992)
Aciculitins Acucilites orientalis C-7-M cytotoxic, antifungal BEWLEYet al. (1996)
Stylostatin Stylotella aurantium c-7 cytotoxic PETTITet al. (1992)
Stylopeptide I Stylotella sp.lPhakellia c-7 PETTITet al. (1995)
costata
Phakellistatin Phakellia costata and c-7 cancer cell growth PETTITet al. (1993,1995)
Stylotella aurantium inhibitor
Axinastatins Axinella spp. c-7 cytotoxic PETI-ITet al. (1993, 1994), KONATet
(1995)
Malaysiatin Pseudoaxinyssa sp.
Cyclodidemnamide Didemnum molle C-7-M
Tab. 6. Continued
Type"
Keramamides Theonella sp. R-P-8-C-6- superoxide generation KOBAYASHI
et al. (1991, 1995)
M(3,CO) response of human
neutrophils inhibitor
Callipeltins Callipelta sp. L-8-C-7-M D'AURIAet al. (1996)
Majusculamide C Ptilocaulis trachys L-9(3) WILLIAMSet al. (1993)
Callipeltin A Callipelta sp. R-P-9-L-6-M anti-HIV ZAMPELLA et al. (1996)
Didemnin B* Trididemnum cyano- R-P-9-L-7(3) ABOU-MANSOUR et al. (1995)
phorum
Keramamide F Theonella sp. R-P-9-C-7- ITAGAKI
et al. (1992)
M(CO)
Theonellamides Theonella sp. C-12(2,3) cytotoxic MATSUNAGA and FUSETANI(1995)
Theonegramide Theonella swinhoei C-l2(2,3)-M antifungal BEWLEYand FAULKNER (1994)
(GLYC)
Polydiscamide A Discoderma sp. R-P-13-L- GULAVITA
et al. (1992)
5-M
Theonellapeptolide IId Theonella swinhoei R-P-13-L- KOBAYASHI
et a1 (1994)
11(M)
Discodermins F-H Discodermia kiiensis R-P-14-L-6 antimicrobial RYUet al. (1994)
Halicylindramides Halichondria cylindrata R-P-14-L-6 antifungal, cytotoxic LI et al. (1995)
Polytheonamides Theonella swinhoei P-48 HAMADA et al. (1994)
Ascidians
Bistratamides C/D Lissoclinum bistratum C-6-M CNS depressant (D) FOSTERet al. (1992)
Patellins Lissoclinum sp. C-6-M(terp), CAROLLet al. (1996)
C-7IC8
Trunkamide A
Patellamides Lissoclinum patella C-6-M cytotoxic ISHIDAet al. (1995)
Discokilides Discoderma kiiensis L-7-M cytotoxic TADAet al. (1992)
Sea Hares
Aurilide Dolabella auricularia C-D-7(5) SUENAGA et al. (1996)
Dolastatin H Dolabella auricularia R-P-4-M cytotoxic SONEet al. (1996)
Dolastatins C, D Dolabella auricularia R-D-4 cytotoxic (D) SONEet al. (1993)
Molluscs
Keenamide Pleurobranchus forskalii C-6-M(terp) cytotoxic WESSONand HAMANN (1996)
Kahalide F Elysia rufescens and R-P-13-L-5 HAMANN and SCHEUER(1993)
Bryopsis sp.
SpiderslScorpions
Nephilatoxins Nephila clavata P2-A-AA neurotoxins MIYASHITA et al. (1992)
Nephilatoxin 7 Nephila clavata R-P-ZA neurotoxins MATSUSHITA et al. (1995)
Clavamine Nephila clavata P-2-A-P-3 insecticidal YOSHIOKA et al. (1992)
Spidamine Nephila clavata R-P-2-A toxin CHIBAet al. (1996)
~~
a Abbreviations used in the description of the structural types are: A amine, AA amino acid, P peptide, C cyclopeptide, L peptidolactone (cyc
structure closed with an ester bond), D depsipeptide (compound containing both peptide and ester bonds); the number associated with each str
tural type (P-n, C-n, L-n, D-n) gives the number of amino acids or hydroxy acids contained in the structure; branched cyclic compounds are given
the total length with the ring size included (P-10-C-5 is thus a branched decapeptide with a pentapeptide cyclic region); acyl compounds, amidati
or modifications by aminoalcohols are not included. Additional information is provided by R acylation by aliphatic or aromatic carboxylic acids,
modifications (largely N-methylations), (GLYC) glycosylation, (TERP) modification by terpenoids, [PYR] ring formation by a pyridine moi
formed by two modified serine residues (thioipeptides), U urea-type of linkage in the peptide chain, CO an additional CO-residue in the pepti
chain, E ether bond cyclization, S thioether link, CN direct C-N linkage for cyclization, (n) gives information on cyclization types differing from t
common a-carboxyl-a-amino-link; 2, 3, and 4 stand for bonds involving p, y or S positioned amino, carboxylic, or hydroxy groups.
yet unidentified cleavage and cyclization reac- BAYER,A., FREUND,S., JUNG,G. (1995), Post-
tions. The restricted number of sources indi- translational heterocyclic backbone modifica-
cates that this field investigations have almost tions in the 43-peptide antibiotic microcin B17,
begun. Eur. J. Biochem. 234,414-426.
Animals as sources have been restricted BENDER,C., PALMER,D., PENALOZAVAZQUEZ,
mainly to marine organisms. The large group A., RANGASWAMY, V., ULLRICH,M. (1996),
Biosynthesis of coronatine, a thermoregulated
of sponges gives rise to the problem of the phytotoxin produced by the phytopathogen
origin of compounds, since sponges are Pseudomonas syringae, Arch. Microbiol. 166,
known to contain large numbers of associated 71-75.
microbes. Recently, the production of a com- BERGEY,D. R., HOWE,G. A., RYAN,C. A. (1996),
plex sponge metabolite has been traced to an Polypeptide signaling for plant defensive genes
associated actinomycete (BEWLEY et al., exhibits analogies to defense signaling in ani-
1996). In the future of screenings more atten- mals, Proc. Natl. Acad. Sci. 93, 12053-12058.
tion will be paid to the interactions of organ- BERNHARD,F., DEMEL, G., SOLTANI,K., VON
isms. In addition, search for biosynthetic DOHREN,H., BLINOV,V. (1996), Identification
genes involved in peptide formation is a of genes encoding for peptide synthetases in the
promising pathway to detect, generate, and gram-negative bacterium Lysobacter sp. ATCC
produce new metabolites. 53042 and the fungus Cyclidrotrichum oligosper-
mum, DNA Sequence 6,319-330.
BEVINS,C. L., ZASLOFF,M. (1990), Peptide from
frog skin, Annu. Rev. Biochem. 59, 395-414.
6 References BEWLEY,C. A., HOLLAND, N. D., FAULKNER,
J. (1996), Two classes of metabolites from The-
D.
onelle swinhoei are localized in distinct popula-

ADEFARATI, A. A., GIACOBBE, R. A., HENSENS, tions of bacterial symbionts, Experientia 52,716-
0. D., TKACZ,J. S. (1991), Biosynthesis of L- 722.
671329, an echinocandin-type antibiotic pro-
duced by Zalerion arboricola: origins of some of BILLICH, A., ZOCHER,R. (1987), N-methyltrans-
the unusual amino acids and the dimethylmyris- ferase function of the multifunctional enzyme
tic acid side chain, J. Am. Chem. SOC. 113,3542- enniatin synthetase, Biochemistry 26, 8417-
3545. 8423.
AHARONOWITZ, Y., COHEN,G., MART~N, J. F. BILLICH,A., ZOCHER,R. (1990), Biosynthesis of
(1992), Penicillin and cephalosporin biosynthetic N-methylated peptides, in: Biochemistry of Pep-
genes: structure, organization, regulation, and tide Antibiotics (KLEINKAUF, H., VON DOHREN,
evolution, Annu. Rev. Microbiol. 46, 461-495. H., Eds.), pp. 57-80. Berlin: de Gruyter.
APARICIO,J. F., MOLNAR,I., SCHWECKE, T., KO- BLANC,V., GIL,P., BAMAS-JACQUES, N., LOREN-
NIG,A., HAYDOCK, S. F., KHAW,L. E., STAUN- ZON, S., ZAGOREC,M., SCHLEUNIGER, J.,
TON,J., LEADLAY, P. F. (1996), Organisation of BISCH, D., BLANCHE,F., DEBUSSCHE,L.,
the biosynthetic gene cluster for rapamycin in CROUZET,J., THIBAUT,D. (1997), Identifica-
Streptomyces hygroscopicus - analysis of the en- tion and analysis of genes from Streptomyces
zymatic domains in the modular polyketide synth- pristinaspiralis encoding enzymes involved in the
ase, Gene 169, 9-16. biosynthesis of the 4-dimethylamino-~-phenylala-
AWAZU,N., IKAI,K., YUMAMOTO, J., NISHIMURA, nine precursor of pristinamycin I, Mol. Micro-
K., MIZUTANI,S., TAKESAKO,K., KATO, I. biol. 23, 191-202.
(1995), Structures and antifungal activities of BOMAN,H. G. (1995), Peptide antibiotics and their
new aureobasidins, J. Antibiot. 48,525-527. role in innate immunity, Annu. Rev. Immunol.
BAILEY,A. M., KERSHAW, M. J., HUNT, B. A., PA- 13,62-92.
TERSON, I. C., CHARNLEY, A. K., REYNOLDS, S.
E., CLARKSON, J. M. (1996), Cloning and se- BOMAN,H. G., MARSH,J., GOODE,J. A. (1994),
quence analysis of an intron-containing domain Antimicrobial Peptides. Chichester, New York:
from a peptide synthetase of the entomopatho- John Wiley & Sons.
genic fungus Metarhizium anisopliae, Gene 113, BORCHERT,S., PATIL, S. S., MARAHIEL,M. A.
195-197. (1992), Identification of putative multifunctional
BALTZ,R. H. (1996), Daptomycin and A54145, in: peptide synthetase genes using highly conserved
Biotechnology of Industrial Antibiotics, 2nd Edn. oligonucleotide sequences derived from known
(STROHL,W., Ed.). New York Marcel Dekker. synthetases, FEMS Microbiol. Lett. 92, 175-180.
6 References 317
BORMANN,C., MOHRLE, V., BRUNTNER,C. parameters in fermentation process, Biotechnol.
(1996), Cloning and heterologous expression of Progr. 9,655-660.
the entire set of structural genes for nikkomycin DE CRECY-LAGARD, v., MARLI~RE, P., SAURIN
synthesis from Streptomyces tendae Tu901 in W. (1995), Multienzymatic nonribosomal pep-
Streptomyces lividans, J. Bacteriol. 178, 1216- tide biosynthesis: identification of the functional
1218. domains catalysing peptide elongation and epi-
BREIL,B. T., BORNEMAN, J., TRIPLETT,E. W. merization, C. R. Acad. Sci. Paris 318, 927-936.
(1996), A newly discovered gene, TfuA, in- DIEz, B., GUTI~RREZ, S., BARREDO,J. L., VAN
volved in the production of the ribosomally syn- SOLINGEN,P., VAN DER VOORT, L. H. M.,
thesized peptide antibiotic trifolitoxin, J. Bacteri- MARTfN, J. F. (1990), The cluster of penicillin
01. 178,4150456. biosynthetic genes. Identification and characteri-
BUADES,C., MOYA,A. (1996), Phylogenetic analy- zation of the pcbAB gene encoding a-aminoadi-
sis of isopenicillin N synthetase horizontal gene pyl-cysteinyl-valine synthetase and linkage to
transfer, J. Mol. Evol. 42, 537-542. the pcbC and penDE genes, J. Biol. Chem. 265,
BURMESTER, J., HAESE,A., ZOCHER,R. (1995), 16358-16 365.
Highly conserved N-methyltransferases as inte- DONADIO,S., STAVER,M. J., MCALPINE,J. B.,
gral part of peptide synthetases, Biochem. Mol. SWANSON, J. B., KATZ,L. (1991), Modular or-
Biol. Znt. 37, 201-207. ganization of genes required for complex poly-
CHEN,Q., WERTHEIMER, A. M., TOLMASKY, M. ketide biosynthesis, Science 252, 675-679.
E., CROSA,J. H. (1996), The AngR protein and D’SOUZA,C., NAKANO, M. M., ZUBER,P. (1994),
the siderophore anguibactin positively regulate Identification of comS, a gene of the srfA ope-
the expression of iron-transport genes in Vibrio ron that regulates the establishment of genetic
anguillarum, Mol. Microbiol, 22, 127-134. competence in Bacillus subtilis, Proc. Natl. Acad.
CHINERY,R., COFFEY,R. J. (1996), Trefoil pep- Sci. USA 91,9397-9401.
tides: less clandestine in the intestine, Science ERSPAMER, V., MELCHIORRI, P., FALCONIERI-ER-
274,204. SPAMER, G., NEGRI,L., CORSI, R., SEVERINI,
CHONG,P. P., PODMORE,S., KIESER, H., REDEN- C., BARRA,D., SIMMACO, M., KREIL,G. (1989),
BACH, M., TURGAY,K., MARAHIEL,M. A., Deltorphins: a family of naturally occurring pep-
HOPWOOD, D., SMITH,C. (1996), Physical iden- tides with high affinity and selectivity for 6
tification of a chromosomal locus encoding bio- opioid binding sites, Proc. Natl. Acad. Sci. USA
synthetic genes for the calcium-dependent anti- 86,5188-5192.
biotic of Streptomyces coelicolor A3(2). Ab- FAULKNER, D. J. (1996), Marine natural products,
stracts Symp. Enzymol. Biosynth. Nat. Prod., Nat. Prod. Rep. 13, 75-125.
Technical University, Berlin. , GUITTON,J. D., HERMAN,F.,
F R ~ C H E TD.,
CLARK,D. P., DURELL,s., MALOY,W. L., ZAS- FAUCHER,D., HELYNCK,G., MONEGIERDE
LOFF,M. (1994), Ranalexin, J. Biol. Chem. 269, SORBIER,B., RIDOUX,J. P., JAMES-SURCOUF,
10849-10 855. E., VUILHORGNE, M. (1994), Solution structure
COQUE,J. J. R., ENGUITA, F. J., MART~N, J. F., LI- of RP-71955, a new 21 amino acid tricyclic pep-
RAS,P. (1995a), A two protein component 7-ceph- tide active against HIV-1 virus, Biochemistry 33,
em-methoxylase encoded by two genes of the 42-50.
cephamycin C cluster converts cephalosporin to FUJITA, K., MINAKATA, H., NOMOTO,K., FURU-
7-methoxycephalosporin C, J. Bacteriol. 177, KAWA, Y., KOBAYASHIE, M. (1995), Structure-
2230-2235. activity relations of fulicin, a peptide containing
COQUE,J. J. R.,PBREZ-LLARENA, F. J., ENGUITA, a D-amino acid residue, Peptides 16, 565-568.
F. J., FUENTE,J. L., MART~N, J. F., LIRAS,P. GEKELER,W., GRILL, E., WINNACKER, E.-L.,
(1995b), Characterization of the cmcH genes of ZENK,M. H. (1989), Survey of the plant king-
Nocardia lactamdurans encoding a functional 3 I - dom for the ability to bind heavy metals through
hydroxymethylcephem 0-carbamoyltransferase phytochelatin, Z. Naturforsch. 44q 361-369.
for cephamycin biosynthesis, Gene 162,21-27. GRILL, E., LOEFFLER,E., WINNACKER, E.-L.,
COSMINA,P., RODRIGUEZ,F., DE FERRA, F., ZENK,M. H. (1989), Phytochelatins, a class of
GRANDI,G., PEREGO,M., VENEMA,G., VAN heavy-metal binding peptides from plants, are
SINDEREN, D. (1993), Sequence and analysis of synthesized from glutathione by a specific y-glut-
the genetic locus responsible for surfactin syn- amylcysteine dipeptididyl transpeptidase (phyto-
thesis in Bacillus subtilis, Mol. Microbiol. 8, 821- chelatin synthase), Proc. Natl. Acad. Sci. USA
831. 86,6838-6842.
DAHOD,S. K. (1993), Dissolved carbon dioxide GUILVOUT, I., MERCEREAU-PUIJALON, O., BON-
measurement and its correlation with operating NEFOY, S., PUGSLEY, A. P., CARNIEL, E. (1993),
High-molecular-weight protein 2 of Yersinia en- from Bacillus subtilis Rb14, J. Ferment. Bioeng.
terocolitica is homologous to AngR of Vibrio an- 76,445450.
guillarum and belongs to a family of proteins in- IHARA,M., FUKUMOTO, K. (1996), Recent pro-
volved in nonribosomal peptide synthesis, J. gress in the chemistry of non-terpenoid indole
Bacteriol. 175, 5488-5504. alkaloids, Nat. Prod. Rep. 13, 241-261.
HAESE, A., SCHUBERT,M., HERRMANN, M., IKAI,K., TAKESAKO, K., SHIOMI, K., MORIGUCHI,
ZOCHER,R. (1993), Molecular characterization M., UMEDA,Y., YAMAMOTO, J., KATO,I., NA-
of the enniatin synthetase gene encoding a mul- GANAWA, H. (1991a), Structure of aureobasidin
tifunctional enzyme catalysing N-methyl-depsi- A, J. Antibiot. 44,925-933.
peptide formation in Fusarium scirpi, Mol. Mi- IKAI,K., SHIOMI,K.,TAKESAKO, K., MIZUTANI,
crobiol. 7, 905-914. S., YAMAMOTO,J., OGAWA,Y., UENO, M.,
HAM,P. J., CHALK,R., SMITHIES, B., ALBUQUER- KATO,I. (1991b), Structures of aureobasidins B
QUE,c., HAGEN,H. (1994), Antibacterial pep- to R, J. Antibiot. 44,1187-1198.
tides in insect vectors of tropical parasitic dis-
JACK,R., TAGG,J. R., RAY,B. (1995), Bacterio-
eases, in: Antimicrobial Peptides (J. GOODE,
cins of gram-positive bacteria, Microbiol. Rev.
Ed.), pp. 140-150, Ciba Foundation Series. New
York: John Wiley & Sons. 59,171-200.
HAMOEN, L. W., ESHUIS,H., JONGBLOED, J., VEN- JORGENSEN,H., NIELSEN, J., VILLADSEN,J.,
EMA, G., VAN SINDEREN, D. (1995), A small MOLLGAARD, H. (1995a), Metabolic flux distri-
gene, designated comS, located within the cod- butions in Penicillium chrysogenum during fed-
ing region of the fourth amino acid-activation batch cultivations, Biotechnol. Bioeng. 46, 117-
domain of srfA, is required for competence de- 131.
velopment in Bacillus subtilis, Mol. Microbiol. JORGENSEN,H., NIELSEN, J., VILLADSEN,J.,
15, 55-63. MOLLGAARD, H. (1995b), Analysis of penicillin
HANCOCK,R. E. W., FALLA,T., BROWN,M. V biosynthesis during fed-batch cultivations with
(1995), Cationic bactericidal peptides, Adv. Mi- a high-yielding strain of Penicillium chrysoge-
crob. Physiol. 37, 135-175. num, Appl. Microbiol. Biotechnol. 43, 123-130.
HANSEN,J. N. (1993), Antibiotics synthesized by KALETTA, C., ENTIAN, K.-D., KELLNER, R., JUNG,
posttranslational modification, Annu. Rev. Mi- G., REIS,M., SAHL,H.-G. (1989), Pep-5, a new
crobiol. 47, 535-564. lantibiotic: structural gene isolation and prepep-
tide sequence, Arch. Microbiol. 152, 16-19.
HECK,S. D., FARACI,W. S., KELBAUGH, P. R.,
SACCOMANO, N. A., THADEIO,P. F., VOLK- KALLOW, W., VON DOHREN,H., KENNEDY,J.,
MA", R. A. (1995), Posttranslational amino TURNER,G. (1996), Abstract, 3rd Eur. Conf.
acid epimerization - enzyme catalyzed isomer- Fungal Genet., Monster, p. 159.
ization of amino acid residues in peptide chains, KELLENBERGER,C., HIETTER,H., Luu, B. (1995),
Proc. Natl. Acad. Sci. USA 93,4036-4039. Regioselective formation of the three disulfide
HERZOG-VELIKONJA, B., PODLESEK,Z., GRAB- bonds of a 35-residue insect peptide, Peptide
NAR, M. (1994), Isolation and characterization Res. 8, 321-327.
of Tn917-generated bacitracin deficient mutants KIMURA,H., MIYASHITA, H., SUMINO,Y. (1996),
of Bacillus licheniformis, FEMS Microbiol. Lett. Organization and expression in Pseudomonas
l21, 147-152. putida of the gene cluster involved in cephalo-
HILTON,M. D., ALAEDDINOGLU, N. G., DEMAIN, sporin biosynthesis from Lysobacter lactamgenus
A. L. (1988), Bacillus subtilis mutant deficient in YK90, Appl. Microbiol. Biotechnol. 45, 490-
the ability to produce the dipeptide antibiotic 501.
bacilysin: isolation and mapping of the mutation, KLAPHECK,S., FLIEGNER,W., ZIMMER, I. (1994),
J. Bacteriol. 170, 1018-1021. Hydroxymethyl-phytochelatins (y-glutamylcys-
HODGSON,J. E., FOSBERRY, A. P., RAWLINSON, teine),-serine) are metal-induced peptides of the
N. S., Ross, H. N. M., NEAL,R. J., ARNELL,J. Pocaceae, Plant Physiol. 104, 1325-1332.
C., EARL,A. J., LAWLOR,E. J. (1995), Clavu- KLEINKAUF, H., VON DOHREN,H. (1982), A sur-
lank acid biosynthesis in Streptomyces clavulige- vey of enzymatic biosynthesis of peptide antibiot-
rus: gene cloning and characterization, Gene 166, ics, in: Trends in Antibiotic Research, pp. 220-
49-55. 232. Tokyo: Japan Antibiotics Res. Ass.
HUANG,C. C., ANO, T., SHODA,M. (1993), Nu- KLEINKAUF, H., VON DOHREN,H. (1986), Peptide
cleotide sequence and characteristics of the antibiotics, in: Biotechnology 1st Edn., Vol. 4
gene, Lpa-14, responsible for biosynthesis of the (REHM, H.-J., REED, G., Eds.), pp. 283-307.
lipopeptide antibiotics iturin a and surfactin Weinheim: VCH.
6 References 319
KLEINKAUF,H., VON DOHREN,H. (1996), A non- binds covalently to cysteine-273 on protein phos-
ribosomal system of peptide biosynthesis, Eur. J. phatase 1, FEBS Lett. 371, 236-240.
Biochem. 236, 335-351. MAGNUSON, R., SOLOMON, J., GROSSMAN, A. D.
KLEINKAUF, H., VON DOHREN,H. (1997), Enzy- (1994), Biochemical and genetic characterization
matic generation of complex peptides, Progr. of a competence pheromone from B. subtilis,
Drug Res. 48, 27-51. Cell 77, 207-216.
KONDO,N., IMAI,K., ISOBE,M., GOTO, T., Mu- MARTfN, J. F., GUTIERREZ, s. (1995), Genes for
RASUGI, A., WADA-NAKAGAWA, C., HAYASHI, beta-lactam antibiotic biosynthesis, Antonie van
Y. (1984), Cadystin A and B, major unit pep- Leeuwenhoek Int. J. Gen. Mol. Microbiol. 67,
tides comprising cadmium binding peptides in- 181-200.
duced in a fission yeast-separation, revision of MEI, B., BUDDE,A. D., LEONG,S. A. (1993), sidl,
structure and synthesis, Tetrahedron Lett. 25, a gene intiating siderophore biosynthesis in Usti-
3869-3872. lago maydis: molecular characterization, regula-
KUBOI, R., MARUKI,T., TANAKA,H., KOMASA- tion by iron, and role in phytopathogenicity,
WA, I. (1994), Fermentation of Bacillus subtilis Proc. Natl. Acad. Sci. USA 90,903-907.
ATCC 6633 and production of subtilin in poly- MEISSNER, N ,BORNER,T. (1996),
K., D I ~ M A NE.,
ethylene glycollphosphate aqueous two-phase Toxic and nontoxic strains of the cyanobacter-
systems, J. Ferment. Bioeng. 78,431-436. ium Microcystis aeruginosa contain sequences
LAMBALOT, R. H., WALSH,C. T. (1995), Cloning, homologous to peptide synthetase genes, FEMS
overproduction, and characterization of the Microbiol. Lett. 135, 295-303.
Echerichia coli holo-acyl carrier protein synth- MEUWLY,P., THIBAULT,P., SCHWAN,A. L.,
ase, J. Biol. Chem. 270,24658-24661. RAUSER,W. E. (1995), Three families of thiol
LAMBALOT, R. H., GEHRING, A. M., FLUGEL,R. peptides are induced by cadmium in maize,
S., ZUBER,P., LACELLE,M., MARAHIEL, M. A., Plant J. 7,391-400.
REID,R., KHOSLA,C., WALSH,C. T. (1996), A MITTENHUBER, G., WECKERMANN, R., MARA-
new enzyme superfamily - the phosphopante- HIEL,M. A. (1989), Gene cluster encoding the
theinyl-transferases, Chemistry & Biology 3, genes for tyrocidine synthetase 1 and 2 from Ba-
923-936. cillus brevis. Evidence for an operon, J. Bacteri-
LAWEN,A., TRABER,R. (1993), Substrate specifi- 01. 171,4881-4887.
cities of cyclosporin synthetase and peptolide Mo, Y. Y., GEIBEL,M., BONSALL,R. F., GROSS,
SDZ 214-103 synthetase, J. Biol. Chem. 268, D. C. (1995), Analysis of sweet cherry (Prunus
20452-20465. avium L.) leaves for plant signal molecules that
LEHRER,R. I., LICHTENSTEIN, A. K., GANZ,T. activate the syrB gene required for synthesis of
(1993), Defensins: antimicrobial and cytotoxic the phytotoxin, syringomycin, by Pseudomonas
peptides of mammalian cells, Annu. Rev. Immu- syringae pv. syringae, Plant Physiol. 107, 603-
nol. 11, 105-128. 612.
LEWIS,J. R. (1996), Muscarine, imidazole, oxazole, MOLNAR,I., APARICIO,J. F., HAYDOCK,S. F.,
thiazole and peptide alkaloids, and other miscel- KHAW, L. E., SCHWECKE,T., KONIG, A.,
laneous alkaloids, Nat. Prod. Rep. 13, 435-467. STAUNTON, J., LEADLAY,P. F. (1996), Organi-
LI, Y. M., MILENE, J. C., MADISON, L. L., KOLTER, sation of the biosynthetic gene cluster for rapa-
R., WALSH,C. T. (1996), From peptide precur- mycin in Streptomyces hygroscopicus - analysis
sors to oxazole and thiazole-containing peptide of genes flanking the polyketide synthase, Gene
antibiotics - microcin B17 synthase, Science 274, 169, 1-7.
1188-1193. MOR,A., DELFOUR,A., NICOLAS,P. (1991), Iden-
LIU, G.-H., CHEN,I.-L., HSIEH,J.-C., LIU, S.-T. tification of a D-alanine-containing polypeptide
(1996), Cloning and sequencing of the peptide precursor for the peptide opioid, dermorphin, J.
synthetase genes encoding the biosynthesis of Biol. Chem. 266,6264-6270.
fengycin, Abstracts, Symp. Enzymol. Biosynth. MORIKAWA, N., HAGIWARA, K., HAKAJIMA, T.
Nat. Prod., Technical University, Berlin. (1992), Brevinin-1 and -2, unique antimicrobial
MACCABE,A. P., RIACH,M. B. R., UNKLES,S. E., peptides from the skin of the frog, Rana brevipo-
KINGHORN, J. R. (1990), The Aspergillus nidu- da porsa, Biochem. Biophys. Res. Commun. 189,
lans npeA locus consists of three contiguous 184-190.
genes required for penicillin biosynthesis, MOTAMEDI,H. (1996), FK506 biosynthetic gene
EMBO J. 9,279-287. cluster. Abstracts, Symp. Enzymol. Biosynth.
MACKINTOSH, R. W., DALBY,K. N., CAMPBELL, Nat. Prod., Technical University, Berlin.
D. G., COHEN,P. T., COHEN,P., MACKINTOSH, NIELSEN,J., JORGENSEN, H. S. (1995), Metabolic
C. (1995), The cyanobacterial toxin microcystin control analysis of the penicillin biosynthetic
pathway in a high-yielding strain of Penicillium PIECQ,M., DEHOTTAY,P., BIOT, A., DUSART,J.

chrysogenum, Biotechnol. Progr. 11,29%305. (1994), Cloning and nucleotide sequence of a re-
NIELSEN,J., KRABBEN,P. (1995), Hyphal growth gion of the Kibdelosporangium aridum genome
and fragmentation of Penicillium chrysogenum homologous to polyketide biosynthetic genes, J.
in submerged cultures, Biotechnol. Bioeng. 46, DNA Sequ. Mapping 4,219-229.
588-598. PIEPER,R., KLEINKAUF, H., ZOCHER,R. (1992),
NIELSEN,J., JOHANSEN, C. L., VILLADSEN,J. Enniatin synthetases from different Fusaria ex-
(1994), Culture fluorescence measurements dur- hibiting distinct amino acid specificities, J. Anti-
ing batch and fed-batch cultivations with Penicil- biot. 45, 1273-1277.
lium chrysogenum, J. Biotechnol. 38, 51-62. PIERS,K. L., BROWN,M. H., HANCOCK,R. E. W.
NIELSEN,J., JOHANSEN,C. L., JACOBSEN,C. L., (1993), Recombinant DNA procedures for pro-
KRABBEN,P., VILLADSEN, J. (1999, Pellet for- ducing small antimicrobial cationic peptides in
mation and fragmentation in submerged cultures bacteria, Gene 134,7-13.
of Penicillium chrysogenum and its relation to PITKIN,J. W., PANACCIONE, D. G., WALTON,J. D.
penicillin production, Biotechnol. Progr. 11, 93- (1996), A putative cyclic peptide efflux pump en-
98. coded by the toxA gene of the plant pathogenic
NOZAKI,S., MURAMATSU, I. (1987), Natural ho- fungus Cochliobolus carbonum, Microbiology
mologs of gramicidin S, J. Antibiot. 37, 689-90. 142,1557-1565.
OFFENZELLER, M., SANTER,G., TOTSCHNIG,K., POSPIECH,A., BIETENHADER, J., SCHUPP, T.
Su, Z., MOSER,H., TRABER,R., SCHNEIDER- (1996), Two multifunctional peptide synthetases
SCHERZER,E. (1996), Biosynthesis of the un- and an 0-methyltransferase are involved in the
usual amino acid (4R)-4-[(E)-2-buteny1]-4-me- biosynthesis of the DNA-binding antibiotic and
thyl-L-threonine of cyclosporin A: enzymatic antitumour agent saframycin Mxl from Myxo-
analysis of the reaction sequence including iden- coccus xanthus, Microbiology 142, 741-746.
tification of the methylation precursor in a poly- PO~TERAT, O., STEPHAN,H., METZGER,J. W.,
ketide pathway, Biochemistry 35, 8401-8412. GNAU,V., ZAHNER,H., JUNG,G. (1994), Abo-
OHNO,A., ANO,T., SHODA,M. (1995), Production rycin-a tricyclic-21-peptide antibiotic isolated
of a lipopeptide antibiotic, surfactin, by recombi- from Streptomyces griseoflavus, Liebigs Ann.
nant Bacillus subtilis in solid state fermentation, Chem. 1994,741-743.
Biotechnol. Bioeng. 47,209-214. RAINEY,P. B., BRODEY,C. L., JOHNSTONE,K.
OHNO,A., ANO,T., SHODA,M. (1996), Use of soy- (1993), Identification of a gene cluster encoding
bean curd residue, okara, for the solid state sub- three high-molecular weight proteins, which is
strate in the production of a lipopeptide anti- required for synthesis of tolaasin by the mu-
biotic, iturin a, by Bacillus subtilis NB22, Process shroom pathogen Pseudomonas tolaasii, Mol.
Biochem. 31,801-806. Microbiol. 0, 643-652.
OKUMURA, Y. (1990), Bioysynthesis of viridogri- RAUSER,W. E. (1990), Phytochelatins, Annu. Rev.
sein, in: Biochemistry of Peptide Antibiotics Biochem. 59,61-86.
(KLEINKAUF, H., VON DOHREN,H., Eds.), pp. REICHERT,J., SAKAITANI,M., WALSH, C. T.
365-378. Berlin: de Gruyter. (1992), Characterization of EntF as a serine-acti-
PARK,Y. S., INOUE,K., YAHIRO,K., OKABE,M. vating enzyme, Protein Sci. 1, 549-555.
(1994a), Improvement of cephamycin C produc- REICHHART,J.-M., PETIT, I., LEGRAIN,M., DI-
tion by a mutant resistant to linoleic acid, J. Fer- MARCQ, J. L., KEPPI,E., LECOCQ,J. P., HOW-
ment. Bioeng. 78, 88-92. MA", J. A., ACHSTETTER, T. (1992), Expres-
PARK,Y. S., MOMOSE,I., TSUNODA,K., OKABE, sion and secretion in yeast of active insect defen-
M. (1994b), Enhancement of cephamycin C pro- sin, an inducible antibacterial peptide from the
duction using soybean oil as the sole carbon fleshfly Formia terranova, Invert. Reprod. Dev.
source, Appl. Microbiol. Biotechnol. 40, 773- 21, 15-24.
779. RICHTER,K., EGGER,R., KREIL,G. (1987), D-Ala-
PAUL,G. C., THOMAS, C. R. (1996), A structured nine in the frog skin dermorphin is derived from
model for hyphal differentiation and penicillin L-alanine in the precursor, Science 238, 200-
production using Penicillium chrysogenum, Bio- 202.
technol. Bioeng. 51, 558-572. RINCE,A., DUFOUR,A., LEPOGAM,S., THUAULT,
PFEIFER,E., PAVELA-VRANCIC, M., VON DOH- D., BOURGEOIS, C. M., LE PENNEC,J. P. (1994),
REN,H., KLEINKAUF, H. (1999, Characteriza- Cloning, expression and nucleotide sequence of
tion of tyrocidine synthetase 1 (TY1): Require- genes involved in production of lactococcin DR,
ment of post-translational modification for pep- a bacteriocin from Lactococcus lactis, Appl. En-
tide biosynthesis, Biochemistry 34, 7450-7459. viron. Microbiol. 60, 1652-1657.
6 References 321
ROLLEMA,H. S., METZGER,J. W., BOTH, P., of L-serine to D-alanine in a ribosomally synthe-
KUIPERS,0. P., SIEZEN,R. J. (1996), Structure sized poylpeptide, J. Biol. Chem. 269, 27183-
and biological activity of chemically modified ni- 27185.
sin species, Eur. J. Biochem. 241, 716-722. SMITH,D. J., BURNHAM, M. K. R., BULL,J. H.,
ROMEO, D., SKERLAVAJ,B., BOLOGNESI,M., HODGSON,J. E., WARD,J. M., BROWNE,P.,
GENNARO, R. (1988), Structure and bactericidal BROWN,J., BARTON,B., EARL,A. J., TURNER,
activity of an antibiotic dodecapeptide purified G. (1990), /?-Lactam antibiotic biosynthetic
from bovine neutrophils, J. Biol. Chem. 263, genes have been conserved in clusters in proka-
9573-9575. ryotes and eukaryotes, EMBO J. 9,741-747.
SAHL,H.-G., JACK,R. W., BIERBAUM, G. (1995), SOLOMON,J. M., MAGNUSON,R., SRIVASTAVA,
Lantibiotics - biosynthesis and biological activi- A., GROSSMAN, A. D. (1995), Convergent sens-
ties of peptides with unique posttranslational ing pathways mediate response to two extracel-
modifications, Eur. J. Biochem. 230, 827-853. Mar competence factors in Bacillus subtilis,
SAKAJOH,M., SOLOMON,N. A., DEMAIN,A. L. Genes Dev. 9,547-558.
(1987), Cell-free synthesis of the dipeptide anti- STACHELHAUS, T., SCHNEIDER, A., MARAHIEL,
biotic bacilysin, J. Ind. Microbiol. 2, 201-208. M. A. (1995), Rational design of peptide antibiot-
SCHAUWECKER, F., PFENNIG,F., KELLER, U. ics by targeted replacement of bacterial and fun-
(1996), Molecular cloning of the actinomycin gal domains, Science 269, 69-72.
synthetase genes acmA and acmB from Strepto- STACHELHAUS, T., HUSER,A., MARAHIEL, M. A.
myces chrysomallus, Abstracts, Symp. Enzymol. (1996a), Biochemical characterization of pep-
Biosynth. Nat. Prod., Technical University, Ber- tides carrier protein (PCP), the thiolation do-
lin. main of multifunctional peptide synthetases,
SCHNELL,N., ENGELKE,G., AUGUSTIN,J., Ro- Chemistry & Biology 3, 913-921.
SENSTEIN, R., UNGERMANN, V., GOTZ, F., EN- STACHELHAUS, T., SCHNEIDER,A., MARAHIEL,
TIAN, K.-D. (1992), Analysis of genes involved M. A. (1996b), Engineered biosynthesis of pep-
in the biosynthesis of the lantibiotic epidermin, tide antibiotics, Biochem. Pharmacol. 52, 177-
Eur. J. Biochem. 204,5743. 186.
SCHWARTZ,D., ALIJAH,R., NUSSBAUMER, B., STEFFENS,J. C. (1990), The heavy metal-binding
PELZER,S., WOHLLEBEN, W. (1996), The pep- peptides of plants, Annu. Rev. Plant Physiol.
tide synthetase gene phsA from Streptomyces Plant Mol. Biol. 41,533-575.
virdochromogenes is not juxtaposed with other STEIN,T., VATER,J. (1996), Amino acid activation
genes involved in nonribosomal biosynthesis of and polymerization at modular multienzymes in
peptides, Appl. Environ. Microbiol. 62, 570- nonribosomal peptide biosynthesis, Amino
577. Acids 10, 201-227.
SCHWECKE, T., APARICIO, J. F., MOLNAR,I., KO- STINTZI, A., CORNELIS, P., HOHNADEL, D.,
NIG,A., KHAW,L. E., HAYDOCK, S. F., OLIY- MEYER,J.-M., DEAN,C., POOLE,K., KOURAM-
NYK,M., CAFFREY,P., CORTES,J., LESTER,J. BAS, S., KRISHNAPILLAI, V. (1996), Novel pyo-
B., BOHM,G. A., STAUNTON,J., LEADLAY,P. verdine biosynthesis gene(s) of Pseudomonas ae-
F. (1995), The biosynthetic gene cluster for the ruginosa PAO, Microbiology 142, 1181-1190.
polyketide immunosuppressant rapamycin, Proc. STROHL,W. R. (Ed.) (1996), Biotechnology of In-
Natl. Acad. Sci. USA 92, 7839-7843. dustrial Antibiotics, 2nd Edn. New York: Marcel
SCOTT-CRAIG,J. S., PANACCIONE, D. G., PA- Dekker.
CARD,J. A., WALTON,J. D. (1992), The cyclic STROHL,W. R., FLOSS,H. G. (1995), Thiopeptides,
peptide synthetase catalysing HC toxin produc- in: Genetics and Biochemistry of Antibiotic Pro-
tion in the filamentous fungus Cochliobolus car- duction (VINING,L. C., STUITARD,C., Eds.),
bonum is encoded by a 15.7 kilobase open read- pp. 223-238. Boston, MA: Butterworth-Heine-
ing frame, J. Biol. Chem. 267,26044-26049. mann.
SELSTED,M. E., OUTLETTE,A. J. (1995), Defen- SUBBALAKSHMI, C., KRISHNAKUMARI, V., NAGA-
sins in granules of phagocytic and non-phagocyt- RAJ, R., SITARAM, N. (1996), Requirements for
ic cells, Trends Cell. Biol. 5, 114-119. antibacterial and hemolytic activities in the bo-
SIMMACO, M., MIGNOGNA,G., BARRA,D., Bos- vine neutrophil derived peptide indolicidin,
SA,F. (1993), Novel antimicrobial peptides from FEBS Lett. 395, 48-52.
skin secretion of the European frog Rana escu- TAKESAKO,K., MIZUTANI,S., SAKAKIBARA, H.,
lenta, FEBS Lett. 324, 159-161. ENDO,M., YOSHIKAWA, Y., MSUDA,T., SONO-
SKAUGEN, M., NISSEN-MEYER, J., JUNG,G., STE- KOYAMA,E., KATO, I. (1996), Precursor di-
VANOVIC, S., SLEITEN, K., INGER, C., ABILD- rected biosynthesis of aureobasidins, J. Antibiot.
GAARD,M., NES,I. F. (1994), In vivo conversion 49,676-681.
TANIGUCHI, M., HOSHINO,K., URASAKI,H., Fu- tion of Penicillium chrysogenum physiology in

JII,M. (1993), Continuous production of an anti- submerged cultures by color and monochrome
biotic polypeptide (nisin) by Lactococcus lactis image analysis, Biotechnol. Bioeng. 48,l-11.
using a bioreactor coupled to a microfiltration VINING,L. C., STUITARD,C. (Eds.) (1995), Ge-
module, J. Ferment. Bioeng. 78, U122. netics and Biochemistry of Antibiotic Production.
THIBAULT,P., FAUBERT,D., KARUNANITHY, S., Boston, M A Butterworth-Heinemann.
BOYD, R. K., HOLMES,C. F. B. (1992), Isola- VON DOHREN,H. (1990), Compilation of peptide
tion, mass spectrometric characterization, and structures - a biogenetic approach, in: Biochem-
protein phosphatase inhibition properties of cyc- istry of Peptide Antibiotics (KLEINKAUF, H.,
lic peptide analogs of grarnicidin S from Bacillus VON DOHREN,H., Eds.), pp. 411-507. Berlin,
brevis (Nagano strain), Biol. Mass Spectrom. 21, New York: de Gruyter.
367-379. WEBER,G., SCHORGENDORFER, K., SCHNEIDER-
THIERICKE, R., ROHR,R. (1993), Biological varia- SCHERZER, E., LEITNER,E. (1994), The peptide
tion of microbial metabolites by precursor-di- synthetase catalyzing cyclosporine production in
rected biosynthesis, Nut. Prod. Rep. 10, 263- Tolypocladium niveum is encoded by a giant
289. 45.8 kilobase open reading frame, Curr. Genet.
TOGNONI, A., FRANCHI, E., MAGISTRELLI, C., Co- 26,120-125.
LOMBO, E., COSMINA, p., GRANDI,G. (1995), A WILLIAMSON, M. P., GAUVREAU, D., WILLIAMS,
putative new peptide synthase operon in Bacil- D. H., WARING,M. J. (1982), Structure and con-
lus subtilis: Partial characterization, Microbiolo- formation of fourteen antibiotics of the quinox-
gy 141,645-648. aline group determined by 'H-NMR, J. Antibiot.
TURGAY,K., MARAHIEL,M. A. (1995), The gtcRS 35, 62-66.
operon coding for two-component system regul- YANG, Y. K., MORIKAWA,M., SHIMIZU,H.,
atory proteins is located adjacent to the grs ope- SHIOYA,S.,SUGA,K., NIHIRA,T., YAMADA,Y.
ron of Bacillus brevis, DNA Sequence 5, 283- (1996), Maximum virginiamycin production by
290. optimization of cultivation conditions in batch
VANHOUTTE,B., PONS, M. N., THOMAS,C. R., culture with autoregulator addition, Biotechnol.
LOUVEL,L., VIVIER,H. (1995), Characteriza- Bioeng. 49,437-444.
Biotechnology
8 Lantibiotics
RALPHJACK
FRIEDRICH
GOTZ
GUNTHERJUNG
Tubingen, Germany
1 Introduction 325
2 The Unique Chemistry and Structure of Lantibiotics 325
2.1 Modified Amino Acids Found in the Lantibiotics 327
2.1.1 Lanthionine (Lan) and 3-Methyllanthionine (MeLan) 328
2.1.2 2d-Didehydroalanine (Dha) and 2,3-Didehydrobutyrine (Dhb) 328
2.1.3 D-Alanine 330
2.2 Type A Lantibiotic Primary Structures 330
2.2.1 Nisin 330
2.2.2 Subtilin 331
2.2.3 Epidermin and Gallidermin 333
2.2.4 Pep5 333
2.2.5 Epilancin K7 333
2.2.6 Strepococcin A-FF22, Lacticin 481 (Lactococcin DR), and Salivaricin A 334
2.2.7 Lactocin S 335
2.2.8 Enterococcal Cytolysin/Bacteriocin 335
2.2.9 Carnocin U149 and Mutacin 336
2.3 Type B Lantibiotic Primary Structures 336
2.3.1 Cinnamycin, Duramycin, Duramycin B, Duramycin C, and Ancovenin 336
2.3.2 Mersacidin and Actagardine 338
3 Lantibiotic Structures in Solution 338
3.1 Type A Lantibiotics 339
3.1.1 Gallidermin 339
3.1.2 Nisin and Subtilin 339
3.1.3 Pep5 340
3.2 Type B Lantibiotics 340
3.2.1 Cinnamycin and the Duramycins 340
324 8 Lantibiotics
4 The GenedProteins Involved in Lantibiotic Biosynthesis and Genetic Regulation 341

4.1 The Requirement for Multiple Gene Products in Lantibiotic Biosynthesis 341
4.2 Structural Genes and Pre-Lantibiotics 341
4.2.1 The Location and Features of the Structural Genes and their Transcription 341
4.2.2 The General Structure of Prepeptides 345
4.2.3 The Possible Role(s) of the Leader Peptide 346
4.3 Amino Acid Modifying Proteins 347
4.3.1 LanB/LanC and LanM Proteins 347
4.3.2 EpiD 348
4.4 Transport Proteins 348
4.5 Leader Peptidases (LanP) 348
4.6 Proteins Regulating Lantibiotic Biosynthesis 350
4.7 Producer Self-Protection (“Immunity”) Mechanisms 351
4.8 The Chain of Events Leading to Lantibiotic Biosynthesis and Maturation 353
5 Biological Activities of Lantibiotics 353
5.1 Type A Lantibiotics 353
5.1.1 Primary Mode of Action 353
5.1.2 Secondary Mode of Action 356
5.2 Type B Lantibiotics 357
5.2.2 Cinnamycin and the Duramycins 357
6 Applications of Lantibiotics 358
6.1 Applications as a Food/Beverage Preservative 358
6.2 MedicaUParamedical and Veterinary Applications 359
7 Conclusions and Future Perspectives 359
8 References 360
1 Introduction ically synthesized by large multi-enzyme com-

plexes in the cell and for which there exists no
structural gene (KATZ and DEMAIN,1977;
The search for novel pharmaceutical com- KLEINKAUFand VON DOHREN,1986, 1987,
pounds with potent new biological activities 1990; NAKANO and ZUBER,1990). Studies of
has only relatively recently begun to look in- the lantibiotics have revealed that they are
ward and away from complex in-lab synthetic genetically encoded and are uniformly pro-
processes toward the huge array of “natural” duced on the ribosome as a precursor peptide
compounds produced by bacteria, offering a which is subsequently modified at specific
plethora of new possibilities. Such studies of points to give rise to the large number of
bacterial-derived proteins in particular have modified amino acids found in these peptides.
unearthed a vast array of naturally produced In addition, the peptides are produced with a
compounds, not the least of which is the leader peptide which is removed during matu-
group of highly modified antimicrobial and ration and are transported by specific trans-
enzyme-inhibitory peptides collectively report-related proteins from the cell. Still other
ferred to as “lantibiotics”. Originally coined to lantibiotic-specific proteins are involved in
describe the rapidly expanding group of the genetic regulation of biosynthesis and
antibiotic-like peptides which were found to generation of the specific producer cell self-
contain the non-protein amino acids lanthion- protection mechanism(s) frequently ob-
ine and 3-methyl lanthionine (SCHNELL et al., served.
1988), the name lantibiotics belies the full ex- The mature lantibiotics have also found a
tent and complexity of this class of bacterially number of potential applications including
synthesized peptides. For example, as the medical and veterinary antibiosis, food, bev-
number of characterized lantibiotics in- erage, and cosmetic preservation and as regu-
creases, novel modified amino acids of inter- lators of both human immune function and
est to chemists and biochemists alike are con- blood pressure (HURST,1981; JUNG, 1991a, b;
stantly being identified, including those with DELVES-BROUGHTON, 1990; MOLITORand
unusual crosslinks or unsaturated R-groups, SAHL, 1991; BIERBAUM and SAHL, 1993; DE
while analysis of the solution structure of the VUYSTand VANDAMME, 1993; JACK et al.,
peptides concerned is offering new perspec- 1995; SAHL et al., 1995). In the following
tives on the relationship between peptide chapter we wish to provide an overview of the
structure and biological function. novel structures, mechanism(s) of biosynthe-
Perhaps more strikingly though, a number sis, genetic regulation, biological activities
of recent studies of the biosynthesis of lanti- and current as well as potential applications
biotics has revealed that they are produced by for this fascinating, novel class of bacterial-
ribosomal biosynthesis and subsequently derived, biologically-active peptides.
modified to generate the mature, biologically
active compound (SCHNELL et al., 1988; BA-
NERJEE and HANSEN,1988; BUCHMANN et
al., 1988; KALETTA and ENTIAN,1989). What
this means is that there are proteins within 2 The Unique Chemistry
the lantibiotic-producing cells which are able
to carry out specific transformations of amino and Structure of
acids, converting them to novel structures
which may play a role in the biological activi- Lantibiotics
ty, stability, etc. of the peptide and, more im-
portantly, may prove useful to those biotech- In order to appreciate the potential of the
nologists interested in creating similar modifi- lantibiotics as biologically active peptides, as
cations of other peptides. well as the possibilities for the application of
It is this last feature of ribosomal biosyn- their respective biosynthetic machinery, it is
thesis which sets the lantibiotics apart from first necessary to discuss the structure of both
the classical peptide antibiotics which are typ- the peptides themselves and the modified
Tab. 1. A Compilation of the Currently Described Lantibiotics" Including Several of their Respective Chemical and Physical Characteristics
Lantibiotic Producing Organism Mass Net Chargeb Number of % Modified Reference

[Da] (at pH 7.0) Rings' Residuesd
Type A
Nisin A Lactococcus lactis 3353 +3 5 38 GROSSand MORRELL(197
Nisin Z Lactococcus lactis 3330 +3 5 38 MULDERSet al. (1991)
Subtilin Bacillus subtilis 3317 +2 5 40 GROSSet al. (1973)
Epidermin Staphylococcus epidermidis 2164 +3 4 41 ALLGAIERet al. (1986)
Gallidermin Staphylococcus gallinarum 2164 +3 4 41 KELLNERet al. (1988)
[lV,6L]-Epidermin Staphylococcus epidermidis 2151 +3 4 41 SAHLet al. (1995)
Pep5 Staphylococcus epidermidis 3488 +7 3 26 KELLNERet al. (1989)
Epilancin K7 Staphylococcus epidermidis 3032 +5 3 32 VAN DE KAMPet al. (1995a
Lactocin S Lactobacillus sake 3764 -1 2 24 SKAUGEN et al. (1994)
SA-FF22 Streptococcus pyogenes 2795 +1 3 27 JACKet al. (1994a)
Lactococcin DR Lactococcus lactis 2901 0 3 26 RINCEet al. (1994)
Salivaricin A Streptococcus salivarius 2315 0 3 27 Ross et al. (1993)
Cytolysin L1 Enterococcus faecalis 4164 0 N.R." N.R. GILMOREet al. (1994)
Cytolysin L2 Enterococcus faecalis 2631 0 N.R. N.R. GILMOREet al. (1994)
Carnocin U149 Carnobacterium piscicola 4635 N.R. N.R. N.R. STOFFELSet al. (1994)
Mutacin Streptococcus mutans 3245 N.R. 3 N.R. NOVAKet al. (1994)
Type B
Cinnamycin Streptomyces cinnamoneus 2042 0 4 47 FREDENHAGEN et al. (1991
Duramycin Streptomyces cinnamoneus 2014 0 4 47 FREDENHAGEN et al. (1991
Duramycin B Streptoverticillium spp. 1951 0 4 47 FREDENHAGEN et al. (1991)
Duramycin C Streptomyces griseoluteus 2008 -1 4 47 FREDENHAGEN et al. (1991
Ancovenin Streptomyces spp. 1959 0 3 37 WAKAMIYA et al. (1985)
Mersacidin Bacillus subtilis 1825 -1 4 42 KOGLERet al. (1991)
Actagardine Actinoplanes spp. 1890 0 4 45 ZIMMERMANN (19951
a Adapted from JACKand SAHL(1995) and SAHLet al. (1995).

Includes those amino- and carboxy termini which remain unmodified following posttranslational modification.
amino acids which they contain. Tab. 1 lists 2.1 Modified Amino Acids Found
all of the currently characterized lantibiotics,
the bacterium which produces them as well as in the Lantibiotics
a number of their physicochemical properties
relevant to the proceeding sections (SAHLet While the lantibiotics take their name from
al., 1995; JACKand SAHL, 1995). the observation that they contain lanthionine
In addition, an earlier review of the lanti- (SCHNELLet al., 1988), this is not the only
biotics has suggested that the peptides can be modified amino acid that they possess. Cur-
divided into two distinct groups (JUNG, rently, a number of modified amino acids and
1991a, b), a subdivision scheme which has other residues have been found in lantibiotic
been maintained in the following chapter. structures including: meso-lanthionine (Lan),
The type A lantibiotics can be defined as lanthionine sulphoxide, fhreo-p-methyllan-
elongated, helical peptides whose primary thionine (MeLan), S-[(Z)-2-aminovinyl]-~-
mode of action appears to be directed at the cysteine (AviCys), S-[(Z)-2-aminovinyl1-3-
depolarization of the bacterial cytoplasmic methyl-D-cysteine, 2,3-didehydroalanine (Dha),
membrane through the formation of voltage- 2,3-didehydrobutyrine (Dhb), D-alanine, 2-
dependent pores. Alternatively, type B lanti- oxopyruvate, 2-oxobutyrate, hydroxypyru-
biotics are generally somewhat smaller, com- vate, erythro-3-hydroxyaspartateand (2S,8S)-
pact, globular structures which either inhibit lysinoalanine, the structures of which are pre-
bacterial cell wall replication, interact with sented for comparison in Fig. 1. Some of
specific enzymes, or activate T-cell prolifera- these residues such as Lan, MeLan, Dha, and
tion (see also Sect. 5 ) . Dhb occur in most, if not all, lantibiotics so-
0 0 0 0 0
II II II - II 0
II
-HN\m,c- -HN \@YC- -"\myC- "\iyC - II -HN\N/c-oH
H--6 -6-H H--6 -6-H -6-H
I I -w\TIyc- I
-S a 2 H&-C --S a 2 H--6. ( ~ 9 3
H' I I
H2C -NH- a 2
0
II
2-
n
-HN c-
ri -HN -HN c-
0
il -m -"\
H
\@Y
- Ca \(Q
G - H
\@Y
H--6
\@'
C. - H
C-H
I
I II I II
H2C -S- CH H3C-C-S- CH
H/
S-[(Z)-Z-aminovinyl]-o-cysteine S[(Z)-2-runinovinyl]-3-methyls-cysteinc r r y t h m - 3 4 1 y d r o ~ ~ @ cacid
0 0 0 0 0 0 0 OH 0
-\/-
II II II II II II I1 I II
-"\ 2- -"\ 7-
c-c-
I
c-c-
I
H-C-C-
I
II II
CH
H-C
I
CH3
P
a33
-3
HlC CH3
H3C
z.3diddlydrcahiw (Z)-2.3-didehydmbutyrine danine 2-oxopynrvyl 2-oxobutyIyl Z-hydroXypyNvyl
Fig. 1. The structures of the modified amino acids and other residues found in lantibiotics. The chirality of
various structures is indicated (where appropriate) above the respective a-carbon atom.
328 8 Lantibiotics
far characterized, while others occur only in rently characterized serine and threonine de-
individual cases. However, due in part to hydratases appear to act on free amino
their novel nature and properties, many of groups via pyridoxal phosphate-dependent
these residues may be of great interest to the Schiff base reactions, suggesting that Ser and
researchers involved in biotechnology. Thr residues incorporated into a peptide
chain would not be appropriate as substrates
for these enzymes.
The subsequent step in lanthionine biosyn-
2.1.1 Lanthionine (Lan) and thesis (Fig. 2) involves the addition of the sul-
3-Methyllanthionine (MeLan) phydryl group of a neighboring Cys residue to
the a,punsaturated amino acid to form a co-
It was the pioneering research of E. GROSS valent thioether bridge (SCHNELLet al., 1988;
and co-workers that finally demonstrated that JUNG,1991a, b). Again, while it is not clear
the peptide antibiotics nisin and subtilin ac- how this step occurs in vivo (i.e., spontaneous
tually contained Lan and MeLan (as well as addition or specific protein-directed addi-
Dha and Dhb), confirming previously held tion), it has some interesting ramifications for
beliefs (BERRIDGEet al., 1952; GROSSand the stereospecificity of the final product.
MORRELL,1971; GROSSet al., 1973). Howev- Since lantibiotics prepeptides are synthesized
er, it was not clear at that time how these ami- on the ribosome, they contain only L-amino
no acid structures should arise. Later studies acids; despite this, Lan and MeLan in the ma-
showed that inhibitors of protein biosynthesis ture lantibiotic are consistently found in the
prevented nisin and subtilin production rneso(i.e., D/L-isomer)-configuration with the
(HURST, 1981), whilst the isolation of the SerlThr-derived “half” in the D-form while
structural genes for a number of lantibiotics the Cys-derived “half” remains in the L-confi-
has finally proved conclusively that these resi- guration. In addition, in the case of type A
dues arise from posttranslational modification lantibiotics the SerKhr-derived “half” of the
of a ribosomally synthesized precursor pep- Lan and MeLan is always closer to the N-ter-
tide (SCHNELLet al., 1988; BUCHMANN et al., minus than is the Cys-derived “half” of the
1988; BANERJEEand HANSEN,1988; KALET- amino acid, however, this is not always true
TA and ENTIAN,1989). In addition, these amongst the type B lantibiotics (JUNG,
studies also showed that Ser, Thr and Cys re- 1991a, b).
sidues found in the prepeptides but not in the
mature lantibiotics, must be the precursors
for the formation of Lan, MeLan, Dha and 2.1.2 2,3-Didehydroalanine (Dha)
Dhb.
While it is not clear how Lan and MeLan and 2,3-Didehydrobutyrine (Dhb)
are synthesized in the cell, the structure of
these thioether-linked, di-carboxy, di-amino The a$-unsaturated amino acids Dha and
acids has been studied in detail and a model Dhb (formed as described above for the first
(Fig. 2) for their biosynthesis has been pro- step in Lan/MeLan biosynthesis) are stable
posed (SCHNELLet al., 1988; JUNG,1991a, b). within the mature lantibiotic, however, their
The first step involves the site-specific dehy- appearance at the amino terminus of the pep-
dration of the a-amino-/3-hydroxy acids Ser tide, such as occurs during amino acid se-
and/or Thr in the propeptide part of the pre- quencing by Edman-degradation, can prove
lantibiotic; such dehydration results in the problematic. The structure elucidation of a
formation of Dha or Dhb, respectively. Pre- number of lantibiotics has previously been
sumably, this reaction is carried out by a spe- hampered by the presence of these amino
cific enzyme within the lantibiotic-producing acids (GROSS and MORRELL, 1971; ALL-
cells, however, neither the enzyme nor the GAIER et al., 1986; KELLNERet al., 1988,
molecular mechanism responsible have yet 1989; JUNG,1991a, b; JACKet al., 1994a; VAN
been identified. Also, the enzyme(s) capable DE KAMPet al., 1995b), the problems result-
of such a reaction should be novel since, curing from the spontaneous oxidative deamina-
(Sert3) (CYS+'l)
H O H O
I I1 I II
[precpidermin(-30to +2)]-N C- [Preepidermin(+4to +6)]-N C- [pre-epidermin(+7to +22)]
wC-H wC-H
I
H-C-OH
I
7%
I
H SH
(L)-serine (Lkcysteine
Site-specific dehydration
@ha+3) (CYS+7)
H O H O
I I1 I I1
[preepid&n(-30 to +2)]-N c- [pre-epidermin(Mto +6)] -N C- [preepidermin(+7to +22)]
\ C/ L I /
C-H
II I
H-C
I YIEL
1
H SH
2,34dehydrodanine (L)-cysteine
Stereospecific addition
H O H O
I I1 I I1
[Precpidermin(-30to +2)]-N C- [pre-epidermin(+4to +6)]-N C- [pre-epidermin(+7to +22)]
\d
H-C
b I /
C-H
I I
S
H-C
H
I 7-"
H
meso-lanthionine
Fig. 2. Proposed pathway for the chemical reactions occurring during the biosynthesis of lanthionine from
the precursor amino acids serine and cysteine using, as an example, the formation of ring A of epidermin.
The chirality of the respective compounds is indicated (where appropriate) above the a-carbon atom.
tion of N-terminally situated Dha or Dhb to thioalcohol, however this has proved unsuc-
produce a sequence-blocking pyruvyl or buty- cessful with other lantibiotics such as SA-
ryl group, respectively (GROSSand MORELL, FF22 (JACKet al., 1994a). Recently, MEYER
1971; JUNG,1991a, b). During the elucidation (1994) has developed a convenient, new ami-
of the structure of Pep5 (KELLNERet al., no acid sequencing regimen for the detection
1989) this problem was circumvented by first of suitably derivatized dehydroamino acids
reducing the two Dhb residues with benzyl- and thioethers which involves thiol addition,
330 8 Lantibiotics
oxidation with per-trifluoroacetic acid and a

second round of thiol addition. Similar oxida-
tive deaminations of N-terminal Dhb and
Dha leads to the formation of the novel se-
quence blocking 2-oxobutyryl and 2-oxopyru- \ad
F-H
vyl residues found at the N-terminus of Pep5
and lactocin S, respectively (KELLNERet al.,
~-6-0~
I
1989; SKAUGENet al., 1994) and will be dis- H
cussed in subsequent sections. (+seine
2.1.3 D-Alanine
Amino acids in bacterially-synthesized pro-
teins and peptides are normally present as L-
isomers, however, the type A lantibiotic lac-
tocin s, which is produced by Lactobacillus
sake strain Lb45, contains D-alanine (SKAUG-
EN et al., 1994). Interestingly, analysis of the
prepeptide sequence showed that the precur-
sor for this amino acid was L-serine, rather
than alanine as might be expected. Thus, it
seems that specific serine residues are dehy- H-C
drated to form Dha and then hydrogenated in I
H
a stereospecific fashion to yield D-alanine
during lactocin S biosynthesis (Fig. 3). Since 2,3-&dehydroalanine
the Dha residues found in other lantibiotics
are generally stable, it seems likely that such a
reaction should be catalyzed by a hitherto un-
known enzyme. In addition, the identification
of an enzyme or enzyme system able to create
D-isomers of alanine in peptides from serine
precursors offers considerable hope for new
approaches in biotechnology and peptide en-
gineering.
H O
I I1
[ I d n S(l to 611 -N C- [ l a d n S(8-3711
2.2 Type A Lantibiotic Primary w
Structures H-’i
H-C-H
I
H
2.2.1 Nisin
(Dkalanine
Since the structure of nisin was the first lan-
tibiotic described (GROSS and MORRELL, Fig. 3. Proposed pathway for the chemical reactions
occurring during the biosynthesis of D - a h i n e from
1971) and nisin was probably the first lanti- a precursor serine residue at position 7 of the type
biotic ever identified (ROGERSand WHIT- A lantibiotic lactocin S. The chirality of the respec-
TIER, 1928), it is fitting to begin any treatise tive compounds is indicated above the a-carbon
of lantibiotic structure with this particular ex- atom.
ample. Nisin is produced by the cheese starter
culture organism Lactococcus lactis ssp. lactis
and is antimicrobial for a broad variety of
gram-positive bacteria (HURST,1981; MOLI-
TOR and SAHL,1991). The peptide itself (Fig. machinery of the cell was able to cope with
4) is a 34 amino acid, 3353 Da, pentacyclic forming a novel dehydrated residue at a posi-
structure formed from its content of one Lan tion not normally occupied by such an amino
and 4 MeLan residues (GROSS and MOR- acid. Interestingly, [Met17Gln/GlylSDhb]-ni-
RELL,1971). While rings A, B, and C are sep- sin also had improved antimicrobial activity
arate the C-terminal pair of rings (D and E) against Bacillus cereus and decreased activity
form a bicycle (in all cases rings have been as- against Micrococcus flavus, while [Metl7Gln/
signed ascending alphabetic characters, start- Glyl8/Thr]-nisin had heightened activity
ing with the A ring closest to the N-terminus). against M . flavus and reduced activity against
In addition, nisin contains one Dha and 2 B. cereus (KUIPERSet al., 1992).
Dhb residues. Recently, MULDERSet al. (1991) identified
In addition to nisin, a number of nisin frag- a naturally occurring analog of nisin which
ments have also been isolated and character- they named nisin Z. The peptide was charac-
ized, many of which have altered antimicro- terized by a single amino acid exchange
bial, activity. BERRIDGEet al. (1952) first re- (His27Asn), resulting from a single base
ported the isolation of nisin fragments gener- change in the 3rd base of the 27th codon of
ated by long-term storage under acidic condi- the structural gene nisZ. Interestingly, while
tions which have subsequently been isolated this exchange appears to have little (if any)
and characterized by CHANet al. (1989a, b). effect on the biological activity of the mutant
Nisinl-32(amide), which results from the hy- nisin, it seems to make the peptide more sol-
drolysis of the Va133-Dha34 peptidyl bond, uble at or around pH 7, presumably as a re-
has similar antimicrobial activity to native ni- sult of the increased hydrophilicity of Asn at
sin, suggesting that the C-terminal two amino neutral pH. Thus, nisin Z might prove to be
acids are dispensable and that Dha33 plays no useful for applications where the low solubili-
particular role in the biological activity of the ty of nisin at neutral pH (HURST,1981) has
peptide (ROLLEMAet al., 1991). However ni- traditionally proved problematic (see also
sinl-32(amide), in which the DhaS-Leu6 pep- Sect. 6).
tidy1 bond had also been hydrolyzed and ring
A was therefore opened, was virtually inac-
tive. 2.2.2 Subtilin
Subsequently these studies have been con-
firmed by site-directed mutagenesis (DODD Like nisin, subtilin is a pentacyclic type A
and GASSON,1994); the exchange Dha33Ala lantibiotic (Fig. 4); in fact, the arrangement
resulted in only slightly decreased activity and nature of all of the thioethers are con-
while the double exchange DhaSAla/ served between nisin and subtilin, even
Dha33Ala essentially destroyed the antimi- though the amino acid sequences of the two
crobial activity of the peptide. Similarly, the peptides varies significantly (GROSS and
presumably conservative exchange DhaSDhb MORRELL,1971; GROSS et al., 1973). First
led to a substantial decrease in activity (be- described by JANSEN and HIRSCHMANN
tween 2- and 10-fold) against some indicators (1944), subtilin is produced by Bacillus subtil-
(KUIPERSet al., 1992), perhaps suggesting is and is active against both the vegetative
that these residues are not only important for cells of a number of gram-positive bacteria as
antimicrobial activity but may also alter the well as the outgrowth of endospores of Bacil-
spectrum of bacterial strains affected. In addi- lus spp. and Clostridium spp. (HURST,1981).
tion, these authors also altered ring C of nisin Interestingly, the two activities are distinguish-
with the dual exchange Metl7Gln/GlylSThr. able, apparently dependent on saturation of
Two products were resolved from this ex- Dha5 which leads to loss of subtilin antispore
change; intact nisin containing either Thrl8 activity, the half-life of which is only 0.8 d
or Dhbl8. The latter product not only con- (HANSENet al., 1991; LIUand HANSEN,1992;
tains a ring with the same sequence as the HANSEN,1993). Using site-directed mutagen-
structurally similar lantibiotic subtilin (Fig. 4), esis, the exchange Dha5Ala was shown to re-
but also demonstrates that the biosynthetic sult in the loss of antispore activity, confirm-
Nisin
Subtilin
-rl S-
I -
Pep5 c -co
II
ii ' S '
Epilancin K7 ~17- - m
dH I S -
Epidennin
c--co
Lactocin S 7
II
SA-FF22
2 The Uniqrxe Chemistry and Structure of Lantibiotics 333
4 Fig. 4. The structure of several type A lantibiotics. Subsequently, a structural analog of epider-
Amino acid residues involved in posttranslational min, called gallidermin and produced by Sta-
modifications have been identified by highlighting phylococcus gallinarum, has also been iso-
while additional modified residues are indicated by lated and characterized (KELLNER et al.,
their respective chemical structures. Dha, 2,3-dide- 1988). While gallidermin appears to have a
hydroalanine; Dhb, 2,3-didehydrobutyrine;Abu, a-
aminobutyric acid; Ala-S-Ala, (2S,6R)-lanthionine; slightly different spectrum of biological activi-
Abu-S-Ala (2S,3S,6R)-3-methyllanthionine. The ar- ty to epidermin (especially against Propioni-
rangement of the thioether bonds in the lantibiotics bacterium acnes), it differs only by the ex-
SA-FF22, lacticin 48l/lactococcin DR, and salivari- change of Ile6 in epidermin for Leu, an ex-
cin A remains unreported. change which is conservative and does not al-
ter the molecular mass. In all other respects,
the two structures are identical.
ing that saturation of Dha5 was the primary
cause of subtilin instability (LIU and HAN- 2.2.4 Pep5
SEN, 1992). In addition, these authors also
showed that the exchange Glu4Ile (i.e., the The type A lantibiotic Pep5 was first iso-
same as is found in this position of nisin; Fig. lated from Staphylococcus epidermidis strain
4) not only increased the half-life of the anti- 5 (SAHLand BRANDIS,1981) and has subse-
spore activity by 57-fold but also increased quently been thoroughly characterized
the sporicidal activity by a factor of 3. From (KELLNERet al., 1989, 1991). The peptide
these results, they suggest that Glu4 may be (Fig. 4) is the largest characterized type A
involved in the spontaneous saturation of lantibiotic (3488 Da) and consists of 34 amino
subtilin, probably acting as a Michael-type ac- acids, 8 of which are basic. In addition, Pep5
ceptor. In similar structure-function studies, is tricyclic as a result of its containing one
subtilin with a succinylated N-terminus has MeLan and two Lan residues; ring A is sepa-
been isolated; this peptide also had reduced rate while, rings B and C overlap. The N-ter-
biological activity (CHANet al., 1993). minus of Pep5 is occupied by a 2-oxobutyryl
residue (KELLNERet al., 1989) and analysis
of the sequence of the structural gene, pepA
2.2.3 Epidermin and Gallidermin (KALEITA et al., 1989), shows that the +1
codon of the prepeptide encodes a threonine
Elucidation of the structure of the Staphy- residue which is subsequently converted to
lococcus epidermidis lantibiotic epidermin Dhb during modification and maturation
(ALLGAIERet al., 1986) has proved impor- (WEIL et al., 1990). It appears that once this
tant in terms of our understanding of lanti- Dhb residue is located at the N-terminus (i.e.,
biotic biosynthesis, as it assisted in the isola- following removal of the leader peptide) it
tion of the structural gene (epiA), proving undergoes oxidative deamination to form 2-
conclusively that lantibiotics were ribosomal- oxobutyryl (Fig. 5), probably by a sponta-
ly synthesized (SCHNELLet al., 1988). Epider- neous mechanism (KELLNERet al., 1989).
min (Fig. 4) is a 22 amino acid (2164 Da), te-
tracyclic peptide which contains one residue 2.2.5 Epilancin K7
each of Dhb and MeLan along with two resi-
dues of Lan. The 4th cyclic structure results Also produced by Staphylococcus epider-
from the novel C-terminal mono-carboxy, di- midis, epilancin K7 is somewhat similar to
amino acid, Cys(Avi) (Fig. 1) between resi- Pep5 (Fig. 4). Recently, VAN DE KAMPet al.
dues 19 and 22, the formation of which results (1995a, b) solved the structure of the peptide
from oxidative decarboxylation of the C-ter- by a variety of chemical and homonuclear
minal Cys residue by the enzyme EpiD (see NMR-based strategies and showed that it is a
also Sect. 4.3.2). In addition, ring B is identi- 3052 Da, tricyclic, type A lantibiotic contain-
cal to ring B of both nisin and subtilin, as is ing one Lan and two MeLan residues. In ad-
the arrangement of both rings A and B dition, epilancin contains two residues each of
among these 3 lantibiotics. Dha and Dhb as well as 6 lysine residues (and
334 8 Lantibiotics
1
Pre-Pep5
Cleavage of
leadex peptide
f7-C-
lfl
FHz
pepS(+2 to +34)] a HP fl
HzNT-C-
7%
[pepS(+210+34)]
+H 0
& bN-C-C-
bC4-H
8 FepS(+2 to +34)1
CH3 CH3
2-oxobutyryl-PepS
Fig. 5. Proposed pathway for the spontaneous oxidative deamination of dehydrobutyrine located at the
N-terminus of Pep5 after cleavage of the leader peptide.
no acidic amino acids), accounting for its ex- RELL, 1971), TAGGet al. (1973a, b) reported
tremely basic nature. Again, like Pep5 that a clinical isolate of Streptococcus pyo-
(KELLNER et al., 1989), the N-terminus of genes designated strain FF22 produced “a dif-
epilancin is blocked; however, while sponta- fusible inhibitor of bacterial growth” which
neous deamination of the N-terminal Dha they called streptococcin A-FF22. Subsequent
should result in the formation of 2-oxopyru- partial purification and characterization
vyl (in analogy to that shown for Pep$ Fig. showed that the inhibitory agent was essen-
5), VAN DE KAMPet al. (1995a) suggest that tially proteinaceous, of low molecular weight
the N-terminal residue is occupied by 2-hy- (TAGGet al., 1973a, b), and was rather similar
droxypyruvate (Fig. 1). The formation of this to nisin (TAGGand WANNAMAKER,1978;
group at the N-terminus probably suggests JACK and TAGG,1992). Recently, streptococ-
that an additional, novel enzyme function is cin A-FF22 was purified to homogeneity and
required for epilancin biosynthesis. In addi- chemically characterized (JACK and TAGG,
tion, this study also isolated a deletion pep- 1991); the lantibiotic was shown to be a 27
tide resulting from hydrolysis of the Ala2- amino acid, 2795 Da type A lantibiotic (Fig. 4)
Dha3 peptidyl bond; epilancin K7(3-33) is which contains one Lan and two MeLan resi-
also blocked, in this case by the expected 2- dues as well as one residue of Dhb (JACK et
oxopyruvyl group, presumably formed by oxi- al., 1994a). Using the sequence information
dative deamination of the N-terminal Dha re- gained from these studies, the structural gene
sidue. Interestingly, this peptide appears to be scnA has been cloned and sequenced and
approximately as active as native epilancin confirms the correct sequence for the peptide
K7, suggesting these two N-terminal amino (HYNESet al., 1993). Unfortunately, because
acids are not essential for its biological activi- of the overlapping nature of the thioether
tY. rings in streptococcin A-FF22, assignment of
the ring order by enzymic and chemical diges-
tions and modifications has proved impossi-
2.2.6 Streptococcin A-FF22, ble (JACK et al., 1994a).
Lacticin 481 (Lactococcin DR), In addition, JACK and TAGG(1991) also
isolated a naturally-occurring derivative of
and Salivaricin A streptococcin A-FF22 devoid of detectable
antibacterial activity, apparently as a result of
At around the same time as the structure of the loss of the N-terminal 4 amino acids
nisin was being reported (GROSSand MOR- which are not involved in formation of ring
structures. While C-terminal amino acid-defi- terized (M0RTVEDT and NES, 1990; M0RT-
cient derivatives have not yet been found, it VEDT et al., 1991). In addition, the structural
would seem that the antibacterial activity of gene encoding the prepeptide has been iso-
streptococcin A-FF22 is dependent on (at lated and characterized in order to gain an in-
least) an intact N-terminus, in direct contrast sight into the likely structure of this lantibiot-
to observations made with the type A lanti- ic (SKAUGENet al., 1994). Taken together,
biotic epilancin K7 (VAN DE KAMP et al., these studies have demonstrated that the pep-
1995a). tide is the largest lantibiotic so-far character-
Another type A lantibiotic which shows ized (3764 Da) and probably contains two
striking similarities to the structure of strepto- Lan residues. In addition, in a situation analo-
coccin A-FF22 has been isolated from Lacto- gous to that observed in Pep5 (Fig. 5; KELL-
coccus lactis ssp. lactis and called lacticin 481 NER et al., 1989) the N-terminus is blocked,
(PIARDet al., 1992). In addition, RINCEet al. probably by a 2-oxopyruvyl group which
(1994) have recently isolated a lantibiotic would arise from the spontaneous oxidative
from Lactococcus lactis ssp. lactis DR which deamination of a N-terminally located Dha
proved to be identical to lacticin 481. Prelimi- residue. However, the nature of the blocking
nary amino acid sequence analysis of lacticin group as well as the arrangement of the
481 has allowed the isolation of the structural thioether bridges has yet to be resolved. Fi-
gene, lcfA (PIARDet al., 1993) and complete nally, lactocin S has also been shown to con-
analysis of the primary structure of the pep- tain D-alanine at positions which contain Ser
tide has shown it to be a 28 amino acid, 2901 in the prepeptide sequence, prompting specu-
Da type A lantibiotic with one MeLan, one lation that these residues arise following ste-
Dhb and two Lan residues (Fig. 4). Interest- reospecific hydrogenation of Dha residues at
ingly the peptide shares both sequence and these sites in the propeptide domain (see also
structural similarities with streptococcin A- Sect. 2.1.3).
FF22 (JACK et al., 1994a), especially in the
presence of the 3 overlapping thioether rings
and the single residue of unmodified serine at 2.2.8 Enterococcal
the C-terminus (PIARDet al., 1993).
Salivaricin A (Fig. 4) is a lantibiotic pro- Cytolysin/Bacteriocin
duced by Streptococcus salivarius strain 20P3
which acts against a number of pathogenic The cytolysin/bacteriocin of Enterococcus
streptococci including the principal human faecalis has long-since been shown to be a vi-
pathogen, Streptococcus pyogenes (DEMPS- rulence determinant which also has the ability
TER and TAGG, 1982). Purification and mi- to kill certain gram-positive bacterial strains
crocharacterization of the peptide responsible (BROCKand DAVIE,1963), however, the re-
showed that it is a small (2315 Da) type A cent isolation of the genetic elements respon-
lantibiotic with one Lan and two MeLan resi- sible for its biosynthesis have demonstrated
dues (Ross et al., 1993). In addition, this par- the relationship between this toxin and the
ticular lantibiotic is apparently the only one lantibiotics (GILMOREet al., 1994). The cyto-
so-far characterized which does not contain lysin/bacteriocin system appears to comprise
either Dha or Dhb. Also, although no at- two peptides both of which are lantibiotics
tempts appear to have been made to deter- and both of which are required for antibacter-
mine the order of the thioether bridges, it ial activity; Cyll is a 4163 Da peptide while
seems likely that all three should overlap. Cy12 has a mass of only 2631 Da, however,
while lanthionine has been identified in these
peptides the exact number of residues etc. is
2.2.7 Lactocin S not yet clear (SAHLet al., 1995). From the de-
rived amino acid sequences obtained through
Lactocin S (Fig. 4) is a large, 37 amino acid analysis of the genes, it appears that the two
lantibiotic produced by Lactobacillus sake peptides have regions of identity, suggesting
which has been purified and partially charac- that they may have arisen through gene dupli-
336 8 Lantibiorics
cation and partial deletion (GILMOREet al., blood pressure regulatory systems, including:
1994). phospholipase A2 and angiotensin-converting
enzyme (see also Sect. 5). Over a period of a
number of years, searches for such inhibitors
2.2.9 Carnocin U149 and Mutacin of immune function and blood pressure regul-
ation have identified a number of these com-
STOFFELSet al. (1992) have reported the pounds (some of which have subsequently
isolation of a large lantibiotic from Curnobac- been shown to be identical) including: cinna-
terium piscicolu which they named carnocin mycin (formerly also known as lanthiopeptin
UI49. The peptide is particularly hydrophobic or Ro09-0198), duramycin (formerly also
as judged by its chromatographic behavior, known as leucopeptin), duramycin B, dura-
has a mass of 4635 Da and a novel N-terminal mycin C, and ancovenin (BENEDICTet al.,
amino acid sequence of G-S-E-I-Q-P-R which 1952; SHOTWELLet al., 1958; GROSS, 1977;
is blocked to further sequencing at the 8th re- WAKAMIYA et al., 1985,1988; KESSLERet al.,
sidue. While the exact number and nature of 1987, 1988, 1991; NARUSEet al., 1989; FRE-
amino acids present in this lantibiotic remains DENHAGEN et al., 1990,1991; HAYASHI et al.,
to be determined, preliminary results suggest 1990), the structures of which are shown in
it is somewhat similar to lactocin S (SAHLet Fig. 6. Each of these compounds has been iso-
al., 1995). In addition, the peptide appears to lated from different strains of Streptomyces
have a propensity for membrane perturbation spp. or Streptoverticillium spp. and all show
and seems to interact with nisin-producing limited antimicrobial activity, principally di-
strains in a specific way, suggesting there may rected against Bacillus spp. (see also Sect. 5).
be an interaction of carnocin U149 with both Since the structures of cinnamycin, duramy-
the cytoplasmic membrane and parts of the cin, duramycin B, duramycin C, and ancove-
nisin-synthesizing machinery of the cell nin are so similar, they will be regarded here
(STOFFELSet al., 1994). as the cinnamycin-group of type B lantibiotics
Mutacin, a lantibiotic from Streptococcus and described together. Indeed, SAHL et al.
mutuns T8, has also been isolated and charac- (1995) have suggested that since these type B
terized (NOVAKet al., 1994). This lantibiotic lantibiotics share such an extent of structural
has a mass of 3245 Da, contains one MeLan and sequence similarity (only 7 conservatively
and two Lan residues and has the unique N- exchanged amino acids overall), that they
terminal amino acid sequence N-R-W-W-Q- ought to be viewed as structural variants of
G-V-V-X, where X indicates a sequence the same compounds.
blocking group. As would be expected, the type B lantibiot-
ics of the cinnamycin group all contain lan-
thionine; indeed, the ring structures and ar-
2.3 Type B Lantibiotic Primary rangements are conserved among all these
peptides (Fig. 6). However, while each con-
Structures tains one Lan and two MeLan residues (FRE-
DENHAGEN et al., 1991; KESSLERet al., 1991;
2.3.1 Cinnamycin, Duramycin, JUNG,1991a, b), there are some notable dif-
Duramycin B, Duramycin C, and ferences between type A and type B lantibiot-
ics. Firstly, these type B lantibiotics all con-
Ancovenin tain a “head-to-tail” bridge, formed by a
MeLan residue spanning residues 1-18 and
Many of the type B lantibiotics described this Melan, along with that formed between
so-far have gained considerable interest, not residues 5 and 11 is in the opposite orienta-
so much because of their unusual structures tion to that observed in all type A lantibiotics
nor their antimicrobial activity (which in most (i.e., the Cys-derived “half” of the di-amino
cases is only moderate), but because they act acid is located toward the N-terminus). In ad-
as inhibitors of essential enzymes involved in dition, cinnamycin-group type B lantibiotics
both the human immune and circulatory (with the exception of ancovenin) also con-
Fig. 6.The structure of several type B lantibiotics. Amino acid residues involved in posttranslational mod-
ifications have been identified by highlighting while additional modified residues are indicated by their
respective chemical structures. Abu, a-aminobutyric acid; Asp-OH, erythro-3-hydroxy-asparticacid Ala-
NH-Lys, (2S,8S)-lysinoalanine;Ala-S-Ala, (2S,6R)-lanthionine; Ala-SO-Ala, lanthionine sulphoxide; Abu-
S-Ala (2S,3S,6R)-3-methyllanthionine.
tain a single residue of hydroxy-aspartic acid of ancovenin) contain the aminoether linked,
(Figs. 1 and 6), however, it is not clear from di-carboxy, di-amino acid, (2S,8S)-lysinoala-
whence or by what mechanism(s) this residue nine (Fig. l),formed between residues 6 and
arises (FREDENHAGEN et al., 1991; JUNG, 19 (Fig. 6). It has been proposed (JUNG,
1991a, b; SAHLet al., 1995). Similarly these 1991a, b) that such a residue could form by
type B lantibiotics (again with the exception the addition of the €-amino group of Lysl9
338 8 Lantibiotics
across the unsaturated Dha6, in a manner correct structure of the peptide has been de-
rather analogous to that observed for the for- termined by ZIMMERMANN and JUNG(1995)
mation of lanthionine shown in Fig. 2. What is after previous attempts (MALABARBA et al.,
not clear about these additional modifications 1985, 1990 KETTENRING et al., 1990). Acta-
is what role (if any) they play in the biological gardine contains 4 sulphide rings (1 Lan and 3
activity of their respective lantibiotic; indeed, MeLan residues) with the rings B, C and D
ancovenin contains neither hydroxyaspartic forming a tricycle. Interestingly, with the ex-
acid nor lysinoalanine, yet is functionally sim- ception of a single, conservative amino acid
ilar to other type B lantibiotics of this group exchange, ring B of actagardine and ring C of
(WAKAMIYA et al., 1985; SHIBAet al., 1986), mersacidin are identical (Fig. 6), a feature
suggesting that they may not be essential for which may be related to their similar biologi-
biological activity. cal activity (see also Sect. 5 ) , which is the inhi-
bition of peptidoglycan biosynthesis (SOMMA
2.3.2 Mersacidin and Actagardine et al., 1977; BROTZ et al., 1995; SAHLet al.,
1995). In addition, actagardine is the only lan-
By contrast to the cinnamycin-like type B tibiotic shown to contain 3-methyllanthionine
lantibiotics described in the preceding sec- sulphoxide (Fig. 6), however, it is not yet
tion, mersacidin and actagardine have been clear whether this residue has arisen as an ar-
studied extensively, in part, because of their tefact of the purification process (KETTEN-
strong antimicrobial activities (see also Sect. RING et al., 1990; ZIMMERMANN and JUNG,
5). Furthermore, as can be seen in Fig. 6 they 1995).
also differ significantly in their structural
characteristics, neither containing the head-
to-tail bridging pattern described above.
Mersacidin (Fig. 6) is produced by Bacillus
spp. (strain HIL Y-85,54728), is the smallest
3 Lantibiotic Structures in
lantibiotic (1825 Da) and consists of 20 amino Solution
acids arranged to give a tetracyclic structure
with rings C and D forming a bicycle (CHA- The lantibiotics, be they either type A or
TERJEE et al., 1992a, b). It is hydrophobic type B, contain a vast number of modified re-
with no net charge, contains a single residue sidues, many of which form bridging struc-
of Dha and 3 MeLan residues and is the only tures along various portions of the respective
lantibiotic to contain the C-terminal unsatu- peptides length (c.f. Figs. 5, 6 and Tab. 1).
rated amino acid (S)-[(Z)-2-aminovinyl]-3- Obviously then, these modified residues and
methyl-D-cysteine. This last residue is proba- unusual bridging patterns should have some
bly formed by a similar process as that re- ramifications on the overall structure adopted
sponsible for Cys(Avi) formation in epider- by the peptides in solution. Initially, unsuc-
min (Fig. 1; ALLGAIERet al., 1986; KUPKEet cessful attempts were made to obtain crystals
al., 1992), except that the dehydrogenated de- of several lantibiotics, in order to determine
carboxylated C-terminal Cys residue should their spatial structures by X-ray crystallogra-
react with a Dhb residue, rather than a Dha. phy (JUNGet al., unpublished data). Subse-
In addition, KOGLERet al. (1991) have shown quently, the advent of sensitive, high resolu-
that mersacidin also contains a residue of tion 2-D, 3-D and 4-D NMR techniques has
MeLan (between residue 1 and 2) formed by allowed a number of laboratories to look at
linkage in the opposite direction to that ob- the solution conformations of various lanti-
served with all type A lantibiotics (i.e., the biotic structures (SLIJPERSet al., 1989; CHAN
Cys-derived "half" of the first ring of mersa- et al., 1989b; PALMER et al., 1989; VAN DE
cidin is located to the N-terminal side with re- VENet al., 1991a,b; LIANet al., 1991; GOOD-
spect to the Dhb-derived "half"). MAN et al., 1991; FREUND et al., 1991a, b, c;
Actagardine (formerly gardimycin) was iso- KESSLERet al., 1991; ZIMMERMANN et al.,
lated from Actinoplunes spp. ATCC 31048 1993; ZIMMERMANN and JUNG,1995; VAN DE
and 31049 by PARENTIet al. (1976) and the KAMPet al., 1995a). In addition, a number of
3 Lantibiotic Structures in Solution 339
these studies have also been able to take ad- Overall, gallidermin adopts a distorted he-
vantage of the nature of NMR and assess the lix-like structure which has some degree of
effects of solutions of differing lipophilicity, flexibility in the central region. In addition, it
micelles, perturbants (e.g., urea) and phos- is amphiphilic with the hydrophobic C-termi-
pholipid vesicles on the solution structures nal residues aligned on one “face”, while the
obtained. N-terminal part contains the hydrophilic resi-
dues also oriented towards the opposite
3.1 Type A Lantibiotics “face” of the cork-screw (FREUNDet al.,
1991a, b). In TFE/water the peptide has an
3.1.1 Gallidermin overall length of about 3 nm, a diameter of
approximately 1 nm and a net dipole moment
The structure of gallidermin (Fig. 7) has of around 75 Debye (JUNG,1991a, b). The
been obtained both in trifluoroethanol amphiphilicity, high dipole moment and
(TFE)/water (95 :5) and dimethylsulphoxide membrane spanning length observed for galli-
(DMSO) where it adopts an extended, cork- dermin in solution may help to explain its
screw-like conformation (FREUND et al., biological activity, which is the formation of
1991a, b). Ring B (residues 8-11), which is voltage-dependent pores in biological mem-
identical to ring B of nisin, forms a p-turn branes (SAHL,1991; BENZet al., 1991).
type 11, while the central domain (residues
11-15) shows the greatest degree of flexibility 3.1.2 Nisin and Subtilin
over the whole molecule, due to its content of
turn-like motifs. Interestingly, this region in- A number of research groups have inde-
corporates a potential trypsin cleavage site pendently investigated the solution structure
(Lysl3-Dhbl4) which appears to be exposed of nisin (and, to some extent, subtilin) and
by this flexibility and has been shown by mo- found that in aqueous solution the peptide
lecular modelling to fit the active site of the adopts a relatively random, unordered struc-
enzyme (FREUNDet al., 1991b). Residues ture (VANDE VEN et al., 1991a, b; GOODMAN
et al., 1991; LIANet al., 1991, 1992; CHANet
al., 1992). Some degree of restraint is shown
in the structures of both ring B and the over-
lapping rings D and E, however, ring A and
C, several of the residues located at the N-
and C-termini of the peptide as well as those
in the central part show considerable confor-
mational freedom.
By contrast, in mixed, lipophilic solvents
such as TFE/water or DMSO the structure of
nisin becomes somewhat more stabilized and
shows that the peptide adopts overall a-heli-
Fig. 7. Stereorepresentation (backbone ribbon) of cal structure. Taken together, nisin can be
the solution structure of the type A lantibiotic galli- viewed as a pair of helical structures (residues
dermin; c.f. Fig. 4 for the primary structure. The 3-19 and 23-28), separated by a flexible hinge
dots represent sulphur atoms, and the N-terminus is region between amino acids 20-22 (VAN DE
at the right side.
VEN et al., 1991a, b; LIANet al., 1992); the re-
maining residues at the N-terminus (amino
Asnl8, Ala19, Ala21, and the C-terminal acids 1-2) and C-terminus (amino acids 29-
Cys(Avi) are oriented inward forming a hy- 34) are extremely flexible and could not be
drophilic core to an otherwise hydrophobic defined, even in the stabilizing, lipophilic so-
cage-like structure consisting of the overlap- lutions. In addition, while it is not so obvious
ping rings at the C-terminus which incorpo- in aqueous solution, nisin, like gallidermin, is
rate the outwardly oriented residues Phel7 amphiphilic. Furthermore, a study by GOOD-
and Tyrl8 (FREUNDet al., 1991a, b). MAN et al. (1991) of nisin in DMSO showed
340 8 Lantibiotics
that it formed a kinked rod-like structure

with an overall length of ca. 5 nm, diameter of
ca. 2 nm and calculated dipole moment of at
least 80 Debye, all features which are consis-
tent with its mode of action (see also Sect. 5).
3.1.3 Pep5
Although they are not yet complete, stud- Fig. 8. Stereorepresentation (backbone ribbon, sul-
ies of the lantibiotic Pep5 in solution suggest phur atoms represented as dots) of the type B lanti-
that it has a similar structure to the type A biotic duramycin B. For the primary structure see
lantibiotics gallidermin, nisin, and subtilin Fig. 6.
previously characterized (FREUND et al.,
1991c; FREUNDand JUNG,1992). Analysis of pect that their solution structures should be
the circular dichroism of Pep5 in water sug- somewhat similar. In addition, all of these
gests it forms a random, disordered structure lantibiotics contain a head-to-tail MeLan
while, on the addition of lipophilic solvents bridge which reduces the possibilities for con-
there is an increasing propensity toward for- formational freedom when compared to the
mation of helical elements. Similarly NMR type A lantibiotics (c.f. Figs. 4 and 6).
studies have shown that in aqueous solution The currently published structures of cin-
Pep5 has no defined structure except in the namycin, duramycin B, and duramycin C (Fig.
region of the four-membered ring B and that 8) as well as the preliminary structure of an-
the unbridged regions (amino acids 1-8) and covenin are all in general agreement as to the
the central region (amino acids 14-23) are overall structure of these lantibiotics (KESS-
particularly flexible and unstructured LER et al., 1991; ZIMMERMANN et al., 1993;
(FREUNDet al., 1991c; FREUND and JUNG, NISHIKAWAet al., 1988). The peptides are
1992). However, as with the CD experiments, bent into a U-shape by a turn induced by
NMR of Pep5 in 90% TFE shows that the Pro9 and stabilized by the three thioether
peptide is able to adopt an overall amphi- rings. In addition, the structure is further sta-
philic helical structure with most of the bilized by antiparallel P-sheets in the N- and
charged, hydrophilic amino acids oriented to C-terminal regions and the planar nature of
one face of the rod. In addition, the central the backbone is slightly distorted by the
region is not bridged however, the presence lysinoalanine ring between residues 6 and 19
of the two dehydroamino acids appear to stq- (ZIMMERMANN et al., 1993). Furthermore,
bilize local structure in this region (FREUND the amino acid exchanges have little or no in-
and JUNG,1992). fluence on the overall structures, but do ap-
pear to influence the degree of mobility of lo-
3.2 Type B Lantibiotics cal structural elements and the overall hydro-
phobicity of the peptides. Like the type A
3.2.1 Cinnamycin and the lantibiotics, the type B lantibiotics are highly
Duramycins amphiphilic with all of the hydrophobic ami-
no acids clustered into the bend of the “U”
In many respects the type B lantibiotics and the hydrophilic residues localized at the
cinnamycin, duramycin, duramycin B, dura- termini. Interestingly, the C-terminal region
mycin C, and ancovenin can be considered of cinnamycin has considerable structural
structural variants of one another (DE Vos et similarity to cyclo-TP5, a thymopoietin ana-
al., 1991) as they have the same principal log able to stimulate T-cell maturation
bridging pattern (apart from ancovenin which (KESSLERet al., 1991). The possibility that
lacks the lysinoalanine bridge) and amino cinnamycin could stimulate T-lymphocytes
acid exchanges are relatively conservative may help to explain the previous observation
(FREDENHAGEN et al., 1991; SAHL et al., that cinnamycin has in vivo anti-Herpes sim-
1995). Therefore, it is not unreasonable to ex- plex activity (WAKAMIYA et al., 1988).
4 The Genes/Proteins Involved in Lantibiotic Biosynthesis and Genetic Regulation 341
3.2.2 Mersacidin and Actagardine biosynthesis are shown for epidermin in Fig. 9
(see also Sect. 4.8).
So-far at least, the spatial structure of mer- In light of these observations, it is not sur-
sacidin has not been reported. However, pre- prising that following the discovery of the
liminary results for actagardine (ZIMMER- first specific genes encoding lantibiotic struc-
MANN and JUNG, unpublished data) show tural genes (SCHNELLet al., 1988; BUCH-
that it is a very rigid peptide, with well-de- MANN et al., 1988; BANERJEE and HANSEN,
fined structure, even though it lacks the head- 1988; KALETTA and ENTIAN,1989), subse-
to-tail bridge of the other type B lantibiotics. quent analysis of the DNA flanking these
In addition, the structure appears to have an genes has revealed the presence of a number
amphiphilic nature. of other open reading frames, often clustered
together, which seem to be involved in the
biosynthesis, genetic regulation and immunity
required during lantibiotic production. A
number of these lantibiotic-synthesizing gene
4 The GenedProteins clusters are compared in Fig. 1 0 wherever
possible, geneslgene products with similar
Involved in Lantibiotic function have been given the same letter de-
signation (DEVos et al., 1991) and the prefix
Biosynthesis and Genetic fanLan (lantibiotic-related) has been used to
Regulation indicate groups of genes (SAHLet al., 1995).
Thus, e.g., the general term fanA indicates all
lantibiotic structural genes while the general
4.1 The Requirement for Multiple term LanA indicates all pre-lantibiotics.
Gene Products in Lantibiotic
Biosynthesis 4.2 Structural Genes and
Lantibiotics are encoded by specific gene Pre-Lantibiotics
sequences and are synthesized on the ribo-
some hence, the prepeptides are limited to 4.2.1 The Location and Features
containing only the 20 amino acids allowed of the Structural Genes and their
for in the genetic code (SCHNELL et al., 1988;
SAHLet al., 1995; JACK et al., in press; JACK Transcription
and SAHL,1995). Therefore, in order to ma-
ture into the biologically active forms which Much debate has surrounded the location
have been isolated outside the producing of the structural gene encoding pre-nisin, with
cells, a number of posttranslational modifica- various research groups suggesting that it was
tion events must occur including: specific ami- located on either the chromosome or a plas-
no acid modifications, formation of intrapep- mid (KOZAKet al., 1974; TSAIand SANDINE,
tide thioether and/or aminoether rings, re- 1987; BUCHMANN et al., 1988; KALETTA and
moval of the leader peptide from the N-ter- ENTIAN, 1989; GIREESH et al., 1992). Howev-
minus and transport of the peptide out of the er, it is now clear that the entire operon re-
cell to the extracellular matrix. In addition to sponsible for nisin production, consisting of
these functions other controlled events are nisABTCIPRKFEG (KUIPERSet al., 1993a;
necessary for lantibiotic production, such as VAN DER MEERet al., 1993; ENGELKE et al.,
the regulation of their biosynthesis and the 1994; SIEGERS and ENTIAN, 1995) is encoded
generation of immunity mechanism(s) to pro- on a 70 kbp conjugative transposon, either
tect the producing cell from the action of its called Tn5301 (DODD et al., 1990, 1991;
own lantibiotic. Although it is not possible to HORNet al., 1991) or Tn5276 (RAUCHet al.,
state a definitive order of events, the general- 1990, 1991; RAUCHand DE Vos, 1992), de-
ized chain of reactions leading to lantibiotic pending on the strain of Lactococcus factis
342 8 Lantibiotics
&- cleavage site Pre-epidermin
* Site-specificdehydration (restricted to propeptide domain)

* Formation oflanthionine and 3-methyllanthioninenngs
* Reductive earboxylation of C-terminal cysteine residue
* Formation of uninovinylcysteine thioether nng
* Cleavage of leader peptide
Epidermin
Fig. 9. General schema for the events occurring during the biosynthesis of the type A lantibiotic epider-
min.
from which it was isolated. Furthermore, nisA and nisB may be co-transcribed and ap-
these authors have shown that these transpos- propriate sized transcripts have been identif-
ons also encode several of the genes necessary ied along with a terminator after nisB (STEEN
for sucrose metabolism in lactococci. There- et al., 1991; ENGELKEet al., 1992; KUIPERSet
fore, the recent observations that sucrose meal., 1993a). Little is known of the events sur-
tabolism-related genes can be found immedi- rounding transcription of the genes down-
ately proceeding the nisA BTCIPRKFEG stream of nisAB, however, a tandem promot-
gene cluster suggests that this represents all of er immediately preceding nisT could suggest
the genes necessary for the production of ni- that the remaining genes are transcribed as
sin (SIEGERSand ENTIAN, 1995). part of a polycistronic message (STEENet al.,
The nisin structural gene, nisA (Fig. lo), is 1991).
the first gene in the cluster and is located very The genes involved in the biosynthesis of
close to the 5' terminus of the transposon epidermin have also been studied extensively.
(DODDet al., 1990). BUCHMANet al. (1988) The gene cluster responsible for epidermin
identified a putative pindependent termina- production consists of epiT'TABCDQP and
tor immediately following the structural gene is carried on the 54 kbp plasmid pTu32
and subsequent studies have mapped the (SCHNELL et al., 1991,1992; AUGUSTIN et al.,
transcription start-site and identified an ap- 1991, 1992). The structural gene, epiA (Fig.
propriate promoter for transcription of nisA lo), is very likely transcribed along with
(KUIPERS et al., 1993a; ENGELKEet al., epiBCD, since epiB does not appear to pos-
1992). In addition, it has been suggested that sess its own promoter and transcripts of ap-
A B T C I P R K F E G
nir + I 1 1 3 x x x x n +
T“ T’A B C D Q P
epi 4 K Hw X I X K l-
T IA P B C
Pep + Hm n n F
lar * A M N
n;r
T UVP W
4
LI L2
-001
M
>I
T
r
P
>
Fig. 10. The arrangement of the gene clusters involved in the biosynthesis of the the lantibiotics nisin (nis),
-
0 1 2 3kb
subtilin (spa), epidermin (epi), Pep5 (pep), lactocin S (las),lactococcin DR (lcn), and the enterococcal
cytolysin/bacteriocin (cyl). In general, homologous genes have been given the same alphabetic suffix. Thus,
A-genes are the structural genes encoding the prepeptides (with the exception of spas), B, C, and M-genes
code for putative modification enzymes, T, E, and F-genes encode ABC superfamily translocator proteins,
P-genes are responsible for the production of leader peptidases, R and K-genes produce histidine kinasel
response regulator proteins (with the exception of epiQ which is equivalent to the R-genes of other gene
clusters), and I-genes encode specific proteins involved in producer self-protection. In addition, the func-
tion of the alternatively named genes in the lactocin S-producing gene cluster are unknown. The arrows
represent the relative direction of transcription of the respective genes.
propriate size for co-transcription (ca. 5 kb) located immediately preceding the -35 region
can be found (SCHNELLet al., 1992; AUGUS- (PESCHELet al., 1993).
TIN et al., 1992). Furthermore, these studies Alternatively, the subtilin structural gene,
have speculated that a putative terminator spas (Fig. lo), is located in the middle
between epiA and epiB may be “leaky”, al- of the subtilin-producing gene cluster
lowing some readthrough and thereby regu- spaBTCSIRKFG, preceded.by an appropriate
lating the levels of transcription of epiB, epic, transcriptional promoter (BANERJEE and
and epiD. In addition, the start-site for epiA HANSEN,1988). These authors were also able
transcription along with an appropriate pro- to map the transcriptional start-site and iden-
moter have been identified; the promoter is tify a 0.5 kb mRNA transcript corresponding
activated by the regulatory protein EpiQ to spas which had an unusually long half-life
which probably binds to an inverted repeat of approximately 45 min. In addition, a pair
344 8 Lantibiotics
of typical pindependent terminator struc- cin DR, lcnDRl (Fig. lo), was located on a 70
tures have been identified immediately pro- kbp plasmid in Lactococcus lactis ssp. lactis
ceeding the structural gene (BANERJEEand DR along with two other genes (RINCEet al.,
HANSEN,1988; HANSENet al., 1991). While 1994). Interestingly, this study showed that
transcriptional data for the genes downstream only lcnDRl and lcnDR2 were required for
of spas has not been reported, the identifica- the production of lactococcin DR, while the
tion of a transcriptional start-site for spaB 3rd gene lcnDR3 which has strong homology
(the first gene in the subtilin-synthesizing with transport proteins was dispensable. Simi-
gene cluster) has prompted speculation that lar genes to lcnDR2, designated lasM and
upstream genes are produced as a large poly- cylM have been located in the gene clusters
cistronic mRNA transcript (CHUNG and responsible for production of lactocin S and
HANSEN,1992; CHUNGet al., 1992; HANSEN, cytolysin/bacteriocin, respectively (SKAUGEN,
1993). 1994; GILMOREet al., 1994). In addition,
PepA, the structural gene for Pep5 (Fig. 10) lasM is surrounded by a number of additional
can be localized to the 18 kbp plasmid ORFs which do not appear to have homology
pED503, harbored within Staphylococcus epi- with known proteins, one (or more) of which
dermidis strain 5 (ERSFELD-DREBEN et al., may encode the novel enzyme responsible for
1984; KALETTAet al., 1989). The gene cluster catalyzing the stereospecific conversion of
responsible for Pep5 production consists of Dha to D-Ala (SKAUGEN,1994). Analysis of
pepTIAPBC, all of which, with the exception the DNA surrounding scnA, the structural
of the putative transporter gene (pepT), ap- gene encoding pre-streptococcin A-FF22, has
pear to be essential for Pep5 production and revealed that scnA is preceded by a putative
immunity (BIERBAUM et al., 1994; MEYERet terminator and proceeded by an inverted re-
al., 1995). Both pepl (responsible for immuni- peat which might act either as a transcription-
ty) and pepA are preceded by promoters, al terminator or as an mRNA processing site
however, while both can be transcribed inde- (HYNESet al., 1993).
pendently, Pep1 is not produced in the ab- So-far at least, there is a distinct paucity of
sence of pepA (REISet al., 1994). In addition, information concerning the genetic elements
REISet al. (1994) were able to identify a weak responsible for the production of type B lanti-
pindependent terminator immediately after biotics and only the structural genes for cin-
pepA and suggested that, analogous to the sit- namycin and mersacidin have yet been re-
uation observed in the epidermin-synthesiz- ported (ENTIANand KALETTA,1991; KA-
ing gene cluster, read-through may occur and LETTA et al., 1991a; BIERBAUM et al., 1995).
regulate downstream transcription. This view In the case of cinnamycin, the structural gene
would be further supported by the observa- (cinA) was identified, cloned and sequenced
tion that the proceeding ORF, pepP, appears from the chromosome of Streptoverticillium
to lack a promoter. Recent analysis of the griseoverticillatum (ENTIAN and KALETTA,
genes responsible for production of epilancin 1991; KALETTAet al., 1991a) while, the mer-
K7 have suggested that it also has a similar sacidin structural gene (mrsA) was identified,
arrangement of genes to that observed for the cloned and sequenced from the producing
Pep5-producing gene cluster except that no strain Bacillus subtilis HIL Y-85,54728
immunity gene (elk4 could be identified be- (BIERBAUM et al., 1995). In both cases it is
tween elkT and elkA, the structural gene apparent that the products (pre-cinnamycin
(VAN DE KAMPet al., 1995b). and pre-mersacidin) are quite different from
The structural genes for a number of other the structural gene products of the type A
lantibiotics including the bacteriocidcytolysin lantibiotics; both have extremely long leader
of enterococci, lactococcin DR (also called peptides with cleavage sites resembling those
lacticin 481), SA-FF22 and lactocin S have found in signal sequences for secreted pro-
also been identified and sequenced (GIL- teins. So-far at least, no information concern-
MORE et al, 1994; RINCEet al., 1994; PIARD ing additional genes involved in the biosyn-
et al., 1993; HYNESet al., 1993; SKAUGEN, thesis of the type B lantibiotics appears to
1994). The structural genes for pre-lactococ- have been reported.
4.2.2 The General Structure of Physical evidence for the structure of the
pre-type A lantibiotics has come from the iso-
Prepeptides lation of pre-Pep5 from the producing strain
Staphylococcus epidermidis strain 5 (WEILet
In general, the information currently avail- al., 1990). In this study, pre-Pep5 was isolated,
able concerning the structure of the lantibiot- albeit in very small quantities due to the ap-
ic prepeptides has been deduced from their parent raid turnover within the cell, and
gene sequences. Overall, lantibiotic prepep- shown to contain modified amino acids only
tides (Fig. 11) consist of two domains: a lead- within the propeptide domain, even though
er peptide and a propeptide domain, sur- the leader peptide contains hydroxyl amino
rounding an appropriate processing site. In acids which could potentially be dehydrated.
addition, the leader peptide domains are gen- Similar results have subsequently been ob-
erally acidic while the propeptide domains tained with pre-nisin (VAN DER MEER et al.,
are generally basic or neutral, both domains 1993). In addition, using a mutant S. epider-
have predicted helical propensity and are sep- midis strain 5 it has subsequently been possi-
arated by a predicted turn structure encom- ble to isolate prepeptides in different stages
passing the cleavage site which may allow the of dehydration, however, the leader peptide
processing protease access to this region alone was never detected, suggesting it is rap-
(JUNG, 1991a, b; JACK et al., 1995; SAHLet idly destroyed after cleavage (SAHLet al.,
al., 1995). Exceptions to these observations 1991).
can be found in the two type B lantibiotic Chemical synthesis of appropriate pre-lan-
prepeptide sequences so-far identified (EN- tibiotics has also provided useful insight into
TIAN and KALETTA, 1991; KALETTA et al., the structure of their fully unmodified form.
1991a; BIERBAUMet al., 1995); the leader Both the propeptide and leader peptide do-
peptides are considerably longer and do not mains (as well as segments of these) of galli-
contain the structural features identified for dermin and Pep5 have been synthesized, as
the type A lantibiotics (Fig. 11). well as the complete pre-gallidermin (BECK-
cleavlge site
Type A
pre-nisin MSTKDFNLDLVSVSKKDSGASPR
1- ITSISLCTPGCKTGCHCSIWSK
prc-subtilin MSKFDDFDLDWKVSKQDSKITPQ - WKSESLCTPGCVTWQTCFLQLTCNCKISK
prcPep5 MKNNKNLFDLEIKKETSQNTDELEPQ - TAGPAIRASVKQCQKTLKATRLFTVSCKGI(NGCK

pre-epidermin MEAVKEKNDLFNLWKVNAKLSNDSGAEPR - IASKFICTPGCAKTGSFNSYCC
pre-SA-FE2 NEKNNEVINSIQEVSLEELDQIIGA - GKNGVFKTISHECHLNTWAFIATCCS
pre-lactmccin DR MKEQNSFNLLQEVTESELDLILGA - KGGSGVIHTISHECNNNSWQRIFTCCS
prc-salivarichA MNAMKNSKDILNNAIEEVSEKELMEVAGG - KRGSGWIATITDDCPNSVFVCC
pre-laetocin S MKTEKKVLDELSLHRSWGARDVESSMNAD - STPVLASVAVSMEUPTASVLYSWAGCmSAKHHC
Type B
p-innunycin -
~ ~ I L ~ S W D A D F R A A L L E N P A A F G A S A A A L P T P V E R Q D a CRQSCSFGPETMCDGNTK
pre-masacidin MSQEAIIRSWKDPFSRENSTQNPAGNPFSELK!?AGXDKLVGAGm - CTFTLPGGGGVCTLTSECIC
Fig. 11. The primary structure of several representative lantibiotic prepeptides, i.e., the primary translation
product of the respective structural genes. The sequences are shown centered around the cleavage site for
leader peptide processing.
346 8 Lantibiotics
SICKINGERand JUNG, 1991). Circular di- might still direct the immature lantibiotic to
chroism analysis of these segments confirmed the identified transporters. Finally, the con-
that the leader peptide formed predominantly served properties of the lantibiotic leader
helical structures. In addition, both synthetic peptides might suggest that they contain spe-
pre-gallidermin and naturally isolated un- cific sequences or motifs which direct the bio-
modified pre-Pep5 showed stronger helical synthetic enzymes to the appropriate site for
propensity than either of their respective do- amino acid modification. Alternatively, they
mains alone, suggesting that the prepeptide may stabilize the conformation of the propep-
helix may be stabilized by interaction be- tide domain such that the biosynthetic ma-
tween the pro- and leader peptide domains chinery can access it and carry out specific
(BECK-SICKINGER and JUNG,1991; SAHLet posttranslational modifications (JUNG,
al., 1991). 1991a, b).
Similarly, both pre-nisin and a large num- Two recent studies may support this last
ber of their fragments have been synthesized proposal. Firstly, VAN DER MEER et al.
and studied (BYCROFTet al., 1991; SURoVOY (1994) have shown that a number of lantibiot-
et al., 1992). Initial results from the NMR ics have conserved motifs at particular posi-
analysis of pre-nisin in water suggests that it is tions among their respective leader peptides.
highly flexible with little or no preference for In this study, they identified the conserved re-
particular conformations (FREUND and sidues Phe-18, Asn or Asp-17, Leu-16, Asp
JUNG,1992). In addition, synthetic pre-nisin or Glu-15, Ser-10, Asp-7, Ser-6, Pro-2, and
has been shown to stoichiometrically bind Arg or Gln-1 and systematically set about al-
Zn2+ ions, an event which may act to stabil- tering these residues in the structural gene for
ize the conformation of the prepeptide and nisin by site-directed mutagenesis. Altera-
may have ramifications in other biosynthetic tions at Arg-1 resulted in failure of the pro-
reactions (SUROVOYet al., 1992). tease to cleave the leader peptide while,
somewhat surprisingly, alterations to the
highly conserved Pro-2 had no effect on
4.2.3 The Possible Role(s) of the either biosynthesis or processing. Similarly,
changing the Ala-4 resulted in production of
Leader Peptide unprocessed, mature nisin, suggesting that the
peptide no longer fit the active site of the
A number of roles for the leader peptide leader peptidase. Alternatively, exchanges
have been suggested and should be consid- with Asp-7 had little effect while those at
ered. Firstly, the leader peptide may serve to Ser-10 resulted in improved production of ni-
keep the prepeptide in an inactive form with- sin. However, exchanges in positions -6, -15,
in the cell. This view may be supported by the -16 and -18 could all be shown to completely
observations that leader peptide cleavage is inhibit nisin biosynthesis, suggesting that (at
generally the last step in lantibiotic biosynthe- least) these residues may be essential for the
sis and that fully modified but unprocessed biosynthesis of this lantibiotic.
pre-nisin and pre-pep5 are not biologically ac- Secondly, gene fusion of the nisin propep-
tive (WEILet al., 1990; SAHLet al., 1991; VAN tide domain to the leader domain for subtilin
DER MEERet al., 1994). Alternatively, it may production resulted in the production of fully
be that the leader peptide signals transport of modified but unprocessed nisin, suggesting
the lantibiotic out of the cell. However, the that the biosynthetic machinery of the cell
leader peptides of lantibiotics share little sim- (with the exception of the processing pro-
ilarity with the signal peptides of the sec-de- tease) recognized the hybrid protein as a suit-
pendent export systems (PUGSLEY,1993) and able substrate (KUIPERSet al., 1993b). Alter-
the identification of specific transport systems natively, a similar hybrid produced by fusion
of the ABC-superfamily (FATH and KOLTER, of the nisin leader domain to the subtilin pro-
1993) within lantibiotic-producing gene clus- peptide region was neither matured nor proc-
ters would suggest that sec-dependent trans- essed in Bacillus subtilis and only a leader
port is not used, although the leader peptide peptide domain which contained the first 7
amino acids of the subtilin leader fused to the to NisB, ENGELKE et al. (1992) were able to
remaining 17 amino acids from the nisin lead- demonstrate that NisB associates with the
er sequence was sufficient to get production membranes of fractionated nisin-producing
of subtilin (RINTALA et al., 1993). Taken to- cells, however they were not able to deter-
gether, these results seem to suggest that mine whether NisB was an integral mem-
there may be structural motifs in the leader brane protein or merely associated loosely
peptide which aid in creating a suitable sub- with this fraction. Similar results have subse-
strate for subsequent modifications. quently been obtained with SpaB analysis,
suggesting that the site of lantibiotic biosyn-
thesis might be organized and localized to the
cytoplasmic membrane (GUTOWSKI-ECKEL
4.3 Amino Acid Modifying et al., 1994).
Proteins The deduced size of the proteins produced
by the fanC genes (Fig. 10) is somewhat
4.3.1 LanB/LanC and LanM smaller at around 450 amino acids (SAHLet
al., 1995) and a number of conserved residues
Proteins can be identified within their sequences. Al-
though localization studies have not yet been
The most obvious and striking feature of reported, the LanC proteins appear to consist
the structure of the lantibiotics is their con- of regularly alternating regions of hydrophob-
tent of the non-protein amino acids lanthion- ic and hydrophilic nature (ENGELKE et al.,
ine and 3-methyllanthionine. Since neither 1992; GUTOWSKI-ECKEL et al., 1994).
codons nor tRNAs for these amino acids are Recently a number of lantibiotic-producing
found in the cell, they must arise by the action gene clusters have been identified which do
of specific enzymes acting on the precursor not contain either the lanB or lanC genes, in-
amino acids found in the prepeptide. Isola- cluding lactococcin DR (also known as lacti-
tion and genetic analysis of the gene clusters cin 481), lactocin S and the enterococcal cyto-
responsible for the production of a number of lysinhacteriocin (RINCEet al., 1994; PIARD
lantibiotics has revealed the presence of a et al., 1993; SKAUGEN, 1994; GILMORE et al.,
number of open reading frames; although the 1994). Interestingly, these gene clusters con-
role of most can be predicted by homology tain instead a different gene, designated lanM
with other proteins, several genes have been (Fig. lo), which share homology in their C-
identified which have no homology with pro- terminal region with the lunC genes. AI-
teins of known function and may be the gene though they do not appear to have any homo-
products responsible for the formation of the logy with the lunB gene products, it is possi-
novel amino acids found in the lantibiotics ble that the LanM proteins represent hybrid
(SCHNELL et al., 1991; AUGUSTIN et al., 1991; modifying proteins, able to carry out the func-
KALETTA et al., 1991). Indeed it has been tions of both LanB and LanC. This hypothe-
clearly shown by gene disruption and comple- sis may be further supported by the observa-
mentation of both spaB and spaC as well as tion that the leader peptides of the respective
with epiB and epic that lantibiotic biosynthe- lantibiotics synthesized by lanM-containing
sis is dependent on the production of these gene clusters are quite different to those syn-
proteins (KLEINet al., 1992). In addition, the thesized by lanB/LanC-containing operons
appearance of subtilin in the culture superna- (Fig. 10). What must still be investigated is
tant correlates directly with the expression of the molecular mechanism(s) involved in the
SpaB (GUTOWSKI-ECKEL et al., 1994). catalysis of dehydration and thioether ring
The lanB genes so-far characterized encode formation, regardless of whether or not it is
proteins of approximately lo00 amino acids carried out by a LanB/LanC or LanM system
(Fig. 10) which are principally hydrophilic, or even by some hitherto unidentified pro-
but which also have several hydrophobic re- tein.
gions which might indicate membrane span-
ning domains. Indeed, using antibodies raised
348 8 Lantibiotics
4.3.2 EpiD 10). Interestingly, all belong to the superfami-

ly of transport complexes known as the ATP-
The gene cluster responsible for the biosyn- binding cassette (ABC) transporters (HIG-
thesis of epidermin contains the gene epiD GINS,1992; FATHand KOLTER,1993). In gen-
(Fig. lo), which is not found in any of the oth- eral terms, ABC transporters consist of a
er lantibiotic-producing operons (AUGUSTIN dimeric complex consisting of two duplicated
et al., 1991, 1992; SCHNELL et al., 1991, 1992); domains, one domain of each monomer con-
similarly, only epidermin and its structural taining the cytoplasmic ATP-binding region
analog gallidermin possess the C-terminal characterized by a consensus Gly-Xaa-Gly-
modified amino acid (S)-[(Z)-2-aminovinyl]- Lys-Ser-Thr sequence (where Xaa indicates
D-cysteine (Figs. 1 and 4; ALLGAIERet al., any amino acid) and the other a membrane
1986 KELLNERet al., 1988). Furthermore, spanning region. In some cases, each domain
expression of epiD has been shown to be es- may be encoded by separate genes, although,
sential for epidermin production (SCHNELLet with the exception of epiT’T’’,all of the char-
al., 1991, 1992; AUGUSTINet al., 1991, 1992). acterized lantibiotic transporters are encoded
Subsequently, the protein EpiD has been ex- by a single gene encoding both domains
tensively studied and represents the first ex- (FATH and KOLTER, 1993; SAHL et al.,
ample of a novel enzyme responsible for ami- 1995).
no acid modification isolated from a lantibiot- In the case of epidermin production, it is
ic-producing strain. KUPKEet al. (1992) were not clear that epiT’T’’ is a functional translo-
able to overexpress EpiD as a fusion protein cator since, a frame shift would be required to
with maltose-binding protein (MBP) and re- transcribe both proteins. Furthermore, heter-
cover the EpiD after removal of the MBP ologous expression of epidermin in Staphylo-
analysis of EpiD revealed that is a 21 kDa fla- coccus carnosus suggests that transport is not
voprotein enzyme which requires the cofactor dependent on epiT’T’’ and that host-encoded
flavinmononucleotide (FMN) and is able to transport systems can efficiently substitute for
catalyze the oxidative decarboxylation of the this putative transport system (SCHNELLet
C-terminal cysteine residue of pre-epidermin al., 1992). Similarly, BIERBAUM et al. (1994)
(Fig. 12). Subsequently, both purified pre-epi- were able to show that disruption of pepT re-
dermin (KUPKE et al., 1993) and synthetic duced yields of Pep5 by about 90% compared
pro-epidermin have been used as substrates to the wild type. The observations of both of
to characterize the catalytic activity of this en- these studies indicate that the gene cluster-
zyme by mass spectrometry (KUPKEet al., encoded transporters in these staphylococci
1994). From this study it could be concluded may well be complemented, at least in part,
that EpiD is responsible for, at least, the oxi- by other cellular transporters. Similar results
dation of the C-terminal Cys residue of epi- have been obtained with other transporters of
dermin (the decarboxylation shown in Fig. 12 the ABC-superfamily, which are generally
might occur simultaneously and that, since thought to have some degree of flexibility in
EpiD acted both on pre-epidermin and the their substrate specificity (FATH and KOLT-
synthetic propeptide domain, its activity is ER, 1993). In direct contrast, interruption of
not “directed” via the leader peptide region the spaT gene results in abolition of the pro-
of the prepeptide form. duction of subtilin by Bacillus subtilis
(CHUNGet al., 1992; KLEIN and ENTIAN,
1994).
4.4 Transport Proteins
4.5 Leader Peptidases (LanP)
Genes which are likely to encode proteins
responsible for the transport of lantibiotics Three genes encoding putative peptidases
out of the cell (designated lanT genes) have (Fig. 10) which are probably responsible for
so-far been identified in each of the charac- the cleavage of the leader peptide of the re-
terized, lantibiotic-producing operons (Fig. spective lantibiotics have so far been se-
(CYS+W
H O
I II
[precpidermin(-30to +21)]-N
\@,/c-oH
C-H
I
YHZ
SH
I
H O
I II
[pre-epidermin(-30to +21))-N
\_7--OH
@ha+l9)
H O H
I II I
[pre-epidermin(-30to +18)] -N C- [praepidemin(+20to +21)]-N
\ / \(Q
CH
C
II II
CH2 CH
I
1
SH
Stereospecific addition
Fig. 12. Proposed pathway for
the formation of the aminovi-
nyl cysteine residue at the C- H O H
terminus of epidermin by the I II I
flavoprotein enzyme E ~ ~ D . [pre-epidemin(-30to +18)] -N C- [precpidemin(+20 to +21)]-N
The chirality of the respective \d
H-C
\m
CH
residues is indicated (where I II
appropriate) above the a-car- H-C S CH
I
bon atom. FMN, flavinmono- H
nucleotide; FMNH2, reduced
FMN. (S)-[(Z)-2-aminovinyl]-o-cysteine
quenced nisP (VANDER MEER et al., 1993; sponsible for cleavage of the leader peptide
ENGELKE et al., 1994), epiP (SCHNELLet al., of epilancin K7. Each of the proteins shares
1992), and pepP (MEYERet al., 1995), and homology with subtilisin-like serine pro-
VAN DE KAMPet al. (1995b) have identified teases, showing conserved active-site residues
the partial sequence of elkP, probably re- and a putative oxyanion hole.
350 8 Lantibiotics
NisP (Fig. 10) encodes a putative 74.7 kDa cific residues, that PepP is essential for pro-
protein which, when expressed in E. coli, can cessing of the immature Pep5 Similarly, the
be detected as a 54 kDa protein, consistent partial sequence of elkP suggests that epilan-
with processing of a prepropeptide to gener- cin K7 processing is carried out inside the
ate a mature protease (VAN DER MEERet al., cell, since the putative peptidase lacks a pre-
1993). Furthermore, NisP contains a C-termi- prosequence required for export (VAN DEN
nal extension not found on most other pro- KAMP et al., 1995b).
teases of this class and which could serve as a
membrane anchor, suggesting that NisP is
both exported from the cell and anchored to 4.6 Proteins Regulating Lantibiotic
the outside of the cytoplasmic membrane. Biosynthesis
This observation would suggest that cleavage
of the nisin leader peptide should be the ulti- Except in the cases of subtilin and mersa-
mate step in maturation, a conclusion con- cidin production, lantibiotics are generally
firmed since modified but unprocessed nisin produced most abundantly during logarithmic
can be found in the culture media after growth when energy sources are at their max-
growth of a NisP-deficient mutant of a nisin- imum, suggesting constitutive production as
producing Lactococcus lactis (VAN DER opposed to regulated biosynthesis (SAHLand
MEER et al., 1993). In addition, this study BRANDIS, 1981; HORNER et al., 1990, DE
showed that NisP produced in E. coli was VUYST and VANDAMME, 1991; JACK and
able to cleave this accumulated unprocessed TAGG, 1992); however, this may not be the
nisin to release the mature, biologically active case since the lantibiotic-producing gene clus-
form. ters appear to encode specific regulatory ele-
The putative protease required for process- ments (Fig. 10). Like the transport proteins
ing of epidermin (EpiP) is also encoded with- which are members of a larger superfamily of
in the epidermin-synthesizing gene cluster transport proteins, so too the gene products
(Fig. 10) and is a 461 amino acid protein with responsible for the regulation of lantibiotic
calculated mass of 51 kDa (SCHNELL et al., biosynthesis appear to part of a large group of
1992). Interestingly, heterologous expression proteins with similar function, that of the two-
of epidermin in Staphylococcus carnosus is component response/regulatory elements. In
not dependent on the presence of a functional general (MSADEKet al., 1993), these regula-
epiP gene, suggesting that host-encoded pro- tory elements are made up of two proteins
tease(s) are capable of substituting for EpiP the first of which is a membrane-bound his-
(AUGUSTIN et al., 1992). Furthermore, no tidine kinase able to respond to an extracellu-
equivalent protease has been found in the lar signal by autophosphorylation of a specific
subtilin-synthesizing gene cluster, suggesting histidine residue in its cytoplasmic domain.
that an intrinsic Bacillus subrilis-encoded pro- Subsequently, the phosphate group is trans-
tease is able to process immature subtilin ferred to the second component, an intracel-
(SAHLet al., 1995). lular regulatory protein which is usually a
By contrast, pepP (Fig. lo), encoding the transcriptional activator, able to bind DNA.
leader peptidase involved in Pep5 maturation, Genes with homology to these two-compo-
encodes a smaller protease which is devoid of nent regulators have been found in the gene
a preprosequence and is therefore not ex- clusters encoding biosynthesis of both nisin
ported from the cell (MEYERet al., 1995). and subtilin (KLEINet al., 1993; KUIPERSet
Thus the site of leader peptide cleavage for al., 1993a; VAN DER MEER et al., 1993; EN-
Pep5 must be intracellular and is therefore GELKE et al., 1994). Both the insertional inac-
not the ultimate step in lantibiotic maturation tivation of spaRK followed by complementa-
in Staphylococcus epidermidis 5. In addition tion of the sapR (KLEINet al., 1993) as well as
PepP lacks some of the conserved residues deletion of nisR (VAN DER MEERet al., 1993)
typical of subtilisin-like serine proteases how- have been used to demonstrate that both of
ever, this study showed, using both gene dis- these genes are essential for regulation of
ruption and site-directed mutagenesis of spe- subtilin and nisin biosynthesis, respectively.
However, so-far at least, none of the respec- elements are responding? In the case of nisin,
tive proteins have been studied in detail, nor DE VUYST and VANDAMME (1991, 1992)
have potential activator binding regions been have suggested that nisin production may be
identified in the gene clusters. associated with carbon source control, since
In contrast, considerable study of the regul- nisin production and sucrose metabolism phe-
ation of epidermin production, both at the notypes are both encoded on the same trans-
DNA and protein level, has been carried out. poson and that high phosphate levels could
The gene cluster responsible for epidermin stimulate nisin biosynthesis (DE VUYSTand
biosynthesis (Fig. 10) encodes EpiQ, which VANDAMME, 1993). However, HORNERet al.
appears to be equivalent to the regulator ele- (1990) demonstrated that high phosphate was
ment and shares significant homology with detrimental for production and that the car-
both SpaR and NisR (AUGUSTIN et al., 1992; bon source had not effect on biosynthesis of
SCHNELLet al., 1992; PESCHELet al., 1993). both epidermin and Pep5 Thus, the signal re-
While a respective histidine kinase has not mains enigmatic however, lantibiotic biosyn-
been identified, overexpression of EpiQ leads thesis seems, in general, programmed to oc-
to increased epidermin production and the cur when there is an abundance of energy-
gene appears to be essential for heterologous providing substrate in the growth medium.
expression of epidermin in Staphylococcus
curnosus, suggesting that this heterologous
host may provide a suitable response ele-
ment. In addition, the purified EpiQ has also 4.7 Producer Self-Protection
been studied; EpiQ is a 25 kDa DNA-binding
transcriptional activator able to bind to one
(“Immunity”) Mechanisms
or more of several putative operator sites up-
stream of the epiA promoter (PESCHELet al., Clearly, lantibiotic synthesizing cells are
1993). producing substances which are potentially
Interestingly, no response/regulator gene detrimental to their own continued well-be-
pair has yet been identified associated with ing, however, such strains generally show a
the gene cluster responsible for Pep5 biosyn- high degree of resistance to the action of their
thesis which is carried on the plasmid own lantibiotic. Recent studies have shown
pED503 in Stuphylococcus epidermidis 5 that this resistance is provided by specific
(ERSFELD-DREBEN et al., 1984; MEYERet al., producer self-protection (immunity) mecha-
1995). However, a vector containing the nisms. A number of early studies using co-
genes epiABCDQ, which are insufficient for elimination or co-transfer demonstrated that
epidermin production in Staphylococcus cur- lantibiotic production and immunity were
nosus, converted a strain of S. epidermidis 5 linked (ERSFELD-DREBEN et al., 1984; GAS-
devoid of pED503 into an epidermin produc- SON,1984; TAGGand WANNAMAKER, 1978).
er (AUGUSTIN, 1991). This gene cluster is in- More recently, the genes encoding specific
sufficient for epidermin production in S. cur- immunity-related polypeptides have been
nosus since epiQ is transcribed from the same identified (Fig. 10) at least in the producers of
promoter as one of the deleted genes (epiP; nisin, subtilin, and Pep5 (KUIPERSet al.,
Fig. 10). In addition, a chromosomal fragment 1993a; KLEIN and ENTIAN,1994; REIS and
isolated from S. epidermidis 5 and containing SAHL,1991; REIS et al., 1994). Interestingly,
a response/regulator could complement this these studies have shown that, while there is a
vector and give epidermin production in S. great deal of similarity between Spa1 and
curnosus (AUGUSTIN,1991). Thus, it may be NisI, the producing strains do not share cross
that since these chromosomal genes from S. immunity with each other, nor are they im-
epidermidis 5 can regulate heterologous epi- mune to the effects of Pep5 Indeed, cross im-
dermin production, they may also be capable munity between lantibiotic producers has
of directing the biosynthesis of Pep5 only been demonstrated where the lantibiot-
What remains unclear is: What is the exter- ics produced can be considered natural var-
nal “signal” to which the response regulatory iants of one another, such as in the case of
352 8 Lantibiotics
nisin A and nisin Z or epidermin and Potential additional mechanism(s) involved

[VallIleI-epidermin (DE Vos et al., 1993; in immunity to lantibiotics have been identi-
SAHLet al., 1995). fied in both nisin-producing and subtilin-pro-
PepI, the immunity protein responsible for ducing bacterial strains. Recently, SIEGERS
protection against the action of Pep5, is a 69 and ENTIAN (1995) have identified the genes
amino acid peptide with a strongly hydro- nisE, nisF and nisG in Lactococcus lactis 6F3
philic C-terminal region and a strongly hydro- and shown that disruption of these genes
phobic N-terminal domain; expression of leads to increased sensitivity of the producing
PepI of which is absolutely essential for the strain to exogenous nisin. Analysis of the
immune phenotype (REIS and SAHL, 1991; gene sequences obtained suggest that NisE
REIS et al., 1994). These studies also showed and NisF together form another ABC-trans-
that production of the immune phenotype is port system but, this transporter is apparently
dependent on co-expression of both pepA not involved in the transport of nisin during
and pepZ and that even partial deletions in maturation. In addition, NisENsF shares ho-
pepA, the structural gene, resulted in com- mology with McbEIMcbF, the transporter re-
plete abolition of immunity, although the rea- ported to be responsible for immunity to mi-
sons for this are not apparent. In addition, crocin B17 in certain strains of E. coli (GAR-
since PepI appears to be accessible to exoge- RIDO et al., 1988). NisG on the other hand is
nously added proteases, it is likely to be lo- an hydrophobic protein with significant simi-
cated outside the membrane, perhaps loosely larities to many of the colicin immunity pro-
associated with it (REISet al., 1994). Further- teins, which are thought to interact directly
more, in this study it was observed that the with the channel-forming colicins (PUGSLEY,
membranes of cells expressing PepI were not 1988; SONGand CRAMER,1991).
depolarized by PepS, suggesting that PepI Similar results have also been obtained
and Pep5 might interact with PepI preventing with analysis of subtilin immunity determi-
formation of transmembrane pores. However, nants (except that the NisE/NisF homolog in
analysis of the possible in vitro interactions Bacillus subtilis is called spaF/SpaG), prompt-
between synthetic PepI and Pep5 was unable ing speculation that the E/F and G-proteins
to demonstrate any direct interaction be- of these two strains are involved in providing
tween these two peptides (SUROVOYand an additional degree of protection to that
JUNG,unpublished data; SAHLet al., 1995). generated by the immunity proteins NisI and
In contrast to PepI, the immunity proteins SpaI (KLEINand ENTIAN,1994; SIEGERSand
NisI and SpaI (Fig. 6) are considerably larger ENTIAN,1995). However, how these proteins
at 245 and 163 amino acids, respectively might achieve this is anything but clear. Ob-
(KUIPERSet al., 1993a; ENGELKEet al., servations of similarities between NisENsF
1994). In addition, both are typical of bacteri- and the microcin B17 immunity proteins
al lipoproteins, possessing both an N-terminal McbE/McbF further confuse the issue be-
signal sequence and a membrane-anchoring cause this bacteriocinkolicin apparently has
Cys residue immediately proceeding the an intracellular target, specifically inhibiting
cleavage site. Moreover, both have relatively DNA gyrase (GARRIDOet al., 1988). Howev-
hydrophilic C-terminal domains, suggesting er, the primary mode of action of nisin and
that this region might be exposed outside the subtilin involves depolarization of the cyto-
cytoplasmic membrane. At least in the case of plasmic membrane through the formation of
NisI, partially immune phenotypes are possi- voltage-dependent pores (see also Sect. 5).
ble, however, as was observed for PepI, full Thus, some authors have speculated that the
immunity to nisin was only achieved when additional transport systems might act either
both nisA and nisZ were co-expressed (KUIP- to transport external nisin into the cell for de-
ERS et al., 1992; REISand SAHL,1991; REIS et gradation or transport internalized subtilin
al., 1994). In addition, expression of NisI in E. out of the cell, thus generating the observed
coli with disrupted outer membranes pro- additional degree of immunity (KLEIN and
tected the cells from the action of exogenous ENTIAN,1994; SIEGERSand ENTIAN,1995).
nisin (KUIPERSet al., 1993a). In any case, it is now apparent that multiple
systems are involved in self-protection of 1991). Thus, it could be concluded that the
Lactococcus lactis and Bacillus subtilis against primary translation product has a very short
nisin and subtilin, respectively. So-far no re- half-life, site-specific dehydration is the first
ports of such additional immunity mecha- step in biosynthesis and that thioether rings
nism(s) have been given for other type A lan- are formed in a separate step. In the case of
tibiotic-producing strains. nisin biosynthesis VAN DER MEER et al.
(1993) showed that nisin secretion is the pen-
ultimate step in processing and that the last
4.8 The Chain of Events Leading step involves cleavage of the leader peptide.
However, this sequence of events may not
to Lantibiotic Biosynthesis and hold for all lantibiotics since Pep5 (at least)
Maturation appears to be processed inside the cell with
transport the ultimate step in production of
From an historical point of view it is inter- this lantibiotic (MEYERet al., 1995).
esting to note that even at the same time as Several modification reactions are likely to
the structures of nisin and subtilin were being occur spontaneously, including the formation
determined (GROSS and MORRELL, 1971; of the N-terminal2-oxopyruvyl and 2-oxobut-
GROSSet al., 1973) other researchers were yryl groups of lactocin S and Peps, respective-
proposing that these lantibiotics were synthe- ly (MORTVEDT et al., 1991; SKAUGEN et al.,
sized by posttranslational modification. Ini- 1994; KELLNERet al., 1989). These reactions
tially, HURST(1966) showed that nisin bio- would require the removal of the leader pep-
synthesis was prevented by inhibitors of pro- tide and should therefore proceed processing.
tein synthesis. On the basis of the incorpora- Similarly, since the N-terminal hydroxypyru-
tion of radiolabeled amino acids INGRAM vyl group at the N-terminus of epilancin K7
(1969, 1970) suggested that didehydroamino probably requires enzymatic reduction, this
acids should result from dehydration of Ser reaction must proceed processing and may
and Thr and that the thioether rings might ar- constitute one of the last steps in biosynthesis
ise from addition of the sulphydryl groups of of this lantibiotic (VAN DE KAMP et al.,
Cys residues. Later, NISHIOet al. (1983) used 1995a, b; SAHL et al., 1995). Alternatively,
anti-subtilin antibodies to isolate precursor since EpiD is able to act on the C-terminal
subtilin and then used cell extracts from sub- Cys residue of both pre-epidermin, synthetic
tilin-producing B. subtilis to convert this pre- pro-epidermin, and other synthetic peptides
cursor into a substance with the same activity not related to epidermin, oxidative decarb-
and electrophoretic mobility as natural subtil- oxylation may well precede thioether ring
in. However, the isolation of the structural formation during Cys(Avi) biosynthesis
genes encoding the precursor peptides for (SCHNELLet al., 1988 KUPKEet al., 1992,
epidermin, nisin, and subtilin provided the fi- 1993,1994, 1995).
nal evidence that both didehydroamino acid
and thioether rings arose from modifications
to genetically encoded Ser, Thr, and Cys resi-
dues (SCHNELLet al., 1988; BANERJEEand
HANSEN,1988; BUCHMANN et al., 1988; KAL-
5 Biological Activities of
ETTA and ENTIAN,1989). Lantibiotics
Subsequently the isolation of pre-peptides
of Pep5 and nisin has shed a great deal of
light on the sequence and subcellular localiza- 5.1 Type A Lantibiotics
tion of modification events required for lanti-
biotic biosynthesis. Cytoplasmic pre-Pep5 has 5.1.1 Primary Mode of Action
been shown to consist of a mixture of dehy-
drated forms (6-fold dehydrated, 5-fold dehy- As earlier mentioned, the type A lantibiot-
drated, etc.) which do not contain L a d ics have been defined as those lantibiotic pep-
MeLan rings (WEILet al., 1990, SAHLet al., tides whose mode of action is principally di-
354 8 Lantibiotics
rected towards the killing of bacterial cells ment of sensitive cells resulted in the cessa-
(JUNG,1991a, b). In general their activity is tion of most macromolecular biosyntheses
confined to gram-positive bacteria. However, (e.g., protein, DNA, RNA, and polysaccha-
provided that the outer membrane is first dis- rides) and that nisin activity was dependent
rupted, several type A lantibiotics have been on external factors such as pH, temperature
shown to also act against a number of gram- and the phase of growth of the target cells
negative bacteria, including E. coli and Sul- (HURST, 1981; SAHL and BRANDIS,1981;
monellu spp. (STEVENSet al., 1991). Thus, it SAHL, 1991). Based on these studies, subse-
would appear that the lipid-rich outer mem- quent analysis of the mechanism of action of
brane protects gram-negative bacteria from the type A lantibiotics has revealed much
type A lantibiotic action, probably by pre- about the way in which they exert their an-
venting access to the inner membrane, the timicrobial activity.
site of their action. In addition, the range of Treatment of susceptible bacterial cells
bacteria affected by the different type A lanti- with micromolar concentrations of type A
biotics varies considerably. Some, such as sali- lantibiotics such as nisin, Pep5, subtilin, epi-
varicin A exhibit a relatively limited antimi- dermin, gallidermin, or streptococcin A-FF22
crobial spectrum (Ross et al., 1993), while arrests amino acid uptake and induces rapid
others such as nisin inhibit a broad range of efflux of preaccumulated amino acids (SAHL
gram-positive bacteria (e.g., many strains of and BRANDIS,1983; RUHRand SAHL,1985;
micrococci, streptococci, lactococci, pediococ- SCHULLERet al., 1989; SAHL,1991; BENZet
ci, staphylococci, lactobacilli, Listeriu spp., al., 1991; JACKet al., 1994b) and, at least in
and mycobacteria) as well as both the vegeta- some cases, has also been shown to cause the
tive cells and spores of Buciffusspp. and Cfos- efflux of the potassium analog R b + (SAHL
tridium spp. (HURST, 1981; DELVES- and BRANDIS,1983; RUHR and SAHL,1985).
BROUGHTON,1990, MOLITOR and SAHL, In addition, following Pep5-treatment of cells,
1991; BIERBAUM and SAHL, 1993; DE VUYST ATP could be found in the external medium
and VANDAMME, 1993). (SAHL and BRANDIS,1983); since there are
Historically, it is interesting that while nisin no known transport systems for ATP, these
has been used in biopreservation for more results, along with the observed efflux of oth-
than 35 years, the mechanism by which it and er low molecular-weight intracellular macro-
other type A lantibiotics kill bacterial cells molecules, suggest that type A lantibiotics
has only recently been clarified (HURST, form discrete pores in the cytoplasmic mem-
1982; MOLITORand SAHL,1991; JACKet al., brane of susceptible bacteria. This mechanism
in press; SAHLet al., 1995). In early studies, is in contrast to the generalized disruption
RAMSEIER(1960) suggested that nisin might that might be expected from the action of a
be acting as a detergent both because of its surfactant. In addition, both ATP and R b +
highly basic pZ and the observation that nisin efflux as well as the arrest of macromolecule
treatment of cells induced leakage of UV-ab- biosynthesis is effected by the growth phase
sorbing intracellular constituents. Later, of the target cells (SAHLand BRANDIS,1983;
GROSSand MORRELL(1971) determined the SAHL, 1991; JACKet al., 1994b), suggesting
structure of nisin and observed that the unsat- that the pore formation occurs in an energy-
urated amino acids Dha and Dhb found in ni- dependent manner.
sin might be able to interact with the sulphy- Further evidence for the mode of action of
dry1 groups of specific enzymes within a cell, type A lantibiotics come from experiments
while other studies suggested that nisin could with cytoplasmic membrane vesicles. Artifi-
interfere with the biosynthesis of the bacterial cially-energized vesicles, which have been
cell wall (LINNETTand STROMINGER, 1973; treated with type A lantibiotics, rapidly efflux
REISINGER et al., 1980). Subsequently, it has preaccumulated radiolabeled amino acids; by
been suggested that type A lantibiotics such contrast, pre-treatment of the vesicles (prior
as nisin kill cells by interfering with energy to energization) induced little or no efflux un-
transduction (SAHL, 1985); this conclusion til after the cells were sufficiently energized
was based on the observation that nisin treat- (SAHL,1985; RUHRand SAHL,1985; SAHLet
al., 1987; KORDEL et al., 1988; SCHULLERet studies have also shown that nisin and Pep5
al., 1989; JACK et al., 1994b). These results form pores only in the presence of a truns-ne-
further demonstrate that type A lantibiotics gative membrane potential, while subtilin,
form transmembrane pores in an energy-de- epidermin, gallidermin, and streptococcin
pendent fashion and allow efflux of preaccu- A-FF22 form pores irrespective of the orien-
mulated intracellular components. However, tation of the applied potential. Analysis of the
these same studies also showed that lantibiot- current voltage curves obtained from these
ic treatment of either whole cells, artificial studies also allowed determination of the
vesicles which had been energized with vali- threshold potential for lantibiotic-induced
nomycin-induced potassium gradients or ar- pore formation; epidermin and gallidermin
tificially energized liposomes (GAO et al., required ca. 50 mV, Pep5, nisin and subtilin
1991; ABEEet al., 1991) resulted in dissipa- require ca. 80 mV, while streptococcin
tion of the membrane potential, suggesting A-FF22 required ca. 100 mV.
that the pores formed are non-specific and al- Further physical properties, including the
low influx of extracellularly accumulated pro- mean diameter and the lifetime of type A lan-
tons and (probably) other ions and small mol- tibiotic channels can be determined from the
ecules. The dissipation of the membrane po- analysis of single channels formed in BLMs.
tential accounts for the observed arrest of en- If the pore is assumed to be a cylinder with a
ergy-dependent macromolecular biosynthesis length equivalent to the thickness of the
and is different to that of the protonophores membrane which is filled with the same solu-
since pore formation is energy-dependent, tion as bathes the membrane (of known con-
potentiated by the membrane potential ductance), then from the observed conduc-
(SAHL, 1991; GARCIA-GARCERA et al., tance it is possible to calculate the diameter
1993). of the pore. Streptococcin A-FF22 pores ap-
Further support for the prediction that type pear to be relatively unstable, appearing as
A lantibiotics kill cells by disruption of ener- millisecond time scale “bursts” and a with a
gy transduction through the formation of en- mean diameter of about 0.5-0.6 nm (JACKet
ergy-dependent pores in the cytoplasmic al., 1994b). Alternatively, nisin and Pep5
membrane have been provided by analysis of pores are somewhat more stable (tens to
the pores formed in artificial bilayers such as hundreds of milliseconds) and have a diam-
the black-lipid membranes (BLM); in addi- eter of ca. 1 nm (SAHLet al., 1987; KORDEL
tion these studies have allowed the measure- et al., 1988), while subtilin pores are larger
ment of several of the physical properties of again at ca. 2 nm diameter (SCHULLERet al.,
the formed pores (BENZ et al., 1978, 1991). 1989). Epidermin and gallidermin pores are
Artificial BLMs can be formed across a small somewhat different; their mean lifetimes may
whole separating two chambers or wells in a extend up to 30 s and their diameter appears
teflon block, both of which are filled with a to increase with the applied potential (BENZ
conducting salt buffer solution. Using elec- et al., 1991).
trodes, it is then possible to apply a defined The determination of many of the physical
potential difference across an artificial bilayer properties of the type A lantibiotic peptides
and measure lantibiotic-induced current flows has allowed some understanding of how they
through the normally insulative model mem- might form pores in the bacterial cytoplasmic
brane. In such a system, the type A lantibiot- membrane. In general, type A lantibiotics
ics can be shown to induce discrete pores have both sufficient length and a sufficiently
since, application of a sufficiently high poten- high dipole moment to be consistent with
tial induces current flow; reduction of the po- voltage-dependent channel formation in
tential allows the pores to close and restores phospholipid bilayers (FREUND et al.,
the insulative properties of the bilayer, indi- 1991a b; BENZet al., 1991) as well as the cen-
cating that membrane disruption is not gener- tral flexible region thought to be essential for
alized (SAHL et al., 1987; KORDELet al., stabilization of the transmembrane structure
1988; SCHULLERet al., 1989; BENZ et al., (VOGELet al., 1993). In addition, NMR stud-
1991; JACK et al., 1994b). In addition, these ies have shown that the peptides themselves
356 8 Lantibiotics
appear to form amphiphilic helical conforma- cation of a sufficiently high membrane poten-
tions in solution and present their hydro- tial the peptides apparently adopt a trans-
phobic residues on one "face" of the helix membrane orientation and create a pore.
and the charged, hydrophilic residues on the However, a number of questions remain
opposite "face" (FREUNDet al., 1991a, b, c; unanswered; e.g., it remains to be seen wheth-
GOODMANet al., 1991; LIAN et al., 1991; er the peptides aggregate before or after ad-
PALMERet al., 1989; SLIJPER et al., 1989; VAN option of the transmembrane orientation,
DE VEN et al., 1991a, b). Clearly, a single lan- how many pores are the minimum required
tibiotic molecule is insufficient to induce pore for transmembrane conductance and which
formation and multiple peptides must some- terminus of the peptide inserts through the
how coalesce in the bilayer to form an aggre- membrane. In addition, the observations that
gate with the properties observed above. In pore diameters fluctuate and the short pore
addition, since type A lantibiotics can form lifetimes observed with most type A lantibiot-
pores in artificial vesicles and bilayers, it is ics suggest that the number of peptide in-
clear they have no requirements for specific volved in pore formation is not static, but
membrane-associated "receptors" as has been rather more dynamic.
suggested for several other channel-forming
antibacterial peptides including lactococcin A
and pediocin PA-1 (VAN BELKUMet al., 5.1.2 Secondary Mode of Action
1991; CHIKINDAS et al., 1993).
Although the exact mechanism remains un- In addition to forming pores in phospholip-
clear, current models (Fig. 13) for type A lan- id bilayers, both Pep5 and nisin have been
tibiotic-induced pore formation suggest that shown to induce autolysis in Staphylococcus
the peptides accumulate at the cytoplasmic simuluns cells (BIERBAUM and SAHL,1991).
membrane, perhaps attracted to the bilayer Characterization of the mechanism by which
by ionic interactions (SAHL, 1991; BENZ et they achieve this suggests that cationic lanti-
al., 1991; JACK et al., 1994b). In the absence biotics such as nisin and Pep5 are able to
of a membrane potential they should remain competitively release cell wall autolytic en-
oriented lateral to the membrane but appar- zymes normally bound to, and regulated by,
ently with their hydrophobic "face" intimate- polyanionic cell wall constituents such as lipo-
ly associated with the bilayer (SCHULLERet teichoic-, teichoic-, and teichuronic acids,
al., 1989; SAHL,1991; BENZet al., 1991; JACK probably by an ion exchange-like mechanism
et al., 1994b; DRIESSEN et al., 1995); on appli- (SAHL, 1985; BIERBAUMand SAHL, 1987,
Fig. 13. Model for the mode of

action of the type A lantibiot-
ics, involving the formation of
voltage-dependent pores in the
cytoplasmic membrane of sen-
sitive bacteria.
1988, 1991). In addition, it has been shown diates also failed to be incorporated into
that activation of the autolytic enzymes oc- cross-linked peptidoglycan in the actagardine-
curred most markedly in the area of the septa treated bacilli and isolation of a labeled inter-
between dividing daughter cells (BIERBAUM mediate (UDP-acetylmuramyl-N-acetylglu-
and SAHL,1985,1991) and that the activation cosamine pentapeptide linked to a C55-iso-
was not specific since similar results could be prenylphosphate carrier) suggested that acta-
obtained using synthetic cationic peptides, gardine treatment was able to inhibit the
such as poly-lysine or poly-arginine (BIER- transfer of cell wall precursor to the peptido-
BAUM and SAHL,1991). Since only small so- glycan receptor during cell wall biosynthesis.
lutes may pass outward (Fig. 13) pores When staphylococci are incubated in the
formed by these lantibiotics in the cytoplas- presence of mersacidin their growth is first in-
mic membrane should allow an influx of wa- hibited, following which the cells begin to
ter into the cell, increasing osmotic pressure. lyse; cessation of growth is accompanied by
This increase in intracellular pressure com- an inability to incorporate D-alanine into the
bined with the interference in energy trans- cell wall and by a marked decrease in the
duction (resulting from depolarization of the thickness of the cell wall (BROTZet al., 1995).
membrane) which should prevent repair of In addition, these authors observed that mer-
the weakened cell wall, probably results in sacidin treatment had no effect on other ma-
the observed lysis (BIERBAUMand SAHL, cromolecular biosynthetic processes such as
1991). DNA, RNA, and protein biosynthesis, in di-
rect contrast to the effects observed with sim-
ilar cells treated with type A lantibiotics. Fur-
5.2 Type B Lantibiotics thermore, whereas both mersacidin and the
glycopeptide antibiotic vancomycin have ap-
proximately the same MIC and both act on
5.2.1 Mersacidin and Actagardine cell wall biosynthesis, mersacidin was not in-
hibited by the tripeptide L-Lys-D-Ala-D-Ala
The type B lantibiotics mersacidin and ac- which is a potent inhibitor of vancomycin ac-
tagardine also kill bacterial cells rather effi- tion. This last result suggests that, while the
ciently, however, their primary mechanism of molecular target of mersacidin is unclear, it
activity appears to be vastly different to that differs from that of vancomycin.
described above for the type A lantibiotics,
directed primarily at the cell wall rather than
the cytoplasmic membrane. Mersacidin acts 5.2.2 Cinnamycin and the
principally against streptococci and staphylo-
cocci; this particular type B lantibiotic has de- Duramycins
monstrated in vivo activity against methicil-
lin-resistant Staphylococcus aureus strains Among other type B lantibiotics, at least ci-
and could therefore offer an alternative to namycin and duramycin have been shown to
vancomycin treatment of such infections inhibit the growth of Bacillus spp.; treated
(CHATERJEEet al., 1992b). Alternatively, ac- cells show increased membrane permeability,
tagardine is primarily effective against obli- undergo a reduction in ATP-dependent pro-
gate anaerobes and streptococci and has been tein translocation and ATP-dependent cal-
used experimentally to treat Streptococcus cium uptake, show marked slowing in the rate
pyogenes infections (ARIOLI et al., 1976; of chloride uptake and show significantly re-
MALABARBA et al., 1990). duced potassium and sodium ATPase activi-
In the late 1970s, SOMMAet al. (1977) ties (RACKERet al., 1983; STONEet al., 1984;
showed that low concentrations of actagar- NAVARRO et al., 1985; CHENand TAI, 1987).
dine inhibited the incorporation of N-acetyl- Duramycin-resistant bacilli remain unaffected
glucosamine and L-alanine into Bacillus sub- and have been shown to possess cytoplasmic
tilis cell wall peptidoglycan. In these experi- membranes with altered phospholipid compo-
ments it was observed that other interme- sitions; whereas sensitive cells contain phos-
358 8 Lantibiotics
phatidyl ethanolamine, resistant cells do not of phospholipase A2, an important enzyme of

(NAVARROet al., 1985; DUNKLEYet al., the immune system involved in the produc-
1988). These results suggest that duramycin tion of prostaglandins and leucotrienes
recognizes and interacts with specific phos- (MARKIand FRANSON,1986; FREDENHAGEN
pholipids in the cytoplasmic membrane of et al., 1990, 1991; MARKI et al., 1991). The
sensitive cells and that alterations to this com- principal substrate of phospholipase A2 is
position, while conferring duramycin resist- phosphatidyl ethanolamine; interaction of
ance, do not seem to effect other membrane- these type B lantibiotics with the substrate
associated cellular activities (DUNKLEY et al., renders it unavailable for lipolysis, thus inter-
1988; CLEJANet al., 1989). fering with immune function.
In addition to inhibiting the growth of ba-
cilli, duramycin also inhibits a number of me-
tabolic properties of isolated mitochondria
and specifically increases the permeability of
the inner membrane, the major component of 6 Applications of
which is phosphatidyl ethanolamine (SOKO-
LOVE et al., 1989). Similarly, cinnamycin has Lantibiotics
been shown to induce hemolysis in isolated
erythrocytes and phosphatidyl ethanolamine- 6.1 Applications as a
containing liposomes; cinnamycin had no ef-
fect on erythrocytes if it was first incubated Food/Beverage Preservative
with phosphatidyl ethanolamine or on lipo-
somes which were prepared from phosphati- Since the type A lantibiotic nisin has been
dyl inositol, phosphatidyl serine, or cardioli- used as a biopreservative for at least 35 years,
pin (CHOUNGet al., 1988a, b). The latter re- there is currently a great deal of information
sults suggest that cinnamycin interacts with regarding its application to foods and bever-
phospholipids containing both a glyceryl ages for which readers are referred to the ex-
backbone and a primary amino group. Using cellent reviews of HURST(1981), DELVES-
electron-spin resonance, these studies also BROUGHTON(1990), MOLITOR and SAHL
showed that the interaction of cinnamycin (1991), and DE VUYST and VANDAMME
and phosphatidyl ethanolamine results in (1993). However, it is worth briefly mention-
reorganization of the membrane bilayer. Tak- ing that already this peptide has been utilized
en together, these results have further con- in a number of areas including: fish and low-
firmed the direct interaction of type B lanti- temperature processed canning, brewing,
biotics with specific phospholipids present in winemaking, cheese and other fermented dai-
the membrane bilayer. ry product manufacture as well as in the pre-
Recently, SHETH et al. (1992) have sug- servation of cosmetics and deodorants. Nisin
gested that the increases in membrane per- has proved particularly useful in a number of
meability associated with duramycin treat- these areas because of its characteristics. Ni-
ment could result from the formation of ion sin, e.g., demonstrates good activity against a
channels. They observed that both the mem- number of food-borne pathogens and spoil-
branes of cultured epithelial cells and artifi- age bacteria including Clostridium botulinum,
cial BLMs treated with duramycin showed Listeria monocytogenes, Lactobacillus spp.
complex conductance states, some of which and Leuconostoc spp. In addition, the peptide
were discrete, and that pores were weakly an- is particularly heat and acid stable making it
ion selective. In addition, this study showed particularly suitable for addition to products
that pore diameter increased with time, per- with low pH (e.g., fermented products) and
haps explaining duramycin-induced increases products which subsequently undergo some
in membrane permeability. sort of pasteurization.
Duramycin, duramycin B, duramycin C,
and cinnamycin may also act as immunoregu-
lators as they are able to influence the activity
7 Conclusions and Future Perspectives 359
6.2 MedicaVParamedical and the threat of these potentially fatal microor-

ganisms (BROTZ et al., 1995).
Veterinary Applications
Interestingly, some of the lantibiotics may
also prove to be useful in medicinal prepara-
tions for both human and animal application. 7 Conclusions and Future
The type A lantibiotics epidermin and galli-
dermin have previously been shown to be ac- Perspectives
tive against Propionibacterium acnes, the cau-
sative agent of the acnes disease. These ob- From the biotechnological standpoint then,
servations have led to speculation that these the lantibiotics represent an exciting class of
lantibiotics might be used in the external peptides for two reasons. Firstly, as antimi-
treatment of acnes as a replacement therapy crobial compounds they may offer not only
for the currently used erythromycinhitamin new opportunities in areas such as food and
A creams. Such application of a lantibiotic beverage preservation (HURST, 1981;
should have several advantages: (1) so-far at DELVES-BROUGHTON, 1990; MOLITORand
least, acquired resistance to these lantibiotics SAHL, 1991; DE VUYST and VANDAMME,
has not been observed, (2) because of their 1993), but may also represent (at least in the
size and nature, the peptides should not be case of mersacidin and gallidermin) novel
absorbed into the body and they should be of therapeutic agents with potential for medici-
low toxicity, and (3) it should be possible to nal use (BROTZ et al., 1995; SAHLet al.,
develop large-scale, low cost production 1995). Secondly, it is clear that the lantibiotics
schemes for these bacterially-derived pep- are produced by novel biosynthetic mecha-
tides (ALLGAIER et al., 1991; JUNG, 1991a, b; nisms; the enzymes which are capable of
UNGERMANN et al., 1991). Similarly, the not- transforming the 20 “protein” amino acids
able acid stability of nisin (HURST, 1981) has into such highly modified forms may well
led several recent studies to suggest that it prove useful for the generation of new pep-
may have efficacy in the treatment of gastric tide compounds and mimetics, previously
ulcers, due to its antimicrobial activity against only produced through time-consuming and
Helicobacter pylori (DE VUYST and VAN- costly in-lab chemical syntheses (SAHLet al.,
DAMME, 1993). In addition, alternative sug- 1995; JACK et al., in press; JACK and SAHL,
gested applications of nisin include: as a 1995).
mouthrinse for the prevention of gingivitis Thus, greater understanding of the biosyn-
and plaque or as a germicidal preparation for thetic mechanisms involved in lantibiotic pro-
the prevention of mastitis in dairy cattle duction may one day allow us to introduce
(SEARSet al., 1992; HOWELLet al., 1993). non-peptide amino acids and enantiomeric
More significantly perhaps, the cinnamycin- amino acids into useful proteins, perhaps im-
group type B lantibiotics have been shown to proving their stability, resistance to proteolyt-
effect human immune function, leucocyte ic activity or even their biological and catalyt-
proliferation and may play a role in protec- ic activity. The continued study of novel pep-
tion against Herpes simplex virus (MARKI and tides and biosynthetic enzymes, such as those
FRANSON, 1986; FREDENHAGEN et al., 1990, involved in lantibiotic production, give the
1991; MARKI et al., 1991); thus the potentials highest prospects for the development of
for application of these compounds in human new, useful, peptide-based compounds
health care are obvious. In addition, the ob- through biotechnology.
servation that mersacidin displays antibacter-
ial activity against methicillin-resistant Sta-
phylococcus aureus strains as well as multi-
drug resistant enterococci by novel mecha-
nisms offers the greatest hope for the devel-
opment of new therapeutic agents to combat
360 8 Lantibiotics
8 References Escherichia coli, Biochim. Biophys. Acta 511,

305-319.
BENZ,R., JUNG,G., SAHL,H.-G. (1991), Mecha-
ABEE,T., GAO,F. H., KONINGS, W. N. (1991), The nism of channel formation by lantibiotics in
mechanism of the lantibiotic nisin in artificial black lipid membranes, in: Nisin and Novel Lan-
membranes, in: Nisin and Novel Lantibiotics tibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 359-
(JUNG,G., SAHL, H.-G., Eds.), pp. 373-385. 372. Leiden: ESCOM Scientific Publishers BV.
Leiden: ESCOM Scientific Publishers BV. BERRIDGE,N. J., NEWTON,G. G., ABRAHAM, E.
ALLGAIER,H., JUNG, G., WERNER, R. G., P. (1952), Purification and nature of the antibiot-
SCHNEIDER, U., ZAHNER,H. (1986), Epider- ic nisin, Biochem. J. 52, 529-535.
min: sequencing of a heterodet tetracyclic 21- BIERBAUM, G., SAHL,H.-G. (1985), Induction of
peptide amide antibiotic, Eur. J. Biochem. 160, autolysis of staphylococci by the basic peptide
9-22. antibiotics Pep5 and nisin and their influence on
ALLGAIER,H., WALTER, J., SCHLUTER,M., the activity of autolytic enzymes, Arch. Micro-
WERNER,R. G. (1991), Strategy for the purifica- biol. 141, 249-254.
tion of lantibiotics, in: Nisin and Novel Lanti- BIERBAUM, G., SAHL,H.-G. (1987), Autolytic sys-
biotics (JUNG,G., SAHL,H.-G., Eds.), pp. 422- tem of Staphylococcus simulans 22: influence of
433. Leiden: ESCOM Scientific Publishers BV. cationic peptides on activity of N-acetylmura-
ARIOLI,V., BERTI,M., SILVESTRI, L. G. (1976), moyl-L-alanine amidase, J. Bacteriol. 169, 5452-
Gardimycin, a new antibiotic from Actinoplanes. 5458.
111. Biological properties, J. Antibiot. 29, 511- BIERBAUM, G., SAHL,H.-G. (1988), Influence of
515. cationic peptides on the activity of the autolytic
AUGUSTIN, J. (1991), Identifizierung und Regula- endo-PN-acetylglucosaminidaseof Staphylococ-
tion der Biosynthesegene des Lantibiotikums cus simulans 22, FEMS Microbiol. Rev. 58, 223-
Epidermin aus Staphylococcus aureus Tii 3298. 228.
PhD Thesis, University of Tiibingen, Germany.
AUGUSTIN,J., ROSENSTEIN,R., KUPKE, T., BIERBAUM, G., SAHL,H.-G. (1991), Induction of
SCHNEIDER, U., SCHNELL,N., ENGELKE,G., autolysis of Staphylococcus simulans 22 by Pep5
ENTIAN,K.-D., GOTZ, F. (1991), Identification and nisin and influence of the cationic peptides
of epidermin biosynthetic genes by complemen- on the activity of the autolytic enzymes, in: Nisin
tation studies and heterologous expression, in: and Novel Lantibiotics (JUNG,G., SAHL,H.-G.,
Nisin and Novel Lantibiotics (JUNG,G., SAHL, Eds.), pp. 386-396. Leiden: ESCOM Scientific
H.-G., Eds.), pp. 277-286. Leiden: ESCOM Publishers BV.
Scientific Publishers BV. BIERBAUM, G., SAHL,H.-G. (1993), Lantibiotics -
AUGUSTIN, J., ROSENSTEIN, R., WIELAND,B., unusually modified bacteriocin-like peptides
SCHNEIDER, U., SCHNELL,N., ENGELKE,G., from Gram-positive bacteria, Zentralbl. Bakter-
ENTAIN,K.-D., GOTZ, F. (1992), Genetic analy- iol. 278, 1-22.
sis of epidermin biosynthetic genes and epider- BIERBAUM, G., REIS,M., SZEKAT,C., SAHL,H.-G.
min-negative mutants of Staphylococcus epider- (1994), Construction of an expression system for
midis, Eur. J. Biochem. 204,1149-1154. engineering of the lantibiotic Peps, Appl. Envi-
BANERJEE, S., HANSEN, J. N. (1988), Structure and ron. Microbiol. 60, 4332-4338.
expression of a gene encoding the precursor of BIERBAUM, G., BROTZ,H., KOLLER,K.-P., SAHL,
subtilin, a small peptide antibiotic, J. Biol. H.-G. (1995), Cloning, sequencing and produc-
Chem. 263,9508-9514. tion of the lantibiotic mersacidin, FEMS Micro-
BECK-SICKINGER, A. G., JUNG,G. (1991), Synthe- biol. Lett. U7,121-126.
sis and conformational analysis of lantibiotic BROCK,T. D., DAVIE,J. M. (1963), Probable iden-
leader-, pro- and pre-peptides, in: Nisin and tity of a group D hemolysin with a bacteriocine,
Novel Lantibiotics (JUNG, G., SAHL, H.-G., J. Bacteriol. 86, 708-712.
Eds.), pp. 218-230. Leiden: ESCOM Scientific BROTZ, H., BIERBAUM, G., MARKUS,A., MOLI-
Publishers BV. TOR,E., SAHL,H.-G. (1995), Mode of action of
BENEDICT, R. G., DVONCH, W., SHOTWELL, 0. L., mersacidin - inhibition of peptidoglycan synthe-
PRIDHAM, T. G., LINDENFELSER, L. A. (1952), sis via a novel mechanism? Antimicrob. Agents
Cinnamycin, an antibiotic from Streptomyces Chemother. 39,714-719.
cinnamonensis nov. sp, Antibiot. Chemother. 2, BUCHMAN, W. B., BANERJEE,S., HANSEN,J. N.
591-594. (1988), Structure, expression and evolution of a
BENZ, R., JANKO,K., Boos, W., LAUGER,P. gene encoding the precursor of nisin, a small
(1978), Formation of large ion-permeable mem- protein antibiotic, J. Biol. Chem. 263, 16260-
brane Channels by the matrix protein (porin) of 16266.
8 References 361
BYCROFT,B. W., CHAN,W. C., ROBERTS,G. C. K. ISHITSUKA, H., INOUE,K. (1988b), Interaction
(1991), Synthesis of pro- and prepeptides related of a cyclic peptide, Ro 09-0198, with phosphati-
to nisin and subtilin, in: Nisin and Novel Lanti- dylethanolamine in liposomal membranes, Bio-
biotics (JUNG,G., SAHL,H.-G., Eds.), pp. 204- chim. Biophys. Acta 940, 180-187.
217. Leiden: ESCOM Scientific Publishers BV. CHUNG,Y.-J., HANSEN,J. N. (1992), Determina-
CHAN,W. C., LIAN,L.-Y., BYCROFT,B. W., Ro- tion of the sequence of spaE and identification
BERTS, G. c. K. (1989a), Isolation and charac- of a promoter in the subtilin (spa) operon in Ba-
terization of two degradation products derived cillus subtilis, J. Bacteriol. 174, 66994702.
from the peptide antibiotic nisin, FEBS Lett. CHUNG,Y.-J., STEEN,M. T., HANSEN, J. N. (1992),
252,29-36. The subtilin gene of Bacillus subtilis ATCC 6633
CHAN,W. C., LIAN,L.-Y., BYCROFT,B. W., Ro- is encoded in an operon that contains a homo-
BERTS,G. c. K. (1989b), Confirmation of the logue of the hemolysin B transport protein, J.
complete structure of nisin by complete 'H- Bacteriol. 174, 1417-1422.
NMR resonance assignments in aqueous and di- CLEJAN,S., GUFFANTI,A. A., COHEN, M. A.,
methylsulphoxide solution, J. Chem. SOC. Perkin KRULWICH, T. A. (1989), Mutations of Bacillus
Trans. 1,2359-2367. jirmus OF4 to duramycin resistance results in
CHAN,W. C., BYCROFT,B. W., LEYLAND, M. L., substantial replacement of membrane lipid phos-
LIAN,L.-Y., YANG,J. C., ROBERTS,G. C. K. phatidylethanolamine by its plasmalogen form,
(1992), Sequence-specific resonance assignment J. Bacteriol. 171, 1744-1746.
and conformational analysis of subtilin by 2D- DELVES-BROUGHTON, J. (1990), Nisin and its uses
NMR, FEBS Lett. 300,56-62. as a food preservative, Food Technol. 44, 100-
CHAN,W. C., BYCROFT,B. W., LEYLAND, M. L., 112.
LIAN,L.-Y., YANG,J. C., ROBERTS,G. C. K. DEMPSTER, R. P., TAGG,J. R. (1982), The produc-
(1993), A novel posttranslational modification of tion of bacteriocin-like substances by the oral
the peptide antibiotic subtilin: isolation and bacterium Streptococcus salivarius, Arch. Oral
characterization of a natural variant from Bacil- Biol. 27, 151-157.
lus subtilis ATCC 6633, Biochem. J. 291,23-27. DE Vos, W. M., JUNG,G., SAHL,H.-G. (1991), Ap-
CHATERJEE,S., CHATERJEE,S., LAD, S. J., PHAN- pendix: definitions and nomenclature of lanti-
SALKAR, M. S., RUPP, R. H., GANGULI, B. N. biotics, in: Nisin and Novel Lantibiotics (JUNG,
FEHLHABER, H.-W., KOGLER,H. (1992a), Mer- G., SAHL, H.-G., Eds.), pp. 457463. Leiden:
sacidin, a new antibiotic from Bacillus: fermenta- ESCOM Scientific Publishers BV.
tion, isolation, purification and chemical charac- DE VOS, W. M., MULDERS, J. W. M., SIEZEN,R. J.,
terization, J. Antibiot. 45,832-838. HUGENHOLTZ, J., KUIPERS,0. P. (1993), Prop-
CHATERJEE,S., CHATERJEE,D. K., JANI,R. H., erties of nisin Z and distribution of its gene nisZ
BLUMBACH, J., GANGULI,B. N. KLESEL,N., in Lactococcus lactis, Appl. Environ. Microbiol.
LIMBERT,M., SEIBERT,G. (1992b), Mersacidin, 59,213-218.
a new antibiotic from Bacillus: in vitro and in DE VUYST,L., VANDAMME, E. J. (1991), Microbial
vivo antibacterial activity, J. Antibiot. 45, 839- manipulation of nisin biosynthesis and fermenta-
845. tion, in: Nisin and Novel Lantibiotics (JUNG,G.,
CHEN,L. L., TAI,P. C. (1987), Effects of antibiotics SAHL, H.-G., Eds.), pp. 397-409. Leiden:
and other inhibitors on ATP-dependent protein ESCOM Scientific Publishers BV.
translocation into membrane vesicles, J. Bacteri- DE VUYST,L., VANDAMME, E. J. (1992), Influence
01. 169,2372-2379. of the carbon source on nisin production in Lac-
CHIKINDAS,M. L., GARCIA-GARCERA, M. J., tococcus Iactis subspp. lactis batch fermentations,
DRIESSEN,A. J. M., LEDERBOER,A. M., NI- J. Gen. Microbiol. 138, 571-578.
SEEN-MEYER, J., NES, I. F., ABEE,T., KONING, DE VUYST,L., VANDAMME, E. J. (1993), Nisin, a
W. N., VENEMA,G. (1993), Pediocin PA-1, a lantibiotic produced by Lactococcus Iactis sub-
bacteriocin from Pediococcus acidilactici spp. lactis: properties, biosynthesis, fermentation
PAC1.O forms hydrophilic pores in the cytoplas- and applications, in: Bacteriocins of Lactic Acid
mic membrane of target cells, Appl. Environ. Bacteria: Microbiology, Genetics and Applica-
Microbiol. 59, 3577-3584. tions (DEVUYST,L., VANDAMME, E., Eds.), pp.
CHOUNG,S.-Y., KOBAYASHI, T., INOUE,J., TAKE- 151-221. London: Chapman and Hall.
MOTO,K., ISHITSUKA,H., INOUE,K. (1988a), DODD, H. M, GASSON,M. J. (1994), Bacteriocins
Haemolytic activity of a cyclic peptide Ro 09- of lactic acid bacteria, in: Genetics and Biotech-
0198 isolated from Streptoverticillium, Biochim. nology of Lactic Acid Bacteria (GASSON,M. J.,
Biophys. Acta 940, 171-179. DE Vos, W. M., Eds.), pp. 211-251. London:
CHOUNG,S.-Y., KOBAYASHI, T., TAKEMOTO, K., Chapman and Hall.
362 8 Lantibiotics
DODD,H. M., HORN,N., GASSON,M. J. (1990), structural revision of duramycin and cinnamycin,
Analysis of the genetic determinant for produc- in: Nisin and Novel Lantibiotics (JUNG, G.,
tion of the peptide antibiotic nisin, J. Gen. Mi- SAHL, H.-G., Eds.), pp. 131-140. Leiden:
crobiol. 136,555-566. ESCOM Scientific Publishers BV.
DODD,H. M., HORN,N., SWINDELL, S., GASSON, FREUND, S., JUNG,G. (1992), Lantibiotics: an over-
M. J. (1991), Physical and genetic analysis of the view and conformational studies on gallidermin
chromosomally located transposon Tn5301, re- and Pep5, in: Bacteriocins, Microcins and Lanti-
sponsible for nisin biosynthesis, in: Nisin and biotics (JAMES,R., LAZDUNSKI, C., PAT~US, F.,
Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 75-92. Berlin: Springer-Verlag.
Eds.), pp. 231-242. Leiden: ESCOM Scientific FREUND,S., JUNG,G., GUTBROD,O., FOLKERS,
Publishers BV. G., GIBBONS, W. A., ALLGAIER, H., WERNER,
DRIESSEN,A. J. M., VAN DEN HOOVEN,H. w., R. (1991a), The solution structure of the lanti-
KUIPER,W., VAN DE KAMP,M. M., SAHL, H.- biotic gallidermin, Biopolymers 31, 803-811.
G., KONINGS, R. N. H, KONINGS, W. N. (1995), FREUND,S., JUNG,G., GUTBROD,O., FOLKERS,
Mechanistic studies of lantibiotic-induced per- G., GIBBONS, W. A. (1991b), The three-dimen-
meabilization of phospholipid vesicles, Biochem- sional solution structure of gallidermin deter-
istry 34, 1606-1614. mined by NMR-based molecular graphics, in:
DUNKLEY, E. A., JR., CLEJAN,S., GUFFANTI,A. Nisin and Novel Lantibiotics (JUNG,G., SAHL,
A., KRULWICH, T. A. (1988), Large decreases in H.-G., Eds.), pp. 91-102. Leiden: ESCOM Scien-
membrane phosphatidylethanolamine and di- tific Publishers BV.
phosphatidylglycerol upon mutation to duramy- FREUND,S., JUNG,G., GIBBONS, W. A., SAHL,H.-
cin resistance do not change the protonophore G. (1991c), NMR and circular dichroism studies
resistance of Bacillus subtilis, Biochim. Biophys. on Pep5, in: Nisin and Novel Lantibiotics (JUNG,
Acta 943,13-18. G., SAHL,H.-G., Eds.), pp. 103-112. Leiden:
ENGELKE,G., GUTOWSKI-ECKEL, Z., HAMMEL- ESCOM Scientific Publishers BV.
MA", M., ENTIAN,K.-D. (1992), Biosynthesis GAO,T., ABEE,T., KONINGS, W. N. (1991), Mech-
of the lantibiotic nisin: genomic organization and anism of the peptide antibiotic nisin in lipo-
membrane localization of the NisB protein, somes and cytochrome c oxidase-containing pro-
Appl. Environ. Microbiol. 58, 3730-3743. teoliposomes, Appl. Environ. Microbiol. 57,
2164-2170.
ENGELKE, G., GUTOWSKI-ECKEL, Z., KIESAU,P., GARCIA-GARCERA, M. G. J., ELFERINK, M. G. L.,
SIEGERS,K., HAMMELMANN, M., ENTIAN,K.- DRIESSEN, A. J. M., KONINGS, W. N. (1993), In
D. (1994), Regulation of nisin biosynthesis and vitro pore-forming activity of the lantibiotic ni-
immuntiy in Lactococcus lactis 6F3, Appl. Envi- sin: role of proton motive force and lipid compo-
ron. Microbiol. 60, 814-825. sition, Eur. J. Biochem. 212,417422.
ENTIAN,K.-D., KALEITA,C. (1991), Isolation and GARRIDO,M. C., HERRERO,M., KOLTER,R.,
characterization of the cinnamycin structural MORRENO,P. (1988), The export of the DNA
gene, in: Nisin and Novel Lantibiotics (JUNG,G., replication inhibitor microcin B17 provides im-
SAHL, H.-G., Eds.), pp. 303-308. Leiden: munity for the host cell, EMBO J. 7, 1853-
ESCOM Scientific Publishers BV. 1862.
ERSFELD-DREBEN, H., SAHL,H.-G., BRANDIS, H. GASSON, M. J. (1984), Transfer of sucrose ferment-
(1984), Plasmid involvement in production of ing ability, nisin resistance and nisin production
and immunity to the staphylococcin-likepeptide into Lactococcus Iactis 712, FEMS Microbiol.
Pep5, J. Gen. Microbiol. 130,3029-3035. Lett. 21, 7-10.
FATH,M. J., KOLTER,R. (1993), ABC transport- GILMORE,M. S., SEGARRA, R. A., BOOTH,M. C.,
ers: bacterial exporters, Microbiol. Rev. 57, 995- BOGIE,C. P., HALL, L. R.,CLEWELL,D. B.
1017. (1994), Genetic structure of the Enterococcus
FREDENHAGEN, A., FENDRICH,G., MARKI,F., faecalis plasmid pAD1-encoded cytolytic toxin
MARKI,W., GRUNER,J., RASCHDORF, F., PE- system and its relationship to lantibiotic deter-
TER, H. H. (1990), Duramycins B and C, two minants, J. Bacteriol. 176, 7335-7344.
new lanthionine-containing antibiotics as inhibi- GIREESH,T., DAVIDSON, B. E., HILLIER,A. J.
tors of phospholipase A2, J. Antibiot. 43, 1403- (1992), Conjugal transfer in Lactococcus lactis of
1412. a 68-kilobase-pair chromosomal fragment con-
FREDENHAGEN, A., MARKI, F., FENDRICH,G., taining the structural gene for the peptide bacte-
MARKI,W., GRUNER,J., VAN OOSTRUM, J., riocin nisin, Appl. Environ. Microbiol. 58, 1670-
RASCHDORF, F., PETER,H. H. (1991), Duramy- 1676.
cin B and C, two new lanthionine-containing an- GOODMAN, M., PALMER,D. E., MIERKE,D., Ro,
tibiotics as inhibitors of phospholipase A2 and S., NUNAMI,K., WAKAMIYA, T., FUKASE, K.,
8 References 363
HORIMOTO,S., KITAZAWA,M., FUJITA, H., HYNES,W. L., FERREITI,J. J., TAGG,J. R. (1993),
KUBO, A., SHIBA,T. (1991), Conformations of Cloning of the gene encoding streptococcin A-
nisin and its fragments using synthesis, NMR FF22, a novel lantibiotic produced by Strepto-
and computer simulations, in: Nisin and Novel coccus pyogenes and determination of its nucleo-
Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. tide sequence, Appl. Environ. Microbiol. 59,
59-75. Leiden: ESCOM Scientific Publishers 1969-1 971.
BV. INGRAM,L. (1969), Synthesis of the antibiotic ni-
GROSS, E. (1977), a,PUnsaturated and related sin: formation of lanthionine and p-methyllan-
amino acids in peptides and proteins, Adv. Exp. thionine, Biochim. Biophys. Acta 184, 216-219.
Med. Biol. 86, 131-153. INGRAM,L. (1970), A ribosomal mechanism for
GROSS,E., MORELL,J. L. (1971), The structure of synthesis of peptides related to nisin, Biochim.
nisin, J. Am. Chem. Soc. 93, 4634-4635. Biophys. Acta 224, 263-265.
GROSS,E., KILTZ,H. H., NEBELIN, E. (1973), Sub- JACK,R. W., SAHL,H.-G. (1995), Unique postrans-
tilin. VI. Die Struktur des Subtilins. Hoppe-Sey- lational modifications involved in lantibiotic bio-
ler’s 2. Physiol. Chem. 354, 810-812. synthesis, Trends Biotechnol. 13,269-278.
GUTOWSKI-ECKEL, G., KLEIN,C., SIEGERS,K., JACK,R. W., TAGG,J. R. (1991), Isolation and par-
BOHM,K., HAMMELMANN, M., ENTIAN,K.-D. tial structure of streptococcin A-FF22, in: Nisin
(1994), Growth phase-dependent regulation and and Novel Lantibiotics (JUNG,G., SAHL,H.-G.,
membrane localization of SpaB, a protein in- Eds.), pp. 171-179. Leiden: ESCOM Scientific
volved in biosynthesis of the lantibiotic subtilin, Publishers BV.
Appl. Environ. Microbiol. 60, 1-11. JACK,R. W., TAGG,J. R. (1992), Factors affecting
HANSEN,J. N. (1993), Antibiotics synthesized by the production of the group A Streptococcus
posttranslational modification, Ann. Rev. Micro- bacteriocin SA-FF22, J. Med. Microbiol. 36,132-
biol. 47, 535-564. 138.
HANSEN,J. N., CHUNG,Y. J., LIU,W., STEEN,M. JACK,R. W., CARNE,A., METZGER,J., STEVA-
T. (1991), Biosynthesis and mechanism of action N O V I ~ ,S., SAHL,H.-G., JUNG,G., TAGG,J. R.
of nisin and subtilin, in: Nisin and Novel Lanti- (1994a), Elucidation of the structure of SA-
biotics (JUNG,G., SAHL,H.-G., Eds.), pp. 287- FF22, a lanthionine-containingantibacterial pep-
302. Leiden: ESCOM Scientific Publishers BV. tide produced by Streptococcus pyogenes strain
HAYASHI, F., NAGASHIMA,K., TERUI,Y., KAWA- FF22, Eur. J. Biochem. 220,455-462.
MURA, Y., MATSUMOTO,K., ITAZAKI, H. JACK,R. W., BENZ,R., TAGG,J. R., SAHL,H.-G.
(1990), The structure of PA48009: the revised (1994b), The mode of action of SA-FF22 a lanti-
structure of duramycin, J. Antibiot. 43, 1421- biotic isolated from Streptococcus pyogenes
1430. strain FF22, Eur. J. Biochem. 219, 699-705.
HIGGINS, C. F. (1992), ABC-transporters: from mi- JACK,R. W., TAGG,J. R., RAY,B. (1995), Bacteri-
croorganisms to man, Ann. Rev. Cell Biol. 8, 67- ocins of Gram-positive bacteria, Microbiol. Rev.
113. 59,171-200.
HORN,N., SWINDELL, S., DODD,H. M., GASSON, JACK,R. W., BIERBAUM, G., HEIDRICH, C., SAHL,
M. J. (1991), Nisin biosynthesis genes are en- H.-G. (in press), The genetics of lantibiotic bio-
coded by a novel conjugative transposon, Mol. synthesis, BioEssays.
Gen. Genet. 228, 129-135. JANSEN, E. F., HIRSCHMANN, D. J. (1944), Subtilin
HORNER, T., UNGERMANN, V., ZAHNER, H., - an antibacterial product of Bacillus subtilis:
FIEDLER,H.-P., UTZ, R., KELLNER,R., JUNG, culturing conditions and properties, Arch. Bio-
G. (1990), Comparative studies on the fermenta- chern. 4,297-304.
tive production of lantibiotics by staphylococci, JUNG,G. (1991a), Lantibiotics - ribosomally syn-
Appl. Microbiol. Biotechnol. 32, 511-517. thesized biologically active polypeptides con-
HOWELL,T. H., FIORELLINI, J. P., BLACKBURN, taining sulphide rings and a$-didehydroamino
P., PROJAN,S. J., DE LA HARPE,J., WILLIAMS, acids, Angew. Chem. (Int. Edn. Engl.) 30,1051-
R. C. (1993), The effect of a mouthrinse based 1068.
on nisin, a bacteriocin, on developing plaque JUNG,G. (1991b), Lantibiotics: a survey, in: Nisin
and gingivitis in beagle dogs, J. Clin. Periodont- and Novel Lantibiotics (JUNG,G., SAHL,H.-G.,
01. 20,335-339. Eds.), pp. 1-34. Leiden: ESCOM Scientific Pub-
HURST,A. (1966), Biosynthesis of the antibiotic ni- lishers BV.
sin by whole Streptococcus lactis organisms, J. KALEITA,C., ENTIAN,K.-D. (1989), Nisin, a pep-
Gen. Microbiol. 44,209-220. tide antibiotic: cloning and sequencing of the
HURST,A. (1981). Nisin, Adv. Appl. Microbiol. 27, nisA gene and posttranslational processing of its
85-123. peptide product, J. Bacteriol. 171, 1597-1601.
364 8 Lantibiotics
KALETTA, C., ENTIAN, K.-D., KELLNER, R., JUNG, produced by Bacillus subtilis ATCC 6633, Appl.
G., REIS,M., SAHL,H.-G. (1989), Peps, a new Environ. Microbiol. 60, 2793-2801.
lantibiotic: structural gene isolation and prepep- KLEIN, C., KALETI-A,C., SCHNELL,N., ENTIAN,
tide sequence, Arch. Microbiol. 152, 16-19. K.-D. (1992), Analysis of the genes involved in
KALETI-A,C., ENTIAN, K.-D., JUNG,G. (1991a), the biosynthesis of the lantibiotic subtilin, Appl.
Prepeptide sequence of cinnamycin (Ro 09- Environ. Microbiol. 58, 132-142.
0198): the first structural gene of a duramycin- KLEIN,C., KALETTA,C., ENTIAN,K.-D. (1993),
type lantibiotic, Eur. J. Biochem. 199, 411415. Biosynthesis of the lantibiotic subtilin is regul-
KALETI-A,C., KLEIN,C., SCHNELL,N., ENTIAN, ated by a histidine kinaselresponse regulator sys-
K.-D. (1991b), An operon-like structure of the tem, Appl. Environ. Microbiol. 59, 296-303.
genes involved in subtilin biosynthesis, in: Nisin KLEINKAUF, H., VON D ~ H R E NH., (1986) Peptide
and Novel Lantibiotics (JUNG,G., SAHL,H.-G., antibiotics, in: Biotechnology, 1st Edn., Vol. 4
Eds.), pp. 309-319. Leiden: ESCOM Scientific (REHM,H. J., REED, G., Eds.), pp. 283-307,
Publishers BV. Weinheim: VCH.
KATZ,E., DEMAIN, A. L. (1977), The peptide anti- KLEINKAUF, H., VON DOHREN,H. (1987), Biosyn-
biotics of Bacilllus: chemistry, biogenesis and thesis of peptide antibiotics, Ann. Rev. Micro-
possible functions, Bacteriol. Rev. 41, 499474. biol. 41,259-289
KELLNER, R., JUNG,G., HORNER,T., ZAHNER, KLEINKAUF, H., VON DOHREN,H. (1990), Non-ri-
H., SCHNELL,N., ENTIAN,K.-D., GOTZ, F. bosomal biosynthesis of peptide antibiotics, Eur.
(1988), Gallidermin, a new lanthionine-contain- J. Biochem. 192,l-15.
ing polypeptide antibiotic, Eur. J. Biochem. 177, KOGLER,H., BAUCH,M., FEHLHABER,H.-W.,
53-59. GRIESINGER, C., SCHUBERT,W., TEETZ, V.
KELLNER, R., JUNG,G., JOSTEN,M.,KALETTA, C., (1991), NMR-spectroscopic investigations on
ENTIAN,K.-D., SAHL,H.-G. (1989), Peps: struc- mersacidin, in: Nisin and Novel Lantibiotics
ture elucidation of a large lantibiotic, Angew. (JUNG,G., SAHL, H.-G., Eds.), pp. 159-170.
Chem. 28,616-619. Leiden: ESCOM Scientific Publishers BV.
KELLNER, R., JUNG,G., SAHL,H.-G. (1991), Struc- KORDEL,M., BENZ,R., SAHL,H.-G. (1988), Mode
ture elucidation of the tricyclic lantibiotic Pep5 of action of the staphylococcin-like peptide
containing eight positively charged amino acids, Peps: voltage-dependent depolarization of bac-
in: Nisin and Novel Lantibiotics (JUNG, G., terial and artificial membranes, J. Bacteriol. 170,
SAHL, H.-G., Eds.), pp. 141-158. Leiden: 84-88.
ESCOM Scientific Publishers BV. KOZAK,W., RAJCHERT-TRZPIL, M., ZAJDEL,J.,
KESSLER,H., STEUERNAGEL, S., GILLESSEN, D., DOBZANSKI, W. T. (1974), The effect of proflav-
KAMIYAMA, T. (1987), Complete sequence de- in, ethidium bromide and elevated temperature
termination and localization of one imino and on the appearance of nisin-negative clones in ni-
three sulphide bridges of the nonadecapeptide sin-producing strains of Streptococcus lactis, J.
Ro 09-0198 by homonuclear 2D-NMR spectros- Gen. Microbiol. 83,295-302.
copy: the DQF-RELAYED-NOESY-experi- KUIPERS,0. P., ROLLEMA,H. S., YAP, W. M. G.
ment, Helv. Chim. Acta 70, 726-741. J., BOOT, H. J., SIEZEN,R. J., DE Vos, W. M.
KESSLER,H., STEUERNAGEL, S., WILL,M., JUNG, (1992), Engineering dehydrated amino acid resi-
G., KELLNER,R., GILLESSEN, D., KAMIYAMA, dues in the antimicrobial peptide nisin, J. Biol.
T. (1988), The structure of the polycylic nonade- Chem. 267,24340-24346.
capeptide Ro 09-0198, Helv. Chim. Acta 71, KUIPERS, 0.P., BEERTHUYZEN, M. M., SIEZBN,R.
1924-1929. J., DE Vos, W. M. (1993a), Characterization of
KESSLER,H., SEIP, S., WEIN,T., STEUERNAGEL, the nisin gene cluster nisABTCIPR of Lactococ-
S., WILL, M. (1991), Structure of cinnamycin cus lactis: requirement of expression of nisA and
(Ro 09-0198) in solution, in: Nisin and Novel nisl genes for development of immunity, Eur. J.
Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. Biochem. 216,281-292.
141-158. Leiden: ESCOM Scientific Publishers KUIPERS,0.P., ROLLEMA,H. S., DE Vos, W. M.,
BV. SIEZEN,R. J. (1993b), Biosynthesis and secre-
KEITENRING,J., MALABARABA, A., VEKEY,K., tion of a precursor of nisin Z by Lactococcus lac-
CAVALLERI, B. (1990), Sequence determination tis directed by the leader peptide of the homolo-
of actagardine, a novel lantibiotic, by homonu- gous lantibiotic subtilin from Bacillus subtilis,
clear 2D NMR spectroscopy, J. Antibiot. 43, FEBS Lett. 330,23-27.
1082-1088. KUPKE, T., STEVANOVI~, S., SAHL,H.-G., GOTZ,
KLEIN,C., ENTIAN,K.-D. (1994), Genes involved F. (1992), Purification and characterziation of
in self-protection against the lantibiotic subtilin EpiD, a flavoprotein involved in the biosynthe-
8 References 365
sis of the lantibiotic epidermin, J. Bacteriol. 174, MEYER,H. E. (1994), Analyzing posttranslational
5354-5361. protein modifications, in: Microcharacterization
KUPKE,T., STEVANOVI~, S., OTTENWALDER,B., of Proteins (KELLNER,R., LOTTSPEICH,F.,
METZGER,J. W., JUNG,G., GOTZ, F. (1993), MEYER,H. E., Eds.). Weinheim: VCH.
Purification and characterization of EpiA, the MEYER,C., BIERBAUM,G., HEIDRICH,C., REIS,
peptide substrate for posttranslational modifica- M., SOLING,J., IGLESIAS-WIND,M. I., KEMP-
tions involved in epidermin biosynthesis, FEMS TER, C., MOLITOR,E., SAHL,H.-G. (1995), Nu-
Microbiol. Lett. 112, 4348. cleotide sequence of the lantibiotic Pep5 biosyn-
KUPKE, T., KEMPTER,C., GNAU,V., JUNG, G., thetic gene cluster and functional analysis of
GOTZ, F. (1994), Mass spectroscopic analysis of PepP and PepC: evidence for a role of PepC in
a novel enzymatic reaction: oxidative decarboxy- thioether formation, Eur. J. Biochem. 232, 478-
lation of the lantibiotic precursor peptide EpiA 489.
catalyzed by the flavoprotein EpiD, J. Biol. MOLITOR,E., SAHL,H.-G. (1991), Applications of
Chem. 269,5653-5659. nisin: a literature survey, in: Nisin and Novel
KUPKE, T., KEMFTER, C., JUNG, G., GOTZ, F. Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp.
(1995), Oxidative decarboxylation of peptides 434439. Leiden: ESCOM Scientific Publishers
catalyzed by flavoprotein EpiD: determination BV.
of substrate specificity using peptide libraries MBRTVEDT, C. I., NES, I. F. (1990), Plasmid-asso-
and neutral loss mass spectrometry, J. Biol. ciated bacteriocin production by Lactobacillus
Chem. 270, 11282-11289. sake, J. Gen. Microbiol. 136, 1601-1607.
LIAN, L.-Y., CHAN,W. C., MORLEY,S. D., Ro- MBRTVEDT, c. I., NISSEN-MEYER, J., SLETTEN, K.,
BERTS,G. c. K., BYCROFT,B. w., JACKSON,D. NES, I. F. (1991), Purification and amino acid se-
(1991), NMR studies of the solution structure of quence of lactocin S, a bacteriocin produced by
nisin A, in: Nisin and Novel Lantibiotics (JUNG, Lactobacillus sake L45, Appl. Environ. Micro-
G., SAHL, H.-G., Eds.), pp. 43-58. Leiden: biol. 57, 1829-1834.
ESCOM Scientific Publishers BV. MSADEK,T., KUNST, F., RAPOPORT,G. (1993),
LIAN, L.-Y., CHAN,W. C., MORLEY,S. D., Ro- Two component regulatory systems, in: Bacillus
BERTS,G. c. K., BYCROFI-,B. w., JACKSON,D. subtilis and Other Gram-Positive Bacteria (SO-
(1992), Solution structures of nisin and its two NENSHEIN, A., HOCH. L., LOSICK,R., Eds.), pp.
major degradation products determined by 729-746. Washington, D C ASM.
NMR, Biochem. J. 283,413-420. MULDERS,J. W. M., BOERRIGTER,I. J., ROLLE-
LINNETT,P. E., STROMINGER, J. L. (1973), Addi- MA,H. S., SIEZEN,R. J., DE Vos, W. M. (1991),
tional inhibitors of peptidoglycan synthesis, An- Identification and characterization of the lanti-
timicrob. Agents Chemother. 4, 231-236. biotic nisin Z, a natural nisin variant, Eur. J. Bio-
LIU, W., HANSEN,J. N. (1992), Enhancement of chem. 201,581-584.
the chemical and antimicrobial properties of NAKANO, M. M., ZUBER, P. (1990), Molecular
subtilin by site-directed mutagenesis, J. Biol. biology of antibiotic production in Bacillus, Crit.
Chem. 267,25078-25085. Rev. Biotechnol. 10, 223-240.
MALABARABA, A., LANDI, M., PALLANZA,R., NARUSE,N., TENMYO,O., TOMITA,K., KONISHI,
CAVALLERI,B. (1985), Physicochemical and M., MIYAKI,T., KAWAGUCHI, H., FUKASE,K.,
biological properties of actagardine and some WAKAMIYA, T., SHIBA,T. (1989), Lanthiopep-
acid hydrolysis products, J. Antibiot. 38, 1506- tin, a new peptide antibiotic. Production, isola-
1511. tion and properties of lanthiopeptin, J. Antibiot.
MALABARABA, A., PALLANZA,R., BERTI, M., 42, 837-845.
CAVALLERI, B. (1990), Synthesis and biological NAVARRO,J., CHABOT,J., SHERRILL,K., ANEJA,
activity of some amide derivatives of the lanti- R., ZAHLER,S. A., RACKER,E. (1985), Interac-
biotic actagardine, J. Antibiot. 43, 1089-1097. tion of duramycin with artificial and natural
MARKI,F., FRANSON,R. (1986), Endogenous sup- membranes, Biochem. 24,4645-4650.
pression of neutral-active and calcium-depend- NISHIKAWA, M., TESHIMA,T., WAKAMIYA,T.,
ent phospholipase A2 in human polymorphonu- SHIBA,T., KOBAYASHI,Y., OKUBO,T., KYO-
clear leucocytes, Biochim. Biophys. Acta 879, GOKU,Y., KIDO, Y. (1988), Chemistry of lanti-
149-156. biotics, in: Peptide Chemistry 1987 (SHIBA,T.,
MARKI, F., HANNI,E., FREDENHAGEN, A., VAN SAKAKIBARA, S., Eds.), pp. 71-74. Osaka: Pro-
OOSTRUM,J. (1991), Mode of action of the lan- tein Research Foundation.
thionine-containing peptide antibiotics duramy- NISHIO,C., KOMURA,S., KURAHASHI, K. (1983),
cin, duramycin B, duramycin C and cinnamycin Peptide antibiotic subtilin is synthesized via pre-
as direct inhibitors of phospholipase A2, Bio- cursor proteins, Biochem. Biophys. Res. Comm.
chem. Pharmacol. 42,2027-2035. 116,751-758.
366 8 Lantibiotics
NOVAK,J., CAULFIELD,P. W., MILLER, E. J. REIS, M., SAHL,H.-G. (1991), Genetic analysis of
(1994), Isolation and biochemical characteriza- the producer self-protection mechanism (im-
tion of a novel lantibiotic mutacin from Strepto- munity) against Peps, in: Nisin and Novel Lanti-
coccus mutans, J. Bacteriol. 176, 4316-4320. biotics (JUNG,G., SAHL,H.-G., Eds.), pp. 320-
PALMER,D. E., MIERKE,D. F., PAITARONI,C., 331. Leiden: ESCOM Scientific Publishers BV.
GOODMAN,M., WAKAMIYA,T., FUKASE,K., REIS, M., ESCHBACH-BLUDAU, M., IGLESIAS-
FUJITA,H., SHIBA,T. (1989). Interactive NMR WIND, M. I., KUPKE,T., SAHL,H.-G. (1994),
and computer simulation studies of lanthionine- Producer immunity towards the lantibiotic Peps:
ring structures, Biopolymers 28, 397408. identification of the immunity gene pepl and lo-
PARENTI,F., PAGANI,H., BEREITA, G. (1976), calization and functional analysis of its gene
Gardimycin, a new antibiotic from Actinoplanes. product, Appl. Environ. Microbiol. 60, 4332-
I. Description of the producer strain and fer- 4338.
mentation studies, J. Antibiot. 24, 501-506. REISINGER, P. SEIDEL, H., TSCHESCHE,H.,
PESCHEL,A., AUGUSTIN,J., KUPKE,T., STEVA- HAMMES,W. P. (1980), The effect of nisin on
N O V I ~S.,, GOTZ, F. (1993), Regulation of epi- murein synthesis, Arch. Microbiol. 127, 187-
dermin biosynthetic genes by EpiQ, Mol. Micro- 193.
biol. 9, 31-39. RINCE,A., DUFOUR,A., LE POGAM,S., THUAULT,
PIARD,J.-C., MURIANA, P. M., DESMAZEAUD M. D., BOURGEOIS, C. M., LE PENNEC, J. P. (1994),
J., KLAENHAMMER, T. R. (1992), Purification Cloning, expression and nucleotide sequence of
and partial characterization of lacticin 481, a lan- genes involved in production of lactococcin DR,
thionine-containing bacteriocin produced by a bacteriocin from Lactococcus lactis, Appl. En-
Lactococcus lactis subspp. lactis CNRZ 481, viron. Microbiol. 60, 1652-1657.
Appl. Environ. Microbiol. 58, 219-284. RINTALA,H., GRAEFFE,T., PAULIN,L., KALKKI-
PIARD,J.-C., KUIPERS,0. P., ROLLEMA,H. S., NEN,N., SARIS,P. E. J. (1993), Biosynthesis of
DESMAZEAUD, M. J., DE VOS, W. M. (1993), nisin in the subtilin producer Bacillus subtilis
Structure, organization and expression of the lct ATCC 6633, Biotechnol. Lett. 15,991-996.
gene for lacticin 481, a novel lantibiotic pro- ROGERS,L. A., WHITTIER,E. 0. (1928), Limiting
duced by Lactococcus lactis, J. Biol. Chem. 268, factors in lactic fermentation, J. Bacteriol. 16,
16361-16368. 211-214.
PUGSLEY,A. P. (1988), The immunity and lysis ROLLEMA,H. S., BOTH, P., SIEZEN,R. J. (1991),
genes of COINplasmid pCHAP4, Mol. Gen. Ge- NMR and activity studies of nisin degradation
net. 211, 335-341. products, in: Nisin and Novel Lantibiotics
PUGSLEY,A. P. (1993), The complete general se- (JUNG, G., SAHL, H.-G., Eds.), pp. 123-130.
cretory pathway in Gram-negative bacteria, Mi- Leiden: ESCOM Scientific Publishers BV.
crobiol. Rev. 57, 50-108. Ross, K. F., RONSON,C. W., TAGG,J. R. (1993),
RACKER,E., RIEGLER,C., ABDEL-GHANY,M. Isolation and characterization of the lantibiotic
(1983), Stimulation of glycolysis by placental po- salivaricin A and its structural gene salA from
lypeptides and inhibition by duramycin, Cancer, Streptococcus salivarius 20P3, Appl. Environ.
Res. 44,1364-1367. Microbiol. 59, 2014-2021.
RAMSEIER,H. R. (1960), Die Wirkung von Nisin RUHR,E., SAHL,H.-G. (1985), Mode of action of
auf Clostridium butyricum, Arch, Microbiol. 37, the peptide antibiotic nisin and influence on the
51-94. membrane potential of whole cells and on artifi-
RAUCH,P. J. G., DE Vos, W. M., (1992), Charac- cial membrane vesicles, Antimicrob. Agents Che-
terization of the novel nisin-sucrose conjugative mother. 27,841-845.
transposon Tn5276 and its insertion in Lactococ- SAHL,H.-G. (1985), Influence of the staphylococ-
cus lactis, J. Bacteriol. 174, 1280-1287. cin-like peptide Pep5 on membrane potential of
RAUCH,P. J. G., BEERTHUYZEN, M. M., DE Vos, bacterial cells and cytoplasmic membrane vesi-
W. M. (1990), Nucleotide sequence of IS904 cles, J. Bacteriol. 162, 833-836.
from Lactococcus lactis subspp. lactis NIZO R5, SAHL, H.-G. (1991), Pore formation in bacterial
Nucleic Acids Res. 18, 4253. membranes by cationic lantibiotics, in: Nisin and
RAUCH,P. J. G., BEERTHUYZEN, M. M., DE Vos, Novel Lantibiotics (JUNG, G., SAHL, H.-G.,
W. M. (1991), Molecular analysis and evolution Eds.), pp. 347-358. Leiden: ESCOM Scientific
of conjugative transposons encoding nisin pro- Publishers BV.
duction and sucrose metabolism in Lactococcus SAHL,H.-G., BRANDIS,H. (1981), Production, pu-
lactis, in: Nisin and Novel Lantibiotics (JUNG, rification and chemical properties of an antista-
G., SAHL, H.-G., Eds.), pp. 243-249. Leiden: phylococcal agent produced by Staphylococcus
ESCOM Scientific Publishers BV. epidermidis, J. Gen. Microbiol. 127, 377-384.
8 References 367
SAHL,H.-G., BRANDIS, H. (1983), Efflux of low Mr SHOTWELL, 0. L., STODOLA,F. H., MICHAEL,W.
substances from the cytoplasm of sensitive cells R., LINDENFELSER, L. A., DWORSCHAK, G.,
caused by the staphylococcin-like agent Pep5, PRIDHAM,T. G. (1958), Antibiotics against
Zentralbl. Bakteriol. Hyg. I. Abt. Orig. A . 252, plant disease. 111. Duramycin, a new antibiotic
166-175. from Streptomyces cinammonensis forma azaco-
SAHL,H.-G., KORDEL,M., BENZ,R. (1987), Volt- luta, J. Am. Chem. SOC.80,3912-3915.
age-dependent depolarization of bacterial mem- SIEGERS,K., ENTIAN,K.-D. (1995), Genes in-
branes and artificial lipid bilayers by the peptide volved in immunity to the lantibiotic nisin pro-
antibiotic nisin, Arch. Microbiol. 149, 120-124. duced by Lactococcus lactis 6F3, Appl. Environ.
SAHL,H.-G., REIS, M.,ESCHBACH,M., SZEKAT, Microbiol. 61, 1082-1089.
C., BECK-SICKINGER, A. G., METZGER,J., STE- SKAUGEN, M., NISSEN-MEYER, J., JUNG,G., STE-
V A N O V I ~ ,S., JUNG,G. (1991), Isolation of Pep5 ~ ,SLETTEN,K., MORTVEDT-ABILD-
V A N O V IS.,
prepeptides in different stages of modification, GAARD,C. I., NES, I. (1994), In vivo conversion
in: Nisin and Novel Lantibiotics (JUNG, G., of L-serine to D-alanine in a ribosomally-synthe-
SAHL, H.-G., Eds.), pp. 332-346. Leiden: sized polypeptide, J. Biol. Chem. 269, 27183-
ESCOM Scientific Publishers BV. 27185.
SAHL,H.-G., JACK,R. W., BIERBAUM, G. (1995), SLIJPERS, M., HILBERS,C. W., KONINGS,R. N. H.,
Lantibiotics: Biosynthesis and biological activi- VAN DE VEN, F. J. M. (1989), NMR studies of
ties of peptides with unique posttranslational lantibiotics: assignment of the 'H-NMR spec-
modifications, Eur. J. Biochem. 230, 827-853. trum of nisin and identification of interresidual
contacts, FEBS Lett. 252, 22-28.
SCHNELL,N., ENTIAN,K.-D., SCHNEIDER,U.,
GOTZ, F., ZAHNER,H., KELLNER,R., JUNG,G. SOKOLOVE,P. M., WESTPHAL,P. A., KESTER,M.
(1988), Prepeptide sequence of epidermin, a ri- B., WIERWILE,R., VAN METER, K. S. (1989),
bosomally-synthesized antibiotic with four sul- Duramycin effects on the structure and function
phide rings, Nature (London) 333, 276-278. of heart mitochondria. I. Structural alterations
and changes in membrane permeability, Bio-
SCHNELL,N., ENGELKE,G., AUGUSTIN, J., Ro- chim. Biophys. Acta 983, 15-22.
SENSTEIN, R., GOTZ, F., ENTIAN, K.-D. (1991), SOMMA,S., MERATI,W., PARENTI,F. (1977), Gar-
The operon-like organization of lantibiotic epi- damycin, a new antibiotic inhibiting peptidogly-
derrnin biosynthesis genes, in: Nisin and Novel can synthesis, Antimicrob. Agents Chemother.
Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 11,396-401.
269-276. Leiden: ESCOM Scientific Publishers
BV. SONG,H. Y., CRAMER,W. A. (1991), Membrane
topography of ColEl gene products: the immun-
SCHNELL,N., ENGELKE,G., AUGUSTIN,J., Ro- ity protein, J. Bacteriol. 173, 2935-2943.
SENSTEIN, R., UNGERMANN, v., GOTZ, F., EN- STEEN,M. T., CHUNG,Y. J., HANSEN, J. N. (1991),
TIAN, K.-D. (1992), Analysis of genes involved
Characterization of the nisin gene as part of a
in the biosynthesis of the lantibiotic epidermin, polycistronic operon in the chromosome of Lac-
Eur. J. Biochem. 204, 57-68. tococcus lactis ATCC 11454, Appl. Environ. Mi-
SCHULLER, F., BENZ,R., SAHL,H.-G. (1989), The crobiol. 57, 1181-1188.
peptide antibiotic subtilin acts by formation of STEVENS,K. A., SHELDON,B. W., KLAPES,N. A.,
voltage-dependent multi-state pores in bacterial KLAENHAMMER, T. R. (1991), Nisin treatment
and artificial membranes, Eur. J. Biochem. 182, for inactivation of Salmonella species and other
181-186. Gram-negative bacteria, Appl. Environ. Micro-
SEARS,P. M., SMITH,B. S., STEWART, W. K., GON- biol. 537,3613-3615.
ZALEZ, R. N., RUBINO, S. D., GUSIK,S. A., KU- STOFFELS,G., NISSEN-MEYER,J., GUDMUNDS-
LISEK, E. s., PROJAN,s. J., BLACKBURN, P. DOTTIR,A., SLETTEN,K., HOLO,H., NES, I. F.
(1992), Evaluation of a nisin-based germicidal (1992), Purification and characterization of a
formulation on teat skin of live cows, J. Dairy new bacteriocin isolated from a Carnobacterium
Sci. 75,3185-3190. spp., Appl. Environ. Microbiol. 58, 1417-1422.
SHETH,T. R., HENDERSON, R. M., HLADKY, S. B., STOFFELS,G., GUDMUNDSDOTTIR, A., ABEE, T.
CUTHBERT,A. W. (1992), Ion-channel forma- (1994), Membrane-associated proteins encoded
tion by durarnycin, Biochim. Biophys. Acta 1107, by the nisin gene cluster may function as a re-
179-185. ceptor for the lantibiotic carnocin UI49, Micro-
SHIBA,T., WAKAMIYA, T., FUKASE,K., SANO,A., biology 140, 1443-1450.
SHIMBO,K., UEKI,Y. (1986), The chemistry of STONE,D. K., XIE, X. S., RACKER,E. (1984), Inhi-
lanthionine-containing peptides, Biopolymers bition of clathrin-coated vesicle acidification by
25, Sll-S19. duramycin, J. Biol. Chem. 259, 2701-2703.
368 8 Lantibiotics
SUROVOY,A., WEIDELICH,D., JUNG,G. (1992), VAN DE VEN,F. J. M., VAN DEN HOOVEN,H. W.,
Electrospray mass spectroscopic analysis of me- KONINGS,R. N. H., HILBERS,C. W. (1991b),
tal-peptide complexes, in: Peptides 1992, Pro- The spatial structure of nisin in aqueous solu-
ceedings of the 22nd European Peptide Sympo- tion, in: Nisin and Novel Lantibiotics (JUNG,G.,
sium (SCHNEIDER,C . H., EBERLE,A. N., Eds.), SAHL,H.-G., Eds.), pp. 3542. Leiden: ESCOM
pp. 563-564. Leiden: ESCOM Scientific Publish- Scientific Publishers BV.
ing BV. VAN DER MEER, J. R., POLMAN, J., BEERTHUY-
TAGG,J. R., WANNAMAKER, L. W. (1978), Strep- ZEN, M. M., SIEZEN,R. J., KUIPERS,0. P., DE
tococcin A-FF22 nisin-like antibiotic substance Vos, W. M. (1993), Characterization of the Lac-
produced by a group A Streptococcus, Antimi- tococcus lactis nisin A operon genes nisP encod-
crob. Agents Chemother. 14,31-39. ing a subtilisin-like serine protease involved in
TAGG,J. R., DAJANI,A. S., WANNAMAKER, L. W., precursor processing and nisR encoding a regul-
GRAY, E. D. (1973a), Group A streptococcal atory protein involved in nisin biosynthesis, J.
bacteriocin: production, purification and mode Bacteriol. 175, 2578-2588.
of action, J. Exp. Med. 138, 1168-1183. VAN DER MEER, J. R., ROLLEMA,H. S., SIEZEN,
TAGG, J. R., READ, R. S. D., MCGIVEN,A. R. R. J., BEERTHUYZEN,M. M., KUIPERS,0. P.,
(1973b), Bacteriocin of group A Streptococcus: DE Vos, W. M. (1994), Influence of amino acid
Partial purification and properties, Antimicrob. substitutions in the nisin leader peptide on bio-
Agents Chemother. 4,214-221. synthesis and secretion of nisin by Lactococcus
TSAI, H.-J., SANDINE,W. E. (1987), Conjugal lactis, J. Biol. Chem. 269, 3555-3562.
transfer of nisin plasmid genes from Streptococ- VOGEL,H., NILSSON,L., RIGLER,R., MEDER, S.,
cus lactis 7962 to Leuconostoc dextranicum 181, BOHEIM,G., BECK, W., KURTH,H.-H., JUNG,
Appl. Environ. Microbiol. 53, 352-357. G. (1993), Structural fluctuations between two
UNGERMANN, V., GOEKE, K., FIEDLER,H.-P., conformational states of a transmembrane heli-
ZAHNER,H. (1991), Optimization of fermenta- cal peptide are related to its channel-forming
tion and purification of gallidermin and epider- properties in planar lipid membranes, Eur. J.
min, in: Nisin and Novel Lantibiotics (JUNG,G., Biochem. 212, 305-313.
SAHL, H.-G., Eds.), pp. 410-421. Leiden: WAKAMIYA, T., UEKI, Y., SHIBA, T., KIDO, Y.,
ESCOM Scientific Publishers BV. MOTOKI,Y. (1985), The structure of ancovenin,
VAN BELKUM, M. J., KOK, J., VENEMA,G., HOLO, a new peptide inhibitor of angiotensin I convert-
H., NES, I. F., KONINGS,W. N., ABEE,T. (1991), ing enzyme, Tetrahedron Lett. 26, 665-668.
The bacteriocin lactococcin A specifically in-
creases the permeability of lactococcal cytoplas- WAKAMIYA, T., FUKASE,K., NARUSE,N., KONI-
mic membranes in a voltage-independent, pro- SHI,M., SHIBA,T. (1988), Lanthiopeptin, a new
tein-mediated manner, J. Bacteriol. 173, 7934- peptide effective against Herpes simplex virus:
structural determination and comparison with
7941.
VAN DE KAMP, M., HORSTINK,L. M., VAN DEN
R o 09-0198, an immunopotentiating peptide,
HOOVEN,H., KONING,R. N. H., HILBERS,C. Tetrahedron Lett. 29, 47714772.
W., FREY,A., SAHL,G.-G., METZGER,J., VAN WEIL, H.-P., BECK-SICKINGER, A. G., METZGER,
DE VEN, F. J. M. (1995a), Sequence analysis by J., STEVNANOVI~, S., JUNG, G., JOSTEN, M.,
NMR spectroscopy of the peptide lantibiotic SAHL, H.-G. (1990), Biosynthesis of the lanti-
epilancin K7 from Staphylococcus epidermidis biotic Pep5: Isolation and characterization of a
K7, Eur. J. Biochem. 227,757-771. prepeptide containing dehydroamino acids, Eur.
VAN DE KAMP,M., VAN DE VEN, F. J. M., KON- J. Biochem. 194,217-223.
INGS,R. H. H., HILBERS,C. W., METZGER,J. ZIMMERMANN, N. (1995), Raumstrukturaufklarung
W., JUNG,G., KUIPERS,0. P., BIERBAUM,G., der Lantibiotika Duramycin B, Duramycin C
SAHL,H.-G. (1995b), Elucidation of the primary und Actagardin durch mehrdimensionale Kern-
structure of the peptide lantibiotic epilancin K7 resonanzspektroskopie, PhD Thesis, University
from Staphylococcus epidermidis: cloning of the of Tubingen, Germany.
epilancin K7-encoding gene and Edman degrad- ZIMMERMANN, N., JUNG,G. (1995), The tetracyclic
ation of the mature peptide, Eur. J. Biochem. lantibiotic actagardine. 'H-NMR and I3C-NMR
230,587-600. assignments and revised primary structure, Eur.
VAN DE VEN, F. J. M., VAN DEN HOOVEN,H. W., J. Biochem. 228,786-797.
KONINGS,R. N. H., HILBERS,C. W. (1991a), ZIMMERMANN, N., FREUND,S., FREDENHAGEN,
NMR-studies of lantibiotics: the structure of ni- A., JUNG,G. (1993), Solution structures of the
sin in aqueous solution, Eur. J. Biochem. 202, lantibiotics duramycin B and C, Eur. J. Biochem.
1181-1188. 216,419-428.
Biotechnology
9 Glycopeptide Antibiotics
(Dalbaheptides)
GIANCARLO
LANCINI
BRUNOCAVALLERI
Gerenzano, Italy
1 Introduction 371
2 Descriptive Chemistry 371
2.1 Physicochemical Properties 376
3 Biological Activity 377
3.1 In v i m Antibacterial Activity 377
3.2 Mechanism of Action 378
3.3 Resistance 379
3.4 In vivo Efficacy and Pharmacology 379
4 Producing Organisms 380
5 Methods of Screening 381
6 Fermentation 381
6.1 Fermentation Media - Carbon Sources 381
6.2 Fermentation Media - Nitrogen Sources 382
6.3 Fermentation Media - Effect of Phosphates 382
6.4 Inhibition by the Final Product 382
6.5 Control of Complex Composition 383
7 Recovery and Purification 383
8 Biosynthesis 384
8.1 Origin of the Uncommon Amino Acids 384
8.2 Origin of Fatty Acids of Lipoglycopeptides 386
8.3 Glycosylation and Final Modifications 387
9 Chemical Modifications 387
310 9 Glycopeptide Antibiotics (Dalbaheptides)
10 Biotransformations 389
10.1 Deglycosylation 389
10.2 Glycosylation 389
10.3 Deacylation 389
10.4 Other Biotransformations 390
11 References 390
1 Introduction planin are used in human medicine, due to

their effect on gram-positive pathogens re-
fractory to established antibiotics, such as
The term “glycopeptide antibiotics” is com- multiresistant Staphylococcus aureus, coagul-
monly used to indicate a family of microbial ase-negative staphylococci, clostridia, and en-
metabolites, active on gram-positive bacteria, terococci. Eremomycin, a relatively recent
and closely related in their chemical structure natural product is under clinical evaluation.
and biological activity to the antibiotics van- Avoparcin is commercially available as
comycin and ristocetin - the first members of growth promoter in animal feed. Finally, ris-
the family isolated in the early 1950s. The ex- tocetin is a diagnostic agent for a particular
pression “glycopeptide antibiotics of the van- disorder of genetic origin (von Willebrand’s
comycin-ristocetin family” is more precise disease) due to the ability to aggregate blood
and is also often used. platelets.
The characteristics common to all members
of the family are:
- a structure composed of a linear heptapep-
tide in which at least five of the amino acid
residues are aromatic, the rings linked to 2 Descriptive Chemistry
form a triphenyl ether moiety and a diphe-
nyl group; The first glycopeptide antibiotic isolated
- a unique mechanism of action, i.e., inhibi- from microbial fermentation was ristocetin in
tion of bacterial growth by binding to the 1953, followed by vancomycin in 1955. Al-
D-Ala-D-Ala terminus of peptidoglycan most immediately their composition of sugar
precursors, thus inhibiting cell wall forma- carrying small peptides was defined, but the
tion. determination of their structures was com-
Taking into account their chemical charac- pleted only 20 years later, mainly by studies
teristics and unusual mechanism of action, the with high-field NMR. In the 1960s and 1970s
name Dalbaheptides, from Dal (anyl-D-ala- the isolation of a small number of new glyco-
nine) B(inding) A(ntibiotics with) Hept(a- peptides including avoparcin and teicoplanin
pept)ide (structure), was proposed for these was reported, but only in the 1980s the intro-
antibiotics .by PARENTI and CAVALLERI duction into screening programs of specific
(1989). methods for the detection of these antibiotics
Several reviews have been published rein fermentation broth resulted in the discov-
porting different aspects of glycopeptide anti- ery of the majority of glycopeptides now
biotics: discovery, isolation, and purification available. Modern techniques of isolation and
(CASSANI,1989; SITRINand FOLENA-WAS- purification such as reverse-phase chromato-
SERMAN,1989; CAVALLERIand PARENTI, graphy and affinity chromatography and
1992), mechanism of action (BARNA and availability of powerful spectroscopic tech-
WILLIAMS, 1984; REYNOLDS,1989), fermen- niques such as 2D-NMR, Fast Atom Bom-
tation and biosynthesis (LANCINI,1989; LAN- bardment Mass Spectrometry (FAB-MS), and
CINI and CAVALLERI, 1990), chemistry and Ion Spray MS have now made it possible to
chemical derivatives (MALABARBA et al., determine the structure of new products in a
1993a), structure-activity relationship (NA- relatively short time.
GARAJAN,1993), pharmacology (CASSETTA A list of natural glycopeptides is shown in
et al., 1991; PHILLIPSand GOLLEDGE,1992; Tabs. 1-4. More detailed references and the
BROGDENand PETERS,1994). Biological and chemical structures are provided in the pre-
chemical aspects are comprehensively dis- viously cited reviews by LANCINI and
cussed in a multiauthor book published re- CAVALLERI(1990) and CAVALLERIand
cently (NAGARAJAN, 1994a). PARENTI(1992). It should be noted that in
Although many natural and semisynthetic some cases different code numbers or names
glycopeptides are provided with relevant anti- were assigned to identical chemical entities
bacterial activity, only vancomycin and teico- produced by different strains, before their
372 9 Glycopeptide Antibiotics (Dalbaheptides}
structure had been elucidated. A striking ex- initially been overlooked (BORGHI et al.,
ample is that of eremomycin that was succes- 1989; NAGARAJAN, 1993).
sively reported as A82846A, MM45289, and The glycopeptide skeleton is shown in Fig. 1.
LY264826. In addition to the products listed The numbering of the amino acid residues is
in Tabs. 1-4 the following compounds with that proposed by BARNAet al. (1984); it has
unknown structure were reported AM374 been widely used because it allows an easy
from Streptomyces ebureosporeus (KUNST- comparison of NMR data of glycopeptides
MANN and PORTER,1974), A477 from Acti- having different core structures. It should be
noplanes sp. (HAMILLet al., 1973), AB65
from Saccharomonospora viride (TAMURA
and TAKEDA,1975). OH
Many of the microorganisms listed produce
families (complexes) of strictly related com-
pounds. The components (factors) of a com-
plex usually differ in the level of methylation
or chlorination of the peptidic skeleton or in
the presence of additional sugars. Analogs
lacking some or all sugar units present in the
parent antibiotic (therefore designated as
"pseudoaglycones" or aglycones, respective-
ly) are either the result of incomplete glycosy-
lation or generated by chemical or enzymatic
deglycosylation during fermentation or recov-
ery and purification. In several cases, further Fig. 1. General lieptapeptide structure of dalbahep-
studies on complex producing strains have re- tides and interaction with the D-alanyl-D-alanine
vealed the presence of additional minor com- peptide terminus. Hydrogen bonds are indicated by
ponents in the fermentation broth that had dottet lines.
Tab. 1. Naturally Occurring Dalbaheptides - Ristocetin Type
Name Producing Strain" Company, Yearb References
Ristocetin A. B Nocardia lurida, NRRL 2430 Abbott. 1953 SZTARICSKAI

and BOGNAR,1984
KATRUKHAand SILAEV,1986
Ristomycin A, B Proactinomyces fructiferi Inst. New SZTARICSKAI
and BOGNAR,1984
Antibiotics, KATRUKHAand SILAEV.1986
Moscow, 1962
Actaplanin Actinoplanes missouriensis, Lilly, 1971 DEBONOet al., 1984
(A4696 complex) ATCC 23342 HUNTet al., 1984a
A35512 complex Streptomyces candidus, Lilly, 1976 MICHELet al., 1980
NRRL 8156 DEBONOet al., 1980
A41030 complex Streptomyces virginiae, Lilly, 1982 BOECKet al., 1985
NRRL 15156 HUNTet al., 1985
A47934 Streptomyces toyocaensis, Lilly, 1982 BOECKand MERTZ,1986
NRRL 15009
UK-68597 Actinoplanes sp., Pfizer, 1987 SKELTON
and WILLIAMS,1990
ATCC 53533 HOLDOMet al., 1988
UK-69542 Saccharothrix aerocoloni- Pfizer, 1991 HOLDENet al., 1991
genes, ATCC 53829 SKELTON
et al., 1991
a The name of the producing strain is according to the original publication; note that several strains have
been subsequently reclassified into different genera.
Year of first paper or patent publication.
HO
CH,
OH
Fig. 3. Vancomycin.
pxz
OH
OH
OH
Fig. 2. Ristocetin A.
noted that the residues are numbered from

the amino terminus, in contrast to the usual
system of peptide numbering starting from
the carboxyl terminus.
Phenylamino acids 2,4,5,6, and 7 are pres-
ent in all glycopeptides and a classification
based on the remaining amino acids 1 and 3 is
commonly used. The compounds wherein the
phenylic moieties of amino acids 1 and 3 are
linked by an oxygen belong to the ristocetin
type (Tab. 1); their structure is exemplified by
that of ristocetin A in Fig. 2. Vancomycin
type includes the glycopeptides listed in Tab. OH
2, where amino acids 1 and 3 are aliphatic.
Synmonicin, also included in Tab. 2, is an ex- Fig. 4. Actinoidin A.
ception, since amino acid 3 is a thioamino
acid and amino acid 1 is aromatic. The struc-
ture of vancomycin is shown in Fig. 3. In the sidic bonds. The sugars vancosamine (and the
actinoidin type (Tab. 3) amino acid 1 is al- related epivancosamine and 4-ketovancosam-
ways a p-hydroxyphenylglycine whereas ami- ine), ristosamine, actinosamine, and acosam-
no acid 3 is a phenylalanine or p-hydroxyphe- ine were first isolated from glycopeptides.
nylglycine (Fig. 4). Only A41030A, B, E and A47934 do not con-
Within a group other characteristics ac- tain any sugar. In a few compounds a phe-
count for the large variety of glycopeptide nolic hydroxyl is esterified as a sulfate mono-
structures with chlorine atoms, methyl and ester (A47934, UK-68597, UK-69542). The
hydroxy groups at different positions of the terminal carboxyl is often a methyl ester and
phenyl residues. Amino acid 6 and occasion- the terminal amino group can be methyl-
ally amino acid 2 carry a hydroxyl group in ated.
@position. Common or unusual sugars are In some ristocetin-type glycopeptides linear
linked at different positions through glyco- or branched (exceptionally unsaturated) fatty
Tab. 2. Naturally Occurring Dalbaheptides - Vancomycin Type
Name Producing Straina Company, Yearb References
Vancomycin Streptomyces orientalis, Lilly, 1955 HARRISet al., 1983

NRRL 2450 SZTARICSKAI and BOGNAR,1984
K-288 Streptomyces haranoma- Tohoku Univer- MATSUMOTO, 1961
chiensis sity, 1961
OA-7653 A, B Streptomyces hygroscopi- Otsuka Pharm., KAMOGASHIRA et al., 1983
cus ssp. hiwasaensis, 1978 ANGet al., 1988
ATCC 31613
A51568A Nocardia orientalis, Lilly, 1982 BOECKet al., 1984
(N-demethyl vanco- NRRL 15232 HUNTet al., 1984b
mycin)
M43 complex N. orientalis, NRRL 2450 Lilly, 1984 HIGGINSet al., 1985
Izupeptin A, B Nocardia sp., Kitasato Insti- SPIRI-NAKAGAWA et al., 1986
FERM P-8656 tute, 1986
Synmonicin A, B, C Synnemomyces mamno- Smith, Kline & ARJUNARAOet al., 1986
(CWI-785) orii, ATCC 53296 French, 1986
Orienticin N. orientalis, PA-42867 Shionogi, 1987 TSUJIet al., 1988b
(PA-42867-A, B, C,
D)
Eremomycin Actinomyces sp., INA-238 Inst. New Anti- BRAZHNIKOVA et al., 1989
biotics, GAUSEet al., 1989
Moscow, 1987
A42867 Nocardia sp., Lepetit, 1987 RIVAet al., 1989
ATCC 53492
A82846 A, B, C Amycolatopsis orientalis, Lilly, 1987 HAMILL
et al., 1988
NRRL 18098, NAGARAJANet al., 1989a
NRRL 18099
Chloroorienticin A, A. orientalis. PA-45052 Shionogi, 1988 TSUJIet al., 1988a
B, C, D, E NAGARAJAN et al., 1989b
(PA-45052)
A80407 A. B Kibdelosporangium phil- Lilly, 1989 DOOLINet al., 1989
ippinensis, NRRL 18198
MM 45289, A . orientalis. NCIB 12531 Beecham, 1989 GOODet al., 1990
MM 47756
MM 47761, A . orientalis. NCIB 12608 Beecham, 1989 Box et al., 1990
MM 49721
Decaplanin Kibdelosporangium dec- Hoechst, 1990 FRANCOet al., 1990
(M 86-1410) caensis, DSM 4763
UK-72 051 A . orientalis Pfizer, 1990 SKELTONet al., 1990
MM 55270, Arnycolatopsis sp., Beecham, 1990 COATESet al.. 1990a
MM 55271, NCIB 40086
MM 55272
Balhimycin Amycolatopsis sp., Hoechst, 1992 NADKARNIet al.. 1994
DSM 5908
A83850 A, B Amycolatopsis albus, Lilly, 1993 HAMILL
and YAO, 1993
NRRL 18532
See footnotes in Tab. 1
acids are linked as amides to the amino group fer in the nature of the aliphatic side chains as
of a glucosamine (or 2-aminoglucuronic acid) illustrated by the structure of teicoplanin
moiety. These antibiotics listed in Tab. 4 are (Fig. 5). The aliphatic moieties cause a certain
thus complexes the components of which dif- lipophilic character of these “lipoglycopep-
Tab. 3. Naturally Occurring Dalbaheptides - Actinoidin Type
Actinoidin A, B Proactinomyces actinoides Inst. New Anti- BERDNIKOVA

et al., 1982
biotics,
Moscow, 1956
Avoparcin Streptomyces candidus, American Cyan- MCGAHREN
et al., 1980
(LL-AV290 com- NRRL 3218 amid, 1966 MCGAHREN
et al., 1983
P W
Chloropolysporin Faenia interjecta, Sankyo, 1983 OKAZAKI et al., 1987
A, B, C FERM BP-583 TAKATSU et al., 1987a
Actinoidin A2 Nocardia sp., Smith, Kline & DINGERDISSEN et al., 1987
SKF-AAJ-193 French, 1987 HEALDet al., 1987
Helvecardin A, B Pseudonocardia compacta Sankyo, 1988 TAKEUCHI et al., 1991a
ssp. helvetica, TAKEUCHI et al., 1991b
SANK 65185
MM47766, Amycolatopsis orientalis, Beecham, 1989 ATHALYEet al., 1989
MM47767, NCIB 40011
MM55256,
MM 55260
Galacardin A. B Saccharothrix sp., Sankyo, 1990 TAKEUCHI
et al., 1992
SANK 64289
Tab. 4. Naturally Occurring Dalbaheptides - Lipoglycopeptides
Teicoplanin Actinoplanes teichomyce- Lepetit, 1975 PARENTIet al., 1978

ticus, ATCC 31121 CORONELLI et al., 1987
Ardacin Kibdelosporangium ari- Smith, Kline & SHEARER et al., 1985
(aridicin, AAD-216 dum. ATCC 39323 French, 1983 SITRINet al., 1985
complex)
A40926 complex Actinomadura sp., Lepetit, 1984 SELVAet al., 1986
ATCC 39727 GOLDSTEIN et al., 1987
Kibdelin K. aridum, ATCC 39922 Smith, Kline & SHEARER et al., 1986
(AAD-609 com- French, 1985 FOLENA-WASSERMAN et al., 1986
P W
Parvodicin Actinomadura parvosata, Smith, Kline & CHRISTENSEN
et al.. 1987
(AAJ-271) ATCC 53463 French, 1986
MM 49728, Amycolatopsis sp., Beecham. 1990 COATESet al., 1990b
MM 55266, NCIB 40089 Box et al., 1991
MM 55267,
MM 55268
A84575 complex Streptosporangium car- Lilly, 1991 MICHELand YAO, 1991
neum, NRRL 18437,
NRRL 18505
MM56597, Amycolatopsis sp., Beecham, 1991 COATESet al., 1991
MM 56598 NCIB 40089

-NH,
-
H;OH
k H
OH
T&-I:R= -
-
T&-2:R=
0
T-A?-3R I
T-M-4:Rr
0
T-M-5: R I Fig. 5. Teicoplanin (teichomycin com-
0 I plex).
tides” that directly influences their biological Tab. 5. Isoelectric Points (I. P.) of Some Dalbahep-
properties and in particular their pharmacoki- tidesa
netic behavior.
Antibiotic I. P.
2.1 Physicochemical Properties A47934 3.2

A40926 3.7-3.9
Ardacin 3.9
Glycopeptides are colorless or whitish A41030 4.9
powders, generally water-soluble, that usually OA-7653 4.9
strongly retain water or crystallization sol- Teicoplanin 5.0
vents. Their molecular weights range from Chloropolysporin 7.5
1150-2300 Da. Some of them as, e.g., teico- A35512B 7.5-7.8
planin can be isolated as an internal salt or as Actaplanin 7.6-8.5
AM374 7.7-8.0
a partial monoalkaline (sodium) salt, depend- Avoparcin 7.7-8.1
ing on the pH value of the aqueous medium Vancomycin 7.7
in the final purification step. Others can be Ristocetin 8.1
isolated as acidic salts: vancomycin and acta- A42867 8.1
planin as hydrochlorides, and ristocetin A, A477 7.9-8.1
avoparcin, and eremomycin as sulfates. Many
pharmaceutically acceptable basic and acidic a Determined by electrofocusing (RIVA,E., SOF-
addition salts are claimed in patent applica- FIENTINI, A., personal communication)
tions. The net charge of several glycopeptides
has been determined by electrofocusing. The
isoelectric points reported in Tab. 5 range
from 3.2 (A47934) to 8.1 (ristocetin, A477,

and A42867).
Retention times from reverse-phase HPLC
give an approximate indication of the relative
lipophilicity. In Tab. 6 the values obtained
with representative compounds are listed.
All glycopeptides exhibit similar UV ab-
sorption spectra with maxima at about
280 nm in acidic or neutral media that shift to
292-300 nm under alkaline conditions. NMR
experiments led to complete assignment of
each hydrogen, carbon, and nitrogen atom for
the majority of the glycopeptides and greatly
contributed to structure elucidation of both
natural and semisynthetic compounds. Al-
though almost 1000 natural and semisynthetic
Fig. 6. Stereo model of teicoplanin aglycone.
glycopeptides are presently known no crystal-
line forms suitable for X-ray analysis have
been reported, except for a degradation prod-
uct of vancomycin (designated as CDP-I) that no acids is lR, 2R, 3S, 4R, 5R, 6S, and 7S, re-
allowed the determination of the absolute spectively. The NMR spectra of vancomycin
configuration of vancomycin (SHELDRICK et and of glycopeptides discovered later show
al., 1978). The configuration of the seven ami- that the stereochemistry and the overall
three-dimensional conformation is the same
in all compounds except for a few epimers at
amino acid residue 1. In teicoplanin five of
Tab. 6. Reverse-Phase HPLC Retention Times (tR) the amide bonds are trans, the 5-6 bond is cis:
of Some Dalbaheptides" modification of the bond conformation often
occurs during chemical reactions. As an ex-
Antibiotic tR [min]
ample, the three-dimensional structure of tei-
AM374 7 coplanin aglycone is presented in Fig. 6.
Vancomycin 10.5
Ristocetin 11
Chloropolysporin 12
A35512B 12.5
Actaplanin
Avoparcin 14
13 3 Biological Activity
A47934 17
OA-7653 20 3.1 In vitro Antibacterial Activity
A41030 21
A477 23
Ardacin 24-26 Glycopeptide antibiotics are active on most
Teicoplanin 24-28 gram-positive aerobic and anaerobic bacteria
A40926 30 (CAMFOLI-RICHARDS et al., 1990). Gram-ne-
gative bacteria are generally insensitive ex-
a On Ultrasphere ODS 5 pm (250-4.6 mm) column cept for Neisseria, Gardnerella, and Branha-
(Beckman); mella strains on which several natural or
mobile phases: semisynthetic glycopeptides exert a weak ef-
(A) 0.02 M aq NaH2P04/CH3CN9: 1, pH 6;
(B) 0.02 M aq NaH2P04/CH3CN3 :7, pH 6; fect. However, most strains of the genera
gradient from 5%-60% of eluent (B) in 40 min; Lactobacillus, Leuconostoc, and Pediococcus
flow rate 1.6 mL.min-'. - although gram-positive - are insensitive
(RIVA,E. and SOFFIENTINI,A., personal communi- (JOHNSON et al., 1990). The minimal inhibito-
cation) ry concentrations (MIC) of some glycopep-
Tab. 7. Comparison of Antibacterial Activity (MIC, pg.mL-') of Some Dalbaheptides (GOLDSTEIN

et al.,
1987)
Antibiotic Staphylo- Staphylo- Strepto- Entero- Propioni- Bacteroides Neisseria

coccus coccus coccus coccus bacterium fragilis gonorrhoeae
aureus epidermidis pyogenes faecalis acnes ATCC ISM 681126
TOUR ATCC C 203 ATCC 7080 ATCC 6919 23745
12228
A40926 0.06 0.06 0.06 0.06 0.016 64 2

A47934 0.06 0.03 0.13 0.13 nd nd 8
Teicoplanin 0.13 0.13 0.06 0.13 0.13 128 32
Vancomycin 0.25 0.5 0.13 0.5 nd nd 32
Ristocetin 4 2 0.25 1 nd nd 64
A35512B 1 0.5 0.13 0.5 nd nd 64
A41030A 0.03 0.008 0.13 0.13 nd nd 64
Aridicin 1 4 0.13 2 nd nd 64
Avoparcin 2 2 0.25 0.25 nd nd 128
Actaplanin 1 2 0.13 0.5 nd nd > 128
nd: not determined
tides on representative pathogens are re- bacterial populations is supported by several

ported in Tab. 7. Since - as discussed later - experimental data (WALLASand STROMIN-
the activity of these antibiotics is due to their GER,1963; SOMMER et al., 1984) the most rel-
ability to bind peptidoglycan precursors that evant of which are:
are similar in gram-positive and gram-nega- - Addition of a glycopeptide to a bacterial
tive bacteria, the insensitivity of the latter is culture blocks the uptake of peptidoglycan
attributed to a lack of penetration through precursors such as acetyl glucosamine or
the outer membrane. It is, however, notewor- diamino pimelic acid several minutes be-
thy that some semisynthetic derivatives exhi- fore any effect can be noticed on the up-
bit a certain activity against Escherichia coli, take of precursors of DNA, RNA, or pro-
Proteus vulgaris, and Pseudomonas aerugino- teins.
sa (MALABARBA et al., 1993a). - Inhibition of uptake is accompanied by ac-
The effect of glycopeptides on staphylococ- cumulation of intermediates of peptidogly-
ci and streptococci is clearly bactericidal at can biosynthesis, notably UDP-muramyl-
concentrations slightly higher than the MICs. pentapeptide, indicating that these anti-
Their bactericidal effect is weaker in Entero- biotics do not interfere with early steps of
coccus faecium or E. faecalis (CAMPOLI- peptidoglycan synthesis but affect a later
RICHARDSet al., 1990). Lethality can be ac- stage of the process.
companied by cell lysis depending on the The observation that vancomycin activity is
strain and on environmental conditions. reversed by cell wall fragments or by late in-
termediates of the peptidoglycan biosynthetic
pathway gave the first indication of the action
of glycopeptides on a molecular level. The
3.2 Mechanism of Action classical work of NIETO and PERKINS
(1971a, b) established clearly that glycopep-
A most interesting property of glycopeptides bind with relatively high affinity to pep-
tide antibiotics is their ability to block cell tides having a D-Ala-D-Ala carboxyl termi-
wall formation in sensitive bacteria by bind- nus. A lower binding affinity is observed
ing to precursors of peptidoglycan synthesis when other amino acids substitute for one of
(REYNOLDS, 1989; NICASand ALLEN,1994). these residues. The tripeptide N,N-diacetyl-L-
Inhibition of cell wall synthesis in growing Lys-D-Ala-D-Ala was used by several authors
as a model of the natural ligand to determine Enterococci resistant to vancomycin or tei-

the binding affinity of different antibiotics. coplanin have been isolated with increasing
An association constant of 1.5.10" L-mol-' frequency in the last years, especially in inten-
was calculated for vancomycin. The values sive care units. Three major phenotypes
obtained with ristocetin and teicoplanin were termed VanA, VanB, VanC can be distin-
5.9-105 and 2.6.10" Lsmol-', respectively guished in resistant Enterococcus species
(SOMMAet al., 1984). (ARTHURand COURVALIN, 1993; NICASand
Experiments performed to establish which ALLEN, 1994). Enterococci carrying the
biochemical reaction is specifically inhibited VanA gene are characterized by high-level re-
in the peptidoglycan biosynthetic pathway sistance to vancomycin and teicoplanin. A
have not achieved decisive results so far. Gly- cluster of genes the expression of which is as-
copeptides are most probably unable to cross sociated with the VanA phenotype has been
the bacterial cytoplasmic membrane as was identified in these strains on a transposon de-
experimentally shown with iodinated vanco- signated Tn2546. Biochemical characteriza-
mycin (PERKINS and NIETO,1970); therefore, tion of the gene products has revealed that
their action appears to be limited to events the resistance is due to an unusual pathway of
occurring at the membrane surface, such as peptidoglycan biosynthesis in which UDP-
transglycosylation or transpeptidation reac- muramyl-tetrapeptide-D-lactateis synthesized
tions (PERKINS,1982). Vancomycin and tei- instead of the normal precursor UDP-mura-
coplanin were reported to inhibit transglyco- myl-pentapeptide. Apparently, all enzymes
sylation in a cell free membrane system. catalyzing the subsequent reactions of the
However, the inhibitory effect was observed pathway accept as substrates the interme-
only at antibiotic concentrations substantially diates carrying this modification. The result is
higher than the MICs of these drugs (SOMMA a peptidoglycan structure in which the residue
et al., 1984). D-Ala-D-Lac substitutes for D-Ala-D-Ala. As
Another aspect difficult to be interpreted is glycopeptides do not form complexes with
the often observed poor correlation between peptides ending with D-Ala-D-Lac the strain
the degree of affinity to model tripeptides results resistant to their action.
and the antimicrobial activity of the antibiot- Phenotype VanB is characterized by mod-
ic. An extreme case is that of erernomycin, erate or high-level resistance to vancomycin
consistently more active than vancomycin but and sensitivity to teicoplanin. The mechanism
demonstrating a significantly lower affinity to of resistance is the same as of the VanA phe-
Ac2-Lys-D-Ala-D-Ala (GOODet al., 1990). notype. However, in VanB strains expression
of resistance genes is inducible by vancomy-
cin but not by teicoplanin which explains the
3.3 Resistance different susceptibility to the two antibiotics.
The VanC phenotype comprises species
To date, despite many years of clinical use, such as Enterococcus gallinarum in which the
vancomycin resistance has not been reported pentapeptides of peptidoglycan and its bio-
for Streptococcus sp. or Staphylococcus aure- synthetic precursors constitutively possess a
us clinical isolates. In vitro selection for van- D-Ala-D-Ser ending. These species show low-
comycin or teicoplanin resistance in S. aureus level resistance to vancomycin and sensitivity
is difficult and strains with only modest in- to teicoplanin.
creases in MIC are obtained (JOHNSON et al.,
1990). A decreased susceptibility to vancorny-
cin and more frequently to teicoplanin has 3.4 In vivo Efficacy and
been observed with coagulase-negative sta-
phylococci, such as S. epidermidis or S. hae- Pharmacology
molyticus. The biochemical mechanism deter-
mining this low-level resistance has not been The efficacy of the principal glycopeptides
elucidated yet (ARDUINOand MURRAY, in curing experimental infections has been as-
1993). sessed in septicemia models in mice (PAL-
Tab. 8. Experimental Septicemia in Mice"
Antibiotic Infecting Organism (Clinical Isolates)
S. aureus L 165 S. pyogenes L 49 S. pneumoniae L 44
MIC ED50 MIC ED50 MIC ED50

[pg * mL - '1 [mg- kg -'I [pg - mL -'I [mg - kg -'I [pg-mL-'I [mg .kg - '1
Teicoplanin 0.4 0.72 0.05 0.11 0.1 0.41
Vancomycin 0.8 7.2 0.8 0.58 0.8 1.9
Ampicillin 0.1 8.1 0.02 0.1 0.02 4.1
Cephaloridine 0.05 2.8 0.01 0.03 0.02 0.93
Erythromycin 0.5 28 0.05 0.44 0.01 26
a Animals treated subcutaneously once daily for three days (PALLANZA

et al., 1983).
LANZA et al., 1983). The results obtained with glycopeptides are produced by streptomy-
teicoplanin and vancomycin are reported in cetes and practically all of them are of the ris-
Tab. 8 in comparison with other antibiotics. tocetin type, although ristocetin itself was iso-
In general, a higher in vivo efficacy correlated from strains originally considered as
sponds to a higher in vitro activity although Nocardia or Proactinomyces and later classi-
no simple correlation is apparent. Eremomy- fied as Amycolatopsis (LECHEVALIER et al.,
cin appears to be more active than vancomy- 1986). A large proportion of known glycopep-
cin both in vitro and in experimental infec- tide is generated by this genus. In fact, the
tions. Ardacins are less active in vivo than vancomycin producing strain has been in turn
vancomycin although having a comparable classified as Streptomyces, then as Nocardia,
antibacterial activity. This also holds for and at present as Amycolatopsis; several gly-
A40926 when compared to teicoplanin. copeptides of the vancomycin type are pro-
A few products have been tested in more duced by A . orientalis and most probably sev-
complex models of experimental infection. eral other strains should be reclassified into
Teicoplanin and vancomycin were found ef- the same genus. Actinoidin producers are
fective in reducing the bacterial load in ex- probably also Amycolatopsis although origi-
perimentally induced endocarditis in rats and nally considered as Proactinomyces or Nocar-
rabbits (GOLDSTEIN et al., 1994). dia. Lipoglycopeptides are produced by a va-
A clinical overview of vancomycin has been riety of rare actinomycetes such as Actino-
published by ZECKELand WOODWORTH madura, Actinoplanes, Streptosporangium,
(1994). Toxicology, pharmacokinetics, phar- and the new genus Kibdelosporangium.
macology, and therapeutic use of teicoplanin Interestingly, glycopeptides that are minor
have been extensively reviewed by GOLD- components of complexes produced by a mi-
STEIN et al. (1994) and BROGDEN and PET- crobial strain are the main products of other
ERS (1994). soil isolates. Partially purified vancomycin
produced by A . orientalis NRRL 2450 con-
tains minor quantities of at least nine struc-
turally related compounds (NAGARAJAN,
1993). One of them, A51586A, is the main
4 Producing Organisms product of N. orientalis NRRL 15232 (HUNT
et al., 1984b).
All known glycopeptides are produced by
microorganisms of the order of Actinomyce-
tales isolated from soil samples collected from
almost everywhere. Surprisingly, only a few
6 Fermentation 381
5 Methods of Screening by testing fermentation broths against vanco-

mycin in a polyclonal antibody assay
(ELISA) (YAO et al., 1988).
The first glycopeptide antibiotics were dis-
covered by conventional screening proce-
dures such as growth inhibition of a test or-
ganism on agar plates. Later, specific tests
were devised to detect inhibitors of cell wall 6 Fermentation
synthesis. In the Lepetit Laboratories, fer-
mentation broths were tested for differential Culture media and fermentation conditions
inhibition of a S.aureus strain and an L-form reported for production of glycopeptide anti-
derived from it. In this way teicoplanin was biotics either at the laboratory level or in in-
discovered. Similarly, izupeptin was found on dustrial manufacturing do not substantially
the basis of its differential activity on cell wall differ from those described for other antibiot-
possessing bacteria and on mycoplasma ics. As usual, precultures (vegetative cultures)
(SPIRI-NAKAGAWA et al., 1986). are prepared in one or more stages using rich
Subsequently, very efficient screening tests media composed of soluble proteins and rap-
were implemented exploiting the unique idly utilizable carbon sources. Fermentation
mechanism of action of the antibiotics. One conditions must be adjusted individually for
of them was based on affinity chromatogra- the different producing strains. In general,
phy; matrix bound D-Ala-D-Ala was prepared optimal temperatures are between 28 and
by forming a peptide bond between amino- 30°C, and the optimal pH is around 7.0. All
caproylSepharose and the amino group of producing organisms are strictly aerobic and a
the dipeptide (CORTIand CASSANI,1985). A consistent oxygen supply must be provided by
different antimicrobial activity of culture fil- adequate aeration and agitation systems.
trates before and after passage on the affinity
resin was taken as evidence for the presence
of a glycopeptide that could easily be isolated 6.1 Fermentation Media - Carbon
by elution. By this method 72 strains (about
0.3% of the fermentation broth tested) were Sources
identified as producers of glycopeptides in a
screening campaign (CASSANI,1989); about Carbon sources frequently used in fermen-
60% were ristocetin producers. Among the tation media include glucose, glycerol, starch,
others teicoplanin, avoparcin or actaplanin or dextrin. The use of oleate or other lipids is
producing strains were identified, and two less common.
new glycopeptides (A40926 and A42867) Antibiotic production is often repressed by
were discovered as well. the presence of rapidly utilizable carbon
RAKEet al. (1986) applied an assay based sources in the production medium. A few
on the affinity of glycopeptides to the tripep- studies of this effect have been reported for
tide N,N-diacetyl-L-Lys-D-Ala-D-Ala to 1936 glycopeptide fermentation. There is evidence
cultures. Reversion of antibacterial activity by that antibiotic production by Amycolatopsis
addition of this receptor analog indicated the species is subjected to this repression, since
presence of a glycopeptide. The test proved higher yields of ristomycin (ristocetin) are ob-
to be very specific: 42 glycopeptides were iso- tained when galactose or glycerol are used as
lated, 6 of which were novel ones including carbon sources in Amycolatopsis lurida fer-
kibdelin, parvodicin, and actinoidin A*. mentations (TOROPOVAet al., 1982). Similar-
A colorimetric assay (SPERA) was pro- ly the production of demethyl vancomycin is
posed (CORTIet al., 1985) based on the com- increased when galactose or dextrin substi-
petition of horseradish peroxidase bound tei- tutes for glucose in the fermentation medium
coplanin and the putative glycopeptide for a (BOECKet al., 1984); dextrin or starch are the
D-Ala-D-Ala peptide attached to the test preferred carbon sources in the fermentations
tube. Glycopeptide A82846 was discovered of vancomycin and of the related antibiotic
izupeptin (MERTZand DOOLIN,1973; SPIRI- 6.3 Fermentation Media - Effect

NAKAGAWAet al., 1986). All these antibiot-
ics are produced by Amycolatopsis strains. A of Phosphates
marked negative effect of glucose is observed
in the production of the antibiotic complex Phosphate concentration is an important
A41030 by cultures of Streptomyces virginiae parameter in antibiotic fermentation pro-
(BOECKet al., 1985). cesses as phosphates are necessary for opti-
In contrast, Actinoplanes species appear to mal growth, but high concentrations inhibit
be less affected by the nature of carbon production.
sources. Only minor differences in actaplanin In vancomycin fermentations antibiotic
yields are observed when dextrin or other production is depressed by phosphate con-
slowly metabolized carbon sources substitute centrations above 0.1 g.L-' (MERTZ and
for glucose in Actinoplanes missouriensis fer- DOOLIN, 1973). Even lower concentrations
mentations. A medium composed of glucose are sufficient to inhibit production of deme-
and yeast extract appears suitable for teico- thy1 vancomycin or ristocetin (BOECKet al.,
planin production by cultures of A . teichomy- 1984; TOROPOVAet al., 1974). The fact that
ceticus. these antibiotics belong to different structural
A substantial increase in yields of ardacins types but are all produced by Amycolatopsis
and kibdelins, complexes produced by Kib- species suggests that phosphate control is
delosporangium strains, is observed up on ad- characteristic for this genus.
dition of oleic acid (as methyl ester) to the Production of ardacins by Kibdelosporan-
fermentation medium (SHEARERet al., 1985). gium aridum and of A47934 by S. toyocaensis
Oleic acid is a good source of acetyl CoA - is also inhibited by relatively low concentra-
one of the starting materials of these antibiot- tions of phosphates (BOECK and MERTZ,
ic biosyntheses. Higher availability of acetyl 1986). In contrast, production of the complex
CoA was therefore proposed to explain the A41030 by S. virginiae was enhanced by in-
oleate effect on yields. However, since arda- creasing phosphate concentrations up to
cins and kibdelins are lipoglycopeptides oleic 1 g.L-' (BOECKet al., 1985).
acid possibly is a direct precursor of their fat-
ty acid chain as it was shown for teicoplanin, a
lipoglycopeptide (see Sect. 8.2).
6.4 Inhibition by the Final Product
6.2 Fermentation Media - Several examples of inhibition of antibiotic

biosynthesis by the final fermentation prod-
Nitrogen Sources uct are reported in the literature. This phe-
nomenon has also been observed in some gly-
Soybean meal or similar soy products are copeptide fermentations. Ristocetin produc-
almost universally used as nitrogen providing tion, e.g., is inhibited by low concentrations of
ingredients in production media; yeast extract this antibiotic (TOROPOVAet al., 1974). It
and corn steep liquor are also used frequent- was observed in the authors' laboratory that
ly. Ammonium salts are never found among A40926 inhibits its own production when
the suitable nitrogen sources. Although spe- present in the growth phase of cultures.
cific studies are lacking it can be assumed that Well documented is the effect of actaplanin
- similarly to other antibiotics - glycopeptide on its producing strain. A. missouriensis cul-
production is repressed by high concentra- tures start producing actaplanin about 40 h
tions of ammonium ions. Nitrates can be uti- after inoculation at the end of mycelial
lized; at least in the case of S. virginiae fer- growth. Depending on the fermentation me-
mentations addition of sodium nitrate to a dium 60-90% of the antibiotic produced is
soybean and corn steep containing medium tightly bound to the mycelium. When myce-
increased production of A41030 by 80% lium was centrifuged and resuspended in sal-
(BOECKet al., 1985). ine no further growth was observed but pro-
7 Recovery and Purification 383
duction continued normally. When mycelium

was centrifuged, washed to completely elimi-
7 Recovery and
nate soluble actaplanin, and resuspended,
growth was resumed and no production was
Purification
observed. Thus, even low concentrations of Although glycopeptides are relatively wa-
actaplanin in the medium apparently inhibit ter-soluble, often a large proportion of the
growth and also have a killing effect as de- product is found in the harvest broth bound
monstrated by viable colony counting in a to mycelium. Solubilization and release can
parallel experiment. Surprisingly the cell normally be obtained by adjusting the pH to
bound antibiotic that can be as much as 50 mg an appropriate value (according to the iso-
per gram of cells has no effect on either electric point of the product acidic or alkaline
growth or production (HUBERet al., 1987). conditions are required). Alternatively ex-
traction with acetone or methanol has been
used. Recovery from the filtered broth may
6.5 Control of Complex include adjustment of pH and use of water
immiscible organic solvents (butanol for tei-
Composition coplanin), or ion-pair extraction (avoparcin).
Other isolation schemes described use ion ex-
An antibiotic composed of several factors change matrices (Dowex, Amberlite IR9),
has to be produced during fermentation in acidic alumina, cross-linked polymeric ad-
constant relative amounts to ensure uniformi- sorbents (Diaion HP, Amberlite XAD), ca-
ty of the finished product. Alternatively, the tion exchange dextran gel (Sephadex) and po-
relative amount of the major component can lyamides in various sequences (SITRINand
be increased to obtain as a final product a sin- FOLENA-WASSERMAN, 1994).
gle substance rather than a complex. The fac- Reverse-phase chromatography with semi-
tors composing the lipoglycopeptide antibiot- preparative and preparative columns, packed
ics teicoplanin and A40926 differ in the struc- with silanized silica gel, allowed purification
ture of their acyl chains, in which the and separation of glycopeptides and of the
branched type, either is0 or anteiso, predomi- single factors of complexes on the basis of dif-
nates. In biosynthesis of branched fatty acids ferential hydrophobicity. Isocratic and gra-
the initiator molecule determines the type of dient elutions were carried out. Examples are
the final product: isobutyric acid gives rise to given by SITRINand FOLENA-WASSERMAN
is0 chains with an even number of carbon (1989) and in almost every publication listed
atoms, 2-methylbutyric acid and isovaleric in Tab. 1-4.
acid give rise to chains with an odd number of The elucidation of the mechanism of action
carbon atoms of the anteiso and the is0 type, at a molecular level allowed the development
respectively. Addition of one of these precur- of affinity chromatography both as a specific
sors to the fermentation broth can selectively discovery tool and as a powerful method of
increase the amount of the fatty acid that it is isolation and purification of glycopeptides
initiating and thus of the corresponding com- from fermentation broths (CORTI and CAS-
ponent of the complex. It has been shown SANI, 1985; FOLENA-WASSERMAN et al.,
that in this way the complex composition of 1987). The affinity adsorbent is prepared by
both teicoplanin and A40926 can be substan- immobilizing D-Ala-D-Ala on commercially
tially altered (BORGHIet al., 1991a; SELVAet available activated supports. The affinity con-
al., 1992). In practice, better results could be stant for the sepharose-D-Ala-D-Ala resin, as
achieved by adding their natural precursors, determined by equilibrium binding experi-
i.e., valine for isobutyric acid, isoleucine for ments, is 8.08.105 Lemol-' for teicoplanin,
2-methylbutyric acid, and leucine for isovaler- 1.57.105 L.mo1-l for vancomycin and
ic acid, to the cultures instead of the acids 3.89.105 Lamol-' for ristocetin A (CORTI
mentioned above. and CASSANI,1985).
An aqueous solution of the glycopeptide is
contacted with the adsorbent in batch mode
or loaded on columns. After washing to re- Experimental studies to elucidate the bio-
move impurities the glycopeptides are eluted synthetic pathway are scanty. In fact, al-
with a small volume of an aqueous buffer so- though it appears to be most probable that as-
lution and an organic solvent (acetonitrile, sembly is performed through the well known
methanol, ethylene glycol) to disrupt the in- thiotemplate system, no experimental evi-
teraction. By varying the pH of the buffer, dence is available on this most important as-
mixtures of different glycopeptides or compo- pect. The origin of the unusualsugars and the
nents of a complex can be resolved, with the reactions leading to the formation of the tri-
additional advantage of obtaining a concen- phenyl ether and diphenyl groups have never
trated solution. The final pure material can be been studied. This lack of information is, in
obtained after desalting by lyophilization. part, due to the fact that several laboratories
failed to isolate blocked mutants accumulat-
ing biosynthetic intermediates. In addition,
the most intensely studied antibiotics, i.e.,
vancomycin, teicoplanin and ristocetin, are
8 Biosynthesis produced by Acfinomyces strains for which a
system of transformation with exogenous
Glycopeptides are complex molecules the DNA is not available. This has severely ham-
biosynthesis of which must include several pered studies based on molecular genetic
steps: methods.
(1) synthesis of the uncommon amino acids
constituting the heptapeptide,
(2) polymerization of the amino acids, 8.1 Origin of the Uncommon
(3) formation of the ether and carbon-car-
bon bonds between the phenylic groups, Amino Acids
(4) synthesis of the unusual sugars and, if
present, of fatty acid chains, A few uncommon amino acids are found as
(5) glycosylation and other final modifica- building blocks of almost all glycopeptide an-
tions. tibiotics: fl-hydroxytyrosine, 3-chloro-p-hy-
The sequence of these steps may, to some droxytyrosine, p-hydroxyphenylglycine and
extent, differ from that indicated; chlorination m-dihydroxyphenylglycine (Fig. 7).
of the phenyl rings, e.g., may occur before or Analysis of 13C NMR spectra of avoparcin
after the amino acid assembly and glycosyla- produced by fermentations of Sfrepfomyces
tion may precede the oxidative ring linking. candidus which had been supplied with 2-13C-
$
HO
OH
&o o& $o
0 0
Fig. 7. Amino acids com-
HO OH HO / OH / posing teicoplanin hepta-
OH OH peptide.
8 Biosynthesis 385
tyrosine demonstrated that tyrosine is the nylglyoxylic acid or p-hydroxy mandelic acid
precursor of the p-hydroxytyrosine and p-hy- depresses the incorporation of radioactivity
droxyphenylglycine residues of the peptide from L-U14C-tyrosine into ardacin indicating
(MCGAHRENet al., 1980). These results were dilution of labeled intermediates (CHUNGet
confirmed and extended by studies on vanco- al., 1986a). It is noteworthy that p-hydroxy-
mycin. Experiments with 13C and 3H labeled phenyl acetic acid has no effect, providing
tyrosine showed that both L- and D-tyrosine evidence that this compound is not an inter-
are precursors of the two 3-chloro-phydroxy- mediate in the conversion. These results and
tyrosine units, although these differ in their the proof that the producing organisms are
configuration at C-2. Moreover, it was found able to convert tyrosine into p-hydroxytyro-
that phydroxylation occurs with retention of sine are consistent with the following reaction
configuration at C-3 (HAMMOND et al., 1982). sequence (Fig. 8):
Similar results were obtained by adding D,L- Tyrosine + p-hydroxytyrosine +
2- 13C-tyrosineto ristocetin fermentations. As pp-dihydroxyphenylpyruvic acid +
expected, the label was incorporated into C-2 p-hydroxymandelicacid +
of the phydroxytyrosine and C-1 of the p-hy- p-hydroxyphenylglyoxylicacid +

droxyphenylglycine residues (HAMMOND et p-hydrox yphenylglycine.
al., 1983). The origin of m-dihydroxyphenylglycine
Although the origin of p-hydroxyphenyl- units has been demonstrated by experiments
glycine from tyrosine is clearly demonstrated, in which 1,2-13C-acetatewas added either to
the sequence of reactions and the interme- vancomycin or ristocetin fermentations
diates of this conversion can be deduced only (HAMMOND et al., 1982). The carbons of the
from indirect evidence. p-Hydroxyphenyl- m-dihydroxyphenylglycine residues resulted
glyoxylic acid appears to be the direct precur- specifically labeled, indicating that this amino
sor of p-hydroxyphenylglycine, since deme- acid backbone originates from cyclization of a
thy1 vancomycin and A47934 yields are in- polyketomethylene chain. A tetraketide chain
creased by addition of either of these com- was originally suggested as the immediate
pounds to A. orientalis or S. toyocaensis cul- precursor. However, this requires a ring clo-
tures, respectively (BOECK et al., 1984; sure involving the initial methyl group of the
BOECKand MERTZ,1986). In K. aridurn fer- chain, a feature never observed in biosynthe-
mentations, addition of cold p-hydroxyphe- sis of polyketide antibiotics. It was, therefore,
0
HOJ COOH
OH OH OH
6
tyrosine
0
HoY
Q-
COOH COOH
-H2$ OH
Fig. 8. Presumptive pathway of p-hydro- OH OH

xyphenyl glycine biosynthesis from tyro-
sine. p-hydroxyphenylglycine
3.5-dihydroxyphenylacetic add
Fig. 9. Hypothetical po-

lyketide chains that can
give rise to 3,5-dihy-
droxyphenyl acetic acid
and to 3,S-dihydroxy-
phenyl glycine.
proposed that a longer chain is formed first, conclusion is based on the following experi-
part of which could be degraded after cycliza- mental evidence:
tion (HAMMOND et al., 1983). An alternative - Production of teicoplanin factor T-A2-1 by
hypothesis can be considered assuming that A. teichomyceticus characterized by a 4-de-
malonate rather than acetate is the initiator cenoyl moiety is entirely dependent on the
molecule of the polymerization process. The presence of linoleic acid in the fermenta-
resulting product, 6-carboxy-3,5-dihydroxy- tion medium.
phenylacetic acid, could be easily converted - The relative amount of factor T-A2-3 pro-
into m-dihydroxyphenylacetic acid by decar- duced that is characterized by a linear deca-
boxylation (Fig. 9). noyl chain is substantially increased by ad-
dition of oleic acid to the medium.
- Factors T-A2-2, T-A2-4, and T-A2-5 bear
8.2 Origin of Fatty Acids of branched acyl chains, namely S-methylnon-
Lipoglycopeptides anoic acid (iso-C10: 0), 8-methyldecanoic
acid (anteiso-C11:0 ) and 9-methyldecanoic
All glycopeptides listed in Tab. 4 are com- acid (iso-C11 :O). Analysis of fatty acid con-
plexes of factors characterized by the pres- stituents of cell lipids revealed the presence
ence of different acyl chains linked, as am- of three major components, 14-methylpen-
ides, to amino sugars. The origin of the fatty tadecanoic acid (iso-C16:0), 14-methyl-
acids constituting the acyl moieties of the tei- hexadecanoic acid (anteiso-C17:0 ) and 13-
coplanin components (see Fig. 5) has been ex- methyltetradecanoic acid (iso-Cl5 :0).
tensively investigated (BORGHIet al., 1991a). These appear to be the logical precursors
Altogether the results obtained indicate that of the iso-C10:0, anteiso-Cll:O, and iso-
these chains are not synthesized de novo but C11 : O moieties of T-A2-2, T-A2-4, and T-
derived from degradation of long-chain fatty A2-5, respectively, assuming the loss of
acid components of cell lipids or from those acetate units by the common p-oxidation
present in the fermentation medium. This mechanism of fatty acid degradation.
9 Chemical Modifications 387
- A mutant strain of A . teichomyceticus pro- cule was never achieved; the aglycone was
duces a novel teicoplanin factor character- rapidly transformed into mannosyl aglycone
ized by a n-nonanoic moiety. Correspond- but no further glycosylation step was ob-
ingly, cell lipids contain heptadecenoic acid served (BORGHIet al., 1991b).
not present in parent strain cells. Antibiotic A47934 produced by S. toyo-
- Addition of 14C-acetate to the culture me- caensis has a chemical structure similar to that
dium at the time of inoculation resulted in of ardacin aglycone from which it differs in
substantial labeling of fatty acid moieties. the presence of a sulfate ester on an aromatic
When 14C-acetatewas added to grown my- ring. Extensive experiments with labeled sub-
celium resuspended in saline, radioactivity strates, blocked mutants, and biochemical in-
of acyl chains was negligible in comparison hibitors indicate that the sulfate is added pri-
to that of teicoplanin aglycone, demonstrat- or to the formation of intermediates that pos-
ing that the fatty acid moieties derive from sess antimicrobial activity. These results ex-
molecules formed during the growth clude that sulfate esterification of the agly-
phase. cone that is antimicrobially active is the last
A similar correspondence between acyl reaction of the biosynthetic pathway (ZMI-
chains of antibiotic factors and cell lipid com- JEWSKI et al., 1987).
position was found for A40926 and its pro-
ducer Actinomadura strain (ZERILLIet al.,
1992).
9 Chemical Modifications
8.3 Glycosylation and Final Early investigations on glycopeptide chem-
Modifications istry were carried out mainly for structural
studies. Hydrolysis reactions yielded the agly-
The last steps of the biosynthetic pathway cones and pseudoaglycones, some of which
were studied in some detail in ardacin pro- exhibited a higher antimicrobial activity than
duction by K. aridum. Sugar substituents in the parent antibiotic (PHILIPet al., 1960).
these antibiotic molecules are mannose and a Later, knowledge of the mechanism of action
N-acyl-2-aminoglucuronic unit. A related on a molecular level, elucidation of the struc-
complex of antibiotics, kibdelins, differs from ture of several compounds, and establishment
ardacins in the presence of N-acylglucosam- of some correlations between pharmacokinet-
ine instead of acylaminoglucuronic acid. Kib- ic and physicochemical properties (PITKINet
delins are readily converted into ardacins by al., 1986) allowed a more rational design of
cultures of K. aridum indicating that oxida- chemical derivatives. In particular, the effica-
tion of C-6 of glucosamine could be the last cy of teicoplanin in experimental infections
biosynthetic step (CHUNGet al., 1986a). pointed out the importance of lipophilic side
Less clear are the results of experiments chains and opened the way to the oriented
aimed at defining the glycosylation sequence. synthesis of a large number of derivatives
When 14C labeled ardacin aglycone or man- both of teicoplanin and other glycopeptides.
nosy1 aglycone were added to K. aridum cul- Products with improved antimicrobial activity
tures, labeled ardacins were produced or better pharmacokinetics have been re-
(CHUNGet al., 1986a). However, since transported, but none of them is in clinical use so
formation required a long incubation period far. Structure-activity relationships of natural
and only a small fraction of the added ra- and semisynthetic glycopeptides have been
dioactivity was incorporated, the possibility of reviewed by NAGARAJAN (1994b). The ideal
a degradation of the molecule and recycling glycopeptide has been outlined in a recent ar-
of the fragments cannot be ruled out. ticle by FELMINGHAM (1993).
Similar attempts were carried out by ad- An extensive review of chemical modifica-
ding teicoplanin aglycone to A. teichomyceti- tions is out of the scope of this chapter, there-
cus cultures. Conversion into the final mole- fore only a condensed summary is given.
Reaction conditions are those conventional molecule with physicochemical properties

in peptide chemistry including suitable pro- similar to those of teicoplanin, and the pep-
tection of active functions not involved in the tidic carboxyl group was condensed with a
reaction. However, their application some- number of amines. Among the derivatives ob-
times requires specific conditions since differ- tained the -NH(CH2)3N(CH3)2 amide (MDL
ent glycopeptides treated with the same rea- 63,246) exhibits a higher antimicrobial activi-
gent under relatively similar conditions yield ty than the parent compound on several pa-
different results depending on the nature of thogens, including vancomycin and teicoplan-
amino acids 1 and 3, on steric factors, and on in resistant strains (MALABARBAet al.,
the presence of sugars or chlorine atoms. 1993b) .
A common modification of glycopeptides is It was suggested that a protonated amino
the selective hydrolysis of sugars leading to group plays an important role in the binding
the preparation of aglycones and pseudoagly- process through an electrostatic interaction
cones. Reports are available, e.g., on avopar- with the carboxyl group of the target peptide
cin (MCGAHRENet al., 1983), teicoplanin (WILLIAMSON et al., 1984). The terminal ami-
(MALABARBAet al., 1984), parvodicin no group of several glycopeptides such as tei-
(CHRISTENSEN et al., 1987), ardacin (SITRIN coplanin, vancomycin, ristocetin, and eremo-
et al., 1986), and eremomycin (KOBRINet al., mycin has been variously alkylated, and acy-
1988). The hydrolyses were performed mainly lated. In vancomycin the higher basicity of
for investigational purposes but also provided the vancosamine amino group compared to
compounds of interest for their intrinsic activ- that of the terminal N-methylleucine allowed
ity or as substrates for further chemical trans- a selective acylation of the former (KANNAN
formations. et al., 1988; NAGARAJAN et al., 1988).
Since some components of certain glyco- To understand the role of the terminal
peptide complexes differ only in the number “carboxylate binding pocket” in glycopep-
or position of C1 atoms, partial or complete tide-substrate interactions binding constants
removal of chlorine converts one glycopep- of glycopeptide fragments (produced by con-
tide into another, e.g., A82846B into orienti- trolled degradations) were determined with
cin A (NAGARAJAN et al., 1989a). Selective peptide models. NAGARAJAN and SCHABEL
dechlorination of vancomycin (HARRISet al., (1988) prepared the “vancomycin hexapep-
1985) and teicoplanin (MALABARBA et al., tide” by removal of the terminal N-methyl-
1989b) and its effect on the affinity of the an- leucine, and MALABARBAand CIABATTI
tibiotics for model peptides and on their anti- (1992) reported a tetrapeptide fragment from
bacterial activity have been reported. Iodov- teicoplanin.
ancomycin (HARRISet al., 1986) and 12’I-la- On the other hand, relevant peptide frag-
beled ristocetin (KIM et al., 1989) were also ments were obtained by stereospecific syn-
described. theses with a view of total synthesis of vanco-
The terminal carboxyl group which may be mycin (RAMARAO et al., 1992) and teico-
free as in teicoplanin or methylated as in ris- planin (CHAKRABORTY and VENKATRED-
tocetin has been submitted to esterification, DY, 1992). Both these fragments and those
reduction to alcohol, or amidation. Teico- prepared by controlled degradation may be
planin, its pseudoaglycones, and its aglycone used as building blocks for the synthesis of
have been amidated with various amines and non-natural glycopeptides.
alkyldiamines (MALABARBA et al., 1989c), a- The various approaches to the synthesis of
amino acids (MALABARBAet al., 1989a), or vancomycin and ristocetin aglycones have
polyamines (MALABARBA et al., 1992). been reviewed by EVANS and DEVRIES
A40926, an antibiotic characterized like ar- (1994).
dacins by an N-acylglucuronic moiety shows a
good antibacterial activity on several patho-
gens, including N. gonorrhoeue (GOLDSTEIN
et al., 1987). The glucuronic carboxyl group
was reduced to hydroxymethyl to obtain a
10 Biotransformations 389
10 Biotransformations 10.2 Glycosylation

Cultures of A. teichomyceticus, the produc-
10.1 Deglycosylation er of teicoplanin, and of K. aridum, the pro-
ducer of ardacins, easily 'transform the agly-
cones of their antibiotics into the mannosyl
Selective hydrolysis of certain sugar substi- derivatives (CHUNGet al., 1986a; BORGHIet
tuents that could not be performed chemical- al., 1991b). Transformation is almost quanti-
ly was carried out enzymatically. tative after 24 h of incubation.
Using naranginase chloropolysporin B and Mannosylation is independent of antibiotic
a-or pavoparcin were converted to the cor- biosynthesis, since with K. aridum it occurs
responding derhamnosyl derivatives with a also in the presence of glyphosate, a biosyn-
40% and 80% yield, respectively. Although thesis inhibitor, and with A. teichomyceticus it
the commercial enzyme used contained pglu- is performed by washed mycelium resus-
cosidase the glucose moiety was not hydro- pended in saline. The reaction is not very spe-
lyzed, probably because of steric hindrance. cific with respect to the substrate: pseudoagly-
a-Mannosidase converted chloropolysporin B cones are also efficiently transformed by both
and C to the corresponding demannosyl pseu- microorganisms.
doaglycones with a 71-50% yield. With this Protoplast preparations of K. aridum were
enzyme it was also possible to obtain the de- tested for their ability to convert ardacin agly-
mannosyl derivative of p-avoparcin which cone into the complete antibiotic. In this case
was found identical with e-avoparcin (TA- the prevalent transformation compound was
KATSU et al., 1987b). also mannosyl aglycone, accompanied by
Enzymatic deglycosylation clarified the re- smaller amounts of poorly defined derivatives
lationship between galacardin A and p-avo- carrying other neutral sugars (CHUNGet al.,
parcin. Treatment of galacardin A with a-gal- 1986b).
actosidase selectively removed the galactose
units linked by a-glycosidic bonds. After in-
cubation either the galactose at the hydroxyl 10.3 Deacylation
of amino acid 3 or the galactose linked to the
L-rhamnose unit at amino acid 1 were re- At present, microbial deacylation seems to
moved leaving the sugar units linked by p- be the only method to remove the fatty acid
glycosidic bonds unaffected (TAKEUCHIet residues from the N-acylamino sugars of lipo-
al., 1992). glycopeptides. The main component of
The acyl glucosamine moiety of teicoplanin A40926 complex is characterized by the pres-
is selectively hydrolyzed by mild acid treat- ence of a 10-methylundecanoyl chain that
ment. More acidic conditions are required to acylates the amino group of the glucuronic
remove the other sugar substituents of the moiety. A few hundred microorganisms were
molecule. Therefore, a selective, chemical hy- tested for their ability to selectively hydrolyze
drolyzation of the mannose moiety was not the amide bond. Positive results were ob-
possible without removing the acyl glucosam- tained with cultures of A. teichomyceticus
ine unit. The single components of the teico- ATCC 31121 (the producer of teicoplanin),
planin complex were converted into the cor- A. missouriensis NRRL 15646 and NRRL
responding demannosyl derivatives by treat- 15647 (SELVAet al., 1988).
ment with cultures of N. orientalis NRRL Ardacins and the components of AAD 609
2450 (the producer of vancomycin-type M43 complex are also deacylated by A. teichomy-
complex) or S. cundidus NRRL 3218 (the ceticus cultures (CHUNGet al., 1986b). Teico-
producer of actinoidin-type avoparcin). The planin is not deacylated by its producer -
time course of the transformation with which is somewhat surprising in view of the
washed mycelium of N. orientalis showed a structure similarity with kibdelins.
40% conversion in 72 h at a concentration of
0.25 g+L-' (BORGHIet al., 1991b).
10.4 Other Biotransformations 11 References

Actaplanin single components or its pseu- ANG,S-G., WILLIAMSON, M. P., WILLIAMS, D. H.
doaglycone are bioconverted by pure cultures (1988), Structure elucidation of a glycopeptide
of A. missouriensis NRRL 15646 or A. mis- antibiotic, OA-7653,J. Chem. SOC. Perkin Trans.
souriensis NRRL 15647 into the correspond- I, 1949-1956.
ing CUC/CSV glycopeptides in which the ARDUINO, R. C., MURRAY, B. E. (1993), Vanco-
-CO-CH(NH2)- group on the terminal ami- mycin resistance in Gram-positive organisms,
no acid 1 is oxidatively deaminated into a Curr. Op. Infect. Dis. 6, 715-724.
-CO-CO- group (CLEM et al., 1985). Anti- ARJUNARAO, V., RAVISHANKAR, D., SADHUK-
HAN, A. K., AHMED, S. M., GOEL,A. K., PRAB-
biotic CUCKSV is also produced by cofer- HU, N. S., VERMA,A. K., VENKATESWARLU,
mentation of A. missouriensis NRRL 15646 A., ALLAUDEEN, H. S., HEDDE,R. D., NISBET,
and NRRL 15647. They are blocked mutants L. J. (1986), Synmonicins: a novel antibiotic
of A. missouriensis ATCC 31683 (producer of complex produced by Synnemomyces mamno-
actaplanin components B3, El, and pseudo- orii gen. et sp. nov. I. Taxonomy of the produc-
aglycone) which in turn derives from muta- ing organism, fermentation and biological prop-
tion of Actinoplunes ATCC 23342 (actaplanin erties, Abstract No. 939, 26rd Interscience Con-
producer). The two strains NRRL 15646 and ference on Antimicrobial Agents and Chemother-
NRRL 15647 do not produce actaplanin aphy, New Orleans: Am. SOC.Microbiol.
(HERSHBERGER, 1985). ARTHUR, M., COURVALIN, P. (1993), Genetics and
mechanisms of glycopeptide resistance in ente-
Bromoactaplanins carrying a Br atom in- rococci, Antimicrob. Agents Chemother. 37,
stead of the C1 on the phenyl residue of ami- 1563-1571.
no acid 6 were obtained by supplying a spon- ATHALYE,M., COATES,N. J., MILNER,P. H.
taneous mutant of the producing strain with (1989), Eur. Patent Appl. 339982.
NaBr (HUBERet al., 1988). As discussed in BARNA,J. C. J., WILLIAMS, D. H. (1984), The
Sect. 8.2 the acyl chains of the teicoplanin fac- structure and mode of action of glycopeptide an-
tors originate from degradation of long-chain tibiotics of the vancomycin group, Ann. Rev. Mi-
fatty acids. This notion was exploited to bio- crobiol. 38,339-357.
synthesize new antibiotic derivatives. Addi- BARNA,J. C. J., WILLIAMS, D. H., STONE,D. J. M.,
LEUNG,T.-W. C., DODDRELL, D. M. (1984),
tion of ricinoleic acid (12-hydroxyoleic acid) Structure elucidation of the teicoplanin antibiot-
to cultures of A. teichomyceticus resulted in ics, J. Am. Chem. SOC. 106,48954902.
the formation of a novel teicoplanin compo- BERDNIKOVA, T. F., LOMAKINA, N. N., POTA-
nent characterized by a 4-hydroxydecanoyl POVA, N. P. (1982), Structure of actinoidins A
chain. Similarly, addition of linolenic acid and B, Antibiotiki 27,252-258.
(which has an 18 carbon linear chain with BOECK, L. D., MERTZ,F. P. (1986), A47934, a nov-
three double bonds) gives raise to a teico- el glycopeptide-aglycone antibiotic produced by
planin derivative with a 4,7-decadienoyl acyl a strain of Streptomyces toyocaensis. Taxonomy
chain (unpublished data from the authors’ la- and fermentation studies, J. Antibiot. 39, 1533-
1540.
boratory). BOECK,L. D., MERTZ,F. P., WOLTER, R. K., HIG-
Selective removal of amino acid 1 of vanco- GENS, C. E. (1984), N-demethylvancomycin, a
mycin (to obtain “vancomycin hexapeptide”) novel antibiotic produced by a strain of Nocar-
was carried out by biotransformation, as an dia orientalis. Taxonomy and fermentation, J.
alternative to the chemical method. Screening Antibiot. 37,446-453.
of soil microorganisms for their ability to BOECK,L. D., MERTZ,F. P., CLEM,G. M. (1985),
grow on a model compound (D-leucyl-D-tyro- A41030, a complex of novel glycopeptide anti-
sine) as the sole carbon source led to the iso- biotics produced by a strain of Streptomyces vir-
lation of a strain (Actinomuduru citreu) that giniae. Taxonomy and fermentation studies, J.
Antibiot. 38, 1-8.
converted vancomycin B into the hexapeptide BORGHI, A., ANTONINI, P., ZANOL,M., FERRARI,
in 90% yields (ZMIJEWSKI et al., 1989). P., ZERILLI, L. F., LANCINI, G. C. (1989), Isola-
tion and structure determination of two new
analogs of teicoplanin, a glycopeptide antibiotic,
J. Antibiot. 42,361-366.
11 References 391
BORGHI,A., EDWARDS, D., ZERILLI,L. F., LAN- VENELLA, A. J., HEALD,s. L., KILLMER, L. B.,
CINI,G. C. (1991a), Factors affecting the normal MICO,B. A., MUELLER,L., PAN,C. H., POEH-
and branched-chain acyl moieties of teicoplanin LAND, B. L., RAKE,J. B., ROBERTS,G. D.,
components produced by Actinoplanes teicho- SHEARER, M. C., SITRIN,R. D., NISBET,L. J.,
myceticus, J. Gen. Microbiol. 137, 587-592. JEFFS,P. W. (1987), Parvodicin, a novel glyco-
BORGHI,A., FERRARI, P., GALLO,G. G., ZANOL, peptide from a new species, Actinomadura par-
M., ZERILLI, L. F., LANCINI, G. G. (1991b), Mi- vosata: discovery, taxonomy, activity and struc-
crobial demannosylation and mannosylation of ture elucidation, J. Antibiot. 40,970-990.
teicoplanin derivatives, J. Antibiot. 44, 1444- CHUNG,S. K., TAYLOR, P., OH, Y. K., DEBROSSE,
1451. C., JEFFS,P. W. (1986a), Biosynthetic studies on
Box, S. J., ELSON,A. L., GILPIN, M. L., WINSTAN- aridicin antibiotics. I. Labeling patterns and
LEY (1990), MM 47761 and MM 49721, glyco- overall pathways, J. Antibiot. 39,642-651.
peptide antibiotics produced by a new strain of CHUNG,S. K., OH, Y. K., TAYLOR,P., GERBER,
Amycolatopsis orientalis. Isolation, purification R., NISBET,L. J. (1986b), Biosynthetic studies
and structure determination, J. Antibiot. 43,931- on aridicin antibiotics. 11. Microbial transforma-
937. tions and glycosylations by protoplasts, J. Anti-
Box, S. J., COATES,N. J., DAVIS,C. J., GILPIN,M. biot. 39,652-659.
L., HOUGE-FRYDRYCH, C. V. S., MILNER,P. H. CLEM,G. M., BOECK,L. D., ANDERSON, M. T.,
(1991), MM-55266 and MM-55268, glycopeptide MICHEL,F. H. (1985), Eur. Patent Appl.
antibiotics produced by a new strain of Amyco- 142285.
latopsis: isolation, purification and structure de- COATES,N. J., ELSON,A. L., ATHALYE, M., CUR-
termination, J. Antibiot. 44,807-813. TIS,L. M., MOORES,L. V. (1990a), Eur. Patent
BRAZHNIKOVA, M. G., LOMAKINA, N. N., BERD- Appl. 375213.
NIKOVA, T. F. (1989), Eremomycin: a new glyco- COATES,N. J., SYKES,R., CHRISTOPHER, D. J.,
peptide antibiotic, in: Bioactive Metabolites from CURTIS, L. M. (1990b), Eur. Patent Appl.
Microorganisms (BUSHELL, M. E., GRAEFE,U., 375448.
Eds.), Vol. 27, pp. 163-165. Amsterdam: Else- COATES,N. J., DAVIS,C. J., CURTIS,L. M., SIKES,
vier. R. (1991), WO Patent 91/16346.
BROGDEN, R. N., PETERS,D. H. (1994), Teicoplan- CORONELLI, C., GALLO,G. G., CAVALLERI, B.
in. A reappraisal of its antimicrobial activity, (1987), Teicoplanin: chemical, physico-chemical
pharmacokinetic properties and therapeutic effi- and biological aspects, I1 Farmaco, Ed. Scient.
cacy, Drugs 47,823-854. 42,767-786.
CAMPOLI-RICHARDS, D. M., BROGDEN,R. N., CORTI,A., CASSANI, G. (1985), Synthesis and char-
FAULDS,D. (1990), Teicoplanin. A review of its acterization of D-alanyl-D-alanine-agarose: a
antibacterial activity, pharmacokinetic proper- new bioselective adsorbent for affinity chroma-
ties and therapeutic potential, Drugs 40, 449- tography of glycopeptide antibiotics, Appl. Bio-
486. chem. Biotechnol. 11,101-109.
CASSANI,G. (1989), Glycopeptides: antibiotic dis- CORTI,A., RURALI,L., BORGHI,A., CASSANI,G.
covery and mechanism of action, in: Bioactive (1985), Solid-phase enzyme-receptor assay
Metabolites from Microorganisms (BUSHELL, M. (SPERA): A competitive binding assay for gly-
E., GRAEFE,U., Eds.), pp. 221-235. Amsterdam: copeptide antibiotics of the vancomycin class,
Elsevier. Clin. Chem. 31, 1606-1610.
CASSETTA, A., BINGEN,E., LAMBERT-ZECHOVS-DEBONO,M., MOLLOY,R. M., BARNHART, M.,
KY, N. (1991), La vancomycine en 1991: actualid DORMAN, D. E. (1980), A35512, a complex of
et perspectives, Path. Biol. 39,700-708. new antibacterial antibiotics produced by Strep-
CAVALLERI, B., PARENTI,F. (1992), Glycopep- tomyces candidus. 11. Chemical studies on
tides (dalbaheptides), in: Kirk-Othmer Encyclo- A35512B, J. Antibiot. 33,1407-1416.
pedia of Chemical Technology, 4le, Vol. 2, pp. DEBONO,M., MERKEL,K. E., MOLLOY,R. M.,
995-1018. New York J. Wiley and Sons. BARNHART, M., PRESTI,E., HUNT,A. H., HAM-
CHAKRABORTY, T. K., VENKAT REDDY, G. ILL,R. L. (1984), Actaplanin, new glycopeptide
(1992), Studies directed toward the synthesis of antibiotics produced by Actinoplanes missou-
glycopeptide antibiotic teicoplanin: first synthe- riensis. The isolation and preliminary chemical
sis of the N-terminal 14-membered ring, J. Org. characterization of actaplanin, J. Antibiot. 37,
Chem. 57,5462-5469. 85-95.
CHRISTENSEN, S . B., ALLAUDEEN, H. S., BURKE, DINGERDISSEN, J. J., SITRIN,R. D., DEPHILLIPS,
M. R., CARR,S. A., CHUNG,S. K., DEPHILLIPS, P. A., GIOVENELLA,A. J., GRAPPEL,S. F.,
P., DINGERDISSEN, J. J., DIPAOLO,M., GIO- MEHTA,R. J., OH, Y. K., PAN,C. H., ROBERTS,
G. D., SHEARER, M. C., NISBET,L. J. (1987), On the biosynthesis of the antibiotic vancomy-
Actinoidin AZ,a novel glycopeptide: production, cin, J. Chem. SOC., Chem. Commun. 1982, 344-
preparative HPLC separation and characteriza- 346.
tion, J. Antibiot. 40, 165-172. HAMMOND, S. J., WILLIAMS, D. H., NIELSEN,R.
DOOLIN,L. E., GALE,R. M., GODFREY JR., 0. W., V. (1983), The biosynthesis of ristocetin, J.
HAMILL,R. L., MAHONEY, D. F., YAO,R. C-F. Chem. SOC., Chem. Commun. 1983,116-117.
(1989), Eur. Patent Appl. 299 707. HARRIS,C. M., KOPECKA,H., HARRIS,T. M.
EVANS,D. E., DEVRIES,K. M. (1994), Approaches (1983), Vancomycin: Structure and transforma-
to the synthesis of the vancomycin aglycones, in: tion to CDP-I, J. Am. Chem. SOC. 105, 6915-
Glycopeptide Antibiotics (NAGARAJAN,R., 6922.
Ed.), pp. 63-104. New York: Marcel Dekker. HARRIS,C. M., KANNAN, R., KOPECKA, H., HAR-
FELMINGHAM, D. (1993), Towards the ideal glyco- RIS,T. M. (1985), The role of chlorine substi-
peptide, J. Antimicrob. Chemother. 32,663-666. tuents in the antibiotic vancomycin: preparation
FOLENA-WASSERMAN, G., POEHLAND,B. L., and characterization of mono- and didechloro-
YEUNG,E. W-K., STAIGER, D., KILLMER, L. B., vancomycin, J. Am. Chem. SOC. 107,6652-6658.
SNADER,K., DINGERDISSEN, J. J., JEWS,P. W. HARRIS,C. M., FESIK,S. W., THOMAS,A. M.,
(1986), Kibdelins (AAD-609), novel glycopep- KANNAN, R., HARRIS,T. M. (1986), Iodination
tide antibiotics. 11. Isolation, purification and of vancomycin, ristocetin A, and ristocetin pseu-
structure, J. Antibiot. 39, 1395-1406. doaglycon, J. Org. Chem. 51,1509-1513.
FOLENA-WASSERMAN, G., SITRIN,R., CHAPIN,F., HEALD,S . L., MUELLER,L., JEFFS,P. W. (1987),
SNADER, K. (1987), Affinity chromatography of Actinoidins A and AP:Structure determination
glycopeptide antibiotics, J. Chromatogr. 392, using 2D NMR methods, J. Antibiot. 40, 630-
225-238. 645.
FRANCO, C. M. M., CHA-ITERJEE, S., VIJAKUMAR, HERSHBERGER, C. L. (1985), Eur. Patent Appl.
E. K. S., CHA-ITERJEE, D. K., GANGULI, B. N., 159436.
RUPP,R. H., FEHLHABER, H-W., KOGLER, H., HIGGINS, H. M., JR., MCCORMICK, M. H., MER-
SELBERT,G., TEETZ, V. (1990), Eur. Parent KEL,K. E., MICHEL,K. H. (1985), Eur. Patent
Appl. 356894. Appl. 159180.
GAUSE,G. F., BRAZHNIKOVA, M. G., LOMAKINA, HOLDEN,K. S., RUDDOCK, J. C., TONE,J., MAE-
N. N., BERDNIKOVA, T. F., FEDOROVA, G. B., DA, H. (1991), Brit. Patent 2243610.
TOKAREVA, N. L., BORISOVA, V. N., BA-ITA,G.
Y (1989), Eremomycin - New glycopeptide anti- HOLDOM,K. S., MAEDA,H., RUDDOCK,J. C.,
biotic: chemical properties and structure, J. Anti- TONE,J. (1988), Eur. Patent Appl. 265 143.
biot. 42, 1790-1799. HUBER,F. M., PIEPER,R. L., TIETZ,A. J. (1987),
GOLDSTEIN,B. P., SELVA,E., GASTALDO,L., Characterization of the process of the biosynthe-
BERTI,M., PALLANZA, R., RIPAMONTI, F., FER- sis of the actaplanin complex by Actinoplanes
RARI, P., DENARO, M., ARIOLI,V., CASSANI, G. missouriensis, J. Ferment. Technol. 65,85-89.
(1987), A40926, a new glycopeptide antibiotic HUBER,F. M., MICHEL,K. L., HUNT,A. H., MAR-
with anti-Neisseria activity, Antimicrob. Agents TIN,J. W., MOLLOY,R. M. (1988), Preparation
Chemother. 31,1961-1966. and characterization of some bromine analogs of
GOLDSTEIN,B. P., ROSINA, R., PARENTI,F. the glycopeptide antibiotic actaplanin, J. Anti-
(1994), Teicoplanin, in: Glycopeptide Antibiotics biot. 41,798-801.
(NAGARAJAN, R., Ed.), pp. 273-307. New York: HUNT, A. H., ELZEY,T. K., MERKEL,K. E., DE-
Marcel Dekker. BONO, M. (1984a), Structures of actaplanins, J.
GOOD,V. M., GWYNN, M. N., KNOWLES, D. J. C. Org. Chem. 49,641-645.
(1990), MM 45289, a potent glycopeptide anti- HUNT, A. H., MARCONI,G. G., ELZEY,T. K.,
biotic which interacts weakly with diacetyl-L-ly- HOEHN,M. M. (1984b), A51568A: N-demethyl-
syl-D-alanyl-D-alanine,J. Antibiot. 43,550-555. vancomycin, J. Antibiot. 37, 917-919.
HAMILL,R. L., YAO, R. C. (1993), U.S. Patent HUNT,A. H., DORMAN, D. E., DEBONO,M., MOL-
5187082. LOY, R. M. (1985), Structure of antibiotic
HAMILL,R. L., HANEY,M. E., JR., STARK,W. M. A41030A, J. Org. Chem. 50,2031-2035.
(1973), Ger. Patent Appl. 2252937. JOHNSON, A. P., U-ITLEY,A. H. C., WOODFORD,
HAMILL,R. L., MAHONEY, D. F., NAKASUKASA, N., GEORGE,R. C. (1990), Resistance to vanco-
W. M., YAO, R. C-F. (1988), Eur. Patent Appl. mycin and teicoplanin: an emerging clinical
265 701. problem, Clin. Microbiol. Rev. 3,280-291.
HAMMOND, S. J., WILLIAMSON, M. P., WILLIAMS, KAMOGASHIRA, T., NISHIDA,T., SUGAWARA, M.
D. H., BOECK,L. D., MARCONI, G. G. (1982), (1983), A new glycopeptide antibiotic, OA-7653,
I1 References 393
produced by Streptomyces hygroscopicus ssp. hi- MALABARBA, A., SPREAFICO,F., FERRARI,P.,
wasaensis, Agric. Biol. Chem. 47, 499-506. KETTENRING, J., STRAZZOLINI, P., TARZIA,G.,
KANNAN, R., HARRIS,C. M., HARRIS,T. M., PALLANZA,R., BERTI, M., CAVALLERI, B.
WALTHO,J. P., SKELTON, N. J., WILLIAMS, D. (1989b), Dechloro teicoplanin antibiotics, J. An-
H. (1988), Function of the amino sugar and N- tibiot. 42, 1684-1697.
terminal amino acid of the antibiotic vancomycin MALABARBA, A., TRANI,A., STRAZZOLINI, P.,
in its complexation with cell wall peptides, J. CIETTO,G., FERRARI, P., TARZIA,G., PALLAN-
Am. Chem. SOC.110,2946-2953. ZA,R., BERTI,M. (1989c), Synthesis and biolog-
KATRUKHA,G. S., SILAEV,A. B. (1986), The ical properties of N63-carboxamidesof teicoplan-
chemistry of glycopeptide antibiotics of the van- in antibiotics. Structure-activity relationships, J.
comycin group, in: Chemistry of Peptides and Med. Chem. 32,2450-2460.
Proteins (VOELTER,W., BAKER,E., OVCHINNI- MALABARBA, A., CIABATTI, R., KETTENRING, J.,
KOV,Y. A., IVANOV, V. T., Eds.), Vol. 3., pp. SCOTTI, R., CANDIANI,G., PALLANZA,R.,
289-306. Berlin: de Gruyter. BERTI,M., GOLDSTEIN, B. P. (1992), Synthesis
KIM,K-H., MARTIN,Y., OTIS,E., MAO,J. (1989), and antibacterial activity of a series of basic am-
Inhibition of '"I-labeled ristocetin binding to ides of teicoplanin and deglucoteicoplanin with
Micrococcus luteus cells by the peptides related polyamines, J. Med. Chem. 35,4054-4060.
to bacterial cell wall mucopeptide precursors: MALABARBA, A., CIABATTI, R., CAVALLERI, B.
quantitative structure-activity relationships, J. (1993a), Semisynthetic dalbaheptides: chemistry
Med. Chem. 32,84-93. and biological activity, Curr. Topics Med. Chem.
KOBRIN,M. B., FEDOROVA, J. B., KATRUKHA, G. 1,359-376.
S. (1988), Study on deglycosylation of certain an- MALABARBA,A., CIABATTI,R., PANZONE,G.,
tibiotics belonging to vancomycin group, Anti- MARAZZI, A. (1993b), WO Patent 93103060.
biotiki i Chemother. 33,331-335. MATSUMOTO, K. (1961), A vancomycin-related an-
KUNSTMANN,M. P., PORTER,J. N. (1974), US.Pa- tibiotic from Streptomyces sp. K-288. Studies on
tent 3 803 306. Streptomyces antibiotics, XXXVIII, J. Antibiot.,
LANCINI, G. C. (1989), Fermentation and biosyn- Ser. A 14,141-146.
thesis of glycopeptide antibiotics, in: Bioactive MCGAHREN,W. J., MARTIN,J. H., MORTON,G.
Metabolites from Microorganisms (BUSHELL, M. O., HARGREAVES, R. T., LEESE,R. A., Lov-
E., GRAEFE,U., Eds.), pp. 283-296. Amsterdam: ELL, F. M., ELLESTAD,G. A., O'BRIEN, E.,
Elsevier. HOLKER,J. S. E. (1980), Structure of avoparcin
LANCINI, G. C., CAVALLERI, B. (1990), Glycopep- components, J. Am. Chem. SOC. 102,1671-1674.
tide antibiotics of the vancomycin group, in: Bio- MCGAHREN, W. J., LEESE,R. A., BARBATSCHI, F.,
chemistry of Peptide Antibiotics, Recent Ad- MORTON,G. O., KUCK,N. A., ELLESTAD, G. A.
vances in the Biotechnology of PLactams and (1983), Components and degradation com-
Microbial Bioactive Peptides (KLEINKAUF, H., pounds of the avoparcin complex, J. Antibiot. 36,
VON DOHREN,H., Eds.), pp. 159-178. Berlin: de 1671-1682.
Gruyter. MERTZ,F. P., DOOLIN,L. E. (1973), The effect of
LECHEVALIER, M. P., PRAUSER, H., LABEDA,D. . inorganic phosphate on the biosynthesis of van-
P., RUAN,J.-S. (1986), Two new genera of no- comycin, Can. J. Microbiol. 19,263-270.
cardioform actinomycetes: Amycolata gen. nov. MICHEL,K. H., SHAH, R. M., HAMILL,R. L.
and Amycolatopsis gen. nov., Int. J. Syst. Bacteri- (1980), A35512, a complex of new antibacterial
01. 36,29-37. antibiotics produced by Streptomyces candidus.
MALABARBA, A., CIABATTI, R. (1992), Int. Patent I. Isolation and characterization, J. Antibiot. 33,
WO 92/10517. 1397-1406.
MALABARBA, A., STRAZZOLINI, P., DEPAOLI,A., MICHEL,K. J., YAO, R. C-F. (1991), Eur. Patent
LANDI,M., BERTI,M., CAVALLERI, B. (1984), Appl. 424051.
Teicoplanin, antibiotics from Actinoplanes tei- NADKARNI, S. R., PATEL,M. V., CHATTERJEE, S.,
chomyceticus nov. sp. VI. Chemical degradation: VIJAYAKUMAR, E. K. S., DESIKAN,K. R.,
physico-chemical and biological properties of BLUMBACH, J., GANGULI, B. N., LIMBERT,M.
acid hydrolysis products, J. Antibiot. 37, 988- (1994), Balhimycin, a new glycopeptide antibiot-
999. ic produced by Amycolatopsis sp. Y-86-21022.
MALABARBA, A., FERRARI, P., CIETTO,G., PAL- Taxonomy, production, isolation and biological
LANZA, R., BERTI, M. (1989a), Synthesis and activity, J. Antibiot. 47, 334-341.
biological activity of N63-carboxypeptidesof tei- NAGARAJAN,R. (1993), Structure-activity rela-
coplanin and teicoplanin aglycone, J. Antibiot. tionships of vancomycin-type glycopeptide anti-
42,1800-1816. biotics, J. Antibiot. 46, 1181-1195.
NAGARAJAN,R. (Ed.) (1994a), Glycopeptide Anti- PERKINS,H. R. (1982), Vancomycin and related
biotics. New York Marcel Dekker. antibiotics, Pharmacol. Ther. 16, 181-197.
NAGARAJAN, R. (1994b), Structure-activity rela- PERKINS, H. R.,NIETO,M. (1970), The preparation
tionships of vancomycin antibiotics, in: Glyco- of iodinated vancomycin and its distribution in
peptide Antibiotics (NAGARAJAN, R., Ed.), pp. bacteria treated with the antibiotic, Biochem. J.
195-218. New York: Marcel Dekker. 116,83-92.
NAGARAJAN, R., SCHABEL, A. A. (1988), Selective P ~ I L I PJ., E., SCHENCK,J. R., HARGIE,M. P.,
cleavage of vancosamine, glucose, and N-methyl HOLPER,J. C., GRUNDY, W. E. (1960), The in-
leucine from vancomycin and related antibiotics, creased activity of ristocetins A and B following
J. Chem. SOC., Chem. Comrnun. 1988, 1306- acid hydrolysis, Antimicrob. Agents Ann. 1960,
1307. 10-16.
NAGARAJAN, R., SCHABEL,A. A., OCCOLOWITZ, PHILLIPS,G., GOLLEDGE,C. L. (1992), Vancomy-
J. L., COUNTER, F. T., O n , J. L. (1988), Synthe- cin and teicoplanin. Something old, something
sis and antibacterial activity of N-acyl vancomy- new, Med. J. Aust. 156,53-57.
cins, J. Antibiot. 41, 1430-1438. PITKIN,D. H., MICO,B. A., SITRIN, R. D., NISBET,
NAGARAJAN, R., BERRY,D. M., HUNT,A. H., Oc- L. J. (1986), Charge and lipophilicity govern the
COLOWITZ, J. L., SCHABEL, A. A. (1989a), Con- pharmacokinetics of glycopeptide antibiotics,
version of antibiotic A82846B to orienticin A Antimicrob. Agents Chemother. 29,440-444.
and structural relationships of related antibiot- RAKE,J. B., GERBER,R., METHA,R. J., NEWMAN,
ics, J. Org. Chem. 54,983-986. D. K., OH, Y. K., PHELEN,C., SHEARER, M. C.,
NAGARAJAN, R., BERRY,D. M., SCHABEL, A. A. SITRIN,R. D., NISBET,L. J. (1986), Glycopep-
(1989b), The structural relationships of tide antibiotics: a mechanism based screen em-
A82846B and its hydrolysis products with chlo- ploying a bacterial cell wall receptor mimetic, J.
roorienticins A, B and C, J. Antibiot. 42, 1438- Antibiot. 39, 58-67.
1440. RAMARAO, A. V., CHAKRABORTY, T. K., JOSHI,
NICAS, T. I., ALLEN,N. E. (1994), Resistance and S. P. (1992), The first synthesis of C-terminal bi-
mode of action, in: Glycopeptide Antibiotics phenyl moiety of vancomycin, Tetrahedron Lett.
(NAGARAJAN, R., Ed.), pp. 219-241. New York 33,4045-4048.
Marcel Dekker. REYNOLDS, P. E. (1989), Structure, biochemistry
NIETO,M., PERKINS,H. R. (1971a), Modification and mechanism of action of glycopeptide anti-
of the acyl-D-alanyl-D-alanine terminus affecting biotics, Eur. J. Clin. Microbiol. Infect. Dis. 8,
complex-formation with vancomycin, Biochem. 943-950.
J. 1239789-803. RIVA,E., GASTALDO,L., BEREITA, M. G., FER-
NIETO,M., PERKINS, H. R. (1971b), The specificity RARI,P., ZERILLI,L. F., CASSANI,G., GOLD-
of combination between ristocetins and peptides STEIN,B. P., BERTI,M., PARENTI, F., DENARO,
related to bacterial cell wall mucopeptide pre- M. (1989), A42867, a novel glycopeptide anti-
cursors, Biochem. J. l24,845-852. biotic, J. Antibiot. 42, 497-505.
OKAZAKI, T., ENOKITA,R., MIYAOKA,H., TA- SELVA,E., GASTALDO, L., BEREITA,G., BORGHI,
KATSU,T., TORIKATA,A. (1987), Chloropoly- A., GOLDSTEIN, B. P., ARIOLI,V., CASSANI, G.,
sporins A, B and C, novel glycopeptide antibiot- PARENTI,F. (1986), Eur. Patent Appl. 177882.
ics from Faenia interjecta sp. nov. I. Taxonomy SELVA,E., BERETTA,G., BORGHI,A., DENARO,
of producing organism, J. Antibiot. 40,917-923. M. (1988), WO Patent Appl. 88102755.
PALLANZA, R., BERTI,M., GOLDSTEIN, B. P., MA- SELVA,E., GASTALDO, L., CASSANI, G., PARENTI,
PELLI, E., RANDISI, E., SCOITI, R., ARIOLI,V. F. (1992), Eur. Patent Appl. 259781.
(1983), Teichomycin: in vitro and in vivo evalua- SHEARER, M. C., ACTOR, P., BOWIE,B. A., GRAP-
tion in comparison with other antibiotics, J. An- PEL, S. F., NASH,C. H., NEWMAN, D. J., OH, Y.
timicrob. Chemother. 11, 419-425. K., PAN,C. H., NISBET,L. J. (1985). Aridicins,
PARENTI,F., CAVALLERI, B. (1989), Proposal to novel glycopeptide antibiotics. I. Taxonomy,
name the vancomycin-ristocetin like glycopep- production and biological activity, J. Antibiot.
tides as dalbaheptides, J. Antibiot. 42, 1882- 38,555-560.
1883. SHEARER, M. C., GIOVENELLA, A. J., GRAPPEL, S.
PARENTI, F., BERETTA,G., BERTI,M., ARIOLI,V. F., HEDDE, R. D., MEHTA,R. J., OH, Y. K.,
(1978), Teichomycins, new antibiotics from Acti- PAN,C. H., PITKIN,D. H., NISBET,L. J. (1986),
noplanes teichomyceticus nov. sp. I. Description Kibdelins, novel glycopeptide antibiotics. I. Dis-
of the producer strain, fermentation and biologi- covery, production, and biological evaluation, J.
cal properties, J. Antibiot. 31, 276-283. Antibiot. 39, 1386-1394.
11 References 395
SHELDRICK, G. M.,JONES,P. G., KENNARD, O., NOSHITA,T. (1987a), Chloropolysporins A, B

WILLIAMS, D. H., SMITH,G. H. (1978), Struc- and C, novel glycopeptide antibiotics from Fae-
ture of vancomycin and its complex with acetyl- nia interjecta sp. nov. 111. Structure elucidation
D-alanyl-D-alanine, Nature (London) 271, 223- of chloropolysporins,J. Antibiot. 40, 933-940.
225. TAKATSU,T., TAKAHASHI, S., TAKAMATSU, Y.,
SITRIN,R. D., FOLENA-WASSERMAN, G. (1989), SHIOIRI,T., IWADO,S., HANEISHI, T. (1987b),
Affinity and HPLC purification of glycopeptide Chloropolysporins A, B and C, novel glycopep-
antibiotics, in: Natural Products Isolation (WAG- tide antibiotics from Faenia interjecta sp. nov.
MAN,G. H., COOPER,R., Eds.), pp. 111-152. IV. Partially deglycosylated derivatives, J. Anti-
Amsterdam: Elsevier. biot. 40,941-945.
SITRIN,R. D., FOLENA-WASSERMAN, G. (1994), TAKEUCHI, M.,ENOKITA,R., OKAZAKI, T., KA-
Separation methodology, in: Glycopeptide Anti- GASAKI, T., INUKAI, M. (1991a), Helvecardins A
biotics (NAGARAJAN, R., Ed.), pp. 29-61. New and B, novel glycopeptide antibiotics. I. Taxono-
York: Marcel Dekker. my, fermentation, isolation and physico-chemi-
SITRIN,R. D., CHAN,G. W., DINGERDISSEN, J. J., cal properties, J. Antibiot. 44,263-270.
HOLL,W., HOOVER,J. R. E., VALENTA,J. R., TAKEUCHI,M., TAKAHASHI,S., INUKAI,M.
WEBB, L., SNADER,K. M. (1985), Aridicins, (1991b), Helvecardins A and B, novel glycopep-
novel glycopeptide antibiotics. 11. Isolation and tide antibiotics. 11. Structural elucidation, J. An-
characterization, J. Antibiot. 38,561-571. tibiot. 44, 271-277.
SITRIN, R. D., CHAN,G. W., CHAPIN, F., GIOVEN-
ELLA, A. J., GRAPPEL, S. F., JEFFS,P. W., PHIL- TAKEUCHI, M., TAKAHASHI, S., ENOKITA, R., SA-
LIPS,L., SNADER, K. M.,NISBET,L. J. (1986), KAIDA, Y., HARUYAMA, H., NAKAMURA, T.,
Aridicins, novel glycopeptide antibiotics. 111. KATAYAMA, T., INUKAI, M.(1992), Galacardins
Preparation, characterization, and biological ac- A and B, new glycopeptide antibiotics, J. Anti-
tivity of aglycone derivatives, J. Antibiot. 39,68- biot. 45, 297-305.
75. TAMURA, A., TAKEDA,I. (1975), Antibiotic AB-
SKELTON, N. J., WILLIAMS, D. H. (1990), Structure 65, a new antibiotic from Saccharornonospora
elucidation of the novel glycopeptide antibiotic viride, J. Antibiot. 28,395-397.
UK-68597, J. Org. Chem. 55,3718-3723. TOROPOVA, E. G., EGOROV,N. S., OBRAZTSOVA,
SKELTON, N. J., WILLIAMS, D. H., RANCE,M.J., A. Y. (1974), Effect of ristomycin and excess of
RUDDOCK, J. C. (1990), Structure elucidation of phosphorus in the medium on activity of en-
UK-72051, a novel member of the vancomycin zymes of Proactinomyces fructiferi var. ristomy-
group of antibiotics, J. Chem. SOC. Perkin Trans. cini, Antibiotiki 19, 418-422.
1,77-81. TOROPOVA, E. G., EGOROV,N. S., TKHAKER, V.,
SKELTON, N. J., WILLIAMS, D. H., RANCE,M.J., BURAKAEVA, A. D. (1982), Effect of various
RUDDOCK, J. C. (1991), Structure elucidation of sources of carbon and nitrogen on biosynthesis
a novel antibiotic of the vancomycin group. The of ristomycin, protease and pigment by Nocardia
influence of ion-dipole interactions on peptide fructiferi var. ristomycini, Antibiotiki 27, 749-
backbone conformation, J. Am. Chem. SOC. 113, 753.
3757-3765. TSUJI,N., KAMIGAUCHI, T., KOBAYASHI, M.,TE-
SOMMA,S., GASTALDO, L., CORTI,A. (1984), Tei- RUI,Y. (1988a), New glycopeptide antibiotics.
coplanin, a new antibiotic from Actinoplanes tei- 11. The isolation and structures of chloroorienti-
chomyceticus nov. sp., Antimicrob. Agents Che- cins, J. Antibiot. 41, 1506-1510.
mother. 26,917-923. TSUJI,N., KOBAYASHI, M.,KAMIGAUCHI, T., Yo-
SPIRI-NAKAGAWA, P., FUKUSHI, Y., MAEBASHI, SHIMURA, Y., TERUI,Y. (1988b), New glycopep-
K., IMAMURA, N., TAKAHASHI, Y., TANAKA, tide antibiotics. I. The structures of orienticins,J.
Y., TANAKA, H., OMURA,S. (1986), Izupeptins Antibiot. 41, 819-822.
A and B, new glycopeptide antibiotics produced
by an actinomycete, J. Antibiot. 39, 1719-1723. WALLAS,C. H., STROMINGER, J. L. (1963), Risto-
cetins, inhibitors of cell wall synthesis in Staphy-
SZTARICSKAI, F., BOGNAR,R. (1984), The chemis-
lococcus aureus, J. Biol. Chem. 238,2264-2266.
try of the vancomycin group of antibiotics, in:
Recent Developments in the Chemistry of Natural WILLIAMSON, M. P., WILLIAMS,D. H., HAM-
Carbon Compounds (BOGNAR,R., SZANTAY, MOND,S. J. (1984), Interactions of vancomycin
Cs.,Eds.), Vol. X, pp. 91-201. Budapest: Aka- and ristocetin with peptides as a model for pro-
demiai Kiad6. tein binding, Tetrahedron 40, 569-577.
TAKATSU, T., TAKAHASHI, S., NAKAJIMA, M.,HA- YAO, R. C., MAHONEY,D. F., BAISDEN,D. K.,
NEISHI,T., NAKAMURA, T., KUWANO,H., KI- MERTZ,F. P., MABE,J. A., NAKATSUKASA, W.
M. (1988), A82846, a new glycopeptide antibiot- moieties of the antibiotic complex A40926 and
ic complex, produced by Arnycolatopsis oriental- their relation with the membrane lipids of the
is. I. Discovery, taxonomy and fermentation producer strain, Rapid Cornrnun. Mass. Spec-
studies, Abstract No. 974,28th Interscience Con- trorn. 6, 109-114.
ference on Antimicrobial Agents and Chemother- ZMIJEWSKI, M. J., BRIGGS,B., LOGAN,R., BOECK,
apy, Los Angeles: Am. SOC.Microbiol. L. D. (1987), Biosynthetic studies on antibiotic
ZECKEL,M. L., WOODWORTH, R. (1994), Vanco- A47934, Antirnicrob. Agents Chernother. 31,
mycin, a clinical overview, in: Glycopeptide Anti- 1497-1501.
biotics (NAGARAJAN, R., Ed.), pp. 309409. ZMIJEWSKI, M. J., LOGAN,R. M., MARCONI,G.,
New York: Marcel Dekker. DEBONO,M., MOLLOY,R. M., CHADWELL, F.,
ZERILLI,L. F., EDWARDS,D. M., BORGHI,A., BRIGGS,B. (1989), Biotransformation of vanco-
GALLO,G. G., SELVA,E., DENARO,M., LAN- mycin B to vancomycin hexapeptide by a soil mi-
CINI, G. C. (1992), Determination of the acyl croorganism, J. Nut. Prod. 52,203-206.
Biotechnology
10 Aminoglycosides and Sugar

Components in Other Secondary
Metabolites
WOLFGANGPIEPERSBERG,
JURGENDISTLER
Wuppertal, Germany
1 Introduction 399
2 Isolation, Distribution, Ecology, and Fermentation 399
3 Biogenesis and Basic Pathways of Cyclitols and Sugar Components 402
3.1 Cyclitols 403
3.2 Sugar Components 405
3.2.1 6-Deoxy- and Other Deoxyhexoses 407
3.2.2 Other Sugar Components 408
4 Genetics and Biochemistry of the Biosynthesis and Functions of Aminoglycosides 409
4.1 Streptomycins and Related Ca Aminoglycosides 415
4.1.1 Streptomycins 416
4.1.1.1 Biosynthesis of Streptidine and Bluensidine 417
4.1.1.2 The L-Dihydrostreptose Pathway 423
4.1.1.3 Hexosamine Pathway 423
4.1.1.4 Condensation of Subunits, Processing, and Export 425
4.1.2 Streptomycin-Related Ca Aminoglycosides 427
4.2 Fortimicins, Istamycins 429
4.3 2-Deoxystreptamine-ContainingAminoglycosides 432
4.4 Other Aminoglycosides 436
4.5 Resistance in Aminoglycoside Producers 440
4.6 Regulation in Streptomycetes 443
4.7 Overview of the Aminoglycoside Pathways 448
4.8 Aminoglycoside-Target Site Interactions and General Effects on Bacterial and
Eukaryotic Cells 449
398 10 Aminoglycosides and Sugar Components in Other Secondary Metabolites
5 Related Sugar Components in Other Secondary Metabolites 451

5.1 Lincosamides 451
5.2 Cyclitols 454
5.3 6-Deoxyhexoses 455
5.4 Other Pentose, Hexose, and Heptose Derivatives 456
6 Evolutionary Aspects 457
7 Involvement of Primary Metabolism in the Delivery of Carbohydrate Precursors 459
8 Pathway Engineering and Other Types of Application in Biotechnology 459
9 Conclusions and Perspectives 462
10 Appendix: Chemical Structures 462
11 References 476
1 Introduction chapter focus mainly on new aspects of the

molecular genetics, biochemistry, and physi-
ology of production of aminoglycosides and
There is no clear chemical or biochemical the similarly modified sugar components of
definition of the term “aminoglycoside” other chemical classes of secondary metabol-
which is traditionally reserved for mono- to ites. In addition, future biotechnological ap-
oligosaccharidic sugar and/or cyclitol deriva- plications of this knowledge are discussed.
tives containing amino nitrogen. Therefore, Many details of the chemistry, mode of ac-
an attempt is made in this chapter to demon- tion, classical strain improvement, fermenta-
strate that in biological and especially genetic tion technology, and pharmacology, e.g., of 2-
terms there are many parallels between ori- deoxystreptamine-containing, streptomycin-
gins and uses of strongly derived sugar com- like and other related aminocyclitol-amino-
ponents in secondary metabolites in general glycosides are well known from many reviews
and with respect to the aminocyclitol amino- published in the past two decades (DAVIES
glycosides in particular. A list of the major and YAGISAWA,1983; GRAFE, 1992; Lo-
chemicalgroups of aminoglycosides (see Tab. 1) RIAN, 1980; MALLAMS, 1988; UMEZAWAand
shows that some new structures have been HOOVER 1982; UMEZAWAet al., 1986; WAL-
described in the past decade. LACE et al., 1979) and will not be extensively
Since the 1st Edition of “Biotechnology ” dealt with here. For the progress made in re-
(cf. UMEZAWAet al., 1986) the field of ami- lated fields, such as the chemical and enzymic
noglycoside research has undergone a signifi- synthesis of carbohydrate components and
cant change characterized by the following their analogs, e.g., cyclitols, hexose and pen-
divergent aspects: tose derivatives, sugar mimics, and their bio-
logical activity and application, the reader is
(1) ceasing interest in the search for and de- referred to the following publications: KEN-
velopment of new antibiotics from this NEDY,1988; HORTON et al., 1989; THIEM,
particular group of compounds accompa- 1990 LUKACSand OHNO,1990 YAMAGUCHI
nied by stagnant market figures and clini- and KAKINUMA, 1992a; OGURAet al., 1992;
cal use; BILLINGTON, 1993; HUDLICKYand CEBU-
(2) increasing interest in basic research on LAK, 1993; COLEMANand FRASER,1993;
the (molecular) biological fundamentals LOOKet al., 1993; KROHNet al., 1993; TESTA
of the production of aminoglycosides and et al., 1993; COOKet a1 1994.
other secondary metabolic carbohydrates,
especially initiated by the new methods of
genetic engineering in actinomycetes, ba-
cilli, and other groups of biotechnological
interest; 2 Isolation, D stribution,
(3) evaluation of new classes of related natu-
ral products for other types of applica- Ecology, and
tions, e.g., inhibitors of glycosidases and
other enzymes. Fermentation
However, this situation might change again In the last decade new metabolites which
due to the currently increasing problems with clearly belong to the aminoglycoside group
antibiotic resistance and the appearance of were detected and described. Tab. 1 gives an
new and reappearence of “old” infectious de- overview of the chronology of publications
seases (see discussion in DAVIES,1992; HOT- and the biochemical grouping of the most im-
TA et al., 1995). Examples are methicillin-re- portant aminoglycosides. Most sugar-contain-
sistant staphylococci, bacterial infections in ing secondary metabolites relevant to this
HIV patients, and the worldwide renaissance chapter are formed by bacteria, mainly acti-
of tuberculosis as a major health concern. In nomycetes, bacilli, and pseudomonads
line with this development the authors of this (Fig. 1). However, the bioactive glycosides
Tab. 1. Chronology of the Detection and Biochemical Classification of the Major Aminoglycosides"
Aminoglycoside Year of First Pathway Formula'

Description
Streptomycin (Al) 1944 Ca(4)-6DOH(2)-HA

Neomycins (All) 1949 HA-(4)Cb(S)-P(3)-HA
5 ' -Hydroxystreptomycin(Al) 1950 Ca(4)-6DOH(2)-HA
Paromomycins (catenulin) (All) 1952 HA-(4)Cb(S)-P(3)-HA
Kanamycins (A12) 1957 HA-(4)Cb(6)-HA
Trehalosamines (A19) 1957 HA( 1)-H
Hygromycin B (A16) 1958 Cb(S)-H(2,3)-HepA
Streptozotocin (AM) 1960 HA
Spectinomycin (A2) 1961 Ca(4,5)-6DOH
Bluensomycin (glebomycin) (Al) 1962 Ca(4)-6DOH(2)-HA
Gentamicins (A14) 1964 HA-(4)Cb(6)-PA
Kasugamycin (A4) 1965 Ca(4)-6DOH
Destomycins (A16) 1965 Cb(S)-H(2,3)-HepA
Nojirimycin (AN) 1966 HA
Apramycin (A15) 1968 Cb(4)-OctA(d)-HA
Tobramycin (A12) 1968 HA-(4)Cb(6)-HA
Sisomicin (A14) 1970 HA-(4)Cb(6)-PA
Ribostamycins (A13) 1970 HA-(4)Cb(S)-P
Validamycins (A21) 1970 Cb(l)-Cb(4)-H
Lividomycins (A11) 1971 HA-(4)Cb(S)-P(3)-HA
Butirosin (A13) 1971 HA-(4)Cb(5)-P
Acarbose (A22) 1972 Cb(l)-6DOH(I)-(H),
Myomycin (AS) 1973 Ca(4)-HA
SS-56-C (A16) 1973 Ca(S)-H(2,3)-HepA
Minosaminomycin (A5) 1974 Ca(4)-6DOH
Amylostatins (A22) 1974 (H)n-(4)Cb(l)-sDOH(l)-(H)"
Siastatin (A18) 1974 HA
Verdamicin (A14) 1975 HA-(4)Cb(6)-PA
Sorbistins (A24) 1976 Ca(2)-HA
Seldomycins (A17) 1977 HA-(4)Cb(6)-PA
LL-BM123a (A6) 1977 Ca(4)-H(4)-HA
Fortimicins (A10) 1977 Ca(6)-HA
1-Desoxynojirimycins (AM) 1978 HA
Adiposins (A22) 1978 (H)n-(4)cb(l)-H(l)-(H)n
Trestatins (A22) 1978 (H)n-(4)Cb(l)-6DOH(l)-(H)n
Istamycins (A10) 1979 Cb(6)-HA
Sporaricins (A10) 1979 Cb(6)-HA
Sannamycin (A10) 1979 Cb(6)-HA
Dactimicin (A10) 1979 Ca(6)-HA
Oligostatins (A22) 1979 (H)n-(4)Cb(1)-6DOH(1)-(H)n
Lysinomicin (A10) 1981 Ca(6)-HA
Spenilomycin (A3) 1982 Ca(4,5)-6DOH
Valiolamine (A21) 1984 Cb
5 '-Hydroxy-N-demethyldihydrostreptomycin(Al) 1985 Ca(4)-6DOH(2)-HA
1-N-Amidino-1-N-demethyl-2-hydroxy- 1985 Ca(S)-H(2,3)-HepA
destomycin (A16)
Neotrehalosadiamine (A19) 1986 HA(1)-HA
AC4437 (Al) 1986 Ca(4)-6DOH
Allosamidin (A25) 1987 Ca(4)-HA(4)-HA
CV-1 (A18) 1987 HA
Boholmycin (A7) 1988 HA-(4)Ca(6)-PA(4)-Hep
Ashimycins (Al) 1989 Ca(4)-6DOH(2)-HA(4)-(H),
Trehazolin (A23) 1991 Ca(l,2)-HA
a For references before 1986 see UMEZAWA and HOOPER(1982) and UMEZAWA et al. (1986), others are
referenced in the text.
Numbers in brackets refer to the figures showing the formulae in Sect. 10.
' See text in Sect. 3; only the basic formula is given if a group of compounds is characterized.
G I 6
&I
their ecological role in the biotopes of the
ZJ %z: producing organisms. There, these substances
z z al=Iz---ou
mcm+cp might be involved in the communication be-
l a ws~-ilr
mmYuzaYuI>I
51” ~tween different organisms and in the hor-
mone-like crosstalk between the differen-
Strept omyces tiated cells of the producer itself (CHADWICK
Strepto verticillium and WHELAN, 1992; PIEPERSBERG,1993).
Streptosporangium Also, the routes of dissemination of amino-
Micromonospora
Dactylosporangium glycoside resistance mechanisms found in
Amycolatopsis both producers and clinically relevant non-
Actinoplanes producing bacteria (cf. Sect. 4.5) and the driv-
Saccharopolyspora ing selective pressures under natural condi-
Corynebacterium tions might become accessible to direct ex-
Bacillus perimental research.
Glycosidic or cyclitol-containing compo-
fig. 1. Aminocyclitol aminoglycoside-producing nents similar to the so-called secondary
bacterial genera (extended according to HASEGA- metabolites of microbes and plants occur in
WA, 1991).
the structures of many pro- and eukaryotic
A C acarbose and related glycosidase inhibitors; extracellular polymers, such as polysacchar-
BM: bluensomycin; BU: butirosins; DM: destomy-
cins; FM: fortimicins (astromicins); GM: gentami- ides, lipopolysaccharides, glycolipids, and
cins; HM-A,B: hygromycins A or B; IM: istamycins; other glycoconjugates. So far, in gram-nega-
K G kasugamycins; KM: kanamycins; LM: livido- tive unicellular bacteria, only very little evi-
mycins; NM: neomycins; PM: paromomycins; RM: dence has been found for the biosynthesis
ribostamycins; SE: seldomycins; SI: sisomicins; SM: and secretion of aminoglycosides or other
streptomycins; SP: spectinomycins; TM: tobramy- carbohydrate-containing diffusible substances
cin; VM: validamycins. (e.g., pseudomonads produce sorbistins).
However, these organisms produce a variable
and abundant range of carbohydrate-based
from plants should also be considered. In extracellular substances, lipopolysaccharides
general, it may be assumed that these glyco- (LPS), and other heteropolymeric polysac-
sides share a common general biochemical charides (e.g., enterobacterial common anti-
basis with those of bacteria and that an equi- gen: ECA). There is also emerging evidence
valent gene pool is used for their cellular pro- that the same gene pool is used for the pro-
duction. However, since little is known about duction of these polymers, which in many in-
their biosynthesis, this chapter concentrates stances resemble polymerized secondary me-
on prokaryotic production systems. tabolites, and for the production of the low
Research covering all aspects of the biology molecular weight end products of, e.g., acti-
of secondary carbohydrate metabolism in or- nomycete secondary metabolism itself (see
ganisms producing secondary extracellular below).
products is still in its initial stages. In contrast, Two recent examples of non-eubacterial
the resistance mechanisms of antibiotic pro- aminoglycoside-like compounds are (1)
ducers for self-protection can be regarded as the glucosaminyl archaetidyl-myo-inositols
having been extensively investigated (see (Fig. 2) produced by some methanogenic Ar-
Sect. 4.5). Even the basis of physiology in nat- chaea (KOGAet al., 1993) and (2) the very
ural biotopes and the ecological role of ami- similar core structure of the glycosyl
noglycosides and other secondary metabolites phosphatidylinositol protein anchors for out-
are unknown (PIEPERSBERG, 1993; MARSH er-surface-attached glycoproteins in euka-
and WELLINGTON, 1994). ryotes (Fig. 3). These are found in a variety of
Aminoglycosides have been shown to be organisms, from trypanosomes to mammals
produced in soils (WELLINGTON et al., 1993). (ENGLUND,1993). Interestingly, they have
Therefore, one of the future aims of amino- the same content of nonacetylated glucosam-
glycoside aminocyclitol research is to study ine and share the 1-phosphoryl-6-glucosami-
used for different purposes in the metabolism

of mainly extracellularly targeted compounds
and that they occur in practically all groups of
organisms.
3 Biogenesis and Basic

Pathways of Cyclitols and
NH*
Sugar Components
Fig. 2. An aminoglycoside-like glycolipid structure
in Archaea. It is unknown what is the general purpose
R archaetidyl residues ( = glyceryl ethers). of sugar modification and incorporation into
homogenous (sugar-based) or mixed-type
(“glycoconjugate-like”) oligomers - mostly
nyl myo-inositol core unit which is synthe- diffusible or membrane-associated substances
sized in the mammalian system from phos- outside the cells - and polymers, which are
phoinositides and UDP-N-acetylglucosamine mostly coating materials for cell surfaces or
at the inner side of the cytoplasmic mem- proteins outside the cells. It might be neces-
brane. Following deacetylation and acylation sary for cell-cell interactions which could be
at the inositol moiety it is transported to the called “communication”, “special purpose”,
outer surface where it is condensed to other or “individualization metabolism” and which
glycosidic residues (mannosyl, and in some might be a common feature of all types of or-
cases galactosyl, N-acetyl-galactosaminyl, ganisms (CHADWICKand WHELAN,1992;
phosphorylethanolamine). The phosphoryl- PIEPERSBERG, 1992, 1993). At least there is
ethanolamine unit connects the short manno- good evidence that most of the sugar metab-
syl chain to the C-terminal carboxyl group of olism beyond glucose, fructose, and some
the protein via formation of a new amide basic pentoses and their incorporation into
bond (reviewed by ENGLUND,1993). These storage and cell wall materials is specific for
examples might suggest that part of the bio- the cell type or strain, rather than for a spe-
chemistry of aminoglycoside biosynthesis cies or higher taxa, and it seems to be based
and, hence, part of the relevant gene pool are on the same common and highly variable and
t?
0-7-0-R
OH
Fig. 3. Structure in glycosyl
phophatidylinositols resem-
bling to aminoglycosides.
Such structures are protein
anchors of proteins bound to
(protein) the outer surface of eukaryot-
0 ic cytoplasmic membranes.
I
Asp-EA-Ph8-Man-a(l->P)Man-a(l
Asp: aspartyl residue in the
protein chain; EA: ethanol-
k
(glycosidic or
I
(glycosidic or
amine; Man: mannosyl resi-
dues; P phosphate; R: diacyl-
other residues) other residues) glycerol.
fluid gene pool. This is reflected by high var- amine (Fig. 6; WIDLANSKIet al., 1989; GODA
iations in glycosylation reactions in all organ- and AKHTAR,1992; RINEHARTet al., 1992;
isms and especially in differentiating organ- YAMAUCHIand KAKINUMA1992c, 1993).
isms. This gives rise to the high antigenic var- This mechanism involves
iation in microbial and higher eukaryotic
cells. Examples of this are the different blood - removal of electrons at C-4 of hexosephos-
group markers in man and the capsular mate- phates (C-5 of heptulose derivatives);
rials, lipopolysaccharides, and various forms - elimination of the phosphoester hydroxyl
of excreted glycosylated products in bacteria, as a leaving group would create a potential
which are described in this chapter. for the formation of a carbanion at C-6
(C-7);
- finally, after the condensation reaction the
3.1 Cyclitols keto group at C-4 (C-5) is reduced to re-
store the hydroxy group (cf. Fig. 4).
According to our present knowledge there
are two different (amino-)cyclitol pathways This step is similar to that which occurs on
(Fig. 4). a C7 sugar acid in the ring closure reaction
during the initial biosynthetic pathway lead-
(I) The myo-inositol pathway. The first route ing to the aromatic amino acids. The main dif-
(designated Ca in this chapter) is via D-myo- ference in the intitial intermediates of the Ca
inositol-3-phosphate which is synthesized by and Cb pathways is that in Cb a first transami-
the NAD +-dependent myo-inositol synthase; nation step can follow immediately whereas
end products are, e.g., D-chiro-inositol, strep- in Ca a phosphatase and an oxidase (dehy-
tidine, bluensidine, actinamine, and D-myo-l- drogenase) reaction has to first create a suita-
deoxy-l-amino-inositol (D-myo-inosamine). ble keto intermediate for a primary amino-
A compilation of Ca cyclitols is given in Fig. 5 transfer (see below, Sect. 4.1.3). The further
(FLOSS and BEALE, 1989 WALKER,1975a; processing of cyclitols to aminocyclitols might
SIPOSand SZABO,1989). The reactions cata- even be at least in part very similar, i.e., the
lyzed by the bacterial and the eukaryotic D- first aminotransfer and other later interme-
myo-inositol-3-phosphate synthases follow diate steps, and might be catalyzed by related
that of a typical intramolecular aldol conden- or even almost identical enzymes. This can be
sation. This involves predicted although the enzymes involved are
not yet known for most of representative
- opening of the pyranose ring and rotation groups of (amino-)cyclitol pathways. Howev-
of the C-4/C-5 bond to place the C-6 group er, there is one detail in which also the later
in an active position, stages of Ca and Cb cyclitol conversion differ:
- the reversible oxidation and reduction of in 2-deoxystreptamine (2DOS) formation the
the C-5 position by NAD+, incorporation of amino groups stereochemi-
- the removal of the pro-R-H atom from C-6 cally proceeds in a direction opposite to that
and its transfer to the C-1 carbonyl, and shown to occur in streptidine biosynthesis
- the use of the p-anomer of ~-glucose-6- (WALKER,1975a; see Sects. 4.1 and 4.3).
phosphate as the preferred substrate The chemical concept of aliphatic and aro-
(WONG and SHERMAN,1985; FLOSS and matic C7 units which basically constitute cy-
BEALE,1989). clohexane derivatives with a nitrogenous
group in the meta position (m-C7N) domi-
(2) The dehydroquinate pathway. The second nates the literature on the biogenesis of build-
route to cyclitols (Cb) follows a NAD+-de- ing blocks in many secondary metabolites
pendent dehydroquinate-synthase-like mech- (RINEHARTet al., 1992; FLOSSand BEALE,
anism on open chain hexose-6- or heptulose- 1989; CHIAOet al., 1995; KIM et al., 1996).
7-phosphates yielding nonphosphorylated 1- These are formally all cyclitol derivatives,
keto-2-deoxy-cyclitols. End products are, e.g., most probably formed via the Cb pathways.
6-deoxystreptamine and valienamine/valiol- An aliphatic m-C7N unit is found in valien-
404 I0 Aminoglycosides and Sugar Components in Other Secondary Metabolites
Enz Enz -
T T
B-H*
H’
H o \w h
OH
HO
OH H
-
OH
[i
rOH
HO
gH)
__c
“AD*]
0
AL 0-Ph OH
(5)
HO
AL 0-Ph
0
- 0
OH b 0 CH2
0
OH’
OH
aminehalidamine-containing substances, such The common dehydroquinatehhikimate

as validoxylamines and validamycins (cf. Sect. pathway of many aromatic compounds should
10, Fig. A21), in acarbose, and in related gly- be regarded as a Cb pathway and is involved
cosidase inhibitors (see Fig. A22; TRUSCHEITin the formation of many aromatic m-C7N
et al., 1984; MULLER,1989). units (cf. Fig. 4). These units occur in many
4 Fig. 4. Routes to the formation of (amino-)cyclitols other molecule derived from the same path-
via myo-inositol-~-3-phosphatesynthase (Ca) and way (e.g., validamycin, acarbose, and other
a dehydroquinate synthase-like enzyme mechanism related compounds; see Sect. 4.4). In extreme
(Cb; cf. reactions 8-7). Substrates can be D-glucose- cases, such as in pactamycin (see Fig. A30),
6-phosphate (1 and 3), sedoheptulose-7-phosphate
(5, or its 5-epimer), or 3-deoxyarabinoheptulosonic they can constitute the central building block
acid (7). The products of Ca or Cb pathways are which is condensed with four or more side
either cyclitolphosphates (2) which have to undergo groups of different origin and varying com-
dephosphorylation and further oxidation before plexity. In pactamycin they consist of two aro-
transamination or 1-keto-2-deoxycyclitols (4,6, and mates (one derived from the dehydroquinate
8) which can be transaminated directly, respective- pathway, a m-C7N unit, the other from the
ly. The numbering of ring atoms in (2) and (4) is polyacetate/polyketide pool), a dimethylated
according to the counting system used in strept- urea group, a hydroxyethyl group derived
amine derivatives (streptidine, actinamine, 2-deoxy- from the methyl groups of methionine, and a
streptamine); numbers in (6) and (8) are given ac- C-bound methyl group. In addition, four of
cording to the nomenclature in valienamine and de-
hydroquinate, respectively. 0 (C-1 or C-2) and A the five C atoms of the pentitol moiety which
(C-6 or C-7) mark the original carbohydrate atoms seems to be a myo-inositol derivative are sub-
forming the new C-C bond in cyclitol products. stituted with nitrogenous group.
The dehydroquinate pathway (7-8) can also be However, it is doubtful whether some of
started with a 5-amino-3,5-deoxyarabinoheptulo- the representatives of a common chemical
sonic acid or an intermediate becomes transamin- group of aminoglycosides share the same cy-
ated after cyclization to yield an aromatic m-C7N clitol pathway. For instance, the destomycin
unit as is found in many secondary metabolites (for group contains two compounds with aminocy-
detail, see FLOSSand BEALE,1989; RINEHARTet clitols derived from streptamine, one of which
al., 1992; YAMAUCHI and KAKINUMA, 1992c,
1993). is identical to 1-N-amidinostreptamine (see
Fig. A 1 6 IKEDAet al., 1985b), an interme-
diate in the streptomycin pathway (Ca path-
way; see Sect. 4.1.1). The other members of
well-known secondary metabolites, such as this family have 2-deoxy-scyllo-streptamine
many ansamycins, e.g., rifamycins, streptovar- (2DOS) as cyclitol moieties (Cb pathway).
icins, geldanamycin (not shown), in candicid- Similarly, the fortimicins (Ca) and the istamy-
ins (not shown), and in pactamycin (cf. Figs. 5 cins (Cb) could be put into separate groups
and A30), which is unusual in that it contains (see Fig. A10 and Sect. 4.2). This might sug-
components seemingly derived from the two gest even two different initial pathways for
alternate cyclitol pathways, one following the cyclitol formation resulting in very similar
Ca and the other an aromatic Cb route differ- end products or, alternatively, a subsequent
ent from that of the ansamycin family (RINE- modification by oxidoreductases (dehydroxyl-
HART et al., 1981, 1992; FLOSSand BEALE, ation of Ca derivatives or hydroxylation of
1989). A detailed account of the aromatic Cb Cb products) as optional routes to produce
pathway is beyond the scope of this chapter; the alternate category of building blocks.
the reader is referred to the literature However, the latter possibility is less likely
(CHIAOet al., 1995; FLOSS,in press; KIM et since in no case were both pathways found in
al., 1996). the same producer.
The cyclitol moieties generally may be re-
garded as aglyca in glycosylation reactions,
e.g., in streptomycins (see Sect. 4.1), fortimi- 3.2 Sugar Components
cins, and istamycins (see Sect. 4.2), most 2-
.deoxystreptamine-(2DOS-) containing ami- The sugar derivatives which can be at-
noglycosides (see Sect. 4.3), and some other tached to the cyclitols of both the Ca and the
products such as the nucleoside antibiotic ad- Cb type, may stem either from nucleotidyl-ac-
enomycin (cf. Fig. A27 and Sect. 5.2). Alter- tivated monomeric precursors (glycosyltrans-
natively, they are (co-)substrates in other fer reactions) or from the pool of polymer-
types of condensation reactions, e.g., with an- ized D-glucose directly (e.g., maltodextrins;
Sy O OH H 2
HO OH
0O H
6H
HO OH
o O
6H
OH HO
bSO3H
OH HO
- OH
rnyeinositol D-chifeinositol (adenomycin) streptamine
(kasugamycin) (SS-56C)
HN=C-NHz
H2N-zo
H3c00
streptidine fortamine rnyeinosamine
(streptomycin) (fortimicin A) (rninosaminomycin)
HzN-C=O
H-CH, OH H~ -!::
HO OH OH
HO
OH OH OH Fig. 5. Structures of (amino-)hexito1
"0:HoooH
actinamine (LL-BM 12%) (myomycins) and pentitol components in secondary
0;
(spectinomycin)
metabolites known or assumed to be
formed via a Ca pathway (see Fig. 4).
CH20H
Some examples of their occurrence are
given in brackets. Known or assumed
positions derived from the original C-1
(0)and C-6 (A) of the glucose precur-
HO-$H sor
hazolin
are labeled
and allosamidin
in the pentitols
no suggestion
of tre-
H3C OH ACH~OH OH NH2
can be given without any direct experi-
(pactamycin) (trehazolin) (allosamidin) mental evidence.
transglycosylations). For a systematic analysis (4) spacing non-carbohydrate-derived resi-

of the pathways involved we define here the dues are unspecified (X).
following general routes by which carbohy-
drate building blocks are preformed: Thus, for the purposes of the discussion
here and for simplification a "pathway formu-
(1) the 6-deoxyhexose pathway (6DOH); la" is defined for each carbohydrate-contain-
(2) the pyranosidic or furanosidic pentose ing compound considered. The postulated
(P), hexose (H), heptose (Hep), octose pathway for the derivation of the mono- and
(Oct), or the hexosamine (HA) pathway; oligosaccharidic structures is given by con-
(3) some glucosyl residues ((H),) are possi- necting the above abbreviations for the indi-
bly incorporated via transglycosylation vidual precursor pathways where the figures
from di- or oligosaccharides such as mal- in brackets indicate the points of glycosidic
todextrins, especially where a-l,4-bound (or other) substitution. For example, strepto-
glucose or maltose moieties are encoun- mycin (Ca(4)-6DOH(2)-HA) or mannosyl-
tered mannosidostreptomycin (Ca(4)-6DOH(2)-
2-deoxy-scylb P-deoxy- (istamycinA: R1,2= NHz, H)

inosose streptamine (istamycinB: R1,2= H, NHz)
CH2OH CHzOH CHeOH
C;H~OH OH OH
1-ketovaliol valienamine validamine valiolamine
OH
no
Fig. 6.Structures of (amino-)hexi- I -
OH 6H O
-.U.
tols and pentitols or benzoic acid 2- hydroxy-
OH
derivatives in secondary metabol- validamine (oligostatins) ~~~~~~~
dehydroquinate
ites known or assumed to be
formed via a Cb pathway (see
Fig. 4). Someare

occurrence examples
given inof brackets.
their HoH$QAd$naAd
Known or assumed positions de- OHA
rived from the original C-1 or C-2
(0)and C-6 or C-7 (A) of the
- -
glucose or heptulose (heptulosonic OH OH OH OH COOH COOH
acid) precursors, respectively, are (aristeromycin) (neplanocinA) 3-aminod-hydroxy- 3-amino-
labeled. benzoic acid benzoic acid
HA(4)-(H)J would be classified together two steps in their pathway is accomplished by

with spectinomycin (Ca(4,5)-6DOH) and set the enzymes dTDP-D-glucose synthetase and
apart from neomycins (HA-(4)Cb(6)-P(3)- dTDP-D-glucose 4,6-dehydratatase yielding 4-
HA), validamycins (Cb(1)-Cb), and amylosta- keto-6-deoxyhexose intermediates (Fig. 7).
tins ((H),,-(4)Cb(l)-6DOH(l)-(H),,) (see Sect. The 4-keto compounds can be used as a com-
10 and Tab. 1 for a compilation). mon precursor for branching the further path-
ways into the D- and L-series of hexose deri-
vatives (Figs. 8 and 9). The ~ - 6 D 0 H are
s iso-
3.2.1 6-Deoxy- and Other merized from the D-configurated precursors
by a 3,5-epimerase (see Fig. 7). In many cases
Deoxyhexoses both biosynthetic routes are followed in the
same producing cell. Frequently, by such
D-glucose, a precursor of most or all 6- branching routes both D- and L-6DOH deri-
deoxyhexoses (6DOH) in prokaryote-formed vatives are formed and are specifically incor-
(antibiotic-like) secondary metabolites, seems porated into a particular complex end prod-
to be generally activated by deoxythymidine uct, e.g., macrolides and some angucyclins. In
diphosphate (dTDP). Catalysis of the first aminoglycosides, 6DOH components are re-
hexose-1 -P NDP-hexose
nucleotidyC 4,g-dehydratase
transferase CH*oH
CHzOH
HOQo-P HO
QO-NDP - O o O - N D p D-GDOH
OH OH OH
1
NDP4keto-
Meoxyhexose
3,S-epimerase
OQHO-NDP L-6DOH
Fig. 7. General pathway for the biosynthesis of 6-deoxyhexoses. Hexose-1-phosphate precursors are either
D-glucose-1-phosphate or D-mannose-1-phosphate.
P: phosphate; NDP dTDP, CDP, or GDP.
latively rare compared with the dominance 6-deoxy hexose via a radical mechanism and
and variability among modifying sugar side by forming a covalently bound PMP-hexose
chains in other chemical classes of microbial intermediate (Fig. 10). The electrons for this
natural products. process are delivered via a second enzyme
Important and interesting types of further (E2) which contains FAD in addition to the
modification are (1) the deoxygenations (for- [2Fe-2S] cluster and uses NADH as an elec-
mally: dehydroxylations) at C-2, C-3, and C-4, tron donor. This type of mechanism could
(2) the transaminations (formally: exchange easily be envisaged also to be involved in the
of hydroxyl for amino groups) at C-2, C-3, 2-deoxygenation of many other 6DOHs in
and C-4, and (3) the isomerization and epi- secondary metabolites, such as the sugar con-
merization steps. Other types of modifica- stituents in the daunorubicin-cytorhodin-
tions, such as C, N, 0, and S methylations or rhodomycin group of anthracyclines or some
transfer reactions for more complex side of the 6DOHs occurring in chromomycins (cf.
groups are also common in the 6DOH path- Figs. 8 and 9).
ways. Here we briefly describe only the first
mechanism; the transaminations and other
types of isomerization (e.g., epimerization) 3.2.2 Other Sugar Components
reactions are discussed below. A mechanism
for the deoxygenation steps was elucidated by As outlined above biogenesis studies in
the recent studies of the CDP-3,6-deoxyhex- early phases of antibiotic research have often
ose pathway (REEVES,1993; SHNAITMAN and misleadingly suggested that D-glucose, D-glu-
KLENA,1993; THORSONet al. 1993; THOR- cosamine, glycerol, or related carbon sources
SON and LIU, 1993a, b; LIU and THORSON, are directly incorporated into sugar consti-
1994). This is an alternative pathway yielding tuents in secondary metabolites. This was in
6DOHs which is abundantly used in gram-ne- most cases interpreted as an indication that
gative bacteria besides the dTDP-pathway, the regular routes of sugar activation and fur-
e.g., in the biosynthesis of lipopolysaccharide ther processing are adopted from' primary
0-chains. The enzyme system involved con- metabolic routes, e.g., the UDP-hexosamine
sists of two iron-sulfur proteins one of which pathway in bacterial cell wall biosynthesis.
(El) catalyzes the pyridoxamine phosphate However, our rapidly increasing knowledge
(PMP)-dependent dehydration of the 4-keto- of the extreme specificity of secondary meta-
3 Biogenesis and Basic Pathways of Cyclitoki and Sugar Components 409
bolic traits does not support this view. Unex- hexose-derived moieties in many chemical
pected modes of sugar activation (e.g., nu- groups of heterogenously composed second-
cleotidylation by CTP instead of UTP), 2-ami- ary metabolites.
no-hexose derivation from precursors other
than D-glucosamine, and unknown mecha-
nisms of modification and condensation have
been found (some examples are given below;
see Sects. 4.1, 4.2, and 5.1). Some of the
known or hypothetical general routes for the
4 Genetics and
biosynthesis of hexosamines are outlined in
Fig. 11; it should be emphasized that 6-amino-
Biochemistry of the
6-deoxyhexoses and 2-amino-2-deoxyhexoses Biosynthesis and
also could be formed in the same way as the
3- and 4-isomers and that the same principal Functions of
route also should yield amino-group-contain-
ing pentoses, heptoses, and octoses. Further Aminoglycosides
variation may come from the transamination
step, before or after sugar activation by nu- Aminoglycosides are largely actinomycete
cleotidylation or even after condensation into products though rare occurrences in other
complex molecules. bacterial groups are known (e.g., butirosin-
Several of the unusual C-5 to C-9 sugar producing Bacillus circulans, sorbistin in
derivatives encountered in many groups of pseudomonads, and N-methyl-scyllo-inos-
microbial products are found in the aminogly- amine in rhizobia; see Fig. 1). Aminoglycoside
coside-related and other microbial secondary production is mainly a property of members
metabolites. Some examples comprise pyra- of the filamentous actinomycete genera Strep-
nosidic 2- or 3-pentosamines (e.g., in seldo- tomyces spp., Streptoverticillium spp., and
mycins and gentamicins, respectively; cf. Figs. Saccharopolyspora spp., and of Micromono-
A14 and A17), 2-deoxyhexoses (e.g., in cytos- spora spp., Amycolatopsis spp., and Actino-
aminomycins; see Fig. A27), 3-deoxyhexos- planes sp. (WILLIAMS et al., 1989). The biolo-
amines (e.g., in lividomycins; Fig. A l l ) , 4- gy, biochemistry, mode of action, biotechnol-
deoxyhexosamines (e.g., in seldomycins; see ogy, and clinical applications of the aminogly-
Fig. A17), L-hexoses (e.g., L-mannose in de- coside antibiotics in addition to aminoglyco-
sertomycin; not shown), and L-hexosamines side resistance have been often reviewed
(e.g., in streptomycin; see Sect. 4.1 and Fig. (e.g., KORZYBSKIet al., 1978; DAVIESand
Al). These moieties are most probably all ac- SMITH,1978; WALLACEet al., 1979; PIEPERS-
tivated and modified via nucleotidylated in- BERG et al., 1980 WALKER,1980 PEARCE
termediates. However, it is thought that sim- and RINEHART,1981; UMEZAWAand Hoo-
ple glucosylations and mannosylations in- PER, 1982; DAVIESand YAGISAWA, 1983;
volve UDP-D-glucose and GDP-D-mannose, FOSTER,1983; UMEZAWA et al., 1986; CUND-
respectively. Furanosidic pentoses, such as ri- LIFFE, 1989, 1990; GRAFE, 1992; PIEPERS-
bose (e.g., in neomycins; see Fig. A l l ) or ara- BERG,1995; HOTTA et al. 1995). Recent in-
binose moieties (e.g., in nucleoside antibiot- vestigations have concentrated mainly on the
ics), might be introduced from 5-phosphori- molecular biology of the clinically relevant
bosyl-l-diphosphate (PRPP) as activated pre- aminoglycoside resistance and molecular as-
cursors or, again, via an unknown activation pects of aminocyclitol aminoglycoside biosyn-
and transfer mechanism. The dehydroxyla- thesis, resistance, and regulation in the pro-
tion reactions of pyranosidic hexoses (other ducing organisms (mainly with respect to
than 6DOH) at positions C-2, C-3, and C-4 streptomycins and fortimicins; see Sects. 4.1
are especially interesting from a mechanistic and 4.2). During the past decade, semisyn-
point of view. Several examples are known in thetic modification, new screening methods,
aminoglycosides (see, e.g., the discussion in and new fields of application have resulted in
Sect. 4.2 on the fortimicin group) and other the discovery of several new aminoglycoside
HO
D O H HZN
D O H 03)(-J# HZN OH
OH 0 OH NHC(NH)NHz
6-deoxy-D-glucose actinospedose glucocinnamoyl-
(acarbose?) (spectinomyan) spermidines
(LL-BM123-complex)
CH3
H3C0
&OH
CH3
0-methyl-D-rhamnose D-mycosamine D-perosamine D-sibirosamine
(antibiotic A201) (amphotherian B; (perimycin) (sibiromyan)
nystatin)
i=)
CHa
HO OH
H3C0 OCH3 OH OH OH
D-mycinose D-mycaminose Ddesosamine Damosamine
(angolamyan) (tylosin) (erythromycin) (amicetin)
N(CH3)2 6H
D-everminose (ravidomyan) D-kasugamine D-digitoxose
(eveminomicin) (kasugamycin) (lipomycin)
"0 OH
OH
OH
H 3 c 0 0 0 H
D-fucose D-digitalose D-ossamine D-chromose A

(benanomian A; (chartreusin) (ossamydn) (chromomycins)
chartreusin)
AHCHI
D-olivose D-forosamine (neocarzinostatin) D-isomycamine
(urdamydns; (forosamydn) (spiramycin)
0- & Cglycosidc) Fig. 8
"Po- NH2HO OH H3CHN

O
OH
O H HO
O O H
D-3-epimycosamine D-thomosamine D-vicenisamine D-2,3,6-tri-
"0
(macrolactam) (pradimidns; (vicenistatin) deoxyglucose
R1.2 = H OT CH3) (dutomycin)
I
OH H O D o H H O GOH
NH2 OH OH
D-baallosamine D-fucose 0-olivose D-oliose
(LPS, 0-chains; (benanomicin A; (olivomycin) (olivomydn)
"0
Pseudomonas) chartreusin)
R OH
OH
Ow0
(avlamycin; (avilamydn) Dnoviose D-vicenisamine
R = acetyl, (novobiodn) (vicenistatin)
hydroxyethyl)
"0HoooH
"0- OH
OCH3 OH
OH
Ddiginose D-sarmentose D-boivinose Dchalcose
bN<
(heart glycosides) (heart glycosides) (stroboside) (chalcomydn)
>
(Io
.H
H3C0 H3C0
O O H
OH HO
HO H3C0
H3CHN OH
6deoxy-D-mannose (notonesomycin) 4-O-methyl-2,3,6-lri-
(staurosporine) (lienomycin) Ddeoxyglucose
(A204A, polyether)
Gr;)
CH3
Fig. 8. Structural variants of D-6-deoxyhexoses (D-

OAc OH 6DOH) found in secondary metabolites. Examples
of their occurrence in particular natural products
[(-)griseusin A] are given in brackets.
Fig. 9
0 0
OH OH
HO HO
OH OH
0
L-rhamnose L-fucose 2-deoxy-L-fume L-rhodinose
aG
(cytorhodins) (cytorhodins)
(oxopropalines) (EPSa)
CHO(or CH@H) CHO (or CH@H)
HO OH
OH OH OH OH NH2
L-(dihydm-)-5- L-daunosamine L-rhodosamine
L-(dihydm-)str*tose h y & o ~ s ~ e p ~ ~ (cytorhodins)
(streptomyans) (daunorubicins)
(streptomyans;
LPS. 0-chains)
OH ' * o O H O eOH @O OH
CH3 OH
L-dneruloseA L-dnerulose B
0
L-mycarose Lchromose B (cytorhodins)
(chromomycins) (cytorhodins)
(macrolides)
HO OH @O - OH
NH-R b
L-amicetose L-awlose
4-methyl-L-
Hou
(glycopeptides: (cytorhodins)
rhodosamine (cytorhodins)
R = -H, -methyl,
-CH&~tyl) (saptomyans)
" U O H HO
D OH O H OH
HoQoH OH
H3C0
CH3 CH3
Lcladinose L-axenose L-ristosamine (calicheamidn)
(maadides) (axenomycins) (ristomycin)
O CH3
O
I
NH2
O H
"VoH OCH3 H3Cb

L-nogalose
OCH3 CH3
Ldecilnitrose
(balhimycin) L-oleandrose (anthracydines)
(anthracyclines)
(avermectin)
Fig. 9
OOH
OOH
HO
QOH SUC-0 H NH2
O V OH
OH OH ( R ) L H H3C0 CH3
6deoxy-L-talose (TAN-1120; (avidinorubicin; avidinosamine
(phenazoviridin) R = cydic ether) SUC= sucdnyl) (avidinorubicin)
H~NOCO OH
L(or D)-noviose 2-O-methyl- (foraminosyl-
(novobiocin) L-rhamnose (+)griseusin A)
(helvecardins)
Fig. 9. Structural variants of L-6-deoxyhexoses (L-6DOH) found in secondary metabolites. Examples of

their occurrence in particular natural products are given in brackets.
0
0 0-NDP
-
El
PMP H
Ph-0
' /
N
\
~ ~ C )
QNDP
CH, 0 H' OH
OH
Ph-0
\
1I OH
>No
HN+20 Ph-0
>
0-NDP
OH
'0-NDP
Fig. 10. Mechanism of 3-de-

hydroxylation in the 3,6-di-
deoxyhexose pathway of
gram-negative bacteria.
El: PMP-dependent, [2Fe-
2S]-containing dehydrase
(AscCIRfbH-type); E2:
NADH-dependent, FAD/
[2Fe-2S]-containing dehy-
drase (for details, see LIU
and THORSON, 1994).
i (R = Acor H)
1
D4HA
owO-:Dp - R
0-NDP DIL3HA (3PA)
NH2
H 2 . 0 ONDP L4HA
(R = CYOH or H )
Fig. 11. General scheme for the derivation of 2-, 3-, or 4-aminated hexoses and pentoses. Also, 2- and
6-aminohexoses (HA), 2-aminopentoses (PA), or aminoheptoses and aminooctoses could be formed in
their furanosidic or pyranosidic forms via the oxidation and transamination pathway.
structures and, in at least one case, in the in- 1989) or be cleaved into two hexose deriva-
troduction into pharmaceutical use. This tives as in AC4437 (dihydrostreptosyl-strept-
chapter will focus mainly on the latter as- idine; AWATAet al., 1986). Compounds of
pects. this group are produced by a broad range of
species (KORZYBSKI et al., 1978). They have
been found in the genera Streptomyces, Strep-
4.1 Streptomycins and Related Ca toverticillium, Amycolatopsis, and may be
present in others. A survey of streptomycin-
Aminoglycosides producing strains from collections and a tax-
onomic study of new isolates that was recent-
The streptomycins and bluensomycin are ly started (PHILLIPSet al., 1992; MARSHand
a relatively homogenous group of basic- WELLINGTON,1994) indicated that they can
ally pseudotrisaccharidic aminoglycosides mainly be grouped into three clusters within
(Fig. A l ) which can be elongated by one or the family of Streptomycetaceae. These clus-
two further glycosidic residues (mannosyl or ters are represented by S. griseus, S. hygro-
ashimosyl; IKEDAet al., 1985a; TOHMA et al., scopicus, and Streptoverticillium mashuense.
Therefore, the presently accepted taxonomy have been clarified, it is known that the com-
of the family Streptomycetaceae (WILLIAMS pounds of the streptomycin family are synthe-
et al., 1989; EMBLEY and STACKEBRANDT,sized via 25-30 enzyme-catalyzed steps
1994) assigns those streptomycin-producing (WALKER, 1975a; RINEHART and STRO-
species that have been investigated in detail SHANE,1976; GRISEBACH, 1978; RINEHART,
to clearly separated species clusters. Recently, 1980; OKUDAand ITO, 1982; PIEPERSBERG,
this view was confirmed and extended by se- 1995). The currently hypothesized streptomy-
quencing the 16s rRNA genes from several cin and bluensomycin biosynthesis pathways
streptomycin-producers (MEHLING et al., are summarized in Fig. 12. They involve the
1995). As with other secondary metabolites, formation of the activated precursors, strep-
the reason for this taxonomically scattered, tidine (bluensidine)-6-phosphate (cf. Sect.
but relatively stable distribution in indiviual 4.1.1.1), dTDP-dihydrostreptose (cf. Sect.
strains worldwide is not known. 4.1.1.2), and NDP-N-(methyl)-L-glucosamine
(cf. Sect. 4.1.1.3) which are made most likely
from ~-glucose-6-phosphate.However, it has
4.1.1 Streptomycins been demonstrated that D-glucosamine can
be incorporated without breakage of its C-C
The biosynthesis of streptomycins, includ- bonds into the N-methyl-L-glucosamine sub-
ing bluensomycin, was intensively studied by unit of streptomycin (reviewed by OKUDA
feeding labeled primary metabolic precursors and ITO, 1982). In a second phase, the precur-
(DEMAINand INAMINE, 1970 MUNROet al., sors are condensed forming dihydro-strepto-
1975) and analyzing streptomycin biosynthet- mycind-phosphate (cf. Sect. 4.1.1.4) which is
ic reactions in cell-free systems and with pu- secreted and probably oxidized while passing
rified biosynthetic enzymes (WALKER,1975a, through the cytoplasmic membrane to give
b; GRISEBACH, 1978; RINEHART,1980). Al- streptomycin-6-phosphate outside the cell (cf.
though not all of the biosynthetic reactions Sect. 4.1.1.4). The biologically active strepto-
F(NH)-NH,
in CM out
dTDPG dTDP-Dihydro- I
streptose
R OH NHCH,
-
NDPG(A)
(R NHZ OT OH)
NDP-&Methyl-
L-glucosamine
Fig. 12. Gerneral outline of the streptomycin (SM) pathway. The three activated intermediates principally
formed from glucose-6-phosphate (G-6-P) are condensed in the cytoplasm to dihydro-SM-6-phosphate
(DHSM-6-P) which is oxidized and dephosphorylated to SM during or after transport through the cyto-
plasmic membrane (CM).
mIP: myo-inositolphosphate; dTDPG: deoxythymidinediphosphate-glucose;NDPG(A): nucleosidediphos-
phate-glucose (or -glucosamine) (NDP = CDP or UDP).
mycins are liberated finally by a specific phos- operons are generally found which may re-
phatase (cf. Sect. 4.1.1.4), and after re-uptake flect the need for a strictly coordinated regul-
they can be phosphorylated by a streptomycin ation of strlsts gene expression in order to
6-phosphotransferase representing the resist- guarantee a coordinated supply of the acti-
ance mechanism which protects the producers vated precursors in streptomycin synthesis.
from their own products (cf. Sect. 4.5). The phenomena involved in genetic regula-
Genetic studies of streptomycin biosynthe- tion of streptomycin biosynthesis are dis-
sis in the various producers started with the cussed below (cf. Sect. 4.6).
cloning of resistance genes (reviewed in
CUNDLIFFE,1989; PIEPERSBERG, 1995). Two
genes were initially cloned, strA (aphD) and 4.1.1.1 Biosynthesis of Streptidine
aphE, which encode two different streptomy-
cin phosphotransferases, APH(6) and and Bluensidine
APH(3 ”), respectively (WALKER, 1975b;
WALKERand WALKER,1975; DISTLERand The biochemistry of the eleven steps in-
PIEPERSBERG, 1985; DISTLERet al., 1987a; volved in streptidine synthesis was evaluated
HEINZELet al., 1988). Only the strA gene by WALKERet al. (WALKER,1975a, 1990) by
could be localized up to now in the produc- means of enzymology and led to the proposed
tion gene clusters of S. griseus and S. gfuu- pathway outlined in Fig. 14. The initial step is
cescens GLA 0. The aphE gene, however, oc- the formation of ~-myo-inositol-3-phosphate
curs only in S. griseus strains and does not ap- from D-glucose-1-phosphate via the Ca route
pear to be linked to the strlsts cluster. Subse- (cf. Sect. 3.1) by an ATP-dependent myo-ino-
quently, further streptomycin biosynthetic sitol phosphate synthase (WALKER,1975a;
genes were identified by chromosome walk- SIPOSand SZABO,1989). Up to now, no strlsts
ing and genetic complementation of mutants gene could be identified which encodes the
from different S. griseus strains deficient in myo-inositol synthetase. Also attempts to pu-
streptomycin biosynthesis (DISTLER et al., rify the enzyme failed because of its instabili-
1985; OHNUKIet al., 1985a, b). About 30 ty (SIPOSand SZABO,1989). The end product
genes for (5 ’-hydroxy-)streptomycin (strlsts) is streptidine (SD)d-phosphate, which also
and bluensomycin (bfu) production have seems to be the first intermediate made from
been cloned and analyzed from various externally supplied SD in SD- mutants (OH-
strains of S. griseus, from S. gfuucescens NUKI et al., 1985a, b; DISTLERet al., 1985).
GLA.0 (ETH 22794), and from other Strepto- SD-6-phosphate is synthesized by two sets of
myces spp., and all were found to be clustered five parallel enzymatic reactions each pre-
in one region of about 3 0 4 0 kb of genomic sumably catalyzed by individual enzymes: two
DNA (Tab. 2; Fig. 13; MANSOURIet al., 1989; cyclitol phosphate phosphatases, two cyclitol
DISTLERet al., 1990, 1992; RETZLAFFet al., dehydrogenases, two aminotransferases, two
1993; PIEPERSBERG,1995). Comparison of phosphotransferases, and two amidinotrans-
the primary structures of the homologous strl ferases in streptomycin producers (cf. Tab. 2,
sts genes and their gene products in S. griseus Fig. 14).
and S. gfuucescens revealed identity values In contrast to this the producer of bluenso-
varying between 58% and 86%. In contrast, mycin S. hygroscopicus forma gfebosus lacks
the corresponding nonencoding intercistronic steps 8 to 11 (see Fig. 14; WALKER,1990)
DNA sections have much less or no signifi- which are replaced by carbamoylation and
cant homology. The order of the strlsts genes phosphorylation reactions at positions 5 and
identified so far within the respective operons 4, respectively, yielding bluensidined-phos-
is similar although the arrangement of the phate (see Fig. 15). The product of the strO
operons differs considerably among produc- gene has significant similarity to eukaryotic
ing species. Usually, the genes encoding en- inositolmonophosphate phosphatases (RETZ-
zymes for the synthesis of a subunit of strep- LAFF et al., 1993). Therefore, this protein is
tomycin (e.g., streptidine) are not arranged in assumed to be the enzyme which catalyzes
subpathway-specific operons. Instead, mixed step 2 in the streptidine pathway (RETZLAFF
Tab. 2. Gene Products and Enzymes Presumed to be Involved in Streptomycin Production
Gene Molecular Datab Enzymatic Function' Remarksd References"

Producta (Preliminary Assignment)
MW [kDa] aa
Unkown 216 (native) synthetase

~-myo-Inositol-3-phosphate ATP-dependent SIPOSand SZABO
(1989)
StrO 28 260 (256) (~-myo-Inositol-3-phosphate phosphatase) A, RETZLAFF et al. (1993)
StrI 37 348 (scyllo-Inosose dehydrogenase) NAD(P)-dependent MANSOURI and PIEPERSBERG (1991)
StsC 45 424 scyllo-Inosose aminotransferase Rel. to StrS and A, RETZLAFFet al. (1993)
StsA;
PLP-dependent
StrN 36 (35) 320 (316) (scyllo-Inosamine 4-phosphotransferase) PISSOWOTZKI et al. (1991)
StrBl 39 347 scyllo-Inosamine-4-phosphate DISTLERet al. (1987b)
amidinotransferase
Unknown N-Amidino-scyllo-inosamine- (cf. WALKER,1975a)
4-phosphate phosphatase
StsB 52 490 (N- Amidino-scyllo-inosamine NAD(P)-dependent A
dehydrogenase)
StsA 43 410 (3-Keto-N-amidino-scyllo-inosamine Rel. to StsC and A, RETZLAFFet al. (1993)
aminotransferase) StrS;
PLP-dependent
StsE 32 312 (N-Amidino-streptamine A
6-phosphotransferase)
StrB2 38 (35) 349 (319) N-Amidino-streptamine-6-phosphate PISSOWOTZKI et al. (1991)
amidinotransferase
StrD 38 355 dTDP-Glucose synthetase PISSOWOTZKI et al. (1991)
StrE 36 328 dTDP-Glucose 4,6-dehydratase NAD-dependent PISSOWOTZKI et al. (1991)
StrM 22 200 dTDP-4-Keto-6-deoxyglucose PISSOWOTZKI et al. (1991)
3,5-epimerase
StrL 32 304 dTDP-4-Keto-~-rhamnosedehydrogenase NADP-dependent PISSOWOTZKI et al. (1991)
[ = dTDP-L-dihydrostreptosesynthase]
StrX 20 182 (NDP-Hexose 3,5-epimerase) Rel. to StrM A, BEYERet al. (1996)
StrU 46 428 (NDP-Hexose oxidoreductase) NAD(P)-dependent A, BEYERet al. (1996)
StrF 32 281 (NDP-Hexose epimerase) MANSOURI and PIEPERSBERG(1991)
StrG 23 199 (NDP-Hexose epimerase) MANSOURI and PIEPERSBERG(1991)
StsG 27 246 (N-Methyltransferase) A
StrS 40 378 (377) (Aminotransferase; unknown function) Rel. to StsC and A, RETZLAFFet al. (1993)
StsA;
PLP-dependent
StrT 33 300 Unknown function A
StsD 24 213 Unknown function A
StsF 26 236 Unknown function A
StrV (ca. 45-50) (inc.) (Exporter for streptomycin ATP-dependent A, BEYERet al. (1996)
6-(or 3 "-)phosphates) (ABC-transporter)
StrW 63 592 (Exporter for streptomycin ATP-dependent A, BEYERet al. (1996)
6-(or 3 "-)phosphates) (ABC-transporter)
StrK 46 449 (462) Streptomycin 6-(or 3 "-)phosphate Extracellular MANSOURI and PIEPERSBERG(1991)
phosphatase protein
StrA 33 307 Streptomycin 6-phosphotransferase ATP-dependent DISTLERet al. (1987a)
(APhD)
StrR 38 (46) 350 (424) DNA-binding protein, activator A, DISTLERet al. (1987b);
of gene expression RETZLAFFand DISTLER(1995)
AphE 29 272 streptomycin 3 " -phosphotransferase ATP-dependent HEINZELet al. (1988)
a Cf. Fig. 13
Genetic data from Streptomyces griseus N2-3-11 andor S. gluucescem GLA.0 (from the latter strain given in brackets if different); inc.: incompl
sequence
Cf. Figs. 14-17
PLP: pyridoxalphosphate
A J. AmERT, S. BEYER,J. DISTLER,K. MANSOURI, G. MAYER,and W. PIEPERSBERG, unpublished data
4 Fig. 13. Gene clusters for the production of strepto- doreductases which form the keto groups pre-
mycins (SM) and the related Ca aminoglycosides ceeding the two transamination steps (see Fig.
bluensomycin (BM) and spectinomycin (SP). The 14). The StrI protein significantly resembles
streptomycete producers investigated most inten- the myo-inositol-2-dehydrogenase enzyme
sively are S. griseus (Sgr) strains N2-3-11 and
DSM40236, S. glaucescens (Sgl) GLA.0 (ETH from Bacillus subtilis (EUJITAet al., 1992)
22794), S. bluensis (Sbl) DSM40564, and S. flavo- and is the enzyme which catalyzes the first cy-
persicus (Sfl) NRRL 2820. Restriction maps for a clitol dehydrogenation step (J. AHLERT,W.
few enzymes are given for orientation; the clusters PIEPERSBERG; unpublished data).
are aligned according to their homologous amidi- The StsC protein is the scyllo-inosose ami-
notransferase [strB( l)] genes. notransferase and catalyzes the transfer of the
a-amino group of glutamine to scyllo-inosose
yielding scyllo-inosamine and a-ketoglutar-
et al., 1993; see Fig. 14). The StrI and StsB amate, an unusual transamination reaction in
proteins are clearly members of the oxidore- bacterial cells (cf. step 4 in Fig. 14; WALKER,
ductase class with an N-terminal dinucleotide 1975a; LUCHER et al., 1989; J. AHLERT,J.
coenzyme binding site (MANSOURIand PIE- DISTLER, and W. PIEPERSBERG, unpublished
PERSBERG, 1991). Therefore, they are good data). This enzyme, like the strlsts-gene prod-
candidates for the step-3 and -8 cyclitol oxi- ucts StrS and StsA, is a member of a new class
(Strl or (StsE or
(3 (StrO) StSB) stsc SWN) SWBl
aKGN
1 5
(11 (2) (3) 4 (4) (5) (6)
(StsB or (StsA or (StrN or YH

(3 SW StrS) StSE) StrB2 C-NH,
HO HO HO 0-P 0-P
8 Ho 9
(6) (I) (8) (9) lo (10) '1
Streptidlne-
&phosphate
HO
O=C-NH, O=C-NH, Bluensldins
ea &phosphate
P)
Fig. 14. The streptidine and bluensidine pathways. Enzymatic steps are numbered; where a particular gene
product is known or postulated to be involved this is given (cf. Tab. 2 and Fig. 13). Known or postulated
intermediates (numbered in brackets) are (1) ~-myo-inositol-3-phosphate, (2) myo-inositol, (3) scyllo-
inosose, (4) scyllo-inososamine, (5) scyllo-inososamine-4-phosphate, (6) N'-amidino-scyllo-inososamine-
4-phosphate, (7) N'-amidino-scyllo-inososamine,(8) 3-keto-N1-amidino-scyllo-inososamine, (9) N'-amidi-
no-streptamine, (10) N1-amidino-streptamine-6-phosphate, (7a) bluensidine; P: phosphate residues.
StrM
0-dTDP
13 14 OH OH
(12) (13) (14) (15) dTDP-Dihydro-
sireptose
Fig. 15. The dTDP-dihydrostreptose pathway. For the numbering and labeling system, see legend of Fig. 14;
intermediates are (12) D-glucose-1-phosphate,(13) dTDP-D-glucose,(14) dTDP-4-keto-6-deoxy-~-glucose,
(15) dTDP-4-keto-~-rhamnose.
of pyridoxalphosphate(PLP)-dependent ami- KANOVA,unpublished data; see below Sect.

notransferases, the so-called secondary meta- 4.1.2). In addition, a spectinomycin phospho-
bolic aminotransferases (SMAT). Several rylating activity was detected in this recombi-
other antibiotic biosynthetic aminotransfer- nant S. lividans strain (J. DISTLER,J. ALTEN-
ases (see the compilation and discussion in BUCHNER,and D. LYUTZKANOVA, unpub-
PIEPERSBERG, 1994), the protein MosB of lished data). These findings support the as-
Rhizobium meliloti involved in the metabol- sumption that sfrN encodes a phosphotrans-
ism of L-3-0-methyl-scyllo-inosamine(MUR- ferase. Two closely related genes, strBI and
PHY et al., 1993), and the PMP-dependent en- strB2, were identified which encode the amid-
zyme El catalyzing the 3-dehydroxylation inotransferase engaged in the biosynthesis of
during the formation of 3,6-dideoxyhexosesin the streptidine moiety (OHNUKIet al., 1985a,
gram-negative bacterial LPS biosynthesis b; DISTLERet al., 1987b; TOHYAMA et al.,
(THORSONet al., 1993; LIU and THORSON, 1987; MAYERet al., 1988).
1994) belong to the SMAT protein family. Both amidinotransferases of S. griseus
Therefore, it seems obvious that one of the cloned in S. lividuns were active in a nonspe-
genes, either sfsA or strS, encodes the N-am- cific assay which did not differentiate be-
indino-scyllo-inosamine L-alanine amino- tween the first and second transamidination
transferase, the second transaminase neces- steps (OHNUKIet al., 1985a, b; DISTLERet
sary for streptidine-6-phosphate formation al., 1987b; TOHYAMA et al., 1987; S. EHRIG-
(cf. Fig. 14, step 9). Alternatively, one of the FRANTZKE,and W. PIEPERSBERG,unpub-
remaining SMAT enzymes, StrS or StsA, may lished data). Since the enzymes encoded by
be involved as a biosynthetic enzyme in the the cloned genes have not as yet been tested
synthesis of the NDP-N-methyl-L-glucosam- with their postulated substrates (WALKER,
ine (NMLGA) subunit (cf. Sect. 4.1.1.3). The 1975a, see steps 6 and 11, Fig. 14) it remains
enzymes catalyzing the phosphotransfer (cf. uncertain whether the assumptions regarding
Steps 5 and 10, Fig. 14) seem to be StrN or their functions (cf. StrB1, step 6, StrB2, step
StsE proteins. Although this function has not 11, Fig. 14) are correct (PISSOWOTZKI et al.,
yet been proven by in vitro assays of the indi- 1991) or whether StrBl carries out both steps
vidually expressed enzymes there is much evi- (OHNUKIet al., 1985b). The latter is sup-
dence for this assumption. Both proteins con- ported by the puzzling finding that extracts of
tain in their C-terminal portion the character- S. bluensis seemed to contain both enzymatic
istic signature motifs which are typical for ami- activities although only StrBl is used in the
noglycoside phosphotransferases and euka- bluesidine pathway (WALKER, 1990) and
ryotic protein kinases (cf. Sect. 4.5; PIEPERS- only the strB1 gene could be detected by hy-
BERG et al., 1988; HEINZELet al., 1988). Re- bridization in two bluensomycin producers, S.
cently, a gene homologous to strN was found bluensis DSM 40564 and S. hygroscopicus ssp.
on a DNA fragment from S. flavopersicus, glebosus DSM 40823 (cf. Fig. 13; G. MAYER,
which confers spectinomycin resistance to S. A. MEHLING,and W. PIEPERSBERG, unpub-
lividans (J. ALTENBUCHNER and D. LYUTZ- lished data).
4 Genetics and Biochemistry of the Biosynthesis and Functions of Arninoglycosides 423
4.1.1.2 The L-Dihydrostreptose tylation of one of the intermediates (if UDP-

N-acetyl-hexosamines are formed), and at
Pathway least three epimerization steps, one of which
could be divided into separate oxidation and
GRISEBACH (1978) postulated a route for reduction steps at C-4 of the hexoseamine,
dTDP-L-dihydrostreptose similar to that lead- followed by N-methylation (RINEHARTand
ing to activated L-rhamnose in gram-negative STROSHANE, 1976; GRISEBACH,1978; OKU-
bacteria: activation of D-glucose in form of DA and ITO, 1982; HIROSE-KUMAGAI et al.,
dTDP-D-glucose (step 12, Fig. 15), dehydrata- 1982; KUMADAet al., 1986). The direct pre-
tion to dTDP-4-keto-6-deoxyglucose (step cursor still remains unknown but could per-
13), epimerization to dTDP-4-keto-~-rham- haps be D-glucose-1-phosphate, D-glucosam-
nose (step 14), and reduction coupled to a he-1-phosphate, N-acetylm-glucosamine-l-
rearrangement of the hexose carbon chain phosphate, or N-methyl-D-glucosamine (S.
yielding dTDP-L-dihydrostreptose (step 15). BEYER and W. PIEPERSBERG,unpublished
The 4 enzymes catalyzing the synthesis of data). Also, the published structures, UDP-
dTDP-L-dihydrostreptose are encoded by the activated and phosphorylated hexosamines
genes strD, strE, strM, and strL (see Tab. 2 (HIROSE-KUMAGAI et al., 1982; KUMADAet
and Fig. 14; DISTLERet al., 1987b; PISSO- al., 1986), of putative intermediates of the V-
WOTZKI et al., 1991; PIEPERSBERG, 1994). methyl-L-glucosamine (NMLGA) pathway
Evidence in support of this is the high level of which were accumulated in the wild type and
homology which they share with the respec- in a mutant blocked in the NMLGA pathway
tive genes encoding the L-rhamnose biosyn- cannot be explained by the currently postu-
thesis enzymes in salmonellae, rfbA, B, C, D , lated pathway. Another unsolved problem is
and their activity in Escherichiu coli (REEVES, the formation of the N-methyl group in N-
1993; PIEPERSBERG, 1994; S. VERSECKand methyl-L-glucosamine which might be intro-
W. PIEPERSBERG, unpublished data). StrD is duced at the D-glucosamine level or later in
related to NDP-hexose synthases (pyrophos- the pathway, perhaps even after condensation
phorylases) (DISTLERet al., 1987b). Genes of the three streptomycin moieties (OKUDA
similar to all or part of the strDELM cluster and ITO, 1982, op. lit.).
are present in many other gene clusters which There is indirect evidence that the genes
encode enzymes for the production of strep- strPQX, isolated and sequenced from S.gluu-
tomycete secondary metabolites containing a cescens, strFG, and stsG, analyzed from S.gri-
6DOH sugar moiety (PIEPERSBERG,1994; sew, are involved in the N-methyl-L-glucos-
LIU and THORSON,1994; VININGand STUT- amine pathway (MANSOURIand PIEPERS-
TARD,1994). In a screening of streptomycete BERG, 1991; J. AHLERT, G. MAYER, S.
strains more than 50 of which produced 6- BAYER,and W. PIEPERSBERG,unpublished
deoxyhexose-containing secondary metabol- data). StrQ is a CDP-D-glucose pyrophospho-
ites a majority seemed to possess genes which rylase which could catalyze the activating step
hybridize to strD, E(L,M) (STOCKMANN and of the N-methyl-L-glucosamine pathway (Fig.
PIEPERSBERG, 1992). Thus, this part of the str 16, step 16; S. BAYERand W. PIEPERSBERG,
gene cluster appears to be widespread among unpublished data). The StrP protein shares a
antibiotic-producing streptomycetes and oth- higher degree of similarity with UDP-glucose
er bacterial groups (cf. Sects. 5.3 and 6). 4-epimerases than with NDP-hexose 4,6-de-
hydratases, which are distantly related, and it
could, therefore, be a C-4 dehydrogenase (or
4.1.1.3 Hexosamine Pathway epimerase; Fig. 16, steps 17,19,18a, or 20) for
NDP-hexose derivatives (PISSOWOTZKI et al.,
The synthesis of the third moiety of strep- 1991). Alternatively, StrP could be an oxido-
tomycin, (NDP-activated) N-methyl-L-glucos- reductase, introducing a keto group at C-2 of
amine (NMLGA), has been intensively stud- a CDP-activated hexose (Fig. 16, step 20).
ied. Several speculative pathways have been This could then be the substrate for an ami-
postulated, including nucleotidylation, deace- notransferase, encoded by either strS or stsA,
A CHPH Cam CH,OH (slrporu?) CHPH (strx?)
HO f&lCD? 0=QiO-mpQQICDP
OH 16 OH 17 OH 18 OH OH
(16) (17) (18) (19)
NDP-N-Methyl-
Lglucosamine
/-
( W (17a) ( W ( w (2W
Fig. 16. The NDP-N-methyl-L-glucosamine (NMLGA) pathway. For the numbering and labeling system,
see legend of Fig. 14. The exact route of formation of NMLGA is unknown and could either procede via
CDP-glucose (A) or an unknown derivative of NDP-D-glucosamine (B) as precursors (for details, see
text). Possible intermediates in (A) are (16) D-glucose-1-phosphate, (17) CDP-D-glucose,(18) CDP-4-keto-
D-glucose, (19) CDP-4-keto-~-mannose, (20) CDP-L-mannose, (21) CDP-2-keto-~-glucose, (22) CDP-L-
glucosamine; possible intermediates in (B) are (16a) NDP-D-glucosamine, (17a) NDP-N-methyl-D-glucos-
amine, (18a) NDP-N-methyl-D-mannosamine, (19a) NDP-4-keto-N-methyl-~-mannosamine, (20a) NDP-
4-keto-N-methyl-~-glucosamine.
which introduces the 2-amino group (Fig. 16, these genetic data, the following routes for
step 21). The strFG genes were mapped in a the synthesis of N-methyl-L-glucosamine can
region which complemented a mutant be proposed among several other possibili-
blocked in the N-methyl-L-glucosamine path- ties:
way (KUMADAet al., i986): Protein compari-
sons suggest that both StrF and StrG could be (1) CDP activation of D-glucose-1-phosphate
members of the group of dinucleotide-inde- (StrQ), epimerization and oxidoreduction
pendent (non-oxidoreductase type) sugar iso- followed by a transamination and N-me-
merases (or epimerases) and might even form thylation yielding CDP-L-glucosamine
a heterodimeric enzyme (MANSOURI and (Fig. 16A);
PIEPERSBERG,1991). A candidate for a 35- (2) activation and modification of D-glucos-
epimerase (Fig. 16, steps 18 and 19a) in the amine (or a derivative thereof) which
N-methyl-L-glucosamine pathway is StrX be- would account for the earlier observa-
cause of its significant similarity -to other 3 5 - tions that D-glucosamine is preferentially
epimerases such as StrM (see Sect. 4.1.1.2; S. incorporated into N-methyl-L-glucos-
BEYER and W. PIEPERSBERG,unpublished amine (Fig. 16B) (OKUDAand ITO, 1982;
data). The StsG protein has three conserved KUMADAet al., 1986).
motifs which are generally found in methyl-
transferases (KAGAN and CLARKE, 1994). However, the sequence of enzymic steps or
Therefore, stsG could encode the N-methyl- the still unknown reactions remain even more
transferase necessary for streptomycin forma- speculative. Corresponding to the cyclitol
tion (Fig. 16, steps 22 and 16a). Based on transaminase reactions of the cyclitol an addi-
tional phosphotransferase could also be in- (MAIERand GRISEBACH, 1979) that this oxid-
volved in the N-methyl-L-glucosamine path- ase is in the particulate (membrane) fraction
way forming the identified, additionally phos- and does not require the addition of an elec-
phorylated NDP-hexosamine (HIROSE-Ku- tron acceptor such as NAD(P)+. Neverthe-
MAGAI et al., 1982; KUMADA et al., 1986). less, these phenomena could be explained by
a hypothetical oxidase/exporter complex for-
mation between the StrU protein and the cy-
4.1.1.4 Condensation of Subunits, toplasmic domain of the StrU/W transmem-
brane complex and a strongly bound dinu-
Processing, and Export cleotide coenzyme in StrU. This would also
explain the coupling of oxidation and trans-
The condensation of the activated precur- port steps which have been interpreted as be-
sors requires two glycosyltransferase steps ing a functional unit (MAIER and GRISE-
which are catalyzed by two enzymes localized BACH,1979). An attractive speculation could
in the soluble cytoplasmic fraction resulting in be that the oxidation is coupled to both the
dihydro-streptomycin-6-phosphate,which is ATP-driven export of a phosphorylated ami-
the last soluble intermediate detected inside noglycoside and a membrane-bound electron
producing cells (KNIEP and GRISEBACH, transport.
1976, 1980). The dihydrostreptosyl transfer- The final dephosphorylation to release the
ase was partially purified and found to be a biologically active antibiotic is catalyzed by
dimeric enzyme (KNIEP and GRISEBACH, StrK, a streptomycin-6-phosphate specific ex-
1980) with a likely subunit molecular weight tracellular phosphatase (step 27, Fig. 17;
of 35kDa. Evidence that the strH gene could WALKER,1975a; MANSOURIand PIEPERS-
be one of the two genes needed for glycosyl- BERG,1991). The gene products of the strK
transfers is weak (OHNUKIet al., 1985a, b; genes of S. griseus and S. glaucescens (orfZ of
MANSOURIand PIEPERSBERG,1991). The VOGTLI and HOTTER, 1987) are highly ho-
condensation product dihydro-streptomycin- mologous to the alkaline phosphatase (PhoA)
6-phosphate is converted, probably coupled of E. coli (MANSOURIand PIEPERSBERG,
with the active transport, to streptomycin-6- 1991). The properties of the StrK phosphat-
phosphate by a membrane-associated dehy- ase of S. griseus when expressed in S. lividans
drogenase (step 26, Fig. 17; MAIERand GRI- are similar to those reported earlier for the
SEBACH. 1979). The proteins encoded by streptomycin phosphate phosphatase (WAL-
strVW and strU recently detected in both S. KER and SKORVAGA, 1973; WALKER,1975a;
glaucescens and S. griseus (PIEPERSBERG, MANSOURIand PIEPERSBERG,1991). Each
1994; BEYERet al., 1996) are able to form this of the five steps (two glycosyltransfers, dehy-
membrane-bound transport/dehydrogenase drogenation, phosphatase reaction, and trans-
complex. The StrU protein has significant port) should have an equivalent in 5 '-hy-
similarity to the alcoholic hydroxyl-group-ox- droxy-streptomycin producers and, except for
idizing dehydrogenases, and the strV(W) gene the dehydrogenation, also in the dihydro-
product(s) represent a new member of the streptomycin and bluensomycin producers
family of the so-called ABC transporters (cf. Fig. Al). The 5 '-hydroxylation reaction
(BEYER et al., 1996) suggesting that they forming 5 '-hydroxy derivatives of streptomy-
could be engaged in the oxidation and export cins is still obscure. It could take place by a
of dihydro-streptomycin-6-phosphate.How- hydroxylase reaction during the release at the
ever, StrU clearly is a member of the dinu- cytoplasmic membrane or at the stage of the
cleotide coenzyme-dependent dehydrogenase dTDP-hexose. However, it seems unlikely
family and does not contain any transmem- that the original 6-hydroxy group of D-glu-
brane domains nor membrane-association cose remains in the intermediates since the
sites. Therefore, the suggested involvement of following steps would then require enzymes
StrU in the oxidation of dihydrostreptomy- with altered substrate specificity or even al-
cin-6-phosphate does not easily correlate to tered reaction mechanisms. D-Mannosylation
the earlier findings of H. GRISEBACH'S group at position 4 of the N-methyl-L-glucosamine
in out
Streptidinr- (7) C(NH)-NH, y(NH)-NHp
6-phosphate HN HN=C-NH,
dTDP-Dihydre
streptose
26
H3C
{
@ I -0
H32: (24) C
,cj
OH OH OH 0 OH 0
+
NDP-N-Methyl-
L-glucosamine
hHCH3
y(NH)-NH,
y(NH)-NH, HN HN=C-
HN HN=C-NH?
AphD (StrA
SM-6-P 28
(or SM-T-P) 7
OH S
4 Genetics and Biochemistry of the Biosynthesis and Functions of Arninoglycosides 427
moiety occurs in some streptomycin-produc- spective clusters of S. griseus and S. gfauces-

ing biovars of S. griseus. This appears to be a cens (cf. Fig. 14). The strB2-related gene is del-
nonspecific peripheral reaction rather than an eted and its truncated protein product (80 aa)
integral step in the biosynthetic pathway, is certainly nonfunctional. The deletion was
since in the same strains a mannosidohydrol- probably accompanied or preceded by an in-
ase is formed under carbon catabolite deple- sertion of an IS112-like IS element part of
tion (INAMINEand DEMAIN,1975). Later, which has been retained. Also, the similarity
also a dimannosylated product was detected pattern among the reading frames is puzzling:
in S. griseus strains (IKEDAet al., 1985a), in the product of the strB2-related gene shows
addition to N-methyl-L-glucosamine amino 83.8% and 92.5%, the product of the strR-re-
group modifications as in the ashimycins pro- lated gene 45.9% and 47.5%, and the product
duced by S. griseus FT3-4 (TOHMA et al., of the strN-related gene 29.5% and 29.4%
1989). Also, pseudodisaccharidic end prod- identity to the StrB1, StrR, and StrN proteins,
ucts of the streptomycin family missing the N- respectively, of S. griseus and S. gfaucescens.
methyl-L-glucosamine moiety have antibiotic It is not yet known whether there are other
activity and occur as natural products in fer- production genes for spectinomycin adjacent
mentations of the Streptomyces sp. strain to this cluster. However, this finding strongly
AC4437 (AWATAet al., 1986). supports the hypothesis that an ancestral pro-
duction gene cluster for streptomycin-like
aminoglycosides could have developed into
4.1.2 Streptomycin-Related Ca another variant by divergent evolution after
degeneration, modification, and, later on, fre-
Aminogl ycosides quent recombination with an intact strlsts
gene cluster thereby creating a new pathway
Spectinomycins (Actinospectacin). The which yields simpler but still effective end
spectinomycin molecule (see Fig. A2) is com- products. More recently, a new spectinomycin
posed of two double condensed hexose deri- derivative, spenolimycin (Fig. A3), has been
vatives, a myo-inositol and a 6DOH-deriva- described (KARWOWSKIet al., 1984). Its
tive (actinospectose, a 4,6-dideoxyhexose; cf. 6DOH moiety is altered in that the 3' posi-
Fig. 8), and therefore is a Ca(4,5)-6DOH tion contains an oxymethyl group and the
compound as is the dihydro-streptosylstrep- C-3 '14'-bond is unsaturated, reminiscent of
tidine (AC4437) produced by some strepto- modifications which occur in other 6DOH re-
mycetes (AWATAet al., 1986). The incorpora- sidues, e.g., in anthracyclines (see Figs. 8 and
tion of radioactively and stable isotope-la- 9 and Sect. 5.3).
beled glucose into both moieties clearly sup-
ports this interpretation (OTSUKA et al., Kasugamycins. Not much work has been
1980). However, no further biochemical stud- done to elucidate the biochemical pathway
ies on the pathway have been reported, ex- for the production of kasugamycin (Fig. A4)
cept for the recent demonstration in a specti- and related Ca(4)-6DOH compounds (minos-
nomycin producer of an L-g1utamine:scyffo- aminomycin; Fig. A5). However, it is reasona-
inosose aminotransferase similar to that ble to postulate basically streptomycin-like
found in streptomycin producers (cf. Sect. pathways also for these aminoglycosides. The
4.1.1.1; WALKER,1995). Recently, a spectino- cyclitols are again clearly derived from myo-
mycin-resistance-conferring DNA segment inositol. However, the postulate by UMEZA-
(3.65 kb) was cloned from the spectinomycin WA et al. (1986) that the 6DOH moiety kasu-
producer S. flawpersicus NRRL 2820 (J. AL- gamine could be derived from UDP-N-acetyl-
TENBUCHNER and D. LYUTZKANOVA, per- D-glucosamine seems unlikely in view of what
sonal communication). The DNA sequence has been learned from the streptomycin path-
suggested that it was derived from a rear- way and from other 6DOH or hexosamine
ranged and degenerated streptomycin gene pathways in the recent past. Rather, it seems
cluster since it showed striking similarity with likely that a dTDP- or a CDP-hexose biosyn-
three genes, strBl, strR, and strN in the re- thetic route is used, since the final product is
a 2,3,4,6-tetradeoxy-2,4-diaminohexose
deri- Hep). Its clear structural relationships with
vative (cf. Sects. 3.2.1 and 5.3). both myomycin (in the cyclitol(4)-3-amino-
hexose pseudodisaccharidic unit; see above)
Myomycin. Myomycin, besides being a possi- and seldomycins (in the cyclitol(6)-2-amino-
ble Ca(4)-HA (Fig. A8) compound modified pentose unit; cf. Sect. 4.3) gives it a clear
by a varying number of plysyl residues, is an- bridging role between the Ca and the '
other streptomycin-related product in that it (2DOS-)Cb aminoglycosides, and it will be

contains some distant structural similarities to interesting to see whether this is reflected by
bluensomycin, such as carbamoyl groups in similarities at the genetidenzymic level. The
the cyclitol moiety in addition to a guanidino incorporation of a D-mannoheptose makes it
group (in the hexosamine moiety) and appar- unique among the cyclitol-containing amino-
ently a mode of action and binding site which glycosides and could be another indication
are identical to those of the streptomycins for the hypothesis that the production of cell
(DAVIESet al., 1988). Since these aminogly- wall components of gram-negative bacteria
cosides are produced by nocardioforms (or (e.g., lipopolysaccharide) and many antibiot-
coryneforms; FRENCHet al., 1973) it will be ics in streptomycetes may be based on the
interesting to see whether genes related to same gene pool (PIEPERSBERG, 1992,1993).
known strlsts genes in the streptomycin or
Ca Aminoglycosides with Monoaminocycli-
bluensomycin producers are also present in tols. Three monoaminocyclitol aminoglyco-
the lower actinomycetes. Examples of such
sidic antibiotics with no importance in phar-
genes which could be found in the myomycin
maceutical application but with interesting
producer are those encoding enzymes of the
structures have been reported which should
following families (see Tab. 2): amidinotrans-
be mentioned here.
ferases (e.g., StrB1, StrB2), carbamoyltrans- (1) Minosaminomycin, a kasugamycin-re-
ferases (bluensomycin pathway), and those
lated Ca(4)-6DOH product of Streptomyces
related to some of the enzymes involved in
sp., has already been mentioned above. Its
the synthesis of the cyclitol (e.g., StrO, StrN) unique structural feature is the presence of a
and in other parts of the pathway, e.g., activa-
tion, synthesis, and transfer of the 3-amino-
1-D-1-amino-1-deoxy-myo-inositol to which a
histidinyl-valine dipeptide is bound via an
hexose precursor (e.g., StrQ, StrS, or StsA),
amide bond (Fig. A5). The introduction of
or involved in export and resistance phenom-
the amino group into the cyclitol could princi-
ena (e.g., StrV, StrW, StrA). Genetic relation-
pally follow the same route as in the streptid-
ships to producers of antibiotics containing /3-
ine pathway, although with altered stereosel-
lysine tails (e.g., streptothricin; cf. Fig. A27;
or viomycin) may also become apparent. The ectivity in the steps catalyzed by the first-step
3-guanidinomannose moiety in myomycin dehydrogenase and transaminase.
(2) Hygromycin A (Fig. A9) was detected
could be derived from an unusual NDP-glu-
prior to hygromycin B in the same strain of
cose or NDP-mannose pathway similar to the
Streptomyces hygroscopicus in the early 1950s
NMLGA unit in streptomycin (cf. Sect.
4.1.1.3). and was later also found as a product of other
Streptomyces sp. and of Corynebacterium
equi. It is also a Ca(l)-[X]-6DOH compound
Boholmycin. The last, structurally new class
and is interesting in several respects:
of aminoglycosides to be described is repre-
sented by the pseudotetrasaccharidic com- - it has an aminocyclitol unit, a 2-amino-neo-
pound boholmycin (Fig. A7; SAITOHet al., inosamine, the derivation of which is totally
1988) and is produced by a strain of Strepto- obscure, but suggesting, however, that both
myces hygroscopicus. It clearly belongs to the a Ca and a Cb (for the biosynthesis of the
Ca aminoglycosides according to our defini- 2DOS moiety in hygromycin B) pathway is
tions. It is composed of a dicarbamoyl scyllo- functional in this strain, and that these two
inositol, two amino sugars condensed via gly- pathways can only coexist because of their
cosidic bonds to the 4- and 6-positions of the completely different stereoselectivity of ox-
cyclitol, and a heptose (Ha-(4)Ca(6)-PA- idationltransarnination steps;
- two of the cis-hydroxyl groups are bridged lyspora hirsuta ATCC 20501 (Fig. 18 and Tab. 3;
by a methylene residue in the cyclitol moie- ITOH et al., 1984; ODAKURAet al., 1984;
ty; DAIRI and HASEGAWA,1989; HASEGAWA,
- the furanosidic 5-ketod-deoxysugar could 1991, 1992; DAIRIet al., 1992a, b, c; OHTAet
be another rare 6DOH unit derived from a al., 1992a, b, 1993a, b; OHTA and HASEGA-
dTDP-glucose (cf. Sects. 3.2.1 and 5.3). WA, 1993a, b; HOTTA et al., 1995). Of the
roughly 20 steps of FTM-A biosynthesis, 14
(3) The aminoglycosides of the LL-BM123 have been identified by various methods, in-
series (Fig. A6), produced by Nocardia sp., cluding induction and analysis of blocked mu-
are again compounds of biosynthetically tants, gene cloning, and feeding of interme-
mixed origin since they also contain amino diates. Most of the production genes (fms)
acid residues such as minosaminomycin. seem to be clustered on a DNA fragment of
Their unique pathway formula is Ca(4)-H(4)- ca. 30 kb or more in M. olivasterospora. The
HA, where the disaccharide D-glucosaminyl- order of identified genes in M. olivasterospo-
(pl,4)-~-rnannoseis glycosidically linked to ra ATCC 21819 is fmsl0, 13, 3, 4, 5, 12, 8, 7,
the 4 position of the 2-amino-2-deoxy-myo- 14,1,11, (orf2), fmr0, (orf4). This DNA seg-
inositol moiety. The latter aminocyclitol ment as a whole only hybridizes with the ge-
should be derived from myo-inositol, which nomic DNA from Micromonospora sp. SF-
again can only be formed via an aminotrans- 2089 and D. matsuzakiense (DAIRI et al.,
ferase with different substrate selectivity rela- 1992b). However, the restriction pattern of
tive to the StsC protein involved in the bio- the hybridizing bands were almost identical
synthesis of streptidine (cf. Sect. 4.1.1.1). only in the strain Micromonospora sp. SF-
2089. In the DNA of the other three produc-
ers investigated no hybridization with the
30 kb fragment was observed, but when indi-
4.2 Fortimicins, Istamycins vidual genes for conserved functions were
taken as probes, e.g., the fmsl3 (smsl3; en-
This group of very similar Ca(6)-HA or coding the N-glycyltransferase) genes, signifi-
Cb(6)-HA compounds (Fig. A10) is wide- cant hybridization was seen with the DNA
spread among filamentous actinomycete gen- from all six producers (OHTA et al., 1992b).
era (cf. Fig. l), namely Micromonospora spp. Thus, it seems likely that all producers of
(fortimicins: FTM; SF-2052), Dactylosporan- FTM-like aminoglycosides contain highly re-
gium spp. (dactimicins), Streptomyces spp. (is- lated gene clusters originating from a com-
tamycins: ISM; sannamycins), and Saccharo- mon evolutionary source with some minor
polyspora spp. (sporaricins). In fact, the only modifications, such as the use of a different
major biosynthetic difference between the pathway for the formation of the (2-deoxy)-
two groups containing either fortamine scyllo-inosose precursor.
(FTM, dactimicins) or 2-deoxyfortamine The difference between the two more dis-
(ISM, sporaricins, and sannamycins) as ami- tant groups (from the hybridization data; see
nocyclitols seems to be the formation of the above) is also reflected by the aquisition of
cyclitol via myo-inositol phosphate or scyllo- two types of resistance genes: (l), f m r 0
inosose, respectively (cf. Sect. 3.1). The FTM (fmrM, fmrD, from M.olivasterospora ATCC
pathway has been most extensively studied in 21819, Micromonospora sp. SF-2098, and
Micromonospora olivasterospora ATCC Dactylosporangium matsuzakiense ATCC
21819 (FTM-A producer). It has been found 31570, respectively); (2) fmrT f i r s , fmrH,
to be almost congruent in substrate specificity from Streptomyces tenjimariensis ATCC
with that of the producers of SF-2051 Micro- 31603, S. sannanensis I F 0 14239, and Saccha-
monospora sp. SF-2089 ATCC 31580, dacti- ropolyspora hirsuta ATCC 20501, respective-
micin Dactylosporangium matsuzakiense ly). Both encode members of the 16s rRNA
ATCC 31570, sannamycin Streptomyces sun- methyltransferases but these enzymes methyl-
nanensis I F 0 14239, istamycin s. tenjimarien- ate different residues (G-1405 and A-1408;
sis ATCC 31603, and sporaricin Saccharopo- CUNDLIFFE,1989; HASEGAWA,1991; OHTA
myeinositd
FTM-A0
1
4.5
(OR.TA. MT)
(OR.TA)
-
1'2
scybinosamine
+
D-glucosamine (GT) OH
HjCHN
3 FTM-FU-10 OH
I 6
(OR.TA)
8.9 FTM-KL1
y 3
(3'-PT,
DHslORs?)
H3CHN
FTM-AP y 3
H3CHN
H3C
10 12 \
(MT)
fH3
OH
FTM-B
13
FTM-KR
FTM-KH H3CHN (GLY)
fH3 v
14
T o 4-
y 3 (FIT)
CHNh H3C
\ OH
HCH=NH FTM-A 0%
OH
dactimicin
0'3%
Fig. 18. The fortimicin (FTM,astromicin) pathway. The same pathway starting from 2-deoxy-scyllo-inos-
amine seems to be established in istamycin/sannarnycin/sporaricin producers. The known intermediates
and postulated enzymatic steps are given; for further details, see HASEGAWA(1992) and HO'ITA et al.
(1995).
and HASEGAWA, 1993a, b; OHTAet al. 1993a, groups (OHTAet al., 1993a). Thus, the set of
b; see Sect. 4.5). In each case the single resist- genes used in the dactimicin producer could
ance gene seems to reside in the production be a mixture of the two extreme evolutionary
gene cluster. However, they seem to be differ- lines found in the other producers of the
ently organized in the gene clusters in both FTMlISM group aminoglycosides: (1) it con-
Tab. 3. Gene Products and Enzymes Known or Presumed to be Involved in the Production and Resistance of Forti
Gene Coding Enzymatic Function or Step‘ Remarksd Organism’

Producta Capacity (Preliminary Assignment)
of Gene aab
Fmsl (D-myo-Inositol 2-dehydrogenase) NAD(P)-dependent? Mol, Msp, (Dma)

Fms2 (scyllo-Inosose aminotransferase) PLP-dependent? Mol, Msp, (Dma)
Fms3 FTM-FU-10 synthesis (D-G~UCOS- Mol, Msp, (Dma)
aminyltransferase)
Fms4 FTM-A0 synthesis Mol, Msp, (Dma)
Fms5 FTM-A0 synthesis Mol, Msp, (Dma)
Fms7 FTM-KKl synthesis Mol, Msp, (Dma)
Fms8 FTM-AP synthesis (FTM-KK1 ATP-dependent, Mol, Msp, (Dma)
phosphotransferase) homologous to
APH(3’)-II
FmslO FTM-KH synthesis Mol, Msp, (Dma)
Fmsll FTM-KR synthesis (FTM-KH Mol, Msp, (Dma)
epimerase)
Fmsl3 (Smsl3) FTM-B N-glycyltransferase Mol, Msp, Dma,
San, Shi
Fmsl4 N-Formimidoyl FTM-A synthase FAD, 4-mer Mol, Msp, (Dma)
(oxidase)
FmrO 16s rRNA (G-1405) methyltrans- SAM-dependent Mol; Msp, Dma
(FmrM, FmrD) ferase
FmrT 16s RRNA (A-1408) methyl- SAM-dependent Ste, (San)
transferase
KamC 16s rRNA (A-1408) methyltrans- SAM-dependent Shi
ferase
Unknown Unknown (upstream f m r 0 ) (FTM-A synthesis?) Mol
ORF (ORF-2)
Unknown Unknown (downstream f m r 0 ) (FTM-A synthesis?) Mol
ORF (ORF-4)
Unknown (Epoxide hydrolase) (Sannamycin synthe- Ste
ORF (ORF-1) sis?)
Unknown Unknown (downstream f m r q (Sarmamycin synthe- Ste
ORF (ORF-3) sis?)
a Cf. Fig. 18 Mol: Micromonospora olivasterospora; M

In brackets: partial sequence data tylosporangium matsuzakiense; San: Strepto
Cf. Fig. 18 mariensis: Shi: SaCCharODOhSDOra hirsuta
P L P pyridoxalphosphate; SAM: S-adenosyl methionine A HASEGAWA (199i) and DAIRIet al. (
tains probably a Cb pathway as the ISM (san- Fig. 10); (2) a 4'5'-dehydratase reaction as is
namycin) type producers; but (2) it has the re- suggested by the occurrence of 4'3'-dehy-
sistance gene and profile as well as the strong- dro-FTM-A; (3) a reductase/dehydrogenase)
er DNA sequence similarity to the FTM(SF- step reducing the 4',5 ' double bond.
2051) type producers.
The last two steps in the formation of
FTM-A (glycyltransfer) and FTM-C (N-for- 4.3 2-Deoxystreptamine-
mimidoylation of the glycyl amino group; cf. Containing Aminoglycosides
Fig. 18), catalyzed by the gene products
Fmsl3 and Fmsl4, respectively, in M. olivu- The large and clinically important group of
sterosporu (Tab. 3) represent the biochemi- 2-deoxystreptamine(2DOS)-containing ami-
cally best investigated phase of the biosyn- noglycosides was extensively studied with re-
thetic pathway for FTM-like aminoglycosides. gard to its biogenesis in wild type and mutant
The genes for these two steps have been strains using 14C-, I3C-, 3H-, and "N-labeled
cloned, analyzed in part, and found to be precursors, such as D-glucose, D-glucosamine,
present in all producers of FIM/ISM type ami- and 2DOS. The resulting data obtained main-
noglycosides either by hybridization or by ac- ly with the producers of neomycin, paromo-
tivity (DAIRI et al., 1992c; OHTA et al., mycin, ribostamycin, butirosin, and the gen-
1992b). In a blocked mutant of the ISM pro- tamicidsagamicin group antibiotics have
ducer S. tenjimuriensis FTM-B was converted been reviewed extensively (RINEHARTand
into 1-epi-FIM-B, dactimicin, and l-epi-dac- STROSHANE,1976; PEARCEand RINEHART,
timicin; M. olivasterosporu in turn converted 1981; KAKINUMA, 1982; KASE et al., 1982;
ISM-A. and ISM-Bo into ISM-A3 and ISM- OKUDAand ITOH, 1982; UMEZAWAet al.,
B3, respectively (cf. Fig. A1 0 HOTTAet al., 1986; GRAFE, 1992). However, since about
1989; DAIRI and HASEGAWA,1989). The 1985 only very few new findings have been
mechanism of the glycyl transfer and the pu- published. Therefore, only a brief summary of
tative activation of the glycyl residue (e.g., what is currently known is given here.
aminoacyl-AMP) has not yet been studied. The 2DOS moiety which is the basic build-
The N-formimidoyl group was shown to be ing block in this family of compounds is made
derived from glycine, the C-2 group of which directly from glucose-6-phosphate via the Cb
is converted via an unusual oxidase mecha- route (see Sect. 3.1; Fig. 19, cf. Fig. 4) as has
nism to the formimidoyl group and probably been clearly demonstrated recently in the
COa in the presence of molecular oxygen neomycin producer S. fradiae (YAMAUCHI
only; this is catalyzed by the FAD-containing and KAKINUMA,1992b, c; 1993; 1995) and
Fmsl4 enzyme (DAIRIet al., 1992~).Another earlier postulated (KAKINUMA, 1982). Pre-
interesting gene product is the Fms8 phos- viously, the biosynthesis of the cyclitol moiety
photransferase, which probably catalyzes the was obscure and was believed to occur either
3 '-OH phosphorylation of the purpuros- via myo-inositol or directly from glucose by
amine moiety in the FTM-KK1 intermediate. an unknown mechanism (RINEHARTand
This enzyme is homologous to the APH(3') STROSHANE, 1976). However, it was already
enzymes encoded by the nmrA and aph genes known from the labeling pattern of the C1
of neomycin-producing Micromonosporu sp. and C, positions of glucose and from the in-
and Streptomyces fradiae, respectively, and corporation of the first amino group, that 2-
can be replaced by the latter gene products deoxy-scyllo-inosamineis not derived from D-
(DAIRI et al., 1992a). However, its involve- glucosamine and that the direction of the sec-
ment in the interesting 3 ' ,4 '-dehydroxylation ond-step transamination is opposite to that of
is unclear (cf. Fig. 18, steps 8, 9). For this the streptidine and actinamine pathways
phase of the pathway probably several steps (PEARCEand RINEHART,1981; UMEZAWA
are required: (1) dehydratation at C-3 'could et al., 1986). The first transamination step
occur via a mechanism similar to that operat- could be catalyzed by an aminotransferase
ing in the 3,6-dideoxyhexose pathway in very similar to the S. griseus StsC enzyme (see
gram-negative bacteria (cf. Sect. 3.2.1 and Sect. 4.1.1.1) since it was shown by WALKER
OH OH
OH
D-Xyb-,
~
Dglucosarnine
INDP. -P, or -H)
HO
boQ7 (R = -NDP. -P, or -
NH2 OH
I
paromamine
/
(5-0-rbosyl)
pyranosyl ?) genlarnicin
NH2 b
HO (y b~ gentamicin
pseudotrisaccharide
sagamicin,
sisomicin,
? verdamicin
G418
et al. (CHENand WALKER,1977; LUCHERet tachment of a furanosidic pentose, xylose

al., 1989; WALKER,1995) that ketocyclitols (forming xylostasins), or ribose (forming ri-
were transaminated by the a-amino group of bostamycins) which finally can be further gly-
glutamine in extracts of the neomycin produc- cosylated by a 2,6-diamino-2,6-dideoxyhexose
er S. fradiae and the gentamicin producer Mi- (neosamine B or C) in the neomycins and pa-
cromonospora purpurea as in streptomycin romomycins. The pseudodisaccharides are
and spectinomycin producers. Therefore, it clearly intermediates and can be converted
will be interesting to see whether stsC probes, directly to the respective end products in mu-
e.g., from S. griseus, will detect a respective tants blocked in their formation, e.g., in the
gene in the production gene clusters of biosynthetic pathway of the aminocyclitol.
2DOS-producing actinomycetes. The later Neamine (neomycin A) has measurable anti-
steps in the formation of 2DOS could also be biotic activity. The pathway of the neosam-
closely related to the respective steps in the ines B and C is as obscure as that of the first
streptidine pathway (cf. Fig. 14), such as de- hexosamine unit, though it is clear that D-glu-
hydrogenation (enzyme StsB- or StrI-like?) cosamine is preferentially incorporated into
and second-step transamination (enzyme them. The type of activation, the glycosyl-
StsA-like?). It is unknown whether 2DOS transferase(s), and the route of incorporation
precursors are phosphorylated at some stage of a second amino-N into the C, position of
or enter the further condensation and secre- the 2-aminohexose remain to be established
tion steps in a form similar to streptomycin. enzymatically. A similarity to the situation in
The presence of aminoglycoside-3’-phospho- streptomycin-producing S. griseus could exist
transferases as resistance mechanisms in some also here: UDP-(N-acetyl-)D-glucosamine
of the 2DOS producers (see below, Sect. 4.5) might not be an immediate precursor as in the
would suggest, however, that the phosphory- cell wall biosynthesis. Therefore, it will be
lation does not occur in the cyclitol moiety. necessary to screen also for a StrQ-related
All major 2DOS-containing aminoglyco- nucleotidylating enzyme among the gene
sides, except for the destomycin-hygromycin products involved in biosynthesis of neomy-
B group and perhaps also the apramycins, cin-like aminoglycosides. Important in this
seem to be formed via a common pseudodi- context are also the results of more recent
saccharidic intermediate, paromamine, which biogenesis studies with l-13C- and 6-13C-la-
is formed as an early intermediate from beled glucose, which suggest that all building
2DOS and a molecule of D-glucosamine, blocks of neomycin seem to be formed from
which is probably nucleotide-activated, via intermediates of the pentose phosphate cycle
the hypothetical pathway outlined in Fig. 19. ,(Fig. 20; RINEHARTet al., 1992). If this
Whether the precursor formation and the at- “equilibration” of all or most glucose mole-
tachment reactions of the D-glucosamine cules through the pentose phosphate cycle
moiety somehow relate to that of the generally occurs for all hexose and other car-
NMLGA unit of streptomycin (see Sect. 4.1) bohydrate components in secondary metabol-
remains to be shown. Further on, the 2DOS ites within the production phase, the earlier
pathways branch into various alternate routes results of isotope-labeling studies would have
resulting in the large variety of end products to be reinterpreted.
(see Figs. All-A17).
4,6-Disubstituted 2DOS Aminoglycosides.
4,5-Disubstituted 2DOS Aminoglycosides. The kanamycins-tobramycin (Fig. A 1 2 HA-
The compounds which are 4,5-substituted at (4)Cb(6)-HA), the gentamicins-sagamicin
the aminocyclitol, such as the neomycins, pa- (Fig. A14; HA-(4)Cb(6)-PA), and probably
romomycins, lividomycins (Fig. A1l ; basically also the seldomycins (Fig. A17; HA-(4)Cb(6)-
Ha-(4)Cb(S)-P(3)-HA) or ribostamycins, and PA) groups of 2DOS-containing aminoglyco-
butirosins (Fig. A13; basically Ha-(4)Cb(5)- sides are also synthesized from paromamine.
P), are probably synthesized directly from pa- However, their further substitution at the C6
romamine or via a second pseudodisaccha- position of the cyclitol and the nature and
ride, neamine (Fig. 19), followed by the at- modification of the second glycosidic residue
Fig. 20. Biogenesis of the

hexose-, pentose- and cycli-
tol-derived components in
neomycin from 6-(CL3)-~-
glucose (according to RINE-
HART et al., 1992). The la- 0
beling patterns measured by OH
(C”)-NMR prove that all
units are built up preferen-
tially from C, (0),C2, or C3
units (thick lines) rearranged
by transketolase- and trans-
aldolase-catalyzed reactions HO OH
in passages through the pen- - -
tosephosphate cycle. The OH
earlier labeling patterns ob-
tained with 1- (*) or 1,6-la-
beled (UlA) D-glucoses are
also given from which a dif-
ferentiation between direct
or indirect incorporation of HO
glucose was not possible.
distinguish the gentamicins from the seldomy- or N-methylation and dehydration, resemble
cins in which a pyranoid C-5 sugar moiety is those of other aminoglycoside pathways; (3)
replaced by a hexosamine. This sugar in the the multiply branching pathway proceeds
gentamicin-related compounds is D-xylose from the first trisaccharidic intermediate GM-
(the pseudodisaccharide unit formed with the A2 to GM-X2 and then branches to yield the
2DOS cyclitol is called garamine) and could two unsaturated intermediates sisomicin (via
be either D-xylose or 2-amino-2-deoxy-~-xy- JI-20A) and verdamicin (via G-418, a com-
lose in the seldomycins. Since the gentamicins pound now frequently used for cloning vector
are only produced in Micromonospora spp., selection in plant and animal cells). In some
and the seldomycins are only found in Srrep- strains these two intermediates can already be
romyces spp. it will be of interest to study the the major end products which are released.
mutual relationships between the respective This biosynthetic phase strongly resembles
sets of biosynthetic genedenzymes relative to the 3,4-dehydroxylation steps in the fortimi-
those involved in the production of the other cin pathway (cf. Fig. 18; steps 8, 9); however,
groups of 2DOS-containing aminoglycosides. the occurrence of a 3,4-unsaturated 2,3,4,6-
The most intensive study on the design of an deoxy-2,6-aminohexose indicates that dehy-
individual 2DOS pathway was carried out on dratation (probably of a 4-hydroxylated pre-
that of the gentamicin(GM)-sisomicin-sa- cursor after 3-dehydroxylation; cf. LIU and
gamicin group (KASEet al., 1982; cf. UMEZA- THORSON, 1994) is a step in this process (see
WA et al., 1986) which is one of the most ver- also Sect. 4.2). The last phase of GM biosyn-
satile with more than 20 end products identi- thesis yielding the final products GM-C,,
fied. Some details should be mentioned for GM-C,,, GM-C2, and sagamicin, which are
comparison (for formulae see Fig. A 14): (1) a mainly used in therapeutics variably involves
minimum of 20 enzymes is probably involved reduction, epimerization, and N-methylation
in the biosynthesis of, e.g., GM-C,; (2) many steps, probably by action of the same en-
of the steps, e.g., dehydrogenation and trans- zymes on different but structurally related in-
amination in positions 3 and 6 of pyranoses, termediates.
Apramycins and Destomycins. A more distant 1-N-amidinostreptamine-6-phosphate inter-

position relative to the other 2DOS aminogly- mediate (cf. Fig. 14) could be used. The strept-
cosides is taken by apramycin (Fig. A15; amine unit in SS-56-C could be formed direct-
CB(4)-OctA(S)-HA) and destomycin-hygro- ly from scyllo-inosamine or via hydrolysis of
mycin B (Fig. A 1 6 Cb/Ca(S)-H(2,3)-HepA) the l-N-amidino group from l-N-amidino-
group. In the apramycins monosubstituted in streptamine. This also could mean that 6-
the 4-position of the 2DOS moiety parom- phosphate intermediates of destomycins are
amine could be an intermediate. The further formed inside the cells as in the case of the
pathway would then require an elongation at streptomycin producers.
the 6'-hydroxymethyl group of the hexos- In summary, compared with the streptomy-
amine moiety by a C-2 unit which, however, cins and the fortimicins the 2DOS aminogly-
seems rather unlikely. An alternative route cosides mentioned so far seem to take an in-
would be the condensation of 2DOS with an termediate position with respect to the distri-
octose derivative, which is reminiscent of the bution of modifying steps relative to the con-
unusual octose pathway in the lincosamides densation reactions. The modifications are
(cf. Sect. 5.1; RINEHART,1980; see also the practically all finished before condensation in
discussion in OKUDAand ITO, 1982). Also, streptomycins, but in the 2DOS antibiotics
the 4-amino-4-deoxy-~-glucosemoiety glyco- much more take place after condensation by
sidically linked to the 8' position is quite unu- glycosyltransfer reactions by which only the
sual among the aminoglycosides; 4 transami- diaminocyclitol is completely preformed.
nation also occurs in the kasugamycin-related
compounds, but on a 6DOH unit (see Sect.
4.1). This unit could be synthesized by trans- 4.4 Other Aminoglycosides
amination of a NDP-4-ketoglucose interme-
diate formed by an enzyme related to the Monomeric sugar derivatives. The number
UDP-glucose 4-epimerase (cf. Sect. 3.2). In of bioactive and stable molecules directly de-
the destomycins and hygromycin B the 2DOS rived from monosaccharidic units or sugar
moiety is 5-substituted with an unusual hex- analogs described in the literature is increas-
ose, D-talose. This in turn is fused via a ing steadily (Fig. A18; cyclitol-related com-
unique type of chemical bonding, an ortho-es- pounds are treated below; cf. Sect. 5.2). Some
ter linkage between the 2'- and 3'-hydroxyls examples are mentioned here:
and the 1"-position of a sugar acid derived (1) A group of glycosidase inhibitors, such
from a 6-amino-6-deoxyheptose, destomic as the nojirimycins (inhibit glucosidases and
acid. These structural details suggest that mannosidases), galactostatin (inhibits galacto-
there is very little resemblance between the sidases), and siastatin (inhibits sialidases), are
pathways of hygromycin B and destomycin sugar analogs with a substituted pyridine ring
production and those of other aminoglyco- and can be regarded as a group of bacterial
sides, except for the biosynthesis of the ami- alkaloids derived from aminohexoses or ami-
nocyclitol. Interestingly, strains of Strepto- nopentoses (GRAFE, 1992). The sugar-like 1-
myces eurocidicus and of Saccharopolyspora deoxynojirimycin (DNJ) and a similar struc-
hirsuta produce the destomycin derivatives, tured plant alkaloid (castanospermine; not
SS-56-C and l-N-amidino-l-N-demethyl-2- shown) can act as anti-HIV drugs by prevent-
hydroxydestomycin A (INOUYEet al., 1973; ing the maturation of the gp120 envelope gly-
IKEDAet al., 1985b; cf. Fig. A16), respective- coprotein. This finding motivated a recent
ly, which are clearly Ca compounds and are study of the biogenesis of these unusual ami-
related to streptidine in their diaminocyclitol no sugars by stable isotope labeling in Strep-
moieties. Here, a DNA recombination event tomyces subrutilus (HARDICKet al., 1992). It
resulting in the fusion of two gene clusters, was found that DNJ is derived from a glucose
those for streptomycin and destomycin pro- molecule converted first via fructose, 6-oxida-
duction, could have created a new mixed tion and reductive 2- or 6-transamination
pathway. From the streptomycin gene cluster steps to mannonojirimycin (Fig. 21). This in-
those genes needed for the production of the termediate can then be dehydrated and re-
also been prepared, two of which, miglitol

and emiglitate (Fig. AM), have reached the
HO +OH 'YOH
phase of clinical development (MULLER,
1989).
tOH
kHpOH
toH
%H*OH
(2) The recently described D-glucosamine
derivative CV-1 (Fig. A18), which is pro-
oxidation/ 1 transamination duced by a Streptomyces spp. and is a weak
[ '(1 ]
transamination / 1 oxidation
antibiotic by itself, has a very interesting co-
operative effect together with spiramycin on
OCHpOH OCHpOH gram-negative bacteria (ICHIMURAet al.,
1987). The inhibitory effect of CV-1 was de-
monstrated to be on LPS synthesis in E. coli
OH Or " I O H OH thereby relieving the barrier effect of the out-
er membrane for spiramycin which by itself is
%H~-NH, A - 0 ineffective on gram-negatives. The biosynthe-
sis of CV-1 involves N-carbamoylation of D-
/
cyclisation glucosamine, probably from L-citrulline,
which conceivably occurs either on a l-phos-
phate- or nucleotide-activated precursor or
the free sugar. Subsequent reorganization of
N-carbamoyl-D-glucosamine to the unique
II
open ring hemiaminal was shown to proceed
epimetisation
spontaneously (YASUZAWA et al., 1987).
(3) Valiolamine (cf. Fig. A21) is a new
CHIOH 1. dehydration DCH,OH monomeric aminocyclitol member of the vali-
- H 2 . r e y *
damycin group of aminoglycosides (KAMEDA
OH OH OH et al., 1984; see below) and has inhibitory ac-
NJ DNJ tivity against a-glucosidases.
Fig. 21. Proposed biogenesis of nojirimycin and re-
Trehalosamines and Other Aminodisaccha-
lated monosaccharide analogs. rides. A larger group of nitrogen-containing
(D)NJ: (1-deoxy-)nojirimycin; (D)MJ: (1-deoxy-)-
and carbohydrate-related actinomycete prod-
mannonojirimycin. The labeling patterns obtained
from 1,6-labeled D-glucose was adopted from ucts are compounds with structural analogy to
RINECHARTet al. (1992) (cf. Fig. 20). the disaccharides trehalose or saccharose.
Most of these have some biological activity
either as antibiotics or as glycosidase inhibi-
tors. The &,a-glycosidic trehalosamines (Fig.
duced to 1-deoxymannonojirimycin or, alter- A19; HA(1)-H) are known for a long time
natively, epimerized at C-2 to nojirimycin and and were isolated on account of their antibac-
subsequently dehydroxylated to DNJ. Thus, terial activity which, however, is only weak
several enzymedgenes related to those used (cf. UMEZAWAet al., 1986; ASANOet al.,
in the formation of other amino and/or deoxy 1989). The biosynthesis of 2-trehalosamine
sugar components in known antibiotics could and mannosyl glucosaminide from a molecule
be used. The same might hold for other each of D-glucosamine and either D-glucose
monomeric amino sugars isolated from cul- or D-mannose, respectively, by enzymes re-
tures of microorganisms such as 3-amino-3- lated to trehalose synthases can easily be en-
deoxy-D-glucose, N-carbamoyl-D-glucosamine, visaged. In contrast, the derivation of 3- and
prumycin, and streptozotocin (Fig. A18; see 4-trehalosamines which also occur as strepto-
also the compilations in UMEZAWAet al., mycete products could follow at least two dif-
1986; GRAFE, 1992). Semisynthetic deriva- ferent routes: (1) formation of an activated 3-
tives of DNJ with improved inhibitory activity or 4-aminohexose which is condensed with D-
on a-glucosidases and pharmacokinetics have glucose or (2) modification of preformed tre-
halose. It would be interesting to know chemo-enzymatic synthesis from the 3-keto

whether these compounds have (auto-)regu- forms of trehalose and sucrose, which were
latory functions in adaptive processes, such as prepared by treatment with D-glucoside 3-de-
osmoregulation, cell differentiation, or sec- hydrogenase from Flavobacterium saccharo-
ondary metabolism; trehalose metabolism philum, and by reductive chemical amination.
could play an important role in those physio- Except for S-I, these compounds are weak an-
logical phenomena in streptomycetes tibiotics; S-I is an inhibitor of invertases.
(CHAMPNESSand CHATER, 1994). A new
member of the trehalosamine family is the Validamycins, Acarbose, and Related Oligo-
qP-glycosidic antibiotic 3,3'-neotrehalosa- saccharidic Aminoglycosides. The cyclitol
diamine (Fig. A19; HA(1)-HA) isolated from moieties formally derived from valiolamine,
a Bacillus pumilus for its antibacterial activity such as valienamine or validamine (cf. Figs. 6
on an aminoglycoside-hypersensitivestrain of and 22), are most probably products formed
Klebsiella pneumoniae (NUMATAet al., 1986; via a Cb pathway and may be regarded as ali-
TSUNO et al., 1986). phatic m-C7N units, though they are not syn-
The 3-amino-3-deoxy analogs of trehalose thesized using erythrose-4-phosphate and
and sucrose and their epimers, T-I, T-I1 (3- phosphoenolpyruvate as the aromatic m-C7N-
trehalosamine), and T-I11 or S-I and S-I1 (3- units (cf. Sect. 3; FLOSSand BEALE,1989; RI-
sucrosamine) should be regarded as members NEHART et al., 1992). Thus, the suggestion
of this family of compounds (Fig. A20; ASA- that valienamine is not derived from the shi-
NO et al., 1989). These have been obtained by kimate pathway (DEGWERT et al., 1987)
13 0-Ph
(-HzOO)
(R-NHd
R=O)
HO
valienamine
HO (-JoHoo
4
CH2OH
OH
HO
OH
HO
OH
0
Fig. 22. Proposed scheme of biogenesis of
C7 sugar-derived cyclitols of the valida-
mycin family. The labeling pattern from
validoxylamine A validoxylamine B validoxylamineG 1,6- Or 6-(C'3)-~-glucoseswas adopted
validamycin A validamycin B vaiidamycin G from RINEHARTet al. (1992) (cf. Fig. 20).
probably hold true only for the origin of the Actinoplanes sp. (DREPPER and PAPE,
C7 precursor (which is either sedoheptulose- 1996).
7-phosphate or an epimer thereof). However,
at least the cyclization and dehydration steps Trehazolin. A new type of aminoglycosidic
are probably catalyzed by enzymes homolo- and very specific trehalase inhibitor, trehazol-
gous to dehydroquinate synthase and dehy- in (or trehalostatin), a product of Micro-
droquinate dehydrase, respectively. This type monospora sp. and Amycolatopsis sp., and
of cyclitol is found in validoxylamines and the chemical synthesis of this compound and
validamycins (Fig. A21; basically Cb(1)- its P-anomer has recently been reported
Cb(4)-H compounds), and in acarbose, amy- (ANDOet al., 1991; KOBAYASHI and SHIOZA-
lostatin, and other structurally related KI, 1994; Fig. A23; Ca(l,2)-HA). The amino-
glycosidase inhibitors (Fig. A22; general for- cyclitol moiety in this molecule is unusual in
mula (H),-Cb( 1)-6DOH/H(l)-(H),; TRUS- that it could be formed in a Ca pathway (cf.
CHEIT et al., 1981; MULLER,1989; YOKOSE et Fig. 4) via myo-inositol and later ring contrac-
al., 1989). Valienamine and the related C7cy- tion as is suggested for pactamycin (see Sect.
clitols are all condensed via an imino group 5.2) or by direct reductive ring closure from a
with either another molecule of the same ori- ketohexose precursor, e.g., fructosed-phos-
gin and similar structure (validamycin family) phate. Also, the presence of a carbamoyl
or at the 4 position with a 4,6-dideoxy-~-glu- group is reminiscent of other groups of ami-
cose or 4-deoxy-~-glucoseunit (acarviosine noglycosides; the incorporation of this group
family of a-glucosidase and trehalase inhibi- may involve enzymes similar to those used in
tors). This suggests that the aminotransfer pathways of other aminoglycosides. However,
reaction takes place on a precursor of the C, the trehazolin molecule exhibits the only
cyclitol moiety and that valienamine might be known true aminoglycosidic linkage in its
an intermediate for all other derivatives (cf. bonding to the am-glucose moiety.
Figs. 6,21 and A21, A22; cf. RINEHART et al.,
1992). Sorbistins. The pseudodisaccharidic amino-
The validamycins are a-D-or P-D-glucosyl- glycosides of the sorbistin family (Fig. A24)
ated at various positions (Fig. A21) which are interesting in two respects: they are pro-
could be achieved by extracellular or cell- duced both by pseudomonads and higher ac-
wall-associated glucosyltransferases. Also, the tinomycetes, and they contain an unusual
acarviosine family of glycosidase inhibitors, open-chained diaminohexitol, 1P-diamino-
comprising acarbose, amylostatins, oligosta- sorbitol, of unknown biosynthetic origin (re-
tins, “amino-oligosaccharides” (epoxy deriva- viewed in UMEZAWAet al., 1986). It is tempt-
tives of amylostatins), adiposins, trestatins, ing to speculate that these compounds could
and AI-5662, are all variably glycosylated, be derived from a pathway similar to that of
mainly by condensation with di- or oligosac- the fortimicins followed by a reductive ring
charidic units such as maltose, oligomaltodex- cleavage between the C-1 and C-2 positions
trins, trehalose, or an acarviosine units of an aminocyclitol similar to fortamine later
(TRUSCHEITet al., 1981; MULLER, 1989). in the pathway (cf. Sect. 4.2; cf. Fig. 18). In
Again, the mechanism of addition of such fact, a nonacylated and nonmethylated fort-
blocks of oligosaccharidic sugars could be a amine precursor could be the direct precursor
membrane-associated extracellular process of 1,4-diaminosorbitol with an additional epi-
similar to the bactoprenol-dependent forma- merization in position c-5 (C-3 of fortarnine)
tion of extracellular heteropolysaccharides, creating the stereochemical configuration of
e.g., lipopolysaccharide 0-chains (GOEKE, sorbitol. The presence of a 4-amino-4-deoxy-
1986; RAETZ, 1996). Alternatively, these glucose can be explained in two ways, as for
compounds could be excreted as inactive pre- the biogenesis of 4-amino-4-deoxytrehalose
cursors by special exporters similar to strepto- (see the discussion on trehalosamines above).
mycin (cf. Sect. 4.1.1.4). This view is sup- In accordance with the similarity to the fort-
ported by the recent identification of an acar- imicin pathway a preformed (NDP-)Camino-
bose 7-phosphotransferase in the producing glucose would be the preferred precursor.
This could be synthesized via reactions simi- producers (WALKER,1975b; COURVALIN et

lar to the initial steps postulated for the bio- al., 1977) and clinical isolates of gram-nega-
synthesis of the NMLGA moiety of strepto- tive and gram-positive bacteria, such as obli-
mycin (cf. Sect. 4.1; cf. Fig. 16), whereby a gate pathogens or from opportunistic infec-
convenient 4-ketohexose intermediate is tions, where they are mostly plasmid-deter-
formed which could be used as an amino- mined (BENVENISTEand DAVIES, 1973;
transferase substrate. UMEZAWA,1974; DAVIESand SMITH,1978
FOSTER,1983), were the first among all anti-
Allosamidin. A new aminoglycosidic product biotic resistance phenomena to be identified
of Streptomyces spp., allosamidin and its de- at the molecular level. In fact, the first resist-
methyl and didemethyl derivatives (Fig. A25; ance gene cloned from an antibiotic producer
Ca(4)-HA(4)-HA), is the first chitinase inhib- was the gene encoding butirosin-3 ‘-phospho-
itor isolated from actinomycetes (SAKUDAet transferase, APH(3 ’)-IV, from Bacillus circu-
al., 1987; ZHOUet al., 1992, 1993). Allos- lans (COURVALIN et al., 1977). Also, the cor-
amidin is composed of unusual components, a relation between the phosphorylation of
branched five-membered C6 aminocyclitol streptomycin and its occurrence in a strepto-
(allosamizoline) and two N-acetyl-D-allos- mycin producers, S. griseus (MILLER and
amine units linked via p-1,4-glycosidic bonds WALKER1969; NIMI et al., 1971) and in clini-
which could be synthesized via a novel Ca cal strains of Escherichia coli and Pseudo-
pathway and via 3-epimerizations from monas aeruginosa (UMEZAWAet al., 1967a,
(N-acety1)-D-glucosamine, respectively. The b), for the first time led to the speculation
exact route of the synthesis of the allosami- that transferable antibiotic resistance could in
zoline cyclitol is unknown, though its C and N general have evolved in the producers of
atoms were shown to be directly derived from these natural products (BENVENISTEand
D-glucosamine. Therefore, it could either be DAVIES,1973). Later, ribosomal target site
synthesized from myo-inosamine formed by modification via specific 16s rRNA methyla-
an enzyme analogous to myo-inositolphos- tion was also identified as a major aminogly-
phate synthase via successive ring contrac- coside resistance mechanism in producers
tion, oxidoreduction/epimerization, and (PIENDL et al., 1984; CUNDLIFFE, 1989,
transamination steps. Alternatively, it could 1992a). For the first time, the transfer of a
be directly cyclized from ~-glucosamined- gene between gram-positive and gram-nega-
phosphate via a Ca type (or other) enzyme tive bacteria in nature was demonstrated to
forming pentacyclitols by activating C-5 and occur with an aminoglycoside resistance gene:
linking it to C-1. The aminooxazoline ring in the gene aphA-3 (kanamycinheomycin 3 ’-
the allosamizoline moiety (Fig. A25) is proba- phosphotransferase) was transferred between
bly introduced via N-amidinotransfer from L- enterococci (also streptococci and staphylo-
arginine, N-monomethylation, cyclization, cocci) and Campylobacter coli (reviewed in
and a second N-methylation step (ZHOU et COURVALIN, 1994). Some of the aminoglyco-
al., 1993). Therefore, also in this pathway sev- side-resistance genes might have spread from
eral steps could be catalyzed by enzymes re- aminoglycoside producers to other bacteria in
lated to strlsts gene products. earlier evolutionary periods, the mechanisms
of which are now becoming apparent (MA-
ZODIER and DAVIES, 1991; COURVAIJN,
4.5 Resistance in Aminoglycoside 1994). In this context, it is also interesting that
related multifactorial systems exist for the in-
Producers tercellular transport of DNA and protein
molecules which function specifically between
Interestingly, aminoglycoside and in parti- bacteria and other, unrelated types, such as
cular streptomycin resistance mechanisms by cells of other bacterial genera, and plant and
mutation (OZAKI et al., 1969; WALLACEet animal cells (POHLMAN et al., 1994).
al., 1979; PIEPERSBERGet al., 1980), and Much progress has been made in the analy-
those encoded by specific resistance genes in sis of resistance mechanisms in bacterial pro-
ducers of aminoglycosides and other carbohy- PERSBERG, unpublished data). Therefore, the
drate-containing self-toxic compounds above view would be additionally supported
(Tab. 4, Fig. 23). Our present knowledge can by the possible existence of an active export
be summarized as follows: system, also for antibiotically inactive precur-
(1) There are two basic resistance-confer- sors of aminoglycosides from the producing
ring biochemical phenomena: first, specific cells. It will be of further interest to investi-
elimination of the inhibitory function inside gate the presence of similar transporter genes
the producing cell (e.g., inactivation by mod- in other aminoglycoside production gene
ification, modification of the target site, or the clusters. A different type of membrane-an-
production of a new, insensitive version of the chored protein has recently been identified as
normal target complex) and second, active the product of the butB gene in Bacillus circu-
transport of the inhibitor out of the cells via lans NRRL B3312 (AUBERT-PIVERTand
energy-driven exporters. The first type could DAVIES,1994; HOTTA et al., 1995). The ButB
be regarded as a mere self-protection mecha- protein is related to the cell-wall-associated S-
nism which employs protective chemical +
layer proteins in low-G C gram-positive
groups or bypass mechanisms, whereas the bacteria and interruption of its gene blocks
second has its main function in transporting butirosin production, thus indicating that. it
the compounds out of the cells in order to en- might also be involved in aminoglycoside ex-
able the bioactive end products to reach their port.
natural destinations, namely other cells. We (3) Some of the aminoglycoside-resistance
only can speculate on the natural functions of mechanisms listed in Tab. 4, e.g., phosphory-
these compounds and the nature of the target lation and acetylation, are also among those
cells. In general these target cells could be which could give us the biochemical basis for
other cells of the same organism (hormone- an understanding of the evolution of this type
like functions: the target cells would be either of resistance determinants. These are sus-
nonproducing or otherwise differently differ- pected to be derived from biosynthetic en-
entiated stages of the life cycle) or cells of zymes or serve both purposes at the same
other, e.g., competitive, organisms (DAVIES time, such as the pac and bar genes encoding
et al., 1992; PIEPERSBERG, 1993). puromycin and phosphinothricin acetyltrans-
(2) Active exporters have been identified ferases, respectively (WALKER,1975a, b; PIE-
so far only for antibiotics such as macrolides, PERSBERG et al., 1988; CUNDLIFE, 1992a;
anthracyclines, tetracyclines (see Tab. 4), but THOMPSON and SETO,1995; TERCERO et al.,
not for aminoglycosides. Interestingly, two 1996). Again, it was the biosynthetic pathway
genes, strV and strW, have recently been de- of streptidine which first supported this spec-
tected in the streptomycin production gene ulation since intermediates in this pathway
clusters of both S. glaucescens and S. griseus; are successively dephosphorylated and re-
these genes encode a new type of ABC trans- phosphorylated twice, the last time at the
porters (PIEPERSBERG, 1995; BEYER et al., same position (C-6 of the aminocyclitol) as is
1996). Since streptomycin is secreted in an phosphorlyated by the resistance enzyme,
inactive, phosphorylated form (WALKER, StrA (AphD or APH(6); see below), in strep-
1975b; MANSOURIand PIEPERSBERG, 1991; tomycin producers (cf. Fig. 17; WALKER,
PIEPERSBERG,1995) these transmembrane 1975b). Since this enzyme also has a phospho-
exporters, if responsible for secretion, would rylating activity for streptidine and its imme-
not give rise to a resistance phenotype diate precursor, N-amidinostreptamine
(RETZLAFFet al., 1993). There is evidence in (WALKER,1975b; however, at much higher
support of this hypothesis: S. lividans 66 K M values; DISTLERand PIEPERSBERG, 1985)
strains carrying a combination of the strA it was suggested to be also a biosynthetic en-
[APH(6)] and strVW transcription units on zyme. However, among the gene products for
plasmids convert added streptomycin to an streptomycin production there are two, StrN
extracellularly accumulated streptomycind- and StsE, with peptide motifs similar to those
phosphate (unfortunately those clones turned of the catalytic centers of antibiotic and pro-
out to be very instable; S. BEYERand W. PIE- tein kinases, especially the H X D X ~ N X , - ~ ~ U D
Tab. 4. Resistance Mechanisms against Glycosidic Antibiotics in Producers (Examples)
Antibiotic Resistance Mechanism" Producing Organismb Refe-

rence
Streptomycin SM-6-PhT; (SM-3 "-PhT); (export: ABC?) S. griseus, S. glaucescens A

Neomycin neomycin-3-AcT, ne0m.d '-ACT, neom.-3 '-PhT S. fradiae A, G
Neomycin neomycin-3-ACT M. chalcea B
Paromomycin paromomycin-3-AcT, par.-3 '-PhT S. rimosus forma paromomycinus A
Ribostamycin ribostamycin-3-AcT, rib.-3 '-PhT S. ribosidificus A
Kanamycin 16s rRNA MT (G-1405) S. kanamyceticus A
Hygromycin B hygromycin B-7-PhT S. hygroscopicus A
Nebramycin nebramycin-3-AcT, par.-3 '-PhT 16s rRNA MT (A-1408, S. tenebrarius A
G-1405); SM-6-PhT
Gentamicin 16s RRNA MT (?) M. purpurea C
Fortimicin (astromicin) (16s rRNA MT?) M. olivasterospora D
1stamycin 16s RRNA MT (A-1408) S. tenjimariensis A
Kasugamycin kasugamycin-2'-ACT fortimicin/istamycin-2'( +2 ")-ACT S. kasugaensis G
Spectinomycin spectinomycin-2" -ACT;spectinom-?-PhT S. spectabilis G
Pactamycin 16s rRNA MT (A-964) S. pactum A
Lincomycin 23s rRNA MT (?); export:ABC; export:ApH S. lincolnensis E
Puromycin puromycin-ACT, export:ApH(?) S. alboniger A
Daunorubicin export:ABC S. peucetius F
Erythromycin 23s rRNA MT (A-2058); export:ABC; export:ApH Saccharopolyspora erythraea A
Tylosin 23s rRNA MT (A-2058); export:ABC, export:ApH S. fradiae A
Novobiocin res. target enzyme: DNA gyrase S. sphaeroides A
a ACT:acetyltransferase; M T methyltransferase; P h T phosphotransferase; A B C ATP-dependent transporter family; ApH: pH gradient-depende

transporter family
S.: Streptomyces; M.: Micromonospora
' A: CUNDLIFFE(1989); B: SALAUZEet al. (1992); C: PIENDLet al. (1984); D: HASEGAWA (1991); E: ZHANGet al. (1992; and unpublish
observations); F GUILFOILE and HUTCHINSON (1991); G: HOTTAet al. (1995)
biosynthesis r*
Y / W
Fig. 23. Schematic representa-

tion of self-protecting resistance 1 modification 3 active export (ATP-driven)
mechanisms in producers of an- 2 resistant target site (ts) 4 active export (ApH-driven)
tibiotically active secondary car-
bohydrates. @,a activehnactive secondary metabolite
= += modifying group
(U = hydrophobic residue, e.g., I, V, L) motif (4) Resistance to aminoglycosides also oc-

probably involved in the phosphate group curs as a cryptic phenotypic property in strep-
transfer (PISSOWOTZKI et al., 1991; RETZ- tomycetes, such as a second streptomycin re-
LAFF et al., 1993; KNIGHTONet al., 1991; cf. sistance enzyme (streptomycin-3 ” -phospho-
also Sect. 6). These are more likely to be the transferase) and kanamycin resistance en-
two phosphotransferases involved in the zyme (kanamycin-3-N-acetyltransferase) in S.
streptidine pathway (see Sect. 4.1.1.1). Re- griseus (HEINZEL, et al. 1988; HOITA et al.,
cently, a spectinomycin resistance gene en- 1988). It can be used either as a basis for the
coding a spectinomycin phosphotransferase, screening for new producers of aminoglyco-
was cloned from S. flavopersicus NRRL 2820 side-like antibiotics (ETIENNE et al., 1991) or
which showed striking similarity to StrN of S. to stimulate aminoglycoside production, e.g.,
griseus and S. glaucescens (J. ALTENBUCH-the 6’-acetyltransferase gene from S. kana-
NER, D. LYUTZKANOVA, and J. DISTLER, myceticus stimulates aminoglycoside produc-
personal communication). This finding tion when several copies are introduced into
strongly supports the hypothesis that resist- kanamycin and neomycin producers (CRA-
ance genes originate from biosynthetic genes. MERI and DAVIES, 1986).
Here, this hypothesis also has to be extended
to include the possibility that an ancestral
production gene could become a resistance
gene by divergent evolution or after degener- 4.6 Regulation in Streptomycetes
ation and modification of a pre-existing path-
way to yield simpler but still effective end The synthesis of aminoglycosides like the
products (this could be the case for spectino- production of other secondary metabolites in
mycin; see Sect. 4.1). The kanamycin-6’-ace- Streptomyces is regulated by a complicated
tyltransferase in S. kanamyceticus could also network of regulator proteins encoded by the
be involved primarily in the acetylation of ka- individual biosynthetic gene clusters and oth-
namycin precursors for several reasons (CRA- er pleiotropic regulatory genes. Besides these
MERI and DAVIES, 1986; see also below). regulatory genes, hormone-like autoregula-
tors (e.g., A-factor), intracellular signal mole- There is conflicting evidence in the litera-
cules (e.g., ppGpp), and nutrients (e.g., phos- ture concerning the influence of the nitrogen
phate, glucose, N sources, etc.) control the source on the production of streptomycin and
production of aminoglycosides at the level of other aminoglycosides. Nevertheless, ammon-
gene expression and/or enzyme activity. ium seems to impede the synthesis of strepto-
Mainly for the streptomycin producing S. gri- mycin, neomycin, and kanamycin (SHAPIRO,
seus as a model system we will here describe 1989) whereas nitrate and some amino acids
some of the most important details known or support the production of aminoglycosides,
postulated (summarized in Fig. 24). e.g., kanamycin (BASAK and MAJUMDAR,
The depression of the production of amino- 1973) and streptomycin. The stimulation of
glycosides by glucose has been reported for streptomycin production by alanine, arginine,
streptomycin, kanamycin, istamycin, and neo- andlor glutamine can be explained in terms of
mycin (VININGand DOULL,1988; DEMAIN, their being direct donors of nitrogenous
1989) and is caused by the repression of anti- groups in enzyme-catalyzed steps during
biotic synthases (e.g., N-acetylkanamycin am- streptomycin biosynthesis (cf. Sect. 4.1). The
inohydrolase; DEMAIN,1989) rather than by mechanisms underlying the positive effect of
an influence on the formation of precursors asparagine and proline (SHAPIRO,1989) or
of secondary metabolites. Carbon catabolite the repression of streptomycin synthesis by
control mechanisms extensively studied in valine (ENSIGN,1988; NEUMANNet al., 1996)
other bacterial systems such as E. coli and Ba- in S. griseus are presently not understood.
cillus subtilis (reviewed by FISCHER,1992) These compounds could influence the expres-
typically use cAMP/CAP-mediated gene reg- sion of biosynthetic key enzymes at the ge-
ulation. Although cAMP relieves glucose re- netic level. This influence could be mediated
pression of N-acetylkanamycin aminohydrol- by a NtrB/NtrC-like system or by a modula-
ase in S. kanamyceticus (SATOHet al., 1976) tion of the activities of primary metabolic en-
there is only weak evidence for the involve- zymes and streptomycin biosynthetic enzymes
ment of cAMP in regulating aminoglycoside at the physiological level via metabolite accu-
production in Streptomyces. The intracellular mulation and excretion by, e.g., feedback re-
cAMP concentration falls sharply in the mid pression or induction of gluconeogenetic
to late vegetative growth phase of S. griseus pathways or metabolic conditions favoring
S104 before the onset of secondary metabol- special routes of intermediary metabolism,
ism (RAGANand VINING,1978). Glucose in- such as the pentosephosphate cycle used in
creases cAMP levels in S. antibioticus while the “reverse direction”.
simultaneously repressing oleandomycin for- The biosynthesis of aminoglycosides (e.g.,
mation (LISHNEVSKAYA et al., 1986). There streptomycin, neomycin, kanamycin) is sensi-
is no evidence that cAMP directly affects tive to a high concentration ( > 5 mM) of inor-
streptomycin production (NEUMANNet al., ganic phosphate (MARTIN,1989). The extra-
1996) or controls secondary metabolism in cellular aminoglycoside phosphate phosphat-
general (MARTINand DEMAIN,1980). It is ase which forms the biologically active anti-
more likely that the carbon source repression biotic, in the case of streptomycin (see Sect.
is mediated by glucose kinase. Glucose kin- 4.1), from an inactive phosphorylated precur-
ase-deficient mutants of S. coelicolor having sor is inhibited by phosphate. Streptomycin-
normal intracellular levels of cAMP do not 6-P is accumulated in cultures of S. griseus
exhibit a glucose repression phenotype (AN- grown at a high concentration of phosphate
GELL et al., 1992; KWAKMAN and POSTMA, or at low pH due to the inhibition of the
1994). In contrast to these results, glucose-in- streptomycin-6-P phosphatase (MANSOURI
sensitive mutants of S. kanamyceticus blocked und PIEPERSBERG, 1991). In addition, there is
in kanamycin production were isolated which evidence that some secondary metabolic
have wild-type levels of glucose kinase genes are controlled by a PhoB/PhoR-like
(FLORESet al., 1993). Thus, a phosphorylated system; so-called pho boxes have been de-
sugar could mediate carbon source repression tected in phosphate-regulated promoter re-
of antibiotic production (DEMAIN,1989). gions (cf. Fig. 24B; MARTIN,1989; MARTIN
0 10 20 30 40
B
50
signal 7
S.
Fig. 24.Model for the regulation of streptomycin production in Streptomyces griseus. A Growth and strep-
tomycin production of S. griseus. The kinetics of biomass accumulation (0)and of streptomycin produc-
tion (0)are shown. The columns indicate the relative expression rates ol StrS, StsA, and StsC as deter-
mined by densitometry of the autoradiographed protein patterns from S. griseus pulse-labeled with (35S)-
methionine after separation on 2-dimensional protein gels. B Regulation of the expression of strlsts genes
by pleiotropic (e.g., A-factor) and pathway-specific (e.g., StrR) factors.
P: promoters; arrows indicate positive regulatory effects; tRNA-molecules are symbolized by cloverleaves.
The dominance of the StrR-dependent regulation is indicated by thick lines, for further explanantions see
the text.
and LIRAS,1989; LIRASet al., 1990). Similar DNA-binding protein (ONAKAet al., 1995).
structures were also found in the promoter In a proposed model the A-factor receptor
regions of the aphDp2 gene of S. griseus and protein acts in the absence of A-factor as a
of the strK (streptomycin-6-P phosphatase) repressor of streptomycin biosynthesis and
gene of S. glaucescens (LIRASet al., 1990; differentiation (MIYAKEet al., 1990). In addi-
DISTLERet al., 1990). tion, A-factor stimulates membrane-bound
The possible role of guanosine tetraphos- GTPases in S. griseus in vivo and in vitro (PE-
phate or guanosine pentaphosphate (ppGpp NYIGE et al., 1992). A-factor induces the ex-
or pppGpp) and GTP pools in controlling pression of StrR, the activator protein of
secondary metabolism in streptomycin-pro- some of the transcription units in the gene
ducing S. griseus and other antibiotic produc- cluster for streptomycin biosynthesis (cf. Sect.
ers was extensively studied by OCHI (1987, 4.1; DISTLERet al., 1987b; RETZLAFFet al.,
1988,1990). Although relC mutants showed a 1993; RETZLAFFand DISTLER,1995). The A-
marked reduction in antibiotic production on factor-dependent transcription of sfrR seems
complete medium it seemed difficult to assess to be controlled via regulation by the A-fac-
whether the effect on streptomycin produc- tor receptor of at least one further DNA-
tion was a direct consequence of the defect in binding protein (pX; cf. Fig. 24) which binds
ppGpp formation or an indirect effect of the to an enhancer-like element located upstream
relC mutation (OCHI,1990). In S. clavuligerus of the strR promoter (HORINOUCHI and BEP-
there is no relationship between ppGpp and PU, 1992; VUJAKLIJA et al., 1991). Three ad-
antibiotic production (BASCARANet al., ditional DNA-binding proteins ( P Y ~ - cf. ~ ; Fig.
1991). The same observation was made for 24B), not dependent on A-factor induction in
streptomycin synthesis in S. griseus on mini- S. griseus, interact with neighboring sites in
mal medium; no ppGpp formation could be the same strR promoter region, suggesting an
detected at any growth phase (NEUMANN et even more complex regulatory network gov-
al., 1996). erning StrR expression (VUJAKLIJAet al.,
A-factor (2-(6’-methylheptanoyl)-3R-hy- 1993).
droxymethyl-4-butanolide), an extracellular The StrR proteins encoded by the strepto-
diffusible autoregulatory molecule, triggers mycin biosynthesis gene cluster of S. griseus
both streptomycin biosynthesis differentia- and S. glaucescens share an identity of 62.8%.
tion and streptomycin biosynthesis in S. gri- The strR gene encodes an activator of strepto-
seus. This factor, which was discovered by mycin and OH-streptomycin production. The
KHOKHLOVet al. (1967, 1988), belongs to a existence of the StrR activator protein was
group of chemically similar y-butyrolactone first postulated in mutants of S. griseus and its
autoregulators which are synthesized by a va- role was suggested from the mutant pheno-
riety of Streptomyces (HORINOUCHI and BEP- type and its suppression by complementation
PU, 1992, 1994). A-factor induces streptomy- with wild type DNA (OHNUKIet al., 1985a,
cin production and sporulation in A-factor- b). Analysis of the mode of action of StrR in
negative mutants of S. griseus at concentra- S. griseus showed that StrR is a DNA-binding
tions as low as 1 nM. The induction by A-fac- protein which activates the expression of the
tor exhibits a strict growth phase dependence, sfrlsts genes strBl and sfsC at the level of
in that A-factor has to be present during the transcription by binding to upstream promot-
first hours of growth (“decision phase” moder sequences (hatched boxes in Fig. 24B;
el, see below; PIEPERSBERG, 1995; NEUMANN RETZLAFFet al., 1993; RETZLAFFand DIST-
et al. 1996). A-factor-induced regulation in S. LER, 1995; BEYERet al., 1996). These and a
griseus depends on an A-factor-binding pro- third identified StrR-binding site within the
tein, which is present in very low amounts (ca. sfrR gene are palindromic sequences with the
37 molecules per genome) and has a binding consensus sequence: GTTCGAnnGn(11)Cnn-
constant (Kd) of 0.7 nM (MIYAKEet al., 1989, CTCAACG (RETZLAFFand DISTLER,1995).
1990). Recently, the gene of the A-factor- The function of the StR-binding site within
binding protein, arpA, was cloned from S. gri- the sfrR gene, e.g., negative feedback regula-
seus and found to encode a repressor type tion of StrR expression or activation of the
promoter aphDp2 located approximately thetic genes could also be dependent on such
400bp downstream of this element, is un- sigma specific for secondary metabolic genes.
clear. In the OH-streptomycin gene cluster of In S. griseus the similarity of the strlsts pro-
S. glaucescens an StrR-binding site was also moters (strRp, aphDP, and strBlp) supports
identified upstream of the StrBl and strX this assumption (DISTLER et al., 1987b).
genes and within the strR gene. Recently, a There are several reports of the possible in-
gene homologous to strR was identified on a volvement of other factors in the regulation
DNA fragment of S. spectabilis conferring of streptomycin production in S. griseus
spectinomycin resistance in S. lividuns (D. strains: the DNA-binding protein ORF1590
LYUTZKANOVA and J. ALTENBUCHNER, unnecessary for sporulation and probably strep-
published data). Therefore, it can be specu- tomycin production in S. griseus (MCCUE et
lated that StrR homologous proteins are al., 1992), ADP-ribosylation of proteins (PE-
widely distributed activators of the synthesis NYIGE et al., 1992) possibly via influencing
of aminoglycoside antibiotics, especially for membrane-bound G proteins (GTPases; PE-
those using the Ca cyclitol pathway, which NYIGE et al., 1992), and C-factor, a cytodiffer-
could have evolved from a common ancestral entiation protein, excreted into the medium
gene cluster. by S. griseus 45H (SZESZAKet al., 1991). Re-
The expression of aminoglycoside biosyn- cently, it has been found that serinelthreonine
thetic genes in Streptomyces is regulated by and tyrosine protein kinases are present in
andlor is dependent on additional gene prod- most or all streptomycetes and, therefore,
ucts having pleiotropic effects and influencing could be part of the complicated regulation
both secondary metabolism and differentia- network necessary for the induction of anti-
tion on a more general level. The bldA gene biotic synthesis and differentiation, also in s.
encodes a leucine-specific t RNA which recog- griseus. Proteins phosphorylated at a tyrosine
nizes the rare codon UUA (cf. Fig. 24; CHAT- residue were detected in S. griseus and other
ER,1989,1992; LESKIWet al., 1991a, b). BldA Streptomyces spp. (WATERSet al., 1994). Spe-
mutants of S. griseus are deficient in aerial cific inhibitors of eukaryotic protein kinases,
mycelium formation and streptomycin pro- such as staurosporin (cf. Fig. A29), inhibit
duction (MCCUE et al., 1992). Recently, it sporulation (HONGet al., 1993) and strepto-
was demonstrated that tRNAUUA-dependent mycin production (NEUMANN et al., 1996) of
control could be a key switching process in S. griseus without affecting vegetative growth.
the onset of differentiation and secondary These results suggest that in Streptomyces a
metabolism (LESKIWet al., 1991a, b). In S. signal transduction pathway similar to that in
griseus, S. glaucescens, and S. spectabilis the eukaryotic organisms may control cell differ-
strR genes contain a respective TTA triplet in entiation and secondary metabolism.
conserved positions in the 5’-section of the S. griseus, like other streptomycetes (HOLT
respective reading frames (DISTLERet al., et al., 1992), undergoes a decision-making
1987b; PISSOWOTZKI et al., 1991; MAYER, process during a rather short period of 1-2 h
1994; D. LUTZKANOVA and J. ALTENBUCH- the mid logarithmic growth phase (Fig. 24A;
NER, unpublished). In the strN genes of S. gri- NEUMANN et al., 1996). This “decision phase”
seus and S. glaucescens additional TTA co- depends on pre-existing factors (such as A-
dons were detected. The strA (sph) gene of S. factor) and can be suppressed by certain me-
glaucescens possesses also a TTA codon, tabolites (e.g., valine) or inhibitors of protein-
which, however, is absent from the counter- modifying processes (e.g., 3-aminobenzam-
part of this gene, strA (uphD), in S. griseus ide). Various changes in physiology, e.g., a
(VOGTLIand HOTTER,1987; DISTLERet al., temporal increase in intracellular CAMP lev-
1987a). In S. coelicolor seven different RNA els and gene expression, can be observed dur-
polymerase sigma factors were identified and ing and after this period (DISTLERet al.,
the role of special sigma factors (e.g., WhiG) 1990 NEUMANN et al., 1996), which altogeth-
in controlling differentiation and antibiotic er seem to be an absolute prerequisite for lat-
production was elucidated (BUTTNER,1989). er cell differentiation and streptomycin pro-
The transcription of aminoglycoside biosyn- duction. It is notable that the A-factor-de-
pendent expression of some strlsts genes synthetic pathways (e.g., for the formation of
(stsC, stsA, stsS) occurs 2-4 h after this deci- the recently detected compound mycothiol,
sion phase, but 10 h before streptomycin is which is a major thiol in most actinomycetes;
detectable in the culture medium for the first NEWTONet al., 1996). Also, part of the en-
time (cf. Fig. 24A). This concept of the induc- zymes for the next steps (e.g., an StrI-like ox-
tion of secondary metabolism during a very idoreductase, cf. Sect. 4.1.1.1) could be used
early growth phase temporally separated in myo-inosotol degradation, a frequent prop-
from the production phase demands a global erty of Streptomyces spp. other than S. gri-
signal transduction network which might in- seus. For these reasons it is an interesting spe-
clude autoregulator molecules, receptors, culation that both the Ca and Cb pathways
transducers, protein phosphorylation/dephos- are used alternatively when a metabolic sort-
phorylation systems, and “second messen- ing mechanism is needed for the alternative
gers”. Although the direct regulation of the use of different anabolic and/or catabolic
strlsts genes and the superimposed regulatory pathways in the same cell.
cascades in S. griseus are still incompletely The various NDP activation mechanisms of
understood, a model for the regulation of monosaccharides before modification and
streptomycin production which includes cur- glycosyltransfers might have a similar meta-
rent hypotheses can be proposed (Fig. 24). bolic sorting effect. However, another expla-
nation might be that they are designed in or-
der to facilitate the horizontal genetic transfer
4.7 Overview of the of whole biosynthetic branches between bac-
teria and maybe even higher taxa. The 6DOH
Aminoglycoside Pathways pathways are used less in aminoglycoside pro-
duction than are the pathways leading to 6-
When we summarize what we have learned hydroxylated hexosamines; however, they are
from the genetics and biochemistry of amino- preferred over other carbohydrate pathways
glycoside synthesis in bacteria and when we in the modification of nonaminoglycosidic
compare the respective pathways with each secondary metabolites (cf. Sect. 3.2 and 5.2).
other, several similarities, but also quite div- In aminocyclitol aminoglycosides the 6DOH
ergent traits become apparent. The initial routes are practically exclusively coupled to
pathways of aminocyclitol formation starting the Ca type, streptomycin-like compounds
at the (deoxy-)scyllo-inosose level, whether (cf. Sect. 4.1). This could reflect a common
via the Ca or Cb routes, and some of the in- evolutionary link between both pathways or a
termediate steps, e.g., the involvement of a D- tight genetic coupling of both sets of genes re-
glucosamine moiety in the formation of a quired (or both). This coupling also implies
pseudodisaccharide in the 2DOS and fortimi- that the sets of functionally identical en-
cidistamycin groups, could use very similar zymedgenes involved should show a higher
enzymes (genes). Also, the resistance mecha- degree of relatedness in producers of com-
nisms may be very similar and based on the pounds such as streptomycins, spectinomy-
evolution of common ancestral genes (see cins, or kasugamycins than in Cb (e.g., 2DOS)
also Sect. 4.5 and 6). Whether the Ca or Cb producers. In contrast to this, amino sugars
routes for aminohexitol formation are used can either be derived from preformed amin-
might depend on the availability of the re- ated precursors (e.g., D-glucosamine), NDP-
spective enzymedgenes. However, other activated hexoses, or formed only after con-
more specific factors might force cells to use densation of a carbohydrate precursor into
either one or the other. The Cb pathway is (pseudo-)oligosaccharidiccompounds and oc-
more economical because it uses fewer steps cur in both Ca and Cb aminoglycosides.
(cf. Sect. 3.1). However, the initial step en- Therefore, it will be interesting to determine
zyme (~-myo-inosotol-3-phosphatesynthase) by which alternative route most of the amino-
for the Ca route might be more wide-spread glycoside hexosamine pathways proceed. In
in streptomycetes because it is probably used particular, the biosynthesis route of the N-
also for other, more generally distributed bio- methyl-L-glucosamine precursor of strepto-
mycin (mode (2), Sect. 4.1.1.3), the details of although evidence is emerging that some of
which are still unknown, could be an example them are based on biosynthetic functions
for a wider distribution of the HA pathway. (e.g., phospho- and acetyltransferases). How-
The streptomycins and probably some of ever, a connection between individual amino-
the more related substances (e.g., spectino- glycoside pathways might exist at a higher
mycin, kasugamycin) are made via the con- level of intracellular regulation or extracellu-
densation of highly modified precursors in lar cell-cell communication between the pro-
the final stage of the pathway. In contrast, the ducing cells and/or other cell types or organ-
larger group of 2DOS-containing aminoglyco- isms (see discussion in PIEPERSBERG, 1993).
sides and the compounds of the fortimicin/is-
tamycin family are condensed first from rela-
tively simple precursors, and the interme- 4.8 Aminoglycoside-Target Site
diates are strongly modified in the later path- Interactions and General Effects
way. Some substances, such as the acarbose-
like a-glucosidase inhibitors might be con- on Bacterial and Eukaryotic Cells
densed with additional subunits only after ex-
port to the cell surface (outer side of the cyto- The interaction of the classic aminoglyco-
plasmic membrane) with the involvement of sides with the small (30s) subunits of eubac-
membrane anchors and carriers, e.g., undeca- teria-type ribosomes (bacterial, mito-
prenyl phosphate residues, for the activated chondrial, and plastidal ribosomes) has been
intermediates. This could also indicate a more well investigated and reviewed several times
general principle in the formation of second- in the past (GORINI,1974; SCHLESSINGER et
ary metabolites where frequently mixtures of al. 1975; WALLACEet al., 1979; VAZQUEZ,
very similar end products formed by a given 1979; PIEPERSBERG et al., 1980; GALEet al.,
strain of the producing microorganism are 1981; HILL et al., 1990; NIERHAUS,1993).
found. Rather than being side products of Two functional groups can be distinguished
“inefficient” or “nonspecific” biosynthetic en- among the aminoglycosidic translational inhi-
zymes, complex product mixtures frequently bitors: (1) translational misreading-enhancing
observed in fermentations might be the result and bactericidal compounds, e.g., the strepto-
of the individual specificities of the export mycins and all 2DOS aminoglycosides; (2)
systems which are responsible for the inter- bacteriostatic compounds which do not affect
mediate/product patterns released from the translational accuracy, e.g., spectinomycin
cells. and kasugamycin. This difference is also re-
Thus, the strategies and routes used by the flected by the alterations in overall transla-
various producers for the production of the tion patterns after aminoglycoside addition to
chemically relatively homogenous class of cultures of E. cofi and B. subtifis, where group
aminocyclitol-containing aminoglycosides de- 1 compounds induce mistranslation and a rap-
scribed so far do not seem to be of similar ho- id loss of overall translation ability and group
mogeneity. Instead, a strong selective advan- 2 aminoglycosides a “quasi-relaxed’’ (Re1 - )
tage could exist in the long-term diversifica- phenotype similar to that induced by chlor-
tion by means of the modulation of the path- amphenicol with an uncoupled residual trans-
way design (e.g., by natural genetic engineer- lation of a special group of proteins for a
ing). Hence, in the long-term possibly from longer period of time (PIEPERSBERG, 1985).
time to time a new pathway for an altered end An exceptional position is taken by hygromy-
product develops which uniformly serves very cin B which exhibits functional traits of both
similar functions of which the individual pro- groups but does not seem to induce misread-
ducers take advantage and at the same time ing of the genetic code (BAKKER,1992).
escape the defense mechanisms of related Originally, it was believed that certain
producers or nonproducers in a competing small subunit ribosomal proteins mediate ami-
situation in natural environments. The resist- noglycoside interaction with the ribosome
ance mechanisms themselves in the aminogly- (OZAKI et al., 1969; PIEPERSBERG et al.,
coside producers do not seem to be similar, 1980). In contrast, an early suggestion of Go-
RINI (1974) that 16s rRNA could be the pri- although they bind at different positions,
mary target site of aminoglycosides was prov- seem to bind within a common domain in the
en more recently; evidence is also accumulat- catalytic center of the 16s rRNA of bacterial
ing that 16s and 23s rRNAs are the catalyti- ribosomes (CUNDLIFFE, 1990; NOLLER,1993;
cally active components in both decoding and BRINKet al., 1994).
peptidyltransfer in the eubacterial ribosome Effects other than the direct interaction of
(MOAZEDand NOLLER,1987; DE STASIOet the group 1 bactericidal aminoglycosides with
al., 1989; CUNDLIFFE,1990; NOLLER,1993). the bacterial ribosome have been suggested
Mistranslation is also observed with some am- to be responsible for causing cell death in eu-
inoglycosides in some but not all Archaea bacteria (DAVIS, 1987). Besides mistransla-
tested (LONDEIet al., 1988). The decoding tion at actively elongating ribosomes (Go-
process involves an interaction of a short RINI,1974; WALLACEet al., 1979; PIEPERS-
RNA duplex (paired codon-anticodon) with BERG et al., 1980), two phenomena were ob-
the P and A sites of the 30s subunits, a pro- served to be coupled with lethality, mainly in
cess similar to that of the self-splicing group I studies with streptomycin and gentamicin: (1)
introns (Fig. 25; VON AHSENand NOLLER, a two-step uptake kinetics of aminoglyco-
1993). Therefore, it is not surprising that the sides, the first phase of which is dependent on
RNA molecules of group I introns specifically the APcomponent of the proton motive force
bind aminoglycosides and that the splicing or driven by ATP (HANCOCK,1981a, b;
reaction is inhibited by these antibiotics (VON BRYANand KWAN,1983; FRAIMOW et al.,
AHSENet al., 1991; SCHROEDER et al., 1993). 1991), and (2) membrane damage, probably
It has also been demonstrated that short via the induction of membrane channels
RNA analogs of the 16s rRNA decoding site (DAVISet al., 1986; BUSSEet al., 1992). Two
specifically interact with both aminoglyco- models could explain the pleiotropy of action
sides and its RNA ligands, tRNA and of these compounds: (1) incorporation of mis-
mRNA, in the absence of ribosomal proteins read proteins into the cytoplasmic membrane,
(PUROHITand STERN,1994). These findings and (2) mistranslation of proteins with a short
initiated speculations that aminoglycosides half-life involved in DNA replication and/or
are “molecular fossils” echoing their possible cell division and made at very few copies in a
functions as ligands of catalytically active distinct phase of the cell cycle. The first mod-
RNAs in a precellular RNA world (DAVIES el would explain also the observed drastic in-
et al., 1992). An argument in favor of this hy- crease of aminoglycoside uptake in the sec-
pothesis is the finding that all known transla- ond, killing phase, where the antibiotics are
tional inhibitors among the aminoglycosides, irreversibly accumulated inside the cells in a
Decoding Splice-site selection

3’ 5’ 5‘ss
Y
16s rRNA-A site
Fig. 25. Similarities between interaction of aminoglycosides in the decoding site of bacterial ribosomes
et a]., 1993).
(left) and self-splicing type I introns (modified according to SCHROEDER
large excess over the number of ribosomes.
Aminoglycosides accumulate by binding elec-
5 Related Sugar
trostatically to anionic groups of cytoplasmic
macromolecules or by being caged into de-
Components in Other
gradation products of mistranslated proteins Secondary Metabolites
(DAVIS,1987; BUSSEet al., 1992). Also, pas-
sage through the outer membrane of gram-
negative bacteria could be affected by direct 5.1 Lincosamides
interaction of aminoglycosides with porins
and/or lipopolysaccharide components (HAN- Characteristic of the lincosamides is an in-
COCK et al., 1991). teresting C-8 aminosugar component which is
In patients treated with aminoglycosides represented by the methylthiolincosaminide
the most prominent adverse side effects en- (MTL) moiety of lincomycin A (Fig. A26;
countered are oto- and nephrotoxicity WRIGHT,1983). Lincomycins (LM) A and B
(PRATTand FEKETY,1986). Aminoglycoside and celesticetin are members of the lincos-
hypersensitivity in humans is a maternally in- amide group of antibiotics produced by S. lin-
herited trait, suggesting a mitochondrion-as- colnensis and several other streptomycete
sociated genetic defect, and a molecular species. Intensive biogenesis studies on LM-A
mechanism has been proposed recently involving measurement of stable isotope la-
(HUTCHINand CORTOPASSI,1994). Hyper- beling patterns led to the proposal of a bio-
sensitive persons have been shown to carry a synthetic pathway for this antibiotic from an
1 5 S Gmutation, i.e., a replacement of the A octulose and L-tyrosine. The former which is
residue which is normally in this position of presumably derived from intermediates of the
the small (12s) mitochondrial rRNA by a G. pentose phosphate cycle is a precursor of the
This area of small, 16s type rRNAs is known MTL moiety and the latter is a precursor of
to interact with aminoglycosides and to be in- the propylproline (PPL) subunit of LM-A
volved in the control of codon-anticodon- (Fig. 26; BRAHMEet al., 1984a, b). However,
pairing (see above; cf. Fig. 25). This defect for the MTL subunit the genetic record seems
leads to frequent death of hair cells in the ear, to suggest a participation of a nucleotide
which is proposed to be due to enhanced pro- (probably dTDP) activation step and a series
duction of superoxide by a mistranslated mi- of modification steps, including dehydrata-
tochondrial complex I, the proteins of which tion, on NDP-activated sugar intermediates
account for more than half of the coding ca- (PESCHKEet al., 1995). Therefore, either a to-
pacity of the mitochondrial genome. Nephro- tally different initial pathway starting from D-
toxicity is primarily accompanied by mem- glucose with the familiar first steps of the 6-
brane damage in the renal tubular cells. The deoxyhexose pathways (see Sect. 3.2.1 and
biochemical basis of this effect in man is still below) or an NDP-activation and modifica-
not fully understood and could be caused tion of a C8 sugar intermediate are the routes
either by an effect on lipid metabolism and for the formation of the MTL subunit which
structure or by mistranslation of proteins in- currently can be postulated (cf. Fig.26;
ducing malfunctions of membrane traffic PESCHKEet al., 1995). Also, it was found that
(KOHLHEPPet al., 1994; and op. cit.). The ~-3,4-dihydroxyphenylalanine (L-DOPA) and
nephrotoxic effect is also seen in neonates 3-propylidene-A -pyrroline-5-carboxylic acid
whose mother was treated with aminoglyco- are probable intermediates in the PPL bio-
sides during gestation (SMAOUIet al., 1993) synthetic branch of the pathway (BRAHMEet
and can be suppressed by polyaspartic acid al., 1984a; K u o et al., 1992). In addition to a
(SWANet al., 1991). specific set of biosynthetic enzymes a special
coenzyme, the so-called co-synthetic factor,
which is structurally identical to the ribode-
azaflavin moiety of the F420 coenzyme of me-
thanogenic bacteria is needed for the PPL
branch of the LM production pathway in S.
A
Xylose-5- (P) Glucose-1- (P) L- Tyrosine
+ 4.
Sedoheptulose-7- (P) Desoxythymidintri- (P)
I
(dl-rP)
H2N’
(LmbO?)
Erythrose - 4 - (P)
+
I
I
bH
dTDP-Glucose
(LmbM?)
ib
(LmbS?)
CH20-P
OH
OH
n steps (?) 3.
(LmbO,M,S?)
HNL
COOH
Methylthio- Propylproline
(PPL)
bH
Fig. 26. Hypothetical pathway of lincomycin A. For the biosynthesis of the Cs sugar moiety, methylthio-
lincosaminide (MTL), two alternative pathways are proposed (A) from intermediates of the pentose phos-
phate cycle via formation of a C8sugar precursor; (B) from dTDP-glucose via a 6DOH pathway an a later
extension of the carbon chain. The possible involvement of three of the gene products of the putative MTL
genes (cf. Fig. 27) is indicated.
Streptomyces lincolnensis 78-1 1 (Lincornycin A)
L M N Z P O S R 0
Q K I KS
KS I
(ImbL) lmbM (ImbN lmbZ ImbP) lmb0 lmbS lmbR lmbQ
dTDP-hexose synthase sugar arninotransferase(SMAT)
dTDP-hexose 4,6-dehydratase other sugar-modifying enzymes

Fig. 27. The, “sugar” subcluster of the biosynthetic gene (lmb) cluster for lincomycin A. Genes marked by
lmr encode resistance (or export) proteins.
lincolnensis (Kuo et al., 1989, 1992). In con- present in a single copy. However, in the in-
trast, no intermediates of the MTL subpath- dustrial strain S. lincolnensis 78-11 the gene
way could be identified so far. The biosynthe- clusters for the production of LM and mela-
sis of the related antibiotic celesticetin (Fig. nin (melc) are duplicated on a large (450-
A26) probably proceeds via a very similar 500 kb) DNA segment by transposition to an-
route. other genomic region accompanied by dele-
The LM-A production gene clusters of two tion events (PESCHKEet al., 1995). This fact
overproducing industrial strains derived from indicates that enhanced gene dosage is one of
Streptomyces lincolnensis NRRL2936 were the factors underlying overproduction in the
cloned and analyzed by mutagenesis and hy- developed industrial strains. Only a minority
bridization: one of them (strain 78-11) has of the putative Lmb proteins belong to
been sequenced (CHUNGand CROSE, 1990; known protein families which among the sup-
ZHANGet al., 1992: PESCHKEet al., 1995). posed PPL biosynthetic enzymes include
The lmbllmr gene cluster is composed of 27 members of the y-glutamyl transferases
open reading frames with putative production (LmbA), an L-tyrosine oxidase (LmbB2), an
functions (biosynthetic or regulatory: lmb L-DOPA oxidase (LmbBl), amino acid acyl-
genes) and three resistance (or export; lmr) adenylate synthetases (LmbC), aromatic ami-
genes, and is flanked by two of them, the no acid aminotransferases (LmbF), and imi-
lmrA and 1mrC genes (Fig. 27). Compared dazoleglycerolphosphate dehydrataseslhistid-
with the respective genome segment of other inolphosphate phosphatases (LmbK)
lincomycin producers, the Imbllmr clusters (PESCHKEct al., 1995; U. PESCHKE, D.
seem to have a very similar overall organiza- NEUSSER,S. KASCHABECK, and W. PIEPERS-
tion in the other LM producers, S. pseudogri- BERG, unpublished data). However, except
seolus NRRL3985, Streptomyces sp. for LmbB1,2, even these proteins do not eas-
NRRL3890, and S. vellosus NRRL8037. ily fit into any hypothetical route from L-Tyr
However, they are embedded in nonhomolo- to PPL, suggesting that the whole pathway is
gous genomic environments and exhibit poly- largely unknown.
morphic restriction patterns. Similarly, in the right hand part of the lmbl
In the wild-type strain (S. lincolnensis lmr cluster a set of eight genes, lmb-
NRRL2936) the lmbllmr cluster is apparently LMNZPOSQ, encode proteins which to var-
ious extents are related to enzymes involved which in addition to an inositol unit contains
in the sugar converting or modifying metabol- an L-hexosamine derivative (Fig. A27).
ism. The existence of these putative enzymes, Therefore, their formation could have several
dTDP-glucose synthase (LmbO), dTDP-glu- steps in common with the streptomycin path-
cose 4,6-dehydratase (LmbM), and (NDP-)- way. In this case equivalent enzymes could be
ketohexose aminotransferase (LmbS), sug- used beyond the cyclitol formation, via D-
gests that in contrast to the earlier proposals myo-inositol-3-phosphate synthase and phos-
of the hypothetical biosynthetic pathway re- phatase (StrO-like enzyme?). The initial nu-
sulting in the MTL moiety (BRAHMEet al., cleotidylation, e.g., part of the oxidoreduction
1984b) this branch of the LM-pathway seems and epimerization, or transamination steps,
to be based on nucleotide-activated sugar in- which might be catalyzed by the gene prod-
termediates. The stable isotope-labeling pat- ucts StrQ, StrP, StrX, and StsA (or StrS), re-
tern found by these authors, suggesting an oc- spectively, during the formation of the N-me-
tulose - phosphate intermediate to be formed thyl-L-glucosamine subunit of streptomycin
first, could also be explained by the same type (see Sect. 4.1.1.3), could also be involved in
of “equilibration” of externally applied glu- the biosynthesis of the hexosamine moiety.
cose through the pentose phosphate cycle as An example is the cyclopentane ring in
was found for the biosyntheses of neomycin pactamycin (Fig. A30) which according to
and validamycin (RINEHARTet al., 1992; cf. earlier biogenesis studies should also be de-
Fig. 20). However, this would probably lead rived from myo-inositol, i.e., via a 1,243-
to another labeling pattern as that found in diaminocyclitol (RINEHART et al., 1981, 1992;
the case of the MTL subunit (BRAHMEet al., GRAFE,1992). Similarly the pentitols in allo-
1984b). At first, for the reaction sequence samidin and trehazolin (Figs. A23 and A25)
NDP-pyranose synthesis/4,6-dehydratation are possible products of the Ca route. The
(cf. Fig. 7) only a hexose seemed likely as a ring contraction of a hexitol to a pentitol
precursor. Therefore, the alternative pathway could be accomplished by a mechanism simi-
given in Fig. 26 (route B) was proposed lar to that operating in the formation of a fu-
(PESCHKEet al., 1995; PIEPERSBERG, 1994). ranoid ring from a pyranose derivative during
The later detection of a gene ZmbR encoding dihydrostreptose biosynthesis (ses Sect.
a transaldolase-like enzyme in the “sugar sub- 4.1.1.2; cf. Fig. 15). In such a mechanism a de-
cluster” of the lmb cluster (cf. Fig. 27) could hydrogenase is believed to reduce a keto
support an alternative route via a pyranosidic group and concomitantly introduces a rear-
octose intermediate which is NDP-activated rangement in the C chain resulting in a C,
and further modified in this form. Finally, a branch. Other possible routes for the synthe-
thiomethyl unit to the C1 position of the pos- sis of pentitols as in pactamycin or in allosam-
tulated (NDP-)6-amino-6,8-deoxyoctose in- idin could be a new and so far unknown Cb
termediate would be added which could be mechanism acting on a ketohexose phosphate
transferred from 5 ‘-thiomethyladenosine, a (e.g., fructose-6-phosphate) as a precursor
side product of polyamine biosynthesis from and subsequent hydroxylation of the 2-deoxy-
S-adenosyl-methionine. pentitol via 2,3-dehydratation and isomeriz-
ing rehydratation such as in the formation of
2-hydroxyvalidamines (cf. Fig. 22).
In contrast to this, as a second example the
5.2 Cyclitols pentitol moieties of some nucleoside homo-
logs, aristeromycin, neplanocin A (inhibitors
Other natural products containing cyclitol of S-adenosylcysteine hydrolase; Fig. A27),
moieties or their (e.g., aromatic, pentitol, and and adecypenol, which are inhibitors of var-
cyclohexane carboxylic acid) derivatives are ious enzymes of nucleoside metabolism, are
more widespread than is at first obvious. For though to be synthesized from a fructose der-
instance, cyclitol derivatives formed via the ivative and via pathways either similar to the
Ca (cf. Sect. 3.1) pathway are components in Ca or analogous to the Cb routes (PARRYet
nucleoside type antibiotics, e.g., adenomycin al., 1989; PARRY,1992). Stable and radioac-
tive isotope labeling experiments have shown tion, and can now be studied in a more direct
that the stereochemistry of the C-6 atom in way.
the cyclopentane ring formation between C-6
and C-2 of the hexose precursor is opposite to
that in enzymatic myo-inositolphosphate syn-
thesis. Also, the inversion of stereochemistry 5.3 6-Deoxyhexoses
is unexplained by a mechanism of the Ca
type. Therefore, it is more likely that the aris- The 6DOH-derivatives, both of D- and L-
teromycin pentitol is formed by a Cb type en- configuration (cf. Sect. 3.2.1; Figs. 8 and 9),
zyme cyclizing fructose-6-phosphate first to a are the most variable group of sugar compo-
2-deoxyketopentitol. Successive reduction of nents in low molecular weight natural prod-
the keto group to hydroxyl, and phosphoryla- ucts. They are widely distributed especially in
tion and pyrophosphorylation in the 5'- and the actinomycete antibiotic groups of the aro-
1 '-positions, respectively, could result in an matic (type I; anthracyclines, angucyclines,
analog of 5 '-phosphoribosyl-1 '-pyrophos- granaticin, etc.) and macrolide type polyke-
phate (PRPP). The addition of the adenine tides (type 11; macrolides, avermectins, PO-
moiety could then be catalyzed by either a lyenemacrolides) and in the glycopeptides
salvage enzyme, e.g., purine-RRPP transfer- (vancomycin family, thiopeptides) (PIEPERS-
ase, or a set of enzymes equivalent to the pu- BERG, 1994; LIU and THORSON, 1994). They
rine biosynthetic enzymes. Indirect evidence also are building blocks in highly modified
has been presented that both could occur in and variable extracellular polysaccharides,
the organism producing aristeromycin and such as the lipopolysaccharides of gram-nega-
neplanocin, i.e., Streptomyces citricolor (PAR- tive bacteria (LIUand THORSON,1994). Their
RY, 1992). biosynthetic pathways have not been studied
An interesting monoaminocyclitol, N-me- extensively in most cases. However, it was
thyl-scyllo-inosamine, is found as a so-called shown that DNA probes taken from the
rhizopine in rhizobia (MURPHYet al., 1987, genes encoding enzymes involved in the basic
1993). Its anabolic metabolism is largely un- steps of the 6DOH pathway (Sects. 3.2.1 and
known so far. However, it will be interesting 4.1.1) detect the gene clusters relevant for
to see whether enzyme-catalyzed steps equi- 6DOH biosynthesis in many streptomycetes
valent to those in the streptidine pathway are and other actinomycetes (PIEPERSBERGet
used (cf. Sect. 4.1.1.1). al., 1991b; STOCKMANN and PIEPERSBERG,
Also, nonglycosylated and neutral cyclitol 1992). Hybridization signals obtained with
derivatives, with unknown biological activi- these cloned fragments were localized in the
ties and biosynthetic origins, such as 2,3,4-tri- predicted production gene clusters. From
hydroxy-6-methylcyclohexanone (Fig. A28; such experiments it was found that the dTDP-
MULLERet al., 1986), seem to be common glucose 4,6-dehydratases, StrE, and related
and stable actinomycete products. Similar enzymes are the most highly conserved ones.
compounds of this group have been found in In contrast, the genedenzymes for the first
the chemical screening of actinomycete -sec- step related to strDIStrD from S. griseus
ondary metabolites (S. GRABLEY,personal (dTDP-glucose synthetase) were more diver-
communication). Cyclophellitol, produced by gent and, interestingly, some of them did not
the mushroom Phellinus sp. and active as a hybridize at all, such as strD or tylAl from ty-
specific inhibitor of almond p-glucosidase losin-producing S. fradiae. The tylAl gene is
(ATSUMIet al., 1990a, b), is another example more closely related to gram-negative bacteri-
for this type of compound. It is plausible that al genes which encode the same enzyme (such
this is synthesized by a basic pathway and cy- as rfbA from salmonellae). Again, this sug-
clization mechanism (Cb) similar to that of gested that these genes can be recruited from
the valienamine-related C, cyclitols (cf. Sect. a common gene pool independent of any tax-
4.4). The exact biosynthetic routes and their onomic distances by all eubacteria whenever
relationships to the known cyclitol pathways a particular secondary carbohydrate metabol-
of all these compounds await further clarifica- ism is required.
The basic enzyme complement and the cin as a single bond only the N-glycosidic
gene and protein families involved in 6DOH linkage exists between the 4 ’-0-methylglu-
biosynthesis is becoming increasingly appar- cose and the indolocarbazole moieties (LAM
ent (see reviews by SHNAITMAN and KLENA, et al., 1989; Fig. A29). The transferring glyco-
1993; PIEPERSBERG, 1994; LIU and THOR- syltransferases, so far postulated to be in-
SON,1994; cf. Sect. 3.2). Enzymatic synthesis volved in 6DOH transfer (LIU and THORSON,
of the activated sugar components dTDP-L- 1994; OTTENet al., 1995; PIEPERSBERG et al.,
oleandrose (MACNEIL,1995; see Fig. 9) and 1995; DICKENSet al., 1996) are related to the
dTDP-L-rhamnose (REEVES,1993; see Fig. 9) macrolide glucosyltransferases (MgtA, MgtB)
which are naturally derived from 6DOH causing resistance to 14- and 16-membered
pathways has been achieved by complete in macrolides (CUNDLIFFE,1992b; VILCHESet
vitro enzymic synthesis from dTDP-D-glu- al., 1992).
cose. Also, total chemical synthesis and intro-
duction into the natural aglyca have been re-
ported for some 6DOH derivatives, e.g., the 5.4 Other Pentose, Hexose, and
6DOHs D-mycosamine (BEAU, 1990) which Heptose Derivatives
occurs in polyene macrolides (see Fig. 8) and
L-daunosamine (THOMAS,1990) which occurs Nitrogen-containing derivatives of pentoses
in anthracyclines (see Fig. 9). and hexoses (other than 6DOHs) are found
The types of chemical bonds involved in in many microbial secondary metabolites, es-
the linkage between 6DOH moieties and pecially in nucleoside type antibiotics, either
their aglyca also vary widely and are worth as (deoxy-)ribose analogs or as additionally
some attention. Besides the usual O-glyco- modifying components in products originat-
sidic bonding C- and N-glycosidic or more ing from pathways with chemically heteroge-
complex linkages are also observed. C-glyco- nous precursors.
sidic bonds occur with both aromatic and ali-
phatic C atoms, e.g., in altromycins (BRILLet Pentosamines. Examples for aminofuranoses
al., 1990). Examples are the 2,6-dideoxy-~- are found in some well-studied aminonucleo-
glucose and 3-amino-3-N-methyl-4-O-methyl- side antibiotics, e.g., puromycin and the re-
2,3,6-trideoxy-~-hexosemoieties in granaticin lated antibiotic A201A (cf. Fig. A27). In the
and staurosporine, respectively (Fig. A29). In puromycin-related aminonucleosides the bio-
granaticin the incorporation of the double- synthetic origin of the 3-amino group of the
(1 ’,4’-)C-bound 6DOH into the benzoiso- ribose derivative, for which an adenine nu-
chromanequinone polyketide aglycone could cleotide is the precursor, is not yet known.
be achieved via two different routes from a However, intensive investigations of the ge-
dTDP-4-keto-2,6-~-hexoseprecursor accord- netics and the enzymology of these com-
ing to earlier proposals (FLOSSand BEALE, pounds have been initiated recently (LA-
1989): via the formation of a C-glycosidic CALLE et al., 1992). Interestingly, the putative
linkage after dTDP elimination or via a reac- products of the two genes pur3 and prgl in
tion of the aromatic C-10 and the 4’-keto the puromycin biosynthetic gene cluster of
group of the sugar. The latter route is consid- Streptomyces alboniger (TERCERO et al.,
ered more likely since other configurations in 1996) are related to the StrO (possible cycli-
the pyrane ring do not alter the regioselectivi- tol-phosphate phosphatase) and StrSIStsAl
ty of the condensation reaction. In staurospo- StsC (SMATs) proteins, respectively, from S.
rin and in the staurosporin-related compound griseus (cf. Sect. 4.1). Besides the 3-amino-3-
K-252a (Fig. A29; KASE et al., 1986) an N- deoxyribose moiety A201A contains two ad-
glycosidic linkage is probably formed first be- ditional rare sugar units: an unusual furano-
tween the 1’-position of the aminodDOH sidic hexose with an unsaturated C-C bond
and one of the indole N atoms of the trypto- between C-1 and C-2 branching off directly
phan-derived indolocarbazole heterocyclic from the furane ring, and a 6DOH, an O-me-
system. This may also be the case for other thyl-D-rhamnose (cf. Fig. A27). Therefore, a
very similar metabolites since in rebeccamy- strE probe was tested for hybridization and
6 Evolutionary Aspects 457
found to give a signal in the genomic DNA of 1992; PIEPERSBERG,1993; EMBLEYand

the A201A producer Streptomyces capreolus STRACKEBRANDT, 1994). Although there are
(A. JIMENEZ,personal communication; cf. no data at present which could unequivocally
Sect. 5.3). prove this hypothesis it nevertheless has a
high likelihood. Evidence is accumulating
Hexosamines. Hexosamines of various struc- that the genes for these pathways have been
tures and positions of amino-N substitution transmitted horizontally between the actino-
occur in many secondary metabolites (see ex- mycetes and have also spread to other micro-
amples in Figs. A7, A8, All-15, A17-19, bial groups. Hence, the butirosin (Fig. A13)
A24, and A27) and many of them may share biosynthetic genes are not likely to have
common biosynthetic traits with aminoglyco- evolved independently in the Bacillaceae.
sides. Most of them have retained a pyrano- Rather, it seems that they have been derived
sidic ring structure, but incorporation into from an actinomycete gene cluster involved in
other heterocyclic ring systems via linearized the biosynthesis of a ribostamycin-like amino-
intermediates can also occur, e.g., the D-glu- glycoside (cf. Fig. A13) which was laterally
cosamine-derived building block in mitomy- transferred to Bacillus circulans or an ances-
cins (Fig. A31; HORNEMANNet al., 1974; tor thereof. The elucidation of the structures
OKADAet al., 1988). Examples of aminopyra- of the gene clusters which encode enzymes in-
noses are encountered in some nucleoside volved in the formation of ribostamycin and
type antibiotics, e.g., streptothricins and blas- butirosin and of other 2DOS aminoglycosides
ticidin S (cf. Fig. A27). The origin of the 2- will clarify these evolutionary aspects. The
amino-D-glucose derivative in streptothricins horizontal dissemination of aminoglycoside
is still unknown. resistance genes among all major eubacterial
The biogenesis of the 4-amino-2,3,4-deoxy- groups is particularly well documented (Fos-
glucuronic acid moiety in blasticidin S (cf. TER, 1983; PIEPERSBERG et al., 1988; CUND-
Fig. A27), which is produced by Streptomyces LIFFE, 1989; SHAWet al., 1993). The primary
griseochromogenes and used against rice blast structure similarities, in particular, between
disease, was studied by isotope labeling the aminoglycoside phosphotransferase en-
(GOULD,1992). The fully retained position- zymes (APH) suggest a common origin of an-
specific labels of fed D-glucose (or even with tibiotic phosphotransferases in bacteria and
higher yield with D-galactose) in this com- eukaryotic protein kinases (DISTLERet al.,
pound indicated that it is directly derived 1987a; HEINZEL et al., 1988; PIEPERSBERG et
from a UDP-D-glucose or UDP-D-galactose al., 1988,1991; KIRBY,1990; RETZLAFFet al.,
precursor with UDP-D-glucuronic acid and 1993). This common evolutionary origin is
cytosylglucuronic acid as intermediates. The also supported by site-directed mutation and
latter intermediates were confirmed by direct gene fusion experiments which yield hybrid
measurements in cell-free extracts of the en- APH enzymes. These experiments have
zymes UDP-D-glucose epimerase, UDP-D- shown that the essential amino acid residues
glucose oxidase, and cytosylglucuronic acid and the two-domain structure of the catalytic
synthase. subunit in the CAMP-dependent protein kin-
ase of eukaryotes are conserved in the APHs
(TAYLORet al., 1990 BLAQUEZet al., 1991;
KNIGHTONet al., 1991; PIEPERSBERG et al.,
1991a; RETZLAFFet al., 1993).
6 Evolutionary Aspects The above-mentioned versatile enzyme
families, which are involved in secondary su-
Most of the aminoglycosides and related gar metabolism, such as aminocyclitol biosyn-
compounds considered here are actinomycete thesis and hexose activation and modification,
products. Therefore, the question of whether are obviously products of a modularly used
the evolution of the respective pathways also gene pool, the products of which are mainly
occurred in this molecularly defined lineage involved in the production of highly variable
of bacterial taxonomy arises (DAVIESet al., and mostly secreted biomolecules not used in
primary cell functions. Instead, they seem to producers. Therefore, in streptomycin pro-
be mainly involved in the extracellular com- ducers gene duplication could have resulted
munication of the producing cell with other during the evolution of two amidinotransfer-
biological systems (PIEPERSBERG, 1992, ase with district substrate specificities from
1993). The secreted low molecular weight one ancestral enzyme exhibiting both activi-
molecules, e.g., antibiotics, enzyme inhibitors, ties. Alternatively, the bluensidine pathway
and autoregulators, could be used for the de- could be a degenerated streptidine pathway
fense against agonistic or competing organ- after loss of the sfrB2 gene and changes in the
isms or as hormone-like signal transmitters in substrate specificity of the StrBl protein.
differentiation processes. The extracellular Also, an open question is whether or not
polysaccharides and other cell surface-bound myo-inositol is a specific precursor and,
compounds (LPS, capsular polymers, etc.) are therefore, its synthase is an intrinsic enzyme
synthesized on the basis of very similar path- of the streptomycin pathway, since measure-
ways and gene clusters; they may be cell sur- ments of this enzyme under various culture
face “individualizing” material for cell-specif- conditions and in various strains suggested a
ic recognition and attachment or protection, rather non-specific distribution (SIPOS and
i.e., these compounds could serve similar SZABO,1989). The recent finding that proba-
functions and be regarded as cell surface- bly all actinomycetes contain a major thiol
bound secondary metabolites. In accordance compound called mycothiol (NEWTONet al.,
with this, both the genes for secondary meta- 1996) which also could be regarded as an ami-
bolites and the genes for LPS biosynthesis noglycoside indicates that the anabolic myo-
have frequently been suggested to be trans- inositol pathway is generally present in this
ferred horizontally (PIEPERSBERG,1993, taxonomic group, in contrast to other bacte-
1994; REEVES, 1993; SHNAITMANand ria. Therefore, the evolution of Ca type path-
KLENA, 1993; LIU and THORSON, 1994). ways could be based on a preferred metabolic
Therefore, the rfa genes and the secondary route in actinomycetes.
carbohydrate biosynthetic genes in actinomy- The protein similarities between StrS,
cetes represent an interesting basis for the StsA, and StsC are in the range of 25% iden-
study of the evolutionary links and dynamics tity and, therefore, too low to suggest a gene
in secondary, highly mobile gene pools. duplication event during the evolution of the
The biosynthesis pathway of streptidine streptomycin pathways which might explain
represents another interesting topic of evolu- the origin of these proteins. The occurrence
tionary studies. The bluensidine pathway, of a third possible aminotransferase is puzz-
e.g., was suggested to be ancestral to that of ling since it was thought that the third amino
streptidine (WALKER,1990). Amidinotrans- group introduced during the biosynthesis of
ferase activities catalyzing both reactions 6 streptomycin into NMLGA was derived via
and 11 of the streptidine pathway (cf. Sect. the primary metabolic pathway from the D-
4.1.1.1 and Fig. 14) were found in the bluen- glucosamine pool (GRISEBACH, 1978). If this
somycin producer S. hygroscopicus ssp. glebo- precursor is not formed de novo under condi-
sus ATCC 14607 (WALKER,1990). However, tions of streptomycin production another ex-
no strB2 gene is present in the two bluenso- planation for the need of an additional trans-
mycin producers, S. bluensk DSM 40564 and amination step would be the regeneration of
S. hygroscopicus ssp. glebosus DSM 40823 glutamine from a-ketoglutamine, the unusual
(MAYER,1994; G. MAYER,A. MEHLING, and by-product of step 4 in S D biogenesis. This
W. PIEPERSBERG,unpublished data.). In aminotransferase reaction is not normally ob-
these strains only one strB gene is conserved served in prokaryotes (WALKER,1975a). The
which clearly belongs to the sfrBl group, N-terminal sequence of the StrS protein is
since the adjacent gene downstream is strF in also identical to that of one of the proteins
both cases (see Fig. 13) and these genes en- expressed at high concentrations in the strep-
code proteins with a much higher amino acid tomycin production phase in S. griseus but
sequences identity (ca. 85%) relative to the not in the mutant M881 (see Tab. 2) (DIST-
StrBl than to StrB2 proteins of streptomycin LER et al., 1992).
7 Involvement of the production phase. In this context it is in-

teresting to note that amino acids seem to be
Primary Metabolism preferred or even essential nutrients in strep-
tomycin or neomycin producers (SHAPIRO,
in the Delivery of 1989; PIEPERSBERG, 1995).
Carbohydrate Precursors
The involvement of primary metabolic
traits in the activation of precursors, the dy-
namics and alterations of precursor pool
8 Pathway Engineering
sizes, and the use of nutritional sources for and Other Types of
precursor formation during the production
phase in the producers of aminoglycosides Application in
and other secondary carbohydrates has not
been studied intensively (SHAPIRO, 1989). Biotechnology
The regulation of streptomycin production
and its growth phase dependence has already The availability of the genes which encode
been discussed above (see Sect. 4.6). These enzymes necessary for the biosynthesis of sec-
aspects will become more accessible for inves- ondary metabolites recently has allowed the
tigation in the future when the genetic pro- initiation of experiments designed for the
grams for the biogenesis of representative production of new hybrid molecules (HOP-
members of this group of compounds are ana- WOOD et al., 1990; KATZ and DONADIO,
lyzed and available for manipulation in differ- 1993; several articles in VININGand STUT-
ent organisms. Biotechnological development TARD, 1995; PIEPERSBERG, 1994). For the ar-
and use of production strains in fermentation ray of aminoglycosides and other carbohy-
could then become more efficient; mathemat- drate-derived natural products this phase of
ical modeling of the producer’s physiology biotechnology has not yet been as successful
would also greatly facilitate the production as it was for the polyketides. The reasons for
processes. It is especially important to gain this are as follows: (1) the pathways involved
specific knowledge about the sources and the are composed of multiple steps catalyzed by
routes of delivery of the carbohydrate precur- highly specific and in general monofunctional
sors and the nitrogenous groups in the case of enzymes; (2) the biochemistry and chemistry
the aminoglycosides. The finding that com- of carbohydrates is complicated, and the in-
pounds such as neomycins and validamycins termediates of biosynthetic pathways are
are synthesized from a precursor pool related mostly quite instable and inaccessible to syn-
to the pentosephosphate (PP) cycle interme- thetic chemistry; and (3) only relatively few
diates (RINEHARTet al., 1992), and that the research groups are engaged in the investiga-
NDP-hexose forming enzymes are encoded tion of secondary carbohydrate metabolism.
by genes in the clusters for aminoglycosides Nevertheless, several fields of biotechnologi-
and other sugar-based secondary metabolites cal application of our increasing knowledge of
(see Sect. 4) in streptomycetes could help to the biochemical and genetic components in-
clarify these aspects. If hexose intermediates volved in the formation of activated and
are derived from the PP cycle mainly, this modified sugar or cyclitol molecules can be
raises the question of whether compartment- envisaged (cf. also PIEPERSBERG,1994): (1)
alization or a difference in the efficiencies of genetic engineering of pathways for improved
the enzymes involved in the conversion of the fermentations, with respect to nutrient con-
primary hexosephosphate are responsible for trol and incorporation, optimization of yield,
this phenomenon. Alternatively, the inver- and restriction of product patterns, or for the
sion of the major route(s) of intermediary C production of new hybrid end products in the
metabolism to the predominant use of the producers themselves (e.g., the hybrid, glyco-
gluconeogenetic pathway could occur during sylated tetracenomycins; DECKER et al.,
1995); (2) development of biotransformation aminobutyryl-dibekacin; cf. Fig. 28B); for

systems for the in vivo glycosylation of fed most of those modifications, natural counter-
aglycones; (3) in vitro enzymatic production parts exist as models.
of activated sugar derivatives and/or glycosyl- (2) Designed mixing of subpathways could
transfer to synthetic aglycones; or (4) transfer be envisaged. For example, first the Cb path-
of the genetic complement for complete pathways could be exchanged for Ca routes and
ways or branching subpathways to new host vice versa. In another line of experimental
systems and/or their redesign for improved models the exchange of the sugar modifying
production characteristics or completely new and transferring subpathways between pro-
metabolites. However, most of the prerequi- ducers of related groups of aminoglycosides
sites for the planned redesign of the pathways could be attempted, e.g., the PA pathways be-
for aminocyclitol aminoglycosides and related tween the producers of gentamicins (cf. Fig.
secondary carbohydrates are not yet availa- A14) and seldomycins (cf. Fig. A17) or the
ble. These prequisites include the complete 6DOH or H A pathways between the produc-
genetic and biochemical analyses of several ers of streptomycins (cf. Fig. Al), spectino-
key pathways and knowledge of substrate mycins (cf. Figs. A2 and A3), kasugamycins
specificities of key enzymes (especially the (cf. Figs. A4 and A5), and boholmycin (cf.
glycosyltransferases and other condensing en- Fig. A7).
zymes) and of the bottleneck steps or the flux (3) Some aminoglycosides are produced in
rates of individual precursors and interme- genera or strains which are not easily accessi-
diates through the pathway. ble for physiological or genetic manipulation
Some interesting goals for pathway engi- or which produce unwanted side products.
neering in aminocyclitol aminoglycoside pro- The ability to produce could, therefore, be
ducers are as follows: transferred to hosts which can be more easily
(1) Transfer of the production and transfer manipulated. For instance, the aminoglyco-
genes for the Nl-a-hydroxy-yaminobutyryl side production genes from Micromospora
moiety of butirosin from Bacillus circulans to spp. could be transferred to Streptomyces spp.
a kanamycin-producing Streptomyces kana- or even to more distantly related “GRAY-
myceticus in order to produce in vivo the organisms (GRAS = “generally regarded as
semisynthetic compound amikacin ( =N I - a - safe”) such as corynebacteria (cf. PIEPERS-
hydroxy-yaminobutyryl-kanamycin A), as al- BERG,1993).
ready suggested much earlier (Fig. 28A; (4) The regulation of production genes and
DAVIESand YAGISAWA,1983). This would the nutrient flow in aminoglycoside producers
require first that the N1-a-hydroxy-yamino- could be further targets of pathway engineer-
butyryltransferase also recognizes the kana- ing. Because of their complicated physiology
mycin A molecule as a substrate and in its and regulation in complex cell differentiation
correct amino acceptor group and, second, cycles (cf. Sect. 4.6) this would be of particu-
that the possible kanamycin exporter in S. ka- lar importance when continuous culture tech-
namyceticus transports the new end product niques are used in production.
with equivalent efficiency. Similarly, it might Another application of the new genetic and
be possible to find enzymic reactions for the biochemical data could be the development
designed modification of other aminoglyco- of genetic screening systems for the search for
sides, e.g., group transfers or dehydroxyla- a particular production ability. For this pur-
tions such as those used in the synthesis of pose, we have to find more genes for key
well-established semisynthetic chemothera- functions which are diagnostically relevant in
peutics with activity against bacteria with clin- the detection of a specific pathway. An exam-
ically important resistance patterns (cf. SHAW ple is the family of strE-related genes encod-
et al., 1993) and with reduced ototoxicity. Ex- ing the dTDP-glucose 4,6-dehydratases char-
amples are (1) netilmicin ( =l-N-ethylsisomi- acteristic for the 6DOH pathways (STOCK-
cin; cf. Fig. A14), (2) isepamicin (cf. Fig. 28B), MA” and PIEPERSBERG, 1992). This type Of
dibekacin ( =3 ’,4 ‘-dideoxykanamycin B; cf. diagnostic material, together with that typical
Fig. A12), and arbekacin ( =N,-a-hydroxy- y- of other chemical classes of secondary meta-
Butirosin A Amikacin
2
(= AHB)
OH
genetic engineering ?
B (= AHP) CH9NH9
H O O o ,:uH3
OH OH
NH-AC
CH2NH2
OH
lsepamicin Arbekacin (Habekacin)

(N1-AHP-3*-N-acetyIgentarnicinB) (N'-AHB-3',4'-dideoxykanamycin6)
Fig. 28. Semisynthetic aminoglycosides as models for the production of modified aminoglycosides by path-
way engineering. A Hypothetical production of the semisynthetic amikacin in a genetically engineered
kanamycin A producer; B structures of two other semisynthetic aminoglycosides, derivatives of gentamicin
B and kanamycin B, with good pharmacological properties and activity against multiple-resistant patho-
gens. The chemical groups that are introduced by semisynthetic additions (given in bold face) or deleted
(arrows) mostly mimic known modifications of related natural components in other molecules.
bolites, could be applied together with classi- groups (e.g., actinomycetes) an unwanted
cal screening methods for the detection of subgroup of producers has to be excluded. In
new compounds. This could be of interest if, the future, when sufficient and highly predic-
e.g., a leading structure is known and new tive pathway-specific gene probes become
derivatives related to that special target group available this method can also be used for a
are searched for, or if in particular microbial more rapid prescreening method.
9 Conclusions and respect to the more restricted group of ami-

nocyclitol aminoglycosides in particular, there
Perspectives still seems to be a potential for finding new
structural classes as is indicated by the recent
detection of new groups via target-directed
The knowledge of all the variants of sec- screenings for glycosidase inhibitors (e.g.,
ondary metabolites composed of (amino-)su- acarbose, trestatins, trehazolin, and allosami-
gars or their derivatives (secondary carbohy- dins). Other fields of application and new
drates) and their biosyntheses is still sparse. structures will also become available by con-
Evidence is accumulating, however, that the tinuing research efforts and the input of intel-
various pathways of aminocyclitol aminogly- ligence and scientific skills.
coside production share several common fea-
tures and that a common gene pool is used Acknowledgements
for the modular design of these pathways. It We thank all our collaborators for their en-
can be predicted that the application of mo- thusiastic participation in aminoglycoside re-
lecular genetics and biochemistry on several search. The work on secondary carbohydrate
production systems for secondary carbohy- genetics and metabolism in the laboratory of
drates will bring some unification to several the authors was generously supported by the
aspects of this field of research. Concomitant- Deutsche Forschungsgemeinschaft, the Bun-
ly, also an intensification of applied research desministerium fur Forschung und Technolo-
on secondary carbohydrates will arise, since gie, the European Commission, and the phar-
these are components in many applied natu- maceutical companies Bayer AG and
ral products where they are essential for Hoechst AG.
bioactivity. Thus, the data reported in this
chapter will influence other fields of research
and development, especially of other chemi-
cal groups of low-molecular weight bioactive
molecules or of polysaccharides and glyco- 10 Appendix
conjugates. Also, the research and develop-
ment in the field of chemo-enzymatic synthe- (Chemical Structures)
sis will be fertilized by providing the bio-
chemical tools such as enzymes and their sub- The figures compiled in the Appendix
strates (e.g., NDP-activated sugars). The basis (Figs. Al-A31) summarize the chemical fami-
for pathway engineering for the designed pro- lies of secondary carbohydrates mentioned in
duction of new variants of secondary meta- the text; their “pathway formulae” are indi-
bolites will also be available very soon. With cated in the legend (cf. Sect. 3.2).
10 Appendix (Chemical Structures) 463
NH
II
streptidine
(bluensidine)
OH
H
6
streptose
N-methyl-L-glucosamine
(demethyl-, 4"-mannosido-,
4"-dimannosido-,4"-ahimosido-,
N-glycolyI-)
OH
R' R2 R3 Ft I+ Ff
streptomycin NH-CNH-NH2 CHO CH3 CH3 H

dihydrostreplomyan NH-CNH-NH2 CHP-OH CHI CH3 H
5'-hydroxystreptomycin NH-CNH-NH, CHO CH2-OH CH3 H
Ndemelhylstreptomycin NH-CNH-NH2 CHO CH3 H H
5'-hydroxy-N-demethyI-
dihydroslreptomycin NH-CNH-NH2 CH2-OH CH2-OH H H H
mannosido-
5'-hydroxyslreptomycin NH-CNH-NH2 CHO CHp-OH CH3 H a-Dmannose (= DM)
dimannosidcstreptomycin NH-CNH-NH2 CHO CH3 CH3 H a-DM-1.6-a-DM
Muensomycin 0-CO-NH2 CH2-OH CH3 CH3 H H
ashmycin A NH-CNH-NH2 CHO CH3 CH3 H 2" '-carboxy-xylo-
furanose (ashimose)
ashmyan 8 NH-CNH-NH2 CHO CH3 CH, CO-CH2-OH a-DM-1,Ga-DM
AC4437 = 5'-hydroxystreplomycin lacking NMLGA
~ ~~~ ~
Fig. Al. Streptomycins; basic pathway formula Ca(4)-6DOH(2)-HA.

u
Fig. A2. Spectinomycin;Ca(4,5)-6DOH.
iH
I
H3C-YH OH CH3
Fig. A7. Boholmycin; HA-(4)Ca(6)-PA(4)-Hep.
Fig. A3. Spenolimycin; Ca(4,5)-6DOH.
HNboo
I
HN=C
I
E
OH
H
HN=C
AH2
I
OH
Fig. AS. Myomycins; Ca(4)-HA.

COOH
Fig. A4. Kasugamycin;Ca(4)-6DOH.
NH
@2
2 NH-CO-CH-NH-CO-NH
HOOC
HZN
OH OH
Fig. AS. Minosanimomycin; Ca(4)-6DOH. Fig. A9. Hygromycin A; Ca(l)-[X]-6DOH.
bH
L-Arg
Fig. A6. LL-BM132a; Ca(4)-H(4)-HA.

d6
A
H3C0
H3CN OH H3CHN OH
I
Lysinomicin(Ca(G)-HA)
A fortimicin A NHp H OH COCH2NHp CH3 H

fortimicin B NHp H OH H CH3 H
1-epkfortimicinB H NHp OH H CH3 H
fortimicin C NH, H OH COCHpNHCONHp CH3 H
fortimicin D NHp H OH COCHpNH2 H H
dactimicin NHp H OH COCHpNHCH=NH CH3 H
1-epi.dactimicin H NHp OH COCHzNHCH=NH CH3 H
sporaricin A H NHp H COCHpNHp CH3 H
sporaricin B H NH2 H H CH3 H
istamycinA, NHp H H H H CH3
istamycinA NHp H H COCHpNHp H CH3
(= sannamycin)
istamycin A3 NHp H H COCHpNHCH=NH H CH3
istamycinBo H NHp H H H CH3
istamycin B H NHp H COCHpNHp H CH3
istamycin B3 H NH, H COCHpNHCHsNH H CH3
istamycin C NHp H H COCHpNHp H CHpCH3
istamycin A2 NHp H H COCHpNHCONHp H CH3
fortimicin KG3 = 4',5'-dehydrofortimicin A
B fortimicin KH OCH3 H OH CH3

fortimicin KR H OCH3 OH CH3
istamycinYo OCH3 H H H
istamycinXo H OCH3 H H
~ ~~~~
Fig. A10. Fortimicinhstamycin family; Ca(6)-HA or Cb(6)-HA.

1”’
HzN
R’ R2 R3 R4 R5
neomycin B OH NH2 H CH2NH2 H

neornycin C OH NH2 H H CH2NH2
paromornycin I OH OH H CH2NHz H
parornomycin II OH OH H H CHzNHz
mannosylparornomycin OH OH a-D-Man CHzNHz H
lividomycin A H OH a-D-Man CHzNHz H
lividomycin B H OH H CH2NHz H
~
Fig. All. Neomycin family; HA-(4)Cb(S)-P(3)-HA.

I0 Appendix (Chemical Structures) 467
Q
CH2R3
HO
R’O
I
R’ R2 R3 R4 R5 R6
kanamycin OH OH NHp NH2 OH H

kanamydn B NHp OH NHp NHp OH H
kanamydn C NH? OH OH NHp OH H
amikacin OH OH NH2 NHp OH ahb
NK-1001 OH OH NH2 OH OH H
NK-1012-1 NH2 OH NHp OH OH H
tobramycin NH2 H NHp NHp OH H
Fig. Al2. Kanamycin family; nebramYCin NH2 OH NH2 NHp OCONHp H
HA-(4)Cb(6)-HA. nebramycin5’ NH2 H NHp NHp OCONHp H
R4 OH
R’ R2 R3 R4 R5
butirosin A ahb OH NH2 H OH
butirosin B ahb OH NH2 OH H
butirosin El ahb OH OH H OH
butirosin E2 ahb OH OH OH H
butirosin C1 ahb H NH2 H OH
butirosin Cp ahb H NH2 OH H
Fig. A13. Butirosinlribos- ribstamycin H OH W 2 OH H
tamycin family; HA- xylostatin H OH NH2 H OH
(4)Cb(S)-P. LL-BM 408a H OH OH OH H
gentamicinA NH2 H OH NHCH3 OH H

gentamicinAl NHz H OH NHCH3 H OH
gentamkin4 NHz H OH OH OH H
gentamicinA3 OH H N H ~ NHCH, H OH
gentamkin& NH, H OH NCH3CH0 OH H
gentamkinB OH H NH, NHCH3 CH3 OH
gentamicinB, OH CH3 NH, NHCH3 CH3 OH
gentamicinXz NH2 H OH NHCH3 CH3 OH
G-418 NHz CH3 OH NHCH3 CH3 OH
JI-20A NHz H NH2 NHCHj CH3 OH
JI-206 NH, CH3 NHZ NHCH3 CH3 OH
0 R'I R'z R'3 R'4 R'5 R'6
gentamicin C1 CH3 H CH3 CH3 OH -

gentamicin C1, H H H CH3 OH -
gentamicin C2 CH3 H H CH3 OH -
gentamicin CZa H CH3 H CH3 OH -
sagamicin H H CH3 CH3 OH -
sisomicin H H H CH3 OH +
verdamicin CH3 H H CH3 OH +
G-52 H H CH3 CH3 OH +
66-408 H H H OH H +
66-40D H H H H OH +
Fig. A14. Gentamicin family; HA-(4)Cb(6)-PA.

R1 R2
Rl R2 R3 R4
apramycin H NH2 H2N
oxyapramycin OH NH2 seldomycin 1 OH OH OH OH
saccharwin H OH seldomycin3 OH OH
OH NH2
seldomycin 5 H NH2 NH2 OCH3
Fig. A15 Apramycin family; Cb(4)-OctA(8)-HA.
Fig. A17. Seldomycin family; HA-(4)Cb(6)-PA.
OH
hygromycin B CH3 H OH H OH H
destomycin A H H O H H OH H
destomycin B CH3 H H OH H OH
destomycin C CH3 H OH H OH H
A-396-1 H H O H H OH H
A-163164 CH3 H H OH OH H
ss-56c H OH OH H OH H
Fig. A16. Hygromycin Bldestro- 1-N-amidino-I-N- H OH OH H OH H
mycin family; Cb(S)-H(2,3)-HepA demethyl-2-hydroxy-
or Ca(S)-H(2,3)-HepA. destomycin A
NH-Ac
siastatin
HGoH
Ri RP R3 R4
nojirimycin A OH OH H H
mannonojirimycin OH H OH H
ldeoxynojirimycin (DNJ) H OH H H
1-deoxymannonojirimycin H H OH H
miglitol H OH H (CH,),OH
emiglitate H OH H (CH2)20R
OH
(R = -gbenzoic acid ethylester)
galactostatin
Hb;;
y- 42
CHgH HmH2 OH NHCOCHCH3 CH2OH
0
HO 0 . QOH HoQoH
OH
3-amino-3-deoxy- cv-1
do prumycin
NH2 NHCOyCH3
streptozotocin
No
Pglucose
Fig. A18. Monomeric aminoglycosides which are products of HA or PA pathways.
trehalosamine H OH NH2 OH
mannosylglucosaminide OH H NH2 OH
4-amino-4-deoxytrehalose H OH OH NH2
3,3'-neotrehalosadiamine
RP (BMY-28251)
OH
Fig. A19. Trehalosamine family; HA(1)-H or HA(1)-HA.
CH2OH CH,OH
Ri R2 R3 R4 Ri R2
T-l H NH2 H OH S-l H NH2
T-ll NH2 H H OH S-ll NH2 H
T-Ill NH2 H OH H
Fig. A20. Trehalosamine-related synthetic disaccharides.
validoxylamine A H H H H H H
validoxylamine B H H OH H H H
validoxy\amine G OH H H H H H
validamycin A H H H P-D-GlC H H
validamycin B H H OH P-D-GlC H H
validamycin C H H H P-D-GlC a-D-Glc H
validamycin D H a-D-Gk H H H H
validamycin E H H H a-D-Glc(l,4) H H
-P-D-Glc
validamycin F H H H P-D-GlC H a-D-Glc
validamycin G OH H H P-D-GlC H H
validamycin H H H H a-D-Glc(1.6) H H
-&D-Glc
CHpOH CH20H CHpOH HOCH2 OH
valienamine validamine valiolamine 2-hydroxy-

validamine
Fig. A21. Validamycin family; Cb(l)-Cb(4)-H. Valiolamine (Cb) was also found as a separate end prod-
uct.
0
CH20H
R3-O NH 0-R2
OH OH
acarbose H a-l.4-maltosyl H, or (a-D-glucopyranosyl),

(amylostatins)
adiposins OH a-l,4-maltosyl H. or (a-D-glucopyranosyl),
trestatin H a-1,Gmahosyl- H. or (core pseudo-
Fig. A22. A carbose/amylostatin family
a-1,rl-trehalosyI trisaccharide), of glycosidase inhibitors; (H),,-Cb(1)-
2.3-epoxyderivatives of acarbose 6DOH(l)-(H). or (H).-Cb(1)-H(1)-
oligostatins= 2,3-dihydro-2-hydroxyderivativesof acarbose
bI= ~
HO o ~ ;
0
0:
HO
OH NH OH NH
HO NH I I
Ac Ac '';J-Rl
I
allosamizoline
R2 ,
(RI = R2 = CH3)
OH R1 R2
Fig. A23. Trehazolin; Ca(l,2)-HA. allosamidin CH3 CH3

demethylallosamidin CH3 H
didemethylallosamidin H H
Fig. A25 Allosamidins; Ca(4)-HA(4)-HA.
co OH
OH
I
R CH20H
sorbistin Al Ethyl
sorbistin A2 Propyl
sorbistin B Methyl
sorbistin D H
Fig. A24. Sorbistins; Ca(2)-HA.

p5
OH
lincomycin A CH3 CH2-CHz-CH3

lincomycin B CH3 CH2-CH3
clindamycin CH3 CHz-CHz-CH3
N-demethyl-
lincomycin A CH3 CH2-CHz-CH3
celesticetin (CHz)z-O-salicylyl H
desalicetin (CHz)z-OH H
celesticetin B (CH2)2-0-isoburyryl H
celesticetin C (CH2)z-O-anthranilyl H
Fig. A26. Lincosamides; HA or celesticetin D (CH,),-O-acetyl H
OctA.
b
HOH2C g h
HO
I OH
Ph-0
OH
Adenophostin A
(Ad = Adenin;
Adenomycin Ph = P03H2)
(P-(l)Ca(3)HA) (P(3FH)
Cytosaminomycinr
(R = acyl residues)
(6DOH(4)6DOH) ro\
CH~
(D-amosamine) OH
(3-h ydroxy-
mOecH2
Damicetose)
" O H 2 Y Y OCH3
CH20H
0 - I H2NOC9~0\~=
A201C
(Ad = Adenin)
(PA(3)-[Xl-H(6)-6DOH)
I
streptothricins
(R = deazapurin)
a
(HA)
Ad Ad COOH
HOH2C
- H 2 C f j J - - ( NH - R
OH OH OH OH
Aristeromycin Neplanocin A Blasticidin S

(Ad = Adenin) (Ad = Adenin) (Cyl = cytosin; R =
(Ca or Cb)
(Ca or Cb) y-N-methyl-P-arginine)
(HA)
Fig. A27. Nucleoside-type secondary metabolites with unusual sugar components or cyclitols.
HQ0 HO$
-(
CH3 OH
2,3,4-trihydroxy-6- cyclophellitol
methylcyclohexanone
Fig. A28. Natural neutral cyclitols; Ca or Cb?
OH
Fig. AM. Pactamycin; Ca?
Granaticin
mitomycinA OCH3CH3 H
mitomycinB OW3 H CH3
mitornycin C NHz CH3 H
porfiromycin NHz CH3 CH3
(thick line, bold face = D-glucosamineunit)
Fig. A31. Mitomycins.
K-252a Rebeccamycin
Fig. A29. Granaticin, staurosporine, and staurospo-

rine-related compounds. Unusual bonds in the link-
age of 6DOH components in microbial secondary
metabolites. C-glycosidic binding in the polyketide
granaticin and N-glycosidic binding in the indolo-
carbazole alkaloids staurosporine, K-252a (same
aglycone as staurosporine), and rebeccamycin
which are all produced by actinomycetes.
11 References of antibiotic resistant bacteria, Proc. Natl. Acad.

Sci. USA 70,2276-2280.
BEYER,S., DISTLER,J., PIEPERSBERG, W. (1996),
ANDO,O., SATAKE,H., ITOI,K., SATOA., NAKA- The str gene cluster for the biosynthesis of 5'-
JIMA,M., TAKAHASHI, S., HARUYAMA, H., OH- hydroxystreptomycin in Streptomyces glauces-
KUMA,Y., KINOSHITA, T., ENOKITA, R. (1991), cens GLA.0 (ETH 22794): new operons and
Trehazolin, a new trehalase inhibitor, J. Anti- evidence for pathway-specific regulation by
biot. 44,1165-1168. StrR, Mol. Gen. Genet. 250,775-784.
ANGELL,S., SCHWARZ,E., BIBB,M. (1992), The BILLINGTON, D. C. (1993), The Inosotol Phos-
glucose kinase gene of Streptomyces coelicolor phates. Weinheim: VCH.
A3(2): its nucleotide sequence, transcriptional BLAZQUEZ,J., DAVIES,J., MORENO,F. (1991),
analysis and role in glucose repression, Mol. Mi- Mutations in the aphA-2 gene of transposon Tn5
crobiol. 6, 2833-2844. mapping with the regions highly conserved in
ASANO,N., KATAYAMA, K., TAKEUCHI,M., Fu- aminoglycoside-phosphotransferases strongly re-
RUMOTO, T., KAMEDA, Y., MATSUI,K. (1989), duce aminoglycoside resistance, Mol. Microbiol.
Preparation of 3-amino-3-deoxy derivatives of 5, 1511-1518.
trehalose and sucrose and their activities, J. An- BRAHME, N. M., GONZALEZ, J. E., ROLLS,J. P.,
tibiot. 42, 585-590. HESSLER,E. J., MIZSAK,S., HURLEY,L. H.
ATSUMI,S., IINUMA, H., NOSAKA,C., UMEZAWA, (1984a), Biosynthesis of the Lincomycins. 1.
K. (1990a), Biological activities of cyclophellitol, Studies using stable isotopes on the biosynthesis
J. Antibiot. 43, 1579-1585. of the propyl- and ethyl-L-hygric acid moieties of
ATSUMI,S., UMEZAWA, K., IINUMA, H., NAGANA- lincomycins A and B, J. Am. Chem. SOC. 106,
WA,H., NAKAMURA, H., IITAKA,Y., TAKEU- 7873-7878.
CHI, T. (1990b), Production, isolation and struc- BRAHME,N. M., GONZALEZ,J. E., MIZSAK,S.,
ture determination of a novel Pglucosidase inhi- ROLLS, J. R., HESSLER,E. J., HURLEY,L. H.
bitor, cyclophellitol, from Phellinus sp., J. Anti- (1984b), Biosynthesis of the Lincomycins. 2.
biot. 43, 49-53. Studies using stable isotopes on the biosynthesis
AUBERT-PIVERT, E., DAVIES,J. (1994), Biosynthe- of the methylthiolincosaminide moiety of Linco-
sis of butirosin in Bacillus circulans NRRL mycin A, J. Am. Chem. SOC. 106,7878-7883.
B3312 identification by sequence analysis and BRILL,G. M., MCALPINE,J. B., WHITTERN, D. N.,
insertional mutagenesis of the butB gene in- BUKO,A. M. (1990), Altromycins, novel plura-
volved in antibiotic production, Gene 147, 1-11. mycin-like antibiotics. 11. Isolation and elucida-
AWATA,M., MUTO,N., HAYASHI, M., YAGINU- tion of structure, J. Antibiot. 43, 229-237.
MA, S. (1986), A new aminoglycoside antibiotic, BRINK,M. F., BRINK,G., VERBEET,M. P., DE-
substance AC4437, J. Antibiot. 39, 724-726. BOER, H. A. (1994), Spectinomycin interacts
BAKKER,E. P. (1992), Aminoglycoside and amino- specifically with the residues G1064 and C1192
cyclitol antibiotics: hygromycin B is an atypical in 16s rRNA, thereby potentially freezing this
bactericidal compound that exerts effects on molecule into an inactive conformation, Nucleic
cells of Escherichia coli characteristic for bacteri- Acids Res. 22, 325-331.
ostatic aminocyclitols, J. Gen Microbiol. 138, BRYAN,L. E., KWAN,S. (1983), Roles of ribosomal
563-569. binding, membrane potential, and electron
BASAK,K., MAJUMDAR, S. K. (1973), Utilization transport in bacterial uptake of streptomycin
of carbon and nitrogen sources by Streptomyces and gentamicin, Antimicrob. Agents Chemother.
kanamyceticus for kanamycin production, An- 23,835-845.
timicrob. Agents Chemother. 4, 6-11. BUSSE, H.-J., WOSTMANN,K., BAKKER,E. P.
BASCARAN,V., SANCHEZ,L., HARDISSON,C., (1992), The bactericidal action of streptomycin:
BRANA,A. F. (1991), Stringent response and in- membrane permeabilization caused by the inser-
itiation of secondary metabolism in Streptomyces tion of mistranslated proteins into the cytoplas-
clavuligerus, J. Gen. Microbiol. 137, 1625-1634. mic membrane of Escherichia coli and subse-
BEAU, J.-M. (1990), Polyene macrolides: stereo- quent caging of the antibiotic inside the cells due
structural elucidation and synthetic studies of a to degradation of the protein, J. Gen Microbiol.
few members, in: Recent Progress in the Chemi- 138,551-556.
cal Synthesis of Antibiotics (LUKACS,G., OHNO, BUTTNER,M. J. (1989), RNA polymerase hetero-
M., Eds.), pp. 135-182. Berlin: Springer-Verlag. geneity in Streptomyces coelicolor A3(2), Mol.
BENVENISTE, R., DAVIES,J. (1973), Aminoglyco- Microbiol. 3, 1653-1659.
side antibiotic inactivating enzymes in actinomy- CHADWICK, D. J., WHELAN,J. (Eds.) (1992), Sec-
cetes similar to those present in clinical isolates ondary Metabolites: Their Function and Evolu-
I1 References 477
tion, Ciba Foundation Symposium 171. Chiches- plified antibiotic resistance genes, J. Antibiot. 39,
ter: John Wiley & Sons. 128-135.
CHAMPNESS, W. C., CHATER,K. F. (1994), Regula- CUNDLIFFE, E. (1989), How antibiotic-producing
tion and interaction of antibiotic production and organisms avoid suicide, Ann. Rev. Microbiol.
morphological differentiation in Streptomyces 43,207-233.
spp., in: Regulation of Bacterial Differentiation CUNDLIFFE, E. (1990), Recognition sites for anti-
(PIGGOT, P., MORANJR., c. P., YOUNGMAN, P., biotics within rRNA, in: The Ribosome. Struc-
Eds.), pp. 61-94. Washington, DC: American ture, Function, and Evolution (HILL, W. E.,
Society for Microbiology. DAHLBERG,A., GAR RE^, R. A,, MOORE,P.
CHATER,K. F. (1989), Multilevel regulation of B., SCHLESSINGER, D., WARNER,J. R., Eds.),
Streptomyces differentiation, Trends Genet. 5, pp. 479490. Washington, DC: American Society
372-377. for Microbiology.
CHATER,K. F. (1992), Genetic regulation of sec- CUNDLIFFE,E. (1992a), Self-protection mecha-
ondary metabolic pathways in Streptomyces, in: nisms in antibiotic producers, in: Secondary Me-
Secondary Metabolites: Their Function and Evo- tabolites: Their Function and Evolution, Ciba
lution (DAVIES,J., CHADWICK, D., WHELAN,J., Foundation Symposium I71 (CHADWICK, D,
WIDDOWS,K., Eds.), pp. 144-156. Chichester: WHELAN,J., Eds.), pp. 199-214. Chichester:
John Wiley & Sons. John Wiley & Sons.
CUNDLIFFE, E. (1992b), Glycosylation of macrolide
CHEN,Y.-M., WALKER, J. B. (1977), Transamina-
tions involving keto- and amino-inositols and antibiotics in extracts of Streptomyces lividans,
glutamine in actinomycetes which produce gen- Antimicrob. Agents Chemother. 36,348-352.
tamicin and neomycin, Biochem. Biophys. Res. DAIRI,T., HASEGAWA, M. (1989), Common bio-
Commun. n,688-692. synthetic feature of fortimicin-group antibiotics,
J. Antibiot. 42, 934-943.
CHIAO,J. S., XIA,T. H., MEI,B. G., JIN,Z. K., Gu, DAIRI,T., OHTA,T., HASHIMOTO, E., HASEGA-
W. L. (1995), Rifamycin SV and related ansamy- WA,M. (1992a), Self cloning in Micromonospora
cins, in: Biochemistry and Genetics of Antibiotic olivasterospora of fms genes for fortimicin A (as-
Biosynthesis (VINING,L., S T U ~ A R D C.,, Eds.), tromicin) biosynthesis, Mol. Gen. Genet. 232,
pp. 477498. Boston, MA: Butterworth-Heine- 262-270.
mann.
DAIRI,T., OHTA,T., HASHIMOTO, E., HASEGA-
CHUNG,S. T., CROSE,L. L. (1990), Transposon WA,M. (1992b), Organization and nature of for-
Tn4556-mediated DNA insertion and site-di- timicin A (astromicin) biosynthetic genes stud-
rected mutagenesis; in: Proc. GIM 1990, Vol. I ied using a cosmid library of Micromonospora
(HESLOT,H., DAVIES,J., FLORENT,J., BOBI- olivasterospora DNA, Mol. Gen. Genet. 236, 39-
CHON,L., DURAND, G., PENASSE,L., Eds.), pp. 48.
207-218. Paris: SociCtC FranGaise de Microbio- DAIRI, T., YAMAGUCHI, K., HASEGAWA,M.
logie. (1992c), N-formimidoyl fortimicin A synthase, a
COLEMAN, R. S., FRASER, J. R. (1993), Acylketene unique oxidase involved in fortimicin A biosyn-
[4 +2]cycloadditions: divergent de novo synthe- thesis: purification, characterization and gene
sis of 2,6-dideoxy sugars, J. Org. Chem. 58, 385- cloning, Mol. Gen. Genet. 236, 49-59.
392. DAVIES, J. (1992), Another look at antibiotic resist-
COOK,G. R., BEHOLZ,L. G., STILLE,J. R. (1994), ance, J. Gen. Microbiol. 138, 1553-1559.
Construction of hydroxylated alkaloids ( + / -)- DAVIES,J., SMITH,D. I. (1978), Plasmid-deter-
mannonolactam, ( + )-deoxymannojirimycin, and mined resistance to antimicrobial agents, Ann.
( +/-)-prosopinine through aza-annulation, J. Rev. Microbiol. 32, 469-518.
Org. Chem. 59,3575-3584. DAVIES,J., YAGISAWA, M. (1983), The aminocycli-
COURVALIN, P. (1994). Transfer of antibiotic re- to1 glycosides (aminoglycosides), in: Biochernis-
sistance genes between gram-positive and gram- try and Genetic Regulation of Commercially Im-
negative bacteria, Antimicrob. Agents Chemo- portant Antibiotics (VINING, L. C., Ed.), pp. 311-
ther. 38, 1447-1451. 354. London: Addison-Wesley Publishing Com-
COURVALIN, P., WEISBLUM, B., DAVIES,J. (1977), pany.
Aminoglycoside-modifying enzyme of an anti- DAVIES,J., CANNON, M., MAUER,M. B. (1988),
biotic-producing bacterium acts as a determinant Myomycin: mode of action and mechanism of re-
of antibiotic resistance in Escherichia coli, Proc. sistance, J. Antibiot. 41, 366-372.
Natl. Acad. Sci. USA 74, 999-1003. DAVIS,B. D. (1987), Mechanism of bactericidal ac-
CRAMERI, R., DAVIES,J. E. (1986), Increased pro- tion of aminoglycosides,Microbiol. Rev. 51,341-
duction of aminoglycosides associated with am- 350.
DAVIS,B. D., CHEN,L., TAI,P. C. (1986), Misread J., FLORENT,J., BOBICHON, L., DURAND,G.,
protein creates membrane channels: an essential PENASSE,L., Eds.), pp. 379-392. Paris: Societ6
step in the bactericidal action of aminoglyco- FranGaise de Microbiologie.
sides, Proc. Natl. Acad. Sci. USA 83, 6164- DISTLER,J., MANSOURI, K., MAYER,G., STOCK-
6168. MANN, M., PIEPERSBERG, W. (1992), Streptomy-
DE STASIO,E. A., MOAZED,D., NOLLER,H. F., cin production and its regulation, Gene 115,105-
DAHLBERG, A. E. (1989), Mutations in 16s ri- 111.
bosomal RNA disrupt antibiotic-RNA interac- DREPPER,A., PAPE,H. (1996), Acarbose 7-phos-
tions, EMBO J. 8, 1213-1216. photransferase from Actinoplanes sp.: purifica-
DECKER,H., HAAG,G., UDVARNORKI, G., ROHR, tion, properties, and possible physiological func-
J. (1995), Novel genetically engineered tetrace- tion, J. Antibiot. 49, 664-668.
nomycins, Angew. Chem. (Int. Ed Engl.) 34, EMBLEY,T. M., STACKEBRANDT, E. (1994), The
1107-1 110. molecular phylogeny and systematics of the acti-
DEGWERT,U., VAN HULST, R.,PAPE, H., HER- nomycetes, Ann. Rev. Microbiol. 48,257-289.
ROLD,R. E., BEALE,J. M., KELLER,P. J., LEE, ENGLUND, P. T. (1993), The structure and biosyn-
J. P., FLOSS,H. G. (1987), Studies on the synthe- thesis of glycosyl phosphatidylinositiol protein
sis of the a-glucosidase inhibitor acarbose: val- anchors, Annu. Rev. Biochem. 62, 112-138.
ienamine, a m-C7N unit not derived from the ENSIGN,J. C. (1988), Physiological regulation of
shikimate pathway, J. Antibiot. 40,855-861. sporulation of Streptomyces griseus, in: Biology
DEMAIN A. L. (1989), Carbon source regulation of of Actinomycetes ’88 (OKAMI,Y., BEPPU,T.,
idiolite biosynthesis in actinomycetes, in: Regul- OGAWARA, H., Eds.), pp. 309-315. Tokyo: Ja-
ation of Secondary Metabolism in Actinomycetes pan Scientific Societies Press.
(SHAPIRO,S., Ed.), pp. 127-134. Boca Raton,
ETIENNE, G., ARMAU, E., DASSIN,M., TIRABY, G.
FL: CRC Press. (1991), A screening method to identify antibiot-
DEMAIN, A. L., INAMINE, E. (1970), Biochemistry ics of the aminoglycoside family, Antibiot. 44,
and regulation of streptomycin and mannosido- 1357-1366.
streptomycinase (a-D-mannosidase) formation,
Bacteriol. Rev. 34, 1-19. FISHER, S. H. (1992), Glutamine synthesis in Strep-
tomyces - a review, Gene 115, 13-17.
DICKENS, M. L., YE, J., STROHL,W. R. (1996),
Cloning, sequencing and analysis of aklavike- FLORES,M. E., PONCE,E., RUBIO,M.,HUITRON,
tone reductase from Streptomyces sp. strain C5, C. (1993), Glucose and glycerol repression of a-
J. Bacteriol. 178, 33843388. amylase in Streptomyces kanamyceticus and iso-
DISTLER,J., PIEPERSBERG, W. (1985), Cloning and lation of deregulated mutants, Biotechnol. Lett.
characterization of a gene from Streptomyces gri- 15,595-600.
seus coding for a streptomycin-phosphorylating FLOSS,H. G. (in press), Biosynthesis of Antibiotics,
activity, FEMS Microbiol, Lett. 28, 113-117. in: Proc. 6th Con$ Genetics and Molecular Bio-
DISTLER,J., KLIER,K., PIENDL,W., WERBITZKI, logy of Industrial Microorganisms, Bloomington,
0.. BOCK, A., KRESZE, G., PIEPERSBERG, W. IN, Oct. 1996 (HEGEMANN, G. D; Baltz, R. H.,
(1985), Streptomycin biosynthesis in Strepto- BALTZ,R. H., Skatrund, P. L., Eds.).
myces griseus. I. Characterization of streptomy- FLOSS,H. G., BEALE,J. M. (1989), Investigation of
cin-idiotrophic mutants, FEMS Microbiol. Lett. the biosynthesis of antibiotics, Angew. Chem.
30,145-150. (Int. Edn.) 28, 146-177.
DISTLER, J., BRAUN,C., EBERT,A., PIEPERSBERG, FOSTER,T. J. (1983), Plasmid-determined resist-
W. (1987a), Gene cluster for streptomycin bio- ance to antimicrobial drugs and toxic metal ions
synthesis in Streptomyces griseus: Analysis of a in bacteria, Microbiol. Rev. 41, 361-409.
central region including the major resistance FRAIMOW, H. S., GREENMAN, J. B., LEVITON,I.
gene, Mol. Gen. Genet. 208,204210. M., DOUGHERTY, T. J., MILLER,M. H. (1991),
DISTLER,J., EBERT,A., MANSOURI,K., PISSO- Tobramycin uptake in Escherichia coli is driven
WOTZKI,K., STOCKMANN, M., PIEPERSBERG, by either electrical potential or ATP, J. Bacteri-
W. (1987b), Gene cluster for streptomycin bio- 01. 173,2800-2808.
synthesis in Streptomyces griseus: nucleotide se- FRENCH,J. C., BARTZ,Q. R., DION,H. W. (1973),
quence of three genes and analysis of transcrip- Myomycin, a new antibiotic, J. Antibiot. 26,272-
tional activity, Nucleic Acids Res. 15, 8041- 283.
8056. FUJITA,Y., SHINDO, K., MIWA,Y., YOSHIDA,K. I.
DISTLER,J., MAYER,G., PIEPERSBERG, W. (1990), (1992), Bacillus subtilis inositol dehydrogenase-
Regulation of biosynthesis of streptomycin, in: encoding gene (idh): sequencing and expression
Proc. GIM 1990, Vol. 1 (HESLOT,H., DAVIES, in Escherichia coli, Gene 108, 121-125.
I1 References 479
GALE, E. F., CUNDLIFFE,E., REYNOLDS, P. E., tion of biosynthetic genes of fortimicin A (as-
RICHMOND, M. H., WARING, M. J. (1981), The tromicin), Actinomycetol. 5, 126131.
Molecular Basis of Antibiotic Action, 2nd Edn. HASEGAWA, M. (1992), A novel, highly efficient
London: Wiley Interscience. gene-cloning system in Micromonospora applied
GODA,S. K., AKHTAR,M. (1992), Neomycin bio- to the genetic analysis of fortimicin biosynthesis,
synthesis: the incorporation of D-6-deoxyglucose Gene 115,85-91.
derivatives and variously labelled glucose into HEINZEL,P., WERBITZKY, O., DISTLER,J., PIE-
the 2-deoxystreptamine ring. Postulated involve- PERSBERG, w. (1988), A second streptomycin
ment of 2-deoxyinosose synthase in the biosyn- resistance gene from Streptomyces griseus codes
thesis, J. Antibiot. 45, 984-994. for streptomycin-3” -phosphotransferase. Rela-
GOEKE,K. (1986), Enzymatische Untersuchungen tionship between antibiotic and protein kinases,
zum Zuckerstoffwechsel und zur Biosynthese Arch. Microbiol. 150, 184-192.
des a-Glucosidase-Inhibitors Acarbose bei Acti- HILL, W. E., DAHLBERG, A., GARRETT,R. A.,
noplanes spec., Dissertation, University of Mun- MOORE,P. B., SCHLESSINGER, D., WARNER, J.
ster. R. (1990), The Ribosome. Structure, Function,
GORINI,L. (1974), Streptomycin and misreading of and Evolution. Washington, DC: American So-
the genetic code, in: Ribosomes (NOMURA, M., ciety for Microbiology.
TISSIERES, A., LENGYEL, P., Eds.), pp. 791-803. HIROSE-KUMAGAI, A., YAGITA,A., AKAMATSU,
New York Cold Spring Harbor Laboratory. N. (1982), UDP-N-methyl-D-glucoseamine-phos-
GOULD,S. J. (1992), Exploring the intricate details phate. A possible intermediate of N-methyl-L-
of antibiotic biosynthesis, in: Secondary-Metab- glucosamine moiety of streptomycin, A . Antibiot
olite Biosynthesis and Metabolism (PETROSKI, R. 35, 1571-1577.
J., MCCORMICK, S. P., Eds.), pp. 11-25. New HOLMES,D. J., DROCOURT, D., TIRABY, G., GUN-
York: Plenum Press. DIFFE, E. (1991), Cloning of an aminoglycoside
GRAFE,U. (1992), Biochemie der Antibiotika. Hei- resistance-endodinggene, kame, from Saccharo-
delberg: Spektrum Akademischer Verlag. polispora hirsuta: Comparison with kamB from
GRISEBACH, H. (1978), Biosynthesis of sugar com- Streptomyces tenebrarius, Gene 102, 19-26.
ponents of antibiotic substances, Adv. Carbo- HOLT, T. G., CHANG,C., LAURENTWINTER, C.,
hydr. Chem. Biochem. 3581-126. MURAKAMI, T., GARRELS, J. I., DAVIES,J. E.,
GUILFOILE, P. G., HUTCHINSON, C. R. (1991), A THOMPSON, C. J. (1992), Global change in gene
bacterial analogue of the mdr gene of mammal- expression related to antibiotic synthesis in
ian tumor cells is present in Streptomyces peuce- Streptomyces hygroscopicus, Mol. Microbiol. 6,
tius, producer of daunorubicin and doxorubicin, 969-980.
Proc. Natl. Acad. Sci. USA 88, 8553-8557. HONG,S. K., MATSUMOTO, A., HORINOUCHI, S.,
ANCOCK, R. E. W. (1981a), Aminoglycoside up- BEPPU,T. (1993), Effects of protein kinase inhi-
take and mode of action with special reference bitors on in vitro protein phosphorylation and
to streptomycin and gentamicin. I. Antagonists cellular differentiation of Streptomyces griseus,
and mutants, J. Antimicrob. Chemother. 8, 249- Mol. Gen. Genet. 236, 347-354.
276. HOPWOOD,D. A., SHERMAN, D. H., KHOSLA, C.,
ANCOCK, R. E. W. (1981b), Aminoglycoside up- BIBB,M. J., SIMPSON, T. J., FERNANDEZ-MORE-
take and mode of action with special reference NO, M. A., MARTINEZ,E., MALPARTIDA, F.
to streptomycin and gentamicin. 11. Effects of (1990), “Hybrid” pathways for the production of
aminoglycosides on cells, J. Antimicrob. Che- secondary metabolites, in: Proc. 6th Int. Symp.
mother. 8, 429-445. Genet. Ind. Microorg. (HESLOT,H., DAVIES,J.,
HANCOCK,R. E. W., FARMER,S. W., LI, Z., FLORENT, J., BOBICHON, L., DURAND, G., PEN-
POOLE,K. (1991), Interaction of aminoglyco- EASSE, L., Eds.), pp. 259-270. Paris: SocietC
sides with outer membranes and purified lipopo- Fransaise de Microbiologie.
lysaccharide and OmpF porin of Escherichia HORINOUCHI, S., BEPPU,T. (1992), Regulation of
coli, Antimicrob. Agents Chemother. 35, 1309- secondary metabolism and cell differentiation in
1314. Streptomyces: A-factor as a microbial hormone
HARDICK, D. J., HUTCHINSON, D. W., TREW,S. J., and the AfsR protein as a component of a two-
WELLINGTON, E. M. H. (1992), Glucose is a pre- component regulator system, Gene 115, 167-
cursor of 1-deoxynojirimycin and l-deoxyman- 172.
nonojirimycin in Streptomyces subrutilus, Tetra- HORINOUCHI, S., BEPPU,T. (1994). A-factor as a
hedron 48,6285-6296. microbial hormone that controls cellular differ-
HASEGAWA, M. (1991), A gene cloning system in entiation and secondary metabolism in Strepto-
Micromonospora which revealed the organiza- myces griseus, Mol. Microbiol. 12, 859-864.
HORNEMANN, U., KEHRER,J. P., NUNEZ,C. S., ITOH,S., ODAKURA, Y., KASE,H., SATHO,S., TA-
RANIERI, A. L. (1974), D-Glucosamine and L-ci- KAHASHI,K., IIDA, T., SHIRAHATA, K., NA-
trulline, precursors in mitomycin biosynthesis in KAYAMA, K. (1984), Biosynthesis of astromicin
Streptomyces verticillatus, J. Am. Chem. Soc. 96, and related antibiotics. I. Biosynthetic studies by
320-322. bioconversion experiments, J. Antibiot. 37,1664-
HORTON,D., HAWKINS, L. D., MCGARVEY,G. J. 1669.
(1989), Trends in synthetic carbohydrate chemis- KAGAN,R. M., CLARKE,S. (1994), Widespread oc-
try, ACS Symp. Ser. 386. Washington, D C currence of three sequence motifs in diverse S-
American Chemical Society. adenosylmethionine-dependent methyltransfer-
HOTTA, K., ISHIKAWA, J., ICHIHARA, M., NAGA- ases suggests a common structure for these en-
NAWA,H., MIZUNO,S. (1988), Mechanism of in- zymes, Arch. Biochem. Biophys. 310,417-427.
creased kanamycin-resistance generated by pro- KAKINUMA, K. (1982), Biosynthesis of ribostamy-
toplast regeneration of Streptomyces griseus. I. cin. Application of the deuterium label, in:
cloning of a gene segment directing a high level Trends in Antibiotic Research. Genetics, Biosyn-
of an aminoglycoside 3-N-acetyltransferase ac- theses, Actions, and New Substances (UMEZA-
tivity, J. Antibiot. 41, 94-103. WA, H., DEMAIN,A. L., HATA, T., HUTCHIN-
HO-ITA,K., MORIOKA, M., OKAMI,Y. (1989), Bio- SON, C. R., Eds.), pp. 185-194. Tokyo: Japan
synthetic similarity between Streptomyces tenji- Antibiotics Research Association.
mariensis and Micromonospora olivasterospora KAMEDA,Y., ASANO,N., YOSHIKAWA, M., TA-
which produce fortimicin-group antibiotics, J. KEUCHI, M., YAMAGUCHI, T., MATSUI,K., SA-
Antibiot. 42, 745-751. TOSHI,H., FUKASE,H. (1984), Valiolamine, a
HO-ITA, K., DAVIES,J., YAGISAWA,M. (1995), new 0-glucosidase inhibiting aminocyclitol pro-
Aminoglycosides and aminocyclitols (other than duced by Streptomyces hygroscopicus, J. Anti-
streptomycin), in: Biochemistry and Genetics of biot. 37, 1301-1307.
Antibiotic Biosynthesis (VINING, L., STU-ITARD, KARWOWSKI, J. P., JACKSON,M., BOBIK,T. A.,
Eds.), pp. 571-595. Stoneham: Butterworth-Hei- PROKOP,J. F., THERIAULT, R. J. (1984), Spenol-
nemann. imycin, a new spectinomycin-type antibiotic. I.
Discovery, taxonomy and fermentation, J. Anti-
HUDLICKY, T., CEBULAK, M. (1993), Cyclitols and biot. 37, 1513-1518.
Their Derivatives. Weinheim: VCH.
KASE,H., ODAKURA, Y., TAKAZAWA, Y., KITA-
HUTCHIN,T., CORTOPASSI, G. (1994), Proposed MURA,s., NAKAYAMA, K. (1982), Biosynthesis
molecular and cellular mechanism for aminogly- of sagamicin and related aminoglycosides, in:
coside ototoxicity, Antimicrob. Agents Chemo- Trends in Antibiotic Research. Genetics, Biosyn-
ther. 38,2517-2520. theses, Actions, and New Substances (UMEZA-
ICHIMURA,M., KOGUCHI,T., YASUZAWA,T., WA,H., DEMAIN,A. L., HATA, T., HUTCHIN-
TOMITA,F. (1987), CV-1, a new antibiotic pro- SON,C. R., Eds.), pp. 195-212. Tokyo: Japan
duced by a strain of Streptomyces sp. I. Fermen- Antibiotics Research Association.
tation, isolation and biological properties of the KASE,H., IWAHASHI, K., MATSUDA,Y. (1986), K-
antibiotic, J. Antibiot. 40,723-726. 252a, a potent inhibitor of protein kinase C of
IKEDA,Y., GOMI,S., YOKOSE,K., NAGANAWA, microbial origin, J. Antibiot. 39, 1059-1065.
H., IKEDA, T., MANABE,M., HAMADA,M., KATZ,L., DONADIO,S. (1993), Polyketide synthe-
KONDO, S., UMEZAWA,H. (1985a), A new sis: prospects for hybrid antibiotics, Ann. Rev.
streptomycin group antibiotic produced by Microbiol. 47, 875-912.
Streptomyces sioyaensis, J. Antibiot. 38, 1803- KENNEDY,J. F. (1988), Carbohydrate Chemistry.
1805. Oxford Clarendon Press.
IKEDA,Y., KONDO,S., KANAI,F., SAWA,T., HA- KHOKHLOV, A. S. (1988), Results and perspectives
MADA, M., TAKEUCHI,T., UMEZAWA,H. of actinomycete autoregulators studies, in: Biol-
(1985b), A new destomycin-family antibiotic ogy of Actinomycetes '88 (OKAMI,Y., BEPPU,T.,
produced by Saccharopolyspora hirsuta, J. Anti- OGAWARA,H., Eds.), pp. 338-345. Tokyo: Ja-
biot. 38, 436-438. pan Scientific Societies Press.
INAMINE, E., DEMAIN,A. L. (1975), Mannosido- KHOKHLOV, A. S., TOVAROVA, I. I., BORISOVA, L.
streptomycin hydrolase, Methods Enzymol. 43, N., PLINER,S. A., SHEVCHENKO, L. A., KOR-
637-640. NITSKAYA, E. Y., IVKINA,N. S., RAPOPORT,I.
INOUYE,S., SHOMURA, T., WATANABE, H., TOT- A. (1967), A-factor responsible for the biosyn-
SUGAWA,K., NIIDA,T. (1973), Isolation and thesis of streptomycin by a mutant strain of Acti-
gross structure of a new antibiotic SS-56C and nomyces streptomycini, Doklady Akad. Nauk
related compounds, J. Antibiot. 26, 374-385. SSSR 177,232-235.
I1 References 481
KIM,C.-G., KIRSCHNING, A., BERGON,P., ZHOU, ase has a regulatory role in carbon catabolite re-
P., Su, E., SAUERBREI, B., NING,S., AHN, Y., pression in Streptomyces coelicolor, J. Bacteriol.
BREUER, M., LEISTNER,E., FLOSS, H. G. 176,2694-2698.
(1996), Biosynthesis of 3-amino-5-hydroxyben- LACALLE,R. A., TERCERO,J. A., JIMENEZ,A.
zoic acid, the precursor of mC7N units in ansa- (1992). Cloning of the complete biosynthetic
mycin antibiotics, J. Am. Chem. SOC. 118, 7486- gene cluster for an aminonucleoside antibiotic,
7491. puromycin, and its regulated expression in hetero-
KIRBY,R. (1990). Evolutionary origin of aminogly- logous host, EMBO J. 11,785-792.
coside phosphotransferase resistance genes, J. LAM, K. S., FLORENZA,S., SCHROEDER,D. R.,
Mol. Evol. 30,489-492. DOYLE,T. W., PEARCE,C. J. (1989). Biosynthe-
KNIEP,B., GRISEBACH, H. (1976), Enzymatic syn- sis of rebeccamycin, a novel antitumor agent, in:
thesis of streptomycin. Transfer of L-dihydro- Novel Microbial Products for Medicine and Ag-
streptose from dTDP-L-dihydrostreptose to riculture (DEMAIN,A. L., SOMKUTI,G. A.,
streptidine-6-phosphate, FEBS Lett. 65, 44-46. HUNTER-CREVA, J. C., ROSSMOORE,H. W.,
KNIEP,B., GRISEBACH, H. (1980), Biosynthesis of Eds.), pp. 63-66. Amsterdam: Elsevier Science
streptomycin. Purification and properties of Publishers.
a dTDP-~-dihydrostreptose:streptidine-6-phos- LESKIW,B. K., BIBB,M. J., CHATER,K. F. (1991a),
phate dihydrostreptosyltransferasefrom Strepto- The use of a rare codon specifically during de-
myces griseus, Eur. J. Biochem. 105, 139-144. velopment, Mol. Microbiol. 5, 2861-2867.
KNIGHTON, D. R., ZENG,J., TENEYCK, L. F., ASH-. LESKIW,B. K., LAWLOR,E. J., FERNANDEZ-ABA-
FORD, v. A., XUONG, N. H., TAYLOR, s. s. LOS, J. M., CHATER,K. F. (1991b), TTA codons
(1991), Crystal structure of the catalytic subunit in some genes prevent their expression in a class
of cyclic adenosine monophosphate-dependent of developmental, antibiotic-negative Strepto-
protein kinase, Science 253, 407-414. myces mutants, Proc. Natl. Acad. Sci. USA 88,
KOBAYASHI, Y., SHIOZAKI, M. (1994), Synthesis of 2461-2465.
trehazolin panomer, J. Antibiot. 47, 243-246. LIRAS,P., ASTURIAS,J. A., MARTIN,J. F. (1990),
KOGA, Y., NISHIHARA, M., MORII,H., AKAGA- Phosphate control sequences involved in tran-
WA-MATSUSHITA, M. (1993), Ether lipids of me- scriptional regulation of antibiotic biosythesis,
thanogenic bacteria: structures, comperative as- TIBTECH 8,184-189.
pects, and biosynthesis, Microbiol. Rev. 57, 164- LISHNEVSKAYA, E. B., KUZINA,Z. A., ASINOOVS-
182. KAYA,N. K., BELOUSOVA, I. I., MALKOV,M.
KOHLHEPP,S. J., Hou, L., GILBERT,D. N. (1994), A., RAVINSKAYA, A. YA. (1986), Cyclic adeno-
Pig kidney (LLC-PKl) cell membrane fluidity sine-3',5 '-monophosphoric acid in Streptomyces
during exposure to gentamicin or tobramycin, antibioticus and its possible role in the regula-
Antimicrob. Agents Chemother. 38,2169-2171. tion of oleandomycin biosynthesis and culture
KORZYBSKI, T., KOWSZYK-GINDIFER, Z., KURY- growth, Mikrobiologiya 55, 350.
LOWICZ,W. (1978), Antibiotics. Origin, Nature, LIU, H.-W., THORSON, J. S. (1994), Pathways and
and Properties. Washington, D C American So- mechanisms in the biogenesis of novel deoxysu-
ciety for Microbiology. gars by bacteria, Ann. Rev. Microbiol. 48, 223-
KROHN,K., KIRST,H. A., MAAG,H. (1993), Anti- 256.
biotics and Antiviral Compounds. Chemical Syn- LONDEI,P., ALTAMURA, S., SANZ,J. L., AMILS,R.
thesis and Modification. Weinheim: VCH. (1988), Aminoglycoside-induced mistranslation
KUMADA,Y., HORINOUCHI, S., UOZUMI,T., BEP- in thermophilic archaebacteria, Mol. Gen. Genet.
PU, T. (1986), Cloning of a streptomycin-produc- 214,48-54.
tion gene directing synthesis of N-methyl-L-glu- LOOK,G. C., FOTSCH,C. H., WONG,C.-H. (1993),
cosamine, Gene 43, 221-224. Enzyme-catalysed organic synthesis: practical
Kuo, M. S., YUREK,D. A., COATS,J. H., LI, G. P. routes to aza sugars and their analogs for use as
(1989), Isolation and identification of 7,8-dide- glycoprocessing inhibitors, Acc. Chem. Res. 26,
methyl-8-hydroxy-5-deazariboflavin,an unusual 182-190.
cosynthetic factor in streptomycetes from Strep- LORIAN,V. (Ed.) (1980), Antibiotics in Laboratory
tomyces lincolnensis, J. Antibiot. 42, 475-478. Medicine. Baltimore, M A Williams & Wilkins.
Kuo, M. S., YUREK,D. A., COATS,J. H., CHUNG, LUCHER,L. A., CHEN, Y.-M., WALKER,J. B.
S. T., LI, G. P. (1992), Isolation and identifica- (1989), Reactions catalysed by purified L-glut-
tion of 3-propylidene-A'-pyrroline-5-carboxylic amine:keto-scyllo-inositol aminotransferase, an
acid, a biosynthetic precursor of lincomycin, J. enzyme required for biosynthesis of aminocycli-
Antibiot. 45,1773-1777. to1 antibiotics, Antimicrob. Agents Chemother.
KWAKMAN, M., POSTMA,W. (1994), Glucose kin- 33,452-459.
LUKACS,G., OHNO,M. (1990), Recent Progress in of streptomycin production genes in two species
the Chemical Synthesis of Antibiotics. Berlin: of Streptomyces. Evidence for occurrence of a
Springer-Verlag. second amidino-transferase gene, Mol. Gen.
MACNEIL,D. J. (1995), Avermectins, in: Biochem- (Life Sci. Adv.) 7, 83-87.
istry and Genetics of Antibiotic Biosynthesis MAZODIER,P., DAVIES,J. (1991), Gene transfer
(VINING,L., STUTTARD,C., Eds.), pp. 421442. between distantly related bacteria, Annu. Rev.
Boston, MA: Butterworth-Heinemann. Genet. 25, 147-171.
MAIER,S., GRISEBACH, H. (1979), Biosynthesis of MCCUE,L. A., KWAK,J., BABCOCK,M. J., KEN-
streptomycin. Enzymic oxidation of dihydro- DRICK,K. F. (1992), Molecular analysis of spo-
streptomycin (6-phosphate) to streptomycin (6- rulation in Streptomyces griseus, Gene 115, 173-
phosphate) with a particulate fraction of Strepto- 179.
myces griseus, Biochim. Biophys. Acta 586, 231- MEHLING,A., WEHMEIER,U. F., PIEPERSBERG,
241. W. (1995), Nucleotide sequence of Streptomyces
MALLAMS,A. K. (1988), The carbohydrate-con- 16s ribosomal DNA: towards a specific identifi-
taining antibiotics, in: Carbohydrate Chemistry cation system for streptomycetes using PCR, Mi-
(KENNEDY,J. F., Ed.), pp. 73-133. Oxford: crobiology 141, 2139-2147.
Claendron Press. MILLER,A. L., WALKER,J. B. (1969), Enzymatic
MANSOURI, K., PIEPERSBERG, W. (1991), Genetics phosphorylation of streptomycin by extracts of
of streptomycin production in Streptomyces gri- streptomycin-producing strains of Streptomyces,
seus: nucleotide sequence of five genes, J. Bacteriol. 99, 401-405.
strFGHIK, including a phosphatase gene, Mol. MIYAKE,K., HORINOUCHI, S., YOSHIDA,M., CHI-
Gen. Genet. 228, 459469. BA, N., MORI,K., NOGAWA,N., MORIKAWA,
MANSOURI,K., PISSOWOTZKI, K., DISTLER,J., N., BEPPU,T. (1989), Detection and properties
MAYER,G., HEINZEL,P., BRAUN,C., EBERT, of A-factor-binding protein from Streptomyces
A., PIEPERSBERG, W. (1989), Genetics of strep- griseus, J. Bacteriol. 171, 4298-4302.
tomycin production, in: Genetics and Molecular MIYAKE,K., KUZUYAMA, T., HORINOUCHI, S.,
Biology of Industrial Microorganisms (HERSH- BEPPU,T. (1990), The A-factor-binding protein
BERGER, c. L., QUEENER,s. w., HEGEMAN, of Streptomyces griseus negatively controls strep-
G., Eds.), pp. 61-67. Washington, D C American tomycin production and sporulation, J. Bacteriol.
Society for Microbiology. 172,3003-3008.
MARSH,P., WELLINGTON, E. M. H. (1994), Molec- MOAZED,D., NOLLER,H. F. (1987), Interactions
ular ecology of filamentous actinomycetes in of antibiotics with 16s rRNA, Nature 327, 389-
soil, in: Molecular Ecology of Rhizosphere Mi- 394.
croorganisms. Biotechnology and the Release of MULLER,L. (1989), Chemistry, biochemistry and
GMOs (O'GARA, F., DOWLING,D. N., BOES- therapeutic potential of microbial a-glucosidase
TEN,B., Eds.), pp. 133-149. Weinheim: VCH. inhibitors, in: Novel Microbial Products for Med-
MARTIN,J. F. (1989), Molecular mechanisms for icine and Agriculture (DEMAIN, A. L., SOMKUTI,
the control by phosphate of the biosynthesis of G. A., HUNTER-CREVA, J. C., ROSSMOORE, H.
antibiotics and other metabolites, in: Regulation W., Eds.), pp. 109-116. Amsterdam: Elsevier
of Secondary Metabolism in Actinomycetes Science Publishers.
(SHAPIRO,S., Ed.), pp. 213-237. Boca Raton, MULLER,A., KELLER-SCHIERLEIN, W., BIELECKI,
FL: CRC Press. J., RAK,G., STUMPFEL, J., ZAHNER,H. (1986),
MARTIN,J. F., DEMAIN,A. L. (1980), Control of (2S,3R,4R,6R) - 2,3,4,- Trihydroxy - 6 - methylhexa-
antibiotic biosynthesis, Microbiol. Rev. 44, 230- non aus zwei Actinomyceten-Stammen, Helv.
251. Chim. Acta 69, 1829-1832.
MARTIN,J. F., LIRAS,P. (1989), Organization and MUNRO,M. H. G., TANIGUCHI, M., RINEHART, K.
expression of genes involved in the biosynthesis L., GOTTLIEB,D., STOUDT,T. H., ROGERS,T.
of antibiotics and other secondary metabolites, 0. (1975), Carbon-13 evidence for the stereo-
Ann. Rev. Microbiol. 43,173-206. chemistry of streptomycin biosynthesis from glu-
MAYER,G. (1994), Molekulare Analyse und Evo- cose, J. Am. Chem. SOC.91,4782-4783.
lution von 5 '-Hydroxystreptomycin- und Bluen- MURPHY,P. J., HEYCKE,N., BANFALVI, Z., TATE,
somycin-Biosynthesegenen aus Streptomyces M. E., DE BRUJIN,F., KONDOROSI, A., TEMPE,
glaucescens GLA.0 und Streptomyces bluensis J., SCHELL,J. (1987), Genes for the catabolism
ISP 5564, Dissertation, Bergische Universitat and synthesis of an opine-like compound in Rhi-
GH, Wuppertal. zobium meliloti are closely linked and on the
MAYER,G., VOGTLI,M., PISSOWOTZKI, K., HUT- Sym plasmid, Proc. Natl. Acad. Sci. USA 84,
TER, R., PIEPERSBERG, W. (1988), Colinearity 493-497.
I 1 References 483
MURPHYP. J., TRENZ,S. P., GRZEMSKI, W., DE cluster for streptomycin biosynthesis and strep-
BRUIJN, F. J., SCHELL,J. (1993), The Rhizobium tomycin resistance, J. Bacteriol. 164, 85-94.
meliloti rhizopine mos locus is a mosaic structure OHNUKI, T., IMANAKA, T., AIBA,S. (1985b), Isola-
facilitating its symbiotic regulation, J. Bacteriol. tion of streptomycin non-producing mutants de-
175,5193-5204. ficient in biosynthesis of the streptidine moiety
NEUMANN, T., PIEPERSBERG, W., DISTLER,J. or linkage between streptidine-6-phosphate and
(1996), The decision phase model of the growth dihydrostreptose, Antimicrob. Agents Chemo-
phase-dependent regulation of biosynthesis of ther. 27, 367-374.
streptomycin in Streptomyces griseus, Microbio- OHTA,T., HASEGAWA, M. (1993a), Analysis of the
logy 142, 1953-1963. self-defense gene ( f m r 0 ) of a fortimicin A (as-
NEWTON,G. L., ARNOLD,K., PRICE,M. S., SHER- tromicin) producer, Micromonospora olivaste-
RILL, c., DELCARDAYRE, s. B., AHARONO- rospora: comparison with other aminoglycoside-
WITZ,Y., COHEN,G., DAVIES, J., FAHEY, R. C., resistance-encoding genes, Gene 127, 63-69.
DAVIS,C. (1996), Distribution of thiols in micro- OHTA,T., HASEGAWA, M. (1993b), Analysis of the
organisms: mycothiol is a major thiol in most ac- nucleotide sequence of fmrT encoding the self-
tinomycetes, J. Bacteriol. 178, 1990-1995. defense gene of the istamycin producer, Strepto-
NIERHAUS, K. H. (Ed.) (1993). The Translational myces tenjimariensis ATCC 31602; comparison
Apparatus. New York: Plenum Press. with the sequences of kamB of Streptomyces ten-
NIMI, O., ITO, G., SUEDA,S., NOMI, R. (1971), ebrarius NCIB 11028 and kamC of Saccharopo-
Phosphorylation of streptomycin at C6-OH of lyspora hirsuta CL102, J. Antibiot. 46, 511-517.
streptidine moiety by an intracellular enzyme of
OHTA, T., HASHIMOTO,E., HASEGAWA,M.
Streptomyces griseus, Agr. Biol. Chem. 35, 848- (1992a), Characterization of sannamycin A-non-
855. producing mutants of Streptomyces sannanensis,
NOLLER,H. F. (1993), On the origin of the ribo- J. Antibiot. 45, 289-291.
some: Coevolution of subdomains of tRNA and
rRNA, in: The RNA World (GESTLAND, R. F., OHTA, T., HASHIMOTO,E., HASEGAWA,M.
ATKINS,J. F., Eds.), pp. 137-184. New York: (1992b), Cloning and analysis of a gene (smsl3)
Cold Spring Harbor Laboratory Press. encoding sannamycin B-glycyltransferase from
NUMATA, K., YAMAMOTO,H., HATORI, M., Streptomyces sannanensis and its distribution
among actinomycetes, J. Antibiot. 45, 1167-
MIYAKI,T., KAWAGUCHI, H. (1986), Isolation
1175.
of an aminoglycoside hypersensitive mutant and
its application in screening, J. Antibiot. 39, 994- OHTA,T., DAIRI,T., HASHIMOTO, E., HASEGA-
1OOO. WA,M. (1993a), Use of a heterologous gene for
&HI, K. (1987), Metabolic initiation of differentia-
molecular breeding of actinomycetes producing
tion and secondary metabolism by Streptomyces structurally related antibiotics: self-defense
griseus: significance of the stringent response genes of producers of fortimicin-A (astromicin)-
(ppGpp) and GTP content in relation to A-fac- group antibiotics, Actinomycetol. 7, 145-155.
tor, J. Bacteriol. 169, 3608-3616. OHTA, T., DAIRI,T., HASEGAWA, M. (1993b),
&HI, K. (1988), Nucleotide pools and stringent re- Characterization of two different types of resist-
sponse in regulation of Streptomyces differentia- ance genes among producers of fortimicin-group
tion, in: Biology of Actinomycetes '88 (OKAMI, antibiotics, J. Gen. Microbiol. 139, 591-599.
Y., BEPPU,T., OGAWARA, H., Eds.), pp. 330- OKADA,H., YAMAMOTO, K., TSUTANO, S., NAKA-
337. Tokyo: Japan Scientific Societies Press. MURA, s. (1988), A new group of antibiotics, hy-
&HI, K. (1990), Streptomyces griseus, as an excel- droxamic acid antimycotic antibiotics. I. Precur-
lent object for studing microbial differentiation, sor-initiated changes in productivity and biosyn-
Actinomycetol. 4, 23-30. thesis of neoenactins NL, and NL,, J. Antibiot.
ODAKURA, Y., KASE,H., ITOH,S., SATHO,S., TA- 41,869-874.
KASAWA, S., TAKAHASHI, K., SHIRAHATA, K., OKUDA,T., ITO,Y. (1982), Biosynthesis and muta-
NAKAYAMA, K. (1984), Biosynthesis of astromi- synthesis of aminoglycoside antibiotics, in: Ami-
cin and related antibiotics. 11. Biosynthetic stud- noglycoside Antibiotics (UMEZAWA, H., HOOP-
ies with blocked mutants of Micromonospora ol- ER, I. R., Eds.), pp. 111-203. Berlin: Springer-
ivasteropora, J. Antibiot. 37, 1670-1680. Verlag.
OGURA,H., HASEGAWA, A., SUAMI,T. (1992), ONAKA,H., ANDO,N., NIHIRA,T., YAMADA, Y.,
Carbohydrates - Synthetic Methods and Applica- BEPPU,T., HORINOUCHI, S. (1995), Cloning and
tions in Medicinal Chemistry. Tokyo: Kodansha. characterization of the A-factor receptor gene
OHNUKI, T., IMANAKA, T., AIBAS. (1985a), Self- from Streptomyces griseus, J. Bacteriol. 177,
cloning in Streptomyces griseus of an str gene 6083-6092.
OTSUKA, H., MASKARETTI, 0. A., HURLEY, L. H., PIEPERSBERG, W. (1993), Streptomycetes and co-
FLOSS,H. G. (1980), Stereochemical aspects of rynebacteria, in: Biotechology, 2nd Edn., Vol. 1
the biosynthesis of spectinomycin, J. Am. Chem. (REHM, H.-J., REED, G., Eds.), pp. 433468.
SOC. 102,6817-6820. Weinheim: VCH.
O ~ E NS., L., LIU, X., FERGUSON, J., HUTCHIN- PIEPERSBERG, W. (1994), Pathway engineering in
SON,C. R. (1995), Cloning and characterization secondary metabolite-producing actinomycetes,
of the Streptomyces peuceticus dnrQS genes en- Crit. Rev. Biotechnol. 14,251-285.
coding a daunosamine biosynthesis enzyme and PIEPERSBERG, W. (1995), Streptomycin and related
a glycosyltransferase involved in daunorubicin aminoglycosides, in: Biochemistry and Genetics
biosynthesis, J. Bacteriol. 177, 6688-6692. of Antibiotic Biosynthesis (VINING,L., STUT-
OZAKI,M., MIZUSHIMA, S., NOMURA,M. (1969), TARD,C., Eds.), pp. 531-570. Boston, M A But-
Identification and functional characterization of terworth-Heinemann.
the protein controlled by the streptomycin-re-
sistant locus in E. coli, Nature 222, 333-339. PIEPERSBERG, W., GEYL,D., HUMMEL, H., BOCK,
PARRY,R. J. (1992), Investigation of the biosynthe- A. (1980), Physiology and biochemistry of bacte-
sis of aristeromycin, in: Secondary-Metabolite rial ribosomal mutants, in: Genetics and Evolu-
Biosynthesis and Metabolism (PETROSKI, R. J., tion of RNA Polymerase, tRNA and Ribosomes
MCCORMICK,S. P., Eds.), pp. 89-104. New (OSAWA, S., OZEKI,H., UCHIDA,H., YURA,T,
York: Plenum Press. Eds.), pp. 359-377. Tokyo: University of Tokyo
PARRY,R. J., BORNEMANN, V., SUBRAMANIAN, Press.
R. (1989), Biosynthesis of the nucleoside anti- PIEPERSBERG, W., DISTLER,J., HEINZEL, P., PER-
biotic aristeromycin, J. Am. Chem. SOC. 111, EZ-GONZALEZ, J.-A. (1988), Antibiotic resist-
5819-5824. ance by modification: many resistance genes
PEARCE,C. J., RINEHART, K. L. (1981), Biosynthe- could be derived from cellular control genes in
sis of aminocyclitol antibiotics, in: Antibiotics, actinomycetes - A hypothesis, Actinomycetol. 2,
Vol. 4, Biosynthesis (CORCORAN, J. W., Ed.), 83-98.
pp. 74-100. Berlin: Springer-Verlag. PIEPERSBERG, W., HEINZEL,P., MANSOURI, K.,
PENYIGE,A., VARGHA, G., ENSIGN,J. C., BARA- MONINGHOFF, U., PISSOWOTZKI, K. (1991a),
BAS,G. (1992), The possible role of ADP ribo- Evolution of antibiotic resistance and produc-
sylation in physiological regulation of sporulation genes in streptomycetes, in: Genetics and
tion in Streptomyces griseus, Gene 115, 181-185. Product Formation in Streptomyces (BAUM-
PESCHKE,U., SCHMIDT,H., ZHANG,H.-Z., PIE- BERG, s., KRUGEL,H., NOVACK,D., Eds.), pp.
PERSBERG, W. (1999, Molecular characteriza- 161-170. New York Plenum Press.
tion of the lincomycin production gene cluster of PIEPERSBERG, W., STOCKMANN, M., MANSOURI,
Streptomyces lincolnensis 78-11, Mol. Microbiol. K., DISTLER,J., GRABLEY,S., SICHEL, P.,
16,1137-1156. BRAu, B. (1991b), German Patent Application
PHILLIPS, L., WELLINGTON, E. M. H., REES,S. B., P4130 967 - HOE 91/F 300.
JUN,L. S., KING,G. P. (1992), The distribution PIEPERSBERG, W., JIMENEZ,A., CUNDLIFFE, E.,
of DNA sequences hybridizing with antibiotic GRABLEY,S., BRAu, B., MARQUARDT,R.
production and resistance gene probes within (1994), Glycosyltransferases from streptomy-
type strains and wild isolates of Streptomyces cetes as tools in biotransformations, BRIDGE
species, J. Antibiot. 45, 1481-1491. Progress Report 1994 (CEC) (VASSAROTTI, A.,
PIENDL,W., BOCK,A., CUNDLIFFE, E. (1984), In- Ed.), pp. 128-133. Brussels: Printeclair.
volvement of 16s ribosomal RNA in resistance
of the aminoglycoside-producers Streptomyces PISSOWOTZKI, K., MANSOURI,K., PIEPERSBERG,
tenjimariensis, Streptomyces tenebrarius, and Mi- W. (1991), Genetics of streptomycin production
cromonospora purpurea, Mol. Gen. Genet. 197, in Streptomyces griseus. Molecular structure and
24-29. putative function of genes strELMB2N, Mol.
PIEPERSBERG, W. (1985), Aminoglycoside Anti- Gen. Genet. 231,113-123.
biotika: Wichtige Therapeutika und Objekte der POHLMAN, R. F., GENETTI,H. D., WINANS,S. C.
Grundlagenforschung, Forum Mikrobiol. 8/85, (1994), Common ancestry between IncN conju-
153-161. gal transfer genes and macromolecular export
PIEPERSBERG, W. (1992), Metabolism and cell indi- systems of plant and animal pathogens, Mol. Mi-
vidualization, in: Secondary Metabolites: Their crobiol. 14,655-668.
Function and Evolution, Ciba Foundation Sym- PRATT,W. B., FEKETY,R. (1986), The Antimicro-
posium 171 (CHADWICK, D., WHELAN,J., Eds.), bial Drugs. New York: Oxford University
pp. 294-299. Chichester: John Wiley & Sons. Press.
I1 References 485
PUROHIT,P., STERN,S. (1994), Interactions of regulators. 11. Allosamidin, a novel insect chitin-
small RNA with antibiotic and RNA ligands of ase inhibitor, J. Antibiot. 40,296-300.
the 30s subunit, Nature 370, 659-662. SALAUZE,D., PEREZ-GONZALEZ, J. A., PIEPERS-
RAETZ,C. R. H. (1996), Bacterial lipopolysacchar- BERG, w., DAVIES,J.’ (1991), Characterisation
ides: a remarkable family of bioactive macroam- of aminoglycoside acetyltransferase-encoding
phiphiles, in: Escherichia coli and Salmonella. genes of neomycin-producing Micromonospora
Cellular and Molecular Biology (NEIDHARDT, F. chalcea and Streptomycesfradiae, Gene 101,143-
C., Ed.), pp. 1035-1063. Washington DC: Ameri- 148.
can Society for Microbiology. SATOH,A., OGAWA,H., SATOMURA, Y. (1976),
RAGAN,C. M., VINING,L. C. (1978), Intracellular Regulation of N-acetylkanamycin amidohydrol-
cyclic adenosine 3 ‘ 3‘-monophosphate levels ase in the idiophase in kanamycin fermentation,
and streptomycin production in cultures of Agric. Biol. Chem. 40,191.
Streptomyces griseus, Can. J. Microbiol. 24, SCHLESSINGER, D., MEDOFF,G. (1975), Strepto-
1012-1015. mycin, dihydrostreptomycin, and the gentami-
REEVES,P. (1993), Evolution of Salmonella 0 an- cins, in: Antibiotics, Vol. 111 (CORCORAN, J. W.,
tigen variation by interspecific gene transfer on a HAHN,F. E., Eds.), pp. 535-550. Berlin: Sprin-
large scale, TIG 9, 17-22. ger-Verlag.
RETZLAFF,L., DISTLER,J. (1995), The regulator of SCHROEDER,R., STREICHER,B., WANK, H.
streptomycin gene expression, StrR, of Strepto- (1993), Splice-site selection and decoding: are
myces griseus is a DNA binding activator protein they related? Science 260, 1443-1444.
with multiple recognition sites, Mol. Microbiol. SHAPIRO, S. (1989), Nitrogen assimilation in acti-
18, 151-162. nomycetes and the influence of nitrogen nutri-
RETZLAFF, L., MAYER,G., BEYER,S., AHLERT,J., tion on actinomycete secondary metabolism, in:
VERSECK,S., DISTLER,J., PIEPERSBERG, W. Regulation of Secondary Metabolism in Actino-
(1993), Streptomycin production in streptomy- mycetes (SHAPIRO,S., Ed.), pp. 135-211. Boca
cetes: a progress report, in: Industrial Microor- Raton, FL: CRC Press, Inc.
ganisms: Basic and Applied Molecular Genetics SHAW,K. J., RATHER,P. N., HARE,R. S., MILLER,
(HEGEMAN, G. D., BALTZ,R. H., SKATRUD, P. G. H. (1993), Molecular genetics of aminoglyco-
L., Eds.), pp. 183-194. Washington: American side resistance genes and familial relationships
Society for Microbiology. of the aminoglycoside-modifying enzymes, Mi-
RINEHART, K. L., JR. (1980), Biosynthesis and mu- crobiol. Rev. 57, 138-163.
tasynthesis of aminocyclitol antibiotics, in: Ami-
nocyclitol Antibiotics, ACS Symp. Ser. 125 SHNAITMAN, C. A., KLENA,J. D. (1993), Genetics
(RINEHART, K. L., SUAMI,T., Eds.). Washing- of lipopolysaccharide biosynthesis in enteric bac-
ton, D C American Chemical Society. teria, Microbiol. Rev. 57, 655-682.
RINEHART, K. L., JR., STROSHANE, R. M. (1976), SIPOS,L., SZABO,G. (1989), Myo-inositol-l-phos-
Biosynthesis of aminocyclitol antibiotics, J. Anti- phate synthase in different Streptomyces griseus
biot. 29, 319-353. variants, FEMS Microbiol. Lett. 65, 339-364.
RINEHARD, K. L., POTGIETER, M., DELAWARE, D. SMAOUI,H., MALLIE,J.-P., SCHAEVERBEKE, M.,
L. (1981), Direct evidence from multiple 13C la- ROBERT,A., SCHAEVERBEKE, J. (1993), Gen-
beling and homonuclear decoupling for the la- tamicin administered during gestation alters
beling pattern by glucose of the m-aminobenzoyl glomerular basement membrane development,
(C7N) unit in pactamycin, J. Am. Chem. SOC. Antimicrob. Agents Chemother. 37, 1510-1517.
103,2099-2101. STOCKMANN, M., PIEPERSBERG, W. (1992), Gene
RINEHART, K. L., JR., SNYDER, W. C., STALEY,A. probes for the detection of 6-deoxyhexose
L., LAU, R. C. M. (1992), Biosynthestic studies metabolism in secondary metabolite-producing
on antibiotics, in: Secondary-Metabolite Biosyn- streptomycetes, FEMS Microbiol. Lett. 90, 185-
thesis and Metabolism (PETROSKI, R. J., 190.
MCCORMICK, S. P., Eds.), pp. 41-60. New York: SWAN,S. K., KOHLHEPP,S. J., KOHNEN,P. W.,
Plenum Press. GILBERT,D. N., BENNETT,W. M. (1991), Long-
SAITOH,K., TSUNAKAWA, M., TOMITA,K., MIAKI, term protection of polyaspartic acid in experi-
T., KONISHI,M., KAWAGUCHI, H. (1988), Bo- mental gentamicin nephrotoxicity, Antimicrob.
holmycin, a new aminoglycoside antibiotic. I. Agents Chemother. 35,2591-2595.
Production, isolation and properties, J. Antibiot. SZESZAK,F., VITALIS,s., BBKEsI, I., SZABb, G.
41,855-861. (1991), Presence of factor C in streptomycetes
SAKUDA, S., ISOGAI,A., MATSUMOTO, s.,SUZUKI, and other bacteria, in: Genetics and Product For-
A. (1987), Search for microbial insect growth mation in Streptomyces (BAUMBERG, S., KRU-
GEL, H., NOVACK,D., Eds.), pp. 11-18. New Neotrehalosadiamine (BMY-28251), a new ami-
York: Plenum Press. nosugar antibiotic, J. Antibiot. 39, 1001-1003.
TAYLOR,S. S., BUECHLER, J. A., YONEMOTO, W. UMEZAWA, H. (1974), Biochemical mechanisms of
(1990), CAMP-dependent protein kinase: Frame- resistance to aminoglycoside antibiotics, Adv.
work for a diverse family of regulatory enzymes, Carbohydr. Chem. Biochem. 30,183-225.
Ann. Rev. Biochem. 59,971-1005. UMEZAWA, H., HOOPER,I. R. (Eds.) (1982), Ami-
TERCERO, J. A., ESPINOSA, J. C., LACALLE,R. A., noglycoside Antibiotics. Berlin: Springer-Verlag.
JIMBNEZ,A. (1996), The biosynthetic pathway UMEZAWA, s.,KONDO,s.,ITO, Y. (1986), Amino-
of the aminonucleoside antibiotic puromycin, as glycoside antibiotics, in: Biotechnology, 2nd
deduced from the molecular analysis of the pur Edn., Vol. 4 (REHM,H.-J., REED,G., Eds.), pp.
cluster of Streptomyces alboniger, J. Biol. Chem. 309-357. Weinheim: VCH.
271, 1579-1590. UMEZAWA,H., OKANISHI,M., UTAHARA,R.,
TESTA,B., KYBURZ,E., FUHRER,W., GIGER,R. MAEDA,K., KONDO,S. (1967a), Isolation and
(1993), Perspectives in Medicinal Chemistry. structure of kanamycin inactivated by a cell-free
Weinheim: VCH. system of kanamycin-resistant Escherichia coli,
THIEM,J. (Ed.) (1990), Carbohydrate Chemistry. J. Antibiot. A20,136-141.
Topics in Current Chemistry 154. Berlin: Sprin- UMEZAWA, H., OKANISHI, M., KONDO,S., HAMA-
ger-Verlag. NA, K., UTAHARA, R., MAEDA,K., MITSUHA-
THOMAS,G. J. (1990), Synthetis of anthracyclines SHI, S. (1967b), Adenylylstreptomycin, a product
related to daunomycin, in: Recent Progress in the of streptomycin inactivated by E. coli carrying
Chemical Synthesis of Antibiotics (LUKAS,G., R-factor, Science 157, 1559-1561.
OHNO,M., Eds.), pp. 467-496. Berlin: Springer- VAZQUEZ,D. (1979), Inhibitors of protein synthe-
Verlag. sis, in: Molecular Biology, Biochemistry, and
THOMPSON, C. J., SETO,H. (1995), Bialaphos, in: Biophysics, Vol. 30 (KLEINZELLER, A., SPRIN-
Biochemistry and Genetics of Antibiotic Biosyn- GER, G. F., WITTMANN, H. G., Eds.). Berlin:
thesis (VINING,L., STUTTARD,C., Eds.), pp. Springer-Verlag.
197-222. Boston, MA: Butterworth-Heinemann. VILCHES,C., HERNANDEZ, C., MENDEZ,C., SAL-
THORSON, J. S., LIU,H.-W. (1993a), Characteriza- AS,J. A. (1992), Role of glycosylation in biosyn-
tion of the first PMP-dependent iron-sulfur-con- thesis of and resistance to oleandomycin in the
taining enzyme which is essential for the biosyn- producer organism, Streptomyces antibioticus, J.
thesis of 3,6-dideoxyhexoses,J. Am. Chem. SOC. Bacteriol. 174, 161-165.
115,7539-7540. VINING, L. C., DOULL,J. L. (1988), Catabolite re-
THORSON, J. S., LIU, H.-W. (1993b), Coenzyme B6 pression of secondary metabolism in actinomy-
as a redox cofactor: a new role for an old coen- cetes, in: Biology of Actinomycetes '88 (OKAMI,
zyme? J. Am. Chem. SOC. 115, 12177-12178. Y., BEPPU,T., OGAWARA, H., Eds.), pp. 406-
THORSON, J. S., Lo, S. F., LIU, H.-W., HUTCHIN- 411. Tokyo: Japan Scientific Societies Press.
SON, C. R. (1993), Biosynthesis of 3,6-dideoxy- VINING,L., STUTTARD,C. (1995), Biochemistry
hexoses: New mechanistic reflections upon 2,6- and Genetics of Antibiotic Biosynthesis. Boston,
dideoxy, 4,6-dideoxy, and amino sugar construc- M A Butterworth-Heinemann.
tion, J. Am. Chem. SOC. 115,69936994. VOGTLI,M., HUTTER,R. (1987), Characterization
TOHMA,S., KONDO, H., YOKOTSUGA, J., IWAMO- of the hydroxystreptomycin phosphotransferase
TO, J., MATSUHASHI, G., ITO, T. (1989), Ashi- gene (sph) of Streptomyces glaucescens: nucleo-
mycins A and B, new streptomycin analogues, J. tide sequencing and promoter analysis, Mol.
Antibiot. 42, 1205-1212. Gen. Genet. 208, 195-203.
TOHYAMA, H., OKAMI,Y., UMEZAWA, H. (1987), VON AHSEN,U., NOLLER,H. F. (1993), Footprint-
Nucleotide sequence of the streptomycin phos- ing the sites of interaction of antibiotics with ca-
photransferase and amidinotransferase of Strep- talytic group I intron RNA, Science 260, 1500-
tomyces griseus, Nucleic Acids Res. 15, 1819- 1503.
1834. VON AHSEN, U., DAVIES,J., SCHROEDER, R.
TRUSCHEIT, E., FROMMER, W., JUNGE,B., MUL- (1991), Antibiotic inhibition of group I ribozyme
LER, L., SCHMIDT,D. D., WINGEDER,W. function, Nature 353, 368-370.
(1981), Chemistry and biochemistry of bacterial VUJAKLIJA, D., UEDA, K., HONG,S., BEPPU,T.,
alpha-glucosidase inhibitors, Angew. Chem. (Int. HORINOUCHI, S. (1991), Identification of an A-
Edn.) 20, 744-761. factor-dependent promotor in the streptomycin
TSUNO,T., IKEDA,C., NUMATA,K.-I., TOMITA, biosynthetic gene cluster of Streptomyces griseus,
K., KONISHI,M., KAWAGUCHI, H. (1986), 3,3'- Mol. Gen. Genet. 229, 119-128.
I 1 References 487
VUJAKLIJA,D., HORINOUCHI, S., BEPPU, T. WONG,Y.-H. H., SHERMAN, W. R. (1985), Anom-
(1993), Detection of an A-factor-responsive pro- eric and other substrate specificity studies with
tein that binds to the upstream activation se- myo-inositol-1-P synthase, J. Biol. Chem. 260,
quence of strR, a regulatory gene for streptomy- 11083-1 1090.
cin biosynthesis in Streptomyces griseus, J. Bacte- WRIGHT,J. L. C. (1983), The lincomycin4elestice-
riol. 175,2652-2661. tin-anthramycin group, in: Biochemistry and Ge-
WALKER,J. B. (1975a), Pathways of the guanidin- netic Regulation of Commercially Important An-
ated inositol moieties of streptomycin and tibiotics (VINING,L. C., Ed.), pp. 311-328. Lon-
bluensomycin, Methods Enzymol. 43,429-470. don: Addison-Wesley.
WALKER,J. B. (1975b), ATP:streptomycin 6-phos- YAMAUCHI, N., KAKINUMA, K. (1992a). Biochemi-
photransferase, Methods Enzymol. 43, 628-632.
cal studies on 2-deoxy-scyllo-inosose an early in-
WALKER,J. B. (1980), Biosynthesis of aminoglyco- termediate in the biosynthesis of 2-deoxystrept-
side antibiotics, Dev. Ind. Microbiol. 21, 105- amine. I. Chemical synthesis of 2-deoxy-scyllo-
113. inosose and [2,2-ZHz]-2-deoxy-scyllo-inosose, J.
WALKER,J. B. (1990), Possible evolutionary rela- Antibiot. 45,756-766.
tionships between streptomycin and bluensomy-
cin biosynthetic pathways: Detection of novel YAMAUCHI, N., KAKINUMA, K. (1992b), Biochemi-
inositol kinase and 0-carbamoyltransferase ac- cal studies on 2-deoxy-scyllo-inosose an early in-
tivities, J. Bacteriol. 172,58445851. termediate in the biosynthesis of 2-deoxystrept-
WALKER,J. B. (1995), Enzymatic synthesis of ami- amine. 11. Quantitative analysis of 2-deoxy-scyl-
nocyclitol moieties of aminoglycoside antibiotics lo-inosose, J. Antibiot. 45, 767-773.
from inositol by Streptomyces spp.: detection of YAMAUCHI, N., KAKINUMA, K. (1992c), Confirma-
glutamine-aminocyclitol aminotransferase and tion of in vitro synthesis of 2-deoxy-scyllo-inos-
diaminocyclitol aminotransferase activities in a ose, the earliest intermediate in the biosynthesis
spectinomycin producer, J. Bacteriol. 172,5844- of 2-deoxystreptamine, using cell free prepara-
5851. tions of Streptomyces fradiae, J. Antibiot. 45,
WALKER,J. B., SKORVAGA, M. (1973), Phosphory- 774780.
lation of streptomycin and dihydrostreptomycin, YAMAUCHI, N., KAKINUMA, K. (1993), Biochemi-
J. Biol. Chem. 248, 2435-2440. cal studies on 2-deoxy-scyllo-inosose an early in-
WALKER,J. B., WALKER,M. S. (1975), ATP:strep- termediate in the biosynthesis of 2-deoxystrept-
tomycin 3 " -phosphotransferase, Methods Enzy- amine. VI. A clue to the similarity of 2-deoxy-
mol. 43, 632-634. scyllo-inosose synthase to dehydroquinate synth-
WALLACE,B. J., TAI, P.-C., DAVIS,B. D. (1979), ase, J. Antibiot. 46, 1916-1918.
Streptomycin and related antibiotics, in: Anti- YAMAUCHI,N., KAKINUMA, K. (1995), Enzymatic
biotics, Vol. V-1, Mechanism of Action of Anti- carbocycle formation in microbial secondary me-
bacterial and Antitumor Agents (CORCORAN,J. tabolism. The mechanism of the 2-deoxy-scyllo-
W., HAHN,F. E., Eds.), pp. 272-303. Berlin: inosose synthase reaction as a crucial step in the
Springer-Verlag. 2-deoxystreptamine biosynthesis in Streptomyces
WATERS,B., VUJAKLIJA,D., GOLD, M. R., DA- fradiae, J. Antibiot. 60, 5614-5619.
VIES,J. (1994), Protein tyrosine phosphorylation
in streptomycetes, FEMS Microbiol. Lett. 120, YASUZAWA,T., YOSHIDA,M., ICHIMURA,M.,
187-190. SHIRAHATA, K., SANO,H. (1987), CV-1, a new
WELLINGTON,E. M. H., MARSH, P., TOTH, I., antibiotic produced by a strain of Streptomyces
CRESSWELL,N., HUDDELSTON, L., SCHILHA- sp. 11. Structure determination, J. Antibiot. 40,
727-731.
BEL, M. B. (1993), The selective effects of anti-
biotics in soil, in: Trends in Microbial Ecology YOKOSE,K., FURUMAI, T., SUHARA,Y., PIRSON,
(GUERRERO,R.,PEDR6S-AL6, C., Eds.), pp. W. (1989), Trestatin: alpha-amylase inhibitor, in:
331-336. Madrid: Spanish Society for Microbiol- Novel Microbial Products for Medicine and Ag-
ogy. riculture (DEMAIN, A. L., SOMKUTI,G. A.,
WIDLANSKI, T., BENDER,S. L., KNOWLES,J. R. HUNTER-CREVA,J. C., ROSSMOORE,H. W.,
(1989), Dehydroquinate synthase: A sheep in Eds.), pp. 117-126. Amsterdam: Elsevier Science
wolf's clothing? J. Am. Chem. SOC. 111, 2299- Publishers.
2300. ZHANG,H., SCHMIDT, H., PIEPERSBERG, W.
WILLIAMS,S. T., SHARPE,M. E., HOLT, J. G. (1992), Molecular cloning and characterization
(1989), Bergey 's Manual of Systematic Bacterio- of two lincomycin-resistance genes, lmrA and
fogy, Vol. 4. Baltimore, ML: Williams and Wil- ImrB, from Streptomyces lincolnensis 78-11, Mol.
kins. Microbiol. 6, 2147-2157.
ZHOU, Z.-Y., SAKUDA, S., YAMADA, Y . (1992), ZHOU, Z.-Y., SAKUDA, S., KINOSHITA, M., YA-
Biosynthetic studies on the chitinase inhibitor, MADA, Y. (1993), Biosynthetic studies of allos-
allosamidin. Origin of the carbon and nitrogen amidin. 2. Isolation of didemethylallosamidin,
atoms, J. Chem. SOC. Perkin Trans. I 1992,1649- and conversion experiments of 14C-labeled de-
1652. methylallosamidin, didemethylallosamidin and
their related compounds, J. Antibiot. 46, 1582-
1588.
Biotechnology
11 Products from Basidiomycetes
GERHARDERKEL
TIMMANKE
Kaiserslautern, Germany
1 Introduction 490
2 Cultivation of Basidiomycetes 490
3 Primary and Secondary Metabolites from Basidiomycetes - Biosyntheses and Possible
Functions 491
4 Screening Methods Used for the Detection of Potentially Useful Metabolites 496
5 Bioactive Metabolites from Basidiomycetes 496
5.1 Pleuromutilin (Tiamulin) 496
5.2 The Strobilurins and Oudemansins 496
5.2.1 Mode of Action - Selective Toxicity 497
5.2.2 Structure-Activity Relationships - Development of Agricultural Fungicides 500
5.2.3 Biosynthesis 500
5.2.4 Possible Functions of Strobilurins and Oudemansins in the Producing
Fungi 501
5.3 Other Antibacterial and Antifungal Metabolites 501
5.4 Cytotoxic and Antitumor Metabolites 503
5.4.1 Antitumor Polysaccharides 508
5.4.2 Immunosuppressive Metabolites 509
5.5 Antiviral Compounds and Inhibitors of Reverse Transcriptases 509
5.6 Inhibitors of Platelet Aggregation 512
5.7 Herbicidal Compounds 513
5.8 Insecticidal and Nematicidal Metabolites 515
5.9 Inhibitors of Cholesterol Biosynthesis 518
5.10 Inhibitors of Aminopeptidases 518
5.11 Inhibitors of Phospholipases C and A2 519
5.12 Inhibitors of (Na+-K+)-ATPases 520
5.13 Addendum 521
5.13.1 Inhibitors of Leukotriene Biosynthesis 522
5.13.2 Inducers of Differentiation of Promyelocytic Leukemia Cells and Inhibitors of
Signal Transduction in Tumor Cells 524
6 Future Perspectives 525
7 References 526
490 11 Products from Basidiomycetes
1 Introduction chemical entities from these organisms. Be-

cause of that and of recent progress in fer-
mentation technology, product recovery, and
The basidiomycetes (mushrooms) consti- spectroscopy for structural analysis new in-
tute a large class of fungi and are estimated to vestigations of other organisms seem to be at-
consist of 30000 species (MOLLER and tractive again. Among these are rare Actino-
LOEFFLER,1982) which is approximately one mycetales, gliding bacteria, marine organisms,
third of all fungi known. Since ancient times and some taxa of higher fungi including ba-
many of them were used as food (e.g., bole- sidiomycetes.
tuses, chantarelles, Agaricus spp.) or for cul- In the following chapter an overview of
tural purposes (hallucinogenic mushrooms). bioactive metabolites from mycelial cultures
One of the first to describe pharmacological of basidiomycetes with special emphasis on
and toxic activities of fruiting bodies was antibiotics, cytotoxic, and antitumor com-
PLINIUSSECUNDUS(A.D. 23-79). Although pounds is given. In some cases secondary me-
a direct use of the fruiting bodies was com- tabolites were obtained from fruiting bodies.
mon practice all over the world, a detailed Normally, basidiomycetes do not form fruit-
study of their contents and the metabolites ing bodies under laboratory conditions.
produced by cultured mycelia started only in Therefore, only the larger mushrooms can be
this century with FLEMING’S discovery of the used in chemical or biological investigations,
imperfect fungus Penicillium notatum as the and they have been studied intensively for the
producer of penicillin, the first antibiotic occurrence of toxins (reviewed by BRESINSKY
metabolite for the treatment of bacterial in- and BESL, 1985), hallucinogens (reviewed by
fections in humans. This led to an intensive SCHULTESand HOFMANN,1980), and pig-
search for new antibiotics produced by other ments (reviewed by GILL, 1994; GILL and
microorganisms, especially easily available STEGLICH,1987).
soil-living forms. The pioneers in the search
for antibiotics from basidiomycetes are AN-
CHEL,HERVEY,WILKINSand their cowork-
ers who investigated extracts from fruiting
bodies and mycelial cultures of approximately
2000 species (for a review, see FLOREY et al.,
1949). Their outstanding work offered a first 2 Cultivation of
glance at the basidiomycete chemistry and
succeeded in the isolation of pleuromutilin Basidiomycetes
(KAVANAGH et al., 1951), the first basidiomy-
cete metabolite to serve later as a leading The life cycle of a typical basidiomycete
structure for the development of a commer- (e.g., Agaricus campestris, which does not
cial antibiotic. The work on antibiotic produc- form conidia and is devoid of a yeast phase)
ing basidiomycetes was almost completely starts with haploid basidiospores germinating
discontinued following WAKSMAN’S discove- to form haploid mycelia, which - if compati-
ry of the streptomycetes as most promising ble - fuse to give rise to dikaryotic mycelia
antibiotic producers. These bacteria can be from which fruiting bodies are derived. Ka-
obtained from soil samples and grow easily in ryogamy and meiosis take place in the basidia
a variety of technical media. Up to now, the located in the hymenium (e.g., lamellae or
worldwide investigation of Streptomyces and pores) of the fruiting body. Usually, 4 haploid
related genera resulted in more than 6000 me- basidiospores are formed.
tabolites, many of them being used as antibio- Basidiomycetes are usually collected as
tics or for other pharmacological purposes. In fruiting bodies from their natural substrate:
spite of the incredible wealth of structures dead or living plants, soil or dung. Cultures
and activities which can to be found in strep- can be derived either from spores (haploid or
tomycetes and imperfect fungi it now has be- dikaryotic mycelia) or tissue plugs (dikaryotic
come increasingly difficult to find novel mycelia) which can germinate and grow on
3 Primary and Secondary Metabolites of Basidiomycetes 491
complex media, typically containing yeast ex-

tract or peptone as a nitrogen source and glu-
3 Primary and Secondary
cose, maltose, or malt extract as a carbon
source. A medium commonly used in the au-
Metabolites from
thors’ laboratory consists of 4 g glucose, 4 g Basidiomycetes -
yeast extract, 10 g malt extract, water to 1 L,
pH adjusted to 5.5. The same media usually Biosyntheses and Possible
are suitable for submerged cultivation either
in Erlenmeyer flasks or in fermenters. Metab- Functions
olite diversity and production are mainly de-
pendent on the biosynthetic capabilities of Secondary metabolites show diverse chemi-
the strain and on the fermentation conditions. cal structures that are often quite different
From the authors’ experience, these condi- from the primary metabolites (such as amino
tions can only be varied to a limited extent acids, acetyl coenzyme A, sugars, mevalonic
since many strains are highly sensitive to acid, and intermediates from the shikimic acid
shear stress imposed by the impellers, and pathway) from which they are synthesized.
media have to be chosen so as to permit suffi- The starting point from primary metabolism
ciently fast growth. These problems were ad- is the basis of classification according to their
dressed by CHENINAet al. (1993), GERMER- biosynthetic precursors (TURNER, 1971;
DONK et al. (1993), and BRAUERand KORN TURNERand ALDRIDGE, 1983). As shown in
(1993). The modeling of a basidiomycete fer- Fig. 1 the main branching points leading to
mentation was achieved by HAS and Mu- secondary products are (HERBERT,1989):
NACK (1993). An example for a detailed de-
scription of a technical-scale process is the - acetyl coenzyme A, leading to polyketides,
production of the antibiotic pleuromutilin polyins, terpenoids, steroids, or carote-
(KNAUSEDERand BRANDL,1975; SCHNEI- noids;
DER and MOSER,1987). Several peculiarities - shikimate, from which aromatic compounds
in the cultivation of basidiomycetes are com- can be derived;
monly encountered: spores of many species, - amino acids, which serve as precursors for
e.g., from the genera Znocybe and Russula, do peptides and alkaloids;
not germinate, and no growth from tissue - glucose, for the biosynthesis of glycosides
plugs can be observed. Many mycelial cul- and aminoglycosides.
tures grow very slowly on solid media or in
submerged cultures. Fermentation times Due to the widespread distribution of the
range from one to several weeks. Suitable biosynthetic pathways mentioned above, re-
methods for the preservation of cultures are lated secondary metabolites, like polyketides,
keeping agar slants at 4°C with periodical steroids, and terpenoids have been isolated
transfers (e.g., once a year) or storage in li- from bacteria, plants, fungi, and animals
quid nitrogen. Most cultures lose viability aft- (BEALE, 1990 JANSEN and DE GROOT,
er freeze-drying. 1991).
The secondary metabolism of basidiomy-
cetes is rich in terpenoids, especially sesqui-
terpenoids (AYERand BROWNE,1981) and
polyacetylenes (JONESand THALLER,1973).
Many of these possess structures which up to
now have only been detected in this class of
fungi, whereas others closely resemble plant
metabolites (FRAGA,1990).
Illudin M and illudin S (1 and 2, Fig. 2)
were two of the first highly antimicrobial and
cytotoxic metabolites of basidiomycetes de-
tected by the screenings of ANCHELet al.
492 I1 Products from Basidiomycetes
Rimm and inteamdim metabolism Secondarv metabolism

................................................................. Glycosides (e. g. Schizonellins)
I 1 1Polysaccharides
1 El +
Teeixe ,
Kojic acid etc.
Trio%-P
I\
Shikimate
1
1
Aromatic amino acids 1
- ,
Aromatic compounds(e. g. Strobilurins)
Alkaloids (e. g. Chalciporon)
Oligopeptides (e. g. Amanitins)
Pyruvate \: , Aminoacids _1 Depsipeptides(e. g. Beauvericin)
I / -
Modified Amino acids (e. g. metabolites
wonylcoL
of Clitocybe acromelalga)
Fatty acids Acetylenes (e. g. Siccayne)
Acetyl-CoA ...................................... Polyketides (e. g. Merulinic acids)
hi ?-
Citric acid cycle
\
MewIonic acid - -
- Fyesy-PP
Geraayl-PP
Sesquitemnoids (e. g. A h c o b )
Diterpenoids (e. g. Striatins)
+ Sesterterpenoids(e. g. Fasciculols)
Amino acids Steroids
Fig. 1. Interrelationship between primary and secondary metabolism.
p-
R = H IlludinM(1)
R = OH Illudin S (2) Fig. 2. Biosynthesis of illudins.
(1950). The producing mushrooms, Clitocybe and xylose involves a loss of protons, intra-
illudens and Lampteromyces japonicus, are molecular cyclization, and acylation. Similar
highly toxic to humans. Illudin S is considered mechanisms may be assumed for the forma-
to be the toxic principle. The first information tion of the related metabolites dihydrostriatal
on stages in sesquiterpenoid biosynthesis C and hericin by H. ramosum.
were obtained by incorporation of radiola- It has been proposed that molecules related
beled mevalonic acid into illudins S and M to secondary metabolites played important
from a head-to-tail condensation of isopente- roles in biochemical evolution as modulators
nyl pyrophosphate and dimethylallyl pyro- or effectors, enhancing or controlling biologi-
phosphate by stationary cultures of Clitocybe cal activities of primitive macromolecules. A
illudens with a humulene-type precursor as number of antibiotics which inhibit transla-
the first cyclic intermediate (MCMORRISand tion, such as aminoglycosides, interact with ri-
ANCHEL, 1965; HANSON and MARTEN, bosomes by directly binding to specific RNA
1973). PRICE and HEINSTEIN(1978) con- conformations. It has been suggested that
firmed this biosynthetic pathway (Fig. 2) by these secondary metabolites served as effec-
using a cell-free homogenate of Clitocybe il- tors of translation and other ribozyme-cata-
ludens which incorporated labeled pyrophos- lyzed reactions in early stages of evolution
phorylated isoprenyl alcohols into illudins. (DAVISet al., 1992). From another viewpoint
Another example of metabolites apparent- secondary metabolism might be an evolution-
ly occurring in basidiomycetes only are the di- ary playground from which new and useful
terpenoids - striatins A, B, C, D - and the biogenetic pathways could evolve (ZAHNER,
corresponding striatals (Fig. 3) which are an- 1982).
tibiotic and cytotoxic products of Cyathus The biological role of secondary metabol-
striatus, C. poeppigii, C, limbatus, C. montag- ites is still a matter of debate. Secondary me-
nei, and Gerronema fibula (ANKE et al., tabolites might be beneficial to producing or-
1977a;HECHTet al., 1978). Striatins and stria- ganisms several ways: improving their ability
tals consist of a cyathan moiety with an at- to grow, reproduce, or disperse under appro-
tached pentose unit. In a resting cell system priate conditions, or affording protection
I4C- and I3C-labeled precursors were incor- against competitors or predators. The majori-
porated into striatins and striatals (RABE, ty of these compounds fit into these catego-
1989). Feeding experiments with l-I3C-glu- ries (VINING,1990). Among the growth-sup-
cose and 2-I3C-glucose and subsequent analy- porting substances are the sideramines (ferric
sis by NMR spectroscopy revealed that the ion-chelating compounds) and related metab-
pentose unit was formed by decarboxylation olites which, in association with specific re-
of C-6 of glucose (70%) and to a smaller ex- ceptors, play an essential role in solubilization
tent (30%) via the pentose phosphate cycle and uptake of iron. Fungal sideramines have
(Fig. 3). The labels observed in the cyathan been isolated from cultures of Penicillium sp.,
skeleton were consistent with the formation Neurospora crassa, some Fusarium strains,
of acetate via glycolysis and subsequent syn- Ustilago sp., and the basidiomycetous yeast
thesis of mevalonate. Further cyclization of Rhodotorula pilimanae (WINKELMANN,
geranyl-geranyl pyrophosphate to the cya- 1986). Their biosynthesis is regulated by the
than skeleton occurred according to the concentration of soluble iron in the substrate.
mechanism of the cyathine formation as de- As has been shown for rhodotorulic acid syn-
scribed by AYERet al. (1979). Herical, an an- thetase (ANKEand DIEKMANN, 1972) and fu-
tibiotic metabolite of Hericium ramosum con- sigen synthetase (ANKEet al., 1973) these key
sists of a cyathan moiety and D-xylose at- enzymes could not be detected as long as suf-
tached to it by a glycosidic bond. When 14C- ficient iron was present in the culture media.
herical was prepared and fed to resting cells The characteristics of fungal sideramine bio-
of C. striatus it was readily incorporated into synthesis is similar to the non-ribosomal bio-
the striatals A and B. Herical is thus consid- synthesis of other peptide antibiotics (KLEIN-
ered as a direct precursor of striatals. The for- KAUF and VON DOHREN,1987). Recently, it
mation of the C-C bond between aglycone has been shown that wood decaying basidio-
PP Acctyl-CoA
Oxid. and Ueeorh. C-6

4Patose phosphotc Cycle
Cyathus striatus 70 30 perccnt
Cyathie
cyathus spp.
I D-Xylose
Striatine A, B, C, Cyathus spp.
OH
Herical, Hericium ramosum
Striatal D, G m n e m a fibula
8327-540, H.ramoaum
I
H
8327-503, H. m o s u m
Fig. 3. Biosynthesis of striatals.

mycetes (brown and white rots) produce side- 1987a). In addition, isovelIeral is a potent an-
rophores of the phenolate type (JELLISON et tifeedant for mammals that normally feed on
al., 1990). mushrooms. Further reduction of an aldehyde
The enzymatic transformation of sesquiter- group converts isovelleral to isovellerol (5,
penes in various species of Russulaceae is an Fig. 4) the mutagenicity, pungency, and an-
example for a proposed chemical defense sys- timicrobial activities of which are diminished
tem that preserves the fruiting bodies from at- or lost upon reduction. It seems probable that
tack by parasites and microorganisms (STER- injured fruiting bodies reduce the unsaturated
NER et al., 1985). The fruiting bodies of Lac- dialdehydes in order to avoid prolonged con-
carius vellereus contain large amounts of stea- tact with their own defense chemicals.
roylvelutinal (3, Fig. 4) which is transformed Another example of a chemical defense
to the unsaturated dialdehyde isovelleral a mechanism in basidiomycetes are azepin deri-
few seconds after injury (4, Fig. 4). While vatives occurring in fruiting bodies of Chalci-
stearoylvelutinal appears to have weak bio- porus piperatus. The main component chalci-
logical activity, isovelleral, like other unsatu- poron (6, Fig. 4) exhibits antibacterial and an-
rated dialdehydes, has strong antimicrobial as tifungal activity. Due to its strong pungency
well as mutagenic properties (STERNERet al., chalciporon is considered to be responsible
Stearoylvelutid(3) Isovelleral(4) Isovellerol(5)
Isochalciporon(7)
Chalciporon (6)
Laccarius vellerus and-

Chalciporus piperatus. Pleuromutiln (8) Tiarnulin (9)
for the antifeedant acitivity. In solution chal- in the development of tiamulin (9, Fig. 4)
ciporon is converted to isochalciporon (7, which exceeds the activity of the parent com-
Fig. 4) which still shows antibiotic activity but pound against gram-positive bacteria and my-
has lost pungency (STERNERet al., 1987b). coplasms by a factor of 10-50. The minimal
inhibitory concentrations (MIC) for different
strains of Mycoplasma were in the range of
0.0039-6.25 p.g/mL-' (DREWSet al., 1975).
Studies on the mode of action revealed that
4 Screening Methods pleuromutilin and its derivatives act as inhibi-
tors of prokaryotic protein synthesis by inter-
Used for the Detection of fering with the activities of the 70 S ribosomal
subunit. The ribosome-bound antibiotics lead
Potentially Useful to the formation of inactive initiation com-
plexes which are unable to enter the peptide
Metabolites chain elongation cycle (HOGENAUER, 1979).
In various bacteria drug resistance is devel-
Culture fluids or mycelial extracts of basid- oped stepwise. In some E. coli mutants the ri-
iomycetes are amenable to all screening bosome has lost its binding ability for tiamu-
methods applied to other microorganisms. lin. Because of its outstanding properties tia-
Routine test systems comprise bacteria, fil- mulin is currently used for the treatment of
amentous fungi or yeasts, human and rodent mycoplasma infections in animals.
cell lines, viruses, plants, and several enzyme Pleuromutilin can be produced by fermen-
assays. tation in a medium composed of 50 g glucose,
50 g autolyzed brewer's yeast, 50 g KH2P04,
0.5 g MgS04x 7 H20, 0.5 g Ca(N03), 0.1 g
NaCl, 0.5 g FeS04 x 7 H 2 0 , water to 1 L, pH
6.0. The yield after 6 d of growth in a 1000 L
5 Bioactive Metabolites fermenter was reported to be 2.2gL-'. It
could be demonstrated that during fermenta-
from Basidiomycetes tion of pleuromutilin derivatives differing in
the acetyl portions attached to the 14-OH
group of mutilin were formed. The biosynthe-
5.1 Pleuromutilin (Tiamulin) sis of these derivatives was strongly stimu-
lated by addition of corn oil as a carbon
So far, the only commercial antibiotic pro- source during fermentation (KNAUSEDER
duced by a basidiomycete is the diterpene and BRANDL,1976). Pleuromutilin overpro-
pleuromutilin (8, Fig. 4). Pleuromutilin was ducers were obtained by conventional muta-
first isolated from Pleurotus mutilus and Pleu- genesis and selection programs as well as by
rotus passeckerianus in a screening for anti- protoplast fusion and genetic studies (STE-
bacterial compounds (KAVANAGHet al., WART,1986).
1951). The structural formula was elucidated
by ARIGONI(1962) and BIRCHet al. (1963,
1966). In 1963, BRANDLet al. (KNAUSEDER 5.2 The Strobilurins and
and BRANDL,1976) isolated an antibiotic
from Clitopilus passeckerianus which was Oudemansins
identical with pleuromutilin. Pleuromutilin is
active against gram-positive bacteria, but the Initially, the strobilurins A (10, Fig. 5) and
most interesting biological activity is its high B (11, Fig. 5) were isolated from cultures of
effectiveness against various forms of myco- Strobilurus tenacellus, a small and very com-
plasms. The preparation of more than 66 der- mon edible mushroom growing on buried
ivatives of pleuromutilin by RIEDL(1976) and pine cones in early spring (ANKE et al.,
EGGERand REINSHAGEN (1976a, b) resulted 1977b). Both compounds showed a remarka-
ble activity against a variety of filamentous (19, Fig. 5) being the most active. In HeLa S3
fungi and yeasts but no antibacterial effects. cells (human) these are accompanied by a
The structure elucidation by W. STEGLICH'S 30% drop of the cellular ATP content and a
group (SCHRAMMet al., 1978; ANKEet al., change in morphology. The observed antivi-
1984) revealed that both compounds be- ral effects of strobilurin E (vesicular stomati-
longed to a new class of antifungal antibiotics. tis virus in baby hamster kidney cells) are
Close similarities between strobilurin A and probably due to an inhibition of host cell
mucidin, an antifungal antibiotic previously growth (WEBERet al., 1990a).
isolated from cultures of the wood-inhabiting
basidiomycete Oudemansiella mucida, were
recognized. Mucidin, however, had been de- 5.2.1 Mode of Action - Selective
scribed as a dextrarotatory crystalline com-
pound (MUSILEK,1969). Fermentations of Toxicity
the Oudemansiella mucida strains used in the
author's laboratory yielded in addition to Respiration in fungi and other eukaryotes
strobilurin A a new antifungal antibiotic, ou- is completely blocked by strobilurin A and
demansin A (23, Fig. 5) (ANKEet al., 1979). oudemansin A. In Ehrlich ascitic carcinoma
Strobilurin A and mucidin were claimed to be cells (ECA, mouse) syntheses of macromole-
identical by SEDMERAet al. in 1982 and this cules (proteins, RNA, DNA) are inhibited
was finally proven by VON JAGOW et al. due to a depletion of their ATP pool caused
(1986) in a direct comparison of both com- by the inhibition of oxidative phosphoryla-
pounds. In the meantime, numerous strobil- tion. Upon addition of glucose this effect is
urins and oudemansins were isolated from completely reversed since ATP supply by gly-
many genera of basidiomycetes (Fig. 5 ) , from colysis seems to be sufficient in these cells. In
tropical as well as from temperate regions; rat liver mitochondrial oxygen uptake and
among them were many Mycena species ATP synthesis were blocked by both a-keto-
(BAUERLEand ANKE,1980; BAUERLE,1981) glutaric acid and succinate as substrates which
(Tab. 1). gave the first evidence of a target within the
Surprisingly, several strobilurins were also respiratory chain (ANKEet al., 1979). This
isolated from an ascomycete, Bolinea lutea molecular target was precisely identified by
(FREDENHAGEN et al., 1990a, b). Strobilurins VON JAGOWand coworkers (BRANDTet al.,
seem to be of worldwide occurrence. Most of 1988, 1993). Strobilurins and oudemansins
their producers grow on wood or decaying specifically inhibit the ubiquinol oxidation
plant material. (Qp center) of the mitochondrial bcl complex
Strobilurins and oudemansins inhibit fun- (Fig. 6).
gal growth at very low concentrations (lo-'- Like other specific inhibitors strobilurins
lo-") (ANKE et al., 1977b, 1979, 1983; and oudemansins have become valuable tools
BACKENS et al., 1988 WEBERet al., 1990a, b; for the development of a more detailed mod-
ANKEet al., 1990; ZAPFet al., 1994) without el of the structure and function of their target.
any significant antibacterial activity. Weak In heterocysts of cyanobacteria the cyto-
phytotoxic activity of several strobilurin and chrome b/f complex, which utilizes light ener-
oudemansin derivatives have been demon- gy to generate the proton gradient used for
strated (SAUTERet al., 1995). Insecticidal ac- ATP synthesis and transfers electrons to ni-
tivity of strobilurin A against adults and lar- trogenase, the key enzyme of nitrogen fixa-
val stages of Epilachna varivestis (Mexican tion, plays a central role. Using heterocysts of
bean beetle), Aphis fabae (aphid), and Tetra- an Anabaena sp. HOUCHINS and HIND(1983)
nychus urticae (mite) were also found at con- found that strobilurin A inhibited the elec-
centrations of 10-4-10-5 M (HOLST,Univer- tron flow from reduced plastoquinone to the
sity of GieSen, personal communication, cytochrome b/f complex in a similar way as in
1978). the mitochondrial bcl complex. The similarity
Reversible cytostatic activity has been de- of the b/f complexes in cyanobacteria and in
scribed for all strobilurins, with strobilurin E higher plants, where the b/f complex plays a
C
H
=
;R2
3" H3CW
'p" O n H3COOC
; C H 3
StrobilurinC (15)
Rl R2
H H, StrobilurinA(10)
Me0 Cl, Strobilurin B (1 1)
HO H, StrobilurinFl (12) L o n ; H3COOC
C H 3
Me0 H, StrobilurinH(13)
H Me0, Strobilurin X (14) StrobilurinF2 (1 6)
StrobilurinD* (17) A HydroxystrobilurinD* (18)
OCH3
StrobilurinE (19) StrobilurinG (20)
0ch3
9-MdhO~StrObildA (2 1) 9-MethoxystrobilurinK* (22)
Oudemansin B (24)
QCH3
H3C0 m ;H3COOC
C H 3
Oudemansin X (25)
Fig. 5. The strobilurins and oudemansins. * The structures of the side chains are currently under investiga-
tion.
Fig. 5 BAS 490 P (2.6) ICW504(27)
Tab. 1. Fungi Producing Strobilurins and Oudernansins

~~
Producer Compound References
Basidiomycetes
Agaricus sp. 89139 10, 12, 17 ZAPF(1994)
Crepidotus 19 WEBERet al. (1990a)
fulvomentosus
Cyphellopsis anomala 10, 12, 17 WEBERet al. (1990b)
Favolaschia sp. 87129 10, 12, 17, 19, 21, 22, 23 ZAPFet al. (1995)
Filoboletus sp. 9054 19 SIMON(1994)
Hydropus scabripes 10 BAUERLE(1981)
Mycena aetites 10 BAUERLE(1981)
M. alkalina 11 BAUERLE(1981)
M. atromarginata 10 BAUERLE(1981)
M. avenacea 11 BAUERLE(1981)
. M. cf. capillaripes 10 BAUERLE(1981)
M. crocata 11 BAUERLE(1981)
M. fagetorurn 10 SCHRAMM et al. (1978)
M. galapoda 10 BAUERLE(1981)
M. galopoda var. alba 10 BAUERLE(1981)
M. oregonensis 10 BAUERLE(1981)
M. polygramma 23 BAUERLE(1981)
M. purpureofima 10 BAUERLE(1981)
M. rosella 10 BAUERLE(1981)
M. sanguinolenta 18 BACKENS et al. (1988)
M. vitilis 11 BAUERLE(1981)
M. zephirus 10 SCHRAMM et al. (1978)
Oudemansiella mucida 10,23 ANKEet al. (1979)
0. radicata 10,25 ANKEet al. (1990)
Strobilurus conigenoides 10 ANKE,unpublished data
S. esculentus 10 ANKEand STEGLICH (1981)
S. tephanocystis 10 ANKE,unpublished data
S.tentacellus 10,ll ANKEet al. (1977b)
Xerula longipes 11, 15 ANKEet al. (1983)
X. melanotricha 10, 11, 24 ANKEet al. (1983)
Ascomycete
Bolina lutea 10, 11, 13, 16, 20 FREDENHAGEN
et al. (1990a, b)
Fig. 6. The Q-cycle mechanism of the bcl-

complex and the mode of action of strobil-
urins and oudemansis (BRANDTet al.,
1993).
Q-Pool: ubichinone pool; Q, Q :, QH,:
ubiquinone, ubisemiquinone, ubihydroqui-
none; c, cl: cytochrome cl; b566: “low po-
tential” heme b; b562:“high potential”
heme b FeS: iron-sulfur protein.
central role in cyclic photophosphorylation 5.2.2 Structure-Activity

and in coupling photosystems 1 and 2, might
contribute, together with an inhibition of mi- Relationships - Development of
tochondrial respiration, to the phytotoxic ac- Agricultural Fungicides
tivity observed for some strobilurin deriva-
tives (SAUTERet al., 1995). Extensive synthetic efforts lead to simple
The mitochondria1 bcl complex, the target mimics and revealed that the E-p-methoxy-
of strobilurins and oudemansins, is common acrylate unit is a prerequisite for the antifun-
to many eukaryotic taxa. Mitochondria1 pre- gal and respiration inhibiting properties of
parations of rat liver, beef heart, house fly, strobilurins and oudemansins (SCHRAMM,
and corn are all sensitive to strobilurins 1980; SCHRAMM et al., 1982; T. ANKEet al.,
(ANKE et al., 1979; BRANDTet al., 1993; 1988). Continuous efforts by STEGLICHand
SAUTERet al., 1995). Surprisingly, strobilurin coworkers (ANKEand STEGLICH,1989) and
A and other synthetic mimics exhibited no by BASF and ICI resulted in compounds with
toxicity to rodents (J. DOUROS, NCI, USA, improved activity and light stability and lead
unpublished SAUTERet al., 1995). In fact, to the development of BAS 490 F (26, Fig. 5)
“mucidermin”, a preparation which apparent- (SAUTER et al., 1995) and ICIA5504 (27,
ly contains strobilurin A (mucidin) was mar- Fig.5) (CLOUGH,1993), which will be com-
keted by Spofa, CSFR, for the treatment of mercialized in the near future by ICI and
dermatomycoses in humans. This lack of tox- BASF.
icity is probably due to enzymic degradation
of the shobilurins by, e.g., mammalian ester-
ases before reaching their target. 5.2.3 Biosynthesis
The biosynthesis of mucidin (28, Fig. 7)
(the E, E, E geometry was revised to E, 2, E
of strobulin A by VON JAGOWet al., 1986)
was investigated by NERUDet al. (1982) by
feeding isotopically labeled phenylalanine,
da a markedly increased rate of respiration is

likely to confer a greater resistance.
Genetic characterization of the exon-in-
tron organization, the deduced amino acid se-
quence of the cytochromes b from S. tenacel-
lus, M . galopoda, and M. viriginata (which
does not produce strobilurin A), and a com-
Mucidin (28) parative sequence analysis of two regions of
cytochrome b contributing to the formation
V of the Qp center as demonstrated by VON JA-
C%C02Na COW and coworkers (KRAICZYet al., 1996)
7 revealed, that the generally lower sensitivity
CH.,C02Na
of all three basidiomycetes was due to the re-
placement of a small amino acid residue in
position 127 by isoleucine. For M. galopoda
replacement of glycine-143 by alanine and
glycine-153 by serine, and for S. tenacellus re-
placement of a small residue in position 254
Fig. 7. Biosynthesis of mucidin (NERUDet al., by glutamine and asparagine-261 by aspartate
1982). E, E, E, geometry has been revised to E, Z, were assumed to cause resistance to
E, of strobilurin A; isotopically labeled carbon E-p-methoxyacrylates. The latter exchange is
atoms in the precursors acetate, benzoic acid, and also found in Schizosaccharomyces pombe
methionine are marked V,V,0, and *. which shows a natural resistance to
E-p-methoxyacrylates.
On the other hand, it was demonstrated
benzoic acid, acetate, and methionine. The that Oudemansiella mucida produces oude-
aromatic part of the molecule and the benzyl- mansin A together with strobilurin A on ster-
ic carbon atom are derived from the shiki- ilized beech wood which is its natural sub-
mate pathway. The side chain consists of ace- strate (SCHWITZGEBEL, 1992). These findings
tate units, and all three methyl groups are de- and the observation that many basidiomy-
rived from methionine (Fig. 7). cetes belonging to different taxa (Tab. 1) use
the same and obviously quite effective princi-
ple to secure their habitat in very different cli-
5.2.4 Possible Functions of mates and locations suggest that strobilurins
Strobilurins and Oudemansins in and oudemansis play an important role in the
producing fungi.
the Producing Fungi
The biological activities in the producers of 5.3 Other Antibacterial and
strobilurins and oudemansins suggest a possi- Antifungal Metabolites
ble role in the defence of habitats and sub-
strates against competing fungi or predatory Lentinellic acid (29, Fig. 8) from Lentinel-
insects. The use of antifungal antibiotics lus omphalodes and Lentinellus ursinus is a
against competing fungi would require the new protoilludine derivative. Interestingly,
producing fungus to be resistant to its own strains from Europe, USA, and Canada
product. This was clearly demonstrated for all produce the same antibiotic. Lentinellic
Strobilurus tenacellus and Mycena galapoda acid shows strong antibacterial activity
by VON JAGOW and coworkers (BRANDTet with minimal inhibitory concentrations of
al., 1993). In the case of S. tenacellus, binding 1-5 pg mL-' for Bacillus brevis, Aerobacter
of strobilurin A and oudemansin A to the bcl aerogenes, and Corynebacterium insidiosum.
complex was reduced by several orders of Compared to lentinellic acid its methyl ester
magnitude, whereas in the case of M.galopo- exhibits much higher antifungal activity. In
CHO
Lentidlic acid (29) Aleurodiscal(30)
Scorodonin (31) l-Hydroxy-2-mnyn-3-one (32)
3.4, I3-Trihydroxy-tetradeca-5,7,9,ll-tetra~c
acid-Y-lactone (33)
R =H Ficolon(34) Hemimycin (36)

R = OH Hydroxyfimicolon (35)
Fig. 8. Other antibacterial and antifungal metabolites.

ECA cells DNA, RNA, and protein syntheses (BAUERLEet al., 1986). Acoranes have not
are inhibited by 50% using 20 pg mL-' lenti- yet been reported from microbial sources.
nellic acid (STARKet al., 1988). Several re-
lated protoilludane orsellinate esters were
isolated from cultures of Armillaria mellea. 5.4 Cytotoxic and Antitumor
These compounds exhibit weak antibacterial Metabolites
and antifungal activity (OBOUCHIet al., 1990;
YANGet al., 1991). Illudins isolated from Clitocybe illudens
Screening for antifungal compounds re- and Lampteromyces japonicus were two of
sulted in the isolation of aleurodiscal (30, Fig.the first known antitumor metabolites. The
8) from mycelial cultures of Aleurodiscus mi- lifetime of Ehrlich ascites tumor mice was
rabilis (LAUERet al., 1989). Aleurodiscal, a prolonged by illudin S (2, Fig. 2) when given
hydroxysesterterpene aldehyde P-D-xyloside at a dose of 166 pg kg-' i.p. Enclosure of il-
with a novel carbon skeleton, is related to re- ludin S into liposomes markedly enhanced
tigeranic acid A which was isolated from lich- this effect, apparently by decreasing the side
ens (KANEDAet al., 1972). Aleurodiscal pos- effects observed under standard experimental
sesses weak antibacterial activity and strongly conditions (SHINOZAWAet al., 1979). 6-
inhibits the growth of several fungi in the agar Deoxyilludin M (37, Fig. 9) was isolated from
diffusion assay at concentrations of 2-10 pg cultures of Pleurotus japonicus. This com-
per disc. In addition, it causes abnormal pound is closely related to illudin M and was
branching of apical hyphae of Mucor miehei effective against murine leukemia P388,
at a concentration of 1 pg mL-'. Acetylenes showing a 24 % increase of life span at a daily
are strong antifungal metabolites commonly dose of 5 mg kg-' i.p. (HARAet al., 1987).
found in basidiomycetes. They also exhibit Several metabolites of basidiomycetes with
antibacterial and cytotoxic activity (TURNER, strong cytotoxic and antifungal activitiy be-
1983). Examples are scorodonin (31, Fig. 8) long to the sesquiterpenoids with a maras-
from cultures of Marasmius scorodonius mane or isolactarane skeleton. Marasmic acid
(ANKEet al., 1980), l-hydroxy-2-nonyn-4-one (38, Fig. 9) was isolated from cultures of Ma-
(32, Fig. 8) from fermentations of Zschnoder- rasmius conigenus in the course of the first
ma benzoinum (ANKE et al., 1982), and extensive screenings for antibacterial com-
3,4,13,-trihydroxy-tetradeca-5,7,9,1l,-tetraynic pounds conducted by KAVANAGHet al.
acid-y-lactone (33, Fig. 8) from cultures of (1949). The sesquiterpenoid structure of ma-
Mycena viridimarginata (BAUERLE et al., rasmic acid was elucidated by DUGANet al. in
1982). 1966. GREENLEEand WOODWARDachieved
Fimicolon (34, Fig. 8) and hydroxyfimico- the first total synthesis in 1976 and several
lon (35, Fig. 8) of Panaeolus fimicola and Psa- new synthetic approaches have been pub-
thyrella orbitarium are antibiotic and cyto- lished since then (MORISAKIet al., 1980). Ma-
toxic guaianes (ANKEet al., 1985a). Guaianes rasmic acid was also isolated from cultures of
are typical metabolites of higher plants. The Lachnella sp. and Peniophora laeta and exhi-
structures responsible for the biological activ- bits pronounced inhibitory action on nucleic
ity of the Pleurotellus metabolites and fimico- acid syntheses in whole mammalian cells and
lons are similar and consist of a five-mem- on some enzymes of nucleoside metabolism.
bered ring with an exomethylene group adja- In isolated rat liver nuclei the guanylation of
cent to an oxirane ring. mRNA was strongly inhibited by 10 pg mL -'
The antibiotic and cytotoxic compound marasmic acid (KUPKAet al., 1983). The life
hemimycin (36, Fig. 8) obtained from Hemi- span of P388 lymphocytic leukemia mice was
mycena cucullata and H. candida is another prolonged by marasmic acid when given at a
example for the occurrence of the same car- total dose of 3.5 mg kg-' i.p. The LDS0 for
bon skeleton in basidiomycetes and higher tumor bearing mice was determined to be
plants (e.g., Acorus calamus). Hemimycin is 28 mg kg-' i.p. (J. DOUROS,National Cancer
highly oxygenated and contains a double Institute, USA, personal communication). It
bond which easily reacts with nucleophiles was proposed that due to the reactive a$-un-
504 I 1 Products from Basidiomycetes
4YH
%,
Marasmic acid (38)

CHO
CHO
Pilatin (39) Merulidial(40)
striatals
R1 R2
H COCH3
OH COCH3
OH H
HO
nc15H31c00
Schizonellin A (44) Schizonellin B (45)
Fig. 9. Cytotoxic and antitumor metabolites.

0 0
Pleurotellol(47)
Pleurotellic acid (48)

Phellodonic acid (49)
Alliicol A (50) Alliacol B (51)
0
Fig. 9 Fulvofemginin (52)
saturated aldehyde function marasmic acid (HEIMet al., 1988). Pilatin inhibits the growth
covalently binds to nucleophilic (e.g., amino) of bacteria and fungi at concentrations of 5-
groups of enzymes or to nucleic acids. The 50 pg mL-'. It strongly interferes with the
hydroxylated derivative of marasmic acid, 9- DNA and RNA syntheses of ECA cells and
Phydroxymarasmic acid, was isolated by both normal and Rous Sarcoma Virus (RSV)-
H. ANKEet al. (1988). Introduction of a hy- transformed chicken embryo fibroblasts
droxyl function reduces the biological activity (CEF). In addition, pilatin causes frameshift
of marasmic acid, but increases mutagenic ac- mutations in Salmonella typhimurium TA 98.
tivity in the Ames test. In vivo no significant antitumor activity on
Pilatin (39, Fig. 9), a new marasmane deri- P388 lymphocytic leukemia mice was ob-
vative, was isolated from fermentations of served for pilatin. The LDSofor tumor bear-
Flagelloscypha pilatii, a cyphelloid fungus ing mice was determined to be 125 mg kg-'
@
H-+o ( ) I l11111<
l
0 ok
Crinipellin B (54)
OWwn3
CrinipellinA (53)
11111<
IIII,
0 oii
0-aoetylcrinipellinA (55) DhydrocrinipellinB (56)
11111<
Tetrahydrwrinipeh A (57)
GWH3
11111
qgH3
Ii 0
Nematolin (58)
11111 Ii
Nematolon (59)
0
Leianafulvene(60) Fig. 9
i.p. (J. DOUROS,National Cancer Institute, merged cultures of the smut fungus Schizon-
USA, personal communication). ella melanogramma (DEMLet al., 1980). Like
The isolactarane merulidial (40, Fig. 9) was ustilagic acids which are glycolipids of a dif-
isolated from submerged cultures of Merulius ferent type obtained from Ustilago maidis
tremellosus (QUACKet al., 1978). The crystal- (LEMIEUXet al., 1951), schizonellins exhibit
line sesquiterpene dialdehyde very strongly weak antibacterial and antifungal activity. In
inhibits DNA synthesis in Ehrlich ascitic car- ECA cells the incorporation of leucine, urid-
cinoma (ECA) cells at 1 pg ml-'. In the assay ine, and thymidine into protein, RNA, and
of AMESet al. (1975) merulidial exhibits mu- DNA is completely inhibited by 25 pg mL-'
tagenic activity. Comparative studies of meru- of schizonellin A or B. Concomitant lysis of
lidial and several hydroxylated and acetylated the cells suggests a detergent-like mode of ac-
derivatives revealed that the molecular mech- tion.
anism responsible for the mutagenicity of me- Hypnophilin, pleurotellol, and pleurotellic
rulidial is different from the mechanism re- acid (46-48, Fig. 9) were isolated from fer-
sulting in antimicrobial and cytotoxic activity. mentations of Pleurotellus hypnophilus (KUP-
Acetylation of merulidial to S-acetylmerulid- KA et a]., 1981a). While pleurotellol and pleu-
ial, e.g., increases antifungal activity but dim- rotellic acid belong to a new group of sesqui-
inishes mutagenic activity. Hydroxylation of terpenoids, hypnophilin is a new member of
merulidial to 9-a-hydroxymerulidia1 and 9-p- the hirsutane family to which a number of
hydroxymerulidial as well as of acetylmeru- typical basidiomycete metabolites belong. All
lidial to 9-a-hydroxyacetylmerulidial does not three antibiotics exhibit antimicrobial and
strongly affect mutagenic activity but drama- very high cytotoxic activity. However, in com-
tically reduces antimicrobial, cytotoxic, and parison to normal cells no selective toxicity
phytotoxic activity (ANKEet al., 1989). for RSV-transformed chicken embryo fibro-
The striatins A, B, and C (4143, Fig. 9) blasts (CEF) could be detected. Hypnophilin
and the corresponding striatals were isolated and pleurotellol also act as plant growth inhi-
from submerged cultures of Cyathus striatus, bitors. In the Avena coleoptile bioassay they
C. poeppigii, C. limbatus, and C. montagnei. strongly inhibit indole-3-acetic acid-induced
They were also detected in the fruiting bodies growth of coleoptile sections. Like other exo-
(ANKE et al., 1977a). In ECA cells DNA, methylene ketones and lactones Pleurotellus
RNA, and protein syntheses are completely antibiotics very readily form adducts with cys-
inhibited by 2 pg mL-' striatins. RSV-trans- teine or other thiols and they are mutagenic.
formed CEF were found to be inhibited at Phellodonic acid (49, Fig. 9), a new hirsu-
lower concentrations as compared to their tane derivative closely related to hypnophilin,
normal counterparts. Studies on the mode of has recently been isolated from cultures of
action revealed that interference with the Phellodon melaleucus (STADLER et al.,
transport of essential precursors was mainly 1993b). Like hypnophilin, phellodonic acid
responsible for their cytotoxic activity (LEE exhibits antimicrobial and strong cytotoxic ac-
and ANKE,1979). Striatins and striatals were tivity. The incorporation of radiolabeled pre-
found to prolong the life span of P388 lym- cursors into DNA, RNA, and protein of L
phocytic leukemia mice and to be inhibitory 1210 cells is almost completely inhibited at a
in the system colon xenograft-athymic mouse. concentration of 5 pg mL-'.
The LDso for tumor bearing mice was deter- The alliacols A and B (50 and 51, Fig. 9)
mined to be 150 mg kg-' i. p. In greenhouse from Marasmius alliaceus are a,punsaturated
experiments striatins exhibited good fungici- sesquiterpene lactones which exhibit rather
dal activity against Plasmopara viticola on low antimicrobial but highly cytotoxic proper-
grape vine, Phytophtora infestans on pota- ties (ANKEet al., 1981). In ECA cells nucleic
toes, Botrytis cinerea on green pepper, and acid biosyntheses are almost completely inhi-
Septoria nodorum on wheat (ANKE et al., bited at concentrations of 2-10 pg mL-'.
1986). Like other a,punsaturated lactones, alliacols
The schizonellins A and B (44 and 45, readily form adducts with nucleophilic thiols.
Fig. 9) are glycolipids produced by sub- It is assumed that a rapid reaction with SH
groups in enzymes or other proteins is re- termined to be >225 mg kg-' (J. DOUROS,
sponsible for most of the biological activity of National Cancer Institute, USA, personal
these compounds. Deduced alliacolide shows communication).
no antibiotic or cytotoxic properties. Leaianafulvene (60, Fig. 9), an orange-yel-
Fulvoferruginin (52, Fig. 9), a sesquiterpe- low pigment, was isolated from mycelial cul-
noid carotane derivative has been isolated tures of Mycena leaiana ( H A R ~ I G et al.,
from Marasmius fulvoferrugineus (KLEINet 1990). The compound is closely related to the
al., 1990). It is closely related to hercynolac- illudins and represents the first example of a
tone which was isolated from liverworths natural "isoilludane" derivative which may be
(HUNECKet al., 1982). Several carotane ses- formed from an illidane precursor by 1,2-mi-
quiterpenes were isolated from Ferula spe- gration of a methyl group. Leaianafulvene ex-
cies. Fulvoferruginin exhibits modest antibac- hibits weak antibacterial activity, whereas its
terial activity and inhibits the growth of sev- cytotoxic activity is quite pronounced. A 50 %
eral fungi at 5-50 pg mL-'. In ECA cells the lysis of ECA cells is observed at 2.5 pg mL-'.
incorporation of leucine, uridine, and thymid- DNA and RNA syntheses are inhibited by
ine into protein, RNA, and DNA was in- 50% at a leaianafulvene concentration of
hibited by 50% at a concentration of 10 pg ml-'. In addition, mutagenic acitvity
10-50 pg mL-'. was also observed.
Crinipellins obtained from fermentations of
Crinipellis stipitaria are the first known natu-
ral tetraquinanes (KUPKAet al., 1979; ANKE 5.4.1 Antitumor Polysaccharides
et al., 1985b). The crinipellins A and B and
0-acetylcrinipellin A (53-55, Fig. 9) contain- Antitumor polysaccharides have been ob-
ing an exomethylene ketone moiety are tained from various kinds of preparations.
strong antibacterials and highly cytotoxic me- They include lentinan (CHIHARAet al., 1970),
tabolites. The reduced compounds dihydro- a high-molecular weight p-1,3 glucan isolated
crinipellin B and tetrahydrocrinipellin A (56 from fruiting bodies of Lentinus edodes (SAS-
and 57, Fig. 9) are inactive. Like striatins and AKI and TAKASUKA, 1976), and schizophyl-
striatals crinipellins exert their cytotoxic ac- Ian (KOMATSUet al., 1969), a high-molecular
tivity mainly by interfering with the transport weight p 1 ,3 1,6 glucan obtained from cul-
of essential nutrients and precursors. The cy- tured mycelia of Schizophyllum commune.
totoxic activity on RSV-transformed CEF These compounds inhibit the growth of var-
seems to be higher than on normal CEF ious transplantable tumors in experimental
(KUPKAet al., 1980). animals, they increase the survival rate and
The caryophyllanes nematolin and nemato- are considered to exert their antitumor activi-
lon (58 and 59, Fig. 9) were isolated from cul- ty by the potentiation of the host animals' de-
tures of Naematoloma capnoides, N. sublateri- fense mechanisms rather than by direct inhi-
tium, N. fasciculare, and N. elongatipes bition of tumor cell growth (SUGA et al.,
(BACKENSet al., 1984). Comparison to the 1984). Lentinan in its sulfated form was also
caryophyllanes of higher plants these basidio- used in conjunction with AZT to suppress
mycete metabolites contain more oxygen HIV (DE CLERCQ,1990). Like several other
functions and one or two a,punsaturated car- sulfated polysaccharides, lentinan interferes
bony1 groups. Nematolon and nematolin are with syncytium formation resulting from fu-
weakly antimicrobial, the cytotoxic activity of sion of HIV-infected and uninfected cells.
nematolon is 5-fold higher than that of nema- KS-2, a peptide containing a-linked mannose
tolin. In ECA cells the incorporation of thy- was extracted from the mycelia of Lentinus
midine into DNA is inhibited by 50% at a ne- edodes (FUJI et al., 1978). KS-2 suppressed
matolon concentration of 2 pg mL - I . In vivo the growth of both Ehrlich tumors and Sarco-
no significant antitumor activity (B-16 mela- ma 180 tumors at dose levels of 1 mg kg-'
nocarcinoma, Lewis lung carcinoma, P-388 and 100 mg kg-' when administered intra-
lymphocytic leukemia) was found for nemato- peritoneally or orally. It was also capable of
lon. The LD5,, for tumor bearing mice was de- inducing interferone in mice.
J7 Bioactive Metabolites from Basidiomycetes 509
PSK (krestin), a polysaccharide prepara- methyl-9-/3-~-ribofuranosylpurine (64-66,

tion isolated from Coriolus versicolor pre- Fig. 11) were isolated from mycelial cultures
dominantly consists of glucan and of ca. 25 % of Collybiu maculutu in a screening for inhibi-
tightly bound protein (TSUGAGOSHIet al., tors of vesicular stomatitis virus (VSV) multi-
1984). Oral administration of PSK increased plication in baby hamster kidney (BHK) cells
the survival rate in several animal cancer (LEONHARDTet al., 1987). 6-methylpurine
models, and PSK is now clinically used in Ja- and 6-methyl-9-/3-~-ribofuranosylpurine had
pan for the treatment of postoperative cancer been obtained before by chemical synthesis.
patients. PSK was also reported to exhibit im- All three nucleosides exhibit modest antifun-
munomodulating acitvity by regulating cyto- gal and cytotoxic activity; the effect on VSV
kine production and effector cell functions multiplication in BHK cells is high and com-
(reviewed by KOBAYASHI, 1993). pares very favorably with that of aruA. Be-
sides their antiviral activity 6-methyl-9-&~-ri-
5.4.2 Immunosuppressive bofuranosylpurine and 6-hydroxymethyl-9-P-
D-ribofuranosylpurhe are inhibitors of ade-
Metabolites nosine desaminase and, therefore, interfere
Only few immunosuppressive compounds with the nucleoside metabolism.
from basidiomycetes have been reported. In Other nucleosides that have been reported
the course of a screening for metabolites sup- as secondary metabolites from basidiomy-
pressing the proliferation of mouse lympho- cetes are nebularine, described as an antibac-
cytes stimulated with mitogens, three geranyl- terial antibiotic from Clitocybe nebuluris
phenols, flavidulol A, B, and C (61-63, (LOFGREN,1954), lentinacin, a hypercholes-
Fig. 10) have been isolated from fruiting bod- terolemic compound from Lentinus edodes
ies of Lucturius fluvidulus (FUJIMOTOet al., (CHIBATAet al., 1969), and the insecticidal
1993). The ICso values for flavidulols A, compound clitocine (67, Fig. 11) from Clito-
B, and C were found to be 8.9 pg mL-', cybe inversu (KUBOet al., 1986). A compila-
4.9 pg mL-', and 36.3 pg mL-', respec- tion of nucleoside antibiotics from microbial
tively, in an assay measuring concanavalin sources and their biological activities was
A-induced proliferation of mouse lympho- published by ISONO(1988) and by ISAACet
cytes, and 6.7 p,g mL-', 3.9 pg mL-', and al. (1991).
28.3 pg mL-', respectively, when the cells A screening for inhibitors of avian myelo-
were stimulated with lipopolysaccharide. blastosis virus (AMV) reverse transcriptase
resulted in the isolation of clavicoronic acid
5.5 Antiviral Compounds and (68, Fig. 11) from fermentations of Cluvicor-
onu pyxidutu (ERKELet al., 1992). Clavicor-
Inhibitors of Reverse onic acid is a non-competitive inhibitor of
Transcriptases AMV (Ki: 130 pM) and Moloney murine leu-
kemia (MMuLV) virus (Ki: 68 pM) reverse
The nucleosides 6-methylpurine, 6-methyl- transcriptases. In permeabilized cells and iso-
9-/3-~-ribofuranosylpurine, and 6-hydroxylated nuclei DNA and RNA synthesis are not
Flavidulol A (61) Flavidulol B (62) Flavidulol C (63)
Fig. 10. Immunosuppressive metabolites.

1
6-Melhylpudne (64)
6-Methyl-9-B-D-nboftuamsyIpurine
(65)
w
Hb dH
Clavicoronic acid (68)
Clitocine (67)
O H 0
Fig. 11. Antiviral com-

O H
pounds and inhibitors
of reverse transcript-
Podoscyphic acid (69) ases.
affected. Clavicoronic acid markedly inhibits RNA syntheses in whole cells and isolated
the multiplication of VSV in BHK cells by in- nuclei are not affected by 100 pg mL-' po-
terfering with the RNA-directed RNA poly- doscyphic acid. Comparison of the ethyl ester
merase of the virus. Clavicoronic acid exhibits and the mono-oxo derivative of podoscyphic
no cytotoxic and very weak antimicrobial ac- acid revealed the importance of the free
tivity. y-oxoacrylate moiety for its biological activity
Podoscyphic acid, (E)-4,5-dioxo-2-hexade- (ERKEL et al., 1991).
cenoic acid (69, Fig. ll), isolated from fer- Several drimane sesquiterpenoids from ba-
mentations of Podoscypha petalodes, is a non- sidiomycetes have been reported as inhibitors
competitive inhibitor of AMV and MMuLV of reverse transcriptases. Drimanes had been
reverse transcriptase. The ICso values for the isolated from a number of other natural
inhibition of AMV reverse transcriptase were sources including higher plants, ascomycetes,
100 pg mL-' and for the MMuLV reverse mollusks, and sponges (reviewed by JANSEN
transcriptase 10-20 pg mL-'. DNA and and DE GROOT,1991). The mniopetals A, €3,
Mniopetal B (71)
Mniopetal c (72) Mniopetal D (73)
Mniopetal E (74) Mniopetal F (75) Kuehneromycin A (76)
Fig. 11 Kuehneromycin B (77) Hyphodontal(78)

C, D, E, and F (70-75, Fig. 11)have been iso- Hyphodontal (78, Fig. l l ) , a new isolacta-
lated from fermentations of a Canadian rane sesquiterpenoid, has been isolated from
Mniopetulum species (KUSCHELet al., 1994). fermentations of a Hyphodontiu species (ER-
They most strongly inhibit the MMuLV KEL et al., 1994). Hyphodontal strongly inhib-
reverse transcriptase at concentrations of its the growth of several yeasts and is a non-
1.7-50 pM, with mniopetal B being the most competitive inhibitor of AMV (Ki: 349 pM)
active compound (ICso: 1.7 pM). The Icso and MMuLV (Ki:112 (LM)reverse transcript-
values for the AMV reverse transcriptase are ase. The ICso for the HIV-1 reverse tran-
much higher. Inhibition of HIV-1 reverse scriptase with the natural heteropolymeric
transcriptase by mniopetals depends on the template was determined to be 77 pM
template primer used. With a natural hetero- (20 pg mL - I ) . The cytotoxic activity of hy-
polymeric template inhibition of HIV-1 re- phodontal is mainly due to the interference
verse transcriptase is most pronounced at with DNA and RNA syntheses in whole cells
concentrations of 30-190 pM. In addition, and isolated nuclei.
mniopetals exhibit cytotoxic properties which Up to now, none of the inhibitors described
may be at least partly due to a lyctic action on above has been shown to inhibit the multipli-
the cytoplasma membrane. The kuehneromy- cation of HIV-1 or HIV-2 viruses in cellular
cins A and B (76 and 77, Fig. 11) have been systems.
isolated from cultures of a Tasmanian Kueh-
neromyces species (ERKELet al., 1995; ANKE
et al., 1993). Like mniopetals the kuehnero- 5.6 Inhibitors of Platelet
mycins A and B preferentially inhibit the
MMuLV reverse transcriptase with an ICsOof Aggregation
36 pM, while inhibition of AMV reverse
transcriptase is much less pronounced. The A screening for antithrombotic compounds
activity of HIV-1 reverse transcriptase with using platelet rich plasma from bovine
the natural heteropolymeric template is re- slaughter blood resulted in the isolation
duced to 50 % at a concentration of 64 pM. In of 2-methoxy-5-methyl-l,4-benzochinone
addition, both compounds exhibit cytotoxic (MMBC 79, Fig. 12) from mycelial cultures
and antimicrobial activity. of Lentinus udhaerens (LAUERet al., 1991).
v
Lagopodin B (80)
Fig. U.Inhibitors of plate-

Ompbalone (81) Panudial (82) let aggregation.
MMBC inhibits aggregation of human blood sidiomycetes Halocyphina villosa (KUPKAet

platelets induced by U46619 a prostaglandine al., 1981b) and Helminthosporium siccans
analog and thromboxane mimic with an ICs0 (ISHIBASHIet al., 1968), frustulosinol, and
of 2.5 pg mL-' (16.45 pM). This effect is frustulosin (85 and 86, Fig. 13) isolated from
completely reversible by the addition of cultures of Stereurn frustolosum (NAIR and
1.8-2.4 pM U46619. Hence, it MMBC is pro- ANCHEL,1977). These compounds are hydro-
posed to act as a competitive thromboxane chinone derivatives with an isopentenyne side
A2 receptor antagonist. Similar effects on chain but without chlorine substitutions. Oth-
platelet aggregation were observed with the er acetylenic substances containing an aro-
benzoquinones lagopodin B (80, Fig. 12) ob- matic ring are the antibacterial and antifungal
tained from Coprinus cinereus and omphalon antibiotics peniophorin B and A (87 and 88,
(81, Fig. 12), an antimicrobial and cytotoxic Fig. 13) (GERBERet al., 1980).
metabolite isolated from cultures of Lentinel- The pereniporins A and B (89 and 90, Fig.
lus omphalodes (STARKet al., 1991). 13) were isolated from cultures of Perennipo-
The drimane panudial (82, Fig. 12) from ria medullaepanis. Pereniporin A inhibits the
cultures of a Panus species was detected as an root elongation of lettuce at 100 pg mL-'
inhibitor of platelet aggregation (LORENZEN and is active against gram-positive bacteria.
et al., 1993). Panudial strongly interferes with Both compounds show cytotoxic activity
the ADP-, collagen-, U46619-, ristocetin-, ara- against Friend leukemia cells at 130 pg mL-'
chidonic acid-, and thrombine-induced aggre- and 3.91 pg mL-', respectively (KIDAet al.,
gation of human and bovine platelets. The 1986).
IC,, values of panudial for all inducers except Fomannosin (91, Fig. 13) was isolated from
for thrombine varied between 5 pM and cultures of the wood-rotting basidiomycete
35 pM. In addition, panudial showed inhibi- Fomes annosus (Heterobasidion annosum). F.
tory activity against type I phospholipase A2 annosus is one of the relatively few basidio-
(ICs0=23 pg mL - ') from Naja mosambique. mycetes that cause the death of host cells in
In HL-60 cells the incorporation of leucine living trees and an extensive decay of heart-
and uridine into protein and RNA is marked- wood of contaminated trees. Fomannosin ex-
ly inhibited by 4-8 pg mL-' panudial. hibits phytotoxic activity in assays with Chlo-
rella pyrenoidosa and against Pinus taeda
seedlings when applied to the stem base or a
5.7 Herbicidal Compounds lateral root at a concentration of 88 pg per
seedling (BASSETTet al., 1967). The absolute
A screening for inhibitors of the key en- configuration and the biosynthesis of foman-
zymes of the glyoxylate cycle, which are po- nosin was elucidated by CAINEand NACH-
tential targets for herbicides, resulted in the BAR (1978). Fomannoxin (92, Fig. 13), a dihy-
isolation of mycenon (83, Fig. 13), a novel drobenzofuran which is 100 times more toxic
chlorinated benzoquinone derivative from to Chlorella pyrenoidosa, was isolated from
cultures of a Mycena species (HAUTZELet al., the same fungus by HIROTANIet al. (1977).
1990). Mycenon inhibits isocitrate lyase prep- Several plant growth inhibitors were iso-
arations from plants, bacteria, and fungi. lated from fruiting bodies of Naematoloma
The Ki values were determined to be 5.2 pM, fasciculare. The fasciculols A, B, and C (93-
11 pM,and 7.4 pM for the enzymes from Rhi- 95, Fig. 13) are tetracyclic triterpenes with a
cinus communis, Acinetobacter calcoaceticus, hydroxylated lanostane skeleton. The fascicu-
and Neurospora crassa, respectively. Malate 101s D, E, and F (96-98, Fig. 13) are the corre-
synthase, the second key enzyme of the glyox- sponding esters with a novel depsipeptide
ylate cycle, was not affected. In addition to its group consisting of 3-hydroxy-3-methylglu-
phytotoxic activities mycenon exhibits antimi- taric acid and glycine (IKEDA et al., 1977).
crobial and cytotoxic properties. The fasciculols and their depsipeptides inhib-
Antibiotically active products which are ited the growth of Chinese cabbage seedlings
structurally related to mycenon are siccayne at concentrations of 100-300 pg mL-'. In ad-
(84, Fig. 13) from cultures of the marine ba- dition, fasciculol D exhibited weak antibac-
Mycenon (83) Siccayne (84) Fmtulosinol(85) Fmtulosin (86)
Peniophorin B (87)
Pereniporin A (89) Pereniporin B (90)
d==
0
cH2 --C %H --C -%OH
I
OCH3 II
0
Peniophorin A (88)
9
0
It
Fomannoxin (92)
HO
Fomannosin (91)
Fig. 13. Herbicidal compounds.

Fig. 14). Ibotenic acid was isolated from fruit-

ing bodies of Amanita muscaria, A . strobili-
formis, and A . pantherina (for a review, see
BRESINSKY and BESL,1985). The cyclodepsi-
peptide beauvericin (101, Fig. 14) which was
isolated from the basidiomycete Polyporus
sulphureus (DEOL et al., 1978) and from the
entomopathogenic fungi Beauveria bassiana
and Paecilomyces fumoso-roseus, exhibits in-
on
0
II
cnjo -C -m2
"ii
-N --c -CH~-C
0
I -cnz -c II- secticidal activity against mosquito larvae,
brine shrimp, houseflies, and cockroach car-
x
I diac cells in vitro (ROBERTS,1981). Beauveri-
cin is a ionophore forming complexes with al-
kali metals. Recently, inhibitory activity
on acyl CoA-cholesterol acyltransferase
(ACAT) in isolated rat liver microsomes has
been described for beauvericin with an IC50
value of 3.0 pM (TOMODAet al., 1992). In a
cell assay using 5774 macrophages the forma-
tion of cholesteryl esters was inhibited by
0.17 pM beauvericin.
More than 150 species of nematophagous
fungi belonging to Zygomycetes, Ascomy-
Fig. 13 cetes, Deuteromycetes, and Basidiomycetes
are known to be capable of capturing nema-
todes, and hence the production of nematici-
dal toxins has been proposed (BARRON,1977;
terial activity against Staphylococcus aureus SAYREand WALTER,1991).
and Klebsiella pneumoniae. 5-Pentyl-2-furaldehyde, 5-(4-pentenyl)-2-
Several new related fasciculol esters, the furaldehyde, and methyl-3-p-anisoloxypro-
fasciculic acids A, B, and C, were isolated pionate (102-104, Fig. 14), were isolated from
from fruiting bodies of Naematoloma fascicu- cultures of Zrpex lacteus (HAYASHIet al.,
lure as inhibitors of a calmodulin-dependent 1981). All three compounds caused 50%
phosphodiesterase (PDE) (TAKAHASHI et al., mortality of the nematode Aphelencoides bes-
1989). The IC50values for the PDE from bo- seyi at 25-50 pg mL-'.
vine heart were 6 pM for fasciculic acid B and 2-Pecenedioic acid (105, Fig. 14), a fatty
10 pM for fasciculic acid A. No data on phy- acid toxic to Panagrellus redivius, was iso-
totoxic activity have been published. lated from a Pleurotus ostreatus strain. The
Related tetracyclic triterpenoids with a la- compound immobilized 95% of the test ne-
nostane skeleton have also been isolated from matode at a concentration of 300 pg mL-'
Gandoderma lucidum, G . applanatum, Pioli- within 1 h (KWOKet al., 1992).
thus tinctorius, P. arrhizus, and Fomes fastuo- A screening using the saprophytic nema-
sus (reviewed by CONNOLLY et al., 1994). tode Caenorhabditis elegans as a test organ-
ism resulted in the isolation of S-coriolic acid,
linoleic acid, p-anisaldehyde, p-anisyl alcohol,
5.8 Insecticidal and Nematicidal l-(4-methoxyphenyl)-1,2-propanediol,and 2-
hydroxy-(4-methoxy)-propiophenone (106-
Metabolites 111, Fig. 14) from cultures of Pleurotus pul-
monaris (STADLERet al., 1993b). The most
Insecticidal activity against houseflies was active metabolites were S-coriolic acid and li-
described for ibotenic acid (99, Fig. 14) and noleic acid with LD50 values between 5 and
its decarboxylation product muscimol (100, 10 pg mL -'. Interestingly, the nematicidal
+ CHa 0. 6%
Ibotenic acid (99)
+ I
P-
Beawericin (101)
I
0 0
5-Pentyl-2-furaldehyde (102) 5-(4-PentenyI)-2-1imldehyde(103)
Methyl-3-pankloxypropionate(104)
rrum-2-Decenoic acid (105)
S-Coriolic acid (106)
Linoleic acid (107)
Fig. 14. Insecticidal and nematicidal metabolites.

p-Anisaldeyde (108) p-Anisyl alcohol (109)
I -(CMethoxyphenyl)-l,2-propanediol(llO) 2-Hydroxy-(4'-methoxy)-propriophenone (11 1)
Cheimonophyllon A (1 12) CheimonophyllonB (113)
Cheiinophyllon C (1 14) CheimonophyllonD (1 15)
Fig. 14 cheimonophyllon E (116) Cheimonophyllal(l17)

activity of different fatty acids depends on the 5.10 Inhibitors of

chain length and the number of double bonds
in the molecule. Furthermore, different spe-
Aminopeptidases
cies of nematodes varied in their sensitivity to
various fatty acids. Among the enzymes bound to the outer
Six different bisabolane sesquiterpenes, the surface of mammalian cells aminopeptidases
cheimonophyllons A-E (112-116, Fig. 14), have been reported to be potential targets for
and cheimonophyllal (117, Fig. 14) from cul- immunomodulating drugs. A prominent ex-
tures of Cheimonophyllum candidissimum ample is bestatin (ubenimex), a strong inhibi-
have recently been described (STADLER et tor of aminopeptidase B and leucine amino-
al., 1994). Cheimonophyllons A, B, D, and peptidase, which was isolated from cultures of
cheimonophyllal exhibit nematicidal, cyto- Streptomyces olivoreticuli (UMEZAWAet al.,
toxic, and antimicrobial activity, whereas 1976). Clinical studies of this drug were pub-
cheimonophyllons C and E show weaker ef- lished by MATHB(1987).
fects. The LDso for Caenorhabditis eleguns Tyromycin A (121, Fig. 16) from fermenta-
was 10 Fg mL-' for cheimonophyllon A and tions of Tyromyces lucteus is the first known
D and 25 pg mL-' for cheimonophyllon B naturally occurring citraconic anhydride deri-
and cheimonophyllal. vative with two 3-methyl maleic anhydride
units in its molecule. Tyromycin A strongly
inhibits both leucine aminopeptidase and cys-
5.9 Inhibitors of Cholesterol teine aminopeptidase bound to the outer sur-
Biosynthesis face of HeLa S3 cells. The Ki values were de-
termined to be 4.10-5 M for leucine amino-
Drugs interfering with the biosynthesis of peptidase and 1.3.10-5 M for cysteine amino-
cholesterol are of potential value in the treat- peptidase. Tyromycin A also inhibits cyto-
ment of hypercholesterolemia which is one of solic and microsomal leucine aminopeptidase
the primary causes of arteriosclerosis and cor- of porcine kidney and carboxypeptidase
onary heart disease. Mevinolin (monacolin of bovine kidney at concentrations of
K), a specific inhibitor of eukaryotic 3-hy- 25-60 Fg mL-'. The inhibitory activity of
droxy-3-methylglutaryl coenzyme A (HMG tyromycin A is due to the two maleic acid
CoA) reductase isolated from cultures of anhydride moieties. Tyromycin amide (122,
Monascus ruber (ENDO,1979) has been intro- Fig. 16) is devoid of any inhibitory activity on
duced into clinical practice. the cell-bound aminopeptidase of HeLa cells
Search for inhibitors of cholesterol biosyn- (WEBERet al., 1992).
thesis in HeLa S3 cells resulted in the isola-
tion of dihydroxerulin, xerulin, and xerulinic
acid (118-120, Fig. 15), from surface cultures
of Xerulu melanotricha (KUHNT et al., 1990). 5.11 Inhibitors of Phospholipases
Dihydroxerulin strongly interferes with the C and A2
incorporation of 14C acetate into cholesterol,
while the incorporation of 14C mevalonate is Caloporoside (123, Fig. 17), an inhibitor of
hardly affected. It was shown that the inhibi- phospholipase C, has been isolated from fer-
tory effect on cholesterol biosynthesis is due mentations of Caloporus dichrous (WEBERet
to a strong inhibition of HMG CoA synthase al., 1994). Caloporoside is a new glycosylated
at concentrations starting from 0.1 pg mL-' salicylic acid derivative which exhibits weak
of dihydroxerulin. Similar results have been antibacterial and antifungal activity. Calopor-
obtained with xerulin. oside shows a strong, selective inhibitory ac-
tivity on phospholipase C from pig brain with
a Ki value of 12.3 pM. Phospholipases C from
Clostridium welchii and Bacillus cereus, which
act on other substrates are inhibited to a
much lesser extent. Phospholipases A2 and D,
Dhydroxedi (118) Xedin (119) Xedinic acid (120)
Fig. 15. Inhibitors of cholesterol biosynthesis.
triglyceride lipases, and acetylcholin esterase spect to the hydroxy-butenolide ring it resem-
are not affected. bles to the manoalides and luffarielloides
Several closely related derivatives of the which were isolated from Luffuriella variabilis
aglycone part of caloporoside were isolated and related sponges ( P o n s and FAULKNER,
from fruiting bodies of Merulius tremellosus 1992). 5-Hydroxy-3-vinyl-2(5H)-furanone
and Phlebia rudiafu. The merulinic acids A, B, specifically inhibits the human synovial phos-
and C (124-126, Fig. 17) exhibit antimicrobial pholipase AZwith ICs0 values of 100 nM. The
and hemolytic activity which may be due to compound also inhibits the aggregation of hu-
their lytic action on the cytoplasma mem- man and bovine platelets stimulated with dif-
brane (GIANNETTIet al., 1978). Merulinic ferent inducers.
acids closely resemble the skin irritants of
Anacardiaceae and of Ginkgo bilobu, anacar-
dic acid 111, pelandjauic acid, and ginkgolic 5.12 Inhibitors of
acid. Interestingly, mycelial cultures of M . tre-
rnellosus do not produce merulinic acids but (Na +-K )-ATPases
+
the antibiotic sesquiterpenoid merulidial.

5-Hydroxy-3-vinyl-2(5H)-furanone (127, The tricyclic sesquiterpenoids coriolin and
Fig. 17) has been isolated from cultures of Ca- coriolin B (128 and 129, Fig. 18) were isolated
lypfella sp. (LORENZEN et al., 1995). With re- from cultures of Coriolus consors (TAKEUCHI
et al., 1969). While coriolin B shows neither
antitumor nor antimicrobial activity, its oxi-
dation product diketocoriolin B (130, Fig. 18)
does. Studies on the mode of action of diketo-
coriolin B revealed that the antitumor activity
is due to the inhibition of (Na+-K +)-ATPase
localized in the cell membrane of tumor cells
Tyromycin A (121) X = 0 which causes a cessation of growth (KUNIMO-
TO et al., 1973). By chemical modification of
Tyromycin amide (122) X = NH
coriolin NISHIMURA et al. (1977) showed that
Fig. 16. Inhibitors of aminopeptidases. the keto group at C-5 and the two epoxy
520 I 1 Products from Basidiomycetes
OH
Caloporoside (123)
i 7
--C =C -(C!H2)6
R2
I
--C
H
I
R1 R1 R2
Merulioic acid A (124) OH H
MedicacidB(125) H OH
MedicacidC(126) H H
f P
S-Hydroxy-3-vinyl-2(5~-tiua~ne
(127) Fig. 17. Inhibitors of phospholipases C and AZ
groups greatly contribute to the antitumor (VSV) in baby hamster kidney cells (BHK-
and antibacterial activity. Diketocoriolin B 21) resulted in the isolation of collybial (131,
augments antibody formation against sheep Fig. 19) from fermentations of Collybiu con-
red blood cells (SRBC) in vivo at a concen- fluens (SIMONet al., 1995). The propagation
tration of 0.1 Fg per mouse or in in vifro using of VSV in BHK-21 cells was reduced by a fac-
spleen cell cultures at 0.01 ng per culture tor of lo3 at 21.5 pM of collybial with cyto-
(ISHIZUKAet al., 1981). toxic effects at 5-fold higher concentrations.
Incorporation of labeled precursors into
DNA, RNA, and proteins revealed that the
5.13 Addendum antiviral effects of collybial are probably due
to an interference with the macromolecular
This section describes compounds isolated syntheses of the host. In addition, antibacter-
from 1995 until November 1996. They are ial activities against gram-positive bacteria
presented under the appropriate section num- were observed.
ber of the main text. Section 5.8 In addition to the recently de-
Section 5.5: A screening for inhibitors of scribed bisabolanes cheimonophyllons A-E
multiplication of vesicular stomatitis virus (112-116, Fig. 14) and cheimonophyllal (117,
5 Bioactive Metabolitesfrom Basidiomycetes 521
OH
Coriolin (128)
OH
Coriolim B (129)
Fig. 18. Inhibitors of (Na+-K+)-ATPases. Diketomriolin B (130)
Fig. 14), the p-menthane 1,Zdihydroxyminth- 5.13.1 Inhibitors of Leukotriene

lactone (132, Fig. 19) was isolated as a minor
nematicidal component from fermentations Biosynthesis
of Cheimonophyllum candidissimum (STAD-
LER et al., 1995). The LD50 against the nema- Leukotrienes are potent biological media-
tode Caenorrhabditis elegans was determined tors derived from arachidonic acid metabol-
to 25 pg mL-’ without any additional antimi- ism and are generated via the 5-lipoxygenase
crobial and cytotoxic activity. The structurally pathway. Leukotriene B4, a dihydroxy deriva-
related minthlactone has been previously re- tive, causes adhesion and chemotactic move-
ported as a constituent of peppermint (Men- ment of leukocytes, enzyme release, and gen-
tha piperita) oil (TAKAHASHI et al., 1980). eration of superoxide in neutrophiles. The
Omphalotin (133, Fig. 19), a new cyclic sulfopeptide leukotrienes C4,D4, and E4, are
dodecapeptide possessing strong and selective known as “slow reacting substances of ana-
nematicidal activity against the plant patho- phylaxis” and induce bronchoconstriction,
genic nematode Meloidogyne incognita was stimulate mucus production, and increase vas-
isolated from mycelial cultures of the basidio- cular permeability (SAMUELSSONet al.,
mycete Omphalotus olearius (MAYERet al., 1987). Due to these effects the leukotrienes
in press; STERNERet al., in press). The LDgO have been implicated as important mediators
against M. incognita was determined to be of inflammation and hypersensitivity reac-
0.76 pg mL-’, whereas the saprophytic ne- tions (FORD-HUTCHINSON et al., 1994; SAL-
matode C. elegans was approximately 50 MON and GARLAND, 1991).
times less sensitive. Omphalotin exhibits no A screening for inhibitors of leukotriene
phytotoxic, antibacterial, or antifungal activi- C4 biosynthesis resulted in the isolation
ties and is only weakly cytotoxic at high con- of ( +)-10a-hydroxy-4-muurolen-3-one (134,
centrations (100 pg mL-’). Fig. 20) from fermentations of an Ethiopian
Favolaschia species (ZAPFet al., 1996). The
IC50 value for the inhibition of leukotriene C4
biosynthesis in rat basophilic leukemia (RBL-
OH
Collybial(l3 1) 1,2-Dihydroxyminthlactone(1 32)
Omphalotin (133)
Fig. 19. Antiviral and nematicidal compounds.
1) cells was determined to be 5-10 kg mL-'. ro-3-(4-methoxypheny1)-2-propen-l-o1 (138,

Related cadinane sesquiterpenes, like ( + )-T- Fig. 20) and lentinellone (139, Fig. 20), a pro-
cadinol and ( -)-3-oxo-T-cadinol, have been toilludane derivative. Blennin A (VIDARI et
reported before (CLAERSON et al., 1991) as al., 1976), blennin C (DE BERNARDI et al.,
constituents of scented myrrh (the resin of the 1976), and deoxylactarorufin A (DANIEWSKI
plant Commiphoru guidotti), and similar ef- et al., 1977; DANIEWSKI and KROL, 1981) are
fects on leukotriene C4 biosynthesis have strong inhibitors of leukotriene C, biosynthe-
been observed (ZAPFet al., 1996). sis in RBL-1 cells with ICso values of
Three known lacterane type sesquiterpe- 5 pgmL-' for blennin A, 4 pgmL-' for
noids, blennin A (135, Fig. 20), blennin C blennin C, and 2 Fg mL-' for deoxylactaro-
(136, Fig. 20), and deoxylactarorufin A (137, rufin A. (Z)-2-chloro-3-(4-methoxyphenyl)-2-
Fig. 20) were obtained from fermentations of propen-1-01 inhibited the leukotriene C4 bio-
Lentinellus cochleutus (WUNDERet al., 1996) synthesis with an IC5"of 15 kg mL -',where-
together with the new metabolites (2)-2-chlo- as lentinellone was inactive.
5 Bioactive Metabolitesfrom Basidiomycetes 523
5.13.2 Inducers of Differentiation possible targets for a rational antitumor ther-

apy (HASS, 1992; LEVITZKI,1994; THOMP-
of Promyelocytic Leukemia Cells SON, 1995; MANNING, 1996).
and Inhibitors of Signal Pinicoloform (140, Fig. 21) an antibiotic
Transduction in Tumor Cells from Resiniciurn pinicolu was detected in a
screening for metabolites inducing the dif-
ferentiation of HL-60 cells to monocytes
Among the phenotypic abnormalities in and macrophages (BECKER et al., 1994)
acute leukemia is a lack of granulocytes, mac- at concentrations between 0.5-1 yg mL -'
rophages, and platelets caused by the inability (1.8-3.5 yM). Cytotoxic as well as antimicro-
of the neoplastic leucocytes to undergo termi- bial activities have also been described at
nal differentiation and eventually apoptosis. concentrations between 10-66 yM.
The human HL-60 leukemia cell line is an ex- The diterpenes lepistal (141, Fig. 21) and
cellent model for a study of functional and lepistol (142, Fig. 21) were isolated from cul-
morphological differentiation in vitro, be- tures of Lepistu sordidu (MAZURet al., 1996).
cause the cells can be induced to differentiate -'
At a concentration of 0.2 yg mL lepistal in-
into granulocytes or monocytes/macrophages. duces the differentiation of 20 % of the HL-60
Differentiation may be followed by apoptosis, cells into granulocyte/monocyte-likecells and
a process of active DNA fragmentation. The of 18% of the human histiocytic lymphoma
induction of differentiation and apoptosis are (U-937) cells into monocyte-like cells. The re-
regulated by a network of signal transduction lated alcohol lepistol is 50-100 times less ac-
pathways and transcription factors which are tive. Cytotoxic activities of lepistal and lepis-
(+)-l~-hydroxy-4-muurolen-3-one
(1 34) Blennin A (135)
Blennin C (136) DeoxylactarorufinA (1 37)
Fig. 20. Inhibitors of leuko-

triene biosynthesis. (Z)-2-chloro-3-(4-mehoxyphenyl)-2-prop~-~-ol(138) Lmtindlone (1 39)
to1 were observed at 1 pg mL-' and ferentiation of HL-60 cells by nidulal is fol-
50 pg mL-', respectively. Lepistal exhibits lowed by apoptosis. In COS-7 cells (African
pronounced antimicrobial activity whereas le- green monkey) nidulal selectively activates
pistol shows no antibacterial and only weak the AP-1 dependent signal transduction path-
antifungal activities. Therefore, the aldehyde ways in a manner similar to the phorbol ester
function is considered to substantially contri- TPA, an activator of protein kinase C. In gel
bute to the biological activity of lepistal. shift assays with extracts of nidulal-treated
Recently, two new bisabolane sesquiter- HL-60 cells a change of binding activities of
penes, nidulal (143, Fig. 21) and niduloic acid the AP-1 transcription factor is observed,
(144, Fig. 21), have been isolated from fer- which may be the result of an altered com-
mentations of Nidulu cundidu (ERKELet al., position of the AP-1 protein complex.
in press). Nidulal and niduloic acid induce The novel norilludane puraquinonic acid
differentiation of 15-25% of HL-60 cells at (145, Fig. 21) was isolated from mycelial cul-
concentrations of 72 p M and 36 pM, respec- tures of Mycenu puru (BECKERet al., in
tively. It has been shown that induction of dif- press) as an inducer of morphological and
pinicolofom (140) Lepistal(141) R==O

Lepistol(142) R= -OH
Nidulal(143) Niduloic acid (144)
Fig. 21. Inducers of differen-

OH tiation and inhibitors of sig-
nal transduction in tumor
F'uraquinonic acid (145) Panepoxydone (146) cells.
6 Future Perspectives 525
physiological differentiation of mammalian

cells. It induces differentiation of 3 0 4 0 % of
6 Future Perspectives
HL-60 cells into granulocyte- or monocyte/
macrophage-like cells at 380 p,M. At the same It is evident that basidiomycetes provide a
concentration U-937 cells, which are blocked rich, yet quite untapped source of compounds
at a later stage of development are affected to with novel structures and in some cases very
a much lesser extent. interesting biological activity. Many of the
NF-KB is an inducible, ubiquitous tran- compounds isolated so far seem to be pro-
scription factor, which regulates the expres- duced exclusively by this class of fungi. From
sion of various cellular genes involved in im- the number of estimated basidiomycetes spe-
mune response, inflammation, acute phase re- cies (>30OOO) it may be assumed that there
sponse, and several viral genes and inhibitors will be many more exciting discoveries made
of NF-KB activation may, therefore, find in the future.
broad application as novel therapeutics
(BAEUERLEand HENKEL,1994; BALDWIN,
1996; MANNING and ANDERSON, 1994).
In a search for new inhibitors of NF-KB-
mediated signal transduction in COS-7 cells
7 References
using the secreted alkaline phosphatase
(SEAP) as reporter gene, panepoxydone AMES,B. N., MCCANN,J., YAMASAKI, E. (1975),
Methods for detecting carcinogens and muta-
(146, Fig. 21) was isolated from fermentations gens with the Salmonellalmammalian mutagenic-
of the basidiomycete Lentinus crinitus (ER- ity test, Mut. Res. 31, 347-364.
KEL et al., 1996). Panepoxydone and several ANCHEL,M., HERVEY,A., ROBBINS, W. J. (1950),
related derivatives have been previously re- Antibiotic substances from basidiomycetes. VII.
ported as secondary metabolites from Punus Clitocybe illudens, Proc. Natl. Acad. Sci. USA 36,
rudis, P. conchatus (KIS et al., 1970), and 300-305.
Penicilliurn urricue (SEGIGUCHIand GAU- ANKE,T., DIEKMANN, H. (1972), Metabolic prod-
CHER,1979).Panepoxydone inhibits the NF-KB ucts of microorganisms. 112. Biosynthesis of si-
activated expression of the SEAP with an deramines in fungi. Rhodotorulic acid synthet-
ase from extracts of Rhodotorula glutinis, FEBS
ICso of 1.5-2 p,g mL-' (7.15-9.52 p,M). No in- Lett. 27, 259-262.
hibition of AP-1-mediated expression of the ANKE,T., STEGLICH,W. (1981). Screening of ba-
reporter gene could be observed at a concen- sidiomycetes for the production of new antibio-
tration up to 5 p,g mL-' panepoxydone. Pan- tics, in: Advances in Biotechnology, Vol. 1
epoxydone strongly reduces the TPA-, TNF-a-, (MOO-YOUNG,M., Ed.), pp 35-40. Pergamon
and ocadaic acid-mediated binding of NF-KB Press, Canada.
to the high affinity consensus sequence in ANKE,T., STEGLICH,W. (1989), P-Methoxyacryl-
COS-7 and HeLa S3 cells as confirmed by ate antibiotics: from biological activity to syn-
electrophoretic mobility shift assays. Pan- thetic analogues, in: Biologically Active Mole-
cules - Identification, Characterization, and Syn-
epoxydone inhibits the phosphorylation of the thesis (SCHLUNEGGER, U. P., Ed.), pp. 9-25.
inhibitory protein IKB and, therefore, seques- Berlin-Heidelberg: Springer-Verlag.
ters the NF-KB complex in an inactive form. ANKE,H., ANKE,T., DIEKMANN H. (1973), Bio-
Recently, the related cycloepoxydon, an inhi- synthesis of sideramines in fungi. Fusigen syn-
bitor of AP-1- and NF-KB-mediated gene ex- thetase from extracts of Fusarium cubense,
pression, has been isolated from fermenta- FEBS Lett. 36,323-325.
tions of a deuteromycete strain (GEHRTet al., ANKE,T., OBERWINKLER, F., STEGLICH,W., HOF-
in press). LE, G. (1977a), The striatins - new antibiotics
from the basidiomycete Cyathus striatus (Huds.
ex Pers.) Willd., J. Antibiot. 30,221-225.
ANKE, T., OBERWINKLER, F., STEGLICH,W.,
SCHRAMM, G. (1977b), The strobilurins - new
antifungal antibiotics from the basidiomycete
Strobilurus tenacellus (Pers. ex. Fr.) Sing., J. An-
tibiot. 30,806-810.
ANKE,T., HECHT,H. J., SCHRAMM, G., STEGLICH, (Relhan ex Fr.) Sing., J. Antibiot. 43, 1010-
W. (1979), Antibiotics from basidiomycetes. IX. 1011.
Oudemansin, an antifungal antibiotic from Ou- ANKE,T., ERKEL,G., KOKSCH,G., KUSCHEL,A.,
demansiella mucida (Schrader ex Fr.) Hoehnel GIMENEZ, A., VELTEN, R., STEGLICH,W.
(Agaricales), J. Antibiot. 32, 1112-1117. (1993), Inhibitoren der HIV-1 Reversen Tran-
ANKE,T., KUPKA,J., SCHRAMM, G., STEGLICH, scriptase und Induktoren der Differenzierung
W. (1980), Antibiotics from basidiomycetes. X. menschlicher Zellinien aus Pilzen, in: Wege zu
Scorodonin, a new antibacterial and antifungal neuen Produkten und Verfahren in der Biotech-
metabolite from Marasmius scorodonius (Fr.) nologie (ANKE,T., ONKEN,U., Eds.). DECHE-
Fr., J. Antibiot. 33, 4 6 3 4 7 . MA Monographien 129, 15-25.
ANKE,T., GIANETTI,B.M., STEGLICHW. (1982), ARIGONI,D. (1962), La struttura di un terpene di
Antibiotika aus Basidiomyceten. XV. 1-Hy- nuovo genere, Gazz. Chim. Itl. 22, 884-901.
droxy-2-nonin-4-on,ein antifungischer und cyto- AYER,W. A., BROWNE,L. M. (1981), Terpenoid
toxischer Metabolit aus Ischnoderma benzoinum metabolites of mushrooms and related basidio-
(Wahl.) Karst., Z. Naturforsch. 37c, 14. mycetes, Tetrahedron Lett. 17, 2199-2248.
ANKE,T., BESL, H., MOCEK,U., STEGLICH,W. AYER, W. A., LEE, S. P., NAKASHIMA,T. T.
(1983), Antibiotics from basidiomycetes. XVIII. (1979), Metabolites of bird’s nest fungi. Part 12.
Strobilurin C and oudemansin B, two new anti- Studies on the biosynthesis of cyathins, Can. J.
fungal metabolites from Xerula species (Agari- Chem. 17,3338-3343.
cales), J. Antibiot. 36,661-666. BACKENS,S., STEFFAN,B., STEGLICH,W., ZECH-
ANKE,T., SCHRAMM, G. SCHWALGE, B., STEFFAN, LIN,L., ANKE,T. (1984), Antibiotika aus Basid-
B., STEGLICH,W. (1984), Antibiotika aus Basi- iomyceten. XIX. Naematolin und Naematolon,
diomyceten. XX. Synthese von Strobilurin A zwei Caryophyllan-Derivate aus Kulturen von
und Revision der Stereochemie der natilrlichen Hypholoma-Arten (Agaricales), Liebigs Ann.
Strobilurine, Liebigs Ann. Chem. 1984 1616- Chem. 1984, 1332-1342.
1625. BACKENS,S., STEGLICH,W., BAUERLE,J., ANKE,
ANKE,T., BACKENS,S., STEGLICH, W. (1985a), A T. (1988), Antibiotika aus Basidimyceten, 28.
new antibiotic from Paneolus and Psathyrella Hydroxystrobilurin D, ein antifungisches Anti-
species. Abstract Annu. Meeting ASM. Washing- biotikum aus Kulturen von Mycena sanguinolen-
ton, D.C.: ASM Press. ta (Agaricales), Liebigs Ann. Chem. 1988, 405-
ANKE,T., HEIM,J., KNOCH,F., MOCEK,U., STEF- 409.
FAN,B., STEGLICH, W. (1985b), Crinipelline, die BALDWIN,A. S. (1996), The NF-KB and IKB pro-
ersten Naturstoffe mit einem Terquinan-Gerust, teins: new discoveries and insights, Annu. Rev.
Angew. Chem. 27, 714-716. Immunol. 14,649-681.
ANKE,T., STEGLICH,W., POMMER,E. H. (1986), BARRON,G. L. (1977), The nematode-destroying
The antifungal activities of striatals and striatins. fungi, in: Topics in Mycobiology, Vol. 1. Guelph,
Abstracts 6th Int. Congr. Pesticide Chem. Ontario: Canadian Biological Publications.
(IUPAC), Ottawa. BASSETT,C., SHERWOOD,R. T., KEPLER,J. A.,
ANKE,T., SCHRAMM, G.,STEGLICH,W., VON JA- HARMILTON, P. B. (1967), Production and bio-
COW,G. (1988), Structure-activity relationships logical activity of famannosin, a toxic sesquiter-
of natural and synthetic E-pmethoxyacrylates pene metabolite of Fomes annosum, Phytopa-
of the strobilurin and oudemansin series, in: The thology 57, 1046-1052.
Roots of Modern Biochemistry (KLEINKAUF, H., BAUERLE,J. (1981), Antibiotika aus Basidiomyce-
VON DOHREN,H., JANICKE,L., Eds.), pp. 657- ten der Gattungen Clitopilus, Hohenbuehelia,
662. Berlin: de Gruyter. Hemimycena, und Mycena (Agaricales). Thesis,
ANKE, H., HILLEN-MASKE,E., STEGLICH,W. Department of Biology, University of Tubingen,
(1988), 9-phydroxymarasmic acid and other ses- FRG.
quiterpenoids from submerged cultures of a ba- BAUERLE,J., ANKE,T. (1980), Antibiotics from
sidiomycete, Z. Naturforsch. 44c,1-6. the genus Mycena and Hydropus scabripes,
ANKE, H., STERNER,O., STEGLICH,W. (1989), Planta Med. 39, 195-196.
Structure-activity relationship for unsaturated BAUERLE,P. A., HENKEL, T. (1994), Function and
dialdehydes. 3. Mutagenic, antimicrobial, cyto- activation of NF-KB in the immune system,
toxic and phytotoxic activities of merulidial deri- Annu. Rev. Immunol. 12,141-179.
vatives, J. Antibiot. 42, 738-744. BAUERLE,J., ANKE,T., JENTE, R., BOSOLD,F.
ANKE,T., WERLE,A., BROSS,M., STEGLICH,W. (1982), Antibiotics from basidiomycetes. XVI.
(1990), Antibiotics from basidiomycetes. Antimicrobial and cytotoxic polyines from M y -
XXXIII. Oudemansin X, a new antifungal E-P- cena viridimarginata, Arch. Microbiol. 132, 194-
methoxyacrylate from Oudemansiella radicata 196.
7 References 527
BAUERLE,J., ANKE,T., HILLEN-MASKE, E., STEG- substance in Lentinus edodes, Experientia 25,
LICH, W. (1986), Hemimycin, a new antibiotic 1237-1238.
from two Hemimycena species (Basidiomycetes), CHIHARA, G., HAMURO, J., MADEA,Y. Y., ARAI,
Planta Med. 5, 418. Y., FUKUOKA, F. (1970), Fractionation and pu-
BEALE,M. H. (1990), The biosynthesis of C5-Cz0 rification of the polysaccharides with marked an-
terpenoid compounds, Nat. Prod. Rep. 7, 25-39, titumor activity, especially lentinan from Lentin-
387-407. us edodes (Berk.) Sing. (an edible mushroom),
BECKER,U., ANKE,T., STERNER,0. (1994), A Cancer Res. 30,2776-2781.
novel halogenated compound possessing anti- CLAERSON, P., ANDERSON,R, SAMUELSSON, G.
biotic and cytotoxic activities isolated from the (1991), T-cadinol: a pharmacologically active
fungus Resinicium pinicola (J. Erikss.) Erikss. & constituent of scented myrrh: introductory phar-
Hjortst., 2. Naturforsch. 49c, 772-774. macological characterization and high field 'H
BECKER,U., ERKEL,G., ANKE,T., STERNER,0. and I3C NMR data, Planta Med. 57, 352-356.
(in press), Puraquinonic acid, a novel inducer of CLOUGH,J. M. (1993), The strobilurins, oudeman-
differentiation of human HL-60 promyelocytic sins, and myxothiazols, fungicidal derivatives of
leukemia cells from Mycena pura (Pers. ex Fr.), Pmethoxyacrylic acid, Nat. Prod. Rep. 10, 565-
Nat. Prod. Lett. 574.
BIRCH,A. J., CAMERON, D. W., HOLZAPFEL,C. CONNOLLY, J. D., HILL, R. A., NGADJUI,B. T.
W., RICKARDS, R. W. (1963). The diterpenoid (1994), Triterpenoids, Nat. Prod. Rep. 11, 91-
nature of pleuromutilin, Chem Ind. (London) 117.
374-375. DANIEWSKI, W. M., KROL, J. (1981), Constituents
BIRCH,A. J., HOLZAPFEL,C. W., RICKARDS, R. of higher fungi. Part XII, Pol. J. Chem. (Rocz.
W. (1966), The structure and some aspects of the Chem.) 55, 1247-1252.
biosynthesis of pleuromutilin, Tetrahedron DANIEWSKI, W. M., KOCOR,M., KROL,J. (1977),
(Suppl.) 8 Part 11, 359-387. Constituents of higher fungi, Pol. J. Chem.
BRANDT,U., SCHLAGGER, H., VON JAGOW,G. (Rocz. Chem.) 51, 1395-1398.
DAVIES,J., VON AHSEN,U., WANK,H., SCHROE-
(1988), Characterisation of binding of the meth-
DER, R. (1992), Evolution of secondary metabol-
oxyacrylate inhibitors to mitochondria1 cyto-
chrome c reductase, Eur. J. Biochem. 173, 499- ite production: potential role for antibiotics as
506. prebiotic effectors of catalytic RNA reactions,
in: Secondary Metabolites: Their Function and
BRANDT,U., HAASE,U., SCHLAGGER, H., VON Evolution (CADWICK, D. J., WHELAN,J., Eds.),
JAGOW,G. (1993), Speziesspezifitat und Wirk- pp. 2444. Ciba Foundation Symposium 171.
mechansimus der Strobilurine, in: Wege zu DE BERNARD],M., FRONZA,G., VIDARI,G.,
neuen Produkten und Verfahren in der Biotech- VITA-FINZI, P. (1976), Fungal metabolites 11:
nologie (ANKE,T., ONKEN,U., Eds). DECHE- new sesquiterpenes from Lactarius scobiculatus
MA Monographien 129,27-38. Scop. (Russulaceae), Chim. Ind. 58, 177-178.
BRAUER, H., KORN, A. (1993), Untersuchung der DE CLERCQ,E. (1990), Perspectives for chemo-
Umsatzleistung im Hubstrahl-Bioreaktor bei der therapy of the HIV infection: An introduction,
biotechnologischen Produktion mit Pilzen, in: Pharmacochem. Libr. 14 (Des. Anti-AIDS
Wege zu neuen Produkten und Verfahren in der Drugs), 1-24.
Biotechnologie (ANKE,T., ONKEN,U., Eds.). DEML,G., ANKE,T., OBERWINCKLER, F., GIAN-
DECHEMA Monographien 129, 147-157. NETTI, B. M., STEGLICH, W. (1980), Schizonellin
BRESINSKY, A., BESL, H. (1985), Giftpilze. Stutt- A and B, new glycolipids from Schizonella mela-
gart: Wissenschaftliche Verlagsgesellschaft. nogramma (Ustilaginales), Phytochemistry 19,
CANE,D. E., NACHBAR, R. B. (1978), Stereochem- 83-87.
ical studies of isoprenoid biosynthesis. Biosyn- DEOL,B. S., RIDLEYD. D., SINGH,P. (1978), Iso-
thesis of fomannosin from [1,2-"CC2]acetate,J. lation of cyclodepsipetides from plant pathogen-
Am. Chem. SOC. 100,3208-3212. ic fungi, Aust. J. Chem. 31, 1397-1399.
CHENINA, R., DOCHNAHL, A., HUFF,T., KUBALL, DREWS, J., GEORGOPOULOS, A., LABER, G.,
H. G., LANGE,B., ANKE,H., ANKE,T. (1993), SCHUTZE,E., UNGER,J. (1975), Antimicrobial
Fluoreszenzspektrometer zur On-line-Verfol- activities of 81.723 hfu, a new pleuromutilin der-
gung von Fermentationsprozessen, in: Wege zu ivative, Antimicrob. Agents Chemother. 7, 507-
neuen Produkten und Verfahren in der Biotech- 516.
nologie (ANKE,T., ONKEN,U., Eds.). DECHE- DUGAN,J. J., DEMAYO,P., NISBET, M., ROBBIN-
MA Monographien 129, 171-182. SON, J. R., ANCHEL, M. (1966), Terpenoids XIV.
CHIBATA,I., OKUMURA, K., TAKEYAMA, S., Ko- The constitution and biogenesis of marasmic
TERA, K. (1969), A new hypercholesterolemic acid, J. Am. Chem SOC. 88,2838-2844.
528 I I Products from Basidiomycetes
EGGER,H., REINSHAGEN, H. (1976a), New pleuro- antitumor polysaccharide, KS-2, extracted from
mutilin derivatives with enhanced antimicrobial culture mycelia of Lentinus edodes, J. Antibiot.
activity. I. Synthesis, J. Antibiot. 29, 915-922. 31,1079-1090.
EGGER,H., REINSHAGEN, H. (1976b), New pleuro- FUJIMOTO,H., NAKAYAMA, Y., YAMAZAKI, M.
mutilin derivatives with enhanced antimicrobial (1993), Identification of immunosuppressive
activity. 11. Structure-activity correlations, J. An- components of a mushroom, Lactarius flavidu-
tibiot. 29, 923-927. lus, Chem. Pharm. Bull. 41,654-658.
ENDO,A. (1979), Monacolin K, a new hypercholes- GEHRT, A., ERKEL, G., ANKE, H., ANKE, T.,
terolemic agent produced by a Monascus spe- STERNER,0. (in press), New hexaketide inhibi-
cies, J. Antibiot. 32, 852-854. tors of eukaryotic signal transduction, Nat. Prod.
ERKEL,G., ANKE,T., VELTEN,R., STEGLICH,W. Lett.
(1991), Podoscyphic acid, a new inhibitor of Av- GERBER,N. N., SHAW,S. A., LECHEVALIER, H.
ian myeloblastosis virus and Moloney murine (1980), Structures and antimicrobial activities of
leukemia virus reverse transcriptase from a Po- peniophorin A and B, two polyacetylenic anti-
doscypha species, Z. Naturforsch. &, 442-450. biotics from Peniophora afjinins Burt., Antimi-
ERKEL,G., ANKE,T., GIMENENZ, A., STEGLICH, crob. Agents Chemother. 17, 636-641.
W. (1992), Antibiotics from basidiomycetes. GERMERDONK, R., BECKER,P., GEHRIG,I. (1993),
XLI. Clavicornic acid, a novel inhibitor of re- Verfahrenstechnische Methoden zur Verbesse-
verse transcriptases from Clavicorona pyxidata rung des Produktionsverhaltens bei der Fermen-
(Pers. ex Fr.) Doty, J. Antibiot. 45, 29-37. tation des Pilzes Cyathus striatus, in: Wege zu
ERKEL,G., ANKE,T., VELTEN,R., GIMENEZ,A,, neuen Produkten und Verfahren in der Biotech-
STEGLICH, W. (1994), Hyphodontal, a new anti- nologie (ANKE,T., ONKEN,U., Eds.). DECHE-
fungal inhibitor of reverse transcriptases from MA Monographien 129, 159-169.
Hyphodontia sp. (Corticaceae, Basidiomycetes), GIANNETTI, B. M., STEGLICH,W., QUACK,W.,
Z. Naturforsch. 49c, 561-570. ANKE,T., OBERWINCKLER, F. (1978), Merulin-
ERKEL,G., LORENZEN, K., ANKE,T., VELTEN,R., sauren A, B und C, neue Antibiotika aus Meru-
GIMENEZ,A., STEGLICH, W. (1995), Kuehnero- lius tremellosus Fr. und Phlebia radiata Fr., Z.
mycins A and B, two new biological active com- Naturforsch. 33c, 807-816.
pounds from a tasmanian Kuehneromyces sp. GILL, M. (1994), Pigments of fungi (Macromy-
(Strophariaceae, Basidiomycetes), 2. Natur- cetes), Nat. Prod. Rep. 11, 67-90.
forsch. 5oe, 1-10. GILL,M., STEGLICH,W. (1987), Pigments of fungi
ERKEL,G., ANKE,T., STERNER,0. (1996), Inhibi- (Macromycetes), in: Progress in the Chemistry of
tion of NF-KBactivation by panepoxydone, Bio- Organic Natural Products, Vol. 51 (ZECHMEI-
chem. Biophys. Res. Commun. 224214221. STER, L., Ed.), pp. 1-317. Wien, New York:
ERKEL,G., BECKER,U., ANKE,T., STERNER,0. Springer-Verlag.
(in press), Nidulal, a novel inducer of differen- GREENLEE,W. J., WOODWARD,R. B. (1976), To-
tiation of human promyelocytic leukemia cells tal synthesis of marasmic acid, J. Am. Chem. SOC.
from Nidula candida, J. Antibiot. 98,6075476.
FLOREY,H. W., CHAIN,W., HEATLEY,A., JEN- HANSON, J. R., MARTEN,T. (1973), Incorporation
NINGS,N. G., ABRAHAM, E. P., FLOREY,M. E. of [2-’H2]- and [4(R)-4-’H]-rnevalonoid hydro-
(1949), Antibiotics. London: Oxford University gen atoms into the sesquiterpenoid illudin M., J.
Press. Chem SOC. Chem. Comm. 171-172.
FORD-HUTCHINSON,A. W., GRESSER, M., HARA,M., YOSHIDA,M., MORIMOTO, M., NAKA-
YOUNG,R. N. (1994), 5-Lipoxygenase, Annu. NO, H. (1987), 6-deoxyilludin M, a new antitu-
Rev. Biochem. 63, 383-417. mor antibiotic: Fermentation, isolation and
FRAGA,B. M. (1990), Natural sesquiterpenoids, structural identification, J. Antibiot. 40, 1673-
Nat. Prod. Rep. 7 , 61-84. 1646.
FREDENHAGEN, A., KUHN, A., PETER, H. H. HARTTIG,U., ANKE,T., SCHERER,A., STEGLICH,
(1990a), Strobilurins F, G, and H, three new an- W. (1990), Leianafulvene, a sesquiterpenoid ful-
tifungal metabolites from Bolinea lutea. I. Fer- vene derivative from cultures of Mycena leaiana,
mentation, isolation and biological activity, J. Phytochemistry 29, 3942-3944.
Antibiot. 43, 655-660. HASS,R. (1992), Retrodifferentiation - an alterna-
FREDENHAGEN, A., HUG, P., PETER, H. H. tive biological pathway in human leukemia cells,
(1990b), Strobilurins F, G and H, three new anti- Eur. J. Cell Biol. 58, 1-11.
fungal metabolites from Bolinea lutea. 11. Struc- HASS, V. C., MUNACK,A. (1993), Modellierung
ture determination, J. Antibiot. 43,661-667. und Regelung der Kultivierung von Cyathius
FUJI, T., MAEDA, H., SUZUKI,F., ISHIDA, N. striatus, in: Wege zu neuen Produkten und Ver-
(1978), Isolation and characterization of a new fahren in der Biotechnologie (ANKE,T., ON-
7 References 529
KEN,U., Eds.). DECHEMA Monographien 129, ISONO,K. (1988), Nucleoside antibiotics: structure,
135-145. biological activity, and biosynthesis, J. Antibiot.
HAUTZEL, R., ANKE, H., SHELDRICK,W. S. 41, 1711-1739.
(1990), Mycenon, a new metabolite from a My- JANSEN,B. J. M., DE GROOT,A. (1991), The occur-
cena species T A 87202 (Basidiomycetes) as an rence and biological activity of drimane sesqui-
inhibitor of isocitrate lyase, J. Antibiot. 43, 1240- terpenoids, Nut. Prod. Rep. 8, 309-318.
1244. JELLISON,J., GOODELL,B., FEKETE,F., CHAND-
HAYASHI,M., WADA,K., MUNAKATA,K. (1981), HOKE, V. (1990), Fungal siderophores and their
New nematicidal metabolites from a fungus, Ir- role in wood biodegradation. The International
pex lacteus, Agric. Biol. Chem. 45, 1527-1529. Research Group On Wood Preservation. Paper
HECHT, H. J., HOFLE, G., STEGLICH,W., ANKE, prepared for the 21st Annual Meeting, Rotorua,
T., OBERWINKLER, F. (1978), Striatin A, B, and New Zealand, 13-18 May, 1990.
C: Novel diterpenoid antibiotics from Cyathus JONES,R. H., THALLER, V. (1973), Microbial poly-
striatus; X-Ray crystal structure of striatin A, J. ynes, in: Handbook of Microbiology, Vol. 111,
Chem. SOC. Chem. Commun. 665-666. Microbial Products (LASKIN,a. I., LECHEVAL-
HEIM, J., ANKE, T., MOECK, U., STEFFAN, B., IER, H. A., Eds.), pp. 63-74. Cleveland, OH:
STEGLICH,W. (1988), Antibiotics from basidio- CRC Press.
mycetes. XXIX. Pilatin, a new biologically active KANEDA,M., TAKAHASHI, R., IITAKA,Y., SHIBA-
marasmane derivative from cultures of Flagello- TA, S. (1972), Retigeranic acid, a novel sester-
scypha pilatii Agerer, J. Antibiot. 41, 1752- pene isolated from the lichens of Lobaria reti-
1757. gera group, Tetrahedron Lett. 1972, 46094611.
HERBERT,R. B. (1989), The Biosynthesis of Sec- KAVANAGH,F., HERVEY,A., ROBBINS,W. J.
ondary Metabolites. London, New York: Chap- (1949), Antibiotic substances from basidiomy-
man and Hall. cetes. IV. Marasmius conigenus, Proc. Natl.
HIROTANI,M., O'REILLY,J., DONELLY,D. M. X. Acad. Sci. USA 35, 343-349.
(1977), Fomannoxin - A toxic metabolite from KAVANAGH,F., HERVEY, A., ROBBINS,W. J.
Fomes annosum, Tertahedron Lett. 7, 651-652. (1951), Antibiotic substances from basidiomy-
cetes. VIII. Pleurotus mutilus (Fr.) Sacc. and
HOGENAUER, G. (1979), Tiamulin and pleuromu-
Pleurotus passeckerianus Pilat, Proc. Natl. Acad.
tilin, in: Antibiotics, Vol. V-1 (HAHN,F. E., Ed.),
Sci. USA 37, 570-574.
pp. 340-360. Berlin, Heidelberg, New York:
KIDA,T., SHIBAI,H., SETO,H. (1986), Structure of
Springer-Verlag.
new antibiotics, pereniporins A and B, from a
HOUCHINS, J., HIND,G. (1983), Flash spectroscop- basidiomycete, J. Antibiot. 39, 613-615.
ic characterization of photosynthetic electron KIS, Z., GLOSSE, A., SIGG, H. P., HRUBAN,L.,
transport in isolated heterocysts, Arch. Biochem. SNATZKE,G. (1970), Die Struktur von Pan-
Biophys. 224,272-282. epoxydon und verwandten Pilzmetaboliten,
HUNECK,S., CAMERON,A. F., CONOLLY,J. D., Helv. Chem. Acta 53, 185-186.
MCLAREN,M., RYCROFT, D. S. (1982), Hercy- KLEIN,K., ANKE,T., SHELDRICK, W. S., BROSS,
nolactone, a new carotene sesquiterpenoid from M., STEFFAN,B., STEGLICH,W. (1990), Fulvo-
the liverworts Barbilophozia lycopodioides and ferruginin, a carotane antibiotic from Marasmius
B. hatcheri. Crystal structure analysis, Tetrahe- fulvoferrugineus Gilliam., Z. Naturforsch. 45c,
dron Lett. 23,3959-3962. 845-850.
IKEDA,M., NIWA,G., TOHAYAMA,K., SASSA,T., KLEINKAUF, H., VON DOHREN,H. (1987), Biosyn-
MIURA,Y. (1977). Structures of fasciculol C and thesis of peptide antibiotics, Annu. Rev. Micro-
its depsipeptides, new biologically active sub- biol. 41, 259-289.
stances from Nematoloma fasciculare, Agric. KNAUSEDER, F., BRANDL,E. (1976), Pleuromutil-
Biol. Chem. 41, 1803-1805. ins: Fermentation, structure and biosynthesis, J.
ISAAC, B. G., AYER, S. W., LETENDRE,L. J., Antibiot. 29, 125-131.
STONARD,R. J. (1991), Herbicidal nucleosides KOBAYASHI H., MATSUNAGA,K., FUJII, M.
from microbial sources, J. Antibiot. 44, 729-732. (1993), PSK as chemopreventive agent, Cancer
ISHIBASHI, K., NOSE, K., SHINDO,T., ARAI, M., Epidemiol. Biomarkers Prev. 2, 271-276.
MISHIMA, H. (1968), Siccayne: a novel acetylen- KOMATSU,N., OKUBO,S., KIKUMOTO,S., KIMU-
ic metabolite of Helminthosporium siccans, Ann. RA, K., SAITO,G., SAKAI,S. (1969), Host me-
Sankyo Res. Lab. 20,76-79. diated antitumor action of schizophyllan, a glu-
ISHIZUKA,M., TAKEUCHI,T., UMEZAWA,H. can produced by Schizophyllum commune,
(1981), Studies on the mechanism of action of Gann 60,137-144.
deketocoriolin B to enhance the antibody for- KRAICZY,P., HAASE,U., GENCIC,S., FLINDT,S.,
mation, J. Antibiot. 34, 95-102. ANKE,T., BRANDT,U., VON JAGOW,G. (1996),
The molecular basis for the natural resistance of LAUER,U., ANKE,T., HANSSKE, F. (1991), Anti-
the cytochrome bc, complex from strobilurin biotics from basidiomycetes. XXXVIII. 2-meth-
producing basidiomycetes to center Q, inhibi- oxy-5-methyl-1,4-benzochinone, a thromboxane
tors, Eur. J. Biochem. 235, 54-63. A2 receptor antagonist from Lentinus adherens,
KUBO, I., KIM, M., WOOD, W. F., NAOKI,H. J. Antibiot. 44,59-65.
(1986), Clitocine, a new insecticidal nucleoside LEE, S. G., ANKE,T. (1979), Die Wirkung von
from the mushroom Clitocybe inversa, Tetrahe- Striatin A, B, und C auf den Glucose-Transport
dron Lett. 27,4277-4280. in normalen und Rous-Sarkom-Virus-transfor-
KUHNT, D., ANKE, T., BESL, H., BROSS, M., mierten Huhner-Fibroblasten, Hoppe-Seyler’s Z.
HERRMANN, R., MOCEK, U., STEFFAN,B., Physiol. Chem. 360,1170.
STEGLICH, W. (1990), Antibiotics from basidio- LEMIEUX, R. U., THORN,J. A., BRICE,C., HAS-
mycetes. XXXVII. New inhibitors of cholesterol KINS,R. H. (1951), Biochemistry of the Ustilagi-
biosynthesis from cultures of Xerula melanotri- nales. 11. Isolation and partial characterization of
cha Dorfelt, J. Antibiot. 43, 1413-1420. ustilagic acid, Can. J. Chem. 29, 409-414.
KUNIMOTO, T., HORI,M., UMEZAWA, H. (1973), LEONHARDT, K., ANKE,T., HILLEN-MASKE, E.,
Mechanism of action of diketocoriolin B, Bio- STEGLICH, W. (1987), 6-Methylpurine,6-methyl-
chim. Biophys. 298,513-525. 9-P-~-ribofuranosylpurine, and 6-hydroxy-
KUPKA, J., ANKE, T., OBERWINCKLER, F., methyl-9-P-~-ribofuranosylpurineas antiviral
SCHRAMM, G., STEGLICH, W. (1979), Antibiot- metabolites of Collybia maculata (Basidiomy-
ics from basidiomycetes. VII. Crinipellin, a new cetes), 2.Naturforsch. 42c, 420-424.
antibiotic from the basidiomyceteous fungus LEVITZKI, A. (1994), Signal-transduction therapy.
Crinipellis stipitaria (Fr.) Pat., J. Antibiot. 32, A novel approach to disease managment, Eur. J.
130-135. Biochem. 226, 1-13.
KUPKA,J., ANKE,T., LIPMANN, F. (1980), Com- LOFGREN,N., LUNING, B., HEDSTROM, H. (1954),
parison of the cytotoxic effect of three antibiot- The isolation of nebularin and the determination
ics from basidiomycetes in different cell systems, of its structure, Acta Chem. Scand. 8, 670-680.
p. 17. Abstract Annu. Meeting ASM. Washing-
LORENZEN, K., ANKE,T., ANDERS,U., HINDER-
ton, D.C.: ASM Press. MAYR,H., HANSSKE, F. (1993), Two inhibitors
KUPKA,J., ANKE,T., GIANETTI, B.M., STEGLICH, of platelet aggregation from a Panus species
W. (1981a), Antibiotics from basidiomycetes. (Basidiomycetes), 2.Naturforsch. 49c, 132-138.
XIV. Isolation and biological characterization of
hypnophilin, pleurotellol, and pleurotellic acid LORENZEN, K., ANKE,T., KONETSCHNY-RAPP, S.,
from Pleurotus hypnophilus (Berk.) Sacc., Arch. SCHEUER, W. (1995), 5-Hydroxy-3-vinyl-2(5H)-
Microbiol. 130, 223-227. furanone - a new inhibitor of human synovial
KUPKA,J., ANKE,T., STEGLICH, W., ZECHLIN, L. phospholipase A2 and platelet aggregation from
(1981b), Antibiotics from basidiomycetes XI. fermentations of a Calyptella species (Basidio-
The biological activity of siccayne, isolated from mycetes), Z. Naturforsch. 5Oc, 403409.
the marine fungus Halocyphina villosa, J. Anti- MANNING, A. M. (1996), Transcription factors: a
biot. 34, 298-304. new frontier for drug discovery, Drug Discovery
KUPKA,J., ANKE,T., MIZUMOTO, K., GIANETTI, Today 1,151-160.
B. M., STEGLICH, W. (1983), Antibiotics from MANNING, A. M., ANDERSON, D. C. (1994), Tran-
basidiomycetes. XVII. The effect of marasmic scription factor NF-KB: an emerging regulator of
acid on nucleic acid metabolism, J. Antibiot. 36, inflammation, in: Annu. Reports Med. Chem. 29
155-160. (BRISTOL,J. A., Ed.), 235-244.
KUSCHEL,A., ANKE,T., VELTEN,R., KLOSTER- MATHB, G. (1987), Bestatin compared to other
MEYER, D., STEGLICH, W. (1994), The mniopet- pharmacologic immunoregulator or modulating
als, new inhibitors of reverse transcriptases from agents, in: Horizons on Antibiotic Research
a Mniopetalum species (Basidiomycetes), J. An- (DAVIS,B. D., ICHIKAWA, T., MAEDA,K.,
tibiot. 47, 733-739. MITSCHER, L. E., Eds.), pp. 44-65. Tokyo: Japan
KWOK,0. C. H., PLATTNER,R., WEISLEDER, D., Antibiotics Research Association.
WICKLOW,D. T. (1992), A nematicidal toxin MAYER,A., ANKE,H., STERNER,0. (in press),
from Pleurotus ostreatus NRRL 3526, J. Chem. Omphalotin, a new cyclic peptide with potent
Ecol. 18, 127-136. nematicidal activity from Omphalotus olearius.
LAUER,U., ANKE,T., SHELDRICK, W. S., SCHE- I. Fermentation and biological activity, Nut.
RER, A., STEGLICH, W. (1989), Antibiotics from Prod. Lett.
basidiomyctes. XXXI. Aleurodiscal: an antifun- MAZUR,X., BECKER,U., ANKE,T., STERNER, 0.
gal sesterpenoid from Aleurodiscus mirabilis (1996), Two new bioactive diterpenes from Le-
(Berk. & Curt.) Hohn, J. Antibiot. 42, 875-882. pista sordida, Phytochemistry 43, 405-407.
7 References 531
MCMORRIS,T. C., ANCHEL,M. (1965), Fungal me- SAMUELSSON, B., DAHLEN,S.-E., LINDGREN,J.
tabolites. The structures of novel sesquiterpe- A., ROUZER,C., SERHAN,C. N. (1987), Leuko-
noids Illudin S and M, J. Am. Chem. SOC.87, trienes and lipoxins: structure, biosynthesis, and
1594-1600. biological effects, Science 237, 1171-1176.
MORISAKI, N., FURUKAWA, J., NOZOE, S., ITAI, SASAKI,T., TAKASUKA,N. (1976), Further study
A., ITAKA,Y. (1980), Synthetic studies on ma- of the structure of lentinan, an antitumor poly-
rasmane and isomarasmane derivatives, Chem. saccharide from Lentinus edodes, Carbohydr.
Pharm. Bull. 28, 500-507. Res. 47, 99-104.
MULLER, E., LOEFFLER,W. (1982), Mykologie. SAUTER, H., AMMERMANN, E., BENOIT, R.,
Stuttgart: Thieme Verlag. BRAND, S., GOLD, R. E., GRAMMENOS, W.,
MUSILEK,V. (1969), Czech. Patent CS 136492, KGHLE, H., LORENZ,G., MULLER,B., ROHL,
Chem. Abstr. 1969, 70, 18900~. F., SCHIRMER, U., SPEAKMAN, J. B., WENDE-
NAIR,M. S. R., ANCHEL,M. (1977), Frustulosinol, ROTH, B., WINGERT,H. (1995), Mitochondria1
an antibiotic metabolite of Stereum frustulosum: respiration as a target for antifungals: lessons
revised structure of frustulosin, Phytochemistry from research on strobilurins, in: Antifungal
16, 390-392. Agents; Discovery and Mode of Action (DIXON,
NERUD,F., SEDMERA, P., ZOUCHOVA,Z., MUSI- G. K., COPPING,L. G., HOLLOMON,D. W.,
LEK,V., VONDRACEK, M. (1982), Biosynthesis Eds.), pp. 173-191. Oxford: BIOS Scientific Pub-
of mucidin, an antifungal antibiotic from basi- lishers.
diomycete Oudemansiella mucida. 2H-, I3C- and SAYRE,R. M., WALTER,D. E. (1991), Factors af-
''C-labelling study, Coll. Czech. Chem. Com- fecting the efficacy of natural enemies of nema-
mun. 47, 1020-1025. todes, Annu. Rev. Phytopathol. 29, 149-166.
NISHIMURA, Y., KOYAMA,Y., UMEZAWA,S., TA- SCHNEIDER, H., MOSER,A. (1987), Process kinetic
KEUCHI,T., ISHIZUKA,M., UMEZAWA,H. analysis of pleuromutilin fermentation, Bioproc.
(1977), Chemical modification of coriolin B, J. Eng. 2, 129-135.
Antibiot. 30,5945. SCHRAMM,G. (1980), Neue Antibiotika aus HO-
heren Pilzen (Basidiomyceten). PhD Thesis, De-
OBUCHI,T., KONDOH,H., WATANABE,N., TA-
MAI, M., O M U R A , s., JUN-SHAN,Y., XIAO-
partment of Chemistry, University of Bonn,
FRG.
TIAN,L. (1990), Armillaric acid, a new antibiotic
SCHRAMM,G., STEGLICH,W., ANKE,T., OBER-
produced by Armillaria mellea, Planta Med. 56,
198-201. WINKLER, F. (1978), Antibiotika aus Basidiomy-
ceten, 111. Strobilurin A und B, antifungische
Po-r-rs, B. C. M., FAULKNER, D. J. (1992), Phos- Stoffwechselprodukte aus Strobilurus tenacellus,
pholipase A2 inhibitors from marine organisms, Chem. Ber. 111, 2779-2784.
J. Nut. Prod. 55, 1701-1717. SCHRAMM,G. STEGLICH,W., ANKE, T. (1982),
PRICE,M., HEINSTEIN, P. (1978), Cell-free biosyn- Structure-activity relationship of strobilurins,
thesis of illudins, Lloydia 41, 574-577. oudemansin, and synthetic analogues, Abstract
QUACK,W., ANKE,T., OBERWINKLER, F., GIAN- 457, in: Abstracts 13th Int. Congr. Microbiol.,
N E ~ IB., M., STEGLICH,W. (1978), Antibiotics August 8-13, Boston.
from basidiomycetes. V. Merulidial, a new anti- SCHULTES,R. E., HOFMANN,A. (1980), Pflanzen
biotic from the basidiomycete Merulius tremello- der Cotter. Bern, Stuttgart: Hallwag.
sus Fr., J. Antibiot. 31, 737-741. SCHWITZGEBEL,K. (1992), Untersuchungen an
RABE,U. (1989), Fermentation von Cyathus stria- holzabbauenden und holzverfarbenden Pilzen,
lus (Huds. ex Pers.) Willd (Basidiomycetes) und M . Sc. Thesis, Department of Biology, Univer-
Biosynthese der Striatine. Dissertation, Univer- sity of Kaiserslautern, FRG.
sity of Kaiserslautern, FRG. SEDMERA,P., MUSILEK,V., NERUD,F. (1981),
RIEDL, K. (1976), Studies on pleuromutilin and Mucidin: Its identity with strobilurin A, J. Anri-
some of its derivatives, J. Antibiot. 29, 132-139. biot. 34, 1069.
ROBERTS,D. W. (1981), Toxins of entomopatho- SEGIGUCHI,J., GAUCHER,M. (1979), Isoepoxy-
genic fungi. In: Microbial Control of Pests and don, a new metabolite of the patulin pathway in
Plant Diseases 1970-1980 (BURGES,H. D., Ed.), Penicillium urticae, Biochem. J. 182, 445-453.
pp. 441-464. London, New York: Academic SHINOZAWA, S., TSUTSUI,K., ODA, T. (1979), En-
Press. hancement of the antitumor effect of illudin S by
SALMON,J. A., GARLAND,L. G. (1991), Leuko- including it into liposomes, Experientia 35,
trienes, antagonists and inhibitors of leukotriene 1102-1103.
biosynthesis as potential therapeutic agents, in: SIMON, B. (1994), Antivirale und cytotoxische
Progress in Drug Research (JUCKER,E., Ed.), Wirkstoffe aus Basidiomyceten. Ph.D. Thesis,
pp. 9-90. Basel: Birkhauser. University of Kaiserslautern, FRG.
SIMON, B., ANKE,T., ANDERS,U., NEUHAUS,M., pressive effect on 3-methylcholantrene-induced

HANSSKE, F. (1995), Collybial, a new antibiotic carcinogenesis, Cancer Res. 44,5132-5137.
sesquiterpenoid from Collybia confluens (Basid- TAKAHASHI, K., SOMEYA, T., MURAKI, S., YOSHI-
iomycetes), 2. Naturforsch. 5Oc, 173-180. DA, T. (1980), A new keto alkohol, (-)-minth-
STADLER,M., ANKE,T., DASENBROCK, J., STEG- lactone, ( + )-isorninthlactone and minor compo-
LICH,W. (1993a), Phellodonic acid, a new bio- nents in peppermint oil, Agaric. Biol. Chem. 44,
logically active hirsutane derivative from Phello- 1535-1543.
don melaleucus (Thelephoraceae, Basidiomy- TAKAHASHI, A., KUSANO,G., OHTA,T., OHIZU-
cetes), Z. Naturforsch. 48c,545-549. MI,Y., NOZOE,S. (1989). Fasciculic acids A, B
STADLER,M., MAYER,A., ANKE,H., STERNER, and C as calmodulin antagonists from the
0. (1993b), Fatty acids and other compounds mushroom Naematoloma fasciculare, Chem.
with nematicidal activity from cultures of basid- Pharm. Bull. 37,3247-3250.
iomyocetes, Planta Med. 60, 128-132. TAKEUCHI, T., IINUMA,H., IWANAGA, J., TAKA-
STADLER, M., ANKEH., STERNER,0. (1994), New HASHI, S., TAKITA, T., UMEZAWA, H. (1969),
nematicidal and antimicrobial compounds from Coriolin, a new basidiomycete antibiotic, J. Anti-
the basidiomycete Cheimonophyllum candidissi- biot. 22, 215-217.
mum, J. Antibiot. 47, 1284-1289. THOMPSON, C. B. (1995), Apoptosis in the patho-
STADLER, M., FOURON, J.-F., STERNER, O., ANKE, genesis and treatment of disease, Science 267,
H. (1995), 1,2-dihydroxyminthlactone, a new 1456-1462.
nematicidal monoterpene isolated from the TOMODA,H., HUANG,X. H., COA, J., NISHIDA,
basidiomycete Cheimonophyllum candidissi- H., NAGAO, R., OKUDA,S.,TANAKA, H., OMU-
mum (Berk & Curt.) Sing., Z. Naturforsch. 5Oc, RA,S., ARAI,H., INOUE,K. (1992), Inhibition of
473-475. acyl-CoA :Cholesterol acyltransferase activity by
STARK,A., ANKE,T., MOECK,U., STEGLICH, W., cyclodepsipeptide antibiotics, J. Antibiot. 45,
KIRFEL,A., WILL,G. (1988), Lentinellic acid, a 1626-1632.
biologically active protoilludane derivative from TSUGAGOSHI, S., HASHIMOTO, Y., FUJI, G., Ko-
Lentinellus species (Basidiomycetes), 2. Natur- BAYASHI, H., NOMOTO,K., ORITA,K. (1984),
forsch. 43c, 177-183. Krestin (PSK), Cancer Treat. Rev. 11,131-155.
STARK,A., ANKE,T., MOCEK,U., STEGLICH,W. TURNER,W. B. (1971), Fungal Metabolites. Lon-
don: Academic Press.
(1991), Omphalone, an antibiotically active ben- TURNER,W. B., ALDRIDGE, D. C. (1983), Fungal
zoquinone derivative from fermentations of
Metabolites 11. London: Academic Press.
Lentinellus omphalodes, Z. Naturforsch. 46c,
989-992. UMEZAWA, H., AOYAGI, T., SUDA,H., HAMADA,
M., TAKEUCHI, T. (1976), Bestatin, an inhibitor
STERNER, O., BERGMAN, R., KIHLBERG,J., WICK- of aminopeptidase B, produced by actinomy-
BERG, B. (1985), The sesquiterpens of Laccarius
cetes, J. Antibiot. 29, 97-99.
vellereus and their role in a proposed chemical VIDARI,G., DE BERNARDI, M., VITA-FINZI, P.,
defense system, J. Nut. Prod. 48, 279-288. FRONZA,G. (1976), Sesquiterpenes from Lac-
STERNER,O., CATRE,R., NILSON,L. (1987a). tarius blennius, Phytochemistry 15, 1953-1955.
Structure-activity relationship for unsaturated VINING, L. C. (1990), Function of secondary
dialdehydes. 1. The mutagenic activity of 18 metabolites, Annu. Rev. Microbiol. 44, 395-427.
compounds in the Salmonellalmicrosome assay, VONJAGOW,G., GRIBBLE, G. W., TRUMPOWER,
Mutat. Rex 188, 169-174. B. L. (1986), Mucidin and strobilurin A are iden-
STERNER, O., STEFFAN, B., STEGLICH, W. (1987b), tical and inhibit electron transfer in the cyto-
Novel azepine derivatives from the pungent chrome bcl complex of the mitochondria1 respi-
mushroom Chalciporus piperatus, Tetrahedron. ratory chain at the same site as myxothiazol,
43,1075-1082. Biochemistry 25, 775-780.
STERNER,O., ETZEL,W., MAYER,A., ANKE,H. WEBER,W., ANKE,T., STEFFAN, B., STEGLICH, W.
(in press), Omphalotin, a new cyclic peptide with (1990a), Antibiotics from basidiomycetes.
potent nematicidal activity from Omphalotus XXXII. Strobilurin E: a new cytostatic and anti-
olearius. 11. Isolation and structure determina- fungal (E)-pmethoxyacrylate antibiotic from
tion, Nut. Prod. Lett. Crepidotus fulvotomentosus Peck, J. Antibiot. 43,
STEWART, K. R. (1986), A method for generating 207-212.
protoplasts from Clitopilus pinsitus, J. Antibiot. WEBER,W., ANKE,T., BROSS,M., STEGLICH, W.
39, 1486-1487. (1990b), Strobilurin D and Strobilurin F: two
SUGA,T., SHIO,T., MAEDA,Y. Y., CHIHARA, G. new cytostatic and antifungal (E)-pmethoxy-
(1984), Antitumor activity of lentinan in murine acrylate antibiotics from Cyphellopsis anomala,
syngeneic and autochthonous hosts and its sup- Planta Med. 56, 446-450.
7 References 533
WEBER,W., SEMAR,M., ANKE, T., BROSS, M., and armillatin, from Armillaria mellea, Planta
STEGLICH,W. (1992), Tyromycin A: a novel in- Med. 57,478-480.
hibitor of leucine and cysteine aminopeptidase ZAHNER,H. (1982), Mikrobieller Sekundarstoff-
from Tyromycetes lacteus, Planta. Med. 58, 56- wechsel, in: Handbuch der Biotechnologie
59. (PRAvE, P., FAUST,U., SITTIG,W., SUKATSCH,
WEBER, W., SCHU, P., ANKE, T., VELTEN, R., D. A., Eds.). Wiesbaden: Akademsiche Verlags-
STEGLICH,W. (1994), Caloporosid, a new inhi- gesellschaft.
bitor of phospholipases C from Caloporus di- ZAPF, S. (1994), Neue antifungische Antibiotika
chrous (Fr.) Ryv., J. Antibiot. 47, 1188-1194. aus Pilzen. Ph.D. Thesis, University of Kaisers-
lautern, FRG.
WINKELMANN, G. (1986), Iron complex products,
ZAPF, S., WERLE,A., ANKE,T., KLOSTERMEYER,
in: Biotechnology 1st Edn., Vol. 4 (REHM,H. J.,
REED, G., Eds.), pp. 216243. Weinheim: VCH. D., STEFFAN,B., STEGLICH,W. (1995), 9-Meth-
oxystrobilurine - Bindeglieder zwischen Strobil-
WUNDER, A., ANKE, T., KLOSTERMEYER,D., urinen und Oudemansinen, Angew. Chem. 107,
STEGLICH,W. (1996), Lactarane type sesquiter- 255-257.
penoids as inhibitors of leukotriene biosynthesis ZAPF, S., WUNDER, A., ANKE, T., KLOSTER-
and other new metabolites from submerged cul- MEYER,D., STEGLICH,W., SHAN,R., STERNER,
tures of Lentinellus cochleatus (pers. ex Fr.) O., SCHEUER,W. (1996), (+)-lOa-hydroxy-4-
Karst., Z. Naturforsch. Slc, 493-499. muurolen-3-one, a new inhibitor of leukotriene
YANG,J. S., Su, Y. L., WANG,Y. L., FENG,X. Z., biosynthesis from a Favolaschia species. Com-
Yu, D. Q., LIANG,X. T. (1991), Two novel pro- parison with other sesquiterpenes, Z. Natur-
toilludane norsesquiterpenoid esters, armillasin forsch. 51c, 487-492.
Biotechnology
12 cyc10sp0rins:
Recent Developments in
Biosynthesis,
Pharmacology and Biology,
and Clinical Applications
JORG KALLEN
VINCENTMIKOL
VALERIEF. J. QUESNIAUX
MALCOLMD. WALKINSHAW
Basel. Switzerland
ELISABETHSCHNEIDER-SCHERZER
KURT SCH~RGENDORFER
GERHARDWEBER
Kufstein-Schaftenau, Austria
HANSG. FLIRI
Vitry-sur-Seine, France
536 12 Cyclosporins: Recent Developments in Biosynthesis, Pharmacology and Biology
1 Introduction 538
2 Clinical Applications of Cyclosporins 539
2.1 Introduction 539
2.2 Transplantation 539
2.3 Autoimmune Diseases 540
2.4 Activity Against Tumor Multidrug Resistance 540
2.5 Anti-HIV Activity 540
3 Mode of Immunosuppressive Action 542
3.1 Introduction to T Cell Activation 542
3.1.1 Signal Recognition 542
3.1.1.1 First Signal 542
3.1.1.2 Costimulatory Signals 542
3.1.2 Signal Transduction 544
3.1.3 Gene Activation 545
3.1.4 Cytokine Receptor Expression 545
3.1.5 Release of Cytokines 545
3.2 Current Knowledge of the Mode of Immunosuppressive Action of CsA 546
3.2.1 Cyclosporin Receptors 547
3.2.2 Active Sites on the Drugs 548
3.2.3 Target Proteins of the Drug-Immunophilin Complexes 549
3.2.4 Downstream Effects in the Nucleus 549
3.3 Molecular Evidence for other Biological Activities of Cyclosporins 550
3.3.1 Antiinflammatory Effects 550
3.3.2 Cyclosporins as Anti-HIV Agents 550
3.3.3 Cyclosporins as Drug Resistance Modifiers 551
3.3.4 Antimalarial Activity of Cyclosporins 551
4 Chemistry 551
4.1 Structural Aspects 551
4.2 Chemical Synthesis and Production 552
4.2.1 MeBmt Transformations and Variations 555
4.2.2 Chemical Transformations of the Cyclic Peptide Backbone 555
4.2.3 Selective Ring Opening Reactions 556
4.2.4 Cyclosporins Incorporating other Non-Proteinogenic Amino Acids 556
4.3 Structure-Activity Relationships 556
5 Structural Investigations of Cyclosporins, Cyclophilins, and their Complexes 559
5.2 Structural Investigations of Uncomplexed Cyclosporins 562
5.2.1 The 3D-Structure of Uncomplexed CsA in Apolar Environment 562
5.2.2 The 3D-Structure of the Uncomplexed Peptolide SDZ 214-103 in Apolar
Environment 562
5.2.3 The 3D-Structure of Uncomplexed CsA in Polar Environment 563
5.3 The 3D-Structure of Cyclophilin A Complexed with a Tetrapeptide 563
5.4 The Peptidyl-Prolyl Isomerase Active Site of Cyclophilin A 564
5.5 The 3D-Structure of Cyclophilin A Complexed with CsA 566
5.6 The X-Ray Structures of Cyclophilin A Complexed with Cyclosporin Derivatives 568
5.6.1 [MeBm2t]-Cyclosporin 568
5.6.2 MeIle4-Cs (SDZ NIM 811) 569
5.6.3 The 3D-Structure of SDZ 214-103 Bound to Cyclophilin A 569
5.7 The X-Ray Structure of Cyclophilin B Complexed with a Cyclosporin Derivative 570
5.8 The X-Ray Structure of Cyclophilin C Complexed with CsA 571
5.9 Conclusions 571
Contents 537
6 Biosynthesis of Cyclosporins 572

6.2 Cyclosporin Synthetase 572
6.2.1 Enzymatic Activities of Cyclosporin Synthetase 572
6.2.2 Characterization of the Enzyme 573
6.2.3 Isolation and Characterization of the Cyclosporin Synthetase Gene 574
6.2.4 Manipulation of the Cloned Cyclosporin Synthetase Gene by Integrative
Transformation 575
6.2.5 Mechanistic Aspects of Biosynthesis 577
6.2.6 Biosynthesis of Cyclosporin Variants 580
6.2.7 Related Enzymes: Peptolide SDZ 214-103 Synthetase 580
6.3 Alanine Racemase 580
6.3.1 Alanine Racemizing Activity 580
6.3.2 Characterization of the Enzyme 581
6.4 Bmt-Synthesizing Enzymes 581
6.4.1 Polyketide Origin of Bmt 581
6.4.2 Identification of the Basic Assembly Product and Characterization of
Bmt-Polyketide Synthase 581
6.4.3 The Transformation Process 583
7 References 583
1 Introduction natural sources. In addition, a close analog

named SDZ 214-103, incorporating a lactone
function in place of a peptide bond and exhi-
Cyclosporins are a family of hydrophobic biting a similar biological profile was also dis-
cyclic undecapeptides with a remarkable covered from natural sources (Fig. 2). The
spectrum of diverse biological activities. The discovery and structure elucidation of cyclo-
first member of this class to be discovered sporins have been amply reviewed (WENGER
was named cyclosporin A (CsA; structure 1986; WENGERet al., 1986; VON WARTBURG
shown in Fig. 1).To this date, some 30 mem- and TRABER,1988 FLIRI and WENGER,
bers of this family have been isolated from 1990).
c
n c
I H
0
n
Fig. 1. Structure of cyclosporin A
(Sandimmun"), including the number-
I ing system.
I I Fig. 2. Structure of SDZ 214-103.

In this chapter, emphasis is given to some and FK506 was greatly facilitated by the
more recent aspects of chemistry, biosynthe- availability of a third compound, rapamycin,
sis, and biological activity. which binds to the same receptors as FK506,
yet exhibits a different spectrum of biological
activities. A detailed account of these aspects
is given in Sects. 3.2. and 3.3. In this section
present clinical applications of Sandimmunm
2 Clinical Applications of are discussed. They include the following in-
dications:
Cyclosporins allograft rejection,
Behset’s uveitis,
2.1 Introduction rheumatoid arthritis,
aplastic anemia (NDA’s pending),
Cyclosporin A was initially isolated as an nephrotic syndrome,
antifungal antibiotic. It was later shown to atopic dermatitis (NDA’s pending),
possess immunosuppressive properties of psoriasis vulgaris.
high therapeutic value. Since 1983, cyclospo-
rin A, under the trade name Sandimmunm,
has been in clinical use worldwide to prevent L.2 Transplantation
rejection of organ transplants. It has subse-
quently been approved for the therapy of cer- Sandimmunm is a reversible inhibitor of the
tain autoimmune diseases. Since the time of transcription of interleukin 2 (IL-2) and sev-
market introduction of SandimmunB, many eral other lymphokines, most notably in help-
additional biological activities of cyclosporins er T lymphocytes (see Sect. 3.2). As a conse-
have been discovered, some of which may quence, it suppresses the activation and/or
lead to novel clinical applications of cyclospo- maturation of various cell types, in particular
rin A or of non-immunosuppressive analogs. those involved in cell-mediated immunity.
At the time of market introduction of Sand- Because of these properties, Sandimmunm has
immunm, the mechanism by which this drug become the first-line immunosuppressant for
mediates immunosuppression was not under- prophylaxis and therapy of transplant rejec-
stood at the molecular level nor was a recep- tion. In fact, the modern era of transplanta-
tor known. Since then, not only was a whole tion surgery was only possible after the avail-
family of receptors discovered (i.e., the cyclo- ability of cyclosporin. The first patient to re-
philins), but a possible role of these proteins ceive a kidney graft under CsA treatment was
for protein folding and cellular protein traffic reported in 1978 (CALNEet al., 1978). Soon
has emerged. Much of what is known today thereafter, transplantations of liver, heart,
about cyclosporins, cyclophilins and their bio- and combined lung-heart commenced
chemistry was greatly aided by the discovery (ERNST,1991; BARRY,1992; KAHAN,1992;
of FK506, an immunosuppressive macrolide. TSANG et al., 1992). In 1991, only in Germany
This compound elicited much interest be- 450 liver transplantations were performed
cause, like cyclosporin A, it was a T cell selec- (HOPFet al., 1992). Organ availability has be-
tive immunosuppressant, but much more po- come a major limiting factor and numerous
tent. This activity was soon shown to be based patients die while awaiting a donor organ.
on a mechanism identical to that of cyclospo- Therefore, organ preservation techniques
rin. Search for FK506 receptors led to the dis- have become an important aspect in the area
covery of the FK506 binding proteins of transplantation surgery. Histocompatibility
(FKBPs), a novel protein family with no ho- matching, besides immunosuppression, is the
mologies to cyclophilins, yet with many prop- key factor contributing to long-term graft sur-
erties in common. Like the cyclophilins, vival. Currently, expected 10-year first graft
FKBPs appear to have a functional role in survival rates for kidneys from HLA-identical
protein folding. Unveiling the mechanism of siblings, l-haplotype-matched relative, and
immunosuppressive activity of Sandimmunm cadaver donors are 74, 51, and 40%, respec-
tively (BARRY,1992). The survival probabili- tage in a number of indications is the steroid-
ty after lung transplantation is approximately sparing effect of Sandimmun". To avoid re-
65% after 1 year and 50% after 3 years lapse after control of active disease, patients
(ERNST,1991). Major problems encountered should continue receiving SandimmunB main-
in transplantation surgery are technical diffi- tenance therapy at the lowest effective dose.
culties during operation, serious infections,
and acute rejection episodes during the first
postoperative period, and chronic (long-term) 2.4 Activity Against Tumor
rejection. Side effects can be classified into Multidrug Resistance
those associated to immunosuppression (lym-
phoproliferative disorders, infectious diseases Cellular resistance to cytotoxic drugs is oft-
caused by bacterial and fungal pathogens as en the cause of inefficient treatment of cancer
well as viruses), and other adverse effects with potent antitumor drugs. While many
which are specific for the immunosuppressive mechanisms of resistance occur, the mecha-
drugs used. For Sandimmun" these include nism of "multidrug resistance" (MDR) has
primarily impairment of renal function, hy- received particular attention (ENDICOTTand
pertension, hirsutism, and gingival hyperpla- LING, 1989). Most often, this type of resist-
sia (MASON, 1989). Neurological and gas- ance extends to several anticancer drugs of
trointestinal effects are also common in Sand- unrelated structural classes and mechanisms
immun" recipients but are usually mild to of action. A common feature of MDR is over-
moderate and resolve on dosage reduction. expression of a particular class of transmem-
brane glycoproteins called P-glycoproteins
(Pgp) which serve as transport proteins rap-
2.3 Autoimmune Diseases idly effluxing antitumor drugs out of the tu-
mor cells as soon as they have entered
Since SandimmunB not only suppresses through the membrane. As a consequence,
cell-mediated immunity but also humoral im- Pgp transporters decrease intracellular drug
mune responses and inhibits chronic inflam- concentrations below their active threshold.
matory reactions it appeared very promising Numerous in vitro studies have described
in the treatment of autoimmune diseases. agents which can restore the sensitivity of
Prospective controlled trials performed in pa- MDR tumor cells, including cyclosporin A at
tients with autoimmune diseases have recent- clinically achievable concentrations (TWEN-
ly been reviewed (FREY,1990; FAULDS et al., TYMAN,1992). Moreover, this effect can be
1993). Efficacy could be proven for the fol- dissociated from immunosuppression, as non-
lowing diseases: Endogenous uveitis, rheuma- immunosuppressive analogs have been shown
toid arthritis, Sjogren's syndrome, myasthenia to retain resistance modifier activity and
gravis, psoriasis, atopic dermatitis, Crohn's some are even more potent than cyclosporin
disease. The drug is considered as a first-line A. One such analog from Sandoz Pharma AG
therapy in patients with moderate or severe called SDZ PSC 833 is approximately tenfold
aplastic anemia who are not eligible for bone more potent than Sandimmun" as a resist-
marrow transplantation. It may also be of ance modifier and is currently undergoing
benefit in patients with primary biliary cirrho- clinical trials (BOESCHet al., 1991). The struc-
sis and intractable pyoderma gangrenosum. ture of SDZ PSC 833 is shown in Fig. 3.
Sandimmun" does not appear to be effective
in patients with allergic contact dermatitis,
multiple sclerosis, or amyotropic lateral scle- 2.5 Anti-HIV Activity
rosis. Successful application in insulin-de-
pendent diabetes will depend on the develop- A possible beneficial effect of Sandimmun"
ment of diagnostic tools indicating early disin HIV disease has been proposed as early as
ease onset before beta cell destruction has 1986 (ANDRIEUet al., 1986). The rationale is
progressed too far and clinically overt dia- that activation of CD4+ cells which is re-
betes is present. The most significant advan- quired for HIV replication (ZACKet al., 1990
Fig. 3. Structure of SDZ PSC 833. I

STEVENSON et al., 1990) is inhibited by CsA. clinical research laboratories at the Sandoz
In addition, CsA would inhibit the initiation Research Institute of Sandoz Pharma AG in
of an autoimmune process involving killing of Vienna, Austria. Compounds were evaluated
HIV infected lymphocytes by cytotoxic cells for antiviral, cytotoxic, and for immunosup-
and may also counteract HIV-induced apop- pressive activity in v i m . It was found that
totic cell death of CD4+ cells (HABESHAW et some non-immunosuppressive analogs of
al., 1990). SandimmunB were equal or even superior in
A thorough investigation of a series of im- their antiviral activity without being cyto-
munosuppressive and non-immunosuppres- toxic. One such analog, SDZ NIM 811
sive cyclosporins was performed in the pre- (Fig. 4), is comparable to Sandimmunm re-
Fig. 4. Structure of SDZ NIM 811.

garding oral bioavailability and pharmacoki- and cytotoxic suppressor functions. The mul-
netics in animals and appears to be of lower tiple steps involved in this process, namely
nephrotoxicity (ROSENWIRTH et al., 1994). signal recognition, signal transduction to the
nucleus, resulting in gene activation, expres-
sion of growth factor receptors, growth factor
synthesis and cell proliferation, are briefly re-
viewed (RYFFEL,1989).
3 Mode of
Immunosuppressive 3.1.1 Signal Recognition
Action
3.1.1.1 First Signal
In addition to its clinical use as an immuno-
suppressant CsA has also been widely em- After uptake and limited proteolysis, the
ployed as an experimental tool for basic re- antigen processed by the antigen presenting
search. It has helped to understand the bio- cell is recognized in the context of the major
chemical events needed to translate a signal histocompatibility complex (MHC) by the an-
from the T cell surface to the nucleus and the tigen receptor on T lymphocytes (Fig. 5). The
pathophysiological processes involving lym- binding of antigen to the T cell receptor is an
phocyte activation in a variety of diseases. absolute requirement for T cell activation un-
Early immunological studies revealed that der physiological conditions. The T cell re-
CsA exerts specific effects on T cell lympho- ceptor is a multicomponent structure consist-
kine transcription (KRONKEet al., 1984). Be- ing of the clonotypic (Y and p (or y and 8)
cause T cells are prominent in the cellular im- chains and the invariant CD3 subunits y, S,E,
mune response, studies on the mechanism of and 7. The complete assembly of all com-
immunosuppression by CsA have mainly fo- ponents is required for cell surface expres-
cused on its role in regulating gene expression sion, and thus for antigen receptor function.
in T lymphocytes. To place CsA activity in The 5 and 7 subunits of CD3 most likely
perspective a summary on T cell activation is transduce to the cytoplasm the activating sig-
first given below. nals originating from antigen recognition by
the T cell receptor. Antibodies against the
CD3 complex can induce T cell functional re-
3.1 Introduction to T Cell sponses that are identical to antigen-induced
responses, regardless of antigen specificity. In
Activation addition, the two transmembrane proteins
CD4 and CD8 expressed on helper and cyto-
A schematic representation of the cellular toxic T cells participate in the interaction be-
immune response with emphasis on the cen- tween the T cell and the antigen presenting
tral role of the activated T lymphocytes is cell by binding to MHC class I1 and I mole-
shown in Fig. 5. For T cell activation, the an- cules, respectively. Originally, they were
tigen receptor on the T cell surface interacts called coreceptors because their association
with the processed antigen exposed in the with an intracellular enzyme facilitates signal-
proper histocompatibility context on the sur- ing during T cell activation (JANEWAYet al.,
face of the antigen presenting cell ( A P C cf. 1989).
Fig. 5). In the presence of additional accesso-
ry interactions between the T cell and the an-
tigen presenting cell, antigen recognition 3.1.1.2 Costimulatory Signals
leads to biochemical events which finally re-
sult in proliferation, differentiation, and ma- In addition to the molecular interactions
turation of the T cell to T effector cells with between the T cell receptor, CD3, CD4, or
specific immunological function, e.g., helper CD8 and the antigen presenting cell, costimu-
W W
I Modulation of calcineurin phosphatase activity I I p70S6kinase I
(other kinases/phosp.hatases ?)
? ? 1
t RNA pol II a
Fig. 5. CsA and FK506 both interfere, by binding to their respective immunophilins, with the function of
intracellular molecules that transmit calcium-associated signals between the T-cell receptor (TCR) and the
activation of lymphokine genes (IL-2) in the nucleus. Transcriptional regulation of IL-2 gene expression is
modulated by the combination of transcription factors (e.g., NF-AT, NFKB, OTF-1) interacting with their
corresponding recognition sites at the IL-2 promoter. These DNA/protein complexes, together with RNA
polymerase I1 (RNA pol 11), result in the antigen-inducible transcription of IL-2. Potential intervention
sites for the pentameric complex (calcineurin A (p61), B (p19), calmodulin (p17), immunophilin, drug),
involving, e.g., modification and translocation of antigen-inducible transcription factors (NF-AT NFKB
(p50, p65)), are indicated (11). CsA and FK506 interfere with the Go to G1 transition of the cell cycle,
whereas raparnycin interferes with the G1 to S transition (for details, see text) (adapted from BAUMANN
and BOREL,1992).
latory signals contribute to the T cell recep- of the T cell receptorKD3 complex reveals
tor-driven proliferative response. These costi- intrinsic catalytic domains, the protein tyro-
mulatory signals are thought to be delivered sine phosphorylation of CD3 is likely to be
by antigen presenting cells but not by tissue mediated by intracellular protein tyrosine
cells and might play an important role in the kinases. Two members of the src family of
selfhon-self discrimination by influencing the protein tyrosine kinases have been identified
consequences of the T cell receptor engage- in this context: ~59""which is physically asso-
ment (LIU and LINSLEY,1992). Conversely, ciated with the T cell receptorKD3 complex,
in v i m stimulation of T cells in the absence and ~ 5 6 ' 'which
~ is associated with CD4 or
of costimulation might lead to functional CD8 and becomes activated upon coligation
inactivation of the T cells, i.e., clonal anergy, of CD4 or CD8 with the T cell receptor. How
or to activation-induced cell death (apopto- this signal is being transmitted from the T cell
sis). receptor-proximal protein tyrosine kinase ac-
The molecular basis for T cell costimula- tivities to the cytoplasmic protein serinel
tion is not yet fully understood. The accessory threonine kinase and phosphatase cascades is
molecules are membrane proteins including less clear. In addition, the phosphorylation of
the lymphoid adhesion molecules. They are tyrosine residues in the cytoplasmic domains
invariant and bind to ligands expressed on the of receptors such as CD3 6 is required for the
surface of other cells such as antigen present- recruitment of several cellular enzymes like
ing cells or target cells, thereby increasing the phospholipase C (PLC) to the cytoplasmic
strength of adhesion between these cells and domain of the receptors. The yl and 2 iso-
the T cells. In addition, accessory molecules forms of the phosphatidyl inositol-specific
may transduce biological signals to the T cell PLC depend on tyrosine phosphorylation for
cytosol. CD2, e.g., which interacts with the their activation. Once activated, PLC in-
leukocyte function-associated antigen3 creases the hydrolysis of inositol phospholip-
(LFA-3, CD58) augments specific signaling ids and produces two second messenger mole-
through the T cell receptor. In contrast, CD28 cules. One, inositol triphosphate (IP3) binds
binding to the B7-1 and B7-2 molecules on to intracellular vesicles that store calcium ions
antigen presenting cells initiate a signaling (Ca'+) causing them to release calcium into
pathway which is distinct from the pathway the cytoplasm. Ca2+ ions then bind to a small
emanating from the T cell receptor. protein, calmodulin, which acts as a regulato-
ry subunit for other enzymes essential for T
cell activation such as the serine/threonine
3.1.2 Signal Transduction phosphatase calcineurin. Another second
messenger molecule is diacylglycerol (DAG),
The antigen binding to the T cell receptor/ a lipid molecule that remains in the mem-
CD3 complex in combination with costimula- brane where it activates protein kinase C.
tory signals triggers a highly complicated sig- Once activated, protein kinase C is translo-
nal transduction pathway which finally results cated from the cytosol to the plasma mem-
in the pleiotropic T cell activation program. brane where it phosphorylates membrane-
Although the molecular details of this pro- bound proteins. This event subsequently di-
gram are still largely unknown, an attempt to rects the modification of a set of other signal
summarize the current state of our knowledge transmitting proteins. Finally, phosphoryla-
with a focus on the mechanism of action of tion and dephosphorylation of certain tran-
known immunosuppressants is schematically scription factors such as NFKB (nuclear factor
represented in Fig. 5 (IZQUIERDOand CAN- for K-light chain expression in B cells) and
TRELL,1992; ABRAHAM et al., 1992). NF-AT (nuclear factor of activated T cells)
One of the earliest events after antigen induce or repress the transcription of their
binding is the activation of protein tyrosine target genes such as the IL-2 gene. The effect
kinases which initiates a cascade of down- of the modification is either direct by increas-
stream biochemical events (SEFTON and ing or decreasing the affinity of the transcrip-
CAMPBELL, 1991). Since none of the subunits tion factors for their specific binding site on
the DNA or indirect via additional protein- DNA synthesis and cell division. Late genes
protein interactions (HUNTER and KARIN, require both DNA synthesis and cell division
1992). for their expression. They are synthesized
during the G2 phase of the cell cycle.
The manyfold mechanisms used by T cells
3.1.3 Gene Activation to regulate their immediate, early, and late
activation genes include alteration of the
The metabolic events resulting from expo- transcriptional rate, of initiation and termina-
sure of T cells to antigen culminate in blast tion of transcription, and of mRNA stability.
transformation and progression through the It is the group of early genes that is most sen-
cell cycle. Activated T cells express novel sitive to immunosuppressants like CsA and
functions such as lymphokine secretion and FK506. The expression of this group of genes
lytic capability which are related to their ef- is mainly regulated at the level of transcrip-
fector role in the immune response. They in- tion initiation. Transcription is initiated once
crease in volume and undergo rapid increases the transcription factors such as NFKB, jun,
in phosphatidyl inositol metabolism, in cyto- fos, or oct-1, -2 that exist as precursors in the
plasmic pH, intracellular free calcium concen- cytoplasm have been translocated into the nu-
tration, and in the serinehhreonine and tyro- cleus and are activated by phosphorylation or
sine phosphorylation pattern of various cellu- dephosphorylation. This enables them to bind
lar proteins. Progression through the cell cy- to their specific regulatory elements and par-
cle is associated with a general increase in ticipate in a functional transcription complex
protein, lipid and RNA synthesis and with an with RNA polymerase I1 (HUNTERand KA-
increase of the mRNA level for a variety of RIN, 1992).
activation-related genes. Many of these genes
encode lymphokines and surface receptors
necessary for the expansion and the immune 3.1.4 Cytokine Receptor
functions of the activated T cells. About 70- Expression
100 genes are activated in T cells during the
differentiation program which is taking place T cells express a series of cytokine recep-
at times ranging from 15 min to 14 d following tors on their membrane which allow them to
stimulation (CRABTREE,1989). respond to various cytokines including IL-1,
By analogy to viral systems the T cell acti- IL-2, IL-4, IL-7, IL-9, IL-10, and IL-12. Some
vated genes can be divided into three groups: of these receptors seem to be upregulated
the immediate, the early, and the late genes. during T cell activation. The receptor for IL-2,
Immediate genes (e.g., c-fos,c-myc) are tran- a major growth factor for T cells, is composed
scribed after activation with no need for proof three subunits (a,p, and y) in its high-
tein synthesis. Their products appear very affinity form (TAKESHITA et al., 1992). The p
soon after stimulation, usually within 10- and y chains associate to form receptors of in-
30 min. Transcription of the immediate genes termediate activity which are expressed on
takes place during the transition from the resting T cells. Upon T cell activation by an-
quiescent state (Go) into the G1 phase of the tigen or by IL-2 itself, the expression of the a
cell cycle. Immediate genes are usually iden- chain is upregulated and contri-
tified by their ability to be transcribed in the butes to form high-affinity receptors. The IL-4
presence of a protein synthesis inhibitor such receptor 130 kDa protein is also upregulated
as cycloheximide or anisomycin. Early genes, during T cell activation by mitogenic ligands
which include the lymphokines (e.g., IL-2, IL-3, or by IL-4 (ARMITAGE et al., 1990).
IL-4, interferon- y, GM-CSF, and IL-2 recep-
tor a) are transcribed after the immediate
genes and require postactivation protein syn- 3.1.5 Release of Cytokines
thesis. The early genes are involved in the
mid to late G1 phase of the cell cycle and in- Several cytokines are produced by T cells
clude genes coding for products required for upon activation, namely IFN-y, IL-2, IL-3,
IL-4, IL-5, IL-6, IL-9, IL-10, TNFa, and GM- acterized as antifungal antibiotics. The struc-
CSF. On the other hand, antigen presenting tures of FK506 and rapamycin, together with
cells release IL-1, IL-6, IL-12, IL-13, TNFa that of cyclosporin A, are shown in Fig. 6.
and p and other cytokines. These cytokines FK506 is a neutral macrolactone produced by
have multiple regulatory effects such as the Streptomyces tsukubaensis, a soil microorgan-
autocrine stimulation of helper T cells and ism collected in the Tsukuba area of northern
the differentiation of cytotoxic and suppres- Japan (KINO et al., 1987). FK506 prolongs the
sor T lymphocytes, natural killer cells, and B survival of skin and organ allografts in experi-
lymphocytes. A prominent cytokine for fur- mental animal models and is active at approx-
ther T cell activation is IL-2 since T cell acti- imately one tenth of the CsA dose required
vation can be induced by IL-2 itself, even in for the same effects. The initial clinical trials
the absence of antigen, in vitro. with FK506 showed a remarkable effect in liv-
er transplantation and in rescuing drug-resist-
ant rejection in organ transplantation. FK506
has recently been approved in the US for the
3.2 Current Knowledge of the indication of transplantation (trade name
Mode of Immunosuppressive PrograF). Like CsA, FK506 interferes with
the process of T cell activation by specifically
Action of CsA inhibiting the transcription of lymphokines
(TOCCIet al., 1989). Rapamycin, a macrolide
In the following section, our current knowl- isolated from cultures of the soil microorgan-
edge as to how CsA interferes with the im- ism Streptomyces hygroscopicus originates
mune response at the molecular level is re- from Easter Island. Although the immuno-
viewed with some reference to two other im- suppressive properties of rapamycin were rec-
munosuppressive drugs of microbial origin, ognized early (MARTEL et al., 1977) they
FK506 and rapamycin. Like CsA, FK506 and have been intensively investigated only re-
rapamycin were initially discovered and char- cently, primarily after the discovery of FK506
Cyclosporin FK506 Rapamycin
Immunophilin: Cyclophilin FKBP FKBP
Potential Calcineurin Calcineurin ?

effector (+ Calmodulin, Ca2+) (+ Calmodulin, Ca2+)
Inhibition
of cell cycle:
Go -..
GI
Fig. 6. Dual domain concept for the immunosuppressants of microbial origin CsA, FK506,and rapamycin
(adapted from BAUMANN and BOREL,1992).
and because of the striking structural similari- and high conservation of the different cyclo-
ty of the two compounds. Rapamycin is effec- philins across species suggests an important
tive in many experimental transplantation role for the normal cell function. All the
models. A single dose of rapamycin leads to members of the cyclophilin family show en-
almost indefinite survival of cardiac allografts zyme activity as peptidyl-prolyl cis-trans iso-
in the rat. It also showed activity in several merases (“rotamases”), a property which en-
murine autoimmune disease models (MOR- ables them to accelerate the cis-trans isomer-
RIS, 1993). Rapamycin inhibits T cell activa- ization of peptide bonds involving a prolyl re-
tion at concentrations comparable to those of sidue. Interconversion of cis and trans con-
FK506 but with a completely different mecha- formers at peptide bonds to proline has been
nism (DUMONTet al., 1990a, b). recognized as a rate determining step of pro-
tein folding in vitro. Therefore, it is not un-
likely that the rotamase activity of cyclophil-
3.2.1 Cyclosporin Receptors ins might also facilitate protein folding in
vivo. This activity of cyclophilins is potently
A specific, saturable and reversible binding inhibited by CsA (FISCHERet al., 1989; KERN
of CsA on mononuclear blood leukocytes was et al., 1993) and some of its analogs, depend-
shown early (RYFFELet al., 1982). Although ing on the respective binding affinity of the
membrane binding specific for CsA was re- compounds. The 40 kDa cyclophilin-like pro-
cently reported (CACALANOet al., 1992) no tein has lower affinity to CsA and its rotam-
membrane-bound receptors have been iso- ase activity is also less sensitive to inhibition
lated to this point. On the other hand, accu- by this drug. FK506 binds to a separate group
mulation of CsA within the cell suggested the of ubiquitous immunophilins termed FK-
existence of an intracellular receptor. Indeed, binding proteins, FKBPs (SIEKIERKA et al.,
CsA, FK506, and rapamycin have been shown 1989b). Both a cytosolic form, FKBP12, and a
to bind to cytosolic proteins termed immuno- membrane-associated form, FKBP13, sharing
philins (SCHREIBER,1991). CsA predomi- the same binding and enzymatic sites were re-
nantly binds to cyclophilins (CUP), a family ported (HAYANOet al., 1991; JINet al., 1991).
of highly conserved proteins comprising both A higher molecular mass protein, FKBP59,
ubiquitous and tissue-specific proteins was shown to be related to and associated
(SCHREIBERand CRABTREE,1992). Cyclo- with heat shock proteins and with the corti-
philin A, the first protein identified, is a ubi- costeroid receptor (TAIet al., 1992). Striking-
quitous cytosolic 18 kDa protein (HAND- ly, the FKBPs also exhibit peptidyl-prolyl cis-
SCHUMACHER et al., 1984). Cyclophilins B, c, trans isomerase activity which is blocked
and D, all with a molecular mass of about when FK506 is bound (SIEKIERKAet al.,
22kDa were reported later (HASELet al., 1989a). Neither CsA nor FK506 appear to
1991; CARONIet al., 1991; PRICEet al., 1991; cross-react for the binding to their respective
BERGSMAet al., 1991). Cyclophilin B con- immunophilins. Since both drugs block the in-
tains an endoplasmic reticulum retention sig- duction of lymphokine gene transcription at
nal and is located in calcium-containing intra- the early stage of antigen-induced helper T
cellular vesicles. Cyclophilin C seems to have cell activation, it has been postulated that T
a more restricted tissue distribution with low cell activation requires the separate activity of
expression in lymphoid tissue. Murine cyclo- both immunophilins. In addition, the rotam-
philin C was reported to be highly expressed ase activity and the resulting protein folding
in kidney and was thus suggested to be in- was thought to be required for nuclear trans-
volved in CsA nephrotoxicity (FRIEDMAN port of some factors involved in signal trans-
and WEISSMAN, 1991). However, this was not duction. However, there seems to be no cor-
observed with human cyclophilin C relation between inhibition of rotamase activ-
(SCHNEIDERet al., 1994). In addition, a ity and inhibition of T cell activation, as de-
40 kDa protein binding to CsA was also monstrated with the antagonists of FK506
shown to share homology with cyclophilin A and CsA, rapamycin and MeVa14-Cs or SDZ
(KIEFERet al., 1992). The relative abundance NIM 811 (ZENKEet al., 1993).
It has been shown by competition experi- domains: an immunophilin(FKBP)-binding

ments that FK506 and rapamycin bind to a domain which is shared between FK506 and
common intracellular receptor since the two rapamycin and an effector domain which is
drugs act as reciprocal antagonists (DUMONT specific for each drug (BIERERet al., 1991)
et al., 1990a). Indeed, rapamycin binds to all and accounts for their different activities. The
the FKBPs described above. In contrast, same dual domain concept has also been de-
FKBP25 which shares homology with monstrated for CsA (Fig. 6). The binding site
FKBP12 in its C-terminal region but differs in for cyclophilin was first mapped by immuno-
the N-terminal part selectively binds rapamy- chemical methods (QUESNIAUX et al., 1987)
cin (GALATet al., 1992). Rapamycin inhibits and later confirmed by NMR analysis (WE-
the rotamase activity of the FKBPs but does BER et al., 1991) and X-ray crystallography
not inhibit lymphokine transcription (Du- (PFLUEGLet al., 1993; MIKOLet al., 1993; see
MONT et al., 1990b). This clearly demon- also Sect. 5.5). Cyclophilin A was shown to
strates that the inhibition of the rotamase ac- bind to residues 1, 2, 9, 10, and 11 of CsA,
tivity of FKBP is insufficient per se to mediate residues which were recognized early on to be
the biological effect of FK506. essential for CsA immunosuppressive activ-
The relevance of the different isoforms of ity. The binding of CsA analogs to cyclophilin
cyclophilins or FKBPs for inhibiting T cell ac- A also showed some correlation with their
tivation was addressed in two different ways. immunosuppressive activity (HANDSCHU-
(1) Overexpression of CYP-A or CYP-B but MACHER et al., 1984; QUESNIAUX et al., 1988;
not of CYP-C increased T cell sensitivity to DURE-I-I-Eet al., 1988). To define the effector
CsA. Similarly, overexpression of FKBP12 site of the cyclosporin A molecule several
but not of FKBP13 or FKBP25 increased T cell-permeable, cyclophilin-binding, but non-
cell sensitivity to FK506 (BRAMet al., 1993). immunosuppressive cyclosporin derivatives
Subcellular localization of CYP-B and CYP-C were selected. These analogs that are modi-
was also essential for their activity (BRAMet fied in the “effector site” of CsA, like
al., 1993). (2) By studying the binding of hu- MeVa14-Cs or SDZ NIM 811, antagonize the
man cyclophilins A, B, and C to a series of immunosuppressive effect of CsA on T cells
cyclosporin derivatives it could be deter- (ZENKEet al., 1993). They effectively inhibit
mined that the three cyclophilins recognize the cis-trans isomerase activity of cyclophilin
very similar sites on the CsA molecule. How- providing again compelling evidence that in-
ever, CYP-C can accommodate larger resi- hibition of rotamase activity is insufficient for
dues at position 2 of the cyclosporin molecule explaining CsA immunosuppression (SIGAL
than CYP-A (SCHNEIDERet al., 1994). et al., 1991; BAUMANN et al., 1992). However,
Therefore, a series of cyclosporin derivatives these results do not suggest that rotamase ac-
modified at position 2 could be selected that tivity in general is irrelevant for the cell. Al-
bind CYP-C up to tenfold better than CYP-A though the biological role of the two highly
in comparison with CsA. Within this series of conserved families of rotamases, cyclophilins
derivatives, the immunosuppressive activity and FKBPS, is poorly understood at this time,
measured by inhibition of IL-2 gene tran- there are a few examples for their role in pro-
scription correlated better with the binding to tein transport. For example, the product of
CYP-A than to CYP-C (QUESNIAUX et al., the Drosophila Nina A gene, a membrane-
unpublished observations). Both approaches bound isomerase homologous to CYP-A, se-
suggest that CYP-C might be less essential lectively transports rhodopsins in photorecep-
than CYP-A for T cell sensitivity to CsA. tor cells (STAMMESet al., 1992). Similarly,
preliminary evidence suggests that cellular cy-
clophilin appears to be required for nuclear
3.2.2 Active Sites on the Drugs transport of DNA transcripts of HIV-1 (Ro-
SENWIRTH et d., 1994; See Sect. 3.3.2).
The current model of the mode of action of
immunosuppressant macrolides suggests that
FK506 and rapamycin act via two regulatory
3.2.3 Target Proteins of the The pharmacological relevance of the inhi-

bition of calcineurin phosphatase activity by
Drug-Immunophilin Complexes CsA or FK506 for immunosuppression has
The structure-activity relationship of a been substantiated by the good correlation
large number of CsA or FK506 derivatives between calcineurin activity and IL-2 produc-
showed that the binding of CsA or FK506 to tion in T cells which is dose-dependently inhi-
their respective immunophilins is necessary bited by CsA or FK506 (FRUMANet al.,
but per se not sufficient to inhibit T cell acti- 1992b). In addition, overexpression of the cal-
vation and thus raised the question of the mo- cineurin catalytic subunit in the Jurkat T cell
lecular target of the drug-immunophilin com- line rendered the transfected cells more re-
plexes, CsAlCYP and FK506/FKBP. Indeed, sistant to the immunosuppressive effects of
CsA and FK506, when attached to their re- CsA and FK506 (O’KEEFEet al., 1992; CLIP-
spective immunophilins could be part of a STONE and CRABTREE, 1992). These results
macromolecular complex and act on different clearly showed that calcineurin is a target for
primary targets or on different sites of a com- drug-immunophilin complexes in vivo and
mon primary target. Such a target would most suggested a physiological role for calcineurin
probably be a component of the signal trans- in T cell activation. Since calcineurin is pres-
duction pathway the activation of which ulti- ent in small amounts in T cells, it might be a
mately leads to lymphokine transcription. Re- rate determining enzyme in the T cell activa-
cent biochemical experiments support this tion pathway. Such a situation would provide
model. The complexes of FK506/FKBP12 or a rationale for the profound effects of CsA
CsA/CYP-A but not of rapamycin/FKBP12 and FK506 on T cell activation.
were found to interact with calcineurin, a cal- A 77 kDa cyclophilin C-associated protein
ciumkalmodulin-dependent protein phos- which binds CYP-C with high affinity in the
phatase (LIUet al., 1991). Calcineurin is com- absence but not in the presence of CsA was
posed of two subunits, the catalytic, calmodu- described (FRIEDMAN et al., 1993). The role
lin binding 61 kDa A subunit and the regula- of this protein is still unknown. A cellular
tory, calcium binding 19 kDa B subunit. The function for FKBP12 has been reported re-
drug-immunophilin complexes bind to and cently, following the observation that
modulate the serinehhreonine phosphatase FKBP12 co-purifies with the ryanodine recep-
activity of calcineurin in vitro. However, calci- tor (BRILLANTESet al., 1994). Four ryano-
neurin does not bind uncomplexed CsA, dine receptors associate to form intracellular
FK506, cyclophilin, or FKBP. In T cell lysates Ca2+ release channels of the sarcoplasmic
containing natural immunophilins both CsA and endoplasmic reticula. FKBP12 was shown
and FK506 inhibit calcineurin activity at con- to stabilize channel gating by improving con-
centrations which effectively block IL-2 pro- ductance increasing mean open time. This ef-
duction in activated T cells (FRUMANet al., fect was reversed by both FK506 and rapamy-
1992a). Rapamycin has no effect on calcineu- cin suggesting that blocking the cis-trans iso-
rin activity. Furthermore, excess concentra- merase catalytic site is sufficient for inhibition
tions of rapamycin compete the effects of by the drugs.
FK506, apparently by displacing FK506 from
FKBP. Cyclophilins A, B, C and CYP-40,
when complexed with CsA, bind calcineurin. 3.2.4 Downstream Effects in the
However, the cyclophilin B complex is 2-5-
fold more potent than that of cyclophilin A to Nucleus
inhibit calcineurin phosphatase activity
(SWANSONet al., 1992). Of the FKBP com- Both CsA and FK506 act specifically on ac-
plexes only FK5061FKBP12 seems to inhibit tivation pathways mediated by the T cell re-
calcineurin. Interestingly, CsA/CYP-A and ceptor that induce an increase in intracellular
FK506/FKBP12 compete for binding to calci- Ca2+ concentration. The actual molecular
neurin despite absence of obvious structural target of the Ca2+-dependent serinekhreo-
similarities. nine phosphatase calcineurin is not known.
Highly attractive candidates include cytosolic intracellular processes that regulate prolifera-
components of the transcription complex that tion (PRICEet al., 1992) and has been identi-
need to be assembled in the nucleus and/or fied recently by SCHREIBER and collaborators
activated for expression of early T cell activa- (BROWNet al., 1994). Previously, it had been
tion genes. Calcineurin may modify the phos- suspected that it might be a proximate up-
phorylation of downstream components such stream activator of the p70 S6 protein kinase
as the antigen-inducible transcription factors such as an activating p70 S6 kinase-kinase or
NFKB (BAUMANN et al., 1991) and NF-AT a regulator of such an enzyme.
(EMMEL et al., 1989) which are essential for
IL-2 transcription. Both NFKB and NF-AT
are present as precursors in the cytoplasm.
They require protein modification such as 3.3 Molecular Evidence for other
phosphorylation or dephosphorylation to be Biological Activities of
translocated into the nucleus where they bind Cyclosporins
to the DNA and participate in the formation
of a functional transcription complex leading
to the transcription of the IL-2 gene. It has 3.3.1 Antiinflammatory Effects
been shown recently that CsA and FK506 in-
hibit dephosphorylation of NF-AT (MCCAFF- Crossreactivity of antiserum against human
REY et al., 1993). CsA and FK506 bound to IL-8 with cyclophilins led to the finding that
their respective immunophilin most probably IL-8 binds CsA but not some of its non-im-
interfere at this level by inhibiting the phos- munosuppressive analogs. Although IL-8
phatase activity of calcineurin. This might di- bears some sequence similarities to cyclophil-
rectly or indirectly alter the protein modifica- in, it has no rotamase activity (BANGet al.,
tion of the transcription factors and conse- 1993) and the relevance of these findings still
quently block lymphokine transcription needs further clarification.
(SCHREIBERand CRABTREE,1992). Rapa- Cyclophilin A shows proinflammatory ac-
mycin is not active on this pathway. tivity and leukocyte chemotactic activity
As implicated by the mutually antagonistic which can be inhibited by CsA but not by a
activity of rapamycin and FK506 the mecha- non-immunosuppressive analog (Xu et al.,
nism of rapamycin action in lymphoid cells is 1992; SHERRYet al., 1992). CYP-A is se-
likely to depend on immunophilin/drug com- creted by macrophages in response to endo-
plex formation. Rapamycin does not affect toxin and it was proposed that CYP-A may
the transcription of genes involved in early function as a cytokine (SHERRYet al., 1992).
activation of T cells, such as IL-2, but appears FKBP was also shown to display some leuko-
instead to block later events leading to T cell cyte chemotactic activity which is inhibited by
activation, such as the signal transduction FK506 (LEIVAand LY-ITLE, 1992). Cyclophil-
pathway driven by the interaction of IL-2 in 40 was shown to share homology with P59,
with the high-affinity IL-2 receptor. More a member of the steroid receptor complex
generally, rapamycin was shown to inhibit (KIEFFERet al., 1993).
growth factor receptor-mediated activation
signals in a number of different cells as well
as in vivo after myelodepression (QUES- 3.3.2 Cyclosporins as Anti-HIV
NIAUX et al., 1994). The rapamycin/FKBP
complex and the putative associated effector Agents
molecules interfere with the activation of the
p70 S6 protein kinase in response to growth The clinical potential of the anti-HIV activ-
factor stimulation (CHUNGet al., 1992). This ity of cyclosporins, in particular of SDZ NIM
is a key regulatory step in the cell cycle pro- 811, was briefly discussed in Sect. 2.4 without
gression from G1 to S phase. The direct target addressing possible mechanisms of this effect
of rapamycin appears to be a crucial element which is clearly different from that of all oth-
linking growth factor receptors to subsequent er anti-HIV agents described to date. In cell-
4 Chemistry 551
free assay SDZ NIM 811 does not inhibit re- vivo was reported as early as 1981 by THOM-
verse transcriptase, protease, or integrase. MEN-SCOTT. Interestingly, treatment was
There is some evidence that the cyclosporin- most effective when started at a time when
sensitive step in HIV replication is an event parasitemia was already established suggest-
after virus penetration, possibly nuclear trans- ing a direct toxic effect on the parasite. Re-
location of viral DNA (ROSENWIRTH et al., cently, with the availability of cyclosporin
1994). It has been reported recently that some analogs binding to cyclophilins but devoid of
cyclophilins bind to the HIV-1 protein p55g”g immunosuppressive activity (such as SDZ
as well as the capsid protein p24 (LUBANet NIM 811) as well as of cyclosporins without
al., 1993). In a cell-free system, this interac- any measurable affinity to cyclophilins, it has
tion is disrupted by both cyclosporin A and become possible to discern between cyclospo-
SDZ NIM 811. Furthermore, inhibition of the rin effects mediated by inhibition of calcineu-
gag-cyclophilin complex formation by cyclo- rin by the cyclosporin-cyclophilin complex
sporins correlates with the cyclophilin binding (e.g., immunosuppression), mediated by cy-
capacity of the compounds but not with their clophilin binding only (such as, e.g., the anti-
immunosuppressive potential. The same cor- HIV activity of SDZ NIM 811), from effects
relation was found for the anti-HIV activity that are independent of cyclophilin (such as,
of these derivatives (ROSENWIRTHet al., e.g., the MDR activity of SDZ PSC 833 which
1994). Whether inhibition of gag-cyclophilin does not bind to cyclophilin). With these rea-
complex formation by cyclosporins explains gents, a re-investigation of antimalarial activi-
the antiviral effect remains to be elucidated. ties of cyclosporins was performed (BELL et
al., 1994). While CsA-sensitive peptidyl-pro-
lyl cis-trans isomerase activity could be de-
3.3.3 Cyclosporins as Drug tected in extracts of P. falciparum the highest
Resistance Modifiers activity against P. falciparum was exhibited
by SDZ PSC 833 suggesting that inhibition of
The clinical potential of the non-immuno- rotamase activity may not be the lethal target
suppressive cyclosporin SDZ PSC 833 to sen- of cyclosporins in P, falciparum.
sitize multidrug resistant tumor cells to che-
motherapeutic agents was described in Sect.
2.3. This resistance phenomenon (MDR) is
thought to be associated with the p-glycopro-
tein transporter. Direct binding of CsA to p-
glycoprotein could be demonstrated by using
a photoaffinity-labeled cyclosporin derivative
4 Chemistry
(FOXWELLet al., 1989). Interestingly, while
the immunosuppressive activity of cyclospo- 4.1 Structural Aspects
rins is mediated by complex formation with
cyclophilin (see Sect. 3.2.3) SDZ PSC 833 has Cyclosporins are composed of 11 aliphatic
no measurable affinity to cyclophilin. Similar- lipophilic amino acids of which four are leu-
ly, other cyclosporins not binding to cyclo- cines and three do not occur in mammalian
philins (e.g., cyclosporin H) also bind to p- proteins. In CsA these are: (4R)-4-[(E)-2-bu-
glycoprotein. Resistance modifier activity of tenyl]-4-methyl-~-threonine (Bmt) in position
cyclosporins appears thus to be a cyclophilin- 1, (L)-cY-amino butyric acid in position 2, and
independent effect. (D)-alanine in position 8. Of the 11 peptide
bonds 7 are N-methylated. This feature has
several important implications: Firstly, the N-
3.3.4 Antimalarial Activity of methylated peptide bonds and the cyclic
Cyclosporins structure of the molecule render cyclosporins
stable towards mammalian digestive and sys-
Activity of cyclosporin A against Plasmo- temic proteases (SandimmunB is metabolized
dium spp. in vitro and in rodent models in extensively in animals and man, however, ex-
clusively by cytochrome P450-mediated oxi-

dative transformations). Therefore, cyclospo-
rins are not only well absorbed when given by
the oral route, high and long lasting plasma
levels are commonly obtained. These proper-
ties are essential prerequisites for a successful
drug. A second consequence of the N-meth-
ylation pattern is a certain conformational ri-
gidity in non-polar environment characterized
by intramolecular hydrogen bonds and the li-
pophilic side chains being oriented towards
the (hydrophobic) environment.
The high number of leucines in the cyclo-
sporins is striking, especially in view of the
well recognized role of leucines in protein-
protein interactions. Their role for biological
activity will be discussed below.
The structures of cyclosporins in apolar as
well as polar solution have been determined
by NMR spectrospic methods as well as by X-
ray crystallography. This work is extensively
discussed in Sect. 5. In non-polar solvents, the
conformation of cyclosporins is very similar
to that found in the crystalline state. Cyclo-
sporin A in polar environment exhibits a dif-
ferent conformation from that found in apo-
Fig. 7. NMR-structure of [D-MeSer3-~-Ser-(O-
lar systems (KO and DALVIT,1992). Cyclo- Gly)R]-cyclosporin,not bound to a receptor pro-
sporin A is not soluble enough in aqueous so- tein, in DMSO (identical to the one in water).
lution to allow a full 3D-structure determina-
tion by NMR but the ‘H-NMR spectra in wa-
ter indicate the presence of a family of con-
formations. A single predominant conforma- mode of binding of cyclosporins to cyclophil-
tion of cyclosporin is observed only in com- in would thus not seem necessary.
plex with certain metal ions (KOECK et al., To date, some 30 cyclosporins have been
1992) or in complex with cyclophilin. There isolated from natural sources. Their struc-
exist derivatives, however, which are more tures are shown in Tab. 1.
water-soluble than CSA and some modifica-
tions (e.g., a methyl group in position 3) of
cyclosporin seem to stabilize one conforma- 4.2 Chemical Synthesis and
tion in water. Recently, the 3D-structure of
such a derivative, namely of [D-MeSer3-D- Production
Ser-(O-Gly)X]-cyclosporin,in DMSO and in
water has been determined by NMR (WEN- The method of choice for the production of
GER et al., 1994). Fig. 7 shows the conforma- cyclosporins on a large scale is fermentation.
tion of this derivative in DMSO which is iden- Invariably, complex mixtures containing the
tical to the one in water. Strikingly, this struc- desired cyclosporin along with a plethora of
ture is very similar to the one found for CsA minor metabolites are obtained. To achieve
bound to cyclophilin (Fig. 8) thus giving evi- sufficient purity of the final product repeated
dence for the first time that CsA adopts the chromatography operations are necessary.
cyclophilin-bound conformation (among For the production of cyclosporins on a large
many other conformations) in aqueous solu- scale (tons), substantial investments in large-
tion. An “induced-fit’’ hypothesis for the scale chromatography facilities are therefore
4 Chemistry 553
Tab. 1. Cyclosporins Isolated from Fermentation Brothsa
Name Amino Acid in Position
1 2 3 4 5 6 7 8 9 10 11
CsA Me Abu Sar Me Val Me Ala (D)-Aia Me Me Me
Bmt Leu Leu Leu Leu Val
CsB Ala
csc Thr
CsD Val
CsE Val
Deoxy-MeBmt
CsF
Abu2
CsG Nva
(D)-Me-
CsH Val
CSI Val Leu
Deoxy-MeBmt
CsK Val2
CSL Bmt
CsM Nva Nva
CsN Nva Leu
Me
cso Nva
Leu
CSP Bmt Thr
CsQ Val
CsR Leu Leu
css Thr Val
CsT Leu
csu Leu
csv Abu
csw Thr Val
csx Nva Leu
CSY Nva Leu
Me
csz
Aocb
Cs26 Nva Leu
Cs27 Bmt Val
Me
Cs28 Leu
Me
Cs29 Ile
Me Val
Cs30
Leu
Cs31 Ile
Cs32 GIY (D)-Ser
FR 901495 Thr Leu Leu
a Only residues different from those in Cyclosporin A are given.
MeAoc = N-Methyl-(L)-amino octanoic acid.
necessary. For structure-activity studies, a

large number of analogs has been prepared
by total synthesis according to the method de-
veloped by Wenger (WENGER,1983) or by
semisynthetic transformations of natural cy-
closporins. In addition, a solid phase-based
methodology has been developed in the pre-
clinical research laboratories of Sandoz Phar-
ma AG allowing the preparation of almost
any analog on a small scale in a relatively
short time (BOBE,unpublished work).
Semisynthesis of cyclosporins involves sev-
eral main approaches: Modification of the
MeBmt aliphatic side chain, chemical trans-
formations of the cyclic peptide backbone,
and modifications based on selective ring
opening reactions.
4.2.1 MeBmt Transformations and

Variations
Fig. 8. X-ray structure of CsA when bound to CYP- The olefinic double bond of MeBmt offers
A, crystallized from an aqueous solution. There is a number of possibilities of selective chemical
only one intramolecular hydrogen bond, namely transformations. In the preclinical research
between the hydroxyl group of MeBmt' and laboratories of Sandoz Pharma AG many
Leu4(CO). All amide bonds are in the trans confor- analogs have been obtained by ozonolysis
mation.
1. 1 ii
Fig. 9. Some chemical trans-

formations of the aliphatic
side chain of MeBmt.
R1 =protective group (most
often acetyl); reagents:
i =ozone,
ii = R2-CH=P(~henyl)~;
k = N-bromo-succinimide.
For references, see text.
o 0 0 CH, 0’ 0 0 CY 0
Me
Fig. 13. Treatment of cyclosporin with phosphorous pentasulfide or Lawesson’s reagent forms mixtures of
mono- and dithio amides. The oxygen atoms replaced by sulfur atoms during these reactions are indicated
with an asterisk. Cleavage of the thioamides can be achieved by alkylation at the sulfur atom, followed by
acid hydrolysis.
As has been outlined in Sect. 3.2, the im- cyclophilin binding and in terms of mediating
munosuppressive activity of cyclosporins is the calcineurin interaction of the complex.
based on an unusually complex mechanism Much insight into the nature of this interac-
which in a first step requires binding of the tion was gained from analyzing the crystal
drug to a receptor of the family of cyclophil- structures of many cyclosporin-cyclophilin
ins and in a second step binding of this com- complexes (Sects. 55-58). As indicated
plex to protein phosphatase 2B (calcineurin), above, both CsA and SDZ 214-103 contain a
thereby inhibiting its catalytic activity. This strikingly high number of leucines. In the cy-
activity is crucially dependent on cyclosporin; clophilin complexes of these molecules the
cyclophilin without any bound drug does not side chains of the leucines in position 4,6, and
interact with calcineurin. There is also evi- 10 are exposed next to each other on the sur-
dence that binding of the cyclosporin-cyclo- face; they form a “leucine cluster” on the pro-
philin complex to calcineurin involves both tein surface. It has long been recognized that
residues of cyclosporin as well as part of the leucine side chains in a specific three-dimen-
cyclophilin molecule. This means that the sional disposition on the surface of a protein
molecule causing immunosuppression is the play a crucial role in mediating protein-pro-
entire cyclosporin-cyclophilin complex. Con- tein interactions, a prominent example being
sequently, cyclosporin structure-activity rela- the dimerization of transcription factors
tionships must be analyzed both in terms of through “leucine zipper” domains (LAND-
IQ
HN J O H HN
I
CH3 O Y
Fig. 14. Cyclosporins incor-

porating rigid dipeptide frag-
ments at the the p t u r n re-
gion in the absence of cyclo-
philin. For references, see
text.
H I
Fig. 15. Design and synthesis of a "calcineurin-bridging" ligand (ALBERGand SCHREIBER, 1993). The
bicyclic amino acid is incorporated into the cyclosporin molecule in place of alanine' and (D)-alaninex.
SCHULTZ et al., 1989). The presence of a leu- chains have crucial functions for the calcineu-
cine cluster on the surface of the cyclosporin- rin interaction; they can be substituted for
cyclophilin complex prompted an analysis of valine or alanine without much affecting the
the role of these leucines by substituting each affinity of the compounds for cyclophilin.
of them for valine or alanine. A shortened However, the affinity of their respective com-
version of the outcome of this analysis is plexes to calcineurin has been lost as evi-
shown in Tab. 2. denced by the lack of immunosuppressive
From these results it is evident that in posi- properties of these derivatives. In line with
tions 4 and 6 the cyclosporin leucine side this interpretation none of these compounds
This protein has 165 amino acids and is found trans isomerase activity is, however, insuffi-
in the cytoplasm. Cyclophilin B has 208 resi- cient to induce immunosuppression (SIGALet
dues and is found in the endoplasmic reticu- al., 1991).
lum. Cyclophilin C has 194 residues and is There is a growing body of available 3D-
thought to have some tissue specificity for structural information on immunophilins and
kidney, at least in the mouse (FRIEDMAN and their ligands. This work is directed at trying
WEISSMAN,1991). There is no sequence ho- to understand both the biological and enzy-
mology between cyclophilins and FKBPs and matic activity. Protein X-ray crystal structures
no obvious three-dimensional structural simi- of cyclophilin A complexed with a tetrapep-
larity. tide substrate (KALLENet al., 1991), a dipep-
Despite the chemical dissimilarity of the li- tide substrate (KE et al., 1993a), without sub-
gands and the structural differences between strate (KE et al., 1991), with CsA (PFLUEGL
the proteins there is an intriguing overlap of et al., 1993; MIKOLet al., 1993), and with a
biochemical and biological activity between cyclosporin derivative modified in position 1
the two immunophilin families (see Sect. 3.2). (KE et al., 1994; MIKOLet al., 1994b) have
The common target of their complexes with also been published as well as the X-ray struc-
the cognate drugs, calcineurin, is a heterodim- ture of a Fab-CsA complex (ALTSCHUHet
er composed of subunit A (61 kDa) and sub- al., 1992). The 3D-structure of the CsA-CYP-
unit B (19 kDa) which has been shown to A complex has also been determined by
have affinity only for the immunophilin-drug NMR techniques (THERIAULT et al., 1993),
complexes but not for the drug alone or the as well as the 3D-structure of a water soluble
immunophilin alone (Lru et al., 1991). Fur- cyclosporin derivative (WENGER et al.,
thermore, the binding of the FKBP12-FK506 1994).
complex to calcineurin competes with the
binding of the CsA-CYP-A complex. Natural
ligands for cyclophilins or FKBPs which could
regulate phosphatase activity have not yet 5.2 Structural Investigations of
been discovered. Another puzzling coinci-
dence between the two immunophilin fami-
Uncomplexed Cyclosporins
lies is their shared peptidyl-prolyl isomerase
(PPIase) activity which led to the re-discovery 5.2.1 The 3D-Structure of
of cyclophilin (TAKAHASHIet al., 1989; Uncomplexed CsA in Apolar
FISCHERet al., 1989) as an enzyme catalyzing
protein folding in vitro. For proteins to adopt Environment
the correctly folded conformation the X-Pro
amide bonds must be in the correct cis or Prior to the studies of the conformation of
trans conformation. Using model peptide sub- CsA bound to CYP-A the 3D-structure of
strates it has been found that immunophilins free CsA had been determined. Because of
lower the energy of activation for isomeriza- the very low solubility of CsA in aqueous so-
tion of this amide bond by over 6 kcal .mol - lution, crystallization and NMR studies were
from about 20 kcal.mol-' down to done using apolar solvents. The NMR struc-
14 kcalemol-' (PARK et al., 1992). Both ture in chloroform at 20°C (Fig. 17) is virtual-
FKBP and cyclophilin have also been shown ly identical to the X-ray crystal structure
to accelerate the refolding of a number of (LOOSLIet al., 1985). The backbone forms a
proteins in vitro, presumably by catalyzing twisted psheet that involves residues 1, 2, 5,
the rate determining step of proline isomer- 6, 7, and 11 and a type 11' turn at Sar3 and
ization (SCHOENBRUNNER et al., 1991; MeLeu4. There is a cis-peptide bond between
FRANSSONet al., 1992). Specific cellular tar- MeLeu' and MeLeu". All the four amide
gets of the PPIases are not yet known but it protons are involved in hydrogen bonds
has been suggested that they play a role in the (three transannular ones):
folding of newly synthesized proteins (GETH- Abu2(NH)- Va15(CO),
ING and SAMBROOK, 1992). Blocking the cis- Va15(NH)-Abu2(CO), and
Fig. 18. X-ray structure of SDZ 214-103 (not bound

to cyclophilin), crystallized from diethyl ether.
Fig. 17. Conformation of cyclosporin A in chloro- Thr2(CO)-Leu5(NH),

form. Ala7(NH)-Leu"(CO), and
Ala7(CO) -Leu "(NH).
The hydroxyl group of MeBmt' is not in-
Ala7(NH)-MeVal"(CO), and one addition- volved in an intramolecular hydrogen bond,
a1 hydrogen bond but makes a packing interaction in the crystal
D-AlaX(NH)-MeLeu6(CO). with Sar3(CO) of a neighboring molecule.
5.2.2 The 3D-Structure of the 5.2.3 The 3D-Structure of

Uncomplexed Peptolide SDZ Uncomplexed CsA in Polar
214-103 in Apolar Environment Environment
The peptolide SDZ 214-103 differs chemi- CsA is not soluble enough in aqueous solu-
cally from CsA in the following way: Thr' in- tion to allow a full 3D-structure determina-
stead of Abu2, Leu' instead of Val', D-Hiv' tion by NMR but the 'H-NMR spectra of
instead of D-Ala*, and Leu" instead of CsA in water indicate the presence of a fami-
MeLeu". The X-ray structure (Fig. 18) of the ly of conformations (CsA adopts a single, pre-
uncomplexed peptolide (crystallized from an dominant conformation only if complexed
organic solvent) is very different from the one with certain metal cations (KOECK et al.,
of uncomplexed CsA (Fig. 30) or CsA bound 1992) or when bound to cyclophilin). There
to CYP-A (Fig. 8): There is a cis amide bond exist cyclosporin derivatives, however, that
between residues 3 and 4 and a Pbend in- are more water soluble than CsA. Further-
volving residues 7, 8, 9, 10. The following in- more, introduction of a substituent in position
tramolecular hydrogen bonds are formed: 3 seems to stabilize one conformation in wa-
ter. Recently, the 3D-structure of such a deri-

vative, namely of ~-MeSer~-~-Ser-(O-Gly)*-
cyclosporin in DMSO and in water, has been
determined by NMR (WENGERet al., 1994).
The derivatization in position 8 makes the
derivative more water-soluble and the substi-
tution of D-MeSer for Sar in position 3 stabil-
izes a single conformation. Fig. 7 shows the
conformation of this derivative in DMSO
which is identical to the one in water. Strik-
ingly, this structure is very similar to the one
found for cyclosporins bound to cyclophilin
(Fig. 8) thus giving evidence for the first time
that cyclosporins preadopt the cyclophilin-
bound conformation (among many other con-
formations) in aqueous solution. An “in- Fig. 19. The architecture of CYP-A consists of an
duced-fit’’ hypothesis for the mode of binding eight-stranded antiparallel /?-barrel the ends of
of cyclosporins to cyclophilin is thus not re- which are closed off by two a-helices. The tetrapep-
quired. tide (a substrate for the PPIase activity of CYP-A)
binds to the outside of the P-barrel. Selected Ca-
positions of CYP-A are indicated by their sequence
numbers.
5.3 The 3D-Structure of
Cyclophilin A Complexed with a
Tetrapeptide
At Sandoz Pharma AG the first three-di-
mensional X-ray structure of CYP-A with a
model substrate (N-acetyl-Ala-Ala-Pro-Ala-
amidomethyl coumarin) bound to its active
site (two complexes per asymmetric unit) was
elucidated (KALLENet al., 1991; KALLENand
WALKINSHAW,1992). The X-ray structure of
unliganded CYP-A (1 molecule per asymmet-
ric unit) was determined by KE (KE et al.,
1991; KE, 1992). The same group also deter-
mined the structure of a CYP-A complex with
the dipeptide substrate Ala-Pro (KE et al.,
1993a). Fig. 20. The complex CYP-A-tetrapeptide rotated
The structure of CYP-A was determined in by 90”with respect to Fig. 19 around a vertical axis
a collaborative effort using NMR and X-ray in the picture plane.
methods: The secondary structure of CYP-A
was determined by NMR methods while
simultaneously the tertiary structure was de-
termined by X-ray methods (KALLEN et al., 1
Cyclophilin is an approximate1 spherical
molecule with a radius of ca. 17 (Figs. 19,
20). The main structural feature is the eight-
1991; KE et al., 1991). Via chemical shift
changes of certain residues in CYP-A upon stranded antiparallel P-barrel consisting of
complexation of the protein with CsA the two roughly perpendicular four-stranded @
NMR technique was also able to identify resi- sheets with a +3, -1, -2, +1, -2, -3 to-
dues probably interacting with cyclosporin pology. Both ends of the barrel are closed off
(KALLENet al., 1991). with an a-helix. The cyclophilin Pbarrel is
similar in some respects to the superfamily of
proteins involved in ligand transport, includ-

ing retinol binding protein (RBP), bilin bind-
ing protein, plactoglobulin, fatty acid binding
protein, and P2 myelin (COWANet al., 1990).
Most of these molecules encapsulate their li-
gand in the /%barrel core. In contrast, the bar-
rel core in CYP-A is tightly packed with hy-
drophobic residues and the ligand binding
site is on the outside of the barrel. The topol-
ogy of CYP-A also differs from the simple
[ + l]n-up-and-down fold found in the RBP
+ +
class of proteins or the [ -3, 1, 11-Greek
key topology that is most frequently found in
/%barrel proteins.
5.4 The Peptidyl-Prolyl Isomerase

Active Site of Cyclophilin A I
I
Two X-ray structures of CYP-A-substrate Fig. 21. Details of the interactions of the tetrapep-
complexes have been published a complex of tide with the enzyme active site of CYP-A. Inter-
CYP-A with the tetrapeptide N-acetyl-Ala- molecular hydrogen bonds are formed between the
Ala-Pro-Ala-amidomethylcoumarin (Nac- main chain N and 0 atoms of Ala' (from the pep-
AAPA-amc) (KALLENet al., 1991) and a tide) and the main chain N and 0 atoms of Asn''
(from CYP-A) as well as between the carbonyl oxy-
complex of CYP-A with the dipeptide Ala- gen of Pro3 (from the peptide) and NH1 and NH2
Pro (KE et al., 1993a). In both cases the Ala- of Arg55 (from CYP-A).
Pro amide bond adopts a cis conformation.
In the following, details of the CYP-A te-
trapeptide structure will be given (Fig. 21).
CYP-A residues which have one atom within The guanidinium group of Arg55 forms hy-
4 8, of any atom of the active site-bound Nac- drogen bonds to carbonyl oxygen of Pro3 in
AAPA-amc are: Arg55, Ile5', Phew, Gln63, the peptide. It also makes a hydrogen bond to
Ala"', AsnIo2,Gln"', Phe'l3, Leu'22, His'26, Gln63 which in turn is hydrogen bonded to
and Arg'48. The enzyme active site is a chan- the side chain of Gln"'. The other two direct
nel sitting on top of two antiparallel P-strands substrate-protein hydrogen bonds are in the
(with contacts from residues Phem, form of a short stretch of antiparallel P-sheet
Gln"', and Phe1I3). Two loops protrude out between the main chain N and 0 atoms of
from the surface of the barrel and provide a Ala2 and the main chain N and 0 atoms of
distinctive grooved protein surface. One loop Asnlo2. The formation of a short stretch of
from residues 101 to 110 contains the contact antiparallel sheet is a common feature in
residues Ala"' and AsnIo2. A narrow pass many enzyme inhibitor complexes including
separates this loop from the second which aspartate proteases and serine proteases
comprises residues 69 to 74. Another impor- (BLUNDELLet al., 1987). The side chain of
tant topological feature of the binding site is His'26 is close to the alanyl-prolyl cis amide
the wall composed of residues 118-126 in a bond, however, the X-ray refinement suggests
close to helical conformation. Leu122 and a conformation in which the histidine side
His'26 are in contact with the peptide sub- chain preferentially hydrogen bonds to sol-
strate. The cis-proline of Nac-AAPA-amc sits vent water and protein main chain rather
in a rather deep pocket made principally by than to the substrate. The mechanism of iso-
Leu'22, His'26, Phe1I3, Phew, and Met6'. merase action is not yet clear, however, the
There are three hydrogen bonds formed Ala-Pro cis amide bond is significantly
between the peptide and CYP-A (Fig. 21). twisted out of plane. Until now, four indepen-
dent determinations of the w angle have been
performed that was found to vary between
20” and 45”. This is consistent with a mecha-
nism of catalysis by distortion (HARRISON
and STEIN,1990; PARKet al., 1992; LIU et al.,
1990; ROSENet al., 1990) in which the immu-
nophilin would bind the X-Pro amide bond of
the substrate with a twisted, high-energy con-
formation. The transition state could also be
stabilized by hydrogen bonding to the proline
amide nitrogen (KOFRONet al., 1991) possi- *RG5
bly via a water molecule or His126or Arg55.
Based on his X-ray structure of the CYP- HIS^
A-AP complex (not showing a significantly
distorted amide bond) KE proposed a mecha-
nism in which the transition state is stabilized
via a hydrogen bond of a water molecule to
the carbonyl oxygen of the amide bond (KE
et al., 1993a).
A comparison of the active site of CYP-A fig. 22. Details of the enzyme active site of CYP-A
when complexed with a substrate (,-f. Fig. 21) in the unliganded form. Only selected water mole-
and in the uncomplexed form (Fig. 22) shows cules with B-factors <40 A’ are indicated
(spheres) as well as their hydrogen bonds to atoms
that there is practically no structural change from CYP-A.
for CYP-A. The biggest movement occurs for
the side chain of Arg55 and shows a move-
ment of its guanidyl group by about 2 A. The
only well-ordered water molecule (with a
crystallographic B-factor c 40 A’) that has to
Fig. 23. A close-up view of the CsA-CYP-

A complex which shows the distinction
between residues of CsA making mostly
contacts with CYP-A (“the CYP-A-bind-
ing region”) and residues available for in-
teractions with calcineurin (“the effector
region”).
be displaced upon binding of the Ala-Pro opposite direction to that observed in the te-
moiety of the tetrapeptide is the one hydro- trapeptide-cyclophilin complex. Argss is also
gen-bonded to the main chain 0 and N of involved as a hydrogen bond donor in both
Asn'02. These hydrogen bonds are replaced, structures, to the carbonyl oxygens of
in the complex, by hydrogen bonds between MeLeu" in CsA and Pro3 in the peptide. An
Ala2 of the tetrapeptide and Asn'02 (cf. Fig. important result of the collaborative X-ray,
21). Interestingly, the water molecules hy- NMR, and modelling work on the CsA-CYP-A
drogen bonded to Hiss4 and His'26 are con- complex was to show conclusively that the cy-
served in the two structures. closporin binding site was identical to the
peptidyl-prolyl isomerase active site (cf. Figs.
19 and 24).
5.5 The 3D-Structure of The NMR structure of the full CsA-CYP-
Cyclophilin A Complexed with A complex (THERIAULT et al., 1993) and the
X-ray structure of a decameric CsA-CYP-A
CsA complex have now been solved (PFLUEGLet
al., 1993). Both structures are very similar and
The NMR structure of CsA when bound to broadly confirm the results of the previous
CYP-A had been determined (WEBERet al., docking studies with the following correc-
1991; NERI et al., 1991) before the 3D-struc- tions: There is actually an intramolecular hy-
ture of CYP-A was known and was found to drogen bond between the hydroxyl group of
be substantially different from the conforma- MeBmt and the carbonyl oxygen of MeLeu4
tion of uncomplexed CsA as determined by and an intermolecular hydrogen bond be-
NMR in chloroform or by X-ray crystallogra- tween Abu2(NH) of CsA and Asnlo2(CO)of
phy. The main features of the CYP-A-bound CYP-A, both not found in the docking model.
conformation as determined by NMR are that The biological relevance of the decameric
all amide bonds are in the rruns conformation form (with a pentamer of CsA-CYP-A com-
and at first no intramolecular hydrogen bonds plexes per asymmetric unit; Fig. 25) is un-
were found (subsequently, the X-ray struc- clear. Direct interactions between the effector
ture of the CsA-CYP-A complex has shown domain of cyclosporin and calcineurin would
that there is an intramolecular hydrogen bond practically be impossible since about 80% of
between the hydroxyl group of MeBmt' and the CsA surface are buried in the decameric
the carbonyl oxygen of MeLeu4). This con- complex.
trasts with the structure of uncomplexed CsA Subsequently, it was possible to grow crys-
in the crystal (cf. Fig. 30) or in chloroform tals and solve the X-ray structure of a mon-
where the amide bond between MeLeu' and omeric CsA-CYP-A complex at 2.1 8, resolu-
MeLeu" is cis and the four free NH groups tion (MIKOLet al., 1993). In contrast to the
are all involved in intramolecular hydrogen decameric structure there are no intermolec-
bonds (LOOSLIet al., 1985). ular contacts between CsA and neighboring
Using this NMR structure of CsA and 32 CYP-A molecules in the crystal. The monom-
intermolecular NOES (nuclear Overhauser eric CsA-CYP-A complex shows the follow-
effects) between CsA and CYP-A as distance ing features (cf. Fig. 26): The binding pocket
constraints CsA was subsequently docked for CsA is a mainly hydrophobic crevice
into CYP-A (SPITZFADEN et al., 1992). The formed by the following 13 residues of CYP-
refined docked structure of the complex A which are within 4.08, of CsA: Arg",
showed that 13 residues of cyclophilin are in Phem, Met6', Gln63, G ~ Y ' ~Ala"', , Asn'OZ,
contact with CsA. This docked structure Ala103, Glu"', Phe113, TrpI2', Leu'22, and
agreed well with the structure-activity hypo- His'26. The binding site rests on three of the
thesis based on the binding of a series of CsA antiparallel strands of the eight-stranded p
derivatives and highlighted the importance of barrel involving Phe6', Met61, Gln63, Phe1I3,
residues 10, 11, 1, 2, 3 (Fig. 23) for CYP-A Gln"l, and Arg". Other clusters of residues
binding (QUESNIAUX et al., 1987). The direc- (Trp12', Leu'22, His'26), (Ala'", Asn'",
tion of the chain in the docked CsA is in the ) located on three sep-
Ala'03), and ( G ~ Y ' ~are
arate loop regions which protrude some

10 A-15 A from the surface of the barrel
forming a deeply grooved surface (cf. Fig. 24).
CsA docks into CYP-A like a coin going part
way into a slot machine; only one rim of the
circular cyclosporin (formed by residues 9,10,
11, 1, 2, 3) is in contact with CYP-A. The
complementarity of fit is particularly good for
MeVal" which lies in the center of this bind-
ing rim. The valine side chain fits snugly into
the deep proline-binding pocket formed by
Phe6", Met61, Phe1I3, and Leu'22 (Fig. 26).
There are five direct hydrogen bonds be-
tween CsA and CYP-A: MeBmt'(C0)-
G ~ ~ I ~ ~ (not
( N Hunambiguously
) identified in
Fig. 24. The monomeric CsA-CYP-A complex the NMR structure of the complex),
viewed in the same orientation as the CYP-A-te- Abu2(NH)-Asn'02(CO), MeLeu'(C0)-
trapeptide complex in Fig. 19 showing that the cy- Trp12'(NE), M ~ L ~ U " ( C O ) - A ~ ~ ~ ~and(NH~),
closporin binding site is identical to the peptidyl- M ~ L ~ U " ( C O ) - A ~ ~ ~(not
~ ( Nreported
H ~ ) in
prolyl active site (but CsA and the tetrapeptide the NMR structure of the complex). There
bind with opposite polarities from N- to C-termin- are five clearly identified water molecules
us). mediating interactions between CsA and
CYP-A: MeBmt'(CO)-Wat'8-His54(NE2),
MeBmt'(C0) - Wat" - Wat" - Asn7'(CO),
A~u'(CO)-W~~~~-T~~~~(CO), and MeLeu6-
(C0)-Watx7-Watlo7-Args5(NH2).
Fig. 25. The pentamer half of the

decameric CsA-CYP-A complex.
Fig. 26. Details of the interactions between CsA Fig. 27. A superposition (using C, of CYP-A) of
and CYP-A in the monomeric complex. CsA (black) and [MeBm2t]-cyclosporin (grey).
5.6 The X-Ray Structures of 1991). The X-ray structure of CYP-A/

([MeBm2t]-Cs) (MIKOLet al., 1994b) shows
Cyclophilin A Complexed with that a change with respect to the structure of
Cyclosporin Derivatives CsA/CYP-A has occurred for the side chains
of residues 1 and 1 0 The C, of MeLeu” and
Using the crystal form with one CsA-CYP- of MeBm2t move apart by about 0.9 8, to ac-
A complex per asymmetric unit and cross- commodate the presence of the additional
seeding techniques it was possible to obtain methyl group on MeBm2t (cf. Fig. 27). This
crystals and to solve the structures of 15 com- small structural change must be beneficial for
plexes of cyclophilin A and different cyclo- the interaction with calcineurin A/B. On the
sporin derivatives. For all derivatives studied other hand, the affinity for CYP-A has de-
(variations in positions 1, 2, 3,4, 8) the back- creased (as compared to CsA), although all
bone conformation of the Cs-ring is practical- the structurally mediated interactions to
ly unchanged (as well as the interactions with CYP-A are conserved. This finding suggests
CYP-A). There are also practically no that for [MeBm2t]-Cs the equilibrium (at-
changes in the CYP-A-structure. In the fol- tained in aqueous solution for the uncom-
lowing, the results from two complexes will plexed ligand) between “nonbinding” and
be summarized. “binding” conformation is shifted to the
“nonbinding” side, indicating a higher kinetic
barrier for complex formation.
5.6.1 [MeBm,t] -Cyclosporin
5.6.2 MeIle4-Cs (SDZ NIM 811)
[MeBm2t]-Cs differs from CsA in having
two (instead of one) methyl groups on the C, MeIle4-Cs has the same affinity for CYP-A
of residue 1 (AEBIet al., 1990). Its affinity for as CsA but no immunosuppressive activity.
CYP-A is about 1% of that of CsA whereas The X-ray structure (J. KALLENet al., to be
its immunosuppressive activity in v i m is only submitted) shows that the only change with
reduced to 30% of that of CsA (SEAL et al., respect to the structure of CsA/CYP-A has
Fig. 28. A superposition (using C, of CYP-A) of

CsA (black) and MeIle4-cyclosporin (SDZ NIM
811) (grey). The substitutionof Ile for Leu in posi-
tion 4, without any other structural change, leads to
a < 1000-fold reduced immunosuppressive activity.
occurred for the side chain of residue 4 in that Fig. 29. The NMR-structure (a family of 20 struc-
Leu is replaced by Ile (Fig. 28). The mere tures with low NOE violations of maximally
presence of a branched (vs. unbranched) C, 0.29 A) of SDZ 214-103 l ~ ~ ton (2YP-A.
d
for residue 4 is thus sufficient to drastically
impair binding to calcineurin A/B (which
must, therefore, have a “tight-binding” pock- structures of free CsA and free SDZ 214-103
et for this region (PAPAGEORGIOU et al., are quite different (cf. Figs. 18 and 30).
1994); cf. the discussion in Sect. 4.3.
5.6.3 The 3D-Structure of SDZ 5.7 The X-Ray Structure of

214-103 Bound to Cyclophilin A Cyclophilin B Complexed with a
Cyclosporin Derivative
The 3D-structure of CYP-A bound SDZ
214-103 has been determined by NMR meth- The affinity of cyclophilin B for CsA is ap-
ods (WIDMERet al., unpublished results). 95 proximately 10-fold higher as that of cyclo-
structurally relevant intramolecular NOES philin A (SCHNEIDER et al., 1994). Similarly,
were used in structure calculations with the the complex CypB-CsA is known to be 5- to
program DIANA. The non-hydrogen atoms 10-fold more effective than CsA-CYP-A in
of a family of 20 structures (with low residual inhibiting the SerlThr phosphatase activity of
NOE-violations of maximally 0.29 A) are de- calcineurin (SWANSONet al., 1992). At San-
picted in Fig. 29. A comparison with the X- doz Pharma AG the X-ray structure of a com-
ray structure of CsA when bound to CYP-A plex of CypB with a cyclosporin derivative
(Fig. 30) shows that the backbone conforma- modified in position 8 ([0-(choliny1ester)-D-
tions are very similar (slight differences might Serx]-cyclosporin) was determined in order to
be due to measurement errors of the two elucidate possible structural origins responsi-
techniques). On the other hand, the X-ray ble for this difference (MIKOLet al., 1994a).
Fig. 31. The architecture of CypB (in complex with

cyclosporin), viewed in the same orientation as
CYP-A in Fig. 8. The major structural differences
with respect to CYP-A occur at the N- and C-ter-
mini and in the loops 19-24 and 152-164 (residue
numbering of CypB).
Fig. 30. The X-ray structure of CsA (not bound to

cyclophilin) crystallized from acetone. C , atoms are
indicated by the residue numbers (1-11) and hy-
drogen bonds are indicated by dashed lines. All the
four amide protons are involved in hydrogen bonds
(three transannular ones: Abu*(NH)-ValS(CO),
Va15(NH)-Abu2(CO), and Ala'(NH)-MeVal'*-
(CO), and one additional hydrogen bond D-
AlaH(NH)-MeLeu6(CO)).
The overall structures of CypB and CYP-A

are relatively similar (cf. Figs. 31, 32, 19, and
20). However, significant differences occur in
two loops (residues 19-24 and 152-164 of
CypB) and at the N- and C-termini. The ac-
tive site regions of CypB and CYP-A are very
similar. Indeed, there are practically no dif-
ferences for any residues of the cyclosporin- Fig. 32. The complex CypB-CsA rotated by 90"
binding pocket within 5.0 A of cyclosporin with respect to Fig. 30 around a vertical axis in the
(rmsd for all these non-hydrogen atoms is less Picture Plane, to be compared with the view of
than 0.15 A). The cholinylester derivatization depicted in Fig. 20.
in position 8 could not be seen in the electron
density map, probably because of too high
conformational mobility. The binding and
conformation of the cyclosporin residues are
practically identical for the two complexes. for an “extreme” derivatization of CsA). The
Candidates for an explanation of the in- cyclosporin structure is also preserved in
creased potency of CypB-Cs complexes are complexes with different receptors (as seen in
the following residues of CypB: Arg”, Lys113, the complex structures with CYP-A, CypB,
Ala’2x, and the loop containing Arg’5X. CypC). The division of the cyclosporin mole-
cule into a “binding region” (making binding
interactions with cyclophilin) and an “effector
5.8 The X-Ray Structure of region” (important for the immunosuppres-
Cyclophilin C Complexed with sive effect by mediating interaction of the cy-
clophilin complex with calcineurin) is visual-
CsA ized in Fig. 23. In the effector region there are
some sensitive positions, i.e., residues in very
The structure of CypC (KE et al., 1993b) close contact with calcineurin such as, e.g., re-
shows that it is similar to CYP-A, with the ex- sidue 4 The mere change from MeLeu4 to
ception of the loops Asp7-Lys9, Met70-Ile76, MeIle4 without any other structural change is
and Gln79-Thr189. The cyclosporin binding sufficient to completely abolish immunosup-
pockets are practically identical as well as the pressive activity (cf. Fig. 28).
conformation of cyclosporin itself for the CsA alone cannot exert an inhibitory effect
complexes CypC-CsA and CsA-CYP-A. on calcineurin, it needs to be complexed with
cyclophilin. Hence, there must be crucial di-
rect interactions between cyclophilin and cal-
5.9 Conclusions cineurin residues also. Evidence for this hy-
pothesis is based on the different inhibitory
Cyclosporins can adopt a variety of confor- potencies of cyclophilins A, B, and C when
mations depending on the molecular environ- complexed with CsA (FLIRIet al., 1993). Ex-
ment. If the environment is hydrophobic, the tensive site-directed mutagenesis experiments
number of intramolecular hydrogen bonds is on CYP-A have been done (ETZKORNet al.,
maximized by adopting a conformation as 1994) in order to find such crucial residues
shown in Fig. 30 (all four amide protons are (e.g., the CYP-A mutant Argl48Glu com-
involved in intramolecular hydrogen bonds). plexed with CsA shows a 20-fold improved
In a polar environment, on the other hand, inhibition of calcineurin). The X-ray structure
the seven N-methyl groups are shielded from of CypB (the complex CypB-CsA is at least
the solvent by adopting a conformation as 10-fold more effective than CsA-CYP-A in
shown in Fig. 8 (there is just one intramolec- inhibiting calcineurin) suggests also that the
ular hydrogen bond between the hydroxyl loop containing ArgISX,which corresponds to
’
group of MeBmt and the carbonyl oxygen of the loop containing Arg’4Xin CYP-A, might
MeLeu4). The transition between these two be important for modulating the interactions
conformations resembles the “inversion of a with calcineurin (apart from the possible can-
glove”. The NMR structure of a water-soluble didates ArggO,Lys113,and Ala’2x of CypB). A
cyclosporin derivative in a polar solvent in final answer to these questions will be given,
the absence of cyclophilin shows that cyclo- of course, by the experimentally determined
sporins can preadopt the cyclophilin-bound 3D-structure of the CsA-CYP-A-calcineurin
conformation in the absence of the protein complex. The additional availability of a 3D-
and there is no need for an “induced-fit” hy- structure of the FKBP12-FK506-calcineurin
pothesis to explain the binding of cyclospo- complex would also answer the question of
rins to cyclophilins. The CYP-A-bound back- how the obviously structurally dissimilar
bone conformation seems to be an energeti- CsA-CYP-A and FKBP12-FK506 complexes
cally favorable one. It is “robust” against per- can both compete for interaction with calci-
turbations of cyclosporin side chains: In all neurin.
cyclosporin derivatives analyzed structurally
until now the backbone conformation is prac-
tically identical in all cases; (see, e.g., Fig. 29
on three enzymes or enzymatic pathways, re-

6 Biosynthesis of spectively: (1) cyclosporin synthetase, (2) ala-
Cyclosporins nine racemase, (3) Bmt-synthesizing en-
zymes.
6.1 Introduction
6.2 Cyclosporin Synthetase
Some structural features of cyclosporin A
are suggestive of a non-ribosomal biosynthet- 6.2.1 Enzymatic Activities of
ic origin: the unusual amino acids (4R)-4-
[(E)-2-butenyl]-4-methyl-~-threonine(Bmt, Cyclosporin Synthetase
position l), a-amino butyric acid (position 2),
and D-alanine (position 8) as well as the cyclic Enzymatic activity in protein fractions of
structure and the N-methylation of seven Tolypocladium niveum that could be ascribed
peptide bonds. Indeed, cyclosporins are bio- to cyclosporin synthetase was first detected
synthesized by an extraordinary large mul- by ZOCHERet al. (1986). Initial in vitro stu-
tienzyme called cyclosporin synthetase. How- dies led to the enzymatic synthesis of cyclo-
ever, efficient cyclosporin biosynthesis not (D-alanyl-N-methylleucyl) diketopiperazine
only needs a high activity of this central en- (D-DKP) rather than to the synthesis of cy-
zyme but also of pathways providing the un- closporin A. The reason for this is not known,
usual components as well as a sufficiently but considering the finding that mutants of T.
high internal pool of the methyl group donor niveum expressing non-functional cyclosporin
S-adenosyl-methionine (Tab. 3). Whereas a- synthetase and, therefore, lacking the compe-
amino butyric acid is most likely derived from tence to produce cyclosporin A produce D-
the common amino acid pool (SENNet al., DKP instead (DITTMANN et al., 1990) it can
1991) Bmt and D-alanine have to be supplied be concluded that D-DKP is a bypass-product
by separate pathways. Hence, general interest of cyclosporin synthetase under non-optimal
in cyclosporin biosynthesis has been focused conditions. First in vifro enzymatic synthesis
Tab. 3. Requirements for Cyclosporin Biosynthesis
Source Enzyme(s) Building Units and Cofactors for

Cs-Biosynthesis
Common amino acid pool Alanine, a-amino butyric acid,

glycine, leucine, valine
(L)-Alanine Alanine racemase (D)-Alanine
Acetate, SAM, NADPH Bmt-polyketide synthase Bmt
0 2 . Nz Transformation enzymes
c,-Pool SAM Synthetase SAM
Glucose ATP (MgCIJ
4'-Phosphopantetheine covalently
Pantothenate attached to cyclosporin synthetase
Cyclosporin
Synthetase
1
Cyclosporins
of the complete cyclosporin molecule from et al., 1990; MACCABEet al., 1991; GUTIER-
the constituent amino acids, ATP, MgC12, and REZ et al., 1991), enniatins (HAESE et al.,
S-adenosyl-methionine as a methyl group do- 1993), gramicidin S (TURGAYet al., 1992;
nor was successful in 1987 (BILLICHand HORI et al., 1989; KRAETZSCHMAR et al.,
ZOCHER,1987). 1989), tyrocidine A (WECKERMANN et al.,
Whereas enzymes synthesizing other D-ami- 1988), HC-toxin (SCOTT-CRAIGet al., 1992),
no acid-containing peptides, e.g., gramicidin S and surfactin (COSMINAet al., 1993) it be-
and tyrocidine (KLEINKAUF and VON D ~ H R - came clear that such enzymes are composed
EN, 1990), harbor an integral epimerase func- of domains. Each of these domains is respon-
tion which epimerizes the respective L-amino sible for the recognition and activation of one
acid into the D-form following its activation, amino acid and for the peptidation reaction
cyclosporin synthetase rather incorporates D- with the amino acid activated by the neigh-
alanine in position 8 only if already supplied boring domain. In the case of methylated en-
in its D-form. Hence, epimerization of t-alan- niatins the corresponding domain harbors an
ine has to be carried out by a distinct enzyme additional module responsible for the methyl-
(see Tab. 3). In contrast, the methyltransfer- ation step (HAESE et al., 1993). Taking the
ase activity for the N-methylation of the pep- molecular masses of all these peptide synthet-
tide bonds is an integral part of the purified ases into account a molecular mass for cyclo-
enzyme (LAWENand ZOCHER,1990). Pro- sporin synthetase of at least 1.6 MDa can be
teolysis experiments indicated the existence extrapolated.
of several methyltransferase domains in one
enzyme molecule, probably one for each of
the seven methylation steps, which has been 6.2.3 Isolation and
confirmed by cloning and sequencing the Characterization of the
gene (see below). Alltogether, cyclosporin
synthetase catalyzes at least 40 partial reac- Cyclosporin Synthetase Gene
tion steps: 11 aminoadenylation reactions, 11
transthiolation reactions, 7 N-methylation The gene coding for cyclosporin synthetase
reactions, 10 elongation reactions, and the fi- has recently been cloned and sequenced
nal cyclization reaction. (LEITNERet al., 1994; WEBERet al., 1994). In
order to obtain partial amino acid sequences
to derive specific oligonucleotide probes the
6.2.2 Characterization of the cyclosporin synthetase was purified from my-
celia of T. niveum and partially digested with
Enzyme endoproteinases. 18 fragments were isolated,
purified, and used for sequence determina-
Purification and analysis by SDS-PAGE tion. One of these fragments was identified by
clearly demonstrated that the complete reac- photoaffinity labeling with S-adenosyl-me-
tion sequence of cyclosporin biosynthesis is thionine and a second fragment by its capaci-
coded by a single multienzyme polypeptide ty to activate L-alanine (LAWENand Zo-
(LAWENand ZOCHER,1990) but lacking ade- CHER, 1990; LEITNERet al., 1994; WEBER et
quate molecular weight markers it was a diffi- al., 1994).
cult task to determine the molecular mass of The cyclosporin synthetase gene (simA)
this giant protein. Analytical ultracentrifuga- was cloned using an oligonucleotide probe
tion indicated that the enzyme most likely has derived from one of these amino acid se-
a discus-like structure with a diameter of quences. A genomic library of T. niveum
about 33 nm, a thickness of 4.6 nm, and a mo- ATCC 34921 in XEMBL3 was screened, al-
lecular mass of about 1.4MDa (SCHMIDTet lowing determination of a nucleotide se-
al., 1992). From cloning and sequencing stu- quence of 46889bp with an ORF of
dies of other peptide synthetases responsible 45823 bp. The gene product (CYSYN) corre-
for biosynthesis of, e.g., 6-(a-L-aminoadipy1)- sponds to 15281 amino acids with a predicted
L-cysteinyl-D-valine (SMITHet al., 1990; DIEZ molecular mass of 1.7MDa. A data bank
search showed characteristic similarities to ity has been shown for other peptide synthet-
known peptide synthetases. 11 regions of ases (LAWENet al., 1992).
CYSYN show high similarity. Two different The C-terminal end of domain 11 is not the
types are found within the 11 domains of C-terminus of CYSYN. There are approxi-
CYSYN. The first type (type I) is about 1000 mately 500 amino acid residues of non-do-
amino acids in size and very similar to the one main character. As the biosynthesis of cyclo-
already detected in other multifunctional sporins also includes ring closure of the pep-
peptide synthetases. This type I is shown in tide one could speculate that this non-domain
Fig. 34 (non-filled rectangle). The second (i.e., showing no significant homology to do-
type (type 11) is larger than the first type and mains 1-11) peptide sequence harbors this
includes an approximately 447 amino acid po- function (WEBERet al., 1994).
lypeptide (Fig. 33 and 34). This polypeptide
sequence has probably N-methyltransferase
activity as demonstrated by several lines of 6.2.4 Manipulation of the Cloned
evidence: (1) An experimentally derived ami- Cyclosporin Synthetase Gene by
no acid sequence of a 45 kDa fragment of
cyclosporin synthetase having methyltransfer- Integrative Transformation
ase activity was found in the deduced amino
acid sequence at a position corresponding to The description of the simA gene and the
one of these additional sequences (Fig. 33). correlation of the order of protein domains
(2) This additional sequence is very similar to and constituent amino acids of cyclosporin
a corresponding sequence found in the ennia- will enable the construction of new fungal
tin synthetase (HAESE et al., 1993). (3) All strains by exchange of domain-specific parts
seven sequences show similarity to a postu- of the simA gene by gene replacement (Fig.
lated consensus sequence for S-adenosyl-me- 35). Necessary for such experiments is a
thionine binding and all seven CYSYN se- transformation system which allows the re-in-
quences are highly conserved in this region troduction of in v i m manipulated DNA by
(LEITNERet al., 1994; WEBERet al., 1994). homologous recombination.
There are seven type I1 and four type I do- Plasmids composed of A. nidulans promot-
mains. All ll domains contain in the same re- ers fused to a bacterial hygromycin phospho-
lative position the putative amino acid bind- transferase gene have been used successfully
ing and phosphopantetheine acid attachment in a number of fungal species (FINCHAM,
sites found in other peptide synthetases. The 1989). Such plasmids can also be used for the
130000 Da cyclosporin synthetase fragment transformation of T. niveum but there is a
could be characterized by its capacity to acti- high proportion of multiple tandem integra-
vate L-alanine. This permits an assignment of tions of the plasmid DNA.
its N-terminal sequence to CYSYN (Fig. 33). Transformation systems for T. niveum were
The fragment corresponds well to the 11th described by LEITNERand WEBER (1994).
domain (LEITNERet al., 1994; WEBERet al., The authors isolated the cyclophilin gene as a
1994). L-Alanine is known as the last amino source of a homologous promoter element
acid added to the growing peptide chain and called the isolated gene cpfA. The entire
(DITTMANN et al., 1994). gene consists of 890 bp, including the three
The order of domains - with or without a introns of 220 bp, 57 bp, and 60 bp respective-
putative methyltransferase activity - corre- ly. The gene codes for a protein (CFT) with a
sponds to the biosynthetic order of methyl- molecular mass of 19569 Da. The cDNA cor-
ated and non-methylated amino acids (DITT- responding to cptA is similar to the cyclophil-
MANN et al., 1994). The authors, therefore, in cDNA of Neurosporu crussu (80%) which
concluded that there is a correspondence of was used as a probe to isolate the T. niveum
the order of the constitutive amino acids of gene. At the amino acid level the similarity is
cyclosporin A and the order of the 11 do- also 80%. The promoter of the T. niveum
mains as shown in Fig. 33 (LEITNERet al., gene was used for plasmid constructions in
1994; WEBERet al., 1994). A similar colinear- which this promoter is fused to a bacterial hy-
Sfi
SPeI f h I i”””
SCaI
i iscaI
‘Xbal
SCaI
NotI
Xba
in lIlmllI
MT Gala
loo00 unnn, 30000 40000
I
Fig. 33. Structure of the cyclosporin synthetase gene (simA) and the derived polypeptide (CYSYN). A
partial restriction map of the 47 kb gene region is shown. The structure of the derived translation product
is illustrated by horizontal boxes. There are two different types of domains which are described in more
detail in Fig. 34. The grey parts identify the N-methyltransferase subdomains. The box labeled with a C
(horizontal lines) indicates the about 500 amino acid long C-terminal part of the translation product. The
two boxes with vertical lines indicate the postujated positions of the N-methyltransferase fragment (MT)
and the L-alanine activating fragment (L-ala) based on the N-terminal sequences and the observed molec-
ular masses (WEBERet al., 1994).
TYPE I
n.
a b c d e f g h i k l
Fig. 34. Relationship of the two do-
main types. The two types of do-
mains are aligned in order to illus-
I I 81
trate their relationships. “a” to “I”
illustrate similar amino acid se-
quences. The grey part of the lower
box stands for the N-methyltrans-
ferase fragment. The small box
within this region symbolizes puta-
tive S-adenosyl methionine binding a b C d e f g
sites. Further putative binding sites
for AMP and DhosDhoDantethein
1 1. B I sl
cofactor are aiso inhicited (AMP, AMP X Y Z PP
PP) (WEBERet al., 1994). TYPE U
gromycin phosphotransferase gene. A frag- transformed with purified plasmid pSIMlO

ment containing 2.1 kb upstream of the puta- DNA in the presence of PEG and hygromy-
tive start codon of the cprA gene was ampli- cin-resistant colonies were obtained (LEIT-
fied by PCR and ligated with a 1.76 kb ClaI- NER and WEBER,1994; WEBERet al., 1994).
XbaI fragment from pCSN44. This 1.76 kb As it was intended to use the transformation
pCSN44 fragment covers a bacterial hygro- system to manipulate the gene for the cyclo-
mycin phosphotransferase gene and the frag- sporin synthetase of T. niveum, derivatives of
ment with the transcriptional terminator of pSIMlO were constructed containing internal
the A. nidulans trpC gene (STABENet al., fragments of this gene. The first example,
1989). The resulting plasmid was called pSIM11, contains a 3.6 kb XhoI restriction
pSIM10. Protoplasts of T. niveum could be fragment of the simA gene. A crossover be-
Cyclosporin A XhoI fragment including the promoter of the

cyclophilin gene, the hygromycin phospho-
MeBmt Abu Sar MeLeu Val transferase gene, and a A. nidulans trpC tran-
scription terminator fragment as inserted into
the central XhoI site of the cloned simA frag-
ment. Following transformation, this con-
struct can be inserted into the genomic DNA
a
by a double crossover (Fig. 36). pSIM13 was
used for transformation of T. niveum ATCC
34921 protoplasts and the transformants were
analyzed for cyclosporin production. DNA
from pSIM13 transformants was analyzed by
MeBmt Abu Sar Val Val Southern hybridization. For the three differ-
ent restriction enzymes used to digest the
DNA the sites expected for the DNA of a
Cyclosporin Q mutant generated by gene replacement were
identified (Fig. 37) (WEBERet al., 1994). The
Fig. 35. Theoretical reprogramming of cyclosporin high frequency of transformants which do not
synthetase. The partial structure of the cyclosporin produce cyclosporin indicates that only one
synthetase gene is shown schematically. The gene copy of the simA gene is present in the ge-
regions coding for domains are shown as boxes and nome of T. niveum. A cyclosporin non-pro-
labeled by the names of the corresponding amino ducing mutant of T. inflatum Cyb156 accumu-
acids. A fragment of the gene coding for a domain
which activates valine (hatched box) is modified in lating the cyclosporin precursor amino acid
vitro by adding the flanking regions of another part Bmt has been described by SANGLIERet al.
of the gene coding for a domain which activates (1990). This mutant showed reduced sporula-
leucine. Due to this sequences the DNA coding for tion and reverted to cyclosporin formation at
the valine-specific domain can replace the DNA high frequency. In contrast, the pSIM11-
coding for the leucine-specific domain following re- transformants show normal morphology and
combination. The recombined gene codes for a cy- growth characteristics. Transformants were
closporin synthetase that preferentially produces also analyzed for accumulation of Bmt but
cyclosporin Q instead of cyclosporin A. This is a only small amounts could be detected even if
hypothetical scheme.
high-yielding T. niveum strains were used for
transformation (LEITNERand WEBER, 1994;
WEBERet al., 1994).
tween the cloned and the genomic version of The transformation system described
DNA should lead to insertion of the plasmid proved to be a powerful tool for gene disrup-
DNA and to a partial duplication of the tar- tion of the cyclosporin synthetase (simA)
get DNA. As the cloned DNA does not con- gene of T. niveum. It is intended to use these
tain the 5 ’ or 3 ’ end of the gene the insertion methods for gene replacement experiments in
inactivates the gene. T. niveum protoplasts which parts of the simA gene coding for ami-
were transformed with pSIMll following lin- no acid specific domains are exchanged. The
earization of the plasmid DNA. Out of 81 mutated genes will direct the synthesis of new
pSIM11 transformants 50 (62%) were found cyclosporins or of cyclosporins which are up
that do not produce cyclosporin. Several of to now only by-products of the biosynthesis.
the transformants which lost the ability to
produce cyclosporin were verified to contain
the expected fragment sizes by Southern hy- 6.2.5 Mechanistic Aspects of
bridizations (LEITNER and WEBER, 1994;
WEBERet al., 1994). Biosynthesis
The second example, pSIM13, includes a
2.1 kb EcoRI fragment of the cyclosporin syn- Detailed studies of the mechanism of cy-
thetase gene cloned into pUC18. A 3.7 kb closporin biosynthesis provided strong evi-
Fig. 36. Gene disruption with

pSIMl1. The upper part of the
figure shows the XbaI fragment
of the Tolypocladium niveum
simA gene. The XhoI fragment
cloned in pSIMll is indicated
"")( I
,Sd
ji XhO
NdI
XhoI
by vertical lines. The middle
part of the figure shows
pSIM11, the hph coding region
is indicated by a box with diag-
onal lines. The lower part shows
the predicted structure of the
DNA following a single recom-
bination event in the region
cloned in pSIM11. The restric- aI
tion fragments hybridizing with
the labeled XhoI fragment are
indicated (WEBERet al., 1994). 10.6kb 4.8b
3.4kb
-2.1kb
0.95kb
Fig. 37. Gene disruption with the Nhel EcoRI
fragment of pSIM13. The upper part of the fig-
ure shows a region of the Tolypocladium niv-
eum simA gene. The EcoRI fragment cloned
and disrupted in pSIM13 is indicated by a box
with vertical lines. The middle part of the fig-
ure shows the NheI EcoRI fragment of pSIM13
n
used for transformation; the hph coding region
is indicated by a box with diagonal lines. The
lower part shows the predicted structure of the
DNA following a double recombination event
in the region cloned in pSIM13. The restriction
V
EcoRV EcoRV
fragments hybridizing with the labeled EcoRI
fragment if the DNA is digested with EcoRI or
EcoRV or SpeI are indicated. For SalI, e.g.,
this is a 3.4 kb fragment for the nondisrupted 7.1kb
gene and a 7.1 kb fragment following gene dis- 2.71rb
ruption (WEBERet al., 1994). 4:R
dence that binding of D-alanine is the initial thetic cycle are still unclear. As early as 1973,
reaction of biosynthesis, followed by the step- a thiotemplate mechanism had been sug-
wise synthesis of a single linear undecapep- gested f o r the nonribosomal biosynthesis of
tide precursor which is finally cyclized to cy- peptides (LALANDand ZIMMER, 1973). Pep-
closporin A (DITTMANN et al., 1994); (cf. Fig. tide synthetases activate their respective sub-
38). However, the details of the whole biosyn- strate amino acids in two steps, involving am-
inoacyl adenylates and thioesters as reactive rin synthetase is rather low. This is reflected
intermediates, as shown in the following by the finding of more than 30 naturally oc-
equation: curring variants of cyclosporin A synthesized
by T.niveurn (TRABERet al., 1987); (see Tab.
E,+aa+MgATP*-+E(aaAMP)+E-S -aa l), the relative amount of the cyclosporin var-
Ea enzyme, adenylation site iants being considerably dependent on the
fermentation conditions (KOBEL and TRA-
E - S - enzyme, amino acid binding site BER, 1982). Strikingly, only some positions,
aa amino acid
especially position 2, show variability in their
amino acid incorporation whereas other posi-
The assembly of the peptide chain is achieved
tions, especially positions 3 and 8, are rela-
by repeated transpeptidation and transthiola-
tively invariant. Detailed in vitro studies
tion reactions facilitated by 4 ’-phosphopante-
showed a somewhat divergent behavior of the
theine as a carrier which interacts with the SH isolated enzyme: Position 1, e.g., was found to
groups of the peripheral amino acid activa-
show substantial flexibility and position 8 ex-
tion centers (KLEINKAUF and VON DOHREN,
hibits low substrate specificity as well (LA-
1990). The 4 ’-phosphopantetheine molecule
has a length of 2 nm and has been postulated WEN et al., 1989,1992; LAWENand TRABER,
to reach up to six of the active centers of a 1993). Even incorporation of p-alanine in po-
synthetase (KLEINKAUFand VON DOHREN, sition 8 (and 7, respectively) was shown to be
1987). Although 4’-phosphopantetheine is possible, causing formation of a 34-membered
also an essential component of cyclosporin ring in contrast to the 33-membered ring of
synthetase (LAWENand ZOCHER,1990), in cyclosporins (LAWENet al., 1994). However,
view of the size of this enzyme the model of a only few of these in vitro accessible com-
central swinging phosphopantetheine arm ap- pounds can be synthesized in vivo by precur-
pears not applicable. Recent results with sor directed biosynthesis (TRABERet al.,
gramicidin S synthetase as well as with surfac- 1988, 1994; HENSENSet al., 1992) perhaps
tin synthetase strongly indicate that the cru- due to unsuitable physiological conditions
cial amino acid for binding of the substrate and/or metabolic utilization of the added pre-
amino acid is serine (SCHLUMBOHM et al., cursors.
1991; VOLLENBROICH et al., 1993; D’SOUZA
et al., 1993), leading the authors to suggest a
modified version of the thiotemplate mecha-
nism in which each substrate amino acid is
bound by the SH group of a phosphopante-
theine arm which in turn is attached to the
crucial serine residue of the respective activa-
tion center. This concept is further supported Fig. 38. Model for cyclosporin biosynthesis by a b
by the finding of the consensus amino acid se- modified thiotemplate mechanism. Cyclosporin
quence for attachment of phosphopante- synthetase activates its respective substrate amino
theine in each of the 11 domains of cyclospo- acids in two steps involving adenylation (indicated
rin synthetase (see Sect. 6.2.3). A model for as a l , a2, etc.) followed by thio-esterification, most
cyclosporin biosynthesis, including the con- probably by a phosphopantetheine arm attached to
a serine symbolized by -0- (SCHLUMBOHM et
cept of a modified thiotemplate mechanism, is al., 1991). 7 out of 11 amino acids are methylated at
shown in Fig. 38. this stage by integral methyl transfer units. Re-
peated transpeptidation and transthiolation reac-
tions, (D)-alanine being the N-terminal amino acid
6.2.6 Biosynthesis of Cyclosporin (DIVMANet al., 1994), lead to the assembly of the
Variants linear undecapeptide precursor which is finally cy-
clized to cyclosporin. The numbers 1-11 do not cor-
relate with the numbering system of the chemical
As common in nonribosomal peptide syn- nomenclature but rather with the sequence of reac-
thesis, the amino acid specificity of cyclospo- tions of the biosynthetic pathway.
N 1 2 ,3 I 4 5 Domains 1-1 1
"
I
a3 S A M - REPEATED ACTIVATION AND

(leu) ELONGATION REACTlONS
N = N - ~ u ~ ~ s al, ...., allaminoacidsl-11 CYCLIZATION

(involvement of
c = c - terminus R1,...., R11 side chains of al, ...., a1 1 possible)
@@ methyltransfaase - unit -
S A M = S - adenosyl methionine .5
C-t~sequence PP = phosphopantotbein arm
ICYCMSPORINI
580 12 Cyclosporins:Recent Developments in Biosynthesis, Pharmacology and Biology
6.2.7 Related Enzymes: Peptolide of C. oligospermum (Corda) Bonorden

(HOFFMANN et al., 1994), the producer of the
SDZ 214-103 Synthetase SDZ 214-103 which contains an ester linked
D-2-hydroxyisovaleric acid instead of D-ala-
In addition to the multitude of cyclosporin nine (cf. Fig. 2). In view of this result, it seems
variants which are synthesized by cyclosporin very likely that the exclusive and indispensa-
synthetase in vitro and/or in vivo, the exis- ble role of the detectable alanine racemase is
tence of many related compounds with more supplying D-alanine for cyclosporin biosyn-
pronounced structural deviations, synthesized thesis. In line with this assumption, inhibition
by homologous multienzymes, can be ex- of alanine racemase activity in vivo destroys
pected. One example is SDZ 214-103 (Fig. 2), the competence of T. niveum for biosynthesis
the cyclosporin-related peptolide produced of cyclosporin A, shown by feeding experi-
by the fungus Cylindrotrichum oligospermum ments with 3-fluoro-(~)-alanine, a well
(Corda) Bonorden (HOFFMANN et al., 1994). known inhibitor of prokaryotic alanine ra-
The multienzyme responsible for biosynthesis cemases (HENSENSet al., 1992). Since activa-
of this compound was isolated from C. oligo- tion of D-alanine is the first step in cyclospo-
spermum, found to have a similar molecular rin biosynthesis (DITTMANNet al., 1994) and
weight as cyclosporin synthetase and to show the affinity of cyclosporin synthetase for D-
cross reactivity with cyclosporin synthetase- alanine is remarkably high (HOFFMANN et al.,
specific antibodies (LAWEN et al., 1991, 1994) the role of this enzyme seems not to be
1992). Despite the obvious similarity of the restricted only to supply one of the structural
enzymes as well as of their respective prod- components of cyclosporin A but it appears
ucts, cyclosporin synthetase and SDZ 214-103 to be rather a rate-limiting pacemaker of the
synthetase can not substitute enzymatically whole biosynthetic cycle. This view is sup-
for each other. Furthermore, detailed in vitro ported by the absence of free D-alanine in ex-
incorporation studies point to a more re- tracts of cyclosporin producing strains of T.
stricted substrate specificity of SDZ 214-103 niveum.
synthetase (LAWENand TRABER,1993).
6.3.2 Characterization of the

6.3 Alanine Racemase Enzyme
6.3.1 Alanine Racemizing Activity Like many of the prokaryotic homologs the
T.niveum enzyme consists of several subunits
As mentioned above, cyclosporin synthet- of a molecular mass of about 40000 Da (esti-
ase is the first example of a peptide synthet- mated by SDS-gel electrophoresis) and de-
ase which lacks an integral activity epimeriz- pends on pyridoxal phosphate as the exclu-
ing L-alanine to D-alanine. Hence, the exis- sive cofactor (HOFFMANN et al., 1994). Con-
tence of a distinct alanine racemase had to be sequently, the enzyme is susceptible to the
postulated. This was somewhat unexpected classical racemase suicide inhibitors in vivo
since the existence of alanine racemases had (HENSENSet al., 1992) as well as in vitro, as
previously been restricted to prokaryotic or- shown for D- and t-(l-aminoethy1)-phos-
ganisms where they are primarily involved in phonate (HOFFMANNet al., 1994). In vitro
cell wall biosynthesis (ADAMS,1972; WALSH studies with a purified enzyme fraction fur-
et al., 1985; SODA et al., 1986). Enzymatic ther revealed that the specificity of the racem-
analysis of crude extracts of T. niveum indeed ase for L- and D-alanine is high but not exclu-
led to the detection of an alanine racemizing sive. Compared to the reaction velocity of the
activity and allowed purification and charac- epimerization of L-alanine, L-serine, 2-~-ami-
terization of the respective protein (HOFF- nobutyric acid, and L-leucine, e.g., are racem-
MANN et al., 1994). Interestingly, no corre- ized with relative reaction rates of 23%, 15%,
sponding activity could be detected in strains and 13%, respectively (HOFFMANNet al.,
1994). Whether the enzyme is related to the 6.4.2 Identification of the Basic
prokaryotic homologs in an evolutionary
sense will be elucidated by cloning and se- Assembly Product and
quencing the corresponding gene which is in Characterization of
progress (K. SCH~RGENDORFER et al., un- Bmt-Polyketide Synthase
published results).
NMR analysis of cyclosporin A produced
by feeding experiments with [l-'3C,'X02]ace-
6.4 Bmt-Synthesizing Enzymes tate demonstrated retention of the acetate-
derived "0-isotope during biosynthesis
(measured as an upfield shift of the *3C-signal
6.4.1 Polyketide Origin of Bmt in position 3 of Bmt; OFFENZELLER et al.,
1993). Assuming that all condensation, reduc-
Bmt with its long aliphatic chain is the most tion and dehydration steps as well as the me-
unusual amino acid of cyclosporins. The thylation reaction occur in the course of the
structure suggested an acetate-derived bio- basic assembly, this result was a first conclu-
synthetic pathway, confirmed by KOBEL et al. sive proof of 3(R)-hydroxy4(R)-methyl-
(1983) who were able to show by feeding ex- 6(E)-octenoic acid to be the biosynthetic key
periments that four "C-labeled acetate units intermediate (Fig. 39). Subsequently, detailed
are coupled in a head-to-tail fashion and enzymatic studies led to the identification of
processed to finally yield Bmt. Methionine this compound as a product of cell free en-
was identified as the source of the methyl zyme fractions from T. niveurn (OFFENZEL-
group in position 4, whereas the origin of the LER et al., 1993). The respective Bmt-polyke-
amino group remained unknown. These re- tide synthase has been characterized and its
sults were later confirmed by feeding experi- isolation, which is in progress, will allow its
ments using selectively I3C-labeled glucose thorough characterization and provide the
(SENNet al., 1991). Hence, Bmt can be clas- basis for cloning the corresponding gene. One
sified as a polyketide. interesting feature of the enzyme is that it re-
These findings led to a model for the bio- leases its product as a coenzyme A thioester
synthetic pathway of Bmt which, in accor- (OFFENZELLER et a]., 1993), a process also
dance with the current concepts of polyketide known from fatty acid synthases of several
biosynthesis, most likely takes place in at fungi (LYNEN,1980). Details of the basic as-
least two distinct phases (for recent reviews, sembly process, especially the stage of the
see KATZ and DONADIO, 1993; JORDAN and methylation reaction (see possible routes in
SPENCER,1993; O'HAGAN,1991; ROBINSON, Fig. 39), remain to be elucidated and are un-
1991; HOPWOODand SHERMAN, 1990; SIMP- der investigation.
SON, 1989): (1) a basic assembly process in-
volving the coupling of four acetate units, the
respective reduction and dehydration steps as 6.4.3 The Transformation Process
well as the methylation reaction; and (2) a
transformation process introducing the amino According to the current understanding of
group (OFFENZELLER et al., 1993). Based on Bmt-biosynthesis transformation most likely
this concept it was obvious that identification takes place by hydroxylation of the basic as-
of the basic assembly product would provide sembly product at the C2-atom with subse-
the first clues for elucidation of the biosyn- quent oxidation and transamination to Bmt
thetic route. (cf. Fig. 39). It seems unlikely that these
transformation reactions are performed by
Bmt polyketide synthase itself suggesting the
involvement of additional enzymes in Bmt
biosynthesis, namely of a hydroxylase, an
oxidase, and a transaminase. This model is
supported by in vivo incorporation of the am-
TRANSFORMATION
I
I $=OH.wl
Fig. 39. Biosyntheticpathway to Bmt. Abbreviations:S-adenosylmethionine, SAM; acyl carrier protein, ACP.
7 References 583
ino group of ”N-labeled glutamate into Bmt BANG, H., BRUNE,K., NAGER, C., FEIGE, U.
which indicates participation of transaminases (1993), Interleukin-8 is a cyclosporin A binding
in the biosynthetic route (R. TRABER and K. protein, Experientia 49, 533-538.
MEMMERT,personal communication). BARRY,J. M. (1992), Immunosuppressive drugs in
renal transplantation. A review of the regimens,
Drugs 44,554-566.
Note added in proof BAUMANN, G., BOREL,J. F. (1992), MCcanismes
molkculaires de I’action des agents immunosup-
presseurs, mtdecine/sciences 8, 366-371.
Enzyme-bound 3-0x0-4-hexenoic acid
BAUMANN, G., GEISSE,S., SULLIVAN, M. (1991),
could be identified as the direct methylation Cyclosporin A and FK-506 both affect DNA
precursor (OFFENZELLER,M., SAUTER,G., binding of regulatory nuclear proteins to the hu-
TOTSCHING,K., Su, Z., MOSER,H., TRABER, man interleukin-2 promoter, New Biologist 3,
R., SCHNEIDER-SCHERZER, E. (1996). Bio- 270-278.
synthesis of the unusual amino acid (4R)-4- BAUMANN, G., ZENKE,G., WENGER,R. M., HIES-
[(E)-2-butenyl]-4-methyl-~-threonineof cy- TAND, P. c.,QUESNIAUX, v. F. J., ANDERSEN,
closporin A: enzymatic analysis of the reac- E., SCHREIER, M. H. (1992), Molecular mecha-
tion sequence including identification of the nisms of immunosuppression, J. Autoimmun. 5
methylation precursor in a polyketide path- (SUPPI.A), 67-72.
way (Biochernisfry 35,84014412). BELL, A., WERNLI,B., FRANKLIN, R. M. (1994).
Roles of peptidyl-prolyl cis-trans isomerase and
calcineurin in the mechanisms of antimalarial ac-
tion of cyclosporin A, FK506, and rapamycin,
Biochem. Pharmacol. 48,495-503.
7 References BERGSMA, D. J., EDER,C., GROSS,M., KERSTEN,
H., SYLVESTER, D., APPELBAUM, E., CUSIMA-
NO, D., LIVI,G. P., MCLAUGHLIN, M. M., KA-
ABRAHAM, R. T., KARNITZ, L. M., SECRIST, J. P., SYAN, K. ET AL. (1991), The cyclophilin multi-
LEIBSON,P. J. (1992), Signal transduction gene family of peptidyl-prolyl isomerases. Char-
through the T-cell antigen receptor, Trends Bio- acterization of three separate human isoforms, J.
chem. Sci. 17,434-438. Biol. Chem. 266,23 204-23 214.
ADAMS,E. (1972), Amino acid racemases and epi- BIERER, B. E., JIN,Y. J., FRUMAN, D. A., CALVO,
merases, in: The Enzymes (BOYER,B. D., Ed.), V., BURAKOFF, S. J. (1991), FK 506 and rapamy-
3rd Edn., pp. 479-507. New York: Academic cin: molecular probes of T-lymphocyte activa-
Press. tion, Transplant. Proc. 23,2850-2855.
AEBI,J. D., DEYO,D. T., SUN,C. Q., GUILLAUME,BILLICH,A., ZOCHER,R. (1987), Enzymatic syn-
D., DUNLAP, B., RICH,D. H. (1990), Synthesis,
thesis of cyclosporin A, J. Biol. Chem. 262,
conformation, and immunosuppressive activities
17258-17259.
of three analogues of cyclosporin A modified in
the 1-position, J. Med. Chem. 33, 999-1009. BLUNDELL, T. L., COOPER,J., FOUNDLING, S. I.,
ALBERG,D. G., SCHREIBER, S. L. (1993), Struc- JONES, D. M., ATRASH, B., SZELKE, M. (1987),
ture-based design of a cyclophilin-calcineurin On the rational design of renin inhibitors: X-ray
bridging ligand, Science 262, 248-250. studies of aspartic proteinases complexed with
ALTSCHUH, D., VIX,O., REES,B., THIERRY, J. C. transition state analogues, Biochemistry 26,
(1992), A conformation of cyclosporin A in 5585-5595.
aqueous environment revealed by the X-ray BOESCH,D., GAVERIAUX, C., JACHEZ,B., POUR-
structure of a cyclosporin-Fab complex, Science TIER MANZANEDO, A., BOLLINGER, P., LOOR,
256,92-94. F. (1991), In vivo circumvention of P-glycopro-
ANDRIEU,J. M., EVEN,P., VENET, A. (1986), tein-mediated multidrug resistance of tumor
AIDS and related syndromes as a viral-induced cells with SDZ PSC 833, Cancer Res. 51, 4226-
autoimmune disease of the immune system: an 4233.
anti-MHC I1 disorder. Therapeutic implications, BRAM,R. J., HUNG,D. T., MARTIN, P. K., SCHREI-
AIDS Res. 2,163-174. BER,S. L., CRABTREE, G. R. (1993), Identifica-
ARMITAGE, R. J., BECKMANN, M. P., IDZERDA,R. tion of the immunophilins capable of mediating
L., ALPERT,A., FANSLOW, W. C. (1990), Regu- inhibition of signal transduction by cyclosporin
lation of interleukin 4 receptors on human T A and FK506: roles of calcineurin binding and
cells, Int. Immunol. 2, 1039-1045. cellular location, Mol. Cell Biol. 13, 47604769.
BRILLANTES, A.-M. B., ONDRIAS, K., SCOTT,A., DIEZ, B., GUTIERREZ,S., BARREDO,J. L., VON
KOBRINSKY, E., ONDRIASOVA, E., MOSCHEL- SOLINGEN,P., VAN DER VOORT, L. H. M.,
LA, M. C., JAYARAMAN, T., LANDERS,M., MARTIN,J. F. (1990), The cluster of Penicillin
EHRLICH, B. E., MARKS,A. R. (1994), Stabiliza- biosynthetic genes, J. Biol. Chem. 265, 16358-
tion of calcium release channel (ryanodine re- 16365.
ceptor) function by FK506-binding protein, Cell DITTMANN, J., LAWEN,A., ZOCHER,R., KLEIN-
77,513-523. KAUF,H. (1990), Isolation and partial character-
BROWN,E. J., ALBERS,M.W., SHIN,T. B., ICHI- ization of cyclosporin synthetase from a cyclo-
KAWA, K., CURTIS,K. T., LANE,W. S., SCHREI- sporin non-producing mutant of Beauveria niv-
BER, S. L. (1994), A mammalian protein tar- ea, Biol. Chem. Hoppe-Seyler 371,829-834.
geted by G1-arresting rapamycin-receptor com- DITTMANN, J., WENGER,R. M., KLEINKAUF, H.,
plex, Nature 369,756-758. LAWEN,A. (1994), Mechanism of cyclosporin A
CACALANO, N. A., CHEN,B. X., CLEVELAND, W. biosynthesis. Evidence for synthesis via a single
L., ERLANGER, B. F. (1992), Evidence for a linear undecapeptide precursor, J. Biol. Chem.
functional receptor for cyclosporin A on the sur- 269,2841-2846.
face of lymphocytes, Proc. Natl. Acad. Sci. USA DUMONT,F. J., MELINO,M. R., STARUCH,M. J.,
89,43534357. KOPRAK,S. L., FISCHER,P. A., SIGAL,N. H.
CALNE,R. Y., WHITE,D. J., THIRU,S., EVANS,D. (1990a), The immunosuppressive macrolides
B., MCMASTER,P., DUNN,D. C., CRADDOCK, FK-506 and rapamycin act as reciprocal antagon-
G. N., PENTLOW,B. D., ROLLES,K. (1978), Cy- ists in murine T cells, J. Immunol. 144, 1418-
closporin A in patients receiving renal allografts 1424.
from cadaver donors, Lancet 2 (8104-5), 1223- DUMONT,F. J., STARUCH,M. J., KOPRAK,S. L.,
1227. MELINO,M. R., SIGAL,N. H. (1990b), Distinct
CARONI,P., ROTHENFLUH,A., MCGLYNN,E., mechanisms of suppression of murine T cell acti-
SCHNEIDER, C. (1991), S-cyclophilin.New mem- vation by the related macrolides FK-506 and ra-
ber of the cyclophilin family associated with the pamycin, J. Immunol. 144,251-258.
secretory pathway, J. Biol. Chem. 266, 10739- DURETTE,P. L., BOGER,J., DUMONT, F. J., FIRE-
10742. STONE,R., FRANKSHUN, R. A,, LIN,C. S., ME-
LINO, M. R., PESSOLANO, A. A. (1988), A study
CHUNG,J., Kuo, C. J., CRABTREE, G. R., BLENIS, of the correlation between cyclophilin binding
J. (1992), Rapamycin-FKBP specifically blocks and in vitro immunosuppressive activity of cy-
growth-dependent activation of and signaling by closporin A and analogues, Transplant. Proc. 20,
the 70 kD S6 protein kinases, Cell 69, 1227- 51-57.
1336.
EBERLE, M. K., NUNINGER, F. (1992), Synthesis of
CLIPSTONE, N. A., CRABTREE, G. R. (1992), Iden- the main metabolite (OL-17) of cyclosporin A, J.
tification of calcineurin as a key signalling en- Org. Chem. 57,2689-2691.
zyme in T-lymphocyte activation, Nature 357, EBERLE,M. K., NUNINGER, F. (1993), Preparation
695-697. of (~-cysteine)8-cyclosporinvia intramolecular
COSMINA,P., RODRIGUEZ,F., DE FERRA,F., sulfur transfer reaction, J. Org. Chem. 58, 673-
GRANDI,G., PEREGO,M., VENEMA,G., VAN 677.
SINDEREN, D. (1993), Sequence and analysis of EMMEL,E. A., VERWEIJ,C. L., DURAND,D. B.,
the genetic locus responsible for surfactin syn- HIGGINS,K. M., LACY,E., CRABTREE,G. R.
thesis in Bacillus subtilis, Mol. Microbiol. 8,821- (1989), Cyclosporin A specifically inhibits func-
831. tion of nuclear proteins involved in T cell activa-
COWAN,S. W., NEWCOMER, M. E., JONES,T. A. tion, Science 246, 1617-1620.
(1990), Crystallographic refinement of human ENDICOTI-, J. A., LING,V. (1989), The biochemis-
serum retinol binding protein at 2 8, resolution, try of P-glycoprotein mediated multidrug resist-
Proteins 8, 44-61. ance, Annu. Rev. Biochem. 58, 137-171.
CRABTREE, G. R. (1989), Contingent genetic regul- ERNST,P. (1991), Lung transplantation: The cur-
atory events in T lymphocyte activation, Science rent state of knowledge, Union Med. Can. 120,
243,355-361. 64-66.
D'SOUZA,C., NAKANO,M. M., CORBELL,N., Zu- ETZKORN, F., CHANG,Z., STOLZ,L., WALSH,C. T.
BER,P. (1993), Amino-acylation site mutations (1994), Cyclophilin residues that affect noncom-
in amino acid-activating domains of surfactin petitive inhibition of the protein serine phos-
synthetase: Effects on surfactin production and phatase activity of calcineurin by the cyclophil-
competence development in Bacillus subtilis, J. in-cyclosporin A complex, Biochemistry 33,
Bacteriol. 175, 3502-3510. 2380-2388.
7 References 585
FAULDS,D., Go, A. S., BENFIELD, P. (1993), Cy- death in murine T cell hybridomas, Eur. J. Im-
closporin. A review of its pharmacodynamic and munol. 22, 2513-2517.
pharmacokinetic properties and therapeutic use GALAT,A., LANE,W. S., STANDAERT,R. F.,
in immunoregulatory disorders, Drugs 45, 953- SCHREIBER, S. L. (1992), A rapamycin-selective
1040. 25-kDa immunophilin, Biochemistry 31, 2427-
FINCHAM (1989), Transformation in fungi, Microb. 2434.
Rev. 53, 148-170. GETHING,M. J., SAMBROOK, J. (1992), Protein
FISCHER,G., WITTMANN-LIEBOLD, B., LANG,K., folding in the cell, Nature 355, 33-45.
KIEFHABER, T., SCHMID,F. X. (1989), Cyclo- GUTIERREZ,S., DIEZ, B., MONTENEGRO,E.,
philin and peptidyl-prolyl cis-trans isomerase are MARTIN,J. F. (1991), Characterization of the
probably identical proteins, Nature 337, 476- Cephalosporium acremonium pcbAB gene en-
478. coding cu-aminoadipyl-cysteinyl-valine synthet-
FLIRI,H. G., WENGER,R. M. (1990), Cyclosporins: ase, a large multidomain peptide synthetase:
synthetic studies, structure-activity relation- Linkage to the pcbC gene as a cluster of early
ships, biosynthesis, and mode of action, in: Bio- cephalosporin biosynthetic genes and evidence
chemistry of Peptide Antibiotics (KLEINKAUF, of multiple functional domains, J. Bacteriol. 173,
H., VON DOHREN,H., Eds.), 1st Edn., pp. 246- 2354-2365.
287. Berlin, New York: Walter de Gruyter. HABESHAW, J. A., DALGLEISH, A. G., BOUNTIFF,
FLIRI,H. G., BAUMANN, G., ENZ,A., KALLEN,J., L., NEWELL,A. L., WILKS,D., WALKER,L. C.,
LUYTEN,M., MIKOL,V., MOVVA,R., QUES- MANCA,F. (1990), AIDS pathogenesis: HIV en-
NIAUX,V. F. J., SCHREIER, M. H., WALKIN- velope and its interaction with cell proteins, Im-
SHAW, M. D., WENGER, R. M., ZENKE,G., ZU- munol. Today 11,418425.
RINI, M. G. (1993), Cyclosporins. Structure-ac- HAESE, A,, SCHUBERT,M., HERRMANN, A.,
tivity relationships, Ann. N. Y. Acad. Sci. 696, ZOCHER,R. (1993), Molecular characterization
47-53. of the enniatin synthetase gene encoding a mul-
FOXWELL, B. M. J., LING,V., RYFFEL,B. (1989), tifunctional enzyme catalyzing N-methyl-depsi-
Identification of the multidrug resistance related peptide formation in Fusarium scirpi, Mol. Mi-
p-glycoprotein as a cyclosporin binding protein, crobiol. 7, 905-914.
Mol. Pharmacol. 36,543-546. HANDSCHUMACHER, R. E., HARDING,M. W.,
FRANSSON, C., FRESKGARD, P. O., HERBERTSSON, RICE, J., DRUGGE,R. J., SPEICHER,D. W.
H., JOHANSSON, A., JONASSON, P., MARTENS- (1984), Cyclophilin: A specific cytosolic binding
SON, L. G., SVENSSON, M., JONSSON,B. H., protein for cyclosporin A, Science 226, 544-547.
CARLSSON, U. (1992), Cis-trans isornerization is HARRISON, R. K., STEIN,R. L. (199O), Mechanistic
rate-determining in the reactivation of dena- studies of peptidyl prolyl cis-trans isomerase:
tured human carbonic anhydrase I1 as evidenced evidence for catalysis by distortion, Biochemistry
by proline isomerase, FEBS Lett. 2%, 90-94. 29, 1684-1689.
FREY,F. J. (199O), Cyclosporin in autoimmune dis- HASEL,K. W., GLASS,J. R., GODBOUT,M., SUT-
eases, Schweiz. Med. Wochenschr. 120, 772-786. CLIFFE,J. G. (1991), An endoplasmic reticulum-
FRIEDMAN, J., WEISSMAN, I. (1991), Two cytoplas- specific cyclophilin, Mol. Cell Biol. 11, 3484-
mic candidates for immunophilin action are re- 3491.
vealed by affinity for a new cyclophilin: one in HAYANO, T., TAKAHASHI, N., KATO,S., MAKI,N.,
the presence and one in the absence of CsA, Cell SUZUKI,M. (1991). Two distinct forms of pepti-
66,799-806. dylprolyl-cis-trans-isomeraseare expressed sepa-
FRIEDMAN, J., TRAHEY, M., WEISSMAN, I. (1993), rately in periplasmic and cytoplasmic compart-
Cloning and characterization of cyclophilin C-as- ments of Escherichia coli cells, Biochemistry 30,
sociated protein: a candidate natural cellular li- 3041-3048.
gand for cyclophilin C, Proc. Natl. Acad. Sci. HENSENS,0. D., WHITE,R. F., GOEGELMAN, R.
USA 90,6815-6819. T., INAMINE, E. S., PATCHETT,A. A. (1992),
FRUMAN, D. A., KLEE,C. B., BIERER,B. E., Bu- The preparation of [2-deutero-3-fluoro-~-
RAKOFF,s. J. (1992a), Calcineurin phosphatase ala*]cyclosporin A by directed biosynthesis, J.
activity in T lymphocytes is inhibited by FK 506 Antibiot. 45,133-135.
and cyclosporin A, Proc. Natl. Acad. Sci. USA HOFFMANN, K., SCHNEIDER-SCHERZER E., KLEIN-
89,3686-3690. KAUF, H., ZOCHER,R. (1994), Purification and
FRUMAN, D. A., MATHER,P. E., BURAKOFF, S. J., characterization of an eukaryotic alanine racem-
BIERER,B. E. (1992b), Correlation of calcineu- ase, acting as a key enzyme in cyclosporin bio-
rin phosphatase activity and programmed cell synthesis. J. Biol. Chem. 269, 12710-12714.
HOPF,U., NEUHAUS, P., KONIG,V., BAUDITZ,J., T-cell cyclophilin A at 2.5 8, resolution, Proc.
KUTHER, S., SCHMIDT,C. A., STEFFEN, R., Natl. Acad. Sci. USA 88,9483-9487.
BLUMHARDT, G., BECHSTEIN,W. O., NEU- KE, H. M., MAYROSE, D., CAO,W. (1993a), Crys-
HAUS,R. ET AL. (1992), Orthotopic liver trans- tal structure of cyclophilin A complexed with
plantation in hepatic cirrhosis: on the problem of substrate Ala-Pro suggests a solvent-assisted
infection of the transplant with persistent hepati- mechanism of cis-trans isomerization, Proc. Natl.
tis viruses, Z. Gastroenterol. 30,576-582. Acad. Sci. USA 90,3324-3328.
HOPWOOD, D. A., SHERMAN, D. H. (1990), Molec- KE, H. M., ZHAO, Y., Luo, F., WEISSMAN, I.,
ular genetics of polyketides and its comparison FRIEDMAN, J. (1993b), Crystal structure of mu-
to fatty acid biosynthesis, Annu. Rev. Genet. 24, rine cyclophilin C complexed with immunosup-
37-66. pressive drug cyclosporin A, Proc. Natl. Acad.
HORI,K., YAMAMOTO, Y., MINETOKI, T., KUROT- Sci. USA 90, 11850-11854.
su, T., KANDA,M., MIURA,S., OKAMURA, K., KE, H. M., MAYROSE,D., BELSHWA,P. J., AL-
FURUYAMA, J., SAITO, Y. (1989), Molecular BERG,D. G., SCHREIBER, S. L., CHANG,Z.,
cloning and nucleotide sequence of the Grami- ETZKORN,F., Ho, S., WALSH,C. T. (1994),
cidin S synthetase 1 gene, J. Biochem. 1%, 639- Crystal structures of cyclophilin A complexed
645. with cyclosporin A and N-methyl-[(E)-2-bute-
HUNTER,T., KARIN,M. (1992), The regulation of nyl]-4,4-dimethylthreoninecyclosporin A, Struc-
transcription by phosphorylation, Cell 70, 777- ture 2, 33-44.
789. KERN, D., DRAKENBERG, T., WIKSTROM,M.,
IZQUIERDO, M., CANTRELL,D. A. (1992), T-cell FORSEN,S., BANG,H., FISCHER,G. (1993), The
activation, Trends Cell Biol. 2 (9), 268-271. cishans interconversion of the calcium regulat-
JANEWAY, C. A., ROJO,J., SAIZAWA, K., DIAN- ing hormone calcitonin is catalyzed by cyclophil-
ZANI,U., PORTOLES,P., TITE, J., HAQUE,S., in, FEBS Lett. 323, 198-202.
JONES,B. (1989), The Co-receptor function of KIEFFER,L. J., THALHAMMER, T. HANDSCHU-
murine CD4, Immunol. Rev. 109,77-92. MACHER, R.E. (1992), Isolation and characteri-
JIN,Y., ALBERS,M. W., LANE,W. S., BIERER,B. zation of a 40-kDa cyclophilin-related protein, J.
E., SCHREIBER, S. L., BURAKOFF, S. J. (1991), Biol. Chem. 267,5503-5507.
Molecular cloning of a membrane-associated hu- KIEFFER,L. J., SENG,T. W., LI, W., OSTERMAN,
man FK506- and rapamycin-binding protein, D. G., HANDSCHUMACHER, R. E., BAYNEY, R.
FKBP-13, Proc. Natl. Acad. Sci. USA 88, 6677- M. (1993), Cyclophilin-40, a protein with homo-
6681. logy to the P59 component of the steroid recep-
JORDAN, P. M., SPENCER, J. B. (1993), The biosyn- tor complex. Cloning of the cDNA and further
thesis of tetraketides: Enzymology, mechanism, characterization, J. Biol. Chem. 268, 12303-
and molecular programming, Biochem. SOC. 12310.
Trans. 21,222-228. KINO,T., HATANAKA, H., HASHIMOTO, M., NI-
KAHAN,B. D. (1992), Immunosuppressive therapy, SHIYAMA, M., GOTO,T., OKUHARA, M., KOH-
Curr. Opin. Immunol. 4, 553-560. SAKA,M., AOKI,H., IMANAKA, H. (1987), FK-
KALLEN,J., WALKINSHAW, M. D. (1992), The X- 506, A novel immunosuppressant isolated from a
ray structure of a tetrapeptide bound to the ac- Streptomyces. I. Fermentation, isolation, and
tive site of human cyclophilin A, FEBS Lett. 300, physico-chemical and biological characteristics,
286-290. J. Antibiot. 40, 1249-1255.
KALLEN,J., SPITZFADEN,C., ZURINI,M. G., KLEINKAUF, H., VON DOHREN,H. (1987), Biosyn-
WIDER,G., WIDMER,H., WUTHRICH, K., WAL- thesis of peptide antibiotics, Annu. Rev. Micro-
KINSHAW, M. D. (1991), Structure Of human Cy- biol. 41, 259-289.
clophilin and its binding site for cyclosporin A KLEINKAUF, H., VON DOHREN,H. (1990), Nonri-
determined by X-ray crystallography and NMR bosomal biosynthesis of peptide antibiotics, Eur.
spectroscopy, Nature 353, 276-279. J. Biochem. 192, 1-15.
KATZ, L., DONADIO,S. (1993), Polyketide synthe- KO, S. Y., DALVIT,C. (1992), Conformation of cy-
sis - prospects for hybrid antibiotics, Annu. Rev. closporin A in polar solvents, Int. J. Pept. Pro-
Microbiol. 47, 875-912. tein Res. 40, 380-382.
KE, H. M. (1992), Similarities and differences be- KOBEL,H., TRABER,R. (1982), Directed biosyn-
tween human cyclophilin A and other beta-bar- thesis of cyclosporins, Eur. J. Appl. Microbiol.
re1 structures. Structural refinement at 1.63 A re- Biotechnol. 14, 237-240.
solution, J. Mol. Biol. 228, 539-550. KOBEL,H., LOOSLI,H. R.,VOGES,R.(1983), Con-
KE, H. M., ZYDOWSKY, L. D., LIU,J. WALSH,C. T. tribution to knowledge of the biosynthesis of cy-
(1991), Crystal structure of recombinant human closporin A, Experientia 39,873-876.
7 References 587
KOECK,M., KESSLER,H., SEEBACH, D., THALER, LEE,J. P., DUNLAP,B. RICH,D. H. (1990), Synthe-
A. (1992), Novel backbone conformation of cy- sis and immunosuppressive activities of confor-
closporin A: The complex with lithium chloride, mationally restricted cyclosporin lactam ana-
J. Am. Chem. SOC. 114, 2676-2686. logues, Int. J. Pept. Protein Res. 35, 481494.
KOFRON, J. L., KUZMIC,P., KISHORE,V., COLON LEITNER,E., WEBER,G. (1994), Disruption of the
BONILLA,E., RICH, D. H. (1991), Determina- cyclosporin synthetase gene of Tolypocladium
tion of kinetic constants for peptidyl prolyl cis- niveum, Curr. Genet. 26, 461467.
trans isomerases by an improved spectropho- LEITNER, E., SCHNEIDER-SCHERZER, E., SCHOER-
tometric assay [published erratum appears in GENDORFER, K., WEBER, G. (1994), CYClOSpO-
Biochemistry 1991 Nov 5; 30 (44): 108181, Bio- rin synthetase, European Patent Application
chemistry 30, 6127-6134. 578616.
KRAETZSCHMAR, J., KRAUSE,M., MARAHIEL, M. LEIVA,M. C., LYTTLE, C. R. (1992), Leukocyte
A. (1989), Gramicidin S biosynthesis operon chemotactic activity of FKBP and inhibition by
containing the structural genes grsA and grsB FK506, Biochem. Biophys. Res. Commun. 186,
has an open reading frame encoding a protein 1178-1 183.
homologous to fatty acid thioesterase, J. Bacteri- LIU,Y., LINSLEY, P. S. (1992), Costimulation of T-
01. 171, 5422-5429, cell growth, Curr. Opin. Immunol. 4, 265-270.
KRONKE,M., LEONARD,W. J., DEPPER,J. M., LIU,J., ALBERS,M. W., CHEN,C., SCHREIBER, S.
ARYA,F., WONG-STAAL,F., GALLO,R. C., L., WALSH,C. T. (1990), Cloning, expression,
WALDMANN, T. A., GREENE,W. C. (1984), Cy- and purification of human cyclophilin in Escher-
closporin A inhibits T-cell growth factor gene ichia coli and assessment of the catalytic role of
expression at the level of mRNA transcription, cysteines by site-directed mutagenesis, Proc.
Proc. Natl. Acad. Sci. USA 81, 5214-5218. Natl. Acad. Sci. USA 87, 2304-2308.
LALAND,S. G., ZIMMER, T.-L. (1973), The protein LIU,J., FARMER,J. D. JR., LANE,W. S., FRIED-
thiotemplate mechanism of synthesis for the MAN, J., WEISSMAN, I., SCHREIBER, S. L. (1991).
peptide antibiotics produced by Bacillus brevis, Calcineurin is a common target of cyclophilin-
Essays Biochem. 9, 31-42. cyclosporin A and FKBP-FK506 complexes,
LANDSCHULTZ,W. H., JOHNSON, P. F., Cell 66,807-815.
MCKNIGHT, S. L. (1989), The DNA binding do- LOOSLI,H. R., KESSLER,H., OSCHKINAT, H., WE-
main of the rat liver nuclear protein C/EBP is BER,H. P., PETCHER,T. (1985), The conforma-
bipartite, Science 243, 1681-1688. tion of cyclosporin A in the crystal and in solu-
LAWEN,A., TRABER,R. (1993), Substrate specific- tion, Helv. Chim. Acta 68, 682-704.
ities of cyclosporin synthetase and peptolide LUBAN,J., BOSSOLT,K. L., FRANKE, E. K., KAL-
SDZ 214-103 synthetase. Comparison of the sub- PANA,G. V., GOFF, S. P. (1993), Human immu-
strate specificities of the related multifunctional nodeficiency virus type 1 Gag protein binds to
polypeptides, J. Biol. Chem. 268,20452-20465. cyclophilins A and B, Cell 73, 1067-1078.
LAWEN,A., ZOCHER,R. (1990), Cyclosporin syn- LYNEN,F. (1980), On the structure of fatty acid
thetase. The most complex peptide synthesizing synthase of yeast, Eur. J. Biochem. 112, 431-
multienzyme polypeptide so far described, J. 442.
Antibiot. 42, 1283-1289. MACCABE,A. P., VAN LIEMPT,H., PALISSA,H.,
LAWEN,A., TRABER,R., GEYL,D., ZOCHER,R., RIACH,M. B. R., PFEIFER,E., VON DOHREN,
KLEINKAUF, H. (1989), Cell-free biosynthesis of H., KINGHORN, J. R. (1991), S(L-a-aminoadi-
new cyclosporins, J. Antibiot. 42, 1283-1289. pyl)-L-cysteinyl-D-valine synthetase from Asper-
LAWEN,A., TRABER, R., GEYL,D. (1991), In vitro gillus nidulans, J. Biol. Chem. 266, 12646-
biosynthesis of [Thr2,Leus,D-Hiv8,Leu'o]-cy- 12954.
closporin A, a cyclosporin-related peptolide, MARTEL,R. R., KLICIUS, J., GALET,S. (1977), In-
with immunosuppressive activity by a multien- hibition of the immune response by rapamycin, a
zyme polypeptide, J. Biol. Chem. 266, 15567- new antifungal antibiotic, Can. J. Physiol. Phar-
15 570. macol. 55,48-51.
LAWEN,A., D I ~ M A NJ., N SCHMIDT,
, B., RIESNER, MASON,J. (1989), Pharmacology of cyclosporine
D., KLEINKAUF, H. (1992), Enzymatic biosyn- (Sandimmune) VII. Pathophysiology and toxi-
thesis of cyclosporin A and analogues, Biochimie cology of cyclosporine in humans and animals,
74,511-516. Pharmacol. Rev. 42,423-434.
LAWEN,A., TRABER,R., REUILLE,R., PONELLE, MCCAFFREY, P. G., PERRINO, B. A,, SODERLING,
M. (1994), In vitro biosynthesis of ring-extended T. R., RAO,A. (1993), NF-ATp, a T lymphocyte
cyclosporins, Biochem. J. 300,395-399. DNA-binding protein that is a target for calci-
neurin and immunosuppressive drugs, J. Biol. PATCHETT,A. A., TAUB, D., HENSENS,0. D.,
Chem. 268,3747-3752. GOEGELMAN, R. T., YANG,L. H., DUMONT,F.
MIKOL,V., KALLEN,J., PFLUEGL,G., WALKIN- J., PETERSON,L., SIGAL,N. H. (1992). Ana-
SHAW, M. D. (1993), X-ray structure of a mon- logues of cyclosporin A modified at the D-Ala8
omeric cyclophilin A-cyclosporin A crystal composition, J. Antibiot. 45, 94-102.
plex at 2.1 8, resolution, J. Mol. Biol. 234, 1119- PFLUEGL,G., KALLEN,J., SCHIRMER, T., JANSON-
1130. IUS,J. N., ZURINI,M. G., WALKINSHAW, M. D.
MIKOL, V., KALLEN,J., WALKINSHAW, M. D. (1993), X-ray structure of a decameric cyclophil-
(1994~),X-ray structure of a cyclophilin B/cy- in-cyclosporin crystal complex, Nature 361, 91-
closporin complex: comparison with cyclophilin 94.
A and delineation of its calcineurin binding do- PRICE,E. R., ZYDOWSKY, L. D., JIN,M. J., BAKER,
main. Proc. Natl. Acad. Sci. USA 91, 5183- C. H., MCKEON,F. D., WALSH,C. T. (1991),
5186. Human cyclophilin B: A second cyclophilin gene
MIKOL, V., KALLEN,J., WALKINSHAW, M. D. encodes a peptidyl-prolyl isomerase with a signal
(1994b), The X-ray structure of (MeBm2t)'-cy- sequence, Proc. Natl. Acad. Sci. USA 88, 1903-
closporin complexed with cyclophilin A provides 1907.
an explanation for its anomalously high immu- PRICE,D. J., GROVE,J. R., CALVO,V., AVRUCH,
nosuppressive activity, Protein Eng. 7 (5), 597- J., BIERER,B. E. (1992), Rapamycin-induced in-
603. hibition of the 70-kilodalton S6 proteinkinase,
MORRIS,R. E. (1993), New small molecule immu- Science 257,973-917.
nosuppressants for transplantation: review of es- QUESNIAUX, V. F. J., SCHREIER,M. H., WENGER,
sential concepts, J. Heart Lung Transplant. 12, R. M., HIESTAND, P. C., HARDING, M. W., VAN
S275-S286. REGENMORTEL, M. H. V. (1987), Cyclophilin
NERI, P., GEMMECKER, G., ZYDOWSKY,L. D., binds to the region of cyclosporine involved in
WALSH,C. T., FESIK,S. W. (1991), NMR studies its immunosuppressive activity, Eur. J. Immunol.
of [U-(13)C]cyclosporin A bound to human cy- 17, 1359-1365.
clophilin B, FEBS Lett. 290, 195-199. QUESNIAUX, V. F. J., SCHREIER, M. H., WENGER,
O'HAGAN,D. (1991), The Polyketide Metabolites. R. M., HIESTAND,P. C., HARDING, M. W., VAN
(MELLOR,J., Ed.). Chichester: Ellis Horwood REGENMORTEL, M. H. V. (1988), Molecular
Series in Organic Chemistry. characteristics of cyclophilin-cyclosporine inter-
O'KEEFE,S. J., TAMURA, J., KINCAID, R. L., Toc- action, Transplantation 46, 23-28.
CI, M. J., O'NEILL, E. A. (1992), FK-506- and QUESNIAUX, V. F. J., WEHRLI,S., WIOLAND,C.,
CsA-sensitive activation of the interleukin-2 SCHULER,W., SCHREIER, M. H. (1994), Effects
promoter by calcineurin, Nature 357, 692-694. of rapamycin on hematopoiesis, Transplant.
OFFENZELLER, M., Su, Z., SANTER,G., MOSER, Proc. 26, 3135-3140.
H., TRABER,R., MEMMERT,K., SCHNEIDER- RICH,D. H., SUN,C. Q., GUILLAUME, D., DUN-
SCHERZER,E. (1993), Biosynthesis of the un- LAP, B., EVANS,D. A., WEBER,A. E. (1989),
usual amino acid (4R)-4-[(E)-2-buteny1]-4-me- Synthesis, biological activity, and conformation-
thyl-L-threonine of cyclosporin A. Identification a1 analysis of (25, 3R, 4s)-MeBmt '-cyclosporin,
of 3(R)-hydroxy-4(R)-methyl-6(E)-octenoic acid a novel 1-position epimer of cyclosporin A, J.
as a key intermediate by enzymatic in vitro syn- Med. Chem. 32, 1982-1987.
thesis and by in vivo labeling techniques, J. Biol. ROBINSON,A. (1991), Polyketide synthase com-
Chem. 268,26 127-26 134. plexes: Their structure and function in antibiotic
PAPAGEORGIOU, C., FLORINETH, A., FRENCH,R. biosynthesis, Philos. Trans. R. Soc. London B
(1994), Calcineurin has a very tight-binding 332,107-114.
pocket for the side chain of residue 4 of cyclo- ROSEN,M. K., STANDAERT, R. F., GALAT,A., NA-
sporin, Bioorg. Med. Chem. Lett. 4, 267-272. KATSUKA, M., SCHREIBER, s. L. ( l w ) , Inhibi-
PARK,S. B., MEIER,G. P. (1989), A semisynthetic tion of FKBP rotamase activity by immunosup-
approach to olefinic analogues of amino acid pressant FK506: twisted amide surrogate,
one (MeBmt) in cyclosporin A, Tetrahedron Science 248, 863-866.
Lett. 30,42154218. ROSENWIRTH, B., BILLICH,A., DATEMA,R., Do-
PARK,S. T., ALDAPE,R. A., FUTER,O., DECEN- NATSCH, P., HAMMERSCHMID, F., HARRISON,
zO, M. T., LIVINGSTON, D. J. (1992), PPIase ca- R. K., HIESTAND,P. C., JAKSCHE,H., MAYER,
talysis by human FK506-binding protein pro- P., PEICHL,P., QUESNIAUX, V. F. J., SCHATZ,
ceeds through a conformational twist mecha- F., WENGER,R. M., WOLFF,B., ZENKE,G., Zu-
nism, J. Biol. Chem. 267, 3316-3324. RINI, M. G. (1994), Inhibition of HIV-1 replica-
7 References 589
tion by SDZ NIM 811, a non-immunosuppres- SEEBACH,D., BECK,A. K., BOSSLER,H. G., GER-
sive cyclosporin A analogue. Antimicrob. Agents BER, c., KO, s. Y., MURTIASHAW, c. W., NAEF,
Chemother. 38, 1763-1772. R., SHODA,S., THALER, A., WENGER,R. M.
RYFFEL,B. (1989), Pharmacology of cyclosporine. (1993), Modification of cyclosporine-A (CS) -
VI. Cellular activation: Regulation of intracellu- Generation of an enolate at the sarcosine resi-
lar events by cyclosporine, Pharmacol. Rev. 41, due and reactions with electrophiles, Helv.
407-423. Chim. Acta 76,1564-1590.
RYFFEL,B., GOETZ, U., HEUBERGER, B. (1982), %!=TON,B. M., CAMPBELL, M. A. (1991),The role
Cyclosporin receptors on human lymphocytes, J. of tyrosine protein phosphorylation in lympho-
Immunol. 129,1978-1982. cyte activation, Annu. Rev. Cell Biol. 7 , 257-
SANGLIER, J. J., TRABER,R., BUCK,R. H., HOF- 274.
MA”, H., KOBEL, H. (1990), Isolation of SENN,H., WEBER, C., KOBEL,H., TRABER,R.
(4R)-4-[(E)-2-butenyl]-4-methyl-~-threonine, the (1991), Selective 13C-labellingof cyclosporin A,
characteristic structure element of cyclosporins, Eur. J. Biochem. 199,653-658.
from a blocked mutant of Tolypocladium infla- SHERRY,B., YARLETT,N., STRUPP,A., CERAMI,
tum, J. Antibiot. 43,707-714. A. (1992), Identification of cyclophilin as a
SCHLUMBOHM, W., STEIN,T., ULLRICH,C., VAT- proinflammatory secretory product of lipopoly-
ER, J., MARAHIEL, M. A., KRUFT,V., WITT- saccharide-activated macrophages, Proc. Natl.
MANN-LIEBOLD, B. (1991), An active serine is Acad. Sci. USA 89, 3511-3515.
involved in covalent substrate amino acid bind- SIEKIERKA, J. J., HUNG,S. H., POE, M., LIN,C. S.,
ing at each reaction center of gramicidin S syn- SIGAL,N. H. (1989a),A cytosolic binding pro-
thetase, J. Biol. Chem. 266,23135-23 141. tein for the immunosuppressant FK506 has pep-
SCHMIDT,B., RIESNER,D., LAWEN,A., KLEIN- tidyl-prolyl isomerase activity but is distinct from
KAuF, H. (1992), Cyclosporin synthetase is a
cyclophilin, Nature 341,755-757.
1.4MDa multienzyme polypeptide. Re-evalua-
tion of the molecular mass of various peptide SIEKIERKA, J. J., STARUCH,M. J., HUNG,S. H. Y.,
synthetases, FEBS Lett. 307,355-360. SIGAL,N. H. (1989b), FK-506, a potent novel
immunosuppressive agent, binds to a cytosolic
SCHNEIDER, H., CHARARA,N., SCHMITZ,R., protein which is distinct from the cyclosporin A-
WEHRLI,S., MIKOL,V., ZURINI,M. G., QUES- binding protein, cyclophilin, J. Immunol. 143,
NIAUX,V. F. J., MOVVA,N. R. (1994), Human 1580-1583.
cyclophilin C: Primary structure, tissue distribu-
tion, and determination of binding specificity for SIGAL,N.H., DUMONT,F. J., DUREITE, P. L., SIE-
cyclosporins. Biochemistry 33,8218-8224. KIERKA, J. J., PETERSON,L., RICH,D. H., DUN-
SCHOENBRUNNER, E. R., MAYER,S., TROPSCHUG, LAP, B. E., STARUCH,M. J., MELINO,M. R.,
M., FISCHER,G., TAKAHASHI, N., SCHMID,F. KOPRAK,S. L. ET AL. (1991),Is cyclophilin in-
X. (1991), Catalysis of protein folding by cyclo- volved in the immunosuppressive and nephro-
philins from different species, J. Biol. Chem. 266, toxic mechanism of action of cyclosporin A? J.
3630-3635. Exp. Med. 173,619-628.
SCHREIBER, S.L. (1991),Chemistry and biology of SIMPSON,T. J. (1989), Biological chemistry. Part
the immunophilins and their immunosuppressive (11) biosynthesis, Annu. Rep. Prog. Chem. Sect.
ligands, Science 251,283-287. B 85,321-351.
SCHREIBER, S . L., CRABTREE, G. R. (1992),The SMITH,D. J., EARL,A. J., TURNER, G. (1990),The
mechanism of action of cyclosporin A and multifunctional peptide synthetase performing
FK506, Imrnunol. Today 13,136-142. the first step of penicillin biosynthesis in Penicil-
SCOTT-CRAIG,J. S., PANACCIONE,D. G., Po- lium chrysogenum is a 421073 Dalton protein
CARD,J.-A., WALTON,J. D. (1992), A large, similar to Bacillus brevis peptide antibiotic syn-
multifunctional cyclic peptide synthetase catalyz- thetases, EMBO J. 9,2743-2750.
ing HC-toxin production in the plant pathogenic SODA,K., TANAKA, H., TANIZAWA, K. (1986),Py-
fungus Cochliobolus carbonum encoded by a ridoxyl phosphate enzymes catalyzing racemiza-
15.7-kb open reading frame, J. Biol. Chem. 267, tion, in: Vitamin B6, Pyridoxyl Phosphate:
26044-26049. Chemical, Biochemical and Medical Aspects.
SEEBACH,D., KO, S. Y., KESSLER,H., KOECK,M., Coenzymes and Cofactors (DOLPHIN, D., POUL-
REGGELIN, M., SCHMIEDER, P., WALKINSHAW, SON, R., AVRAMOVIC, O., Eds.), pp. 223-251.
M. D., BOELSTERLI,H. J., BEVEC,D. (1991), New York: Wiley & Sons.
Thiocyclosporins: Preparation, solution and SPITZFADEN,C., WEBER, H. P., BRAUN, W.,
crystal structure, and immunosuppressive activi- KALLEN,J., WIDER,G., WIDMER,H., WALKIN-
ty, Helv. Chim. Acta 74,1953-1990. SHAW, M. D., WUTHRICH, K. (1992),Cyclospo-
rin A-cyclophilin complex formation. A model [MeIle4]cyclosporin,a novel natural cyclosporin

based on X-ray and NMR data, FEBS Lett. 300, with anti-HIV activity: Structural elucidation,
291-300. biosynthesis, and biological properties, Antiviral
STABEN,C., JENSEN,P. B., SINGER,M., POLLOCK, Chem. Chemother. 5,331-339.
J., SCHECHTMAN, M., KINSEY,J., SELKER,E. TSANG, V., HODSON, M. E., YACOUB,M. H.
(1989), Use of a bacterial hygromycin resistance (1992), Lung transplantation for cystic fibrosis,
gene as a dominant selectable marker in Neuro- Br. Med. Bull. 48,949-971.
spora crassa transformation, Fung. Genet. News- TURGAY,K., KRAUSE, M., MARAHIEL,M. A.
lett. 36, 79-81. (1992), Four homologous domains in the prima-
STAMNES, M. A,, RUTHERFORD, S. L., ZUKER,C. ry structure of GrsB are related to domains in a
S. (1992), Cyclophilins: A new family of proteins superfamily of adenylate-forming enzymes, Mol.
involved in intracellular folding, Trends Cell Microbiol. 6, 529-546.
Biol. 2, 272-276. TWENTYMAN, P. R. (1992), Cyclosporins as drug
STEVENSON, M., STANWICK, T. L., DEMPSEY,M. resistance modifiers, Biochem. Pharmacol. 43,
P., LAMONICA, C. A. (1990), HIV-1 replication 109-117.
is controlled at the level of T cell activation and VOLLENBROICH, D., KLUGE, B., D’SOUZA, C.,
proviral integration, EMBO J. 9, 1551-1560. ZUBER,P., VATER,J. (1993), Analysis of a mu-
SWANSON,S. K., BORN, T., ZYDOWSKY,L. D., tant amino acid-activating domain of surfactin
CHO, H., CHANG,H. Y., WALSH,C. T. RUSNAK, synthetase bearing a serine-to-alanine substitu-
F. (1992), Cyclosporin-mediated inhibition of tion at the site of carboxyl thioester formation,
bovine calcineurin by cyclophilins A and B, FEBS Lett. 325, 220-224.
Proc. Natl. Acad. Sci. USA 89,3741-3745.
VON WARTBURG, A., TRABER,R. (1988), Cyclo-
TAI,P., ALBERS,M. W., CHANG,H., FABER,L. E., sporins. Fungal metabolites with immunosup-
SCHREIBER, S . L. (1992), Association of a 59- pressive activities, Progr. Med. Chem. 25 (ELLIS,
kilodalton immunophilin with the glucocorticoid G. P., WEST,G. B., Eds.), pp. 1-33. Amsterdam:
receptor complex, Science 256, 1315-1318. Elsevier Science Publishers, B.V., Biomedical
TAKAHASHI, N., HAYANO, T., SUZUKI,M. (1989), Division.
Peptidyl-prolyl cis-trans isomerase is the cyclo-
sporin A-binding protein cyclophilin, Nature WALSH,C. T., BADET,R., DAUB,E., ESAKI,N.,
337,473-475. GALAKATOS,N. (1985), Bacterial alanine ra-
TAKESHITA, T., ASAO,H., OHTANI,K., ISHII,N., cemases: Targets for antibacterial agents, Spec.
Publ. - R. SOC.Chem. 55 (SCI-RSC Med. Chem.
KUMAKI, S., TANAKA, N., MUNAKATA, H., NA-
Symp. 3rd), 193-209.
KAMURA, M., SUGAMURA, K. (1992), Cloning
of the gamma chain of the human IL-2 receptor, WEBER, C., WIDER,G., VON FREYBERG, B., TRA-
Science 257,379-382. BER,R., BRAUN,W., WIDMER,H., WUTHRICH,
THERIAULT, Y., LOGAN,T. M., MEADOWS,R., K. (1991), The NMR structure of cyclosporin A
Yu, L., OLEJNICZAK, E. T., HOLZMAN,T. F., bound to cyclophilin in aqueous solution, Bio-
SIMMER,R. L., FESIK,S. W. (1993), Solution chemistry 30,65634574,
structure of the cyclosporin Ncyclophilin com- WEBER,G., SCHOERGENDORFER, K., SCHNEIDER-
plex by NMR, Nature 361, 88-91. SCHERZER,E., LEITNER, E. (1994), The peptide
THOMMEN-SCOTT, K. (1981), Antimalarial activity synthetase catalyzing cyclosporin production in
of cyclosporin A, Agents Actions 11,770-773. Tolypocladium niveum is encoded by a giant
TOCCI,M. J., MATKOVICH, D. A., COLLIER,K. A., 45.8-kilobase open reading frame, Curr. Genet.
KWOK,P., DUMONT,F. J., LIN, S., DEGUDICI- 26,120-125.
BUS, S., SIEKIERKA, J. J., CHIN,J., HUTCHIN- WECKERMANN, R., FUERBASS,R., MARAHIEL, M.
SON, N. I. (1989), The immunosuppressant A. (1988), Complete nucleotide sequence of the
FK.506 selectively inhibits expression of early T tycA gene encoding the tyrocidine synthetase 1
cell activation genes, J. Immunol. 143, 718-726. from Bacillus brevis, Nucleic Acids Res. 16,
TRABER,R., HOFMANN,H., LOOSLI,H. R., PON- 11841.
ELLE, M., VON WARTBURG,A. (1987), Neue WENGER,R. M. (1983), Synthesis of cyclosporine,
Cyclosporine aus Tolypocladium inflatum. Die Helv. Chim. Acta 67, 502-525.
Cyclosporine K-Z, Helv. Chim. Acta 70, 13-36. WENGER,R. M. (1989), Pharmacology of cyclospo-
TRABER,R., HOFMANN, H., KOBEL,H. (1988), Cy- rin (Sandimmune). 11. Chemistry, Pharmacol.
closporins - new analogues by precursor di- Rev. 41, 243-247.
rected biosynthesis, J. Antibiot. 42, 591-597. WENGER,R. M. (1986), Cyclosporine and ana-
TRABER,R., KOBEL,H., LOOSLI,H. R., SENN,H., logues: Structural requirements for immunosup-
ROSENWIRTH, B., LAWEN, A. (1994), pressive activity, Transplant. Proc. 18, 213-218.
7 References 591
WENGER,R. M., PAYNE,T. G., SCHREIER,M. H. ZACK,J. A., ARRIGO,S. J., WEITSMAN,S. R., Go,
(1986), Cyclosporine: Chemistry, structure-ac- A. S., HAISLIP,A., CHEN,I. S. Y. (1990), HIV-
tivity relationships and mode of action, Progr. entry into quiescent primary lymphocytes: mo-
Clin. Biochem. Med. 3, 157-191. lecular analysis reveals a labile, latent viral struc-
WENGER,R. M., FRANCE,J., BOVERMANN, G., ture, Cell 61, 213-222.
WALLISER, L., WIDMER, A., WIDMER,H.
(1994), The 3D structure of a cyclosporin ana- ZENKE,G., BAUMANN, G., WENGER,R. M., HIES-
logue in water is nearly identical to the cyclo- TAND,P. C., QUESNIAUX, V. F. J., ANDERSEN,
philin-bound conformation, FEBS Lett. 340, E., SCHREIER,M. H. (1993), Molecular mecha-
255-259. nisms of immunosuppression by cyclosporins,
WITZEL, B. (1989), New cyclosporin analogues Ann. N. Y. Acad. Sci. 685, 330-335.
with modified “C-9 amino acids”, US Patent no.
4798823. ZOCHER, R., NIHIRA,T., PAUL, E., MADRY,N.,
Xu, Q., LEIVA,M. C., FISCHKOFF, S . A., HAND- PEETERS, H., KLEINKAUF, H. (1986), Biosyn-
SCHUMACHER, R. E., LYTTLE, C. R. (1992), thesis of cyclosporin A: Partial purification and
Leukocyte chemotactic activity of cyclophilin, J. properties of a multifunctional enzyme from To-
Biol. Chem. 267, 11968-1 1971. Iypocladium inflatum, Biochemistry 25, 550-553.
Biotechnology
l3 Secondary Products from Plant

Cell Cultures
JOCHEN BERLIN
Braunschweig, Federal Republic of Germany
1 Introduction 595
2 Some General Conclusions and Suggestions Based on the Present Biotechnological Impact
of Plant Cell Cultures as Producers 596
3 Secondary Product Formation in Suspension Cultures 598
3.1 Products Accumulating at High or Good Levels in Suspension Cultures 598
3.1.1 Cinnamic Acid Derivatives 598
3.1.2 Naphthoquinones and Anthraquinones 601
3.1.3 Protoberberines and Benzophenanthridine Alkaloids 603
3.1.4 Monoterpene Indole Alkaloids 606
3.1.5 Anthocyanins and Betalains 609
3.1.6 Steroidal Compounds 611
3.1.7 Immunologically Active Polysaccharides 612
3.2 Products of Commercial Interest Accumulating in Traces or not at all in Suspension
Cultures 612
3.2.1 Morphinan Alkaloids 613
3.2.2 Tropane Alkaloids 613
3.2.3 Quinoline Alkaloids 614
3.2.4 Antitumor Compounds 614
3.2.5 Cardiac Glycosides 615
3.2.6 Vanillin and Vanilla Aroma 615
4 Secondary Product Formation in Hairy Root Cultures 619
5 Plant Tissue Cultures as a Source of New Chemicals? 621
6 Biotransformations with Cultured Plant Cells 621
6.1 Arbutin 622
6.2 Biotransformation of Cardiac Glycosides 622
7 Metabolic Engineering of Secondary Pathways in Cultured Cells 623
7.1 Serotonin Biosynthesis in Peganum harmala 624
7.2 Affecting Nicotine Alkaloid Biosynthesis in Tobacco 624
594 13 Secondary Products from Plant Cell Cultures
7.3 Enhancing Scopolamine Production in Atropa belladonna 625

7.4 Does Genetic Engineering Improve the Potential of Plant Tissue Cultures as Producers
of Interesting Metabolites? 626
8 Conclusions and Outlook 627
9 References 628
1 Introduction 595
1 Introduction tabolites have been produced from some

plant cell cultures it seems obvious to suggest
that commercially interesting compounds
Higher plants produce a great variety of might be produced in a factory type produc-
secondary metabolites, some of which are an tion in large bioreactors, such as those used in
indispensable source of commercially impor- the production of microbial drugs. The advo-
tant compounds. These include pharmaceuti- cates of the new technology have greatly
cals (such as steroids, alkaloids, and gluco- overestimated the problems with agricultural-
sides), natural flavors, fragrances, dyes, and ly grown plant material while the biological,
gums (such as natural rubber). As only a technical, legal, and financial problems of tis-
small portion of plants have as yet been ana- sue culture technology have been generally
lyzed for secondary products, it is not surpris- underestimated. Interestingly, advocates of
ing that isolation and characterization of new the tissue culture technology have come
compounds continue unabatedly. The array mainly from universities and public research
of assays for detecting biological activities of centers, while in industry (perhaps with the
natural compounds has been enlarged and exception of some Japanese industries) the
simplified, and has been paralleled by an in- interest was in general very meager world-
crease in assay sensitivity. Thus, the screening wide.
of plant extracts for new biologically active Thus, one can name immediately two rea-
compounds still seems to be a promising ap- sons why the tissue culture technology was in
proach today (BALANDRIN et al., 1985). a difficult position from the beginning: Firstly,
However, it should also be mentioned that the lack of new plant compounds with very
the screening of some 40000 plants by the special, extraordinary pharmacological char-
NCI (National Cancer Institute, USA) acteristics for which it would be worth-while
yielded only three really novel anticancer developing novel tissue culture processes; sec-
compounds (vincristine, vinblastine, and tax- ondly, the difficulty of replacing an estab-
01). It is doubtful whether this result helps to lished and legally approved production pro-
attract more industrial investments to this cess by a new technique. It is indeed difficult
area of plant research. Five Rauwolfia alka- to replace a conventional, approved produc-
loids with tranquilizing and antihypertensive tion process based on the extraction of field-
activities and the three anticancer substances grown plants by a plant tissue culture produc-
were the only plant-derived substances with tion process. This could only happen if the
therapeutic efficacy and utitility, proven calculation of costs showed an extreme ad-
beyond doubt (TYLER, 1988), which have vantage for tissue cultures, and if the proce-
been introduced to the American market dur- dure of obtaining a new legal approval re-
ing the last 40 years. All other plant-derived quired for a product from a new source is not
compounds used as drugs today have been too time-consuming and expensive. If a new
known for a much longer time and their pro- product is under consideration, the superiori-
duction is well established. Thus, a general ty of the new drug must be clear and evident
concern of pharmacognosists is how to make to justify the long and expensive procedures
plant analyses more efficient with respect to needed nowadays for introducing a new phar-
drug development (TYLER,1988). maceutical to the market. In view of these
If the plant product under consideration two obstacles we have to leave it to industry
cannot be synthesized chemically it has to be and biotechnological companies to decide
isolated from wild or field-grown plants. Con- whether they consider production by tissue
sequently, plants synthesizing products in cultures as an attractive alternative. For each
high demand are usually grown in large-scale product under consideration the decision may
plantations. Since the establishment of cell be quite different from company to company
cultures of any plant species is nowadays rou- and from country to country.
tine in most cases, plant cell biomass can also Updates of almost all the products found in
be produced in huge fermentors. Since in ad- plant tissue cultures were published by ELLIS
dition, reasonable amounts of secondary me- (1988), CONSTABELand VASIL (1988), and
BANTHORPE(1994). However, for many of cultures and the various culture types and
the compounds mentioned in these reviews it techniques in great detail in Vol. 1 of this
is not readily clear whether they are of any multi-volume comprehensive treatise. Due to
biotechnological importance. Therefore, I will their recent review and those of other (PARR,
concentrate on some groups of compounds 1989; CHARLWOODand RHODES, 1990;
about which sufficient information is avail- PAYNEet al., 1991; BUITELAAR et al., 1992;
able in order to evaluate the biotechnological BANTHORPE, 1994; MISAWA,1994) it is not
impact of plant cell cultures as producers. The necessary to repeat here the general aspects
main objective of my previous review (BER- and specific characteristics of plant cell cul-
LIN, 1986) was to describe the experimental tures and plant secondary metabolism out-
methods by which high yielding cultures were lined in my previous article (BERLIN,1986).
obtained and to demonstrate that the same However, before reviewing individual culture
techniques that could be applied to some systems and documenting the state of the art,
pathways failed for others. Thus the aim was I would like to make some general statements
to provide the reader with a better under- which may be regarded as suggestions for fu-
standing of what is possible today and what ture research programs and which gain sup-
might become possible tomorrow. port from the analyses of the individual sys-
In this updated review I will expand on this tems.
aspect, explaining why certain compounds are As indicated in Sect. 1, plant tissue cultures
easily produced in cultured cells while other are presently not well accepted as a source of
metabolites have proved recalcitrant to all commercially interesting products. Indeed,
known techniques for improved production. the impact of tissue culture technology has
As I will describe mainly the same groups of not increased, but rather decreased despite all
compounds as in 1986 in the first edition of efforts and scientific progress. Some think
“Biotechnology”, the extent of progress with that this technique is a futile approach; others
respect to production levels during the last 10 believe that the scientific breakthrough has
years will become apparent. Some com- not yet been achieved for a true evaluation of
pounds have been replaced by others which the potential of plant cell cultures as produc-
are presently of more biotechnological inter- ers. This prevents a larger engagement of
est; the new developments are included in this companies. Thus, a critical review should ana-
review. The possibility of altering production lyze in which areas research efforts must be
characteristics of pathways by genetic engi- intensified and identify which approaches can
neering opened a new area. This will be ana- be reduced today. The analysis of the field
lyzed critically to whether it will improve the after 20 years of rather intensive research al-
situation of tissue cultures as producers of lows indeed some clear conclusions to be
commercially interesting compounds. made about promising directions and about
what should no longer be tried.
Today it is possible to analyze the scientific
usefulness of systems for improving the gen-
2 Some General eral standing of plant tissue cultures as tools
of biotechnology. The term “scientific useful-
Conclusions and ness” means that biotechnologically relevant
studies must not be restricted to cultures of
Suggestions Based on the plant species synthesizing commercially im-
Present Biotechnological portant compounds. Meaningful studies on
model systems include studies on the regula-
Impact of Plant Cell tion and expression of metabolic pathways
and the possibilities of their manipulation.
Cultures as Producers These are presently at least as important for
the future of this field as working on so-called
PETERSENand ALFERMANN (1993) have commercially attractive pathways. The seem-
already described special features of plant cell ingly most attractive pathways yielding com-
2 Some General Conclusions and Suggestions Based on the Present Biotechnological Impact 597
mercially useful products can presently not be ences. If this is a biotechnologically relevant
expressed very well in morphologically undif- finding, e.g., a true variant line, similar lines
ferentiated cells, and thus such cultures are can easily be isolated by other researchers by
not very suitable for biochemical or molecu- employing a corresponding screening, selec-
lar studies. tion, or an extended culture initiation pro-
In the past the efforts of most tissue culture gram (BERLIN,1988). If seemingly unique
groups were aimed at establishing highly pro- lines with surprisingly good or different pro-
ductive plant cell cultures by conventional duction characteristics cannot be detected by
techniques. The product level was the most other groups despite all efforts it is certainly a
important goal. These studies included the in- transient trait, not useful for any biotechno-
itiation of many individual cultures from dif- logical purposes. Important and stable pro-
ferent explants of one or more plants, of one duction improvements are usually confirmed
or more species or varieties. The cultures by independent laboratories.
were grown on media with different phyto- As pointed out, some pathways are sponta-
hormone compositions. Screening or selec- neously well expressed in cultured cells allow-
tion, as well as trials for enhancing productiv- ing product formation rates in the range of
ity by media variation or by the use of induc- g L-' in a rather short culture period. A pre-
er compounds have been part of a good pro- requisite for such biotechnologically relevant
gram for optimizing a culture system. The ap- levels is that product formation occurs in rap-
plication of the same techniques led in some idly growing cells, parallels growth, or is eas-
cases to product levels in the range of g L-' ily induced in a production medium. If a well
within a few days, while in other cases expressed product is of commercial interest
-'
amounts from zero to a few pg L were pro- (shikonin, berberines), industry can substan-
duced within a month. Some products were tially improve the production rates in shake
generally found at high levels while others flasks or small fermentors by optimizing all
were always found at low levels or even lack- process parameters. If industry has taken
ing in a culture. These findings were con- over a process, it does not make much sense
firmed by laboratories from all over the for research groups to continue with that cul-
world. Indeed, most tissue cultures of one ture system in shake flask systems or small
plant species, after full adaptation to the cul- bioreactors if the only aim is to demonstrate
ture conditions, have very similar production that another medium or elicitor, or an altered
characteristics, independent of the laboratory fermentation protocol improves productivity.
where they were established. The extent of Unfortunately, most products found at bio-
expression of a pathway is usually not re- technologically relevant levels in cultured
stricted to one plant species. If, for example, a cells are not of great commercial importance.
pathway is highly repressed in the cultures of Thus, it is even more questionable to spend
one plant species, it is most likely that cul- too much time on further productivity im-
tures of other plant species containing the provements for such compounds.
same or related pathways will also be poor Nowadays, the main question should be
producers. If there are clear indications that a whether a previously established product ex-
certain pathway is poorly expressed in cul- pression level, or its manipulation by chang-
tured cells, it is unlikely that one can change ing the culture conditions, allows meaningful
this through the analyses of many more indi- biochemical and molecular studies. Without
vidually established cultures from various ex- knowledge of the enzymes involved in the
plants, by the application of screening and se- pathways, without recognizing the rate-limit-
lection, by media variation, or by the use of ing steps within complex and branched path-
elicitors (BERLIN,1988). Thus, it does not ways, and without identifying the factors con-
make sense to waste time with such poorly trolling the often organ-specific expression of
producing cultures, unless new approaches pathways, it is unlikely that plant tissue cul-
can be applied. There are some reports in the ture based production processes (especially in
literature claiming production characteristics view of the general obstacles discussed in
which seem to contradict all other experi- Sect. 1) will more often be considered as an
attractive alternative. From this point of view, has been gathered for an evaluation. Subse-
all research which aims to elucidate specific quently, I will give an overview of the seem-
or general features of pathway regulation ingly commercially most important pathways
with the ultimate goal of manipulating path- which have remained recalcitrant to improve-
way expression is important in keeping the ments despite all efforts. For information
tissue culture technology alive. Thus, it is about pathways not mentioned here the read-
hoped that more and more biotechnologically er is referred to 34 special reviews in the book
orientated groups use their established tissue of CONSTABEL and VASIL(1988) which co-
culture systems for such purposes during the vers nearly all groups of secondary com-
next decade. pounds found in plant tissue cultures.
3.1 Products Accumulating at

3 Secondary Product High or Good Levels in
Formation in Suspension Suspension Cultures
Cultures 3.1.1 Cinnamic Acid Derivatives
Well-growing cell suspension cultures have Esters and amides of cinnamic acids, main-
been regarded for a long time as the only rel- ly of caffeic acid, form a major, widespread
evant system for biotechnological production group of phenylpropanoid metabolites in
processes. Most recently, rapidly growing hairy plants. This group of compounds accumulates
root cultures with their root-specific pro- spontaneously, often at high levels, in cell sus-
duction characteristics have also been grown pension cultures. Two groups independently
in larger bioreactors (CURTIS,1993). There reported the accumulation of high levels of
are also a few groups who believe that immo- rosmarinic acid in Coleus blurnei (Fig. 1)
bilized plant cells are better suited for plant (RAZZAQUEand ELLIS, 1977; ZENKet al.,
cell production processes than freely sus- 1977a). The cultures were initiated on the
pended cells (TANAKA, 1994). However, only widely used B5 medium with 2,4-D and kine-
cell suspension cultures have been scaled up tin as phytohormones. These cultures were
to volumes of industrial relevance. For exam- maintained in suspension for several years
ple, suspension cultures of Echinuceu purpur- without loosing their capacity for synthesizing
eu have successfully been grown in bioreac- and accumulating rosmarinic acid. RAZZA-
tors of up to 75 m3 (WESTPHAL,1990). Thus, QUE and ELLIS(1977) reported the accumu-
without a doubt they remain the most attrac- lation of 8-1 1% rosmarinic acid (correspond-
tive system from a technological point of ing to 1.2-1.5gL-') on the B5 growth me-
view. The disadvantage of rapidly growing, dium. The synthesis of rosmarinic acid can be
morphologically undifferentiated callus cul- stimulated by increasing the sucrose levels to
tures and cell suspensions is that in such a cell 5 4 % . Under these conditions, the content of
state only a small part of the biosynthetic po- rosmarinic acid increased up to 15% (ZENK
tential is expressed. Nevertheless, some prod- et ai., 1977a). The specific values exceed
ucts are found in suspension cultures at levels those of the various organs of the plants by a
which exceed those of the differentiated factor of 5. From their experiments in a vol-
plant. The few pathways which are extremely ume of 25 mL ZENK'S group calculated a
well expressed in suspension cultures will be -'
yield of 3.6 g L rosmarinic acid within 13
described first, followed by a few examples of days (0.3 g L-l d-'). However, their first at-
pathways which yield reasonable product lev- tempts at scaling up rosmarinic acid produc-
els. For the latter group I have chosen com- tion in a 3 0 L airlift bioreactor showed that
pounds of at least some biotechnological in- shake flask experiments are not necessarily
terest and about which sufficient information transferable to larger-scale fermentations
9 COOH
tures. This production level justifies a serious
consideration of establishing a culture process
as an alternative to harvesting from lower
producing plants. However, the pharmacolog-
ical efficacy of rosmarinic acid was evidently
not high enough to encourage the Natter-
mann Company to further continue work on
Rosmarinic acid this compound. Indeed, the company has giv-
Coleus blumei
en up all previous activities in plant tissue cul-
Anchusa officinalis tures, and presently there seems to be no
commercial interest in rosmarinic acid.
Returning to the discussion in Sect. 2,
which of the points raised are applicable to
rosmarinic acid? (1) Rosmarinic acid belongs
to the group of compounds which is sponta-
neously accumulated at reasonably high to
high levels in typical plant cell culture media.
Verbascoside R = Rhamnose It is not only produced at high levels by C.
Syringa vulgaris blumei but also by several other plant species
which are able to biosynthesize rosmarinic
0 acid (WHITAKER et al., 1984). Cultures of
Anchusa officinalis spontaneously accumu-
lated 6% rosmarinic acid (DE-EKNAMKUL
and ELLIS,1984). (2) The production of ros-
Caffeoyl putrescine marinic acid can be greatly improved by very
Nicotiana tabacum simple media variation (DE-EKNAMKUL and
ELLIS,1985a, b). (3) As with most sponta-
Fig. 1. Caffeic acid derivatives accumulating at very neous high producer lines production is rath-
high levels in cultured cells (see text). er stable or can easily be re-established by
controlled culture conditions.
The fact that high levels of rosmarinic acid
since productivity dropped to 15% of the can be easily produced by suitable tissue cul-
shake flask experiments. As rosmarinic acid tures was well established in 1985. The hope
has recently been shown to have good anti- for a commercial process was abandoned two
phlogistic activity, the Nattermann Company, or three years later. Thus, even higher pro-
Cologne, investigated this culture further. duction levels than those reported by UL-
They established new lines of C. blumei which BRICH et al. (1985) will not change the situa-
spontaneously synthesized rosmarinic acid tion. A recent report that an optimized me-
with yields of 200 mg L-' on the growth medium resulted in product levels of 6.4 g L-' in
dium. This level could be enhanced 20-fold by a culture of Salvia officinalis (HIPPOLYTE et
transferring the cells to a simple production al., 1992) is thus only a strong confirmation of
medium with an optimized concentration of previous findings. In terms of biotechnolog-
sucrose as the main stimulator for maximum ical relevance, it is clear that the importance
production (ULBRICH et al., 1985). As ex- of rosmarinic acid for the field is not so much
pected from the experience with microbial a question of product levels. More important
fermentations, these yields were further in- is how this system is used to determine at the
creased in 30 L airlift or stirred reactors up to molecular level why this pathway is so well
21% or 5.6 g L-'. The output of rosmarinic expressed and why, for example, sucrose
acid was thus raised to 0.93 g L d -' -' (UL- sometimes acts as a strong inducer. Media
BRICH et al., 1985). This was at that time the variation experiments aimed elucidating the
highest amount of a defined secondary me- physiological or biochemical basis of ob-
tabolite ever produced with plant cell cul- served production increases are of course also
still important. Thus, it has recently been been confirmed. Cultures of Hydrophifa erec-
noted that the effect of sucrose depends on tu (HENRYet al., 1987) or Leucosceptrum ju-
the carbohydrate level in the medium at the ponicum (INAGAKI et al., 1991) yielded prod-
time when phosphate limitation occurs uct levels in the range of 2 g L-' without op-
(GERTLOWSKIand PETERSEN,1993). This timization. The analyses of cultures of several
observation might explain why not all rosmar- other plant species at the callus level indicate
inic acid cell cultures react to the same extent that many more highly effective systems for
to increased sucrose supply. At least two the production of verbascoside can be estab-
groups have used cultures of A. officinalis and lished (DELL et al., 1989; INAGAKIet al.,
C. bfumei for identifying all enzymes involved 1991). However, there is currently no com-
in rosmarinic acid biosynthesis (MIZUKAMI mercial interest in the production of these
and ELLIS,1991; PETERSENet al., 1993) and compounds by tissue cultures. Though the
some of these enzymes have been purified so pathway is well expressed in cultured cells it
that cloning of the corresponding genes could has not yet been used for biochemical and
now proceed. Culture systems expressing a regulatory studies.
biosynthetic pathway quite well under all cul- The presence of various hydroxycinnamoyl
ture conditions may not be very suitable for putrescines (Fig. 1) in cultured cells of Nico-
the identification of regulatory factors con- tiunu tabacum was first reported by MIZUSA-
trolling expression. Recently, it has been KI et al. (1971). That these compounds have
shown that rosmarinic acid and the enzymes received some more attention during the last
involved in its biosynthesis can be induced in years is due to accidental finding. When se-
Lithospermum erythrorhizon and Orthosi- lecting the p-fluorophenylalanine resistant
phon aristutus by elicitors such as yeast and cell line TX4 from a widely distributed XD
methyl jasmonate from nearly zero up to ca. (TX1) line (N. tubucum cv. Xanthi) PALMER
1.5% of dry mass (MIZUKAMI et al., 1993; Su- and WIDHOLM(1975) noted that the levels of
MARYONO et al., 1991). Identification of the phenolics were increased manifold in TX4
factors which allow spontaneous overproduc- cells. The phenolics were identified as hy-
tion of rosmarinic acid in cultures of some droxycinnamoyl putrescines with caffeoyl pu-
plant species, and clarification of why other trescine as main component (Fig. 1) (BERLIN
plant cells require inducer compounds will et al., 1982). TX1 cells accumulated between
hopefully provide some useful hints for ma- 0.6-1 % hydroxycinnamoyl putrescines on the
nipulating pathway controls. growth medium, while TX4 cells contained up
Another caffeoyl derivative is verbascoside to 10% of these compounds on a dry mass ba-
(acteoside) (Fig. 1). ELLIS(1983) established sis. This system has been used in two direc-
various lines of Syringa vulgaris on B5 me- tions: optimization of product formation and
dium, all of which are produced spontaneous- comparison of the biochemical differences
'
ly high levels of up to 1.4 g L - verbascoside. which lead to the different productivities.
Rapidly growing suspension cultures of S. Growth limiting conditions (e.g., phosphate
vulgaris were found to contain a higher spe- limitation) stimulated product formation
cific content of verbascoside (15%) than cal- (SCHIELet al., 1984a). It is a frequently ob-
lus cultures ( 5 4 % ) (ELLIS1983, 1985). Due served phenomenon that growth and second-
to reports that verbascoside is a biologically ary metabolism are countercurrent processes
active compound with antibacterial, antiviral, in cultured cells. Thus, even the formation of
antihypertensive, and immunosuppressive products accumulating at high levels on the
properties, the interest in verbascoside-pro- growth medium can often be strongly en-
ducing tissue cultures has increased, especial- hanced by growth limiting conditions. The ne-
ly because plants generally contain only low gative effect of accumulated phosphate on the
amounts of this compound (see literature synthesis of hydroxycinnamoyl putrescines
cited by INAGAKI et al., 1991). The initial ob- suggested that the employment of fedbatch
servation of ELLISthat verbascoside belongs fermentation would give highest yields. The
to the group of compounds whose production productivity of the high producing variant
is favored under cell culture conditions has TX4 was indeed increased to 1.5 g L-' by a
70 L fed-batch fermentation with phosphate did not enhance the content of the major
as a limiting nutrient (SCHIELet al., 1984b). compound, the flavonoid rutin, but instead
Shake flask and batch fermentation of TX4 increased the levels of many other phenolics
-’
cells usually yielded 0.8-1.2 g L (BERLINet which were only present in trace amounts in
al., 1982). The yield of TX1 cells was in- wild-type callus (BATE et al., 1994).
creased by the fedbatch techniques from 160-
200mgL-’ to 300-400mgL-’. This shows
the importance of using the best possible line 3.1.2 Naphthoquinones and
for product optimizations. TX4 cells were the
first biochemically selected variant line which Anthraquinones
showed overproduction of secondary meta-
bolites and were thus interesting for biotech- Quinones comprise a large group of sec-
nological studies. It is also noteworthy that ondary metabolites that are widely distri-
this highly productive variant line has main- buted in the plant kingdom. Since they are
tained its production potential for more than colored and, therefore, visible compounds,
15 years (MEURER-GRIMES et al., 1989). they have been a favored target of tissue cul-
Since the last review (BERLIN,1986) no fur- ture research. Naphthoquinones and anthra-
ther studies regarding production improve- quinones sometimes accumulate in cultured
ments of hydroxycinnamoyl putrescines have cells at levels far exceeding the amounts
been performed. found in the intact plant. Some of these struc-
Though hydroxycinnamoyl putrescines are tures represent the active components of
of no commercial interest the results obtained drugs. They are also important as natural
with TX4 cells during product optimization dyes.
stimulated the analysis of other culture sys- The red shikonin pigments of the cork
tems. A biochemical comparison of the low layer of the roots of Lithospermum erythro-
and high producing tobacco lines showed that rhizon are derivatives of 1,4-naphthoquin-
enhanced hydroxycinnamoyl putrescine for- ones (Fig. 2). These compounds have been
mation was due to distinctly enhanced activi- used medicinally in Japan for the treatment of
ties of biosynthetic enzymes providing the burns and skin disease and are now mainly
cinnamoyl and amine moieties in TX4 cells used as a dye for lipsticks and for staining
(BERLINet al., 1982) while the activity of the
conjugating enzyme was similar in TX1 and
TX4 cells (MEURER-GRIMES et al., 1989). OH 0
Though the biochemical comparison of par-
ent and variant lines provided some clues as
to the biochemical requirements of overpro-
ducing lines, we did not analyze the system at
the molecular level. One reason for this was Shikonins R = H, or aliphatic acids
mo-R
the finding that not only the hydroxycinna- Lithospermum erythrorhizon
moyl putrescine pathway but also other en-
zyme activities such as tyrosine decarboxylase 0
(WALKERet al., 1986) were altered in the p -
fluorophenylalanine resistant cell line. An- CHzOH
other reason was the fact that not only one 0 OH
but several enzyme activities should be “co-
Anthraquinones (e.g. Lucidin
enhanced” in order to improve hydroxycinna- primveroside, R = Glucose-Xylose)
moyl putrescine production by genetic tech-
Morinda citrifolia
niques. It has recently been shown that phe- Galium mollugo
nylalanine ammonia lyase is the rate-deter-
mining step in phenyl propanoid biosynthesis Fig. 2. Naphthoquinones and anthraquinones accu-
(BATE et al., 1994). Overexpression of this mulating at very high levels in cultured cells (see
enzyme in transgenic tobacco callus, however, text).
silk. The plants have to be grown for 3-4 individual shikonins was different (FUJITA,
years before a yield of 1-2% shikonin is 1988). The altered ratio of the various shikon-
achieved in the roots (FUJITA,1988). The to- in derivatives in the cell cultures is no prob-
tal amount of Lithospermum roots used each lem if the mixture of compounds is to be used
year in Japan is 10000 kg. From this an an- for cosmetic purposes or as a dye. However, if
nual demand of 150 kg shikonins can be cal- one wants to replace a known approved plant
culated. As the plant cannot be grown in drug by a tissue culture extract extensive eval-
commercial quantities in Japan it has to be uation of the pharmacological properties and
imported from Korea and China. It is often equivalence studies are required. Such studies
argued by Japanese scientists in industry (Ko- have recently been initiated for L. erythrorhi-
MAMINE et al., 1991) that due to the geo- zon cultures by the Mitsui Company (OZAKI
graphic situation the indigenous supply of et al., 1990; SUZUKIet al., 1991).
plant material is not sufficient, and that plant There are scattered reports in the literature
tissue culture technology is, therefore, per- of increased shikonin production in L. ery-
haps a more attractive alternative in Japan throrhizon cultures by in situ extraction, elici-
than elsewhere. This would explain the com- tation, or media variation. Though the ob-
mercial production of shikonins from L. ery- served effects are well demonstrated their fi-
throrhizon cell suspension cultures by the nal levels remain far below those reported by
Mitsui Company. Callus cultures of L. ery- FUJITAet al. Such studies would deserve
throrhizon were found to accumulate shikon- more attention if the production levels at the
in derivatives (TABATAet al., 1974). By re- Mitsui Company could additionally be im-
peated analytical screening over a period of proved by the newly recommended tech-
two years two highly productive strains con- niques. This remark seems to be especially
taining 20-fold increased levels of 1mg g -' valid in view of the report of the Mitsui Com-
fresh mass (ca. 10 mg g -' dry mass) were iso- pany that two-phase cultures did not improve
lated (MIZUKAMIet al., 1978). FUJITAet al. productivity of their high yielding line (DENO
(1981a, b) investigated the effects of all media et al., 1987). In general, naphthoquinones and
constituents on growth and production. They benzoquinones seem to belong to the groups
developed a production medium yielding of compounds which might readily be formed
1.4 g L-' shikonin derivatives within 23 d or in cultured cells of various plant species (Fu-
12% on a dry mass basis. Combining the two KUI et al., 1983; INOUEet al., 1984). It is evi-
media in a two-stage process (1st stage 200 L dent that the highly expressed naphthoquin-
growth medium, 2nd stage 750 L production one biosynthetic pathway would be a good
medium) the yield was increased to 3.7 g g -' system for biochemical and molecular studies.
dry mass inoculum within 23 d (FUJITAet al., The first results on the regulation of shikonin
1982). By screening protoplast derived clones biosynthesis have been presented by HEIDE
they isolated lines with an improved growth et al. (1989), showing that the ratio of p-hy-
and higher productivity. The best line had a droxybenzoic acid geranyltransferase and p -
specific content of 23.2% and yielded 6.45 g hydroxybenzoic acid glucosyltransferase ac-
shikonin per g inoculum (FUJITAet al., 1985). tivities is one of the regulatory controls of shi-
Production characteristics during the fermen- konin biosynthesis.
tation process were optimized by high density Anthraquinones in higher plants are
cultivation, fedbatch technique, controlled formed either via the acetate polymalonate
oxygen supply, and by the development of a pathway or via the o-succinylbenzoic acid
rotating cylindrical bioreactor (FUJITA and pathway. Their production in cell cultures has
HARA,1985; FUJITA,1988; TAKAHASHI and been reviewed in great detail by KOBLITZ
FUJITA,1991). Since 1983 the Mitsui Compa- (1988). Highly productive suspension cultures
ny produces shikonins by this technology for of plant species (e.g., of Cassia spp. or Rham-
cosmetics and dyes (TAKAHASHI and FUJITA, nus spp.) producing anthraquinones via the
1991). Comparison of the composition of shi- acetate polymalonate pathway have never
konins extracted from cell cultures with that been reported (VANDEN BERGet al., 1988).
of various roots showed that the ratio of the However, it has been documented in numer-
ous publications that the o-succinylbenzoic nylbenzoic acid derived anthraquinones. 17 of
acid derived anthraquinones are well ex- the cultures yielded anthraquinone levels
pressed in rapidly growing cell suspension higher than those found in the corresponding
cultures and accumulate sometimes at explants. It was shown that nutritional and hor-
traordinary levels. A good example are cell monal requirements of the various anthraqui-
cultures of Morinda citrifolia. ZENKet al. none producing cultures, even those of the
(1975) tested a large variety of nutritional fac- one family, may be quite different. Cultures
tors for their effect on growth and anthraqui- of Rubia cordifolia maintained their high pro-
none production. They established a produc- ductivity when scaled up to 75 L (SUZUKIand
tion medium yielding 2.5 g L-' anthraqui- MATSUMOTO,1988), and cultures of Rubia
nones corresponding to more than 10% of tinctorum have been studied at San-Ei Chem-
dry mass which exceeds the concentration of ical Industries (ODAKEet al., 1991) for the
the root by a factor of 10. Important for high development of a commercial production
production levels were the replacement of the process of anthraquinone pigments (e.g., of
phytohormone 2,4-D by NAA and an in- alizarin, purpurin). It is clear that these cul-
crease of the sucrose level to 7%. The anthra- ture systems are also suitable for biochemical
quinones of cell cultures of M. citrifolia con- and regulatory studies. The first enzymes in-
sist of a mixture of at least 12 aglyca and glu- volved in the biosynthesis of o-succinylbenzo-
cosides (LEISTNER,1975; INOUEet al., 1981). ic acid (SIMANTIRAS and LEISTNER,1989)
The main components are lucidin derivatives and its further metabolism (SIEWEKEand
(Fig. 2). Some of them have not yet been LEISTNER,1992) have been studied in cell
found in the intact plant (INOUEet al., 1981). cultures of Galium spp. The cultures of G.
The spontaneous high production of these an- mollugo have also been used to overproduce
thraquinones were maintained in various shikimic acid. The addition of glyphosate in-
bioreactors (WAGNER and VOGELMANN,hibits the formation of o-succinylbenzoic acid
1977). The yields of anthraquinones in an air- and anthraquinones in these cultures thus
lift reactor were 30% higher than in experi- causing the accumulation of the biosynthetic
ments with shake flasks. An interesting exam- precursor, shikimic acid (10% of dry mass,
ple of manipulating the expression of path- 1.2 g L-') (STEINRUCKENand AMRHEIN,
ways in cultured cell was reported for cell sus- 1980).
pension cultures of Morinda lucida. In pho-
toautotrophic cell cultures (chlorophyllous,
no sugar in the medium) lipoquinones were 3.1.3 Protoberberines and
the main components while anthraquinones
were not found. When these cultures were Benzophenanthridine Alkaloids
transferred into the dark and sugar was added
to the medium, anthraquinone biosynthesis Isoquinoline alkaloids represent one of the
was induced and lipoquinone formation was largest groups of alkaloids in the plant king-
repressed (IGBAVBOA et al., 1985). dom. Common to all these alkaloids is that
The results achieved with M. citrifolia cul- they are derived from tyrosine via (S)-norco-
tures can also be obtained with cultures of claurine as a central intermediate (RUEFFER
Galium mollugo (BAUCH and LEISTNER, and ZENK,1987) and not, as initially thought,
1978). A BS-NAA medium with 7% sucrose via (S)-norlaudanosoline synthase (RUEFFER
gave the highest yields with ca. 2 g L-' within et al., 1981). Since the pathway is highly
14 days. Lucidin primveroside (Fig. 2) was the branched, a great variety of very different
main component. While the pathway of an- structures results from this central interme-
thraquinones was readily expressed in cul- diate. Some of the branches are readily ex-
tured cells, the biosynthesis of iridoids repressed in cultured cells, while others such as
mained repressed under all culture condi- the morphinan alkaloids remain mostly re-
tions. SCHULTEet al. (1984) optimized cell pressed.
suspension culture media of 19 different Ru- There are many reports showing that pro-
biaceae species for optimal yields of o-succi- toberberine alkaloids (Fig. 3) spontaneously
els of up to 1.7 g L-' were readily achieved in

normal growth medium. The Coptis line de-
scribed by YAMADAand SATO (1982) con-
tained mainly berberine and only trace
amounts of other alkaloids. By screening
small cell aggregates for high berberine (yel-
Protoberberine Alkaloids low) producing clones cell lines were estab-
R'+ R2= CH2 Berberine lished which produced more than 1g berber-
R'= CH,, R2= H Columbamine -'
ine L or 10% on a dry weight basis (SATO
R' = H, R2= CH, Jatrorhizine and YAMADA,1984).
e.g., Coptis japonica Another important species spontaneously
Berberis species accumulating high levels of protoberberines is
Thalictrum rugosum
Thulictrum. While cell cultures of Thulictrum
minus secrete most berberine into the me-
dium (NAGAKAWA et al., 1984) those of T.
flavum, T. dipterocarpum or T. rugosum accu-
mulate the alkaloids within the cells (SUZUKI
et al., 1988; PIEHLet al., 1988). Interestingly,
0
when cells of T. rugosum were transferred to
R3 fresh medium lacking phosphate, protober-
Benzophenanthridine alkaloids berines were released into the medium (BER-
R' + R2= R3+ R4= CH2 Dihydro- LIN et al., 1988a). There are other reports in-
sanguinarine dicating that cultures of plant species with the
R' + R2= CH2, R3+ R4= CH, Dihydro- capability to biosynthesize protoberberines
chelerythrine will spontaneously yield cultures accumulat-
e.g., Papaver somniferum ing these colored alkaloids in high amounts
Eschscholtzia californica (0.2-1 g L-'). The levels of protoberberines
Fig. 3. Structures of protoberberine and benzo- are so high that the Mitsui Company has de-
phenanthridine alkaloids. cided to develop a production process for
berberine with Coptis juponicu cell cultures
(MATSUBARAand FUJITA,1991). As in the
case of shikonins the medium was first optim-
accumulate at high levels in cultured cells. ized. For example, a 10-fold increase of Cu2+
Various cultures of Berberis spp. accumulated increased the berberine content by 40% (Mo-
between 0.2 and 1.7 g protoberberine alka- RIMOTO et al., 1988). The addition of very
loids, mainly jatrorrhizine, on a growth me- low levels of gibberelic acid increased the
dium with 3.5% sucrose (HINZ and ZENK, content by 30% (HARA et al., 1988). Since
1981). BREULING et al. (1985) optimized the cell aggregate screening was regarded as very
production to 3 g L-' in a 20 L airlift bioreac- time-consuming and not very efficient, proto-
tor. Independently, two Japanese groups re- plasts with high berberine content were iso-
ported the accumulation of high levels of pro- lated with a cell sorter (HARAet al., 1989).
toberberines in cultures of Coptis juponicu All selected lines contained higher levels than
(FUKUIet al., 1982; YAMADAand SATO, the original culture. The next step was to de-
1982). FUKUIet al. established a line accumu- velop culture conditions for a high density
lating berberine, jatrorrhizine, palmatine, and culture (MATSUBARA et al., 1989) so that 70 g
coptisine at a ratio of 50: 22 :22 :6. Interesting- dry mass per L (corresponding to ca. 700g
ly, the content of alkaloids increased gradual- fresh weight) were obtained. This required in-
ly during subculturing from 8-15% which was creasing the nutrition and oxygen supply by
paralled by increased growth (FUKUIet al., fedbatch-perfusion cultivation (MATSUBARA
1982). This shows that protoberberine forma- and FUJITA,1991). Today, the Mitsui Compa-
tion is indeed a favored pathway in these rap- ny produces berberines by a high density con-
idly growing cell cultures. Thus, alkaloid lev- tinuous culture method over a period of sev-
era1 months with a yield of 0.65 g L-’ d-’. rum. When they analyzed the alkaloid pattern
This is 8 times more than the value obtained of callus cultures of 11 other species of Papav-
in batch cultures (MATSUBARA and FUJITA, eraceae, almost identical alkaloid spectra
1991). This demonstration of the potential of were found (IKUTAet al., 1974). Since then,
plant cell culture process optimization by the the occurrence of these alkaloids- has been
bioengineers suggests that further attempts to confirmed in numerous reports. Cell culture
optimize protoberberine production should conditions seem to favor the synthesis of ben-
start from initial product levels as high as zophenanthridine alkaloids which are found
those published by the Mitsui Company. It in the corresponding intact plants often at
was speculated that cultures which release rather low levels (WILLIAMSand ELLIS,
their compounds into the medium might be 1993). However, there is a significant differ-
favored by industry because this would dimin- ence between the production of benzophen-
ish the problem of sacrificing the slowly grow- anthridine and protoberberine alkaloids in
ing plant cells for product extraction. In the cultured cells. While in C. juponicu, e.g., pro-
case of protoberberines, an alternative protoberberine production increased with in-
duction process seemed to be possible with creasing growth rate (FUKUIet al., 1982),
the highly productive T. minus line. T. minus benzophenanthridine synthesis decreased
cells were probably tested by the Mitsui Com- with better growth (BERLINet al., 1985). A
pany in an attempt to develop a new bioas- suspension culture of P. somniferum with a
say-based screening method for high berber- growth cycle of 28 d contained nearly 6% san-
ine-producing cell colonies (SUZUKI et al., guinarines (360 mg L-’) after the 5th subcul-
1987). The real advantage of T. minus cells tivation. During further subcultivation the
can only be exploited if cells are immobilized -’
yield decreased to 200 mg L after 20 sub-
for berberine production. KOBAYASHI et al. cultures and to 20mgL-’ after 35 subcul-
(1988) obtained a production rate of tures. Biomass production of this culture in-
’
50 mg L d - in batch and semicontinuous creased from 6 g in 28 d to 11 g in 10 d (BER-
bioreactors over a period of at least 60 days LIN et al., 1985). The “biotechnological” in-
of cultivation. This was undoubtedly an excit- terest in benzophenanthridine alkaloids in-
ing result. However, in view of the improve- creased only when it was noted that these al-
ments obtained with the C. juponicu culture, kaloids can be induced greatly by stress fac-
it is clear that T. minus cultures are not yet tors and elicitors. Thus, the yields of benzo-
competitive for use in a commercial process. phenanthridine alkaloids of suspension cul-
Berberine chloride is used in Japan as a med- tures of Eschscholtziu culifornicu were in-
icine for intestinal disorders and treatments -’
creased 10-fold to 150 mg L by increasing
of abnormal zymosis and has previously been the sucrose concentration in the medium to
obtained by extraction from the roots of C. 8% (BERLINet al., 1983). The observation of
juponica and the cortex of Phellodendendron EILERTet al. (1985) that sanguinarine levels
umurens (MATSUBARAand FUJITA, 1991). of P. somniferum were enhanced from 0.01%
Berberine produced with tissue culture has to 2.9% by treatment with fungal elicitors re-
not yet been approved as a drug in Japan. ceived even more interest. The advantage of
However, pharmacological studies towards the elicitor treatment in comparison with the
this goal are underway (SUZUKI et al., altered production media is that the cells re-
1993a, b). spond more rapidly, and the induced alka-
Another group of isoquinoline alkaloids loids in Pupuver cell cultures are released into
which accumulate spontaneously in cultured the medium (EILERTet al., 1985; CLINEand
cells are benzophenanthridine (Fig. 3) and COSCIA,1988). In the case of E. culifornicu,
the related protopine alkaloids. Cell cultures the alkaloids, induced by various fungal and
of most Papaveraceae contain alkaloids of yeast preparations evidently remained within
this group. FURUYA et al. (1972) were the the cells (SCHUMACHER et a]., 1987). These
first to describe the occurrence of sanguinar- authors reported product levels in the same
ines, protopines, and the aporphine magno- range (160 mg L-’) as were found in the su-
florine in callus cultures of Pupuver somnife- crose induced cells (BERLIN et al., 1983).
There was, however, one significant differ- other characteristics important for the suit-
ence: the elicited cells contained the quarter- ability of a selected line for scale tlp purposes
nary alkaloids while the sucrose-induced cells remain to be shown. Although a commercial
contained the corresponding dihydro forms. enterprise (Vipont) has been involved in the
It is interesting to note that ZENK’Sgroup did development of this process, it is unclear
not consider elicitor technology to be helpful whether an industrial production will result.
for increasing productivity for commercial Further attempts to optimize sanguinarine
purposes (SCHUMACHER et al., 1987). This production with cultures of P. somniferum, E.
was based on the finding that lines containing californica, or Sanguinaria canadensis should
high levels of the alkaloids in the unelicited be left to industry if product level improve-
state did not produce higher levels after elici- ments are the only goal. They know how
tation than low yielding lines. This technology much the productivity has to be enhanced for
has nevertheless been used in efforts of devel- a tissue culture process to be superior to con-
oping a commercial production process for ventional extraction of field grown plant ma-
sanguinarine. This compound has an antibiot- terial.
ic activity against oral microorganisms caus- It is evident that the culture systems pro-
ing periodontal disease (SOUTHARDet al., ducing protoberberines and benzophenan-
1984). The chances of plant cell cultures to be thridines are useful systems for studying the
used as a source were improved by the fact enzymology of these pathways. All 13 en-
that cultured cells contained much higher lev- zymes required for the synthesis of berberine
els of sanguinarine than tissues of intact have been identified. A summary of the pres-
plants and that an established production ent knowledge, derived mainly from studies
process did not exist. of ZENK’Sgroup at the University of Munich
Consequently, research has concentrated and from YAMADA’S group at the University
on the optimization and scale-up of sanguin- of Kyoto, has been presented by HASHIMOTO
arine production. It was shown that the mildly and YAMADA(1994). The enzymology and
elicited P. somniferum could be re-elicited molecular biology of benzophenanthridine al-
after a period of regrowth suggesting that a kaloid biosynthesis has been reviewed by
semicontinuous production process with re- KUTCHAN and ZENK (1993). Thus, the bio-
elicitation could be established (TYLERet al., technological value of the cultures producing
1988). Due to the fact that a great portion of these alkaloids does not only lie in their im-
the elicited alkaloids were released, the suita- pressive productivities but also in their poten-
bility of surface-immobilized cells with ad- tial to provide a deep insight into the regula-
sorption of the released alkaloid to a resin tion of the expression of these complex and
was tested (KURZ et al., 1990). The total branched pathways.
yields, however, were reduced. The highest
yield published up to now is 300 mg L-’ in a
300L airlift reactor (PARKet al., 1992). The 3.1.4 Monoterpene Indole
cultures were grown from ca. 20 to 180 g fresh
mass per L and were elicited after 7 d or Alkaloids
when ca. 2 g L-’ glucose were left in the me-
dium, and were harvested two days later. The About 1200 alkaloids derived from trypto-
authors believed that further improvements phan have been isolated from higher plants,
might be possible by optimizing the amount which corresponds to about one quarter of all
of dissolved oxygen in the medium. If, in ad- alkaloids (GRBGER,1980). Several of these
dition, conditions for elicitation of a high den- alkaloids from Rauwolfia, Catharanthus, and
sity culture could be established, a commer- Cinchona are used medicinally. Therefore, it
cial production would be possible. was of great interest to see whether these al-
There have also been efforts to select cell kaloids also accumulate in cultured cells. The
lines with higher production potential for san- dimeric monoterpene indole alkaloids, vin-
guinarine (SONGSTADet al., 1990). The sta- blastine and vincristine, used as antileukemia
bility of these cell lines during production and agents, are only present in trace amounts in
the whole Catharanthus roseus plant. Indeed, Therefore, this claim cannot be regarded as
these two compounds were initially the main being biotechnologically relevant. Cell cul-
reason why cultures of C. roseus received so tures of R. serpentina were scaled up to 75 m3
much attention. These two alkaloids as well by the Diversa Company and no alkaloids
as the medicinally used alkaloids of Cinchona were found in the rapidly growing cell cul-
spp. will be discussed later in Sect. 3.2 in con- tures (WESTPHAL,1990). Indeed, it is some-
nection with poorly expressed compounds. times very difficult to evaluate the impact of
More than 50 monoterpene indole alka- product level claims, and it is futile to discuss
loids have been isolated from various Apocy- such claims here with respect to their biotech-
naceae, e.g., Catharanthus and Rauwolfia spp. nological impact. Information regarding this
The chemical identification of their complex issue can be found in several special reviews
alkaloid mixtures was achieved mainly by covering the monoterpene alkaloids in cul-
three groups (STOCKIGTand SOLL, 1980; tured cells (BALSEVICH,1988; DELUCAand
KOHLet al., 1982; KUTNEYet al., 1983). A list KURZ, 1988; VAN DEN HEIJDENet al., 1989;
of almost all monoterpene indole alkaloids MORENOet al., 1995) or in the article of EL-
found in plant tissue cultures of various plant LIS (1988). The biotechnological progress on
species is given by ELLIS(1988). Most of the the improvement of product levels of ajmali-
alkaloids accumulate at low levels in the cul- cinelserpentine and catharanthine, about
tures. Sufficiently confirmed quantitative data which reports from many different laborato-
allowing an evaluation of the biotechnolog- ries are available, is discussed below.
ical relevance are most often not available. Callus cultures of C. roseus were found to
Most studies regarding product levels have contain low levels of monoterpene indole al-
concentrated on the optimization of ajmali- kaloids (CAREW,1975). However, suspension
cine, serpentine (oxidized form of ajmalicine) cultures accumulated no or only trace
and catharanthine (Fig. 4). There has been amounts of these compounds when grown on
one report that a cell suspension culture of a growth medium with 2,4-D as phytohor-
Rauwolfia serpentina accumulated 1.6 g rau- mone. ZENK’Sgroup was the first to develop
caffricine per liter of medium (SCHOBELet a production medium (ZENKet al., 1977b).
al., 1989). However, this level dramatically When the cells were transferred from the
exceeds the levels of all other monoterpene growth to a production medium, alkaloid for-
indole alkaloids ever measured in cultured mation was resumed after 3-5 d. This medium
cells and has not yet been confirmed in fur- has successfully been used by several other la-
ther publications by this or any other group boratories. Of 458 independently established
working with R. serpentina cell cultures. cell lines, 312 (approx. 75%) produced alka-
loids when transferred to ZENK’Sproduction
medium (KURZet al., 1980). This production
medium, however, had no special composi-
tion; the sole transfer of the cells into a 2,4-
D-free medium allows the accumulation of
reasonable levels of indole alkaloids (ZENK
et al., 1977b; KNOBLOCHand BERLIN,1980;
PAREILLEUX and VINAS,1984). The increase
of sucrose to 5 4 % had an additional benefi-
Ajmalicine cial effect. The phytohormone composition
(indole acetic acid and benzylaminopurine) of
ZENK’Smedium may have an additional stim-
ulatory effect. The stimulatory effect of the
production medium probably depends on the
Catharanthine
physiological state of the cells at the time of
transfer from the growth to the production
Fig. 4. Monoterpene indole alkaloids of Catharan- medium. The level of phosphate accumulated
thus roseus. in the cells seems to play an important role in
the inducibility of alkaloid formation in a analytical screening of cultures growing on

phosphate-free production medium (KNo- NAA/kinetin medium. Again, researchers of
BLOCH and BERLIN,1983; SCHIEL et al., the Mitsui Company (FUJITAet al., 1990) ob-
1987). tained this success. Repeated analytical
Suspension cultures of C. roseus grew rap- screening improved the specific catharanthine
idly, were easy to scale up in fermentors and content from 0.1% to 0.7%, and these lines
'
yields of 30-70 mg L - serpentine were have maintained their superior productivity
usually obtained within 20-30 days (ZENKet for at least 10 subcultures. Additionally, a
al., 1977b WAGNERand VOGELMANN, 1977; biochemical selection for 5-methyl-trypto-
SMARTet al., 1982). There have been several phan-tolerant cell lines was applied. (FUJITA
efforts to isolate higher producing cell lines et al., 1990). Although the specific catharan-
using screening methods. The group at the thine content was further increased, it is not
NRC in Saskatoon established more than clear why these cell lines form higher levels of
2000 individual lines from three C. roseus cul- alkaloids. The productivity of the isolated
tivars and screened them for alkaloid patterns lines was further optimized by optimizing the
and levels (KURZet al., 1985). They found a medium composition, inoculum density, and
substantial variation among the lines, and one fermentation parameters. Finally, the 5-MT-
variety seemed to be more suitable for catha- '
tolerant line produced 230 mg L - catharan-
ranthine formation. However, this tremen- thine in a 1.7 L bioreactor within one week
dous effort did not evidently result in the iso- (specific content 1.1% of dry mass).
lation of an exeptionally high producing line. There have been tremendous efforts to
ZENK'Sgroup has screened established lines stimulate the production of ajmalicine and
for high yielding clones using specific ra- serpentine in cultures of C. roseus. Many ap-
dioimmunoassays and the fluorescence micro- proaches have been used such as the addition
scope (ZENKet al., 1977b; DEUS and ZENK, of various biotic and abiotic elicitors, the use
1982). Indeed, they found clones having accu- of high and low density culture, feeding and
mulated high levels of serpentine or ajmali- elicitation, alteration of media composition,
cine of which theoretical levels of up to immobilization, etc. (for reviews, see VAN
400 mg L-' were calculated. The isolated DEN HEIJDENet al., 1989; MORENOet al.,
clones were, however, unstable, and alkaloid 1995). Despite the many reports claiming to
levels rapidly dropped to the values of the un- have demonstrated increased alkaloid pro-
screened culture (DEUS-NEUMANNand duction, one has to realize that the product
ZENK,1984). The behavior of the clones indi- levels have not been improved since the first
cates that the they were not true variants as findings 15-20 years ago. Thus, the product
claimed but instead were cell colonies in a levels of bioreactor studies remain at 10-
transient physiological state. Reasons for the 50 mg L-' ajmalicine or serpentine (corre-
failure to select true variants in this system by sponding to 0.1 to 0.5% of dry mass) (SCHIEL
the applied screening method have been dis- and BERLIN, 1987; SCRAGG et al., 1989;
cussed in detail (BERLINand SASSE, 1985; SCRAGGet al., 1990; SCHLATMANN et al.,
BERLIN,1986). 1994). One reason for this undoubtedly poor
For many years cultures of C. roseus were progress could be the fact that most often
grown on media with 2,4-D. Alkaloid forma- only a single factor was analyzed at any one
tion thus required transfer to a production time. Furthermore, analysis of the various
medium. MORRIS(1986) found that cultures factors did not take advantage of the pre-
of C. roseus can not only be established and viously improved culture levels. A compari-
maintained on a medium with NANkinetin son of the many studies done in universities
lacking 2,4-D but that they also produce alka- and research centers in an attempt to optim-
loids on this medium. These findings have ize monoterpene alkaloid production in C. 10-
also improved the chances of selecting clones seus with the industrial approach, e.g., of the
with higher production. As predicted at a Mitsui Company reveals that the former
meeting in 1988 (BERLIN,1990), high yielding group of scientists is interested in detecting
clones were isolated by consecutive repeated factors influencing alkaloid levels, but not re-
ally in improving product levels. Despite (DELUCAet al., 1989; GODDIJN,1992). The
nearly 20 years of bioreactor studies with C. negative effect of phytohormones on alkaloid
roseus it has been reported that the inconsis- formation can perhaps now be explained by
tencies and the often contradictory results are the finding that auxins down-regulate tran-
likely due to insufficient characterization of scription levels of the tryptophan decarboxyl-
the inoculum material (VAN GULIKet al., ase gene (GODDIJNet al., 1992). Geraniol-lO-
1994). The authors suggested to use identical hydroxylase is regarded as a regulatory en-
inoculum material when analyzing the effects zyme for the monoterpene moiety (secolo-
of factors on growth and alkaloid production. ganin) of the indole alkaloids. Indeed, this en-
For industrial scientists, productivity im- zyme is also induced during initiation of alka-
provement is a step-by-step enhancement of loid biosynthesis (SCHIELet al., 1987). It has
product levels per volume and time and this already been purified and its cloning is in pro-
requires that each step is reproducible. Thus, gress (MEIJERet al., 1993a). The enzyme and
it seems that the maximum potential of C. ro- the gene coding for strictosidine synthase
seus cultures for serpentine and ajmalicine which connects tryptamine and secologanin
production may only be found when a com- have been isolated (KUTCHAN,1993). in
mercial enterprise takes over. A first step some lines the enzyme was present in non-
would be a screening of cultures growing on producing cells, in others it was induced when
NAA kinetin medium for true variants over- alkaloid formation was elicited. Molecular
producing ajmalicine or serpentine. However, studies suggest a coordinated regulation of
a comparison of the production level improv- the genes coding for tryptophan decarboxyl-
ements for berberine, shikonins, and catha- ase and strictosidine synthase (PASQUALIet
ranthine by the Mitsui Company shows that al., 1992). Detailed overviews of the enzymol-
the maximum yield possible is mainly deter- ogy and regulation of monoterpene indole al-
mined by the biology of the cell. The rather kaloids have been given recently (DELUCA,
low levels of the monoterpene indole alka- 1993; MEIJER et al., 1993b). The results of
loids found to date suggest that at best levels molecular studies should show whether ge-
in the range of 200-500 mg L-' within 1-2 netic improvements of the biosynthesis of cer-
weeks might be produced by an optimized tain monoterpene alkaloids are an attainable
culture. According to an economic assess- goal.
ment of the production of ajmalicine by C. ro-
sew cultures, the present levels have to be in-
creased at least by a factor of 40 to make the 3.1.5 Anthocyanins and Betalains
process competitive (DRAPEAUet al., 1987).
In contrast to the poor progress at the Anthocyanins have been found in many
product level cell cultures of C. roseus have cultured cells of many plants (Daucus, Haplo-
been used quite well for studies on the regul- pappus, Catharanthus, Petunia, Mathiola, Eu-
ation of alkaloid formation. The correlation phorbia, Perilla, Vitis, Aralia, and many oth-
of enzyme activity pattern and product accu- ers). For some cultures the individual antho-
mulation curves and the conclusions drawn by cyanin components have been identified, for
various groups are not, however, always con- others only the total amount of the pigments
sistent. Therefore, only facts which are gener- has been given (ELLIS,1988; SEITZand HIN-
ally accepted are mentioned here. It has re- DERER,1988; YAMAMOTO,1991). In a few
peatedly been shown that tryptophan decar- culture systems anthocyanins are formed in
boxylase is induced when alkaloids are pro- dark grown cells, but in most cases light is re-
duced. However, it has also been shown that quired for optimal production (SAKAMOTO et
induction of this enzyme does not necessarily al., 1994). Due to their low toxicity, antho-
lead to enhanced alkaloid formation (for a re- cyanins are widely used in food additives and
view see VAN DEN HEIJDENet al., 1989). also as a dye for silk (YAMAMOTO,1991).
Nevertheless, since tryptophan decarboxylase While anthocyanin extracts are rather inex-
is readily inducible two groups have cloned pensive the prices for individual anthocyanins
the cDNA of this enzyme from C. roseus seem to be very high which would justify the
use of tissue culture based processes. Again,

Japanese scientists and industries are leaders
in the development of biotechnologically rel-
evant culture systems.
Since anthocyanin producing cells are eas-
ily detected in callus cultures they have be-
come an early target for visual screening R = arabinoside
(KINNERSLEYand DOUGALL, 1980). The Cyanidin-3-arabinoside(Euphorbia millii)
most impressive success was reported by YA-
MAMOTO et al. (1982) when they established
high yielding cell lines of Euphorbiu milli by
consecutive repeated screening. Only after 24 HO
clonal selections were stable lines found; this
required permanent screening over a period
of many months. NOZUE et al. (1987) were
also successful in selecting high anthocyanin HOOC
A!k
N COOH
yielding cell lines from sweet potato by conse- I
H
cutive repeated screening. This shows that
even for compounds formed in the growth Betanidin, R = H
Betanin, R = Glucose
medium, stable clones are only obtained by
repeated screenings. This result casts some (various Centrospermae)
doubt to claims of the instability of high pro- Fig. 5. Structures of anthocyanin and betacyanin.
ducing variant lines if permanent screening
was not applied over a very long period
(BERLIN,1990). As for many other secondary.
metabolites, the culture medium can also be the ratio NO,/NH,+, and the total nitrogen
optimized for anthocyanin production. The levels improved the anthocyanin content to
optimization of E. miflii cultures by the Nip- 10.3% of dry mass. This culture was scaled up
pon Paint Corporation improved the produc- to 300 L in a 500 L jar fermentor (KOBAYA-
tivity 4.5-fold to 32 mg L-' d-' (YAMAMOTO SHI et al., 1993). After 1 6 d of cultivation
et al., 1989). The main component of the E. 69 kg fresh mass with 545 g anthocyanin (cor-
rniffii cell culture is cyanidin-3-arabinoside responding to 17.2% of dry mass) were ob-
(Fig. 5 ) (YAMAMOTO, 1991). A great disad- tained. The further improvement of produc-
vantage of the Euphorbiu system in commer- tivity during scale-up was evidently due to the
cial production is the requirement of light for controlled supply of C 0 2 to the cultures. Ef-
pigment production. The same holds true for forts to identify the anthocyanins present in
a highly productive culture of Perrillu frutes- this high yielding culture system are under-
cens (ZHONG and YOSHIDA, 1995) which way. As is typical in Japan the project is being
produces up to 5.8 g L-' anthocyanins in Iod done with close cooperation between re-
under optimized culture conditions and light. searchers from university and industry (Kyo-
Therefore, the best line for the commercial wa Hakko Kogyo Co. and Tonen Co.).
production of anthocyanins may be cultures Tissue culture systems may only become
of Aruliu cordutu which produce high levels in competitive for the production of individual
the dark. As with the cultures of Euphorbiu anthocyanins and not for mixtures of antho-
and Zpomeu, continuous cell aggregate clon- cyanins. Therefore, it might be useful to look
ing led to a highly productive and fast grow- not only at total yields during selection and
ing line (SAKAMOTO et al., 1994). 90% of all media optimization but also at whether such
cells of the cloned line produced anthocyan- methods affect the percentage of any one an-
ins, a result which is only possible if product thocyanin in a complex mixture ( D o and
formation and growth are closely interrelated. CORMIER,1991).
Optimization of the basic inorganic salt con- As shown in the previous sections optim-
centration, the carbohydrate concentration, ized culture systems have often successfully
been applied for biochemical and regulatory Beta vulgaris contain higher levels of beta-
studies by those interested in their biotechno- cyanins than all plant tissues tested, it is un-
logical application. Although this has not yet likely that cell cultures will be used for the
been the case, our knowledge about their bioproduction of betacyanins since the marked
synthesis and the regulation of this pathway price for red beet extracts is extremely low
far exceeds that of all other secondary path- (LEATHERSet al., 1992).
ways (for reviews see DIXON and LAMB,
1990; FORKMANN, 1993; DOONERet al., 1991;
KOES et al., 1994). Thus several genetic ma- 3.1.6 Steroidal Compounds
nipulations of these pathways have been
made with the aim of altering pigmentation of There are numerous reports on the occur-
flowers (MEYERet al., 1987; VAN DER KROL, rence of sterols, steroids, sapogenins, sapon-
1990). ins, and steroidal alkaloids (ELLIS,1988). In
Other pigments often found in cell cultures my previous review (BERLIN,1986) the ef-
of Centrospermae are the betalains which in- forts to optimize diosgenin production in cul-
clude red betacyanins and yellow betaxan- tures of Dioscorea deltoidea were described in
thins (BOHM and RINK, 1988). Cultures of some detail. As no significant progress to-
Chenopodium rubrum (BERLINet al., 1986), wards a commercial production has been re-
Phytolaca americana (SAKUTAet al., 1987), ported since then, I will concentrate here on
Amaranthus tricolor (BIANCO-COLOMAS and the ginseng saponins, which were only briefly
HUGUES,1990), Beta vulgaris (GIROD and mentioned in the last review.
ZRYD,1991), and Portulaca (KISHIMAet al., The roots of Panax ginseng C.A. Meyer
1991) produce rather high levels of betacyan- have widely been used as a tonic and precious
ins under light conditions. Betacyanin forma- medicine in oriental countries since ancient
tion is more or less growth-rela#ed, which ex- times. It is regarded as an adjuvant to prevent
plains the spontaneous accumubtion of these health disorders and is considered to be a
pigments in cell cultures. By visual screening “miraculous” drug for preserving health and
higher productive lines were isolated from the promoting longevity (MISAWA,1994). The ac-
parent callus line and media compositions tive component with proven pharmacological
were optimized. Often, quantification was effects are saponins, e.g., ginsenoside-Rb and
based only on the absorption at 535nm. In -Rg (Fig. 6). Due to the worldwide increased
the case of Chenopodium rubrum we identi- demand for ginseng extracts and the fact that
fied amaranthin, celosianin, and betanin as a cultivation period of 5-6 years is necessary
the main betacyanins (Fig. 5 ) (BERLINet al., before harvest, production of ginseng extracts
1986). Under optimized cultured conditions or saponins using tissue cultures was regarded
-’
35-45 mg L betacyanins were produced in as an alternative source of supply. FURUYA’S
a modified Murashige & Skoog growth me- group analyzed a number of callus cultures of
dium supplemented with tyrosine (BERLINet Panax ginseng with different phythohormone
&
al., 1986). From this line a superior subline requirements (FURUYA, 1988), optimized
was selected by screening. This line produced
not only 3 4 times more betacyanins but also
had an altered pigmentation due to an altered
ratio of betacyanins (with celosianin as the
main compound) (BERLIN,unpublished re-
sults). This line has not yet been optimized
for highest product levels but was successfully
used for the detection of enzymes involved in
celosianin biosynthesis (BOKERNet al., 1991). R3 -- I
The question of whether further optimiza- RZ

tions are justified arises if the only aim is to Fig. 6. Saponin aglycone of Panax ginseng. R, and
make betacyanin tissue culture production R3 are often diglycosides in Rb-ginsenosides. In Rg-
more competitive. Although cell cultures of ginsenoides R2 is glucosylated.
their growth conditions, and compared their of the immunologically active components of
saponin patterns with those of P. ginseng Echinacea drugs, two polysaccharides were
roots. The patterns were quite comparable isolated and characterized (PROKSCHand
and levels of total saponins reached up to WAGNER,1987). The large amounts of these
0.7% of dry mass or 50mgL-' within 4 polysaccharides needed for further in vivo ex-
weeks in a 30 L jar fermenter. It was shown periments were difficult to obtain from the
that some morphological differentiation in- plant material. Therefore, cell cultures of E.
creased saponin formation, and thus a rapidly purpurea were established. The cultures also
growing embryo-like cell line was selected for produced two major active polysaccharides
scale-up to 20 m3 (USHIYAMA, 1991). A bio- and released them into the culture medium;
mass production of 700 mg L-' d -* was ob- however, they were not identical with the po-
tained. lysaccharides of the intact plant (WAGNERet
Since 1988 the use of the tissue culture ma- al., 1989). These cultures were optimized on a
terial has been approved for the Nitto Denko laboratory scale and then scaled up to 60m3
Co. by the Japan Ministry of Welfare and in a cascade of bioreactors (WESTPHAL,
Health. The extracts from the tissue culture 1990). Especially noteworthy was that the
material have been added to wines, tonic specific yield was improved 2-3-fold and that
drinks, soups, herbal liqueurs, and other food the cultivation periods in various fermentors
preparations since 1989 (USHIYAMA,1991). were greatly reduced during the optimization
For the approval, a comparison of the consti- by the process engineers (WAGNERet al.,
tuents of plant extracts and plant tissue cul- 1989; WESTPHAL,1990). The cooperation be-
ture derived material was necessary. In addi- tween researchers of the University of Mu-
tion, some biological assays (Ames test, acute nich, the pharmaceutical company Loma-
virulence test) and dietary tests with livestock pharm, Emmenthal, and the bioengineering
feeding containing high amounts of the tissue company DIVERSA, Hamburg, led to the
culture material had to be performed. It is a production of commercially relevant levels of
little astonishing that the chemical compari- a tissue culture specific drug which entered
son of plant and marketed tissue culture ex- clinical trials. However, DIVERSA has re-
tracts does not contain details about the sa- cently been taken over by Phyton, mainly
ponins and the ratio of Rb:Rg saponins known for its engagement in developing a
(USHIYAMA, 1991). In view of the large-scale production process for taxol@(palitaxel) and
production of ginseng extract using an em- thus it is not clear whether production of the
bryo-like suspension culture originally estab- immunologically active polysaccharides will
lished by FURUYA'S group (USHIYAMA, continue.
1991), it is also somewhat surprising that one
of the latest reports from FURUYA'Sgroup
described the growth of an embryo culture in
various 3 L bioreactors (ASAKAet al., 1993). 3.2 Products of Commercial
-'
This culture produced 0.1 mg L-' d sapon- Interest Accumulating in Traces or
ins.
not at all in Suspension Cultures
3.1.7 Immunologically Active Large efforts have been made in the past to
express compounds which are widely used as
Poly saccharides pharmaceuticals in cell cultures. These in-
clude morphinan alkaloids, tropane alkaloids,
There are a few reports that plant cell cul- quinoline alkaloids, and cardiac glycosides. In
tures sometimes release considerable addition, substantial efforts have been made
amounts of polysaccharides into the culture to develop cell culture systems producing an-
medium. These polysaccharides however, titumor compounds. In this section the pro-
have rarely been analyzed for biological activ- gress in increasing the product levels of these
ity. During investigations on the identification compounds will be analyzed. In contrast to
the previous review (BERLIN,1986) a special claims a production of 2.5 mg morphine and
section on monoterpenoids will not be given 3.0 mg codeine per g dry mass (SIAHand Do-
here. Production of these compounds has re- RAN, 1991). However, considering that these
mained low with callus and suspension cul- very high levels correspond only to 10-
tures (for a review see MULDER-KRIEGER et 14 mg L-' after a period'of 56 d using phyto-
al., 1988; BANTHORPE, 1994). So far, tissue hormone-free medium which must be re-
culture systems have not contributed very placed several times, it is clear that such find-
much to the analysis of lower terpenoid bio- ings are of no biotechnological relevance.
synthesis and its regulation, despite the fact They only show that under extreme stress
that even a low producing tissue culture can conditions the morphinan branch is weakly
be a good source for related enzymes (BAN- expressed. Progress in the production of mor-
THORPE, 1994). Organ cultures are usually phinan alkaloids by undifferentiated, rapidly
much better producers of lower terpenoids growing cell suspension cultures will, there-
(CHARLWOODet al., 1990; BANTHORPE, fore, be made only if factors are identified
1994). Nevertheless, the development of a which prevent expression of the pathway in
biotechnologically viable process for these better growing cultured cells. A breakthrough
compounds is unlikely. could indeed result from the recent detection
of the enzyme channelling (R)-reticuline via
salutaridine into the morphinan alkaloid
3.2.1 Morphinan Alkaloids branch (GERARDYand ZENK, 1993). This
membrane-bound microsomal enzyme is evi-
A number of reports have claimed that dently only expressed in tissues and cells able
morphinan alkaloids are present in callus and to synthesize morphinan alkaloids. This
cell suspension cultures of Papaver somnife- might, however, be changed in the near fu-
rum and P. bracteatum (for a review see ture by genetic engineering.
KAMO and MAHLBERG,1988; KAMIMURA,
1991). However, there are also several reports
where no morphinan alkaloids were detected 3.2.2 Tropane Alkaloids
(BERLIN, 1986; KAMO and MAHLBERG,
1988). The elicitation of poppy cell cultures Hyoscyamine and scopolamine are the
was not initially planned for the production of most important medicinal tropane alkaloids.
sanguinarines but instead for the induction of Hence, several groups have worked on the
morphinan alkaloids. EILERTet al. (1985) production of these compounds in cultured
clearly stated that morphinan alkaloids could cells. However, the alkaloid levels were found
not be induced by any of the elicitors tested. to be extremely low in callus and suspension
Claims in the literature of substantial morphi- cultures of Hyoscyamus, Atropa, Datura, Du-
nan alkaloid levels of up to 0.15% dry mass boisia, and Scopolia. By using a squash tech-
are of no biotechnological relevance since nique in screening for alkaloid producing
such data were not confirmed in further in- clones, YAMADAand HASHIMOTO (1982) se-
vestigations. Evidently, some morphological lected a cell line of Hyoscyamus niger con-
differentiation is required for expression of taining just 0.01-0.02% hyoscyamine per g
the morphinan alkaloid pathway. Thus, fresh- dry weight in suspension cultures. Scopolam-
ly initiated cultures having retained the ca- ine levels were 10-fold lower. Also media var-
pacity to redifferentiate when stressed by eli- iation experiments were not successful in in-
citors or altered phytohormone composition, creasing tropane alkaloid levels. Evidently,
are able to synthesize and accumulate low some root formation is required to express
levels of these alkaloids (KAMO and MAHL- the tropane alkaloid pathway to a reasonable
BERG, 1988). Such cultures usually exhibit extent (see Sect. 4).
poor growth and are not very stable, and thus There has only been one report (BALLICA
no efforts have been made to produce these and RYU, 1994) describing a selected cell sus-
alkaloids in larger volumes. The most recent pension culture derived from stem cells of
report on morphinan alkaloids in cell cultures Datura stramonium which seems to produce
-'
up to 80 mg L total tropane alkaloids with- 3.2.4 Antitumor Compounds
in 25 days. According to the authors, an inte-
grated approach (selection, media optimiza- The search for antitumor compounds in tis-
tion, elicitation, precursor feeding, optimiza- sue cultures has been a favored goal of many
tion of process parameters) is necessary for researchers. The great interest in cell cultures
enhancing productivity in cell cultures. How- of Cathurunthus roseus resulted from the fact
ever, it was not clearly shown that a combina- that this plant species synthesizes in very low
tion of all factors positively affecting produc- amounts (0.0005% of dry mass) the expensive
tivity was indeed necessary to obtain the dimeric monoterpene indole alkaloids vin-
above yield. In any case it does not make blastine (Fig. 7) and vincristine which are the
much sense to continue studies on the optimi- most active agents used in the treatment of
zation of tropane alkaloid formation in sus- certain forms of cancer (MISAWAand ENDO,
pension cultures for production purposes. For 1988). HIRATAet al. (1989) determined a vin-
studying the biosynthesis and regulation of -'
blastine content of 1.4 mg g of dry mass in
this pathway, highly productive hairy root leaves of C. roseus, a level which would re-
cultures of all these species are available (see duce the need for a tissue culture derived
Sect. 4). Using hairy root cultures, the initial process. It is now generally accepted that cell
steps of tropanelnicotine alkaloid formation suspension cultures of C. roseus are not able
have been identified, enzymes have been pu- to synthesize the dimeric alkaloids; vindoline,
rified, and gene cloning is in progress (HA- one of the monomeric precursors, is not
SHIMOTO and YAMADA, 1994). formed in cultured cells but in leaves (DECA-
ROLIS and DELUCA, 1993). Research has
gone into two directions - development of
3.2.3 Quinoline Alkaloids shoot culture systems and semi-syntheses
from the two monomers. However, as with
Cinchona trees have been grown in planta- many shoot culture systems, productivity is
tions since more than 130 years for the pro- presently 100-fold lower than in the leaves of
duction of bark containing the antimalarial, intact plants (HIRATAet al., 1989). More effi-
antifever compound quinine. The related cient is the approach of coupling vindoline
compund quinidine is used as a drug against and catharanthine (derived from plant tissue
cardiac arrhythmia. Though the compounds culture) by chemical or enzymatic means to
have repeatedly been found in cell suspension form 3 ',4 '-anhydrovinblastine and vinblas-
cultures of Cinchona succirubra, C. pubes- tine (GOODBODYet al., 1988). The coupling
cens, and C. ledgeriana at levels of up to 0,9% of catharanthine and vinblastine with peroxi-
(KOBLITZet al., 1983), suspension cultures of dases or'.ferric ions and subsequent reduction
Cinchona could not be further developed for yielded 50-70% anhydrovinblastine and 12-
biotechnological purposes. Attempts to in- 28% vinblastine (DICOSMO,1990). The re-
crease the usually rather low levels by screen- sults obtained by GOODBODYand coworkers
ing and media variation were not very suc- at Allelix looked promising from a commer-
cessful (HARKESet al., 1985). In addition, cial point of view. Nevertheless, the project
growth of cultures of this species is generally was discontinued. Evidently, there was no
slow (WJNSMA and VERPOORTE, 1988). need for developing a tissue-culture-based
Highest alkaloid production was obtained commercial process. High levels of catharan-
with cultures showing some degree of differ- thine and vindoline for chemical coupling are
entiation, e.g., roots, shoots (WIJNSMAand readily isolated as by-products from C. roseus
VERPOORTE,1988), and compact globular plants. Since vindoline is not found in cell sus-
structures (HOEKSTRAet al., 1990). Due to pension cultures, tissue cultures are not
the difficulties of establishing reasonably pro- needed for efficient synthesis of vinblastine.
ductive culture systems, the progress in the Presently, there is a tremendous interest in
elucidation of the biosynthetic pathway lead- developing a tissue culture process for the
ing to quinoline alkaloids has remained slow production of taxole (taxol is a registered
(WIJNSMAand VERPOORTE,1988). trademark of Bristol Myers-Squibb for pacli-
proven to be a very effective drug against sev-

eral forms of cancer. It was approved by the
US Food and Drug Administration (FDA)
for the treatment of ovarian cancer in 1992.
The chemistry and the mbde of action of tax-
01 has been reviewed recently (NICOLAOUet
al., 1994; KINGSTON,1994). A significant
problem is a shortage in taxol supply (CRAGG
et al., 1993). To isolate 1 kg taxol loo00 kg of
Vin blastine (Catharanthus roseus)
dried bark from 3000 Taxus brevifolia trees
have to be extracted. As nearly 2 g taxol are
0 needed for the treatment of one patient it is
evident that research has been initiated to,im-
prove taxol production in trees, to look for al-
ternative sources of taxol, and to develop a
total or semi-synthetic chemical synthesis (NI-
COLAOU et al., 1994; KINGSTON,1994). Pres-
ently, taxol is produced commercially by
semi-synthesis from 10-deacteyl baccatin 111
from needles of young Taxus trees grown in
\=/ plantations. One alternative is to produce tax-
Taxol (Taxus spp.) 01 by using tissue cultures and, indeed, as
pointed out by EDGINGTON (1991) “if ever
there was a plain target for plant tissue cul-
ture, this is it”. In 1991, two American compa-
nies, Phyton Inc. and ESCAgenetics, an-
nounced that the development of their tissue
culture-based processes was to be completed
within 2-5 years. It was impossible to esti-
mate how realistic these claims were, because
both companies did not release scientifically
OCH3 sound information about their processes.
5-rnethoxypodopyllotoxin-4~-D-glucoside Thus, conclusions could only be drawn from
(Linum flavum) the results published by researchers from uni-
versities and from the improvements of sys-
tems with similar productivity obtained when
industry was involved. In the meantime, ES-
CAgenetics dropped out of the race, while
Phyton licensed their tissue culture process to
Bristol-Myers Squibb. It has to be seen
whether the huge fermentation facilities of
Triptolide (Tripterygiumwilfordii) Phyton in Germany will be used for commer-
cial taxol production. In contrast to the
Fig. 7. Antitumor compounds of some higher American companies, the Mitsui Company
plants. has published some of their research results
on taxol production in Taxus cultures (YuKI-
MUNE et al., 1996). According to the pub-
taxel) (Fig. 7). Detected during the screening lished data, the Japanese company again is
program of the National Cancer Institute ahead of all others.
(NCI), it was considered the most interesting Early publications on taxol production
compound among 11OOOO tested structures in tissue cultures of various Taxus species
(NICOLAOUet al., 1994). Taxol has been looked not very promising and did not sup-
port the taxol levels (153 mg L-' after 6 or 1.1 mg L-'d-' taxol were found of which
weeks) claimed in the patent of BRINGIand 90% were extracellular (PESTCHANKER et al.,
KADKADE(1993). WICKREMESINHE and AR- 1996). A breakthrough maybe was the finding
TECA (1993) induced callus cultures of var- that methyl jasmonate is an inducer of taxol
ious Taxus spp. and screened them for taxol biosynthesis in cultures of various Taxus spe-
production. The levels ranged from O.OOO1- cies (MIRJALILIand LINDEN,1996). Addition
0.0131% of dry mass, and older browning cal- of 10pM methyl jasmonate increased taxol
lus produced more taxol than young pale cal- '
productivity 19-fold, from 0.2 mg L - to 3.4
lus. When they established cell suspension mg L-'. YUKIMUNEand coworkers (1996)
cultures from taxol producing callus, the spe- added higher concentrations of methyl jas-
cific taxol content could not be determined monate (100 pM), used a cell line of T. media
because the taxol peak was overlaid by other with a higher capacity for taxol biosynthesis,
compounds (WICKREMESINHE and ARTECA, and observed taxol accumulation after induc-
1994a). From 28 g dry mass 120 pg pure taxol tion for a longer period. Although taxol levels
(4.3 x 10-'%) were extracted. Thus quantita- increased only 5-fold compared with controls,
tive data based only on HPLC determinations a productivity of 110 mg L-' within 14 d was
should be regarded with caution. The group obtained. The specific content of taxol ob-
of DICOSMOat Toronto University published tained was 0,6% and thus exceeded by far the
several articles regarding the optimization of specific contents found in any tissue of the in-
taxol production in Taxus callus and suspen- tact plant.
sion cultures. They reported initial yields of Recently, transformed phytohormone-in-
0.02% for a culture of Taxus cuspidata calli dependent callus cultures were found to con-
(FETT-NETO et al., 1992). Subsequently, it tain taxol levels up to 16 pg g-' dry mass
was shown that the medium composition af- (HAN et al., 1994). Embryo culture has also
fects taxol production and growth. As only been tested for taxol production (FLORES et
percent of controls were given, it is not clear al., 1993). However, taxol and taxane yields
whether the experiments resulted in really were not very different from those reported
higher productivity (FETT-NETOet al., 1993). initially for callus and suspension cultures. A
When the cells were cultivated in shake flasks recent publication suggests roots of hydro-
productivity went down by a factor of 10 in ponically grown Taxus plants as a source for
rapidly growing cultures (FETT-NETOet al., taxol and related taxanes (WICKREMESINHE
1993), and in a more productive suspension and ARTECA,1994b), which would indicate
culture levels of 0.15 mg L-' were detected that transformed root cultures should also be
(FETT-NETOet al., 1994a). Taxol production a suitable source for these compounds. In-
could only be improved a little by feeding of deed, a recent patent application of Celex-La-
potential precursors and thus, levels of 0.01 % boratories (1994) describes the production of
dry mass were found (FETT-NETO et al., taxol via hairy root cultures. Studying the
1994b). Substantially higher levels were re- published literature on taxol production in
ported by two groups from Cornell Univer- cell culture systems (for a review, see JAZIRI
sity, Ithaca (the city where Phyton is located); et al., in press), one might conclude that most
during culture media optimization experi- cell lines of all Taxus species contain low lev-
ments they reached levels of up to 15 mg L-' els of taxol, that some more productive lines
taxol using lines of T. baccata, T. canadensis, can be obtained by screening, that methyl jas-
and T. cuspidata (HIRASUNAet al., 1996; monate is presently the only unambigously
KETCHUMand GIBSON,in press). However, proven enhancer of taxol synthesis, and that
substantial variations of yields in individual production levels of higher producing lines
experiments were observed making interpre- often oscillate to an unacceptable extent.
tations quite difficult. To overcome the oscil- Nevertheless, if the published production
lations of taxol production might be a severe rates can reproducibly be obtained in large
problem in developing a large-scale process. volumes of several thousand liters and if they
In a bioreactor experiment with a working are optimized further by biochemical engi-
volume of 600 mL levels of up to 22 mg L - ' neers, culture systems may compete with
semi-synthesis. More knowledge about the Other podophyllotoxin producing cell cul-
biochemistry, regulation and limiting steps of tures are those of L. album (SMOLLNY et al.,
the taxol pathway may eventually help to en- 1993) and P. hexandrum (WOERDENBAG et
gineer lines with improved productivity al., 1990). The levels of podophyllotoxins vary
(SRINIVASAN et al., 1996). They provide from 0.05 to 0.3% of dry mass depending on
some evidence that taxol biosynthesis is lo- the culture conditions and the tendency to
cated in plastids. They also conclude that the differentiate. Thus root cultures (producing at
conversion of phenylalanine to phenylisoser- least 1% podophyllotoxins in dry weight)
ine is a rate-limiting step of taxol biosynthe- seem to be most suitable for future research
sis. It might be important to stimulate the flux efforts, e.g., for biochemical studies (OOST-
of the primary precursors into the taxane HAM et al., 1993).
skeleton. Previously, it was assumed that me- A recent review by MISAWAand ENDO
valonate is the precursor of the isoprenoids. (1988) describes the productivity of cell cul-
However, the findings of EISENREICH et al. ture systems for other potential antitumor
(1996) suggest that the isoprenoid moieties of compounds produced by plants and plant cell
the taxane molecule are derived from a yet cultures. These include camptothecine, homo-
unknown precursor which is likely to be de- harringtonine, and maytansine. It must be
rived from a novel pathway of isoprenoid bio- concluded that the levels found in the culture
synthesis (using triose phosphate-type com- systems are far too low to be attractive from a
pounds and activated acetaldehyde), recently biotechnological point of view. The same
detected by ROHMERet al. (1993). CRO- holds true for tripdiolide and triptolide
TEAU’S group demonstrated that taxa- (Fig. 7) produced at 0.01% of dry mass by
4(5),11(12)-diene is the first intermediate in Tripterygium species (TAKAYAMA, 1994).
taxane biosynthesis, and recently they have
purified the corresponding enzyme taxadiene
synthase cyclizing the universal diterpene 3.2.5 Cardiac Glycosides
precursor geranylgeranyl pyrophosphate to
taxa-4(5),11(12)-diene in a single step (HE- The formation of cardiac glycosides in un-
ZARI et al., 1995). Due to the importance of differentiated cell cultures of various Digitalis
taxol it is predicted that further progress in species was found to be very low or even
the elucidation of taxol biosynthesis and its lacking in rapidly growing suspension cul-
regulation will soon be made and that geneti- tures. In slowly growing green callus cultures
cally engineered lines might be created in the or in embryogenic cultures some cardiac gly-
near future. coside accumulation was observed (for review
Another important, commercially used an- see LUCKNERand DIETTRICH,1988). How-
titumor compound is etoposide, a semisyn- ever, low- or non-producing plant cell suspen-
thetic podophyllotoxin. Production of podo- sion cultures of Digitalis fanata are effective
phyllotoxin by tissue cultures of Podophyllurn in the biotransformation of cardenolides and
peltatum was first attempted by KADKADE for this reason they are still rather attractive
(1982). However, the levels remained disap- from a biotechnological point of view (see
pointingly low. The interest in podophyllo- Sect. 6.2).
toxins was revived when root cultures of Lin-
um flavum were found to contain high levels
(1% of dry mass) of 5-methoxypodophyllo- 3.2.6 Vanillin and Vanilla Aroma
toxin (Fig. 7) and its glucoside (BERLINet al.,
1988b). Two Dutch groups confirmed this re- Vanilla is probably the most widely used
sult and showed that suspension cultures of L. flavor in food industries with a worldwide an-
flavum also contained some podophyllotoxins nual consumption of 1200-1 500 t in 1993
(VAN UDEN et al., 1990; WICHERSet al., (HAVKIN-FRENKEL, 1994). The prices for one
1991). However, reasonable productivity re- kg natural vanillin from cured beans of Vunil-
quired some morphological differentiation, laplanifolia amount to $3000-4000 while the
e.g., root formation (VANUDENet al., 1991). price for synthetic vanillin is less than $20 per
kg. The flavor of natural vanilla extracts from tered commercial development”. Unfortu-
beans of different origin varies. Though vanil- nately, the company was not willing to sup-
lin is the most important constituent of the port their announcements by providing any
aroma other compounds seem to affect sub- scientific data or comments of their “mar-
stantially the taste of this flavor. Due to the keted” product for this review. Thus, some
high price of natural vanilla flavor plant tissue doubts remained as to whether “PhytoVanil-
cultures, mainly of V. plunifoliu, have been laTMis considerably more economic to pro-
regarded as an alternative source for the produce than natural vanilla extract” (GOLD-
duction of vanilla aroma and vanillin. The STEIN, ESCAgenetics). It has to be seen
first publication on phenylpropanoid metab- whether another company will continue the
olism in V.plunifolia showed that vanillin was process of vanilla aroma production after the
not produced under normal culture condi- shutting down of ESCAgenetics.
tions (FUNKand BRODELIUS,1990)”Forma- Based on the literature and the patent it is
tion of vanillic acid, however, was observed now clear that vanillin production can be in-
when 3,4-(methy1enedioxy)-cinnamic acid, an duced and optimized in cultured cells of Vu-
inhibitor of p-coumarate CoA ligase, was ad- nilla spp. by media composition, selection, eli-
ded to the suspension cultures. This led to a citation, and feeding of suitable precursors.
shift from lignin to benzoate biosynthesis. The highest specific yields (0.16% vanillin on
After feeding potential cinnamic acid precur- a dry mass basis) were reported for an em-
sors there were indications that the cell cul- bryo culture of V. plunifoliu grown in small
ture must contain enzymes involved in ben- bioreactors (KNORR et al., 1993; HAVKIN-
zoate biosynthesis. Finally, it was found that FRENKEL,1994). A comparison of an HPLC
kinetin is an efficient elicitor of vanillic acid extract of the embryo culture and a Bourbons
formation (FUNK and BRODELIUS,1992). vanilla bean extract exhibited much more
Vanillic acid levels of up to 0.1% of dry mass similarities of the components than shown in
were detected. Thus, according to the litera- the patents of ESCAgenetics. HAVKIN-FREN-
ture, one would assume that cultures of V. KEL (1994) also compared the production
plunifoliu are of no commercial interest. It costs. If the cultures produce 2% vanillin in
was, therefore, somewhat surprising that large vessels, the estimated costs for the pro-
ESCAgenetics (KNUTH and SAHAI, 1991) duction of 1 kg vanillin would be between
filed a patent in 1988 (issued in 1991)for the $500-1000 (investment for bioreactors not in-
production of vanilla aroma (PhytoVanil- cluded). Although the embryo culture system
laTM)and biosynthesized vanillin (Phytovan- contains the highest published content of va-
illinTM)by suspension cultures of V. frugruns. nillin reported up to now it is commercially not
The aroma compounds are secreted into the yet competitive. In this context it should be
medium and bind to absorbents. The patent mentioned that alternative biotechnological
states that 16-18 mg vanillin per L of medium approaches for the production of vanillin
were produced within ca. 45 days. A compari- have been investigated (CHEETHAM,1993).
son of the vanilla flavor profiles of beans with For example, root tissue of V. pfunifofiucon-
that of tissue cultures as presented in the pa- verts ferulic acid to vanillin at production
tents reveals that vanillin is the main compo- -’ -’
rates of 400 mg kg d root tissue and con-
nent in both extracts, but that otherwise the centrations of 7 g kg-’ roots were obtained
composition of the two extracts is quite differ- (WESTCOTTet al., 1994). This suggests that
ent. Thus, one can conclude that the suspen- hairy root cultures should be tested for the
sion cultures produce a new vanilla aroma for production of vanillin by biotransformation.
which a market may or may not exist. ESCA- The production of vanillin and related com-
genetics announced that they would be able pounds may also be improved in the near fu-
to produce tissue culture flavors such as vanil- ture on the basis of present studies on the en-
la at an incredibly low price of $50-100 kg-’ zymology of the formation of benzoic acids
(MOSHY et al., 1989). STAHLHUT(1993) re- from cinnamic acids (LOSCHERand HEIDE,
ported that the vanilla flavor from suspension 1994).
cultures of V. fragruns “exited the lab and en-
4 Secondary Product Formation in Hairy Root Cultures 619
4 Secondary Product are accessible to biochemical and molecular

studies. Indeed, they are also amenable to ge-
Formation in Hairy Root netic manipulations.
The first report on the initiation of hairy
Cultures root cultures for production purposes was in
1985 when FLORESand FILNERshowed that
Hyoscyamus spp. hairy root cultures pro-
The attempts at enhancing product forma- duced 0.5% tropane alkaloids on a dry mass
tion in rapidly growing suspension cultures basis and those of tobacco contained more
showed that some pathways are spontaneous- than 3% nicotine. This was followed by a
ly well expressed under such conditions. Fur- huge number of reports on the initiation of
thermore, their product formation can easily metabolite producing hairy root cultures.
be optimized by rather simple methods. Poor- Thus, this review cannot cover all the produc-
ly expressed pathways, however, do not re- tive hairy root culture systems which have
spond well to manipulations aimed at obtain- been described in the recent years (for a re-
ing productivity improvements. There are un- view or literature see SIGNSand FLORES,
fortunately many more compounds, not men- 1990; RHODESet al., 1990 PAYNEet al., 1991;
tioned in the previous sections, which are SAITOet al., 1992; TOIVONEN,1993; BAN-
only formed at low levels in undifferentiated THORPE, 1994). Naturally, plant species which
cell cultures. After induction of some mor- do not produce the desired commercially im-
phological differentiation, however, enhanced portant compounds in suspension cultures
biosynthesis of many of these compounds was were the main target. The greatest biotechno-
observed. Organ cultures of plant cells, e.g., logical efforts have been put into hairy root
root and shoot cultures, are known for quite cultures of tropane alkaloid producing Sola-
some time. However, from a biotechnological naceae. According to the literature several
point of view these cultures were of low inter- hundred individually transformed hairy root
est because they are difficult to establish, they cultures have been established and analyzed
often grow slowly and are sometimes very un- with respect to their ability to form tropane
stable. An overview of the nevertheless sub- alkaloid (KNOPPet al., 1988; PARR et al.,
stantial studies on the accumulation of sec- 1990) and hairy root cultures of all known
ondary compounds by organized shoot and plants (e.g., Arropa, Datura, Hyoscyamus,
root plant cultures has been given by CHARL- Brugmansia, Duboisia, Scopolia) producing
WOOD et al. (1990). The image of organ cul- these alkaloids have been established. In gen-
tures has greatly changed since the finding eral, it was found that hairy root cultures con-
that for most dicotyledonous plants rapidly tained roughly the same amount of alkaloids
growing hairy root cultures can be initiated as that found in the roots of the correspond-
by transformation with Agrobacterium rhizo- ing intact plant species. Root cultures may
genes. Besides the progress in the biochemis- show some variation in growth rate, alkaloid
try and molecular biology of plant secondary content, and productivity (MANO et al.,
pathways the broad application of hairy root 1989). Thus, screening for a high producing
cultures has been the most important stimulus line seems to be worthwhile. However, the
for the field. The special characteristics of productivity differences between 60 different
transformed hairy root cultures in compari- lines of one species were only 2-3-fold on av-
son with normal root cultures have been re- erage. The medium composition is also im-
viewed by RHODESet al. (1990). They are portant for growth and alkaloid production
phytohormone-independent, exhibit rapid (MANOet al., 1989; BERLINet al., 1990). In
growth and a high degree of genetic and bio- particular, the amount of nitrogen and the ra-
chemical stability. Their biosynthetic capacity tio NOq/NH$ seem to affect branching of
is equivalent to the corresponding plant root, roots and product formation. With the addi-
and they can be grown in bioreactors. The tion of phytohormones to the growth medium
most important biotechnological value of of root cultures, more callus-like root cultures
these cultures is perhaps that many pathways are formed which produce much lower levels
of alkaloids. In contrast to suspension cul- been used for growing hairy root cultures.
tures where improved production is often ob- Good results were obtained with mechanical-
served under growth limiting conditions hairy ly agitated fermentors (KONDOet al., 1989;
root cultures produce best when root forma- BUITELAARet al., 1991). High growth rates
tion and growth is optimal. In agreement with were also achieved in reactors in which the
this elicitors (stress factors) had no great ef- roots were fixed to supports of stainless steel
fect on product levels although it induced or polyurethane foam and sprayed with nu-
phytoalexin production (FURZEet al., 1991). trient (WILSONet al., 1990; Dr IORIO et al.,
In general, it was found that the specific pro- 1992). A technical problem is the scale-up of
duction of secondary metabolites cannot be hairy root cultures from small to very large
manipulated as readily as in productive sus- bioreactors. Hairy roots form usually large
pension cultures (BERLINet al., 1990 TOI- aggregates which cannot be pumped through
VONEN,1993). pipes from one reactor to another. We sug-
In theory, all dicot plants should be suscep- gested (BERLINet al., 1990) to grow the phy-
tible to transformation by A. rhitogenes and tohormone-independent hairy root cultures
should yield good growing hairy root cultures. initially in the presence of phytohormones.
However, some plant species have been The resulting callus-root culture with substan-
found to be rather recalcitrant. For example, tially shorter roots would be used for scale
transformation of Pupuver somniferum with up. In the last bioreactor full root formation
A. rhizogenes resulted in phytohormone-in- could then be reinduced by omission of the
dependent cell suspension cultures rather phytohormone. TAKAYAMA et al. (1994) de-
than root cultures (WILLIAMSand ELLIS, monstrated the feasibility of this approach for
1993). Since normal root cultures of Linum a hairy root culture of Hyoscyumus niger
fravum produce high levels of 5-methoxypo- which was scaled up in the presence of
dophylotoxin (BERLINet al., 1988b) it was 0.3mgL-' NAA in an agitated bioreactor.
logical to initiate hairy root cultures of this They obtained a root suspension which could
species. A successful transformation has been be transferred through pipes (13 mm internal
reported only once (OOSTHAMet al., 1993). diameter) from one reactor to the next. Tro-
However, this culture exhibited a lower pane alkaloid production (3% of dry mass)
growth rate than normal root cultures. Our reached the original levels after transfer to
own efforts to establish hairy root cultures of the phytohormone-free medium.
L. flavum have not yet been successful (BER- Sometimes the products of hairy root cul-
LIN and KUZOVKINA, unpublished results). tures are spontaneously released into the me-
In addition to the usefulness of hairy root dium, from which they can easily be recov-
cultures in biochemical and molecular studies, ered. A scopolamine secreting line of Duboi-
they might also be useful for the production siu leichhurdtii was cultivated in a bioreactor
of desired metabolites in large-scale volumes with continuous exchange of the medium.
due to their good productivity. A recent re- The product was recovered by using a XAD-2
view on the cultivation of root cultures in column. Under optimized conditions (stain-
bioreactors (CURTIS, 1993) shows that the less steel mesh as a support, turbine-blade
process engineers are just at the beginning of reactor, two-stage culture for growth and
developing suitable reactors and operation product release) a total of 1.3 g L-' scopol-
schemes. Several hairy root cultures have amine was recovered from the XAD-column
been grown in bioreactors with capacities of during 11 weeks of continuous operation
1-20L. For example, hairy root cultures of (MURANAKAet al., 1993). High levels of
Coleus forskohlii grown in 20 L glass jar fer- thiophenes are only produced in root cultures
mentors produced forskolin (a novel heart ac- of Tugetes ssp. BUITELAARet al. (1991) grew
tive and blood pressure-lowering compound) hairy roots of 7'.putulu with 1.6% thiophenes
with a yield of 14mgL-' after 3 weeks on a dry mass basis in various bioreactors and
(KROMBHOLZ et al., 1992). In suspension cul- continuously harvested 70% of the lipophilic
tures the forskolin content was extremely low. compounds in a two-liquid-phase system. In-
Many different reactor configurations have terestingly, product formation and secretion
6 Biotransformations with Cultured Plant Cells 621
of secondary products into the medium was chances of a successful outcome of this ap-
often enhanced when two-phase culture or proach were questionable from the beginning.
absorption to resins was used (TOIVONEN, In the meantime, the Nattermann Company
1993). has given up this approach. Although the
number of “new” compounds found exclu-
sively or originally in tissue cultures has
steadily increased (RUYTERand STOCKIGT,
1989), the importance of cell cultures as a
5 Plant Tissue Cultures as source of novel compounds has not in-
creased.
a Source of New
Chemicals?
Plant tissue cultures are not usually used as
producers for commercially interesting com- 6 Biotransformations with
pounds since alternative production methods
(field grown plants, chemical synthesis) have Cultured Plant Cells
been established by industry. The observation
that under cell culture conditions new metab- The previous sections dealt mainly with the
olites sometimes accumulate which had not de novo synthesis of secondary metabolites by
previously been detected in the intact plant cell suspension and hairy root cultures. The
suggested that plant cell cultures may be used production of valuable plant products from
as a source of new chemicals. In addition, cheap precursors by biotransformation is an-
components which are present in the intact other possibility of using plant tissue cultures.
plant at low levels were found to be present Indeed, the 12-p-hydroxylation of p-methyl-
at high levels in cultured cells in some cases, digitoxin to P-methyldigoxin by cell suspen-
for example, sanguinarine in poppy plants sion cultures of Digitalis lanata was expected
(WILLIAMSand ELLIS, 1993). Dedifferentia- to become the first commercial tissue culture
tion of the cultured plant cell leads to a regu- process (REINHARD and ALFERMANN, 1980).
latory state which is not present in the devel- This example showed that even cultures un-
oping plant, and this may lead to the expres- able to synthesize a particular compound de
sion of novel product patterns (ARENSet al., novo can carry out specific reactions of the
1982). The idea of screening plant cell cul- corresponding pathway. Many other exam-
tures for novel pharmacologically active com- ples demonstrate the potential of plant cell
pounds was mainly followed by the Natter- cultures to biotransform exogenous substrates
mann Company in Cologne and researchers (for a review see SUGAand HIRATA,1990).
of the Pharmaceutical Institute of the Univer- However, the main question remains as to
sity of Munich. Indeed, some new biologically whether plant cell cultures can compete with
active compounds were detected by such the well-known capabilities of microorgan-
screenings (ARENSet al., 1986). However, isms in the field of biotransformations. Thus,
from a biotechnological point of view this ap- one should look for plant-specific biotransfor-
proach would only be successful if the new mation reactions which are not performed by
compound had unique (superior) pharmaco- microorganisms. Glycosylation of foreign
logical characteristics and if product levels products, for example, is a reaction which is
were high enough to encourage further devel- not performed by bacteria. Plant cells should
opment. Taking into account the number of also be superior to microorganisms if stereos-
extracts which had to be screened by the NCI pecific enzymatic reactions of plant specific
program before a unique compound such as pathways are required. The two examples giv-
taxol was detected, and that product patterns en below demonstrate how plant cell cultures
and formation are much lower in cultured can be used for biotechnologically relevant
cells than in intact plants (BERLIN,1986), the biotransformations.
6.1 Arbutin tent in therapy, and therefore conversion of

digitoxin to digoxin by 12-P-hydroxylation
Arbutin has been used as a urethral disin- was regarded as a biotechnologically useful
fectant for many years and has been shown to biotransformation reaction. Cell cultures of
be a potent suppressor of melanin synthesis in D. lanata, although unable to produce cardiac
human skin. It is thus used as a skin depig- glycosides de novo, were found to contain
mentation agent. Although arbutin is found high 12-phydroxylation activity (REINHARD
at rather high levels (54%) in several mem- and ALFERMANN, 1980). However, p-methyl-
bers of the Ericaceae, it is presently produced digitoxin was initially found to be the only
by chemical synthesis. The Japanese Shiseido substrate which was exclusively converted
Company is trying to develop an alternative into a digoxin derivative. With all other digi-
production process by the glucosylation of hy- toxins, side reactions were observed. The use
droquinone added to cell suspension cultures of pmethyldigitoxin as a substrate seemed to
of Catharanthus roseus, although arbutin is be feasible since it was already being pro-
not a natural constituent of this plant (Yo- duced by Boehringer Mannheim. This compa-
KOYAMA and YANAGI,1991). First the cul- ny initially was interested in applying cell cul-
ture stage during which highest glucosylation tures of D. lanata in the biotransformation of
occurs was determined. Then the medium pmethyldigitoxin. However, the idea of using
composition was optimized, and the biotrans- this technology for production was dropped
formation efficiency of various strains and in the early 1980s. Nevertheless, research on
cultures were compared. By fed-batch cultiva- this system has continued. REINHARD and co-
tion and continuous feeding of the high densi- workers at the University of Tubingen optim-
ty cultures with substrate (14 d old, 5 L im- ized the biotransformation up to a 210 L
peller driven bioreactor) arbutin formation working volume in an airlift reactor (REIN-
was improved to nearly 1OgL-' within 4 HARD et al., 1989). During a semicontinuous
days. A specific content of 45% arbutin per g process (6 repeated batch cultivations) 513 g
dry mass was tolerated by the cells which is pmethyldigoxin was produced from 641 g P-
the highest specific content of a secondary methyldigitoxin within 3 months. Since the
metabolite reported for cultured plant cells. It product is released into the medium further
was stated that the production costs for arbu- improvements in shortening the production
tin by chemical synthesis or biotransforma- period are expected. Optimization of the bio-
tion in tissue cultures are very similar (Yo- transformation process did not only include
KOYAMA and YANAGI,1991). They are rath- studies on optimal substrate supply and me-
er optimistic that the biotechnological pro- dium composition but also selection of an im-
duction of useful glucosides will become in- proved strain for highest hydroxylation ca-
creasingly inexpensive relative to chemical pacity by repeated cell aggregate cloning
synthesis. Recently, it was shown that cell sus- (REINHARD et al., 1989).
pension cultures of Rauwolfia serpentina are When digitoxin was added to cell cultures
also able to form efficiently arbutin from hy- of D. lanata two reactions occurred 12-phy-
droquinone (18 g L-' in 7 d) (LU'TTERBACH droxylation and 16'-O-glucosylation yielding
and STOCKIGT,1992). deacetyllanatoside C as the main product.
KREIS and REINHARD(1990) developed a
two-stage semicontinuous process by transfer-
6.2 Biotransformation of Cardiac ring a selected cell line into a production me-
dium (8% glucose) for the biotransformation
Glycosides reaction. On average 400 mg L-' deacetylla-
natoside C were produced from 600mg digi-
Digitalis lanata plants contain two main toxin in a 20 L airlift fermentor within 7 d. To
cardiac glycosides which upon hydrolysis achieve a high biotransformation of digitoxin
yield digoxin and digitoxin, drugs which are into one major cardenolide, it was important
widely used in the treatment of heart dis- to change from a growth to a production me-
eases. Digitoxin is used to a much lesser ex- dium. In contrast to pmethyldigoxin, deace-
tyllanatoside C is mainly found within the genetic engineering techniques. Indeed, sev-
cells. The main goal, however, was to develop eral groups worldwide have used their culture
a biotransformation production process from systems (as indicated in the previous sections)
digitoxin to digoxin by preventing the 16'-0- not only for optimizing product levels but
glucosylation. Since the optimum tempera- also for biochemical and molecular studies.
ture of 12-p-hydroxylase is around 20°C while Thus, our knowledge of the enzymology of
that of glucosyltransferase is 37"C, lower tem- several pathways has greatly increased during
peratures (e.g., 19OC) would therefore favor the last few years. The genes coding for sev-
hydroxylation (KREISand REINHARD, 1992). eral of these enzymes have been cloned. For
Deacetyllanatoside is mainly stored in the the phenylpropanoid/flavonoid pathway,
vacuoles of the cells while digoxin is released which has been analyzed the most extensive-
into the medium. The lower the cell density ly, the first regulatory elements have been de-
was, the lower was the storage capacity for tected (FORKMANN, 1993; KOES et al., 1994).
deacetyllanatoside and the higher was the Thus, we seem to be at the beginning of an
percentage of digoxin formation. Based on era when aimed manipulations of secondary
this knowledge a semicontinuous production pathways might become possible. This might
process in a 300 L airlift reactor was devel- also tremendously increase the production
oped with 8% glucose as a production me- potential of plant cell cultures and thus their
dium and incubation times of 40-60 h. From biotechnological impact. However, progress
0.8 mmol digitoxin, ca. 0.6 mmol digoxin were in this direction is likely to be slow since even
obtained, 80% of which was found in the me- in microbial systems the success of metabolic
dium (KREISand REINHARD,1992). The lat- engineering is not yet overwhelming (CAME-
ter result demonstrates in particular the im- RON and TONG,1993; STEPHANOPOLOUS and
portance of understanding the physiology and SINSKY,1993). In order to devise rational ap-
biochemistry of a pathway if natural products proaches for metabolic engineering a knowl-
are used in biotransformation processes edge of the limiting steps of the pathways is
(KREISet al., 1993). required (STEPHANOPOULOS and SINSKY,
1993). Locating such critical steps is often
rather difficult. In the case of non-expressed
pathways in plant cells, a one-step manipula-
tion will usually not be sufficient. Neverthe-
7 Metabolic Engineering less, the first genetic manipulations of second-
ary pathways in plant cells have been per-
of Secondary Pathways in formed. However, only a few of them have
been directed towards the production of sec-
Cultured Cells ondary metabolites. In most cases, the estab-
lishment of transgenic plants rather than cul-
Many reviews have pointed out that more tures with altered characteristics has been at-
knowledge of the regulation of secondary tempted. As the expression of secondary
pathways at the enzyme and gene level is pathways is quite different in intact plants
needed before the true potential of cultured and in cultured cells, the metabolic effects of
plant cells for metabolite production can be a genetic transformation in these systems will
recognized and exploited. The above analysis also be different. This review focusses on the
of the field clearly supports this view. Despite manipulation of product formation in cul-
all the progress, the fact remains that the biol- tured cells which might later be grown in
ogy of the cell is the limiting factor since only bioreactors. Thus, only the results of trans-
those pathways which are spontaneously well genic cell cultures which were created for the
expressed under culture conditions or easily overproduction of desired compounds by var-
inducible by simple modifications are accessi- ious biosynthetic pathways will be described
ble for the development of biotechnological in some detail. Nevertheless, it should be
processes. This can only be changed if we suc- mentioned that genetic engineering of trans-
ceed in altering the expression of pathways by genic plants with desired characteristics is in
full progress. For example, Calgene and Sun- of Catharanthus roseus (GODDIJN, 1992) was
tory have announced the development of blue introduced into P. harmala cells under the
roses (HOLTONand TANAKA,1994). A gen- control of the 35s promoter of the Cauliflow-
eral overview of metabolic engineering of er Mosaic Virus via Agrobacterium tumefa-
commercially useful biosynthetic pathways in ciens (BERLINet al., 1993). Several cell sus-
transgenic plants and plant cells were given pension cultures were obtained with constitu-
by KISHOREand SOMERVILLE (1993) and tively expressed TDC activities of around
NESSLER(1994). 30 pkat/mg protein and serotonin levels of 1-
2% of dry mass; the serotonin levels of the
controls were below 0.1%. It was also shown
7.1 Serotonin Biosynthesis in that in these transgenic cultures tryptophan
supply is the next rate-limiting factor (BER-
Peganum harmala LIN et al., 1993) which has to be overcome to
obtain even higher levels of serotonin. Al-
Cell cultures of Peganum harmala loose though serotonin biosynthesis is not repre-
their capability of synthesizing serotonin sentative of a typical secondary pathway be-
(Fig. 8) during prolonged subcultivation. It cause of its simplicity, this example shows
was found that this was due to the lack of that removal of a rate-limiting step (clearly
tryptophan decarboxylase (TDC) activity in identified by the feeding experiments with
fully undifferentiated cell suspension cultures. tryptamine) can help to increase the produc-
The second and final enzymatic step of sero- tion rate.
tonin biosynthesis, the 5-hydroxylation of The same tdc gene was also overexpressed
tryptamine, however, remains highly ex- in C. roseus, the plant from which it was de-
pressed in all P. harmala cell cultures (SASSE rived. However, as expected from literature
et al., 1987; COURTOIS et al., 1988). Thus, it data, the constitutive expression of TDC only
was a rational approach to introduce a consti- led to a large accumulation of tryptamine and
tutively expressed tdc gene into P. harmala did not affect alkaloid levels (GODDIJN,
cell suspension cultures to restore or maintain 1992). Feeding of tryptamine does not en-
serotonin formation. A corresponding cDNA hance alkaloid formation in Catharanthus cell
suspension cultures. The overproduction of
tryptamine by the constitutive expression of
the tdc gene had evidently no negative effect
on tryptophan supply for protein synthesis.
Anthranilate synthase levels (the proposed
H regulatory enzyme of tryptophan biosynthe-
Serotonin sis) were unchanged in transgenic tobacco
d?
plants (POULSENet al., 1994). An interesting,
yet inexplicable observation was made by
SONGSTAD et al. (1991). They found that rdc-
transgenic tobacco plants not only overpro-
N duce tryptamine but also produce high levels
of tyramine, although the engineered enzyme
Anabasine
does not accept tyrosine as substrate.
7.2 Affecting Nicotine Alkaloid

- Biosynthesis in Tobacco
FH~OH
Scopolamine
The regulation of nicotine biosynthesis is
Fig. 8. Secondary metabolites whose levels were en- relatively well understood (for a review see
hanced in cultured cells by metabolic engineering. HASHIMOTO and YAMADA,1994). The activ-
ity levels of putrescine N-methyltransferase zyme had been targeted to leucoplasts, the
(PMT) determine whether reasonable levels site of lysine biosynthesis in root cells. High
of nicotine can be formed in tobacco suspen- levels of cadaverine (0.5% of dry mass; con-
sion cultures. As this enzyme is also a key en- trol cultures, ca. 0.01%) and anabasine (0.5%
zyme in tropane alkaloid biosynthesis, great of dry mass; controls, 04.02%) were found in
efforts have been made to clone this gene, the most productive root culture (HER-
even by researchers from industry. Philip MINGHAUS et al., 1996). Feeding of lysine en-
Morris (NAKATANIand MALIK,1992) filed a hanced both cadaverine and anabasine levels
patent on the production of transgenic tobac- to more than 1% of dry mass while metabol-
co plants overexpressing PMT. However, re- ite levels of control cultures were hardly af-
ports on such transgenic plants overproducing fected (HERMINGHAUS et al., 1996). Nicotine
PMT have not yet been published. was still the main alkaloid in these cultures.
The first attempts to affect nicotine biosyn- However, anabasine became the second most
thesis genetically have used the gene of a abundant alkaloid. Improving the internal
yeast ornithine decarboxylase (odc) (HAMILL supply of lysine by altering the feedback con-
et al., 1990). This enzyme activity, however, is trol of lysine biosynthesis (SHAULand GAL-
not generally regarded as a rate-limiting step ILI, 1991) might lead to even higher levels of
of the pathway since feeding putrescine does anabasine. Lysine decarboxylase expressed in
not affect nicotine levels. Transgenic root cul- the leaves resulted in the production of cad-
tures of Nicotiana rustica constitutively ex- averine only (HERMINGHAUS et al., 1996).
pressing the yeast gene under the control of This shows that genetic manipulation at one
the CaMV35S promoter contained distinctly biosynthetic node will lead to significant im-
enhanced ODC activity. However, the high provements of product levels only when sev-
enzyme activity did not result in correspond- eral prerequisites are fulfilled. Firstly, the tar-
ing high levels of nicotine as product levels get enzyme must be the rate-limiting step;
were only increased 2-fold. In view of the fact secondly, the engineered enzyme must be suf-
that putrescine is not a limiting intermediate ficiently supplied with substrate; the reaction
in nicotine biosynthesis, the outcome was not product must be transported to the site from
so surprising. It is not yet clear whether the where it is channelled into the related path-
2-fold increase is really due to the increased way. The latter means that the other enzymes
enzyme activity. Feeding of the cultures with of the pathway should be expressed quite
ornithine could provide more insight into the well. Therefore, genetic manipulations in un-
influence of the engineered ODC on the nico- differentiated cells, in organ cultures, or in
tine levels. whole plants might yield completely different
Anabasine (Fig. 8) is an analog of nicotine results depending on the overall expression of
in which the N-methyl pyrrolinium ring is re- a biosynthetic pathway in various tissues.
placed by a A l-piperideinium ring. This ring is
derived from lysinekadaverine, and it has
clearly been shown that cadaverine is the 7.3 Enhancing Scopolamine
rate-limiting intermediate in anabasine bio-
synthesis (WALTON and BELSHAW,1988). Production in Atropa belladonna
Therefore, a bacterial lysine decarboxylase
(Idc) gene was introduced into tobacco root Hyoscyamine 6p-hydroxylase catalyzes the
cultures (BERLINet al., 1994, HERMINGHAUS oxidative reactions in the biosynthetic path-
et al., 1996). When the gene was placed under way leading from hyoscyamine to scopolam-
the control of the CaMV35S promoter and ine (Fig. 8). A cDNA coding for this enzyme
fused to the coding sequence of the transit with epoxidating and hydroxylating activities
peptide of the small subunit of ribulose di- was cloned using RNA from a root culture of
phosphate carboxylase, many root cultures Hyoscyamus niger (reviewed by HASHIMOTO
with LDC activity were obtained. This en- and YAMADA,1994). The cDNA was cloned
zyme activity was not detected in any of the into a binary vector for plant transformation
control cultures. It was shown that the en- under the control of the CaMV35S promoter
and integrated into Agrobacterium rhizogenes formation. Thus, one can understand and
15834 (HASHIMOTO et al., 1993). Hairy root share the sceptical view of HASHIMOTO and
cultures were obtained by infecting Atropa YAMADA(1994) that single site manipula-
belladonna leaf disks. Hyoscyamine 6P-hy- tions of secondary pathways will usually not
droxylase activities were 3- to 5-fold in- be sufficient for high increases of pathway
creased in the transgenic root cultures com- end product levels. On the other hand, one
pared to the controls. Scopolamine levels of should realize that we are just at the begin-
the controls were ca. 0.05% of dry mass, hyo- ning of a new era. Therefore, conclusions
scyamine being the main alkaloid. In the best should not be drawn before the first double
transgenic lines scopolamine levels were up to transformants have been obtained. In addi-
5-fold higher while the hyoscyamine content tion, the product levels which are not over-
was greatly reduced (HASHIMOTOet al., whelming so far may increase further when
1993). However, there is evidence that the ex- expression and targeting of the engineered
pression of the engineered enzyme is not yet protein is optimized.
optimal in root tissues. When the vector was The greatest potential of metabolic engi-
introduced into an Agrobacterium tumefa- neering for improving the standing of plant
ciens strain transgenic A. belladonna plants cell cultures will, however, only become ap-
which accumulated high levels of scopol- parent if we succeed in identifying regulatory
amine in leaves and stems were obtained; the genes which control the expression of whole
amount of scopolamine formation in the roots pathways in cultured cells. It is possible that
of the same plants was not very high (YUN et pathways which are not usually in operation
al., 1992). might become active in undifferentiated cells.
We are still far away from the identification
of elements which regulate the tissue- and or-
gan-specific initiation of a whole secondary
7.4 Does Genetic Engineering pathway. However, a promoter analysis and
Improve the Potential of Plant the search for transduction signals and tran-
Tissue Cultures as Producers of scription factors involved in the regulation of
secondary pathways have already been
Interesting Metabolites? started (DIXONand LAMB,1990 KOESet al.,
1994). Thus, it may be possible to express
' The examples given above show that the more pathways in cultured cells in the not too
levels of end products of biosynthetic path- distant future.
ways can be modified by genetic engineeriqg In some situations it may be desirable to
methods. Best results will be obtained when turn off or fortify one branch of a pathway so
the overproduced enzyme activity is clearly that more or less carbon enters that branch.
the rate-limiting step of the pathway. How- Overexpression of a tryptophan decarboxyl-
ever, usually a pathway has a series of rate- ase gene in Brassica napus led to the accumu-
limiting steps. Thus, if one step has been relation of high levels of tryptamine, while the
moved by genetic engineering, the next (e.g., levels of unwanted indole glucosinolates in
substrate supply) may soon become active seeds of the transgenic plants were only 3%
and limit the extent of overproduction. The of that found in seeds of untransformed
systems we have been working on (serotonin plants (CHAVADEJ et al., 1994). Reduction of
and anabasine biosynthesis) could be used to metabolite levels has also been achieved by
check whether by a second transformation expressing antisense RNA which is comple-
the second rate-limiting step might also be mentary to the mRNA encoding a pathway
overcome. The genes controlling lysine and enzyme. Antisense expression of a chalcone
tryptophan biosynthesis have been cloned. synthase gene resulted in distinctly decreased
On the other hand, it is clear that the identifi- flower pigmentation (VAN DER KROL et al.,
cation of rate-limiting steps of pathways is 1988).
usually not as easy as in our systems where Overexpression of an engineered enzyme
simple feeding experiments provided the in- may lead to an accumulation of intermediates
8 Conclusions and Outlook 627
of a pathway to high levels. Examples are the
overexpression of the amino acid decarboxyl- 8 Conclusions and
ases leading to high accumulation levels of
the corresponding amines (BERLIN et al.,
Outlook
1994). In addition new products may be
formed in cultured cells. For example, forma- Substantial progress has been made in the
tion of resveratrol in tobacco occurred only if biochemistry and molecular biology of sec-
stilbene synthase genes from Aruchis hypo- ondary pathways of higher plants since the
gueu or Vitis viniferu were expressed (HAINet first edition of “Biotechnology”. Many novel
al., 1990 1993). Expression of a coriander de- enzyme reactions have been detected, and a
saturase results in the formation of petrose- number of genes encoding components of
linic acid, a novel product of tobacco (CA- secondary metabolism have been cloned. Suc-
HOON et al., 1992). However, engineering of cessful alterations of secondary pathways by
novel enzyme reactions in a plant cell will genetic transformation have been achieved.
only be useful if a corresponding substrate is With the availability of stable and rapidly
naturally formed in the transgenic cells. Ex- growing hairy root cultures, the number of
pression of a strictosidine synthase gene, cod- products found at high levels in cultured cells
ing for a key enzyme of monoterpene alka- has increased greatly. The technological feasi-
loid biosynthesis, in tobacco, led to an active bility of large-scale fermentations of suspen-
enzyme without function because its sub- sion cultures has been clearly demonstrated.
strates tryptamine and secologanin are not The technical problems of growing hairy root
formed in tobacco (MCKNIGHTet al., 1991). cultures in huge bioreactors are surmounta-
The enzymes of secondary metabolism in ble. However, despite all scientific progress,
plant cell cultures have been regarded as a the enthusiasm of industry for plant cell cul-
“pot of gold” (ZENK, 1991). There is no tures as producers has not yet corresponding-
doubt that certain plant cell culture systems ly increased. Indeed, there is no reason to be-
have been most important for recent progress lieve that industries will change their reserved
in the enzymology of secondary pathways. It attitude in the foreseeable future. One can
remains to be seen whether these enzymes only hope that this conclusion is wrong, and
are also a “pot of gold” for biotechnology that the following statement made at the In-
purposes. However, it is now possible to ternational Tissue Culture Congress is cor-
create plant cell cultures which overproduce rect: “Scientists from universities and re-
certain enzymes of secondary metabolism by search centers have not yet realized what is
genetic engineering. When the enzyme itself really going on industry with respect to the
is the target compound for overproduction, tissue culture technology” (SMITH,1995). Al-
plant cell cultures are not considered to be though this optimistic view is shared by some
competitive (KUTCHANet al., 1994). These other American colleagues (TATICEKet al.,
authors recommended the heterologous ex- 1994) the future does not look so promising.
pression of plant genes in insect cultures as a As a consequence of the overall low interest
more convenient approach, because of their of the industrial sector in plant secondary me-
higher productivity and secretion of the en- tabolites and plant tissue culture technology,
zymes into the medium from which they are the governmental support for this field of bio-
easily purified. Using a cell culture of Spo- technology was reduced. Several leading
dopteru frugiperdu (Sf9) up to 4 mg L-’ stric- groups of biotechnologically orientated gov-
tosidine synthase of berberine bridge enzyme ernmental research centers, e.g., in England
was produced (KUTCHANet al., 1994). The and Canada, were recently forced to leave the
real potential of transgenic plant cell cultures field. Thus, this field will remain an exciting
as enzyme producers, however, has not yet area of basic research but not of biotechnolo-
been exploited. Using targeting signals and gy. It remains to be seen whether the future
suitable promoters it may be possible to pro- progress in understanding and manipulating
duce comparable levels of a desired protein in regulatory controls of secondary pathways
plant cells. will eventually enhance the biotechnological
importance of plant cell cultures as produc- BANTHORPE, D. V. (1994), Secondary metabolism

ers. The decisive question, which can only be in plant tissue culture: scope and limitations,
answered by each individual company, is Nut. Prod. Rep. 11, 303-328.
whether the commercial importance of any BATE,N. J., ORR,J., NI, W.,MEROMI,A., NAD-
plant-tissue-culture-derived product is high LER-HASSAR,T., DOERNER,P. W., DIXON,R.
A., LAMB,C. J, ELKIND,Y. (1994), Quantitative
enough to justify the undoubtedly high in- relationship between phenylalanine ammonia-
vestments needed for large-scale production. lyase levels and phenylpropanoid accumulation
in transgenic tobacco identifies rate-determining
step in natural product synthesis, Proc. Natl.
Acknowledgement Acad. Sci. USA 91,7608-7612.
BAUCH,H. J., LEISTNER,E. (1978), Aromatic me-
1 would like to thank Dr. HARA,Mitsui Co., tabolites in cell suspension cultures of Galium
Dr. SMITH,Urbana, and Dr. TAKAYAMA, mollugo, Planta Med. 33, 105-127.
Shizuoka, for providing me with new and un- BERLIN,J. (1986), Secondary products from plant
published information. Special thanks are due cell cultures, in: Biotechnology, Vol. 4, 1st Edn.
to Dr. HAVKIN-FRENKEL, David Michael & (REHM, H.-J., REED, G., Eds.), pp. 630-658.
Co,, for her helpful advice regarding flavor Weinheim: VCH.
production in tissue cultures. BERLIN,J. (1988), Formation of secondary meta-
bolites in cultured plant cells and its impact on
pharmacy, in: Biotechnology in Agriculture and
Forestry, Vol. 4 (BAJAJ,Y. P. S., Ed.), pp. 37-59.
Berlin: Springer-Verlag.
BERLIN,J. (1990), Screening and selection for var-
References iant cell lines with increased levels of secondary
metabolites, in: Secondary Products from Plant
ARENS, H., BORBE, H. O., ULBRICH, B., Tissue Culture (CHARLWOOD, B. V., RHODES,
STOCKIGT,J. (1982), Detection of pericine, a M. J. C., Eds.), pp. 119-137. Oxford: Claren-
new CNS-active indole alkaloid from Picralima don.
nitida cell suspension culture by opiate receptor BERLIN,J., SASE, F. (1985), Selection and screen-
binding studies, Planta Med 40,218-223. ing techniques for plant cell cultures, Adv. Bio-
ARENS,H., ULBRICH,B., FISCHER,H., PARNHAM, chem. Eng. 31, 99-132.
M. J., ROMER,A. (1986), Novel antiinflammato- BERLIN,J., KNOBLOCH, K. H., HOKE, G., WITTE,
ry flavonoids from Podophyllum verspille cell L. (1982), Biochemical characterization of two
culture, Planta Med. 52, 468-473. tobacco cell lines with high and low yields of cin-
ASAKA,I., 11, I., HIROTANI,M., ASADA,Y., Fu- namoyl putrescines, J. Nut. Prod. 45, 83-87.
RUYA, T. (1993), Production of ginsenoside sa- BERLIN,J., FORCHE,E., WRAY,V., HAMMER,J.,
ponins by culturing ginseng (Panax ginseng) em- HOSEL,W. (1983), Formation of benzophenan-
bryogenic tissues in bioreactors, Biotechnol. thridine alkaloids by suspension cultures of
Lett. 15, 1259-1264. Eschscholtzia californica Z. Naturforsch. 38c,
BALANDRIN, M. F., KLOCKE,J. A., WURTELE,E. 346-352.
S., BOLLINGER,W. H. (1985), Natural plant BERLIN,J., BEIER,H., FECKER,L., FORCHE,E.,
chemicals: Sources of industrial and medicinal Noe, W., SASSE, F., SCHIEL,O., WRAY, V.
materials, Science 228, 1154-1160. (1985), Conventional and new approaches to in-
BALLICA,R., RYU,D. D. Y. (1994), Tropane alka- crease alkaloid production of plant cell cultures,
loid production from Datura stramonium: An in- in: Primary and Secondary Metabolism of Plant
tegrated approach to bioprocess optimization of Cell Cultures (NEUMANN,K. H., BARZ, W.,
plant cultivation, in: Advances in Plant Biotech- REINHARD, E., Eds.), pp. 272-280. Berlin:
nology. Studies in Plant Sciences, Vol. 4 (RYu, Springer-Verlag.
D. D. Y., FURASAKI, S., Eds.), pp. 221-254. Am- BERLIN,J., SIEG, S., STRACK,D., BOKERN,M.,
sterdam: Elsevier. HARMS,H. (1986). Production of betalains by
BALSEVICH,J. (1988), Monoterpene indole alka- suspension cultures of Chenopodium rubrum,
loids from Apocynaceae other than Catharan- Plant Cell Tissue Organ Cult. 5, 163-174.
thus roseus, in: Cell Culture and Somatic Cell Ge- BERLIN, J., MOLLENSCHOTT,C., WRAY, V.
netics, Vol. 5 Phytochemicals in Plant Cell Cul- (1988a), Triggered efflux of protoberberine af-
tures (CONSTABEL,F., VASIL, I. K., Eds.), kaloids from cell suspension cultures of Thalic-
pp. 371-384. San Diego: Academic Press. trum rugosum, Biotechnol. Lett. 10, 193-198.
9 References 629
BERLIN, J., BEDORF, N., MOLLENSCHOTT,C., bolites with plant cell cultures: a literature re-
WRAY,V., SASSE,F., HOFLE, G. (1988b), On view, J. Biotechnol. 23, 111-141.
the podophyllotoxins of root cultures of Linum CAHOON,E. B., SHANKLIN, J., OHLROGGE, J. B.
flavum, Planta Med. 54, 204-206. (1992), Expression of a coriander saturase re-
BERLIN,J., MOLLENSCHOTT, C., GREIDZIAK, N., sults in petroselinic acid production in transgenic
ERDOGAN, S., KUZOVKINA, I. (1990), Affecting tobacco, Proc. Natl. Acad. Sci. USA 89, 11184-
secondary product formation in suspension and 11 188.
hairy root cultures - a comparison, in: Progress CAMERON, D. D., TONG,I. T. (1993), Cellular and
in Plant Cellular and Molecular Biology (NIJ- metabolic engineering - an overview, Appl. Bio-
KAMP,H. J. J., VAN DER PLAS,L. H. W., VAN chem. Biotechnol. 38, 105-140.
AARTRIJK,J., Eds.), pp. 763-768. Dordrecht: CAREW,D. P. (1975), Tissue culture studies of Ca-
Kluwer. tharanthus roseus, in: The Catharanthus Alka-
BERLIN,J., RUGENHAGEN, C., DIETZE,P., FECK- loids (TAYLOR,W. I., FARNSWORTH, N. R.,
ER, L. F., GODDIJN,0. J. M., HOGE,J. H. C. Eds.), pp. 193-208. New York: Marcel Dekker.
(1993), Increased production of serotonin by Celex-Laboratories (1994), Production of Taxus sp.
suspension and root cultures of Peganum har- hairy root culture - taxol, taxane and diterpene
mala transformed with a tryptophan decarboxy- production by Agrobacterium rhizogenes trans-
lase cDNA from Catharanthus roseus, Transgen. formed yew culture, WO 94/20606.
Res. 2, 336-344. CHARLWOOD, B. V., RHODES,M. J. C. (1990), Sec-
BERLIN,J., FECKER,S., HERMINGHAUS, C., RU- ondary products from plant tissue culture, Proc.
GENHAGEN (1994), Genetic modification of Phytochem. Soc., Vol. 30, Oxford Clarendon.
plant secondary metabolism: alteration of prod- CHARLWOOD, B. V., CHARLWOOD, K. A,, MOLI-
uct levels by overexpression of amino acid de- NA-TORRES, J. (1990), Accumulation of second-
carboxylases, in: Advances in Plant Biotechnolo- ary compounds by organized plant cultures, in:
gy. Studies in Plant Science, Vol. 4 (RYu, D. D. Secondary Products from Plant Tissue Culture
Y., FURASAKI, S., Eds.), pp. 57-81. Amsterdam: (CHARLWOOD, B. V., RHODES,M. J. C., Eds.),
Elsevier. pp. 167-200. Oxford: Clarendon.
BIANCO-COLOMAS, J., HUGUES,M. (1990), Estab- CHAVADEJ,S., BRISSON,N., MCNEIL,J. N., DE-
lishment and characterization of a betacyanin LUCA, V. (1994), Re-direction of tryptophan
producing cell line of Amaranthus tricolor: In- leads to production of low indole glucosinolate
ductive effects of light and cytokinins, J. Plant canola, Proc. Natl. Aad. Sci. USA 91, 2166-
Physiol. 136,734-739. 2170.
BOHM,H., RINK,E. (1988), Betalains, in: Cell Cul- CHEETHAM, P. S. J. (1993), The use of biotransfor-
ture and Somatic Cell Genetics of Plants, Vol. 5: mations for the production of flavours and fran-
Phytochemicals in Plant Cell Cultures (CONSTA- grances, TIBTECH 11,478-488.
BEL, F., VASIL,I. K., Eds.), pp. 449-463. San CLINE,S . D., COSCIA,C. J. (1988), Stimulation of
Diego: Academic Press. sanguinarine production by combined fungal eli-
BOKERN,M., WRAY,V., STRACK,D. (1991), Accu- citation and hormonal deprivation in cell sus-
mulation of phenolic conjugates and betacyan- pension cultures of Papaver bracteatum, Plant
ins, and changes in the activities of enzymes in- Physiol. 86, 161-165.
volved in feruloylglucose metabolism in cell sus- CONSTABEL, F., VASIL,I. K. (Eds.) (1988), Cell
pension cultures of Chenopodium rubrum, Plan- Culture and Somatic Cell Genetics of Plants, Vol.
ta 184, 261-270. 5: Phytochernicals in Plant Cell Cultures. San
BREULING, M., ALFERMANN, A. W., REINHARD, Diego: Academic Press.
E. (1985), Cultivation of cell cultures of Berberis COURTOIS,D., YVERNEL,D., FLORIN,B., PE-
wilsonae in 20-L airlift bioreactors, Plant Cell TIARD, V. (1988), Conversion of tryptamine to
Rep. 4,220-223. serotonin by cell supension cultures of Peganum
BRINGI,V., KADKADE,P. (1993), Enhanced pro- harmala, Phytochemistry 27, 3137-3141.
duction of taxol and taxanes by cell cultures of CRAGG,G. M., SCHEPARTZ,S. A., SUFFNESS, M.,
T a u s species, WO 93/17121. GREVER,M. R. (1993), The taxol supply crisis.
BUITELAAR,R. M., LANGENHOFF,A. A. M., New NCI policies for handling the large-scale
HEIDSTRA,R., TRAMPER, J. (1991), Growth and production of novel anticancer and anti-HIV
thiophene production by hairy root cultures of agents, J. Nut. Prod. 56, 1657-1668.
Tagetes patula in various two-liquid-phase bio- CURTIS,W. R. (1993), Cultivation of roots in bio-
reactors, Enzyme Microb. Technol. 13, 487-494. reactors, Curr. Opin. Biotechnol. 4, 205-210.
BUITELAAR, R. M., TRAMPER, J. (1992), Strategies DE CAROLIS,E., DE LUCA,V. (1993), Purification,
to improve the production of secondary meta- characterization, and kinetic analysis of a 2-0x0-
glutarate-dependent dioxygenase involved in and evaluation, Appl. Microbiol. Biotechnol. 37,

vindoline biosynthesis from Catharanthus ro- 457-462.
seus, J. Biol. Chem. 268,5504-5511. DIXON,R. A., LAMB,C. J. (1990), Regulation of
DE-EKNAMKUL, W., ELLIS,B. E. (1984), Rosmar- secondary metabolism at the biochemical and
inic acid production and growth characteristics genetic levels, in: Secondary Products from Plant
of Anchusa officinalis cell suspension cultures, Tissue Culture (CHARLWOOD, B. V., RHODES,
Planta Med. 50, 346-350. M. J. C., Eds.), pp.103-118. Oxford Claren-
DE-EKNAMKUL, W., ELLIS,B. E. (1985a), Effects don.
of auxins and cytokinins on growth and rosmar- Do, C. B., CORMIER, F. (1991), Effects of high am-
inic acid formation in cell suspension cultures of monium concentrations on growth and antho-
Anchusa offcinalis, Plant Cell Rep. 4, 50-53. cyanin formation in grape (Vitis vinifera L.) cell
DE-EKNAMKUL, W., ELLIS,B. E. (1985b), Effects suspension cultured in a production medium,
of macronutrients on growth and rosmarinic acid Plant Cell Tissue Organ Cult. 27, 169-174.
formation in cell suspension cultures of Anchusa DOONER,H. K., ROBBINS,T. P., JORGENSEN, R.
officinalis, Plant Cell Rep. 4, 46-49. A. (1991), Genetic and developmental control of
DELL,B., ELSEGOOD,C. L., GHISALBERTI, E. L. anthocyanin biosynthesis, Annu. Rev. Genet. 25,
(1989), Production of verbascoside in callus tis- 173-1 99.
sue of Eremophila spp., Phytochemistry 28, DRAPEAU,D., BLANCH,H. W., WILKE, C. R.
1871-1872. (1987), Economic assessment of plant cell cul-
ture for the production of ajmalicine, Biotech-
DELUCA,V. (1993), Enzymology of indole alka-
nol. Bioeng. 30,946-953.
loids, in: Methods of Plant Biochemistry, Vol. 9
(DEY,P. M., HARBORNE, J. B., Eds.), pp. 345- EDGINGTON, S. M. (1991), Taxol - out of the
368. London: Academic Press. woods, Bio/Technology 9,933-938.
EISENREICH, W., MENHARD,B., HYLANDS, P. J.,
DELUCA,V., KURZ,W. G. W. (1988), Monoter- ZENK,M. H., BACHER,A. (1996), Studies on
pene indole alkaloids (Catharanthus alkaloids), the biosynthesis of taxol: the taxane carbon skel-
in: Cell Culture and Somatic Cell Genetics of eton is not of mevalonoid origin, Proc. Natl.
Plants, Vol. 5: Phytochemicals in Plant Cell Cul- Acad. Sci. USA 93,6431-6436.
tures (CONSTABEL,F., VASIL, I. K., Eds.),
EILERT,U., KURZ, W. G. W., CONSTABEL, F.
pp. 385-402. San Diego: Academic Press.
(1985), Stimulation of sanguinarine accumula-
DELUCA,V., MARINEAU, C., BRISSON,N. (1989), tion in Papaver somniferum cells by fungal ho-
Molecular cloning and analysis of a cDNA en- mogenates, J. Plant Physiol. 119, 65-76.
coding plant tryptophan decarboxylase: compar- ELLIS,B. E. (1983), Production of hydroxyphenyl-
ison with animal dopa decarboxylase, Proc. Natl. ethanol glycosides in suspension cultures of Sy-
Acad. Sci. USA 86, 9969-9973. ringa vulgaris, Phytochemistry 22, 1941-1943.
DENO,H., SUGA,C., MORIMOTO, T., FUJITA, Y. ELLIS,B. E. (1985), Metabolism of caffeoyl deriva-
(1987), Production of shikonin derivatives by tives in plant cell cultures, in: Primary and Sec-
cell suspension cultures of Lithospermurn ery- ondary Metabolism of Plant Cell Cultures (NEU-
throrhizon. VI. Production of shikonin deriva- MANN, K. H., BARZ,W., REINHARD, E., Eds.),
tives by a two layer culture containing an or- pp. 164-173, Berlin: Springer-Verlag.
ganic solvent, Plant Cell Rep. 6, 197-199. ELLIS,B. E. (1988), Natural products from plant
DEUS, B., ZENK,M. H. (1982), Exploitation of tissue culture, Nut. Prod. Rep. 5 , 581-612.
plant cells for the production of natural com- FEIT-NETO, A. G., DICOSMO,F., REYNOLDS, W.
pounds, Biotechnol. Bioeng. 24, 1965-1974. F., SAKATA, K. (1992), Cell culture of Taxus as a
DEUS-NEUMANN, B., ZENK,M. H. (1984), Instabil- source of the antineoplastic drug taxol and re-
ity of alkaloid production in Catharanthus roseus lated taxanes, Bio/Technology 10, 1572-1575.
cell suspension cultures, Planta Med. 50, 427- FETT-NETO,A. G., MELANSON, S. J., SAKATA,K.,
431. DICOSMO,F. (1993), Improved growth and taxol
DICOSMO,F. (1990), Strategies to improve yields yield in developing calli of Taxus cuspidata by
of secondary metabolites to industrially interest- medium composition modification, Bio/Technol-
ing levels, in: Progress in Plant Cellular and Mo- Ogy 11,731-734.
lecular Biology (NIJKAMP,H. J. J., VAN DER FEIT-NETO, A. G., ZHANG,W. Y., DICOSMO,F.
PLAS, L. H. W., VAN AARTRIJK,J., Eds.), (1994a), Kinetics of taxol production, growth
pp. 717-724. Dordrecht: Kluwer. and nutrient uptake in cell suspensions of Taxus
DI IORIO,A. A., CHEETHAM, R. D., WEATHERS, cuspidata, Biotechnol. Bioeng. 44,205-210.
P. J. (1992), Growth of transformed roots in a FEIT-NETO,A. G., MELANSON, S. J., NICHOLSON,
nutrient mist bioreactor: Reactor performance S. A., PENNINGTON, J. J., DICOSMO,F. (1994b),
9 References 631
Improved taxol yield by aromatic and amino Plant Cell Culture 1982 (FUJIWARA, A., Ed.),
acid feeding to cell cultures of Taxus cuspidata, pp. 313-314. Tokyo: Maruzen Press.
Biotechnol. Bioeng. 44,967-971. FUKUI,H., TSUKADA, M., MIZUKAMI, H., TABA-
FLORES,H. E., FILNER,P. (1985), Metabolic rela- TA, M.(1983), Formation of stereoisomeric mix-
tionships of putrescine, GABA and alkaloids in tures of naphthoquinone derivatives in Echium
cell and root cultures of Solanaceae, in: Primary lycopsis callus cultures, Phytochemistry 22,453-
and Secondary Metabolism of Plant Cell Cultures 456.
(NEUMANN, K. H., BARZ,W., REINHARD, E., FUNK, C., BRODELIUS,P. (1990), Influence of
Eds.), pp. 174-185. Berlin: Springer-Verlag. growth regulators and elicitor on phenylpropan-
FLORES,T., WAGNER, L. J., FLORES,H. E. (1993), oid metabolism in suspension cultures of Vanilla
Embryo culture and taxane production in Taxus planifolia, Phytochemistry 29,845-848.
spp., In Vitro Cell. Dev. Biol. 29P,160-165. FUNK,C., BRODELIUS, P. (1992), Phenylpropanoid
FORKMANN, G. (1993), Control of pigmentation in metabolism in suspension cultures of Vanilla
natural and transgenic plants, Curr. Opin. Bio- planifolia. IV. Induction of vanillic acid forma-
technol. 4, 159-165. tion, Plant Physiol. 99, 256-262.
FUJITA,Y. (1988), Shikonin: Production by (Li- FURUYA, T. (1988), Saponins (Ginseng saponins),
thospermum erythrorhizon) cell cultures, in: Bio- in: Cell Culture and Somatic Cell Genetics of
technology in Agriculture and Forestry, Vol. 4 Plants, Vol. 5: Phytochemicals in Plant Cell Cul-
(BAJAJ, Y. P. S., Ed.), pp. 225-236. Berlin: tures (CONSTABEL,F., VASIL, I. K., Eds.),
Springer-Verlag. pp. 213-236. San Diego: Academic Press.
FUJITA,Y., HARA,Y. (1985), The effective produc- FURUYA, T., IKUTA,A., SYONO,K. (1972). Alka-
tion of shikonin by cultures with an increased loids from callus tissue of Papaver somniferum,
cell population, Agric. Biol. Chem. 49, 2071- Phytochemistry 11,3041-3044.
2075. FURZE,J. M., RHODES,M. J. C., PARR,A. J., Ro-
BINS,R. J., WHITEHEAD, I. M., THRELFALL, D.
FUJITA,Y., HARA, Y., OGINO, T., SUGA, C. R. (1991), Abiotic factors elicit sesquiterpenoid
(1981a), Production of shikonin derivatives by phytoalexin production but not alkaloid produc-
cell suspension cultures of Lithospermum ery- tion in transformed root cultures of Datura stra-
throrhizon. I. Effects of nitrogen sources on the monium, Plant Cell Rep. 10, 111-114.
production of shikonin derivatives, Plant Cell GERARDY, R., ZENK,M. H. (1993), Formation of
Rep. 1, 59-60. salutaridine from (R)-reticuline by a membrane
FUJITA,Y., HARA,Y., SUGA,C., MORIMOTO, T. bound cytochrome-P-450 enzyme from Papaver
(1981b), Production of shikonin derivatives by somniferurn, Phytochemistry 32,79-86.
cell suspension cultures of Lithospermum ery- GERTLOWSKI, C.,PETERSEN, M. (1993), Influence
throrhizon. 11. New medium for the production of the carbon source on growth and rosmarinic
of shikonin derivatives, Plant Cell Rep. 1, 61- acid production in suspension cultures of Coleus
63. blumei, Plant Cell Tissue Organ Cult. 34, 183-
FUJITA,Y., TABATA,M., NISHI,A., YAMADA, Y. 190.
(1982), New medium and production of second- GIROD,P. A., ZRYDJ. P. (1991), Secondary metab-
ary compounds with the two-stage culture meth- olism in cultured red beet (Beta vulgaris L.)
ods, in: Plant Tissue Culture 1982 (FUJIWARA, cells. Differential regulation of betaxanthins and
A., Ed.), pp. 399-400. Tokyo: Maruzen Press. betacyanin biosynthesis, Plant Cell Tissue Organ
FUJITA,Y., TAKAHASHI, S., YAMADA, Y. (1985), Cult. 25, 1-12.
Selection of cell lines with high productivity of GODDIJN, 0.J. M. (1992), Regulation of terpenoid
shikonin derivatives by protoplast culture of Li- indole alkaloid biosynthesis in Catharanthus ro-
thospermum erythrorhizon cells, Agric. Biol. seus: The tryptophan decarboxylase gene. PhD
Chem. 49,1755-1759. Thesis, Rijksuniversiteit Leiden.
FUJITA,Y., HARA,Y., MORIMOTO, T., MISAWA, GODDIJN, 0. J. M., DE KAM, R. J., ZANETTI,A.,
M. (1990), Semisynthetic production of vinblas- SCHILPEROORT, R. A., HOGE,J. H. C. (1992),
tine involving cell cultures of Catharanthus ro- Auxin rapidly down-regulates transcription of
seus, in: Progress in Plant Cellular and Molecular the tryptophan decarboxylase gene from Catha-
Biology (NIJKAMP, H. J. J., VAN DER PLAS, L. ranthus roseus Plant Mol. Biol. 18, 1113-1120.
H. W., VAN AARTRIJK, J., Eds.), pp.763-768. GOODBODY, A. E., ENDO,T., VUKOVIC, J., KUT-
Dordrecht: Kluwer. NEY, J. P., CHOI, L. s. L., MISAWA,M. (1988),
FUKUI,H., NAGAKAWA,K., TSUDA,S., TABATA, Enzymic coupling of catharanthine and vindol-
M. (1982), Production of isoquinoline alkaloids ine to form 3 ’,4 ’-anhydrovinblastineby horsera-
by cell suspension cultures of Coptis japonica, in: dish peroxidase, Planta Med. 54, 136-140.
632 I3 Secondary Products from Plant Cell CIultures
GROCER,D. (1980), Alkaloids derived from tryp- HEIDE,L., NISHIOKA, N., FUKUI,H., TABATA,M.
tophan and anthranilic acid, in: Secondary Plant (1989), Enzymatic regulation of shikonin biosyn-
Products. Encyclopedia of Plant Physiology, thesis in Lithospermum erythrorhizon cell cul-
Vol. 8 (BELL,E. A., CHARLWOOD, B. V., Eds.), tures, Phytochemistry 28, 1873-1877.
pp. 128-159. Berlin: Springer-Verlag. HENRY,M., ROUSSEL,J. L., ANDARY,C. (1987),
HAIN,R., BIESELER,B., KINDL,H., SCHRODER, Verbascoside production in callus and suspen-
G., STOCKER,R. (1990), Expression of a stil- sion cultures of Hygrophila erecta, Phytochemis-
bene synthase gene in Nicotiana tabacum results try 26, 1961-1963.
in synthesis of the phytoalexin resveratrol, Plant HERMINGHAUS, S., THOLL,D., RUGENHAGEN, C.,
Mol. Biol. 15, 325-335. FECKER, L. F., LEUSCHNER,C., BERLIN, J.
HAIN, R., REIF, H. J., KRAUSE,E., LANGEBAR- (1996), Improved metabolic action of a bacterial
TELS, R., KINDL,H., VORNAM,B., WIESE, W., lysine decarboxylase gene in tobacco hairy root
SCHMELZER, E., SCHREIER,P. H., STOCKER,R. cultures by its fusion to a rbcS transit peptide
H., STENZEL,K. (1993), Disease resistance re- coding sequence, Transgen. Res. 5, 193-201.
sults from foreign phytoalexin expression in a HEZARI,M., LEWIS,N. G., CROTEAU,R. (1995),
novel plant, Nature 361, 153-156. Purification and characterization of taxa-4
HAMILL,J. D., ROBINS,R. J., PARR,A. J., EVANS, (5),11(12)-diene synthase from Pacific yew (Tax-
D. M., FURZE,J. M., RHODES,M. J. C. (1990), us brevifolia) that catalyzes the first committed
Overexpressing a yeast ornithine decarboxylase step of taxol biosynthesis, Arch. Biochem. Bio-
gene in transgenic roots of Nicotiana rustica can phys. 322,437444.
lead to enhanced nicotine accumulation, Plant HINZ,H., ZENK,M. H. (1981), Production of pro-
Mol. Biol. 15, 27-38. toberberine alkaloids by cell suspension cultures
HAN, K. H., FLEMING,P., WALKER,K., LOPER, of Berberis species, Naturwissenschafien 68,620-
M., CHILTON,W. S., MOCEK,U., GORDON,M. 621.
P., FLOSS,H. G. (1994), Genetic transformation HIPPOLYTE,I., MARIN, B., BACCOU,J. C., JON-
of mature Taxus: an approach to genetically con- ARD, R. (1992), Growth and rosmarinic acid
trol the in vitro production of the anticancer production in cell suspension cultures of Salvia
drug taxol, Plant Sci. 95, 187-196. officinalis, Plant Cell Rep. 11, 109-112.
HARA,Y., YOSHIOKA,T., MORIMOTO,T., FUJITA,
Y., YAMADA,Y. (1988), Enhancement of ber- HIRASUNA, T. J., PESTCHANKER, L. J., SRINIVAS-
berine production in suspension cultures of Cop- AN, V., SHULER,M. L. (1996), Taxol production
tis japonica by gibberellic acid treatment, J. in suspension cultures of Taxus baccata, Plant
Plant Physiol. 133, 12-15. Cell Tiss. Org. Cult. 44, 95-102.
HARA, Y., YAMAGATA,H., MORIMOTO,T., HI- HIRATA,K., KOBAYASHI, M., MIYAMOTO, K., HA-
RATSUKA, J., YOSHIOKA,T., FUJITA,Y., YA-
SHI,T., OKAZAKI,M., MIURA,Y. (1989), Quan-
MADA, Y. (1989), Flow cytometric analysis of titative determination of vinblastine in tissue
cellular berberine contents in high- and low-pro- cultures of Catharanthus roseus by radioimmu-
ducing cell lines of Coptis japonica, Planta Med. noassay, Planta Med. 55, 262-264.
55,151-154. HOEKSTRA,S. S., HARKES,P. A. A., VERPOORTE,
HARKES,P. A. A., KRIJBOLDER, L., LIBBENGA,K. R., LIBBENGA, K. R. (1990), Effect of auxin on
R., WJNSMA,R., NSENGIYAREMGE, T., VER- cytodifferentiation and production of quinoline
POORTE, R. (1985), Influence of various media alkaloid in compact globular structures of Cin-
constituents on the growth of Cinchona ledgeria- chona ledgeriana, Plant Cell Rep. 8, 571-574.
nu tissue cultures and the production of alka- HOLTON,T. A., TANAKA,Y. (1994), Blue roses - a
loids and anthraquinones therein, Plant Cell Tis- pigment of our imagination? TIBTECH l2,40-
sue Organ Cult. 4, 199-214. 41.
HASHIMOTO,T., YAMADA,Y. (1994), Alkaloid IGBAVBOA,U., SIEWEKE,H. J., LEISTNER,E.,
biogenesis: Molecular aspects, Annu. Rev. Plant ROWER,I., HUSEMANN,W., BARZ, W. (1985),
Physiol. Plant Mol. Biol. 45, 257-285. Alternative formation of anthraquinones and li-
HASHIMOTO,T., YUN, D. J., YAMADA,Y. (1993), poquinones in heterotrophic and photoauto-
Production of tropane alkaloids in genetically trophic cell suspension cultures of Morinda luci-
engineered root cultures, Phytochernistry 32, da, Planta 166, 537-544.
713-718. IKUTA,A., SYONO,K., FURUYA, T. (1974), Alka-
HAVKIN-FRENKEL, D. (1994), Vanilla flavor pro- loids of callus tissues and redifferentiated plant-
duction: Tissue culture and the whole plant. lets in the Papaveraceae, Phytochemistry 13,
VIII. IA PTC Congress, Firenze (Abstract), S18- 2175-2179.
56. INAGAKI, N., NISHIMURA, H., OKADA,M., MITSU-
9 References 633
HASHI, H. (1991), Verbascoside production by cultures of Catharanthus roseus. I. Comparison
plant cell cultures, Plant Cell Rep. 9, 484-487. of enzyme activities and product accumulation,
INOUE,K., NAYESHIRO, H., INOUYE, H., ZENK,M. Plant Cell Tissue Organ Cult. 2, 333-340.
H. (1981), Anthraquinones in cell suspension KNOPP,E., STRAUSS,A., WEHRLI,W. (1988), Root
cultures of Morinda citrifolia, Phytochemistry 20, induction on several Solanaceae species by
1693-1700. Agrobacterium rhizogenes and determination of
INOUE,K., UEDA,S. NAYESHIRO, H., MORITOME, root tropane alkaloid content, Plant Cell Rep. 7 ,
N., INOUYE, H. (1984), Biosynthesis of naphtho- 590-593.
quinones and anthraquinones in Streptocarpus KNORR, D., CASTER, C., DORNENBURG,H.,
dunnii cell cultures, Phytochemistry 23, 313- GRAF, S., HAVKIN-FRENKEL, D., PODSTOLSKI,
318. A., WERRMANN, U. (1993), Biosynthesis and
JAZIRI,M., ZHIRI,A., Guo, Y. W., DUPONT,J. P., yield improvement of food ingredients from
SHIMOMURA, K., HAMADA,H., VANHAELEN, plant cell and tissue cultures, Food Technol. 47
M., HOMES,J. (in press), Taxus sp. cell, tissue (12), 57-63.
and organ cultures as alternative sources for tax- KNUTH,M. E., SAHAI,0. P. (1991), Flavor compo-
oid production: a literature survey, Plant Cell sition and method, US Patents No. 5,057,424 and
Tiss. Org. Cult. 5,068,184.
KADKADE,P. G. (1982), Growth and podopyllo- KOBAYASHI, Y., FUKUI,H., TABATA, M. (1988),
toxin production in callus tissues of Podophyl- Berberine production by batch and semi-contin-
lum peltatum, Plant Sci. Lett. 25, 107-115. uous cultures of immobilized Thalictrum cells in
KAMIMURA, S. (1991), Production of morphinan an improved bioreactor, Plant Cell Rep. 7 , 249-
alkaloids, in: Plant Cell Culture in Japan. Pro- 252.
gress in Production of Useful Plant Metabolites KOBAYASHI, Y., AKITA,M., SAKAMOTO, K., LIU,
by Japanese Enterprises Using Plant Cell Culture H., SHIGEOKA, T., KOYANA,T., KAWAMURA,
Technology (KOMAMINE, A., MISAWA,M., DI- M., FURUYA,T. (1993), Large-scale production
COSMO,F., Eds.), pp. 27-38. Tokyo: CMC Co. of anthocyanin by Aralia cordata cell suspension
KAMO,K. K., MAHLBERG, P. G. (1988), Morphi- cultures, Appl. Microbiol. Biotechnol. 40, 215-
nan alkaloids: Biosynthesis in plant (Papaver 218.
spp.) tissue cultures, in: Biotechnology in Agri- KOBLITZ,H. (1988), Anthraquinones, in: Cell Cul-
culture and Forestry, Vol. 4 (BAJAJ,Y. P. S., ture and Somatic Cell Genetics of Plants, Vol. 5:
Ed.), pp. 251-263. Berlin: Springer-Verlag. Production of Phytochemicals in Plant Cell Cul-
KETCHUM,R. E. B., GIBSON,D. M. (in press) Pa- tures (CONSTABEL,F., VASIL, I. K., Eds.),
clitaxel production in suspension cultures of pp. 113-142. San Diego: Academic Press.
Taxus, Plant Cell Tiss. Org. Cult. KOBLITZ,H., KOBLITZ,D., SCHMAUDER, H. P.,
KINGSTON,D. G. I. (1994), Taxol: the chemistry GROCER,D. (1983), Studies on tissue cultures of
and structure relationships of a novel anticancer genus Cinchona L.: Alkaloid production in cell
agent, TIBTECH 12, 222-227. suspension cultures, Plant Cell Rep. 2, 122-125.
KINNERSLEY, A. M., DOUGALL,D. K. (1980), In- KOES, R. E., QUATTROCCHIO, F., MOL, J. N. M.
crease in anthocyanin yield from wild carrot cell (1994), The flavonoid biosynthetic pathway in
cultures by a selection based on cell-aggregate plants: function and evolution, BioEssays 16,
size, Planta 149, 200-204. 123-132.
KISHIMA, Y., NOZALI,K., AKASHI,R., ADACHI,T. KOHL,W., WITTE,B., HOFLE,G. (1982), Alkaloids
(1991), Light-inducible pigmentation in Portula- from Catharanthus roseus tissue cultures 111, 2.
ca callus; selection of a betalain producing cell Naturforsch. 37b,1346-1351.
line, Plant Cell Rep. 10, 304-307. KOMAMIME, A., MISAWA,M., DICOSMO,F. (Eds.)
KISHORE,G. M., SOMERVILLE, C. R. (1993), Ge- (1991), Plant Cell Culture in Japan. Progress in
netic engineering of commercially useful biosyn- Production of Useful Plant Metabolites by Japa-
thetic pathways in transgenic plants, Curr. Opin. nese Enterprises Using Plant Cell Culture Tech-
Biotechnol. 4, 152-158. nology. Tokyo: CMC Co.
KNOBLOCH, K. H., BERLIN,J. (1980), Influence of KONDO,O., HONDA,H., TAYA,M., KOBAYASHI,
the medium composition on the formation of T. (1989), Comparison of growth properties of
secondary compounds in cell suspension cultures carrot hairy root in various bioreactors, Appl.
of Catharanthus roseus, Z. Naturforsch. 35c, Microbiol. Biotechnol. 32,291-294.
551-556. KREIS,W., REINHARD, E. (1990), Two-stage culti-
KNOBLOCH, K. H., BERLIN,J. (1983), Influence of vation of Digitalis lanata cells: semicontinuous
phosphate on the formation of the indole alka- production of desacetyllanatoside C in 20-litre
loids and phenolic compounds in cell suspension airlift bioreactors, J. Biotechnol. 16, 123-126.
KREIS,W., REINHARD, E. (1992), 12PHydroxyla- L., var. bikores monogerm (red beet). I n Vitro
tion of digitoxin by suspension-culturedDigitalis Dev. Biol. Plant. 28, 39-45.
lanata cells: Production of digoxin in 20-litre and LEISTNER,E. (1975), Isolierung, Identifizierung
300-litre air-lift bioreactors, J. Biotechnol. 26, und Biosynthese von Anthrachinonen in Zell-
257-273. suspensionskulturen von Morinda citrifolia,
KREIS,W., HOELZ,H., SUTOR,R., REINHARD, E. Planta Med. (Suppl.) 214-224.
(1993), Cellular organization of cardenolide bio- LOSCHER,R., HEIDE,L. (1994), Biosynthesis of p -
transformation in Digitalis grandiflora, Planta hydroxybenzoate from p-coumarate and p-cou-
191,246-251. marate-coenzyme A in cell-free extracts of Li-
KROMBHOLZ,R., MERSINGER,R., KREIS, W., thospermum erythrorhizon cell cultures, Plant
REINHARD, E. (1992), Production of forskolin Physiol. 106, 271-279.
by axenic Coleus forskohlii roots cultivated in LUCKNER, M., DIETTRICH,B. (1988), Cardenol-
shake flasks and 20-L glass jar bioreactors, Plan- ides, in: Cell Culture and Somatic Cell Genetics
ta Med. 58, 328-333. of Plants, Vol. 5: Phytochemicals in Plant Cell
IRZ, W. G. W., CHATSON, K. B., CONSTABEL, F., Cultures (CONSTABEL, F., VASIL,I. K., Eds.),
KUTNEY,J. P., CHOI,L. s. L., KOLODZIEJCZYK, pp. 193-212. San Diego: Academic Press.
P., SLEIGH,S. K., STUART,K. L., WORTH,B. R. LUTTERBACH, R., STOCKIGT, J. (1992), High-yield
(1980), Alkaloid production in Catharanthus ro- formation of arbutin from hydroquinone by cell
seus cell cultures: initial studies on cell lines and suspension cultures of Rauwolfia serpentina,
their alkaloid content, Phytochemistry 19, 2583- Helv. Chim. Acta 75, 2009-2011.
2587. MANO, Y., OHKAWA,H., YAMADA,Y. (1989),
KURZ, W. G. W., CHATSON, R. B., CONSTABEL, F. Production of tropane alkaloids by hairy root
(1985), Biosynthesis and accumulation of indole cultures of Duboisia leichhardtii transformed by
alkaloids in Catharanthus roseus cultivars, in: Agrobacterium rhizogenes, Plant Sci. 59, 191-
Primary and Secondary Metabolism in Plant Cell 201.
Cultures (NEUMANN, K. H., BARZ, W., REIN- MATSUBARA, K., FUJITA,Y. (1991), Production of
HARD,E., Eds.), pp. 143-153. Berlin: Springer- berberine, in: Plant Cell Culture in Japan. Pro-
Verlag. gress in Production of Useful Plant Metabolites
KURZ, W. G. W., PAIVA,N. L., TYLER,R. T. by Japanese Enterprises Using Plant Cell Culture
(1990), Biosynthesis of sanguinarine by elicita- Technology (KOMANINE, A., MISAWA,M., DI-
tion of surface immobilized cells of Papaver COSMO,F., Eds.), pp. 3 9 4 . Tokyo: CMC Co.
somniferum, in: Progress in Plant Cellular and MATSUBARA, K., KITANI,S., YOSHIOKA, Y., Mo-
Molecular Biology (NIJKAMP, H. J. J., VAN DER RIMOTO, T., FUJITA,Y., YAMADA, Y. (1989),
PLAS, L. H. W., VAN AARTRIJK,J., Eds.), High density culture of Coptis japonica cells in-
pp. 682-688. Dordrecht: Kluwer. creases berberine production, J. Chem. Technol.
Biotechnol. 46, 61-69.
KUTCHAN, T. M. (1993), Strictosidine: From alka-
loid to enzyme to gene, Phytochemistry 32,493- MCKNIGHT, T. D., BERGEY,D. R., BURNETTE,R.
506. J., NESSLER, C. L. (1991), Expression of an en-
zymatically-active and correctly targeted stricto-
KUTCHAN, T. M., ZENK,M. H. (1993), Enzymol- sidine synthase in transgenic tobacco plants,
ogy and molecular biology of benzophenanthri- Planta 185, 148-152.
dine biosynthesis, J. Plant Res. (Spec. Issue) 3, MEIJER,A. H., LOPESCARDOSO,M. I., VOSKUIL-
165-1 73. EN, J. T., DE WAAL,A., VERPOORTE, R., HOGE,
KUTCHAN,T. M., BOCK, A., DIETTRICH,H. J. H. C. (1993a), Isolation and characterization
(1994), Heterologous expression of the plant of a cDNA clone from Catharanthus roseus en-
proteins strictosidine synthase and berberine coding NADPH: cytochrome P-450 mono-oxy-
bridge enzyme in insect culture, Phytochemistry genase in plants, Plant J. 4,4740.
35,353-360. MEIJER,A. H., VERPOORTE,R., HOGE,J. H. C.
KUTNEY,J. P., AWERYN, B., CHOI,L. S. L., HON- (1993b), Regulation of enzymes and genes in-
DA, T., KOLDZIEJZCYK, P., LEWIS,N. G., SATO, volved in monoterpene indole alkaloid biosyn-
T., SLEIGH,S. K., STUART,K. L., WORTH,B. R., thesis in Catharanthus roseus, J. Plant Res.
KURZ,W. G. W., CHATSON, K. B., CONSTABEL, (Spec. Issue) 3, 145-164.
F. (1983), Studies in plant tissue culture: The MEURER-GRIMES, B., BERLIN,J., STRACK,D.
synthesis and biosynthesis of indole alkaloids, (1989), Hydroxycinnamoyl-CoA:putrescine hy-
Tetrahedron 39,3781-3795. droxycinnamoyltransferase in tobacco cell cul-
LEATHERS, R. R., DAVIN,C., ZRYD,J. P. (1992), tures with high and low levels of caffeoylputres-
Betalain producing cell cultures of Beta vulgaris cine, Plant Physiol. 89, 488-492.
9 References 635
MEYER, P., HEIDMANN,I., FORKMANN,G., MULDER-KRIEGER, T., VERPOORTE,R., BAER-
SAEDLER,H. (1987), A new petunia flower col- HEIM-SVENDSEN, A., SCHEFFER,J. J. C. (1988),
our generated by transformation of a mutant Production of essential oils and flavours in plant
with a maize gene, Nature 330, 677-678. cell and tissue cultures. A review, Plant Cell Tis-
MIRJALILI,N., LINDEN,J. C. (1996) Methyl jas- sue Organ Cult. 13, 85-154.
monate induced production of taxol in suspen- MURANAKA,T., OHKAWA, H., YAMADA, Y.
sion cultures of Taxus cuspidata: ethylene inter- (1993), Continuous production of scopolamine
action and induction models, Biotechnol. Prog. by a culture of Duboisia leichardtii hairy root
12,110-118. clone in a bioreactor system, Appl. Microbiol.
MISAWA,M. (1994), Plant tissue culture: an alter- Biotechnol. 40,219-223.
native for production of useful metabolites, NAGAKAWA, K., KONAGAI,A., FUKUI,H., TABA-
F A 0 Agric. Serv. Bull. 108. Rome: FAO. TA, M. (1984), Release and crystallization of
MISAWA,M., ENDO, T. (1988), Antitumor com- berberine in liquid medium of Thalictrum minus
pounds, in: Cell Culture and Somatic Cell Ge- cell suspension cultures, Plant Cell Rep. 3, 254-
netics, Vol. 5: Phytochemicals in Plant Cell Cul- 257.
tures (CONSTABEL,F., VASIL, I. K., Eds.), NAKATANI, H., MALIK,V. S. (1992), Putrescine N-
pp. 553-568. San Diego: Academic Press. methyltransferase, recombinant DNA molecules
MIZUKAMI, H., ELLIS, B. E. (1991), Rosmarinic encoding putrescine N-methyltransferase, and
acid formation and differential expression of ty- transgenic tobacco plants with altered nicotine
rosine arninotransferase isoforms in Anchusa of- content, Eur. Patent Application 9 113283.6,
ficinalis cell suspension cultures, Plant Cell Rep. Publication Nr. 0486214 A2.
10,321-324. NESSLER,C. R. (1994), Metabolic engineering of
MIZUKAMI,H., KONOSHIMA,M., TABATA,M. plant secondary products, Transgen. Res. 3, 109-
(1978), Variation in pigment production in Li- 115.
thospermurn erythrorhizon callus cultures, Phy- NICOLAOU, K. C., DAI,W. M., GUY,R. K. (1994),
tochemistry 17, 95-97. Chemie und Biologie von Taxol, Angew. Chem.
106,3849.
MIZUKAMI,H., TABIRA, Y., ELLIS,B. E. (1993), NOZUE,M., KAWAI,J., YOSHITAMA, K. (1987), Se-
Methyl jasmonate-induced rosmarinic acid bio-
lection of a high anthocyanin-producing cell line
synthesis in Lithospermum erythrorhizon cell
of sweet potato cell cultures and identification of
suspension cultures, Plant Cell Rep. 12, 706-
the pigment, J. Plant Physiol. US,81-88.
709.
ODAKE,K., ICHI,T., KUSUHARA,K. (1991), Pro-
MIZUSAKI,S., TANABE,Y., NOGUCHI,M., TAMA- duction of madder colorants, in: Plant Cell Cul-
KI, E. (1971), p-coumaroyl putrescine, caffeoyl ture in Japan. Progress in the Production of Use-
putrescine and feruloyl putrescine from callus ful Plant Metabolites by Japanese Enterprises Us-
tissue culture of Nicotiana tabacum, Phytochem- ing Plant Tissue Culture Technology (KOMAM-
istry 10, 1347-1350. INE, A., MISAWA,M., DICOSMO, F., Eds.),
MORENO,P. R. H., VAN DER HEIJDEN,R., VER- pp. 13&146. Tokyo: CMC Co.
POORTE, R. (1995), Cell and tissue cultures of OOSTHAM, A., M o t , J. N. M., VAN DER PLAS,L.
Catharanthus roseus: a literature survey. 11. Up- H. W. (1993), Establishment of hairy root cul-
dating from 1988-1993, Plant Cell Tiss. Org. tures of Linum flavum producing the lignan 5-
Cult. 42, 1-25. methoxypodophyllotoxin, Plant Cell Rep. 12,
MORIMOTO,M., HARA,Y., KATO, Y., HIRATSU- 474-477.
KA, J., YOSHIOKA, T., FUJITA.,Y., YAMADA,Y. OZAKI,Y., SUGA,C., YOSHIOKA,T., MORIMOTO,
(1988), Berberine production by cultured Coptis T., HARADA, M. (1990), Evaluation of equival-
japonica cells in a one stage culture using me- ence on pharmacological properties between
dium with a high copper concentration, Agric. natural crude drugs and their cultured cells
Biol. Chem. 52, 1835-1836. based on their components. Accelerative effect
MORRIS,P. (1986), Regulation of product synthesis of Lithospermi radix and inhibitory effect of
in cell cultures of Catharanthus roseus. 11. Com- Coptidis rhizoma on proliferation and granula-
parison of production media, Planta Med. 52, tion tissue, Yakugaku Zasshi 110, 268-272.
121-126. PALMER,J. E., WIDHOLM,J. M. (1975), Character-
MOSHY,R. J., NIEDER,M. H., SAHAI,0. P. (1989), ization of carrot and tobacco cell cultures resist-
Biotechnology in the flavor and food industry of ant to p-fluorophenylalanine, Plant Physiol. 56,
the USA, in: Biotechnology. Challenges for the 233-238.
Flavor and Food Industry (LINDSEY,R. L., WIL- PAREILLEUX, A., VINAS,R. (1984), A study on the
LIS,B. J., Eds.), pp. 145-163. London: Elsevier. alkaloid production of resting cell suspensions of
Catharanthus roseus in a continuous flow reac- REINHARD,E., ALFERMANN, A. W. (1980), Bio-

tor, Appl. Microbiol. Biotechnol. 19, 316-320. transformation of plant cell cultures, Adv. Bio-
PARK,J. M., YOONS. Y., GILES,K. L., SONGSTAD, chem. Eng. 16,49-83.
D., EPPSTEIN,D., NOVAKOVSKI, D., FRIESEN, REINHARD, E., KREIS,W., BARTHLEN, U., HELM-
L., ROEWER,I. (1992), Production of sanguinar- BOLD,U. (1989), Semicontinuous cultivation of
ine by suspension cultures of Papaver somnife- Digitalis lanata cells: Production of @methyldig-
rum in bioreactors, J. Ferment. Bioeng. 74, 292- oxin in a 300-L airlift bioreactor, Biotechnol.
296. Bioeng. 34,502-508.
PARR,A. J. (1989), The production of secondary RHODES,M. J. C., ROBINS,R. J., HAMILL,J. D.,
metabolites by plant cell cultures, J. Biotechnol. PARR, A. J., HILTON,M. G., WALTON,N. J.
10, 1-26. (1990), Properties of transformed roots, in: Sec-
PARR,A. J., PAYNE,J., EAGLES,J., CHAPMAN, B. ondary Products of Plant Tissue Culture
T., ROBINS,R. J., RHODES,M. J. C. (1990), Var- (CHARLWOOD, B. V., RHODES,M. J. C., Eds.),
iation in tropane alkaloid accumulation within pp. 201-225. Oxford: Clarendon.
the Solanaceae and strategies for its exploita- ROHMER,M., KNANI,M., SIMONIN,P., SUTTER,
tion, Phytochemistry 29, 2545-2550. B., SAHM,H. (1993), Isoprenoid biosynthesis in
PASQUALI,G., GODDIJN,0. J. M., DE WALL,A., bacteria: a novel pathway for the early steps
VERPOORTE, R., SCHILPEROORT, R. A., HOGE, leading to isopentenyl diphosphate, Biochem. J.
J. H. C., MEMELINK, J. (1992), Coordinated re- 295,517-524.
gulation of two indole alkaloid biosynthetic RUEFFER,M., ZENK,M. H. (1987), Distant precur-
genes from Catharanthus roseus by auxin and sors of benzylisoquinoline alkaloids and their
elicitors, Plant Mol. Biol. 18, 1121-1131. enzymatic formation, 2. Naturforsch. 4212, 319-
PAYNE,G. F., BRINGI,V., PRINCE,C., SHULER,M. 332.
L. (1991), Plant Cell and Tissue Culture in Li- RUEFFER,M., EL-SHAGI, H., NAGAKURA,N.,
quid Systems. Munich: Hanser. ZENK,M. H (1981), S-Norlaudanosoline synth-
PESTCHANKER, L. J., ROBERTS,S. C., SHULER,M. ase: the first enzyme in the benzylisoquinoline
L. (1996), Kinetics of taxol production and nu- biosynthetic pathway, FEBS Lett. 129, 5-9.
trient use in suspension cultures of Taxus cuspi- RUYTER,C. M., STOCKIGT, J. (1989), Neue Natur-
data in shake flasks and a Wilson-type reactor, stoffe aus pflanzlichen Zellkulturen, GIT Fachz.
Enzyme Microb. Technol. 19,256-260. Lab. 33,283-293.
PETERSEN, M., ALFERMANN, A. W. (1993), Plant
cell cultures, in: Biotechnology, Vol. 1,2nd Edn. SAITO,K., YAMAZAKI, M., MURAKOSHI, I. (1992),
(REHM, H.-J., REED, G., Eds.), pp. 578-614. Transgenic medicinal plants: Agrobacterium-me-
diated foreign gene transfer and production of
Weinheim: VCH.
secondary metabolites, J. Nut. Prod. 55, 149-
PETERSEN,M., HAUSLER,E., KARWATZKI,B., 162.
MEINHARD,J. (1993), Proposed biosynthetic
pathway for rosmarinic acid in cell suspension SAKAMOTO, K., IIDA,K., SAWAMURA, K., HAJI-
RO, K., ASADA,Y., YOSHIKAWA, T., FURUYA,
cultures of Coleus blumei, Planta 189, 10-14.
PIEHL,G. W., BERLIN,J., MOLLENSCHOTT,C., T. (1994), Anthocyanin production in cultured
LEHMANN, J. (1988), Growth and alkaloid pro- cells of Aralia cordata, Plant Cell Tissue Organ
duction of a cell suspension culture of Thalic- Cult. 36, 21-26.
trum rugosum in shake flasks and membrane- SAKUTA,M., TAKAGI,T., KOMAMINE, A. (1987),
stirrer reactors with bubble free aeration, Appl. Effects of nitrogen source on betacyanin accu-
Microbiol. Biotechnol. 29,456-461. mulation and growth in suspension cultures of
POULSEN,C., GODDIJN,0. J. M., HOGE,J. H. C., Phytolaca americana, Physiol. Plant. 71, 459-
VERPOORTE,R. (1994), Anthranilate synthase 463.
and chorismate mutase activities in transgenic SASSE,F., WITTE,L., BERLIN,J. (1987), Biotrans-
tobacco plants overexpressing tryptophan decar- formation of tryptamine to serotonin by cell sus-
boxylase from Catharanthus roseus, Transgen. pension cultures of Peganum harmala, Planta
Res. 3, 43-49. Med. 53, 354-359.
PROKSCH, A., WAGNER,H. (1987), Structural anal- SATO,F., YAMADA,Y. (1984), High berberine-pro-
ysis of a 4-0-methylglucuronoarabinoxylanwith ducing cultures of Coptis juponica cells, Phyto-
immuno-stimulating activity from Echinacea chemistry 23, 281-285.
purpurea, Phytochemistry 26, 1989-1993. SCHIEL,O., BERLIN,J. (1987), Large-scale fermen-
RAZZAQUE,A., ELLIS,B. E. (1977), Rosmarinic tation and alkaloid production of cell suspension
acid production in Coleus blumei, Planta 137, cultures of Catharanthus roseus, Plant Cell Tis-
287-291. sue Organ Cult. 8, 153-161.
9 References 637
SCHIEL,O., JARCHOW-REDECKER, K., PIEHL,G. eine and morphine production in suspended Pa-
W., LEHMANN, J., BERLIN, J. (1984a), Increased paver somniferum cultures after removal of exo-
formation of cinnamoyl putrescines by fed-batch genous hormones, Plant Cell Rep. 10,349-353.
fermentation of cell suspension cultures of Nico- SIEWEKE,H. J., LEISTNER,E. (1992), o-Succinyl-
tiana tabacum, Plant Cell Rep. 3, 18-20. benzoate :CoA ligase from anthraquinone pro-
SCHIEL,O., MARTIN,B., PIEHL,G. W., NOWAK,J., ducing cell cultures of Galium mollugo, Phyto-
HAMMER,J., SASSE, F., SCHAER,W., LEH- chemistry 31, 2329-2335.
MANN, J., BERLIN, J. (1984b), Some technologi- SIGNS,M. W., FLORES,H. E. (1990), The biosyn-
cal aspects on the production of cinnamoyl pu- thetic potential of plant roots, BioEssays 12, 7-
trescines by cell suspension cultures of Nicotiana 13.
tabacum, in: 3rd Eur. Congr. Biotechnol., Vol. 1, SIMANTIRAS, M., LEISTNER, E. (1989), Formation
pp. 167-172. Weinheim: VCH. of o-succinylbenzoic acid from isochorismic acid
SCHIEL,O., WITTE, L., BERLIN,J. (1987), Geran- in protein from anthraquinone-producing plant
iol-10-hydroxylase activity and its relation to cell suspension cultures, Phytochemistry 28,
monoterpene indole alkaloid accumulation in 1381-1382.
cell suspension cultures of Catharanthus roseus, SMART,N. J., MORRIS,P., FOWLER,M. W. (1982),
2. Naturforsch. 42c, 1075-1081. Alkaloid production by cells of Catharanthus ro-
SCHLATMANN, J. E., MORENO,P. R. H., VINKE,J. seus grown in airlift fermenter systems, in: Plant
L., TEN HOOPEN,H. J. G., VERPOORTE,R., Tissue Culture 1982 (FUJIWARAA., Ed.),
HEIJNEN,J. J. (1994), Effect of oxygen and nu- pp. 397-398. Tokyo: Maruzen Press.
trient limitation on ajmalicine production and SMITH,M. A. L. (1995), Large-scale production of
related enzyme activities in high density cultures secondary metabolites, in: Current Issues in
of Catharanthus roseus, Biotechnol. Bioeng. 44, Plant Molecular and Cellular Biology (TERZI,
461-468. M., CELLA,R., FALAVIGNA, A., Eds.), pp. 669-
SCHUBEL,H., RUYTER, C. M., STOCKIGT,J. 674. Dordrecht: Kluwer.
(1989), Improved production of raucaffricine by SMOLLNY, T., WICHERS, H., DE RIJK, T., VAN
cultivated Rauwolfia cells, Phytochemistry 28, ZWAM,A., SHASAVARI, A., ALFERMANN, A.
491494. W. (1993), Formation of lignans in suspension
SCHULTE, U., EL-SHAGI,H., ZENK,M. H. (1984), cultures of Linum album. 3rd Workshop on
Optimization of 19 Rubiaceae species in cell cul- “Primary and Secondary Metabolism of Plants
ture for the production of anthraquinones, Plant and Plant Cell Cultures”. Abstract P13, Leiden
Cell Rep. 3, 51-54. University.
SCHUMACHER, H. M., GUNDLACH, H., FIEDLER, SONGSTAD, D. D., GILES,K. L., PARK,J. M., Nov-
F., ZENK,M. H. (1987), Elicitation of benzo- AKOVSKI, D., EPPSTEIN, D., FRIESEN, L. (1990),
phenanthridine alkaloid biosynthesis in Esch- Use of nurse cultures to select for Papaver som-
scholtzia californica cell cultures, Plant Cell Rep. niferum cell lines capable of enhanced sanguin-
6,410-413. arine production, J. Plant Physiol. 136,236-239.
SCRAGG,A. H., CRESSWELL, R. C., ASHTON,S., SONGSTAD, D. D., KURZ,W. G. W., NESSLER, C.
YORK,A., BOND,P. A., FOWLER,M. W. (1989), L. (1991), Tyramine accumulation in Nicotiana
Growth and alkaloid production in bioreactors tabacum transformed with a chimeric tryptophan
by a selected Catharanthus roseus cell line, En- decarboxylase gene, Phytochemistry 30, 3245-
zyme Microb. Technol. 11, 329-333. 3246.
SCRAGG,A. H., ASHTON,S., YORK,A., STEPAN- SOUTHARD, G. L., GROZNIK, W. J., BOULWARE,
SARKISSIAN, G., GREY,D. (1990), Growth of R. T., THORNE,E. E., WALBORN, D. E., YAN-
Catharanthus roseus suspensions for maximum KELL, S. L. (1984), Sanguinarine, a new anti-
biomass and alkaloid accumulation, Enzyme Mi- plaque agent: retentention and plaque specifici-
crob. Technol. 12,292-298. ty, J. Am. Dent. Assoc. 108, 339-341.
SEITZ,H. U., HINDERER, W. (1988), Anthocyan- SRINIVASAN, V., CIDDI,V., BRINGI,V., SHULER,
ins, in: Cell Culture and Somatic Cell Genetics of M. (1996), Metabolic inhibitors, elicitors, and
Plants, Vol. 5: Phytochemicals in Plant Cell Cul- precursors as tools for probing yield limitation in
tures (CONSTABEL,F., VASIL, I. K., Eds.), taxane production by Taxus chinensis cell cul-
pp. 49-76. San Diego: Academic Press. tures, Biotechnol. Prog. 12,457466.
SHAUL,O., GALILI,G. (1991), Increased lysine STAHLHUT, R. (1993), ESCAgenetics Corporation,
synthesis in transgenic tobacco plants expressing IAPTC Newslett. 73, 12-15.
a bacterial dihydrodipicolinate synthase in their STEINRUCKEN, H. C., AMRHEIN, N. (1980), The
chloroplasts, Plant J. 2, 203-209. herbicide glyphosate is a potent inhibitor of 5-
SIAH,C. L., DORAN,P. M. (1991), Enhanced cod- enolpyruvylshikimic acid 3-phosphate synthase,
Biochem. Biophys. Res. Commun. 94, 1207- TAKAYAMA, S. (1994), Tripterygium wilfordii: In
1212. vitro culture and the production of anticancer
STEPHANOPOULOS, G., SINSKY, A. J. (1993), Meta- compounds tripdiolide and tripdolide, in: Bio-
bolic engineering - methodologies and future technology in Agriculture and Forestry, Vol. 28
prospects, TIBTECH 11, 392-396. (BAJAJ, Y. P. S., Ed.), pp. 457-468. Berlin:
STOCKIGT,J., SOLL,H. J. (1980), Indole alkaloids Springer-Verlag.
from cell suspension cultures of Catharanthus TAKAYAMA, S., TAKIZAWA, N., KUROYANAGI, M.
roseus and C. ovalis, Planta Med. 40,22-30. (1994), The method and system for large-scale
SUGA,T., HIRATA,T. (1990), Biotransformation of culture of plant roots using aeration-agitation
exogenous substrates by plant cell cultures, Phy- bioreactor. Strategies for scale-up. VIlI. IA PTC
tochemistry 29, 2393-2406. Congr., Firenze (Abstract) S20-22.
SUMARYONO, W., PROKSCH,P., HARTMANN, T., TANAKA,A. (1994), Immobilization of plant cells,
NIMTZ,M., WRAY,V. (1991), Induction of ros- in: Advances in Plant Biotechnology. Studies in
marinic acid accumulation in cell suspension cul- Plant Science, Vol. 4 (Ruu, D. D. Y., FURASAKI,
tures of Orthosiphon aristatus after treatment S., Eds.), pp. 209-220. Amsterdam: Elsevier.
with yeast extract, Phytochemistry 30, 3267- TATICEK,R. A., LEE, C. W. T., SHULER,M. L.
3271. (1994), Large-scale insect and plant cell culture,
SUZUKI,H., MATSUMOTO, T. (1988), Anthraquin- Curr. Opin. Biotechnol. 5, 165-174.
one: Production by plant cell culture, in: Bio- TOIVONEN, L. (1993), Utilization of hairy root cul-
technology in Agriculture and Forestry, Vol. 4 tures for production of secondary metabolites,
(BAJAJ, Y. P. S., Ed.), pp. 237-250. Berlin: Biotechnol. Prog. 9, 12-20.
Springer-Verlag. TYLER,V. E. (1988), Medicinal Plant Research:
SUZUKI,T., YOSHIOKA,T., HARA,Y., TABATA, 1953-1987, Planta Med. 54, 95-100.
M., FUJITA,Y. (1987), A new bioassay system TYLER,R. T., EILERT,U., RIJNDERS,C. 0. M.,
for screening high berberine-producing cell colo- ROEWER,I. A., KURZ,W. G. W. (1988), Semi-
nies of Thalictrum minus, Plant Cell Rep. 6, 194- continuous production of sanguinarine and dihy-
196. drosanguinarine by Papaver somniferum L. cell
SUZUKI,M., NAKAGAWA, K., FUKUI,H., TABATA, suspension cultures treated with fungal homo-
M. (1988), Alkaloid production in cell suspen- genate, Plant Cell Rep. 7,410-413.
sion cultures of Thalictrum flavum and T.dipte- ULBRICH,B., WIESNER,W., ARENS, H. (1985),
rocarpum, Plant Cell Rep. 7, 26-29. Large-scale production of rosmarinic acid from
SUZUKI,H., SUGA,C., MORIMOTO, T., HARADA, plant cell cultures of Coleus blumei, in: Primary
M. (1991), Quantitative analysis of plant hor- and Secondary Metabolism of Plant Cell Cultures
mones, auxins, in biotechnologically cultured (NEUMANN, K. H., BARZ, W., REINHARD, E.,
products of medicinal plants, Shoyakugaku Zas- Eds.), pp. 293-303. Berlin: Springer-Verlag.
shi 45, 137-141. USHIYAMA, K. (1991), Large-scale culture of gin-
SUZUKI,H., OZAKI,Y., SUGA,C., MORIMOTO, T., seng, in: Plant Cell Culture in Japan. Progress in
SATAKE,M., HARADA,M. (1993a), Dissolution the Production of Useful Plant Metabolites by Ja-
tests of Coptis rhizome and cultured cells of panese Enterprises Using Plant Tissue Culture
Coptis japonica, Shoyakugaku Zasshi 47, 311- Technology (KOMAMINE, A., MISAWA,M., DI-
315. COSMO, F., Eds.), pp. 92-98. Tokyo: CMC CO.
SUZUKI, H., OZAKI, Y., SATAKE,M. (1993b), VAN DEN BERG,A. J. J., RADEMA,M. H., LABA-
Changes of components - dissolution from pre- DIE, R. P. (1988), Effects of light on anthraqui-
scription containing Coptis rhizome or cultured none production in Rhamnus purshiana suspen-
cells of Coptis japonica into the detection, sion cultures, Phytochemistry 27, 415417.
Shoyakugaku Zasshi 47, 396-401. VAN DEN HEIJDEN,R., VERPOORTE,R., TEN
TABATA, M., MIZUKAMI, H., HIRAOKA,N., KON- HOOPEN,H. J. G. (1989), Cell and tissue cul-
OSHIMA, M. (1974), Pigment formation in callus tures of Catharanthus roseus, Plant Cell Tissue
cultures of Lithospermum erythrorhizon, Phyto- Organ Cult. 18,231-280.
chemistry 13, 927-932. VAN DER KROL, A. R., LENTING,P. E., VEEN-
TAKAHASHI, S., FUJITA,Y. (1991), Production of STRA, J., VAN DER MEER,I. M., KOES, R. E.,
shikonin, in: Plant Cell Culture in Japan. Pro- GERATS,A. G. M., MOL, J. N. M., STUITJE,A.
gress in the Production of Useful Plant Metabol- R. (1988), An antisense chalcone synthase gene
ites by Japanese Enterprises Using Plant Cell in transgenic plants inhibits flower pigmentation,
Culture Technology (KOMANINE, A., MISAWA, Nature 333, 866-869.
M., DICOSMO,F., Eds.), pp. 72-78. Tokyo: CMC VAN DER KROL, A. R., MUR, L. A., BELD, M.,
co. MOL,J. N. M., STUITJE,A. R. (1990), Flavonoid
9 References 639
genes in Petunia: addition of a limited number of WICKREMESINHE, E. R. M., ARTECA,R. N. (1993),
gene copies may lead to a suppression of gene Taxus callus cultures: Initiation, growth, optimi-
expression, Plant Cell 2, 218-221. zation, characterization and taxol production,
VAN GULIK,W. M., NUUTILA,A. M., VINKE,K. Plant Cell Tissue Organ Cult. 35, 181-193.
L., TEN HOOPEN,H. J. G., HEIJNEN, J. J. (1994), WICKREMESINHE, E. R. M., ARTECA, R. N.
Effects of carbon dioxide, air flow rate, and inoc- (1994a), Taxus cell suspension cultures: optimi-
ulation density on batch growth of Catharanthus zation of growth and production of taxol, J.
roseus cell suspensions in stirred fermenters, Plant Physiol. 144,183-188.
Biotechnol. Prog. 10, 335-339. WICKREMESINHE, E. R. M., ARTECA, R. N.
VAN UDEN, W., PRAS, N., VOSSEBELD,E. M., (1994b), Roots of hydroponically plants as a
MOL, J. N. M., MALINGRE,T. M. (1990), Pro- source of taxol and related taxanes, Plant Sci.
duction of 5-methoxypodophyllotoxin in cell 101,125-135.
suspension cultures of Linum jlavum, Plant Cell WJNSMA,R., VERPOORTE,R. (1988), Quinoline al-
Tissue Organ Cult. 20,81-87. kaloids of Cinchona, in: In Cell Culture and So-
VAN UDEN, W., PRAS, N., HOMAN, B., MAL- matic Cell Genetics of Plants, Vol. 5: Phytochem-
INGRE,T. M. (1991), Improvement of the pro- icals in Plant Cell Cultures (CONSTABEL, F., VA-
duction of 5-methoxypodophyllotoxin using a SIL, I. K., Eds.), pp. 337-356. San Diego: Aca-
new selected root culture of Linum flavum, demic Press.
Plant Cell Tissue Organ Cult. 27, 115-121. WILLIAMS,R. D., ELLIS,B. E. (1993), Alkaloids
WAGNER,F., VOGELMANN, H. (1977), Cultivation from Agrobacterium rhizogenes-transformedPa-
of plant cell cultures in bioreactors and forma- paver somniferum cultures, Phytochemistry 32,
tion of secondary metabolites, in: PIant Tissue 719-723.
Culture and its Biotechnological Application WILSON,P. D. G., HILTON,M. G., MEEHAN,P. T.
(BARZ,W., REINHARD, E., ZENK,M. H., Eds.), H., WASPE,C. R., RHODES,M. J. C. (1990), The
pp. 245-252. Heidelberg: Springer-Verlag. cultivation of transformed roots from laboratory
WAGNER, H., STUPPNER, H., PUHLMANN,J., to pilot plant, in: Progress in Plant CelIular and
BRUMMER,B., DEPPE,K., ZENK,M. H. (1989), Molecular Biology (NIJKAMP,H. J. J., VAN DER
Gewinnung von immunologisch aktiven Polysac- PLAS, L. H. W., VAN AARTRIJK,J., Eds.),
chariden aus Echinacea-Drogen und -Gewebe- pp. 700-705. Dordrecht: Kluwer.
kulturen, Z. Phytother. 10, 35-38. WOERDENBAG, H. J., VAN UDEN, W., FRIJLINK,
WALKER,M. A.,ELLIS,B. E., DUMBROFF,E. B., H. W., LERK,C. F., PRAS,N., MALINGRE, T. M.
DOWNER,R. G., MARTIN,R. J. (1986), Changes (199O), Increased podophyllotoxin production in
in amines and biosynthetic enzyme activities in Podophyllum hexandrum cell suspension cul-
p-fluorophenylalanine resistant and wild type to- tures after feeding coniferyl alcohol as a p-cy-
bacco cell cultures, Plant Physiol. 80, 825-828. clodextrin complex, Plant Cell Rep. 9,97-100.
WALTON,N. J., BELSHAW, N. J. (1988), The effect YAMADA,Y., HASHIMOTO,T. (1982), Production
of cadaverine on the formation of anabasine of tropane alkaloids in cultured cells of Hyoscya-
from lysine in hairy root cultures of Nicotiana mus niger, Plant Cell Rep. 1,101-103.
hesperis, Plant Cell Rep. 7,115-118. YAMADA,Y.,SATO, F. (1982), Production of ber-
WESTCOTT,R. J., CHEETHAM,P. J. S., BARRA- berine in cultured cells of Coptis japonica, Phy-
CLOUGH,R. J. (1994), Use of organized viable tochemistry 20, 545-547.
plant vanilla aerial roots for the production of YAMAMOTO, Y. (1991), Anthocyanin production in
natural vanilla, Phytochemistry 35, 135-138. plant cell cultures, in: Plant Cell Culture in Ja-
WESTPHAL,K. (1990), Large-scale production of pan. Progress in the Production of Useful Plant
new biologically active compounds in plant cell Metabolites by Japanese Enterprises Using Plant
cultures, in: Progress in Plant Cellular and Mo- Cell Culture Technology (KOMAMINE, A., MIS-
lecular Biology (NIJKAMP,H.J. J., VAN DER AWA,M., DICOSMO, F., Eds.), pp. 114-126. To-
PLAS, L. H. W., VAN AARTRIJK,J., Eds.), kyo: CMC Co.
pp. 601408. Dordrecht: Kluwer. YAMAMOTO,Y., MIZUGUCHI,R., YAMADA,Y.
WHITAKER,R. J., HASHIMOTO,T., EVANS,D. A. (1982), Selection of a high and stable-pigment
(1984), Production of secondary metabolite ros- producing strain in cultured Euphorbia millii
marinic acid by plant cell suspension cultures, cells, Theor. Appl. Genet. 61, 113-116.
Ann. N. Y. Acad. Sci. 435, 364-366. YAMAMOTO,Y., KINOSHITA,Y. WATANABE,S.,
WICHERS, H. J., VERSLUIS-DEHAAN, G. G., YAMADA,Y. (1989), Anthocyanin production in
MARSMAN, J. W., HARKES,M. P. (1991), Podo- suspension cultures of high producing cells of
phyllotoxins in plants and cell cultures of Linum Euphorbia millii, Agric. Biol. Chem. 53, 417-
jlavum, Phytochemistry 30,3601-3604. 423.
YOKOYAMA, M., YANAGI,M. (1991), High-level ZENK,M. H., EL-SHAGI, H., SCHULTE, U. (1975),
production of arbutin by biotransformation, in: Anthraquinone production by cell suspension
Plant Cell Culture in Japan. Progress in the Pro- cultures of Morinda citrifolia, Planta Med.
duction of Useful Plant Metabolites by Japanese (Suppl.), 79-101.
Enterprises Using Plant Cell Culture Technology ZENK,M. H., EL-SHAGI,H., ULBRICH, B. (1977a),
(KOMAMINE, A., MISAWA,M.,DICOSMO,F., Production of rosmarinic acid by cell suspension
Eds.), ..
pp. 78-91. Tokyo: CMC Co. cultures of Coleus blumei, Naturwissenschaften
YUKIMUNE,Y., TABATA,H., HIGASHI,Y., HARA, 64,585-586.
Y. (1996), Methyl jasmonate-induced overpro- ZENK,M. H., EL-SHAGI,H., ARENS,H., STOCK-
IGT, J., WEILER,E. W., DEUS,B. (1977b), For-
duction of paclitaxel and baccatin 111 in Taxus
cell suspension cultures, Nut. Biotechnol. 14, mation of indole alkaloids serpentine and ajmal-
1129-1132. icine in cell suspension cultures of Catharanthus
roseus, in: Plant Tissue Culture and its Biotech-
YUN,D. J., HASHIMOTO, T., YAMADA, Y. (1992), nological Application (BARZ, W., REINHARD,
Metabolic engineering of medicinal plants: E., ZENK,M. H., Eds.), pp. 2743. Heidelberg:
Transgenic Atropa belladonna with an improved Springer-Verlag.
alkaloid composition, Proc. Natl. Acad. Sci. U S A ZHONG,J. J., YOSHIDA,T. (1995), High density
89,11799-1 1803. cultivation of Perilla frutescens cell suspensions
ZENK,M. H. (1991), Chasing enzymes of secondary for anthocyanin production: Effects of Sucrose
metabolism - Plant cell culture a pot of gold, concentration and inoculum size, Biotechnol.
Phytochemistry 30,3861-3863. Bioeng. 38,6534158.
Biotechnology
14 Biotechnical Drugs as Antitumor

Agents
UDO GRAFE,KLAUSJURGENDORNBERGER,
HANS-PETERSALUZ
Jena, Germany
1 Introduction 643
2 Prescreening for Antineoplastic Agents 645
3 Classical Anticancer Drugs 647
3.1 Nucleoside Antibiotics and Analogs of Nucleobases 647
3.2 Drugs Binding Non-Covalently to the DNA 649
3.3 Intercalating Antibiotics 650
3.3.1 Ellipticin 650
3.3.2 Actinomycins 650
3.3.3 Quinoxaline Antibiotics 652
3.3.4 Anthracyclines and Related p-Quinones 652
3.4 Inhibitors of Enzymes of DNA Replication and Transcription: DNA Topo-
isomerases 653
3.5 Agents Forming Covalent Bonds with DNA 656
3.5.1 Mitomycin C 656
3.5.2 Anthramycins 656
3.5.3 Cyclopropane, Aziridines, and Epoxide Compounds 656
3.5.4 Streptonigrin 658
3.5.5 Anthracyclines 658
3.5.6 Bleomycin 661
3.5.7 Macromolecular Antitumor Antibiotics and the Enedyine Family of Cytotoxic
Drugs 664
3.5.8 Other Agents Forming Active Oxygen Radicals in the Cells 667
3.6 Inhibitors of Mitosis and the Microtubular System 667
3.7 Reduction of the Side Effects of Highly Toxic Anticancer Agents 669
3.8 Cytotoxic Compounds with Poorly Characterized Mode of Activity 669
642 14 Biotechnical Drugs as Antitumor Agents
4 Non-Classical Approaches to Antitumor Drugs 670

4.1 Potentiators of Cytotoxic Antitumor Agents 670
4.2 Inhibitors of Glutathione S Transferase as Enhancers of the Antitumor Activity of
Drugs 673
4.3 Antimetastasis Drugs and Inhibitors of Angiogenesis 673
4.4 Antitumor Effects of Immunomodulators 675
4.5 Inhibitors of the Cellular Mitogenic Signal Transduction Pathway 676
4.5.1 Inhibitors of Protein Kinases and Protein Phosphatases 677
4.5.1.1 Inhibitors of Protein Kinase C 679
4.5.1.2 Inhibitors of Other Protein Kinases 679
4.5.1.3 Inhibitors of Tyrosine Protein Kinases 679
4.5.2 Inhibitors Affecting the Metabolism of Phosphoinositols 679
4.5.3 Substances Changing the Morphology of Oncogene-Transformed Cells or Showing
Selective Toxicity 682
4.5.4 Inhibitors of rus-Farnesyltransferase 685
4.5.5 Inhibitors of Sexual Hormone Production and Hormone-Receptor
Interactions 685
4.5.6 Miscellaneous Drugs with Potential Antitumor Activity 685
5 Closing Remarks 687
6 References 688
I Introduction 643
1 Introduction covered (e.g., vincristine, camptothecin, may-

tansin, podophyllotoxins, taxol, and taxoids);
however, only a few of these have found ap-
Malignant tumors are a major cause of plication (Tab. 1).
mortality, ranking only second to cardiovas- The high toxicity of most of these' drugs
cular diseases in industrialized countries. The and the nonspecific interaction with normal
frequency of cancer increases with age, and and tumor cells were the major hurdles for
the general increase in life span in this cen- their therapeutic application. Today, a small
tury renders the prevention and treatment of spectrum of natural products (antibiotics,
cancer a serious problem. plant alkaloids, and terpenes), synthetic alkyl-
The aim of this chapter is to review the ating agents (nitrosoureas, busulfan), heavy
structures and modes of action of drugs from metal complexes (cis-diammine-dichloro-pla-
microorganisms, such as doxorubicin, bleo- tinum, etc.), and antifolates (5-fluorouracil,
mycin, mitomycin C, and others which have methothexate) are used in the treatment of
been used in the past as anticancer chemo- cancer (GALEet al., 1981; WILMAN,1990). In
therapeutics. The treatment of cancer by mi- general, the use of cytotoxic drugs in chemo-
crobial metabolites was proposed already in therapy is accompanied by severe side effects
1955 when the cytotoxic effect of actinomycin such as gastrointestinal disorders, cardiovas-
D on tumor cells was first described. Subse- cular toxicity, nausea, and vomiting. Many tu-
quently, a plethora of other cytotoxic metab- mors are not susceptible (while slowly divid-
olites of microbial and plant origin was dis- ing) or even become resistant to cytostatic
Tab. 1. Biotechnical and Plant-Derived Drugs as Therapeutic Antitumor Agents
Name Structural Type Mode of Activity

_ _ _ _ ~~ ~ ~
Actinomycin D Chromopeptide Intercalation into the DNA, inhibition of

(Dactinomycin) RNA polymerase
Daunorubicin Anthracycline Intercalation into the DNA, formation of
Doxorubicin free radicals, induction of DNA strand
Carminomycin breaks, inhibition of topoisomerases
Nogalam ycin
Bleomycin Glycopeptide Formation of free radicals, DNA strand
(Peplomycin, Phleomycin) breaking agent
Anthramycins Benzodiazepine Formation of free radicals and induction of
DNA strand breaks
Mitomycin C, Qu in one Formation of free radicals and induction of
porfiromycin DNA strand breaks
Endynamicin Enediyne Formation of free radicals and induction of
DNA strand breaks
Taxol, taxoids Diterpene Inhibition of tubulin depolymerization
Vincistine Alkaloids Inhibition of mitosis by interference with tu-
Vinblastin bulin polymerization
Vindesin
Camptothecin Terpenoid plant toxins Topoisomerase I1 inhibitor
Etoposid, teniposid Podophyllum toxins Topoisomerase I inhibitor
Maytanosides, Ansamacrolide Inhibition of mitosis
ansamitocin,
geldanamycin
drugs under therapeutic conditions (ASZA- occurs. Cell division is accomplished during
LOS, 1988). In the 1970s it became evident the subsequent G2 and M phases. Subse-
that cytotoxic drugs were not the only answer quently, the two cell copies either enter a new
to the cancer problem. New efforts were Go phase or differentiate to a mature non-div-
needed to uncover the molecular causes of iding cell. Cancer cells are unable to differen-
malignant cellular growth and to develop tiate and hence they undergo a new cell cycle
more specifically acting anticancer agents. and divide repeatedly. Tumor cells can regul-
The similarity of the metabolic pattern of nor- ate their growth autonomously by secreting
mal and cancer cells, the high diversity and hormone-like proteins such as bombesin.
different sensitivity of human tumors to anti- Bombesin is a tetradecapeptide produced by
cancer agents made the search for an anti- most of the small-cell lung carcinomas. It is a
cancer “wonder” drug a rather hopeless en- potent mitogen which stimulates growth of
terprise. This is illustrated by the moderate small-cell lung cancer (SCLC) cells in serum
effects of chemical derivatives on the im- cultures (MULLER, 1986). Several of the
provement of the activity of common anti- “classical” antitumor agents were shown to
cancer drugs and the reduction of their toxic affect individual phases of the cell cycle in a
side effects. More recent efforts towards a specific manner, e.g., daunorubicin and cis-
more specific delivery of cytoxic drugs, such platinum act on GI phase cells; arabinosyl cy-
as the development of immunotoxins or drug tosine, thioguanine, and doxorubicin act on S-
encapsulations into liposornes, did not im- phase cells; bleomycin, and cis-diammine-
prove this general picture. A major aim of dichloro-platinum act on GZphase cells; vin-
modern anticancer chemotherapy is to reduce cristine, doxorubicin, colchicine, etoposid,
the side effects, e.g., nausea and vomiting. and bleomycin act on M phase cells. This sug-
Major approaches to both the elucidation gested that they could change specific signal-
of the causes of malignant diseases and the ing pathways regulating cell division. The re-
search for new anticancer drugs have been cent discovery of the cycline family of euka-
promoted by recent advances of molecular ryotic regulatory proteins unraveled specific
biology and biomedical pharmacy. Although signaling factors which trigger the coordi-
it is not within the scope of this survey to dis- nated events of cell division in yeast
cuss these in detail a short reference is given (SCHWOBand NASMYTH,1993).
below. Identical copies of the parental cell are
Dividing mammalian cells undergo a cell formed in mammals in response to outer sig-
cycle (mitosis). Somatic cells in Go phase en- nals such as growth hormones (Fig. 1) (KAHN
ter the GI phase and subsequently the S and GRAF,1986; BURCKet al., 1988). These
phase during which chromosome replication exogenous signals are recognized by mem-
Fig. 1. Signal trans-

duction pathway in
cellular growth reg-
ulation (KAHNand
GRAF,1986).
2 Prescreening for Antineoplastic Agents 645
brane receptors which transmit the message tinent, deregulated cellular growth function.
via GTP-dependent membrane proteins (G A series of receptor proteins and growth-reg-
proteins) to proteins located at the inner side ulating enzymes has been well characterized
of the cytoplasmic membrane. In response, and in vitro screening assays have been devel-
the latter become activated and modify (e.g., oped (Tab. 2). Since the 1980s, many new
phosphorylate, farnesylate, etc.) proteins structures which interfere with cellular signal
which by switching on subsequent events of transduction have been discovered. These are
the cellular growth-regulating pathway, act as regarded as “soft” anticancer drugs. Their
a kind of an operational amplifier (YARDIN particular mode of action renders them inva-
and ULLRICH,1988). A normal cell only div- luable “biochemical tools” in detailed investi-
ides in response to a given signal, but a cancer gations of malignant growth. The first part of
cell has lost all growth control and repeatedly this article surveys microbial products which
enters a new cell cycle. Hence, a deregulated are used as “classical”, cytotoxic antitumor
expression of potential cancer genes (proto- agents. They interfere with DNA replication,
oncogenes) encoding growth factors, mem- transcription, and mitosis. In the second part,
brane receptors, transcriptional factors, and “non-classical” approaches with new and low-
regulatory enzymes has been suggested. In toxicity anticancer agents are discussed
turn, autonomous secretion of growth factors, (UMEZAWA,1989; WILMAN,1990). Some of
deregulated responses to growth factors, and these new drugs amplify the cytotoxicity of
continuous cell division will occur. The exis- classical anticancer agents, even in resistant
tence of tumor suppressor genes has been cell lines, and inhibit metastasis and angio-
proven, the products of which prevent the genesis of tumors. In addition, inhibitors of
outbreak of malignant growth. growth regulatory enzymes such as protein ki-
Moreover, it became evident that carcino- nases, inositol kinases, and phospholipases
genesis can be caused by chemical DNA will be reviewed.
strand-breaking and/or alkylating agents, and
transforming oncogenic viruses (BRADSHAW
and PRENTIS,1987; PIMENTEL,1987). Inser-
tion of the latter into the DNA causes partic-
ular genes (proto-oncogenes) involved in
growth control to become cancer genes (“on- 2 Prescreening for
cogenes”). Usually, oncogene-transformed
cells can be distinguished from the ancestral Antineoplastic Agents
type by
In order to screen microbial cultures for
- the overproduction of growth factors (auto- antitumor drugs, predictive prescreens are re-
crine secretion of growth hormones), quired because animal antitumor assays have
- altered structures of membrane receptors poor sensitivity. For DNA-damaging and cy-
rendering them permanently activated even totoxic agents a series of specific high capaci-
in the absence of exogenous growth fac- ty assays has been developed (FOX, 1991).
tors, Promising results have been obtained with
- altered structures of growth factors giving microorganisms which express special genes
rise to permanent (irreversible) receptor (such as recA) in response to DNA damage.
activation, and Moreover, enhanced activity against repair-
- altered production rates, structures, and deficient microbial mutants is another indica-
functions of intracellular signaling proteins tion for cytotoxic drugs interfering with
such as protein kinases, phospholipases, DNA. Studies on DNA-replication in a cell-
inositol kinase. free system have led to the detection of coval-
ently modifying agents (GREENSTEINand
Different types of oncogene-expressing MAIESE,1984).
(transformed) cells are now available as tools Neoplastic cell lines have been widely used
in the search for specific inhibitors of the per- in many laboratories to screen microbial cul-
Tab. 2. Inhibitors of Enzymes of the Cellular Growth-Regulating Signal Transmission Pathway and Drugs
Exerting a Non-Classical Action on Tumors
EnzymelScreening Feature Inhibitor
Protein kinase C Calphostin, staurosporin, RK-286c, K-252a, UCN-01, BE-

13793c, Sch45752, balanol, MS-282a
Protein tyrosin kinase BE-23372M, emodin, lavendustin, erbstatin, genistein, epider-
statin
Protein phosphatase Tautomycin, okadaic acid, dephostatin, microcystin, calyculin
Calmodulin and CAMP-
Phosphodiesterase KS-505a
Phospholipase D Sch49210, Sch53514, Sch53517
Inositol-specific phospholipase C Hispidospermidin, psi-tectorigen, Q12713
Inositol phosphate receptor Adenophostins
agonist
Inositol kinase Herbimycin, isoflavones, echiguanin, inostamycin, piericidin
B1-N-oxide, echigramins, piericidin, benzaldehyde
Diacylglycerol kinase Cochlioquinone, temphone
Inositol monophosphatase L-671-776
DNA topoisomerase I Dotriacolide, camptothecins (from plants), rebeccamycin
DNA topoisomerase I1 UCT4B, saintopin, BE-10988, streptonigrin, anthracyclines,
BE-22179, cyclothialidin
Poly(ADP-ribose)polymerase 2-Methylquinazolines, benzamides (synth.)
Glyoxylase I Glyo I and I1
Induction of differentiation of Hygrolidine, cycloheximide, trichostatins, reveromycin, differ-
cancer cell lines anisol, differenol A, microcystilide A, herbimycin, indolecar-
bazoles (staurosporin), lavanducyanin, oxanosin, kasuzamycin
Metastasis inhibitors and inhibitors TAN-1120, TAN-1323, U-77863, U-77864, matlystatins, siasta-
of angiogenesis tin B, WF-16775, erbstatin, herbimycin, lactoquinomycin
Glutathion S-transferase TA-3037A, benastatins, rishirilid, cysfluoretin, bequinostatin
Enhancer of the cytotoxicity of Piperafizins, bisucaberin, verapamil (synth.), BE-l3793c, cy-
antitumor agents closporin A, resorthiomycin, rubiginon, polyethers (laidlomy-
cin), hatomarubigin
Reduction of side effects of
highly toxic anticancer agents Conagenin
Ras-farnesyltransferase Pepticinnamine, streptonigrin, indolocarbazoles
Immunostimulators Forphenicinol, FK-156, bestatin
Activity against tumor cells, Kazusamycin, lactoquinomycin
resistant to other antitumor drugs
Immunosuppressants Spergualin, 15-deoxyspergualin
Adenosine deaminase Adecypenol, pentostatin
5 ‘-Nucleotidase Nucleoticidin
Inhibitors of the eukaryotic cell cycle Reveromycins
Melanogenesis OH-3981
ture or plant extracts; however, better puri-
fied fractions and components are required 3 Classical Anticancer
for results which are more reliable. Thus, the
test program of the National Cancer Institute
Drugs
of the USA operates in vitro disease-oriented
screens which are directed specifically to cell 3.1 Nucleoside Antibiotics and
type specific agents. New compounds are Analogs of Nucleobases
tested against a panel of approximately 60 cell
lines derived from human solid tumors such
as colon tumors, melanoma, kidney tumors, The structural class of the naturally occur-
ovarian tumors, brain tumors, and leukemia ring nucleoside antbiotics comprises approxi-
(GREVERet al., 1992; KAO and COLLINS, mately 150 representatives (ISONO,1988; As-
1989). A more recent approach is the usage of ZALOS and BERDY,1978). Numerous analogs
cell cultures in the screening for inducers of have been obtained by chemical synthesis.
tumor cell apoptosis (YAMAZAKIet al., They exert their antimicrobial and cytotoxic
1995). activity as “pseudo” nucleobases or nucleo-
Increasing use is been made of transformed sides which are activated to “pseudo” nucleo-
cells (such as those transformed by retrovi- tides. Thereafter, they act via a negative feed-
ruses, e.g., the Rous sarcoma virus) to search back mechanism on nucleotide-forming bio-
for agents which display a selective toxicity or synthetic pathways or inhibit DNA- and
which alter tumor cell morphology. In addi- RNA-dependent polymerases. However, the
tion, these cell lines are useful as tools in incorporation of “pseudo” nucleotides may
screening for inhibitors of special oncogene- also diminish the template function of DNA.
encoded cellular functions (UMEZAWA, Antagonists of folic acid, purines, and pyrim-
1989). Another more recent, promising ap- idines have been used in anticancer chemo-
proach concerns the protein targets of drugs therapy.
interfering with the regulation of the cell cy- Nucleobase analogs such as 5-fluorouracil
cle. High-throughput screening assays have (synthetic), 6-mercaptopurine (synthetic),
been developed for a series of receptors and thioguanine, thiouridine (NISHIKORIet al.,
enzymes of the cellular signaling cascade 1992), 7-hydroxy-guanine (Streptomyces sp.)
which are crucial in the development of mal- (KITAHARAet al., 1985), and “pseudo” nu-
ignancy. cleosides such as arabinosyl cytosine (synthet-
Despite these promising in vitro ap- ic), cadeguomycin (TSUCHIYAet al., 1992),
proaches trials with animals will finally be re- oxanosin (Streptomyces hygroscopicus)
quired to reduce the number of prescreening- (YUANet al., 1985), neplanocin (Ampullariel-
positive compounds to an amount which justi- la reguluns) (YAGINUMA et al., 1981), spica-
fies continuing investigations, including clini- mycin, its derivative SPM VIII (KAMISHOHA-
cal trials. In this context, the use of human tu- RA et al., 1994) and others (see, e.g., ISONO,
mor xenograft models represents a new ap- 1988) display antitumor and antiviral activi-
proach to the evaluation of putative antican- ties against various cells and viruses. The gen-
cer agents. eral therapeutic problem associated with
An effective screening program for new an- these agents is their poor selectivity and rela-
ticancer agents will need new sources of drugs tively high toxicity. However, several exam-
such as rare microorganisms, plants, and ani- ples of a more selective action of nucleosides
mals. During the last ten years, marine organ- against viruses are known; synthetic drugs
isms such as tunicates, molluscs, dinoflagel- such as acyclovir and its derivatives are pow-
lates, and bryozoa have become the subject of erful inhibitors of herpes virus thymidylate kin-
an expanding field of research in the search ase, and hence invaluable therapeutic agents
for new antitumor agents (see, e.g., dolasta- (Fig. 2, see also the therapeutic effect of 2,3-
tins, spongistatins, bryostatins, okadaic acid, dideoxynucleosides on HIV virus infections).
and other cytotoxic metabolites) (JENSENand A more specific antitumor activity has been
FENICAL,1994). demonstrated for inhibitors of enzymes asso-
H2N
H
7~mxy#wnutn
T h ~ ~ i l
* HO-CH,
0
OH
&y
OH
NH
KLy
va.oXO(lin
FH2
y
COOH
2
Fig. 2. Structures of
neoplasm inhibitory
nucleobase analogs,
nucleosides, and anti-
folates.
ciated with special kinds of malignancies, Inhibitors of this enzyme such as pentostatin
such as adenosine deaminase. (2-deoxy-coformycin; Streptomyces antibioti-
The importance of purine metabolism for cus) (HOLLIS-SHOEWALTER et al., 1992), and
the immune system was recognized due to the adechlorin (Actinornadura sp. OMR-37)
observable associaton of deficiencies in pu- (OMURAet al., 1985) (Fig. 2) were aimed to
rine salvaging enzymes, such as adenosine prevent the degradation of arabinosyl adeno-
deaminase and purine nucleoside phosphoryl- sine by this enzyme (AGARWALet al., 1983;
ase, with heritable immunodeficiency. Coad- OMURAet al., 1986).
ministration of adenosine deaminase inhibi- In eukaryotic cells, 5 '-nucleotidase is nec-
tors in the treatment of leukemia by arabino- essary for the formation of the 5'cap struc-
syl adenosine caused syndromes similar to ture of mRNA. Hence, inhibitors of this en-
those observed in immunodeficient patients. zyme such as the polysaccharide nucleoticidin
(Pseudomonus sp.) (OGAWARAet al., 1985) (MURATA,et al., 1987). Diazaquinomycin A

inhibit malignant cell growth. Moreover, fo- (MURATAet al., 1985) and vanoxonin (KA-
late metabolism was suggested as an addition- MA1 et al., 1985) are microbial inhibitors of
al promising target for antitumor drugs (see, thymidylate synthase which have cytotoxic ef-
e.g., the role of methotrexate as an inhibitor fects on cell cultures (Fig. 2).
of tumor cell dihydrofolate reductase; Ros- Moreover, a series of nucleotide analogs
OWSKI et al., 1992) (Fig. 2). More recently, has been isolated from microorganisms,
thymidylate synthase (VANDER WILT et al., which are active as antitumor and antimetas-
1994; KALMAN,1989) and serine hydroxyme- tatic agents due to their interaction with sugar
thyltransferase (RAo, 1991) have been sug- nucleotide glycosyltransferase (see below)
gested as targets for new anticancer drug de- (KHANand MALTA,1992).
velopment.
The search for inhibitors of folic acid me-
tabolism from microbial cultures (e.g., by the 3.2 Drugs Binding Non-Covalently
use of Entercoccus facium as an assay organ- to the DNA
ism) led to the discovery of analogs and inhib-
itors of thymidylate synthase as “antifol- Netropsin and distamycin (pyrrolamidine
ates”. Thus, 7-hydro-8-methylpteroylglutami- antbiotics from Streptomyces netropsis and S.
nylglutamic acid (HMPGG) was isolated as a dimtallus, respectively) are invaluable bio-
folic acid analog from an Actinomyces culture chemical tools in recent DNA research
Nelropsin
Distamycin
Fig. 3. Structures of antibiotics binding

non-covalently to the DNA.
Hm:
HO
0
Charbeusin
(Fig. 3). Both antibiotics bind to the minor tions with DNA have been studied extensive-
groove preferably at regions rich in adenosine ly by physicochemical methods (WANGet al.,
and thymidine. The complex formation of ne- 1987; WANG,1992).
tropsin and the duplex DNA (d(5‘CGCGA- The term intercalation refers to the inser-
ATTCGCG3 ’)) suggested that at the tion of flat, planar, aromatic molecules (qui-
5’AATT sequence in the center the three nolines, acridines, phenanthridines, phenoxa-
NH groups of netropsin form hydrogen bonds zines, anthraquinones, fluorenes) into the mi-
with 0-2 of thymine and N-3 of adenine nor groove of DNA, between two adjacent
(KOPKAet al., 1985) (see Fig. 3). Distamycin base pairs (WANG,1992; WARING,1975). Po-
which is a closely related antibiotic forms van lar substituents such as amino acids or sugars
der Waals bonds between its 5 NH-groups (see the structures of actinomycin D and dox-
and 0 - 2 of thymidine and the N-3 of adenine orubicin, Fig. 4) stabilize the intercalation
in the DNA minor groove (COLLet al., 1987; complex by the formation of hydrogen bonds
DERVANet al., 1987) (see Fig. 3). with the phosphodiester groups of the deoxy-
As a consequence of binding, the unwind- ribose moieties of DNA (MANGER,1980;
ing of DNA during replication and transcrip- KNUGHet al., 1980). Intercalation results in
tion as well as the access of the pertinent en- the distortion of the DNA and steric stress
zymes and proteins to specific DNA sites are whereby neighboring bonds are weakened.
prevented (ZIMMERet al., 1990). Chemical As a consequence, the template function of
modifications of distamycin were done to im- the DNA is reduced, and single and double
prove the antiviral properties of its structures strand breaks as well as frameshift mutations
(ARCAMONE, 1994). may occur. In addition to some synthetic and
Non-covalent binding to DNA at different plant-derived drugs (flavines, acridine dyes,
sequences was also shown for a series of poly- ellipticin), the actinomycins (KNUGHet al.,
cyclic aromatic antbiotics such as chartreusin 1980; ADAMSONet al., 1979) and the anthra-
(KRUGERet al., 1986) (Fig. 3), chromomycin cyclines (EL KHADEM,1982; CASSIDYand
(KOAMURA et al., 1988; UCHIDAet al., 1985), DOUROS,1988; LOWN,1988) are well known
angucyclins (saquayamycin), and saframycin intercalators of microbial origin. The latter
(see Fig. 8a). However, none of these com- type of antbiotics differs from the intercalat-
pounds are used in anticancer therapy. ing drugs by its radical-forming properties
Both chromomycin A3 and olivomycin in- and its inhibitory effect on DNA topoisomer-
hibit the synthesis of DNA and RNA in vitro ase I1 (see below).
and the function of RNA polymerase and
DNA polymerase I in vivo. Binding of chro-
momycin to DNA is magnesium-dependent 3.3.1 Ellipticin
and occurs preferably at regions with a high
G/C content. The complex formed is stable to Ellipticin and its 9-methoxy derivative were
nuclease digestion. Similar to chromomycin isolated from plants such as Excavatia cocci-
A3, an antibiotic from Streptomyces plicatus, nea and Chrosia moorie. They inhibit DNA
mithramycin binds preferably to d(5 ‘ATG- synthesis by binding to the minor grove of
CAT3’)2 regions (BEAUVILLEet al., 1990). double-stranded DNA.
Olivomycin interferes with the elongation
step of RNA polymerase and inhibits DNA
polymerase I by binding to guanine- and cyto- 3.3.2 Actinomycins
sine-containing DNA regions.
Already in 1955 the antitumor activity of
actinomycin D (Fig. 4) (from Streptomyces
3.3 Intercalating Antibiotics antibioticus) was observed (TSCHAGOSHIet
al., 1986). Representatives of the actinomycin
Intercalating drugs have been important chromopeptide family are extremely potent
both as tools of biochemical research and as inhibitors of RNA polymerase, and they have
anticancer agents. Their molecular interac- widely been used as a biochemical tool to
0 OH 0
Mel-Val-0 MA-Val-
MeGly
t
L+
D-Val
4
MGJOH
Me Me
Doxorubicin
Actinomycin D
Echinomycin
UK 63.052
CH.
NHyN'3
I
Mitoxantron(X=oH)
Fig. 4. Structures of intercalat- Bisanthrene
ing drugs. Ametantron (X=H)
NH
study RNA synthesis (GALE et al., 1981; sandramycin, a related 3-hydroxy-quinoxal-

FORNICA,1977). Actinomycins are composed inic acid cyclopeptide antibiotic was isolated,
of a phenoxazinone chromophore, which is which lacks a sulphide bridge (MATSONet al.,
the intercalating unit, and two linked penta- 1993). The compound UK-63.052 (Sfrepto-
peptidolactone moieties (MAYERand KATZ, myces bruegensis) (Fig. 4) is a new represent-
1978; KATAGIRI,1975). The homologs are ative of the quinomycin group of antibiotics
distinguishable by their amino acid composi- (RANCEet al., 1989).
tion. The compositon of the actinomycin com- In every case, the alanine residue of the oc-
plex depends on the available amino acid pre- tapeptide ring of echinomycin type antibiotics
cursors (FORNICA,1977). Intercalation of the is a critical determinant of the observed spe-
actinomycins requires the 2-NH2 group of the cificity for CpG dinucleotide sequences
phenoxazinone chromophore and the quinoid (WARING1979, 1990). This is due to hydro-
2-C 02;DNA binding occurs at (d(CAT- gen binding interactions which involve NH
GAT))2 sequences (ZHOUet al., 1989). Inter- and CO groups of the alanines together with
calation of actinomycin D and its analogs in the 2-amino group and N-3 of guanine in the
DNA was extensively studied by spectroscop- minor groove of the helix. The binding of
ic methods (KNUGHet al., 1980). Computa- echinomycin type antibiotics to DNA is very
tional techniques have been employed to cal- tight and remains stable even during gel elec-
culate the substituent effect on free energy of trophoresis of DNA fragments in footprinting
binding to DNA (LYBRAND,1988). In the analysis (Fox, 1990).
past, actinomycin D (dactinomycin, Fig. 4)
was used in clinical anticancer trials but due
to its high toxicity its therapeutic use in 3.3.4 Anthracyclines and Related
Wilms tumor, choriocarcinoma, testis carcino- p-Quinones
ma, neuroblastoma, and sarcoma is disputed.
In order to reduce toxic side effects and to Both, daunomycin and doxorubicin (Fig. 4)
improve the efficacy against certain tumor (a semisynthetic derivative of daunomycin)
cells, a series of analogs of actinomycin D was bind to DNA sites which contain either only
obtained by semisynthesis, but none of them (G:C) or (A:T) base pairs (ASZALOSand
appeared to be more promising than the par- BERDY,1978; SALUZand WIEBAUER,1995;
ent compound (ADAMSONet al., 1979; FOR- NEIDLYand WARING,1983; WILMAN,1990).
NICA, 1977). In general, a three base pair binding site is re-
quired for the intercalation. The aminosugar
at C-7-0 of the benzo[a]naphthacene quinoid
3.3.3 Quinoxaline Antibiotics aglycone is needed to replace water from the
minor groove and to anchor the molecule to
The anticancer activity of echinomycin the phosphodiester linkages between the ad-
(Fig. 4) was rediscovered due to investiga- jacent deoxyribose moieties (WANG et al.,
tions of its mode of action at the molecular 1987). Due to these properties some repre-
level. It was the first DNA bis-intercalator to sentatives of the anthracyclines induce a small
be identified (WARING,1992). degree of helical unwinding.
Bis-intercalation involves the binding of a Doxorubicin and less frequently daunoru-
drug to DNA via the quinoxaline rings, with bicin are used in the treatment of acute lym-
the peptide moieties binding to the DNA mi- phatic leukemia, acute myeloid leukemia,
nor groove (ADDESet al., 1992). Representa- lymphoma, sarcoma, Wilms tumor, and can-
tives of this group of antibiotics display differ- cer of the breast, lung, bladder, thyroid, and
ent sequence specificity (ADDESS et al., prostate. After recognizing the anthracyclines
1992), whereby CpG sequences are preferred. as lead structures a series of intercalating an-
Triostin and luzopeptin are related antibiotics thraquinone type anticancer agents such as
which display similar bis-intercalating proper- mitoxantron, ametantron (Fig. 4), anthropy-
ties (WANG,1992) as detected by X-ray dif- razol, and bisanthrene (Fig. 4) was developed
fraction and NMR spectroscopy. Recently, as anticancer drugs and used therapeutically,
e.g., in the treatment of breast cancer (ASZA- SHEN, 1994; CHAKRABORTY et al., 1994;
LOS,1988; HOLLIS-SHOEWALTER, 1988). CHENand LEROY,1994; HSIEH,1992), either
by interfering with ATP hydrolysis or with
the DNA cleaving subunit A.
3.4 Inhibitors of Enzymes of DNA Drugs introducing lesions in the DNA trap
Replication and Transcription: the transient topoisomerase 11-DNA inter-
mediate which is held together by two coval-
DNA Topoisomerases ent bonds. Anthracyclines (see Fig. 7), ellipti-
cin, and epipodophyllotoxins (etoposid and
Dramatic advances have been achieved in tenoposid) (Fig. 5b) are the major antitumor
the field of DNA topoisomerases. These in- drugs that form cleavable complexes in euka-
clude the cloning of new topoisomerase genes ryotes.
involved in DNA recombination and segrega- In a search for new inhibitors of topoisom-
tion of replicated DNA (WATT and HICK- erase I1 (HECHTet al., 1992) a benzoanthra-
SON, 1994; POOT and HOEHN, 1993; GIAC- quinone (UCE 1022) has recently been iso-
CONE, 1994). Hence, topoisomerases have be- lated from Puecilomyces sp. (FIJI] et al.,
come promising targets of modern anticancer 1994). Other inhibitors of microbial origin are
agents (PAOLETIT,1993). Antitumor agents dotriacolide (FIJII et al., 1994) (inhibitor of
such as epipodopyllotoxins (etoposid, tenipo- topoisomerase I and DNAse), TAN-1496A (a
sid), acridine dyes, anthracyclines, and ellipti- diketopiperazine, Strepfomyces sp.; Fig. 5a)
cins were found to be specific inhibitors of to- (FUNABASHI et al., 1994), and UCE6 (a non-
poisomerase 11, and the plant alkaloid camp- glycosylated anthraq uinone, Streptomyces sp.;
tothecin was recognized as a specific inhibitor Fig. 5a) (FIJII et al., 1993). Anthracyclines
of mammalian topoisomerase I (Fig. 5a) were also shown to inhibit toposomerase I in
(FOSTELand SHEN,1994). addition to topoisomerase I1 (CROW and
Topoisomerases as targets of antibiotics CROTHERS,1994).
and antitumor drugs, their relation to the to- Recent screening approaches in the search
pological stage of intracellular DNA, and for microbial inhibitors of type I1 topoisomer-
their function in replication have been dis- ases (PAOLETTI,1993) revealed new inhibi-
cussed in a series of reviews (ZIMMERet al., tors, such as BE-22179 (a peptide from Strep-
1990 FOSTEL and SHEN,1994; CHENand LE- fomyces) (OKADAet al., 1994), saintopin (a
ROY, 1994). The enzymes catalyze the con- benzo[a]anthraquinone from Paecilomyces
certed breakage and rejoining of the DNA sp.; Fig. 5b) (YAMASHITA et al., 1990a), BE-
backbone, and hence they are indispensable 10988 (Sfrepfomyces xanthocidicus; Fig. 5b)
for DNA replication, transcription, chromo- (OKAet al., 1991), streptonigrin (TOLSTIKOV
some segregation, and recombination. Mam- et al., 1992; YAMASHITA et al., 1990b) and its
malian type I topoisomerases mediate the re- derivatives, and UCT4B (terpentecine type,
laxation of negatively or positively super- Strepfomyces sp.; Fig. 6b) (UOSAKI et al.,
coiled DNA by transiently breaking and re- 1993; KAWADAet al., 1992a). The majority of
leasing one DNA strand in such a manner these inhibitors exhibit potent antitumor ac-
that the linking number changes by steps of tivity. Some also affect bacterial topoisomer-
one. Type I1 topoisomerases break and reli- ase I1 (gyrase) which is the target of several
gate phosphodiester bonds in double- important antibacterial agents. A cytotoxic
stranded DNA passing another duplex region quinoid system, popolahuanone E (Fig. Sb),
through the break, thereby altering DNA to- was isolated from a Pohnpei sponge. It is se-
pology. In addition, they are structural pro- lectively toxic for the A549 non-small cell hu-
teins involved in the spatial organization of man lung cancer cell line (CARNEY and
chromatin (ZUNINOand CAPRANICO, 1990). SCHEUER,1993). In a series of comparative
One group of agents such as coumarin deriva- studies it was shown that the interaction with
tives (novobiocin, coumermycin) and the syn- mammalian topoisomerase I1 is one of the
thetic quinolone antibacterial agents act at major causes of cytotoxicity of the anthracy-
the level of the enzyme subunits (FOSTELand cline type antibiotics (CAPRANICO and ZUNI-
H
%
o Me
OH OH OH 0
UCT 48
"$+,pCoN4
Sainbpin
N
0 Me Popolohuanone E
BE409BB
OH
I
0
4
MeO
0%
e
Y
Me'CH-O
O
\ 4
Pdophyllotoxin(Rl: H; R2 Me)
b Teniposid (R1:X ;R2: H); Etoposid (Rl: Y; R2: Me)
Fig. 5. Structures of topoisomerase inhibitors. a Inhibitors of topoisomerase I, b inhibitors of topoisomer-

ase 11.
NO, 1990 ZUNINO and CAPRANICO,1990; are active against a broad spectrum of human
GREVERet al., 1992). Although interference tumors (SLICHENMYER et al., 1993). They
with the functions of DNA topoisomerases I slow down the religation step of topoisomer-
and I1 holds great promise for the treatment ase I and stabilize the covalent adduct be-
of cancer, these drugs may themselves be tween the enzyme and DNA. In S phase cells
dangerous due to their mutagenic and carci- double-stranded breaks occur at the position
nogenic potential (ANDERSONand BERGER, of the topoisomerase I-DNA adducts which
1994). DNA helicase has recently been pro- is probably the reason for the high cytotoxici-
posed as a target involved in DNA replica- ty of these agents. Both drugs are now under
tion, repair, recombination and transcription clinical investigation (SLICHENMYER et al.,
by unwinding of double-stranded DNA. Heli- 1993; CURRAUet al., 1993).
quinomycin as a new spirocyclic aromatic
compound from Streptomyces sp. effectively
inhibits the enzyme (CHINOet al., 1996). 3.5 Agents Forming Covalent
Bonds with DNA
Podophyllotoxins as inhibitors of Covalent modification of DNA by interac-
topoisomerase 11: tion of reactive molecules with bases or su-
gars blocks its template function and induces
The antitumor drugs podophyllotoxin the onset of repair processes. As an initial
(Fig. 5b), peltatins, and other lignane type step, radical species are generated by the drug
compounds have been extracted from Podo- molecule due to its interaction with metabolic
phyllum peltatum L. These compounds inter- enzymes such as cytochrome-dependent and
act non-covalently with tubulin and inhibit other oxidoreductases. The result is inter-
mitosis during metaphase (see Sect. 3.6). Due strand crosslinking and single- or double-
to their high toxicity for mammalian organ- stranded DNA cleavages. Some of the promi-
isms they cannot be used for therapeutic pur- nent representatives of this type of commer-
poses. Semisynthetic derivatives, such as te- cially available agent will be discussed (PINE-
noposid, etoposid (Fig. 5b), and mitopodosid DO, 1980).
appear to be more promising. They poorly in-
terfere with tubulin and mitosis but instead,
inhibit the transport of nucleosides in cells 3.5.1 Mitomycin C
and the incorporation of thymine and uridine
into DNA and RNA due to interaction with Mitomycin C (Streptomyces caespitosus)
topoisomerase I (FRANZ,1990). (Fig. 6) is used in anticancer chemotherapy by
virtue of its inhibitory effect on DNA synthe-
sis and its ability to induce strand cleavages
Camptothecin type compounds as inhibitors (TOMASZ,1994). It is known as one of the
of topoisomerase 11: most potent antitumor antibiotics. The molec-
ular mechanism of action of this highly toxic
Camptothecins are plant dimeric indole al- antibiotic involves the reduction of the qui-
kaloids which were discovered as selective none and subsequent activation of C-1 (PINE-
growth regulators for several species of DO, 1992; TOMASZ,1994; SARTORELLI et al.,
mono- and dicotyledonous plants (BUTAand 1993). Mitomycin C and its analogs, such as
KALINSKI, 1988; WALLand WANI,1993; SLI- porfiromycin and mitiromycin (KASAIet al.,
CHENMYER et al., 1993; CURRAU, 1993) 1991; ARAIet al., 1994) are used as an alter-
(Fig. 5a). Chemical studies on inhibitors of native to radiation during the treatment of
DNA topoisomerase I which lead to the pro- hypoxic solid tumors which have acquired en-
duction of semisynthetic derivatives, such as hanced sensitivity to this bioreductive alkylat-
CPT-11 and topotecan suggested a therapeu- ing agent. It has been suggested that in hyp-
tical potential for these types of compounds oxic cells, one-electron transfer enzymes, such
(ANDOHet al., 1993). Topotecan and CPT-11 as diaphorase, control bioreductive alkyl-
ation. At least six different enzymes are capa- 3.5.3 Cyclopropane, Aziridines,
ble of activating mitomycin C and other simi-
lar drugs. The nature of the activated molecu- and Epoxide Compounds
lar species and the resultant biological lesions
can vary with the activating enzyme. This, in The compound CC-1065 (Sfreptomyceszel-
ensis; Fig. 6) (MARTINet al., 1980) is a hetero-
turn, causes variations in toxicity for different
cell types. The process of activation is accom-cyclic antibiotic which forms a DNA adduct
panied by the opening of the conformational- via the initial interaction of the DNA guanine
residues with a cyclopropane ring attached to
ly constrained aziridine ring to yield free radi-
cals as intermediates. The double-bonded a cryptic p-quinoid structure. This is followed
structure formed interacts with guanine bases by molecular rearrangements and cleavage of
in opposite DNA strands, thereby crosslink- the phosphodiester bond from deoxyribose.
ing them in a bifunctional manner (RAUTH The compound CC-1065 displays sequence
and RAYMOND,1993; TOMASZ,1994; ROCK- specificity in that it binds to regions rich in
WELL et al., 1993; WARDMAN, 1990; BUTLER guanine (SCAHILLet al., 1990).
and HAEY, 1987). Frameshift mutations ap- A series of microbial products containing
pear to be another consequence of DNA re- aziridine and epoxide structures, such as azi-
pair and cause of potent mutagenic and tera- nomycins (Strepfomyces griseofuscus), WF-
togenic activity of the mitomycins. In order to3405 (Amauroascus aureus), and duocarmy-
obtain less toxic and more potent mitomycin cins A and SA (S. zelensis) (Fig. 6; ICHIMURA
C analogs, both congeners of mitomycin fer- et al., 1991) was suggested to form covalent
mentations (see, e.g., albomitomycin KW- bonds with DNA. The exhibit extremely po-
2149) and synthetic derivatives have been iso- tent cytotoxic activity (ZCso=lo-'* M-
lated (KONOet al., 1991, 1995; KASAIet al., 10-9M for Hela S3 cells). The constrained
1991). DNA damage appears to be critical for three-membered ring structure in addition to
the cytotoxicity of mitomycins. Both inter- particular steric and electronic conditions
strand and intrastrand crosslinking occur in permits a nucleophilic attack on DNA bases.
mitomycin-treated cells. Duocarmycin SA (ICHIMURA et al., 1990)
was found to be the most stable and most po-
tent cytotoxin of these agents. They alkylate
3.5.2 Anthramycins DNA in mechanistically similar manner to
CC-1065 (ICHIMURA et al., 1990,1991). DNA
Anthramycins (pyrrolo(l,4)-benzodiaze- alkylation occurs via addition of adenine N3
pines) bind tightly via their C-11 atom to the to the least substituted carbon of the acti-
2-amino group of guanine bases with the vated cyclopropane within AT-rich minor
elimination of water (Fig. 6). The amino groove sites (BOGERand JOHNSON,1996). In
groups of these antibiotics play an important this context, scirpene type mycotoxins such as
role in their molecular mechanism of action trichothecin (Fig. 6), trichodermole, anguid-
(BARKLEYet al., 1986 MORRISet al., 1990). ine, nivalenol, crotocin, and their analogs
Representatives of these agents include an- were tentatively used as antitumor agents.
thramycin (Fig. 6), sibiromycin, tomaymycin, They are frequently occurring products of
neothramycin, and porothramycin (TSUNA- Fungi imperfecti (Trichothecium, Fusarium,
KAWA et al., 1988). Due to the high risk of Myrothecium, Trichoderma)and inhibit DNA
hazardous side effects, the anthramycins offer and protein synthesis.
no advantage in cancer treatment compared
to other cytotoxic antitumor drugs. Some an-
thramycins were evaluated in clinical trials 3.5.4 Streptonigrin
but they have not come to general use.
Streptonigrin (bruneomycin) (Fig. 6) and
its natural analogs (LIN et al., 1992) are cyto-
toxic radical-forming anticancer drugs which
modify DNA covalently and cause strand
H$cyJ& OH H
p::
H2n* ,,ocy Anthramycin
NH
hlt
N
-
0 /
0 CONH,
MitomyeinC
OCH,
I
0
NH
'0
0 H.CO
Pynolo~,l-cl[lnbcnz~i~epine Duocarmycin A
DNA adduct (41
Streptonigrin
hwo
Me0
cc-lo65
crotonyl
Triehothccin
Fig. 6. Structures of drugs binding covalently to the DNA.
breaks. Interactions with topoisomerase I1 35.5 Anthracyclines

appear to be involved in this activity (YAMA-
SHITA et al., 1990b). Streptonigrin and a se- The anthracyclines (Fig. 7a) belong to a
ries of chemical derivatives were shown to in- class of antitumor drugs which exhibit activity
hibit HIV reverse transcriptase (TAKEet al., against a spectrum of human cancers. Only a
1989). few tumors, such as colon cancer, melanoma,
HO wF2Rp OH
O-(sugar)i-3
COOMe
H CH3
\ 0
--
OH O OH 0(sugar)l-3
@&;
OMe
a Osugar
\ / &\~ H z o H / OH
- OMe 0 OH
OMe 0 OH -
- --
DounoruMcin Doxorubicin
cH3&OH
0 cH3@0 OH
cH3@
OH
Aclacinomycin A 0 Nogalamycin
(AcbruMcin)
IL-rhodospminyl-
L-fucosyl-
Lcinerulose
b
Fig. 7. Structures of anthracycline type antibiotics.
a General substitution pattern of anthracyclinone aglycones of anthracyclines, b anthracycline structures,
c recently discovered new anthracycline structures.
Me@NMe,
Yellamycin B
OH
Alldimycin OH
COOCH,
OH 0 OH 0
L-rhodosamin-
Epelmycin A L-rhodosaminyl-
Fig. 7c Lcinerulose A
chronic leukemias, and renal cancer are un- (1) Synthesis of new anthracycline analogs:
responsive to them. The first clinically effec- This process has continued from the
tive anthracycline, daunorubicin (DNR), was 1960s up to the present and might also
discovered independently in 1963 at Farm- continue in the future (LOWN,1993). No
italia (daunomycin) and RhGne-Poulonc (ru- one has kept a count of the anthracycline
bidomycin). Later on, adriamycin (doxorubi- analogs synthesized over the past 25
cin, DOX) was discovered as the product of a years, but their number probably exceeds
mutant strain and was also obtained semisyn- 2000, and more are being reported every
thetically. month.
In 1969, it was demonstrated that DOX was (2) Coadministration of other agents with
less toxic and more active against a much DOX
broader spectrum of tumors than DNR. The Researchers have attempted worldwide
former rapidly replaced the older drug in clin- to administer doxorubicin in conjunction
ical applications. In general, cumulative car- with other substances that will mitigate
diotoxicity limits treatment with DOX to ap- cardiotoxicity or overcome drug resist-
proximately nine months at usual dosages, ance of the cancer cells (STEINHERZand
and most cancers develop resistance to this STEINHERZ,1991; VAN KALKENet al.,
agent. 1991).
DOX differs from DNR only by an addi- Moreover, immunoconjugates of DOX
tional hydroxyl group. This fact has encour- and different anthracyclines incorporated
aged researchers worldwide to search for ana- into liposomes are being evaluated in
logs of DOX that display lower acute toxicity clinical trials to minimize heart exposure
or cardiomyopathy, that can be administered to the drug while maintaining antitumor
orally, and that have different or greater anti- efficacy (PEREZ-SOLERet al., 1995).
tumor efficacy. The aim of this research was (3) Screening for new anthracycline com-
three-fold: pounds of microbial origin:
DNR and DOX as well as their biosyn- the earliest detectable intermediates of an-
thetic congeners are all derived from acti- thracycline biosynthesis (WAGNER et al.,
nomycetes, particularly from the genus 1991; ECKARDT and WAGNER,1988). More
Streptomyces. They are obtained by using recently, the biosynthesis of anthracycline
the well known fermentation and recove- aglycones was studied in detail by a genomic
ry techniques generally used in antibiotic analysis (STROHLet al., 1989; BARTELet al.,
technology. Many of the more than 400 1990). The biosynthetic origin of many of the
anthracyclines isolated and characterized known anthracycline aglycones, such as E-
to date have been isolated from blocked and ppyrromycinones, E-, p, a-,and &,-rho-
mutants of a variety of strains. The main domycinones, €-iso-rhodomycinones, dauno-
anthracyclines selected for clinical evalu- mycinone, and adriamycinone could be as-
ation are shown in Fig. 7b. Alldimycins cribed to aklanonic acid as a single interme-
(JOHDOet al., 1991a), yellamycins (JoH- diate.
DO et al., 1991b), epelmycins (Fig. 7c) The pigmented antibiotics are formed in
(JOHDO et al., 1991c), respinomycin the mycelium of microorganisms grown in
(UBUKATA et al., 1993), cororubicin shake flasks or in stirred aerated fermenters
(ISHIGAMIet al., 1994), mutactimycin on media containing the usual organic nu-
(MAEDA et al., 1992), betaclamycin B trients and inorganic salts. For the biotechni-
(YOSHIMOTOet al., 1992), cinerubin R cal production of daunorubicin, the fermen-
(NAKATAet al., 1992), and rubomycins F ters are operated usually for up to 10 days.
and H (FOMICHOWA et al., 1992) are ex- Because anthracyclines are hazardous chemi-
amples of the previously discovered new cals, due to their cardiotoxicity and/or muta-
anthracycline type structures. genicity, special care is needed when isolating
and purifying these agents from the fermenta-
Anthracyclines consist of a tetracyclic agly- tion broth (UMEZAWAet al., 1987). Recovery
cone (anthracyclinone, tetrahydro-naphtha- and purification of the anthracyclines from
cene-quinone) which is linked with up to ten large-scale fermentations is described else-
sugars usually attached at positions C-7 or where (WHITEand STROSHANE, 1984).
C-10 C-Cglycosylated derivatives have also
been reported. The general substitution pat-
tern of the anthracyclinones is shown in New anthracycline analogs:
Fig. 7a. The nogalamycin family is distingui- the next generation:
shable by the unusual substitution pattern of
the left-side aromatic ring by a In the 1970s and the early 1980s the main
bis-oxa-bicyclo[3.3.l.]nonane ring system 'aims were to expand the antitumor spectrum
(DUMITRIU,1996). of DOX, to reduce its cardiotoxicity, and to
Fig. 7c shows a series of anthracycline develop an orally administrable compound.
structures representing various building pat- The outcome of this research was a few ana-
terns. The aglycones differ in the hydroxyla- logs which display moderate clinical advan-
tion of the aromatic nucleus and in the sub- tages (WEISS, 1992). In the 1980s develop-
stituents at positions 7 and 10. Modifications mental efforts in this field also turned to the
of the side chain (e.g., by oxidations, reduc- problem of multi-drug resistance of cancer
tions) andor methylations are a consequence cells. For instance, F860191, AD-198,
of numerous enzymatic and non-enzymatic FCE23762, ME2303, and MX-2 were re-
reactions occurring during biosynthesis. The ported to be active against DOX resistant and
anthracyclinone part of the anthracyclines is multi-drug resistant cell lines and even to ex-
produced via a polyketide synthase system by ceed DOX activity sometimes by a factor of
a series of reactions similar to that of fatty up to ten (for a survey, see WEISS,1992).
acid biosynthesis (VANEKet al., 1977). The
intermediate undergoes hydroxylations, me-
thylations, glycosylations, and other modifica-
tions. Aklanonic acid and aklaviketone are
Formation of oxygen radicals by feeding of amino acids has been reported

anthracyclines: (TAKITA,1984). Peplomycin and liblomycin
(A, B, C) (KURAMOCHI et al., 1988; KURA-
As a general feature, p-quinoid structures MOCHI-MOTEGI et al., 1991; TAKAHASHIet
such as mitomycin C, streptonigrin, tetraceno- al., 1987c) are semisynthetic derivatives of
mycin type antibiotics, angucyclines, and the bleomycins which contain a more space-filling
representatives of the anthracycline family side chain. The aim of their synthesis was to
(BUTLERand HAEY,1987; KONOet al., 1991; improve the stability against hydrolysis and to
KASAI et al., 1991; BARKLEYet al., 1986; increase lipophilicity (OTSUKAet al., 1988;
FEIGand LIPPARD,1994) can be reduced en- SEBTI and LAZO, 1988; NISHIMURA et al.,
zymatically by cytochrome-dependent en- 1987; OHNO,1989). A general characteristic
zymes to yield semiquinone radicals (Fig. 8). of these agents is that they form heavy metal
These will combine with molecular oxygen to complexes with divalent cations such as
form superoxide anion and hydroxyl ion radi- Cu2+, Co2+, Zn2+, Fe2+, and can even be
cals. Thereafter, the free radical species react isolated as heavy metal (Cu”) chelates from
preferably with the deoxyribose moieties of the aqueous solution (Fig. 9).
DNA splitting single bonds between neigh- The mode of antitumor action of bleomycin
boring sugars. Although the anthracyclines has been subject to detailed investigations
could be considered as intercalating agents (MATSHURA,1988; STUBBEand KOZARICH,
and inhibitors of topoisomerase I1 (see 1987; STREKOWSKI, 1992). The antibiotic and
above) they are also potent producers of free its derivatves are commonly thought to exert
radicals. The anthracycline semiquinones are their biological effects as metal-drug com-
subsequently stabilized by deglycosylation of plexes which bind to the DNA. Thereafter,
the C-7-0-linked sugar. As a consequence of the adduct causes oxidative strand cleavage,
free radical formation, frameshift mutations and hence may be regarded as a low-molecu-
are induced. Thus the anthracyclines, bleomy- lar weight “DNAse” (HECHT,1994). The di-
cin (see below), mitomycin C, and similar an- saccharide moiety appears to be necessary to
ticancer agents are hazardous chemicals enable penetration of the antibiotic through
(UMEZAWAet al., 1987). the cytoplasmic membrane barrier, while the
thiazolopeptide side chain is essential for the
contact to the DNA (POVIRKand AUSTIN,
3.5.6 Bleomycin 1991). The cytotoxicity of bleomycin for some
kind of tumor cells results from the reductive
Bleomycin is a glycopeptide antbiotic pro- activation of dioxygen by metallobleomycins.
duced by Streptomyces verticillatus as a com- Iron-I1 ions are able to transfer electrons to
plex of not less than sixteen related antitumor molecular oxygen to form reactive and dam-
antibiotics (TAKITA,1984; MURADA,1988) aging intermediates of oxygen, such as super-
(Fig. 9). The main component bleomycin A2 oxide anion radical. Radical generation (02)
was originally used clinically in the treatment is initiated by the binding of iron-I1 ions to
of lung carcinoma, plate epithelium carcino- the paminoalanine-pyrimidine core of bleo-
ma, Hodgkins lymphoma, glioma, skin carci- mycins. Interaction of Fe-I1 with molecular
noma, and testis carcinoma (GIRIand WANG, oxygen generates the superoxide anion radi-
1989; SOLAIMAN, 1988; POVIRKand AUSTIN, cal to form Fe(II1)-bound bleomycin. The
1991; TAKITAet al., 1989). Pulmonary toxici- formation of radicals probably occurs in close
ty is dose-limiting. More than 200 semisyn- proximity to the DNA, since there is intercal-
thetic derivatives have been obtained in order ative binding of the bisthiazole moiety of
to improve activity and reduce toxic side ef- bleomycin to special minor groove DNA se-
fects (MURAOKAet al., 1988; TAKAHASHI et quences. It was also shown that the bleomy-
al., 1987a, b; TAKITA and OGINO,1987). The cin-Fe(II)-02 complex splits double-stranded
naturally occurring bleomycins are distin- DNA specifically at the GC(5’-3‘) and
guishable by the amino side chain. Directed GT(5’-3’) sequences. It has been proposed
biosynthesis of individual components by the that upon cleavage of the DNA the bleomy-
a
0
OMe
0 Me
0
Menoxymycin A R=NO(CHd2
II
Medermycin (R= N(CHd2 d' 'c,H,
Safnmycin A
OH
&:
Fig. 8. Drugs forming oxy-
gen radicals in the cells.
a Suggested mode of radi-
cal formation by anthracy-
cline drugs (see GOOR-
MAGHTIGH and RUYS-
SCHAERT, 1984),
b selected structures of \ 0 0 Me
drugs forming oxygen ra-
dicals (MYERSet al., H ~ V ~ H Atram
OH o ycin
1986). b Hydromycin
Fig. 9. Structure of bleomycin and the bleomycin-Fe(II)-O2 complex (STUBBE and KOZARICH,1987;
MATSUHARA,1988).
cin-Fe( 111) complex dissociates from the tar- 3.5.7 Macromolecular Antitumor
get and is reduced to the (Fe-11) complex. En-
zymatic modification of the antibiotic occurs Antibiotics and the Enedyine
by serum proteases which hydrolyze the Family of Cytotoxic Drugs
amide structure (SEBTI and LAZO, 1988).
Hence, inhibitors of proteases can be used to The most recent discovery in the field of ra-
improve the efficacy of bleomycin. Moreover, dical-forming, cytotoxic antitumor antibiotics
bleomycin-Fe(I1)-0, and bleomycin- concerned the extremely potent structural
Fe( 111)-O2 complexes catalyze lipid peroxida- group of the enedyines. In general, these anti-
tion concomitantly with singlet oxygen evolu- biotics are highly toxic for tumor cells, such as
tion. Thus, oxidative damage of the cellular P388 leukemia and B16 melanoma. Enedyine
membranes may be one of the mechanisms chromophores are frequent constituents of
determining the cytotoxic activity of this anti- macromolecular antitumor antbiotics, such as
biotic (KIKUCHIand TETSUKA,1992). neocarzinostatin and C-1027 (HOFSTEADet
al., 1992; ISHIDAet al., 1965; OTANI,1993).
The enedyine subunits display antitumor ac-
tivity even in the absence of the pertinent
apoprotein. Other members of this protein mechanism of neocarzinostatin action was

group are auromomycin (YAMASHITA et al.,shown to involve the formation of bisradicals
1979), macromomycin (CHIMURAet al., and subsequent damage at C-5’ of deoxyri-
1968), kedarcidin (LEET, 1992; HOFSTEADet bose of thymidylate in the DNA. The neocar-
al., 1992), largomycin (YAMAGUCHIet al., zinostatin chromophore binds to DNA in a
1970; MURAMATSU et al., 1991), maduropep- two-step process. The first is an external bind-
tin (HANADAet al., 1991a), and actinoxan- ing and the second involves the intercalation
thin (KHOKHLOV et al., 1969), but the chemi- of the naphthoate moiety between adjacent
cal nature of their chromophores has not al- DNA base pairs in the minor groove and the
ways been explored in detail. Similar to neo- electrostatic interaction of the amino sugar
carzinostatin, madurapeptin is composed of a with the charged phosphate backbone of the
large peptide backbone and an aromatic chro- DNA. Upon addition of thiol agents (R-SH)
mophore which is responsible for the biologi- free radicals are produced which attack the
cal activity (HANADAet al., 1991a, b). DNA to cause double-stranded breaks
The agent C-1027 from Streptomyces glo- (GOLDBERG,1991; HENSENS and GOLD-
bisporus is the most recently discovered rep- BERG, 1989; DUMITRIU, 1996).
resentative of the high molecular-weight anti- The esperamicins (LAM et al., 1993)
tumor antibiotics. It is composed of a 110 (Fig. 10a) as the main representatives of the
amino acid backbone and an unstable ene- non-protein group are composed of a bicyclic
dyine chromophore (OTANI,1993) (Fig. 10a). core (an enedyine, an allylic trisulfide, and an
A synthetic route to C-1027 was recently ela- enone) which is bound to a trisaccharide and
borated (IIDAet al., 1993). The apoprotein of a substituted 2-deoxy-~-fucose. Esperamicin
C-1027 displays some amino peptidase activi- Al (Actinomadura verrucosospora) was pro-
ty which is inhibited by amastatin and besta- duced on a large scale for clinical trials in can-
tin (SAKATAet al., 1992). cer treatment (GOLDBERGet al., 1989). Bio-
More recent research in the field of high synthetic studies on esperamicin Al (BENT-
molecular weight antitumor antibiotics has fo- LER et al., 1994) (Fig. lob) investigated the
cussed on the nucleotide sequences of apo- incorporation of the single and double I3C-la-
protein genes (SAKATAet al., 1989, 1992). beled acetates ~-(methyl-I~C) methionine and
Advances in physicochemical analysis permit- Na”S04. The CI5 bicyclic enedyine core is
ted the structural elucidation of the enedyine derived by head-to-tail condensation of seven
chromophore of neocarzinostatin from Strep- acetate units. The S-methyl groups of the tri-
tomyces carzinostaticus (EDO et al., 1988). sulfide, the thiosugar, and 0-methyl sugars
Several non-protein-bound representatives of are incorporated from S-adenosyl-methion-
the unique enedyine structure such as espera- ine. A minor congener of esperamicin A], es-
micins (MAGNUSand BENNETT, 1989), cali- peramicin P, was isolated from the fermenta-
cheamicin (LEE, 1992), and dynemicins tion broth of Actinomadura verrucospora
(KONISHet al., 1991) have been isolated from (GOLDBERGet al., 1989). It differs frorri es-
Actinomyces strains (Fig. 10a). peramicin Al in that it contains a methyl te-
trasulfide moiety instead of a methyl trisul-
fide.
Mode of action of the enedyines: The dynemicins (Micromonospora chersi-
nu) lack the apoprotein constituent of the en-
Neocarzinostatin (NCS) was the first natu- edyine chromophore. Dynemycin A contains
ral enedyine described which directly attacks the bicycl0[7.3.1]-1,5-diyn-3-eneand 1,4,6-tri-
the sugar moiety of DNA residues (GOLD- hydroxy-anthraquinone functions (Fig. 10a).
BERG, 1986; GOLDBERG et al., 1989). In Con- It exhibits a potent antibacterial and antitu-
trast to ionizing radiation or bleomycin which mor activity against a wide range of bacteria
damage deoxyribose through reactive oxygen and cells, respectively. Satellite compounds,
species, NCS produces DNA sugar lesions in such as L, M, N, 0, P, and Q have recently
a sequence-specific manner at closely corre- been isolated as shunt metabolites in fermen-
sponding sites of the two DNA strands. The tations of Micromonospora chersina and from
mutants of this strain (KONISHIet al., 1991; 3.6 Inhibitors of Mitosis and the
KAMEIet al., 1991; MIYOSHI-SAITOH et al.,
1991). Microtubular System
The calicheamicins (Fig. 10a) belong to a
family of seven glycosylated compounds from The early discovery of the antimitotic activ-
Micromonospora echinaspora ssp. calichensis ity of plant products such as colchicine
which demonstrate potent activity in vivo (Fig. 11) led to some clinical trials for cancer
against the murine tumors P388 and B16 treatment, but due to their high toxicity and
(LEE et al., 1989, 1992). Enedyines, such as relatively poor efficacy they have not found
esperimicin have previously been synthesized therapeutic application.
via several routes (LAM et al., 1993; LEE, The complex indole type vinca alkaloids
1992). (Catharanthus roseus) (GUNDAet al., 1994;
HEINSTEIN and CHANG,1994; KINGSTON,
1994), such as vincristine and its derivatives
3.5.8 Other Agents Forming (POITIERet al., 1994; JOEL, 1994), vinblas-
Active Oxygen Radicals in the tine, and the taxol type agents (from Taxus
brevifolia; ATTA-UR-RACHMAN et al., 1994;
Cells NOBLE, 1990) are plant-derived antitumor
drugs which prevent mitotic cell division
All normal cells have various defense sys- (Fig. 11). While the vinca alkaloids inhibit tu-
tems against active oxygen species. However, bulin polymerization, the latter (taxoids) ap-
several tumor cell types have lost a part of pear as particularly promising due to their op-
these defense mechanisms. Hence, substances posite effect of promoting the polymerization
generating active oxygen radicals such as of tubulin.
quinoid structures have been suggested to ex- In folk medicine, leaf extracts of the sub-
hibit selective cytotoxicity against such tumor tropical plant Catharanthus roseus were
cells (CURRAUet al., 1993). In screening for known to be effective in the treatment of dia-
such antitumor agents Sfrepfomycessp. KBlO betes. Attempts to isolate the active hypo-
was found to produce the new menoxymycins glycemic principle led, instead, to the discove-
A and B (related to medermycin) which ac- ry and isolation of cytotoxic vinblastine and
tively generate superoxide radicals in N18- vincristine. These complex indole alkaloids
RE-105 cell lysates (HAYAKAWA et al., 1994) are now used in the clinical treatment of a va-
(Fig. 8b). Numerous p-quinoid structures riety of cancers (NOBLE,1990) and in thera-
such as dioxamycin (SAWAet al., 1991), hy- peutic tumor cell synchronization (CAMPLE-
dramycin (HANADAet al., 1991b), the glyco- JOHN, 1980). A major problem in cancer
sylated saptomycins (ABE et al., 1993a), tetra- treatment by vincristine is the development of
cenomycins, and angucycline type aromatic resistance due to P glycoprotein-mediated
polycycles, such as saframycins (KANEDAet drug efflux (HILL, 1986). 500 kg of Catharan-
al., 1987) and atramycins (ABE et al., 1991) tus roseus (Vinca rosea) have to be extracted
possibly owe their cytotoxic and cytostatic ef- and the extract fractionated to obtain 1 g of
fects to the formation of active oxygen radi- vincristine. Hence, partial synthesis of vincris-
cals during aerobic metabolism (Fig. 8b). Sa- tine, vinblastine and vindesin is done using
framycin A was also reported to bind cova- vindolin. Taxol (paclitaxel) (Fig. 11) was dis-
lently to duplex DNA (KANEDAet al., 1987). covered in 1964 during a large screening for
In addition, various polycyclic aromatic com- antitumor drugs at the National Cancer Insti-
pounds with quinone structures or an epoxy tute of the USA. It is an exciting new anti-
side chain display cytotoxic activities. This cancer drug, exhibiting clinical activity in the
feature was also ascribed to the formation of treatment of ovary and breast cancer (KING-
free radicals or other reactive structures. Rep- STON, 1994; MARTYet al., 1994; ROTHEN-
resentatives of this kind of agents are the sa- BERG,1993). Taxol was first isolated from the
purimycins (Streptomyces sp.) (UOSAKIet al., pacific yew tree (Taxus brevifolia). To over-
1991). come the problems of raw material supply,
+
OH
OMe
H OMe
M on
a OMe
Callcheamlcln a1Br
C-1027
Chromophore
NH2
SSSMe
"YL+d
OMe y
p OMe
Esperamicin A,
OH o on
R = CH(Me)2 M a d o Dynemicin A
Al
b
R=Et MeO
AlC R=Me .&me
a CH2
Fig. 10. Structures of enedyine drugs.

a Chromophore structures of esperamicin, calicheamicin, dynemicin, and C-1027,
b enrichment pattern of esperamicin A1 supplemented with 13C-labeledsingle and double-labeled acetate.
13C labeling is indicated by 0, W, and WO.
-
a) CH3COOH
Ri OH
CH,O,CNH rz,
0
CM,S,
I
b) C!H3COOH
R1
OH
OH
CH,O,CNH 0 5
Fig. lob
paclitaxel and docetaxel (taxotere) (Fig. 11) brevifoliu was reported to form low amounts
were developed (POITIER et al., 1994; MARTY of taxol, but so far a biotechnical procedure
et al., 1994). In particular, the semisynthesis for the fermentative production of taxol has
from the related structures of the baccatins still not been developed (STIERLEet al.,
appear to be a feasible way of taxane produc- 1994). Taxoids are particularly promising an-
tion (HEINSTEINand CHANG,1994; KING- titumor agents due to their novel cancerostat-
STON, 1994). Baccatins are produced by the ic mechanism of action (ROTHENBERG,
leaves of the European yew and can be chem- 1993). These compounds interfere with essen-
ically transformed to taxotere (docetaxel) tial cellular processes such as mitosis, cellular
(POITIERet al., 1994; JOEL,1994). Recently, motility, transport, and maintenance of cellu-
an endophytic hyphomyces fungus of Taxus lar structures. The mechanism of action of
"q
Me HCOCH,
Me0
OMe
. 0
Colchicin
R1 R2 R3
Vinblaslin CH3 OCH3 C W H 3

Wncrlstln CHO OCH3 COCH3
CI CH,
Vindesin CH3 NH2 H
CH, OCH,
Maytansine R= COCH(CH3)N(CH3)COCH3
Ansamitocin P-3 R= COCH(CH3)*
Paclitaxel (TaxolR)
Wo
Fig. 11. Inhibitors

of the microtubu-
lar system.
taxanes is to promote the formaton of extent against proliferating cells than non-proli-
tremely stable non-functional microtubular ferating cells in v i m (ROTHENBERG, 1993).
aggregates. The cells are thus prevented from Myelosuppression is a dose-limiting side ef-
entering the G, or M phase of the cell cycle. fect. Current research is centered now around
Both paclitaxel and docetaxel are more PO- improving drug extraction, the semisynthetic
production of the parent taxane structures 1992). To overcome myelosuppression as one

(PAQUE-ITE, 1993; GUERITTE-VOEGELEIN et of the limiting factors in cancer chemotherapy
al., 1994), the enhancement of water solubili- by anthracyclines, a search was initiated for
ty, and the identification of taxane analogs immunomodulatory stimulators of leukocyte
which are active against resistant cancer cells and platelet formation (KAWATSUet al.,
(ROTHENBERG, 1993; VYAS, 1993). Due to 1993,1994). As a result of this screening, con-
the impressive clinical activity of paclitaxel agenin was detected in cultures of Strepto-
and docetaxel, the taxanes will be subject to myces roseosporus. It binds exclusively to T
preclinical and clinical development in the fu- cells activated by concanavalin A or cytokines
ture (GUERI-ITE-VOEGELEIN et al., 1994). and enhances both T cell proliferation and
Moreover, structure-activity relationships lymphokine formation. Due to the induction
(SAR) of drug-tubulin interactions are under of cytokines, the production of leukocytes
investigation (GUERI-ITE-VOEGELEIN et al., and platelets is subsequently stimulated.
1994). Chemical modifications such as C-2
deoxygenation result in a total loss of tubulin
function. A relatively small number of micro- 3.8 Cytotoxic Compounds with
bial drugs inhibits mitosis in tumor cells. Ex- Poorly Characterized Mode of
amples are rhizoxin (Rhizopus chinensis), a
macrolide antibiotic and tubulin inhibitor of Activity
Aspergillus nidulans (KATOet al., 1991), an-
samitocin from Acfinomyces sp. (similar There are continuously references in the
to maytansine from plant Maytanus sp.) literature to new cytotoxic structures (see,
(Fig. ll), and curacin A from the marine cya- e.g., reports on structures such as russuphe-
nobacterium Lyngbya mujuscula (PIRRUNG lins (TAKAHASHI, 1993), heptelidic acid (KI-
and NAUHANS,1994). Rhizoxin exhibits po- WASHIMA et al., 1994), and stubomycin
tent antimitotic activity by binding to Ptubu- (KOMIYAMA et al., 1983). Cytotoxicity ap-
lin, and is active against most eukaryotic cells. pears to be a very frequent property of new
It shares the same binding site with maytan- natural products. In a personal literature
sine and ansamitocin P-3 (Fig. 11) on porcine, compilation, out of 4,000 microbial drugs, 600
brain, and fungal tubulin which is different were indicated to suppress cell growth. For
from that of colchicine and vinblastine (IWAS- the majority of these agents the mechanism of
AKI, 1989). action has not been investigated. In the past,
much interest has been focussed on new me-
tabolites which might inhibit tumor cell lines
3.7 Reduction of the Side Effects that are intrinsically resistant to the known
set of anticancer drugs or that have acquired
of Highly Toxic Anticancer Agents multi-drug resistance. In the course of screen-
ing using adriamycin-resistant HL-60 cells,
Highly effective cytotoxic anticancer homooligomycins A and B (Sfrepfomycesbot-
agents, such as the anthracyclines, vincristine, tropensis) were detected as a result of their
and bleomycin, usually have severe side ef- particular effect on colon-26 carcinoma (YA-
fects such as cardiotoxicity, myelosuppression MAZAKI et al., 1992).
(FURKet al., 1989), release of interleukin 2, A further focus of interest have been the
and others which are dosage limiting. Prein- numerous neoplasm inhibitors which are se-
duction of metallothein synthesis, e.g., by bis- lectively toxic against a special type of tumor
muth salt was reported to exert a protective cell. For instance, the homooligomycins A
effect against various anticancer agents (e.g., and B (Sfreptomyces boffropensis) were de-
bleomycin, cis-diammine-dichloro-platinum) tected due to their particular effect against
(NAGAMURA et al., 1993). colon-26 carcinoma (MAGAEet al., 1993). In
Iron chelators and antioxidants have also a screening program for immunomodulators,
been studied with regard to brain protection the melastins (Sfrepfomycessp.) (MAGAEet
against neoplasm inhibitors (GU~TERIDGE, al., 1993) were similarly found to selectively
inhibit the growth of leukemia cells and the 4.1 Potentiators of Cytotoxic
lipopolysaccharide-inducedblastogenesis of T
cells. Antitumor Agents
DNA repair mechanisms serve as a useful
target for modulating the cytotoxic and che-
motherapeutic effects of those agents. Their
4 Non-Classical mechanism of action causes the induction of
Approaches to Antitumor DNA damage. Poly(ADP-ribose) polymerase
responds to DNA breaks by cleaving the sub-
Drugs strate NAD and using the resultant ADP-
+
ribose moieties to synthesize homopolymers

Much interest has been centered on new of ADP-ribose. Inhibitors of this enzyme such
anticancer agents which exert their antitumor as benzamide derivatives and benadrostin
activity via a hitherto unknown interaction (Fig. 12) (YOSHIDAet al., 1988) prevent
with cellular regulation and cellular transduc- DNA repair and potentiate the tumoricidal
tion pathways. Among these new agents, inhib- effects of DNA strand-breaking agents such
itors of tumor cell resistance to anticancer as bleomycin (BERGERet al., 1987; YOSHIDA
agents, tumor metastasis, and angiogenesis of et al., 1988; GAAL and PEARSON,1986; Su-
tumors appear promising. ZUKI et al., 1990). The simple compound, 2-
Otherwise, the discovery of the oncogenes methyl-4[3H]-quinazoline, was discovered in
and their role in malignant cell growth has Bacillus cereus cultures and found to be an in-
opened new horizons for target-directed hibitor of poly(ADP-ribose) synthetase from
screening of new anticancer agents. This de- calf thymus (YOSHIDAet al., 1988). In addi-
velopment was initated by the exploration of tion, calcium channel antagonists such as
particular oncogene-encoded functions such verapamil and fendiline are known to poten-
as those of erb, sis, and ras in tumor cells. On- tiate the activity and toxicity of mitomycin C
cogenes are modified forms of normal genes in mammalian and bacterial cells (SCHEIDet
(proto-oncogenes) which have been altered al., 1991). The mechanism of this effect has
either by chemical carcinogenesis or retrovi- not yet been explored in detail; however, it
ruses in such a manner that they escape the seems reasonable to suggest that P glycopro-
normal pattern of control (FERDINAND, tein-mediated efflux of the drug is prevented.
1989). These alterations may result in the Similarly, a series of other agents were re-
overexpression of the genes, the loss of down- ported to increase the cytotoxic effect of neo-
regulation of the genes, and the formation of plasm inhibitors. A major obstacle in cancer
altered proteins. Proteins encoded by these chemotherapy is the appearance of resistant
genes, such as the epidermal growth factor re- tumor cells which are non-treatable by cyto-
ceptor (EGF receptor) and membrane-asso- static drugs such as vincristine and doxorubi-
ciated Rus-proteins are abnormal derivatives cin. Several reasons for this are known (HILL,
of the normal proteins involved in cellular 1986; BIEDLERet al., 1993). Overexpression
signaling and cell cycle regulation. The pres- of metallothein in cancer cell lines, e.g., con-
ently available information on cellular regula- fers resistance to anticancer drugs such as pla-
tion appears to be the “tip of the iceberg”. tinum complexes (KELLEYet al., 1988). De-
Recent additions to our knowledge were pro- toxification of cytotoxic drugs may also occur
vided by the discovery of the cyclins, the tu- by glucuronidation (BURCHELLet al., 1991).
mor suppressor genes and their proteins. Other cancer cell lines become resistant to a
These findings will doubtlessly promote the series of anticancer drugs such as taxol, col-
discovery of new agents which interfere with chicine, doxorubicin, actinomycin, and vin-
these targets and module their activity. The cristine (HILL, 1986). This so-called multi-
following survey summarizes the results of re- drug resistance (MDR) is frequently asso-
cent screening approaches to these non-classi- ciated with increased drug efflux and de-
cal anticancer drugs. creased accumulation within the cells. The
a
b
?!
F OMe
Me
Me
Me
On M ~ O on
o on
Me Me Me
FD895 Smbomycin
HnbXlUNblglIlA
cocn. n
Rubiginon A,
Me
672 I4 Biotechnical Drugs as Antitumor Agents
phenomenon is due to membrane proteins, version of the multi-drug resistance of cancer

such as P 170 glycoprotein, which is capable cells) (WAKUSAWA et al., 1993), 5-N-actylar-
of conveying drugs out of the cells. This gly- deemin (Fig. 12) (Aspergillus fischeri, rever-
coprotein was detected in the mouse macro- sion of multi-drug resistance in tumor cells)
phage-like cell line J 774.2 and in other cells, (HOCHLOWSKI et al., 1993) and BE-12406 A
and its role as mediator of the active efflux of and B (Sfreptomyces sp., inhibition of vincris-
cytotoxic drugs was studied (GUNICKEand tine- and doxorubicin-resistant P388 murine
HOFMANN, 1992; BORREL et al., 1994; leukemia cells) (KOGIRIet al., 1991). A par-
REICHLEet al., 1991). Mobile ionophoric car- ticular feature of the potentation of antitumor
riers, such as valinomycin, lasalocid, monen- drug activity is the potentiating effect of inhi-
sin, calcimycin, and others inhibit the efflux bitors of bleomycin hydrolases (leupeptin,
of anthracyclines, whereas channel-forming pepstatin) on bleomycin activity (NISHIMURA
ionophores (e.g., gramicidin A) do not. Cy- et al., 1987). Some tumor cells resistant to in-
closporine, a calcium chelator, also interfered hibitors of topoisomerase I1 display a type of
with /3-glycoprotein-mediated efflux of these MDR that differs from P glycoprotein-asso-
drugs (BORRELet al., 1994; DIETEL,1991). ciated MDR and is restricted to drugs that in-
Directed screening for drugs which inhibit hibit topoisomerase I1 by stabilizing cleavable
the growth of multiresistant tumor cells DNA-protein complexes (BECK et al.,
(MDR) or which could restore and potentiate 1993).
the antitumor activity of known agents thus
appeared to yield promising substances. Ex-
amples of these agents (Fig. 12) are the hato- 4.2 Inhibitors of Glutathione S
marubigins (Streptomyces sp., inhibitors of Transferase as Enhancers of
multi-drug resistant tumor cell lines) (HAYA-
KAWA et al., 1991), BE 137932~ Antitumor Activity of Drugs
(Streptornyces sp., inhibits the growth of doxo-
rubicin-resistant cancer cells) (KEFRI et al., Glutathione conjugate formation may con-
1991), the piperaficins (Streptoverticillium as- stitute an important drug detoxification and
pergilloides, potentiate vincristine activity bioactivation mechanism for several classes of
against resistant murine P388 cell lines) (KA- mutagenic and carcinogenic compounds
MEI et al., 1990), bestatin (Streptoverticillium (SATO,1989). For instance, the metabolism of
olivoreticuli, an immunostimulator, increases nephrotoxic and nephrocarcinogenic haloal-
the in vivo vincristine cytotoxicity against colo- kenes involves glutathione S conjugate for-
rectal K562 carcinoma cells) (UMEZAWA, mation. Glutathione was suggested also to
1989), sekothrixid (an antitumor drug) (KIM play a role in the toxicity or action of bleomy-
et al., 1991), rubiginon (Sfrepfomycesgriseo- cin, cyclophosphamide, and neocarcinostatin
rubigenosus, restores colchicine sensitvity to (ANDERS,1991; SATO, 1989). An increased
resistant tumor cells) (OKAet al., 1990,1991), glutathione S transferase activity may create
resorthiomycin (potentiates the antitumor ac- resistance to some anticancer drugs such as
tivity of vincristine) (TAKARAet al., 1990), alkylating agents, doxorubicin, and cis-diam-
laidlomycin (a polyether antibiotic from mine-dichloro-platinum. Consequently, inhib-
Streptomyces sp., potentiates anticancer drug itors of glutathione S transferase may im-
activity against MDR carcinoma KB-4 cells) prove the activity of antitumor drugs. More-
(KAWADAet al., 1992b), bisucaberin (Altero- over, they could also act as antiinflammatory
monus haloplunctis, sensitizes tumor cells to and antiallergic agents due to their effect on
cytolysis mediated by macrophages) (TAKA- the prostaglandin and eicosanoid metabolism.
HASHI et al., 1987a), kazusamycin (see In a worldwide screening approach several
Fig. 18) (Sfrepfomycessp.) (YOSHIDAet al., new inhibitors were discovered, such as ben-
1987), lactoquinomycin (Sfrepfomyces funa- astatins C and D (Streptomyces sp.) (AOYA-
shiriensis, active against resistant L 51178~ M A et al., 1993a), cysfluoretins (Streptomyces
lymphoblastoma) (TANAKAet al., 1989), and sp.) (AOYAMAet al., 1993b), TA-3037 A
semisynthetic derivatives of staurosporin (re- (KOMAGATA et al., 1992a), rishirilide B (KO-
magata et al., 1992b), and bequinostatins C nojirimycins and their chemically derived der-
and D (YAMAZAKI et al., 1993) (Fig. 13). ivatives have recently been reported to pre-
vent metastasis by the inhibition of p-glucu-
ronidase, heparanase, and P-D-mannosidase
4.3 Antimetastasis Drugs and of tumor cells (SATOHet al., 1996 TSURUO-
Inhibitors of Angiogenesis KA et al., 1996; KAWASEet al., 1996; TATSU-
DA et al., 1996).
Prevention of metastasis of cancer cells and Using the chicken embryo chorioallantoic
angiogenesis is a major aim of cancer chemo- membrane assay system, a series of angiogen-
therapy. Angiogenesis is a tissue differentia- esis inhibitors were disclosed, such as herbi-
tion process which connects a growing tumor mycin A (Streptomyces sp.) (YAMASHITAet
with the blood vessels. Recently, phospholip- al., 1989; SAKAI et al., 1989), irsogladin
ase D was proposed as a target for tumor in- (SATO et al., 1993), erbstatin (Streptomyces
vasion inhibitors. Drugs, such as Sch 49210, sp.) (OIKAWAet al., 1993), TAN-1120 (an an-
Sch 53514 and Sch 53517 from the fungus thracycline, Streptomyces sp.) (NOZAKIet al.,
Nattrasia mangifera displayed potent activity 1993), TAN-1323 (Streptomyces sp. S-45628)
in the antitumor invasion chamber assay (MUROI,1990), analogs of siastatin B, WF-
(CHU, 1994). Among the rnetalloproteinases 16775 A,, A2 (derivatives of pyridazine from
involved in tumor invasion, angiogenesis, and Chaetosbolisia erysiophoides) (OTSUKAet al.,
rheumatoid arthritis, the type “N” enzymes 1992), 15-deoxyspergualin (NISHIKAWA et al.,
may play a particular role in the degradation 1991a,b), and staurosporin (OIKAWAet al.,
of basement membranes. Searching for inhib- 1992). These drugs are expected to prevent
itors of this enzyme, such as the matlystatins neovascularization and thus seem to be prom-
from Actinomadura atramentaria, has been ising not only in the prevention of metastasis
proposed as a promising approach towards but also in the treatment of retinopathy and
new antitumor drugs (HARUYAMA et al., rheumatoid arthritis (OTSUKAet al., 1992).
1994). Antimetastasis activity was reported
for U-77863 (TROLLet al., 1987) and U-77864
(methylphenyl-propenyl-carboxamides)from 4.4 Antitumor Effects of
Streptomyces griseoluteus (HARPER and Immunomodulators
WELCH, 1992). These compounds were
shown to inhibit the growth of tumor xeno- In the healthy human, the immune system
graft models (Fig. 14). Nitrogen-containing recognizes both microbes and malignant cells
pseudo-sugars such as siastatin B, nagstatin, and destroys them. The outbreak of cancer
I
6 H 0 OH
Cysfluoretin
Rishirilide
Benadrostin
COOH
H H
TDD OH0 OH
Bequinostatin
Fig. 13. Inhibitors of glutathione S transferase.

0 on
NHCHO
6H
Erbstatin U-77863
Me Me
TAN-I1
NMC
,+OH
ii
Matlystatin lrsogladin WF-I 6775 A1
OH Me
I I
Me -ClKONH-CH-CO-
D
NH- CHCOOH
D L I
(CH,),CONHCHCONHCH,COOH NH, NH,
FK-606
I ' Forphenicin Forphenicinol
Spemidin (=OH) Deoxyspetmidin(R=H) Conagenin
Fig. 14. Selected structures of antimetastatic inhibitors and anti-angiogenetic and immunomodulatory
drugs.
may be promoted by insufficient immunologi- glucans secreted into the medium, such as shi-
cal defense and poor recognition of abnormal zophyllan and lentinan, belong to a group of
cells. Hence, immunostimulating and immu- immunostimulatory compounds which exhibit
noregulatory agents have been considered as antitumor activity. They are derived from
potential anticancer therapeutic agents. The higher fungi and mushrooms. Their antitumor
immunostimulatory and antineoplastic prop- activity has been attributed to the stimulatory
erties of bacterial wall preparations, especial- effect on the immune sysem (LAATSCH,
ly from Mycobacteriurn spp. were discovered 1992). Subsequently, the chemical structures
already in the 1970s. High molecular weight of other naturally occurring immunomodula-
HerbimycinA Herbimycin B
OH 0 Me
&NH2
\ N OH" Me
M.
Fig. 14 U-7786 W-16775
tors were elucidated. These compounds may agents such as spergualin (Bacillus laterospo-
provide the pharmaceutical scientist with an rus) and its semisynthetic 15-deoxy deriva-
armamentarium that can be used to mount a tive, antitumor activities were demonstrated,
rational treatment of immunopathologies in- e.g., against the murine leukemia P 388 (NI-
cluding cancer. For several immunomodula- SHIKAWA et al., 1991a, b). Retardation of
tors of microbial origin, protective effects malignant cell growth was ascribed to the cy-
against experimental tumors have been totoxic and cytostatic effects of these drugs,
established in mice (KLEGEMANN,1993) but the mode of action of these and other im-
(Fig. 14). munosuppressants (see, e.g., cyclosporine A
Best atin (Strepto verticillium olivoreticuli) and FK-506) has not been explored in detail.
which was discovered by H. UMEZAWA'S
group in Japan (ABE et al., 1985) enhances
the activity of antitumor drugs and is used in
clinical applications. The drug was initially 4.5 Inhibitors of the Cellular
found to be an inhibitor of leucine aminopep- Mitogenic Signal Transduction
tidase. Immunostimulatory and antitumor ef-
fects have been reported also for some other Pathway
microbial protease inhibitors (BILLINGS,
1993). The compound FK-156 (Streptomyces Fig. 15 shows the role of protein kinases,
olivaceogriseus) (IZUMIet al., 1983) and con- phosphoprotein phosphatases (NEER and
agenin (Streptomyces roseosporus) (YAMA- CHAPHAM,1988), phospholipases, and farne-
SHITA et al., 1993; KAWATSU et al., 1993) are syltransferases in the transduction of an ex-
additional examples of microbial metabolites tracellular signal, such as a growth hormone,
which may exert anticancer activity due to into an intracellular response of DNA repli-
their stimulatory effect on cytokine produc- cation and mitogenesis. In oncogene-trans-
tion by T cells and macrophage activity. For- formed cells these enzymes are often involved
phenicin (Streptomyces sp.) and its biotrans- in neoplasmic cell growth. Much effort in the
formation product forphenicinol inhibit search for new anticancer agents is now fo-
chicken alkaline phosphatase. Forphenicinol cussed on specifically acting inhibitors of the
induces yinterferon production in mice cell cycle such as acetophthalidin from a ma-
which were sensitized by the BCG vaccine, rine Penicillium strain which arrests the mam-
and enhances macrophage activity. It also dis- malian cell cycle in the G2/M phase (CUI et
plays an antitumor effect on MetA fibrosar- al., 1996a, b).
coma and adenocarcinoma (OKURAet al., So far, more than 40 distinguishable onco-
1986). For a series of immunosuppressing genes have been identified. They can be clas-
sified into four main groups according to their 4.5.1 Inhibitors of Protein Kinases
function:
and Protein Phosphatases
(1) oncogenes encoding tyrosine protein kin-
ases such as src (similar to insulin recep- Protein kinases are ubiquitous regulatory
tor and catalytic chain of mammalian proteins in microbial, plant, and animal cells
CAMP-dependent protein kinase from which exert their activity by the ATP-depend-
Rous sarcoma virus), erbB, fgr, and fes ent phosphorylation of serine, threonine, and
(see also Fig. l), tyrosine residues (Fig. 16). The phosphoryl-
(2) oncogenes encoding proteins involved in ated proteins display an altered function (ac-
the metabolic regulation of GTP-binding tivated or inactivated) (BASU and LAZO,
proteins as essential parts of G protein-1994; LEVITZKI,1994). Phosphorylation and
dependent membrane receptors (K-Ras, dephosphorylation of proteins can result in
H-Ras, N-Ras), the intracellular amplification of a mitotic sig-
(3) oncogenes encoding proteins which act at nal, generated by interaction of a growth hor-
the level of gene regulation as transcrip-
mone with a membrane receptor. The protein
tion factors such as Myc and Myb, and kinases share at least some common features
(4) oncogenes encoding growth factor-like in their secondary and tertiary structures
(TAYLOR,1989; TOWBRIDGE,
proteins such as Sis (similar to platelet- 1991). Myosin
derived growth factor). light chain kinase, a cytoplasmic enzyme, has
a Ca2+kalmodulin-binding domain carboxy
To escape the normal pattern of control terminal to the catalytic core, and binding of
“proto-oncogenes” have to be transformed ligands activates the kinase. Membrane-asso-
into oncogenes either by structural altera- ciated protein kinase C is activated by Ca2+,
tions, due to chemical carcinogenesis (GUEN- diacylglycerol or phorbol ester type com-
GRICH,1988) or by integration into retrovi- pounds, and phospholipids, and the recogni-
ruses and reintegration into the host genome tion sites for these ligands lie amino terminal
(HART and TURTURRO,1988 BERTRAM, to the catalytic core. Tyrosine-protein kinase
1990 WEINBERG,1985). activity was first detected as a characteristic
Fig. 15. Key steps in the

cellular mitotic signal
transduction pathway.
of the transforming protein from Rous sarco- The phorbol esters and similar structures
ma virus, pp60’”“. Removal of a phosphoryl- (Fig. 16) activate protein kinase C. This effect
ation site of this protein converts the “proto- has been visualized as bleb formation on the
oncoprotein” into a transforming protein. cell surface of K562 chronic myeloic leukemia
Following myristylation (farnesylation) cells which is induced in the presence of phor-
pp60’-”“ moves to the plasma membrane. Re- bol dibutyrate. Bleb induction can be pre-
ceptors of growth factors, such as the epider- vented by specific inhibitors of protein kinase
mal growth factor (EGF) receptor, span the C and thus a practicable high-throughput
membrane via a single membrane-spanning screening assay was developed (OSADAet al.,
element. Binding of EGF to the outer domain 1988). The same bleb formation assay has
activates the cytoplasmic tyrosine protein kin- also been used to search for inhibitors of pro-
ase activity. Another prototype of a tyrosine- tein phosphatases. These enzymes provoke
protein kinase transmembrane protein is the same type of stimulaton of bleb formation
CD45 (T200or leukocyte common antigen) which is a characteristic of the phorbol esters
(TROWBRIDGE, 1991). (MAGAEet al., 1988, 1992; BLUMBERG et al.,
Another type of protein kinase is activated 1989). On the other hand, differences in the
by cyclic 3,5-adenosine monophosphate effect on tetrazolium blue reduction of HL-60
(CAMP). The activating ligand (CAMP)binds cells of phorbol esters and tautomycin sug-
to a distinct regulatory subunit thereby induc- gested that the latter has a different mode of
ing conformational changes that provoke dis- action (MAGAEet al., 1992). For the sake of
sociation of the holoenzyme (TAYLOR, completeness, the bryostatins (from marine
1989). Bryozoa) should be mentioned since they are
Reversible phosphorylation/dephosphory- similar tumor promotors (Fig. 16). They are
lation of proteins is involved in many cellular antitumor and antileukemic compounds iso-
activities (KAHNand GRAF,1986). For exam- lated from marine animal organisms as a type
ple, the M phase promoting factor (MPF) in of cyclic polyether which also mimics the
mammalian cells is composed of a catalytical structure of 1,2-diacylglycerol (RAMSDELL
~ 3 4 protein
~ ~ ~ which
’ has protein kinase ac- and PETTIT, 1986; WOLF and BAGGIOLINI,
tivity and a regulatory protein (cyclin B). The 1988; TAKAHASHI et al., 1987b). Similarly,
activity of MPF is regulated by reversible novel inhibitor structures of protein kinase C
phosphorylation/dephosphorylation.Recent- and phosphoprotein phosphatases were dis-
ly, tautomycin (Streptomyces sp.) was found covered, such as staurosporin, UCN-1, UCN-
to inhibit dephosphorylation of MPF thereby 2 (TAKASHIet al., 1989), tautomycin (Strepto-
preventing mitosis (MAGAE et al., 1988, myces spiroverticillatus) (MAGAE et al.,
1992). 1992), calyculin, microcystin (Fig. 16), and
In the search for inhibitors of protein kin- okadaic acid (from a marine organism) (Su-
ase C, the morphogenic effect of phorbol es- GANUMA et al., 1988). Calyculin A affects
ters on mammalian cells provided a useful phosphoprotein phosphatase 1 (PP1) 30 to
screening feature (TAKAHASHI et al., 1987, 250fold more than okadaic acid, and also in-
1989). Phorbol esters and indolactams (teleo- hibits protein phosphatase 2A (PP2A). Tau-
cidin and blastmycetin from Streptoverticilium tomycin inhibits several protein phospha-
sp.) (HAGIWARA et al., 1988) are known as tases, even those of smooth muscles. No activ-
tumor promoters. They stimulate growth of ity was found on myosin light chain kinase or
tumors but are not carcinogenic. The discove- protein kinase C. Okadaic acid, a polyether-
ry of tumor promoters from Euphorbiaceae, like shellfish toxin, inhibits serine-threonine
such as phorbol esters contributed to a better specific protein phosphatases, particularly
understanding of the cellular mechanism of protein phosphatase 2A. Moreover, isopali-
growth regulation. In addition to tumor pro- murin, a mild protein phosphatase inhibitor,
motion, other processes such as inflamma- has recently been isolated from a sponge
tion, mitogenesis of lymphocytes, platelet ac- (MURRAYet al., 1993) (Fig. 16).
tivation, etc. were affected by structural ana- Presently, much interest is focussed on
logs of the second messenger diacylglycerol. phosphotyrosyl protein phosphatases which
0
II
Phorbol Teieocidin 84
12,13dibutyraI
BRYOSTATIN 1 & I 2
Cocn, QCQ
Dephostatin
Epiderstatin isoflavonoids
on
Me
H
O
H
Q 0
0 lsopalimurin
BE33372M
4 Non -Classical Approaches to Antitumor Drugs 679
4 Fig. 16. Selected structures of inhibitors of protein-

product) (NAKANOet al., 1987; CAI et al.,
kinase C, tyrosine protein kinase, and protein phos-
1996).
phatases. The antitumor antibiotic calphostin (UCN-
1028) (Fig. 16) consists of five components
(A, B, C, D, I) (IIDAet’al., 1989). It is pro-
are a particular family of transmembrane en- duced by the fungus Cladosporium cladospor-
zymes (LAN et al., 1989; KRUEGERet al., oides and strongly inhibits protein kinase C.
1990; FISCHER et al., 1991). Another inhibitory structure is balanol (aze-
pinostatin) (Fig. 16) isolated from Verficillium
balanoides and Fusarium merismoides (BOR-
4.5.1.1 Inhibitors of Protein 0s et al., 1994; OHSHIMA et al., 1994). In ad-
dition nucleoside antibiotics, such as sangiva-
Kinase C mycin (Fig. 16), and membrane-active com-
pounds, such as polymyxin B, were reported
Members of the indolocarbazole family of to inhibit protein kinases including protein
antibiotics such as staurosporin (Sfreptomyces kinase C (LOOMISand BELL, 1988; ALL-
sp.), UCN-01, UCN-02 (Sfrepfomyces sp.) GAIER et al., 1986).
(AKINAGAet al., 1993), K-252a (Nocardiop-
sis sp.) (NAKANISHI et al., 1986), RK-286C
(OSADA et al., 1990), KT 6006 and 4.5.1.2 Inhibitors of Other Protein
CGP41251, are potent inhibitors of protein Kinases
kinase C (Fig. 16). These unique structures
are produced by actinomycetes; however, The pamamycin-type agents MS-282a and
staurosporin-like compounds have recently MS-282b, are macrodiolide-type inhibitors of
been discovered in extracts from marine tuni- calmodulin-activated myosin light chain kin-
cates. Derivatives of staurosporin such as ase from Strepfomyces fauricus ATCC 27470
MLR-52, display immunosuppressive activity. (NAKANISHI, 1994). Other protein kinases
This effect is comparable to that of other im- are not inhibited by these agents. The interac-
munosuppressors, such as cyclosporine A and tion of MS-282 compounds with the enzyme
FK-506, which affect protein phosphoryla- occurs at the calmodulin-binding site. Calmo-
tions and phosphoprotein dephosphoryla- dulin-dependent nucleotide phosphodiester-
tions (MCALPINEet al., 1994). ase is also affected. The compound Sch45752
Moreover, indolocarbazoles display re- (Fig. 16) from the fungus SCF-125 is an inhib-
markable antimicrobial activities against bac- itor of various protein kinases (IIDA et al.,
teria and fungi including Candida albicans 1989); however, it also inhibits calmodulin-
and Botryfis cinerea, but no correlation was sensitive cyclic nucleotide phosphodiester-
found between protein kinase inhibitory po- ase.
tencies and inhibition of Strepfomyces sporu-
lation or growth inhibition (SANCELME et al.,
1994). The indolocarbazole compounds dis- 4.5.1.3 Inhibitors of Tyrosine
play antitumor activity against human and Protein Kinases
murine tumor cell lines both in vitro and in
vivo (AKINAGAet al., 1993; TAKAHASHI et Erbstatin (Fig. 14), which was isolated from
al., 1987b; WOLF et al., 1988). This type of the culture broth of Strepfomyces sp. MH 435-
structure (SANCELMEet al., 1994) inhibits UF3, is a potent inhibitor of the EGF asso-
either protein kinase C (see above) or topo- ciated protein tyrosine kinase (IMOTOet al.,
isomerase I (rebeccamycin, AT 2433, and der- 1987; SONODAet al., 1989). It specifically in-
ivatives) (OSADAet al., 1990 KANEKOet al., hibits the autophosphorylation of EGF recep-
1990). AT 2433 has an inhibitory effect on tor but does not inhibit either cyclic AMP-de-
protein kinase C. Staurosporin and its analogs pendent protein kinases or protein kinase C
also inhibit the activity of the Rous sarcoma (IMOTOet al., 1987; UMEZAWAet al., 1986).
virus-transformed protein p60 (scr oncogene Also, synthetic peptide sequences were found
to be good substrates for the EGF tyrosine 4.5.2 Inhibitors Affecting the
kinase and have been used in kinetic studies.
According to these investigations, erbstatin Metabolism of Phosphoinositols
competes with the peptide substrate but not
with ATP (UMEZAWAet al., 1986). 4',7,8-tri- Phosphatidylinositol turnover appears to
hydroxyisoflavone, 3 ' ,4 ',7-trihydroxyisofla- be correlated with cell transformation by
vone, 8-chloro-3',4 ',5,7-tetrahydroxyisofla- some types of oncogenes (rus) and also with
vone, and orobol were isolated from the cul- cellular responses to growth factors (EGF,
ture broth of Sfrepfomyces sp. OH-1049 are PDGF) (FLEISCHMAN et al., 1986; JACHOWS-
isoflavones (Fig. 17) which also inhibit EGF KI et al., 1986; BERRIDGEand IRVINE, 1984,
receptor tyrosine-protein kinase. This effect 1989). The signal generation induced by the
coincided with their increasing effect on the enzymatic breakdown of phosphatidylinosi-
life span of tumor-bearing mice (S180, P388) tol-4,5-diphosphate involves phospholipase C.
(KOMIYAMA et al., 1989; OGAWARAet al., This key enzyme is capable of forming two
1989a). This compounds do compete with second messenger molecules, inositol-1,4,5-
ATP. The interaction of flavonoids with triphosphate and 1,2-diacylglycerol. Inositol-
mammalian protein kinases was later found 1,4,5-triphosphate mobilizes the intracellular
to be non-specific (END, 1987). Although calcium pool and 1,2-diacylglycerol activates
they occur in microbial cultures, isoflavone protein kinase C (Fig. 15). Consequently,
glycosides such as genistein and daidzin ap- phospholipase C, the phosphatidylinositol
pear to stem from the plant-derived nutrients. turnover, the phosphatidylinositol receptor,
Microbial biotransformation results in the protein kinase C (see above), and diacyglyc-
formation of substituted or even glycosylated erol kinase are potential targets in the search
products (ANGANWUTAKU et al., 1992). of cellular growth regulators. Q 12713 (Acfi-
More recently discovered structures with nomuduru sp.) (OGAWARA et al., 1992), (his-
inhibitory activity against EGF receptor me- pidospermidins (Chuefomiu sp.) (YANAGISA-
diated protein-tyrosine kinase are, e.g., de- WA, 1994a, b), and caloporoside (Culoporus
phostatin (Sfreptomyces sp.) (KAKEYAet al., dichrous) (WEBERet al., 1994; TATSUDAand
1993), BE-13793C (TANAKAet al., 1992), YASUDA,1996) which are inhibitors of phos-
BE-23372M (Rhizoctoniu sp., TANAKA et al., pholipase C have recently been discovered
1994), tyrphostins (synthetic) (Fig. 16), and p- (Fig. 17). The search for inhibitors of the ino-
quinoid structures like paeciloquinones A-F sitol-triphosphate turnover provided a series
(FREDENHAGEN et al., 1995). Epiderstatin (a of different chemical structures, such as psi-
substituted 2,6-piperidine-dione) (Fig. 16) tectorigen (an isoflavone from Actinomyces
was isolated from Sfreptomyces pulveruceus cultures and plants, Fig. 17) (IMOTOet al.,
ssp. epidersfugenes (SONODAet al., 1989). 1988, 1991), inostamycin A (a polyether from
The compound inhibits incorporation of a Sfreptomyces strain) (IMOTOet al., 1990),
[3H]thymidine into quiescent cells stimulated echiguanins (TAKEUCHI,1992), and piericid-
by EGF and also reverts the morphology of ins B1, B5 and their "-oxides (NISHIOKAet
"src-transformed cells to normal. al., 1991). Recently, automated screens have
Contact inhibition of growth by neighbor- been developed for inhibitors of myo-inositol
ing cells is an essential characteristic of nor- monophosphatase such as ATCC 20928 A-C
mal cells. However, tumor cells lack this be- (diterpene types; STEFANELLI et al., 1996).
havior. Herbimycin A (Fig. 14) (Sfrepfomyces Piericidin B1 N-oxide (Fig. 17) was isolated
sp.) is a benzoquinoid ansamacrolide which from a Sfrepfomyces strain as an inhibitor of
was shown to prevent the interaction of v-src phosphatdylinositol phosphate turnover (Nr-
oncogene-expressed cells due to the inhibi- SHIOKA,1994). The drug did not inhibit the
tion of pp60""" tyrosyl protein kinase and synthesis of DNA, RNA, or proteins. Howev-
phosphatidylinositol kinase (IWAIet al., 1980; er, it reversibly reduced growth of A431 cells
SUZUKAKE-TSUCHIYA et al., 1989). This anti- and Ehrlich carcinoma (NISHIOKA,1994).
biotic was originally screened for its herbici- Echiguanin (TAKEUCHI,1992) was shown to
dal activity. inhibit phosphatidylinositol kinase. In con-
-
w
Me
Me
Hiapidospennin
pni-lectorigen R1=OMe R2=
H
Orobol R,=H Rt-
-0
Womolonln
v
Trkhostann A
R- NHofl; PLerMdrO B, N-oxide
C: R= NKO-bDglucose
Me
OH
Adenospbomn L671.776
Fig. 17. Inhibitors of the phosphoinositol metabolism.
trast to inostamycin, the inostamycins B and stance altering the transformed morphology
C displayed no inhibitory effect on phosphati- of v-src-expressed cells to normal morpholo-
dylinositol turnover despite their structural gy. These changes were accompanied by al-
similarity (ODAI et al., 1994; IMOTOet al., terations in cytoskeletal organization, synthe-
1990). sis of fibronectin, and colony-forming ability
Herbimycin A (Fig. 14), a benzoquinoid on distinct media. The compound was shown
antibiotic (IWAIet al., 1980; SUZUKAKE-TSU- to inhibit phosphoinositol kinase (Suzu-
CHIYA et al., 1990) was reisolated as a sub- KAKE-TSUCHIYA et al., 1989). The trichosta-
tins (Streptomyces toyocaensis) (Fig. 17) are stance, Sch52900 and Sch52901 are new dim-
antifungal antibiotics but also potent inducers eric diketopiperazine structures from Gliocla-
of erythroid differentiaton in mouse Friend dium sp. which inhibit c-fos protooncogene
leukemia cells. In low concentration tricho- induction as an early event in the transition of
statin A arrested the cell cycle of normal fi- cells from the quiescent to the growing stage
broblast cells in both GI and GS, and induced (CHUet al., 1995). Other fungal products, try-
the formation of proliferative tetraploid cells prostatins A and B (diketopiperazins from
after release from the G2 arrest. Inhibition of Aspergillus fumigatus), inhibit the cell cycle
phosphoinositol kinase was demonstrated for of murine tsFT210 cells by inhibition of Cdc-
this agent (YOSHIDAet al., 1990). 2-kinase (CUI et al., 1996a, b). Moreover, in-
The opposite effect of inositol phosphokin- dolocarbazoles inhibited the cell cycle pro-
ase inhibitors, i.e., an increase in cellular gression of ras-transformed rat fibroblasts
phosphatidylinositol concentration can be (AKINAGA et al., 1993).
achieved by inhibitors of inositol phosphate Ras-related proteins control a wide variety
phosphatases. This idea led to the discovery of cellular processes. They are associated with
of L-671,776 from a hyphomycete (Memnon- the cell membrane, and their function and ac-
iella echinata). It was shown to inhibit both tivity is modified by prenylation, proteolysis,
the myo-inositol-l,4-disphosphatephosphat- and carboxymethylation of the carboxyl ter-
ase and the 1,4,5-triphosphate 5-phosphatase minus (KHOSRAVI-FAR et al., 1992). The dor-
(LAMet al., 1992). The adenophostins A and rigocins A and B (Streptomycesplatensis) and
B (Fig. 17) occur as phosphorylated adeno- depudecins (Fig. 18) have recently been re-
sine analogs in cultures of Penicillium brevi- ported as new antifungal antibiotics that
compactum (TAKAHASHI et al., 1994 ). Theychange the morphology of ras-transformed
are potent agonists of the inositol-1,4,5-tri- N/H/3T3 cells to that of normal cells by in-
phosphate receptor and mimic second mes- hibiting protein carboxymethylation (KAR-
senger activity. WOWSKI et al., 1994; MATSUMOTOet al.
Diacylglycerol kinase phosphorylates di- 1992). Differanisol A (Fig. 18), from a Chae-
acylglycerol to form phosphatidic acid. This tomium strain induces the differentiation of
enzyme is affected by extracellular stimula- Friend leukemic cells in mice. Interestingly,
tors and is involved in the regulation of pro- the chemical structure of this substance is
tein kinase C by decreasing the intracellular very similar to that of the stalk cell differen-
concentration of diacylglycerol. In a search tiation inducing factor of the cellular slime
for inhibitors of this enzyme cochlioquinone mold Dictyostelium discoideum (KUBOHARA
and stemphone from Drechslera sacchari have et al., 1993; MORI, 1989). The reveromycins
recently been discovered in fungal strains A-D (Fig. 18) interrupt eukaryotic cell
(OGAWARA et al., 1994) (Fig. 17). growth due to their antagonistic effect on the
mitogenic activity of the growth hormone,
EGF (KOSHINOet al., 1992). They cause mor-
4.5.3 Substances Changing the phological reversion of src+'-NRK cells and
Morphology of exhibit antiproliferative activity for human tu-
mor cell lines. Rhizopodin (Fig. 18) is a cyto-
Oncogene-Transformed Cells or static compound from Myxococcus stipitatus
Showing Selective Toxicity which inhibits growth of various animal cul-
tures without killing the cells. Fibroblast cells
Oncogene-transformed (ras, myc, fos, erb, become larger and form long branches in an
etc.) cell lines are now available. They are irreversible manner. Protein phosphorylation
used to select for specific inhibitors of signal has been suggested to be the initial effect of
transduction pathways. The selection of these this compound (SASSEet al., 1993).
inhibitors is based on their capacity to restore Redifferentiation of cancer cells was also
normal cell morphology or selectively kill on- reported for the anthracycline type cosmomy-
cogene-expressing cells (SUZUKAKE-TSU- cins (Streptomyces cosmosus, redifferentiaton
CHIYA et al., 1990; UMEZAWA, 1989). For in- inducer of Friend leukemia cells) (KHOS-
RAVI-FAR et al., 1992), latosillan (Alculigenes changes of HL-60 cells at low concentrations
lutus, inducer of differentiation of mouse and display cytocidal activity at higher dos-
myeloic leukemia cells; high-molecular weight ages (SEKI-ASANOet al., 1994). Antitumor
polysaccharide) (HAYAKAWA et al., 1982), activities are also a characteristic of macrolac-
spicamycin (Streptomyces ulunosinicus, differ- tams, such as leinamycin and hitachimycin
entiation inducer of myeloic leukemia cells; (Fig. 18) (HARAet al., 1990; SHIBATAet al.,
Fig. 18) (HAYAKAWA et al., 1983), citrinin 1988). The antibiotic FR 901228 from Chro-
(Monosporuscus carpounus, differentiation mobucterium violuceum (Fig. 18) also rev-
inducer of myeloic leukemia cells; Fig. 18) erted the transformed morphology of a rus
(KAWASHIMAet al., 1983), canacelunin transformant cell line to normal and exhibited
(Streptomyces sp., inhibitor of cancer cell ag- antitumor activity against murine and human
glutination, inducer of lymphoblastoid trans- tumors (UEDA et al., 1994). Melastin (mol.
formation of mouse brain cells) (IKEKAWA et wt. 5 OOO f 3 000) suppresses lipopolysacchar-
al., 1980), microcystilide A (Microcystis seru- ide-induced blastogenesis of B cells more ef-
ginosu) (TSUKAMOTO, 1993), hygrolidin (Su- fectively than concanavalin A and selectively
ZUKAKE-TSUCHIYA et al., 1991), differenol A inhibits growth of several leukemia cells
(Streotomyces sp., redifferentiation inducer of (MAEDA,1993). Leptomycin B and leptolsta-
murine leukemia cells) (ASAKIet al., 1981), tin from Streptomyces strains (Fig. 18) have
lavanducyanin (Streptomyces ulriufer, stimu- recently been found to inhibit proliferation of
lation of Hela cell proliferation; Fig. 18) cultured animal cells by blocking progression
(MATSUMOTO and SETO,1991), kasuzamycin of the cell cycle in G1 and G2 phases (ABEet
(Streptomyces sp., antitumor antibiotic, af- al., 1993b). Phosmidosin C nucleosides from
fects L1210 cell cycle; Fig. 18) (TAKAMIYA et Streptomyces sp. display morphology rever-
al., 1988), trichostatins (Streptomyces sp., sim- sion activity against src transformed NRD
ilar to leptomycin, inducer of erythroid differ- cells (MATSUURAet al., 1996). Last but not
entiation in murine Friend leukemia cells, in- least, reductioleptomycin A (Streptomyces
hibition of the cell phase transfer G1 to G2; sp.) (Fig. 18) was shown to flatten the mor-
Fig. 18) (YOSHIDAet al., 1990), and hygrolid- phology of v-rusts NRK cells (HOSAKAWA et
in (SUZUKAKE-TSUCHIYA et al., 1991). al., 1993).
Modulation of the actin filament network
and/or protein phosphorylation is involved in
tautomycin-induced bleb formation. Protein 4.5.4 Inhibitors of
phosphorylation is also involved in recycling
cell surface receptors for transferrin and EGF ras-Farnesyltransferase
(KURISAKI et al., 1992).
Redifferentiation of rus-transformed cells The rus genes are mutated in 50% of colon
to normal morphology is also a characteristic and 90% of pancreatic carcinoma. Farnesyl-
of flavones and some glutarimide antibiotics protein transferase catalyzes farnesylation of
such as acetoxycycloheximide (Fig. 18) and a protein encoded by rus (Ras or p21). This is
cycloheximide (OGAWARAet al., 1989a, b). essential for the associaton of Ras proteins
Herbimycin (Fig. 14) and sparoxomycins re- with the cell membrane and rus-mediated
verse the transformed morphology of temper- transformation. Inhibition of the prenylating
ature-sensitve Rous sarcoma virus-infected enzyme, rus-farnesyltransferase, was there-
rat cells (ts/NRK) to the normal morphology fore proposed as a specific target of "soft" an-
concomitant with a drastic reduction in intra- ticancer agents. In screening for representa-
cellular p60"" kinase activity. Semisynthetic tives of such drugs, streptonigrin (Strepto-
derivatives of herbimycin prolonged the life myces albus) and its 10 '-desmethyl derivative
span of tumor-bearing mice (UEHARAet al., (VAN DER PYL et al., 1992), representatives
1988; UBUKATA et al., 1996). Some macrolide of the indolcarbazole type inhibitors of pro-
antibiotics such as FD-891 (Fig. 18) and FD- tein kinase C (CGP 412511, staurosporin,
892, which possess polyhydroxy-alkane-sub- UCN-01, K-252a), the pepticinnamins (lipo-
stituted side chains, induce morphological peptide, Streptomyces sp.) (OMURAet al.,
OH 0
Me Me Me me
0
MW no 0
o w on 0 a CY
0 Differanisol A
Dorrigocin A CY
Leinamycin
0 OR Y
FD-891
Hichimycin Reveromycin A
NHOH
Me Me
MPW
M;
Trichostatin
_.
Fr 901228
OH AH
Splcamycln tnrlnln Kazuramycln B
OH
Me Me Me Me Me Me Me Me 0
OH
0 OH
Leptolatath
bpudain
LIV.II~UCY.II~I
Fig. 18. Examples of drugs changing the morphology of oncogene-transformed cells or showing selective
toxicity against oncogene-expressedcells.
1993; SHIOMIet al., 1993), saquayamicins (SE- similar to pyocyanin, WS-9659A and B, were
KIZAWA et al., 1996), and andrastins (Penicil- also obtained from Streptomyces cultures
lium sp., OMURAet al., 1996) were found to (NAKAYAMA et al., 1989) (Fig. 20).
be new inhibitors of this enzyme and blockers
of the cell cycle progression of ras-trans-
formed cells (Fig. 19). More recently, a series 4.5.6 Miscellaneous Drugs with
of other structures from Actinoplanes sp. such Potential Antitumor Activity
as actinoplanic acid A and its derivatives have
been reported to inhibit ras-farnesyltransfer- The discovery of new cytotoxic compounds
ase (SINGHet al., 1994a, b, c). In this context, will continue. New structures such as duacins
fungal structures such as fusidienol (Fusidium (HIDAet al., 1994), leptosins (TAKAHASHI et
griseum) (SINGHet al., 1994a): kurasoins A al., 1994), cochleamycin (SHINDOet al., 1996),
and B (Paecilomyces sp. FO-3648, UCHIDAet and himastatin (LEET et al., 1996) may be
al., 1996), preussomerins, deoxypreussomer- suitable for studying structure-activity rela-
ins, Sch 49210, Sch 53514-53517 (Preusseria tionships and modes of action. A therapeutic
isomera, Harmonema demativides) (SINGHet potential was suggested for inhibitors of me-
al., 1994c), chaetomellic acid (Chaetomella lanin synthesis such as OH-3984K1 and -K2
acuseta) (SINGHet al., 1993) (Fig. 19), glio- (Streptoymces sp.), members of the albocy-
toxin, and acetylgliotoxin (VAN DER PYL et cline family of antibiotics (KOMIYAMA et al.,
al., 1992) provided non-conventional leads in 1993) (Fig. 21) and melanoxazal from the he-
the search for inhibitors of this enzyme. molymphe of the silkworm Bombyx mori
(TAKAHASHI et al., 1996) as well as for in-
hibitors of the CAMP-phosphodiesterase
4.5.5 Inhibitors of Sexual (HEDGEet al., 1993), in addition to glyoxy-
Hormone Production and lase inhibitors such as glyo I and I1 (ALLENet
al., 1993; TORNALLAY, 1990).
Hormone-Receptor Interactions Marine organisms have contributed many
new and promising antitumor agents
Suppression of the production of sexual (Fig. 21). The spongistatins from the Eastern
hormones (estrogens and androgens) and of Indian Ocean sponge species are exceptional-
receptor interaction of these hormones were ly potent growth inhibitors for human cancer
proposed as a therapeutical strategy in the cells. They comprise a series of homologous
treatment of hormone-dependent tumors. compounds distinguishable by chlorine sub-
The therapeutic efficacy of synthetic aromat- stitutions (PETTITet al., 1993a, b). Similarly,
ase inhibitors such as aminoglutethimides, im- Dolabella auricularia from the Indian Ocean
idazoles, and triazoles (VANDENBOSCHEet contains the cytotoxic dolastatins (see the do-
al., 1994; BRANDet al., 1988) for breast can- lastatin D) and its congeners (SAE et al.,
cer and tetrazoles motivated a search for non- 1993). Other recent examples of cytotoxic
steroidal microbial compounds which could compounds from marine organisms are
reduce the effects of estrogen by antagonizing phloeodictines Al-A7 and Cl-C2 which are
its receptor in a manner similar to the syn- guanidine alkaloids from the New Caledonian
thetic drug tamoxifen. The compound R1128, sponge Phloeodictyon sp. (KOURANY-LE-
an estrogen receptor antagonist, is an alky- FOLL et al., 1994), phakellistatins from a Co-
lated trihydroxyanthraquinone from Strepto- moros marine sponge (PETTIT,1994), agelas-
myces sp. 1128 (HORI et al., 1993a, b). Re- phins (sponge Agelas mauritianus) (NATORI
lated polycyclic phenolic structures are the et al., 1994; HARADAet al., 1993), amphidi-
napyradiomycins A and B (HORI et al., nolide F (KOBAYASHI et al., 1993) and bryos-
1993c) which have also recently been isolated tatins 16-18 from the marine bryozoan Bugu-
from a Streptomyces culture. A promising ap- la neritina (PETTITet al., 1996). Cyanobacter-
proach to the treatment of prostata carcino- ia also supply a rich source of cytotoxic meta-
ma is the use of inhibitors of testosterone5a- bolites. However, due to their general toxicity
reductase. Non-steroidal inhibitor structures against human cells a therapeutic application
-Q$-
y.Q 0
M*, Ml
COT 41251
UWMr
CO,H
OH 0
AdkphkwidA Fu~ldkul
Fig. 19. Examples of inhibitors of ras-farnesyltransferase.
o on
on
on on Me
lMe
n
Fig. 20. Estrogen re-
ceptor antagonists and
a#+ testosterone Sa-reduct-
R 1129 (R= Pmp,Butpent) Napyradlmycin A WS-9859A ase inhibitors.
of these compounds not is foreseeable at marine animals, by using new target-oriented

present. screening assays. Both in the field of classical
cytotoxic drugs and in the recognition of spe-
cific inhibitors of the mammalian cell cycle,
major advances have been made. Investiga-
5 Closing Remarks tions with enedyines and taxol have led to
new types of neoplasm inhibitors which are
now being introduced to clinical therapy.
During the past ten years many new drug Various novel effectors of the cell cycle and
structures have been discovered from new the signaling cascade are now used as invalua-
sources, such as new microbes, plants, and ble biochemical tools which may help to gain
5 Closing Remarks 681
2Me
hlldnA
Me
AphldlnolfdeF Lephla
Fig. 21. Selected structures of cytotoxic metabolites of marine organisms.
a deeper insight into the mechanism of malig- ABE, N., ENOKI,N., NAKAKITA, Y., UCHIDA,H.,
nant cell growth. Moreover, their therapeutic NAKAMURA,T., MUREKATA, N. (1993a), Novel
potential is presently under investigation. It is antitumor antibiotics septomycins. 11. Isolation,
likely that new antitumor drugs will be dis- physico-chemical properties and structure eluci-
dation, J. Antibiot. 46, 1536-1549.
covered in the future when new screening as-
says and natural sources become available. ABE, K., YOSHIDA,M., NAOKI,H., HORINOUCHI,
S., BEPPU,T. (1993b), Leptolstatin from Strepto-
myces sp. SAM 1595, a new GAP phase-specific
Acknowledgement inhibitor of the mammalian cell cycle. 11. Physi-
co-chemical properties and structure, J. Antibiot.
We are deeply indebted to Mrs. HILTRUD 46,735-740.
KLOSE for typing and preparing the manu- ADAMSON,R. H., SIEBER,S. M., DOUROS,J. D.
script. (1979), Development of actinomycin analogs.
Adv. Med. Oncol. Res. Educ. Proc. Int. Cancer
Congr. 12th (1979), Meeting Date 1978, Vol. 5,
pp. 93-99. Oxford: Edn. B.W. Fox, Pergamon.
6 References ADDESS,K. J., GILBERT, D. E., FEIGOU,J. (1992),

Sequence specificity of quinoxaline antibiotics: a
comparison of the binding of [N-MeCyS3] [N-
ABE, F., SHIBUYA, K., ASHIZAWA, Y., TAKAHA- MeCys71 tandem and echinomycin to DNA
SHI, K., HORINISHI, H., MATSUDA,A., ISHIZU- studied by proton NMR spectroscopy. Struct.
KA,M., TAKEUCHI, T., UMEZAWA,H. (1985), Funct. Proc. Conversation Discip. Biomol. Ste-
Enhancement of antitumor effect of cytotoxic reodyn. 7th (1992), Meeting Date 1991, Vol. 1,
agents by bestatin, J. Antibiot. 38, 411-414. pp. 147-164. (SARMA,R. H., SARMA,M. H.,
ABE, N., ENOKI,N., NAKAKITA, Y., UCHIDA,H., Eds.). New York: Adenine Press.
NAKAMURA, T. (1991), Isolation and characteri- AGARWAL,R. P., BELL, R., LILLQUIST,A.,
zation of atramycin B, new isotetracenone type MCCAFFEY,R. (1983), Purine metabolism in
antitumor antbiotic, J. Antibiot. 44,1025-1027. leukemia, Ann. Natl. Acad. Sci. 2, 160-168.
AKINAGA, S., NOMURA,K., GOMI,K., OKABE,M. A, a differentiation inducing substance against

(1993), Diverse effects of indolecarbazole com- mouse leukemia cells, J. Antibiot. '34, 919-920.
pounds on the cell cycle progression of ras-trans- ASZALOS,A. (Ed.) (1988), Antitumor compounds
formed rat fibroblast cells, J. Antibiot. 46, 1767- of natural origin: chemistry and biology, Vol. 11.
1770. Boca Raton, FL: CRC Press.
ALLEN,R. E., Lo, T. W. C., THOMALLY, P. J. ATTA-UR-RACHMAN, A., IQBAL,Z. NASIR,H.
(1993), Inhibitors of glyoxylase. I. Design, syn- (1994), Synthetic approaches to vinblastine and
thesis, inhibitory characteristcs and biological vincristine - anticancer alkaloids of Catharan-
evaluation, Biochem. SOC. Trans. 21, 535-540. thus roseus. Stud. Nut. Prod. Chem. 14 (Stereo-
ALLGAIER,C., HERTTING,G. (1986), Polymyxin selective Synthesis) (Pt. I.), 805-877.
B, a selective inhibitor of protein kinase C, dim- BARKLEY, M. D., CHEATHAM, S., THURSTON, D.
inishes the release of noradrenaline and the en- E., HURLEY, L. H. (1986), Pyrrolo[l,4]benzodia-
hancement of release caused by phorbol 12,13- zepine antitumor antbiotics: evidence for two
dibutyrate, Naunyn-Schmiedeberg's Arch. Phar- forms of tomaymycin bound to DNA, Biochem-
macol. 333,218-221. istry 25, 3021-3031.
ANDERS,M. W. (1991), Glutathione-dependent BARTEL,P. L., ZHU, C.-B., CAMPEL, Z. s.,DOSCH,
bioactivation of xenobiotics: implications for D. C., CONNORS, N. C., STROHL,W. R., BEALE,
mutagenicity and carcinogenicity, Proc. Int. J. M., FLOSS,H. G. (1990), Biosynthesis of an-
Syrnp. Princess Takamatsu Cancer Res. Fund thraquinones by interspecies cloning of actino-
(1991) Vol. Date 190,21 (Xenobiot. Cancer) 89- rhodin biosynthesis genes in streptomycetes:
99. clarification of actinorhodin gene functions, J.
ANDERSON, R. D., BERGER,N. A. (1994), Mutag- Bact. 172,48164826.
enicity and carcinogenicity of topoisomerase in- BASU,A., LAZO,J. S. (1994), Protein kinase C, in:
teractive agents, Mutat. Res. 309,109-142. New Molecular Targets in Cancer Chemotherapy.
ANDOH,T., KJELDSEN, E., BOUVEN,B. J., WEST- (KERR,D. J., WORKMAN, P., Eds.), pp. 121-14.
ERGAARD,O., ISHII, K., OKADA,K. (1993), Boca Raton, FL CRC Press.
Camptothecin-resistant DNA topoisomerase I, BEAUVILLE, D. L., KENIRY,M. A., SHAFER,R. H.
in: DNA Topoisomerase in Cancer (POTMESIL, (1990), NMR investigation of mithramycin A
M., KOHN,K. W., Eds.), pp. 249-295. New York: binding to d(ATGCAT)2: a comparative study
Oxford Univ. Press. with chromomycin Al, Biochemistry 29, 9294-
ANGANWUTAKU, I. O., ZIRBES,E., ROSAZZA,J. P. 9304.
N. (1992), Isoflavonoids from streptomycetes: BECK,W. T., DANKS,M. K., WOLVERTON, J. S.,
origins of genistein, 8-chlorogenistein, and 6,8- KIM,R., CHEN,M. (1993) Drug resistance asso-
dichlorogenistein, J. Nut. Prod. 55, 1498-1504. ciated with altered DNA topoisomerase 11, Adv.
AOYAMA, T., KOMIYAMA, F., YAMAZAKI, T., TA- Enzyme Regul. 33, 113-127.
TEE, T., ABE, F., MURAOKA, Y., NAGANAWA,
H., AOYAGI,T., TAKEUCHI, T. (1993a), Bena- BENTLER,J. A., CLARK,P., ALVAREDO,A. B.
statins C and D, new inhibitors of glutathione S- (1994), Esperamicin P, the tetrasulfide analog of
transferase, produced by Streptomyces sp. esperamicin Al, J. Nut. Prod. 57,629-633.
MI384-DF12: production, isolation, structure de- BERGER,N. A., BERGER,S. J., GERSON,S. L.
termination and biological activities, J. Antibiot. (1987), DNA repair, ADP-ribosylation and pyr-
46,712-718. idine nucleotide metabolism as targets for cancer
AOYAMA, T., ZHAO,W., KOJIMA,F., MURAOKA, chemotherapy, Anti-Cancer Drug Res. 2, 203-
Y., NAGANAWA, H., TAKEUCHI, T., AOYAGI, T. 210.
(1993b), Cysfluoretin, a new inhibitor of gluta- BERRIDGE, M. J., IRVINE,R. F. (1984), Inositol
thione S-transferase produced by Streptomyces trisphosphate, a novel second messenger in cel-
sp. MI384-DF-12, J. Antibiot. 46, 1471-1474. lular signal transduction, Nature 312, 315-321.
ARAI,H., KANDA,Y., ASHIZAWA, T., MORIMOTO, BERRIDGE,M. J., IRVINE,R. F. (1989), Inositol
M., GOMI,K., KONO,M., KASAI, M. (1994), Mi- phosphates and cell signalling, Nature 341
tomycin derivatives having unique condensed- (6239), 197-205.
ring structures, J. Antibiot. 47, 1312-1321. BERTRAM,J. S. (1990), The chemoprevention of
ARCAMONE, F. (1994), Design and synthesis of an- human cancer: an overview, in: Effects of Thera-
thracycline and distamycin derivatives as new, p y on Biology and Kinetics of the Residual Tu-
sequence-specific DNA-binding pharmacologi- mor. Part A: Preclinical Aspects, pp. 345-360.
cal agents, Gene 149,57-61. New York, London: Wiley-Alan Liss.
ASAKI,K., ONO, I., KUSUKABE, H., NAKAMURA, BIEDLER, J. L., MYERS,M. B. (1993), Multidrug re-
G., ISONO,K. (1981), Studies on differentiation- sistance (vinca alkaloids, actinomycin D, and an-
inducing substances of animal cells. I. Differenol thracycline antibiotics), in: Drug Resistance of
6 References 689
Mammalian Cells, Vol. 2 (GUPTA,R. S., Ed.), CAMPLEJOHN, R. S. (1980), A critical review of the
pp. 57-88. Boca Raton, FL: CRC Press. use of vincristin, Cell Tissue Kinet. 13(3), 327-
BILLINGS,C. P. (1993), Approaches to studying the 335.
target enzyme of anticarcinogenic protease in- CAPRANICO, G., ZUNINO,F. (1990), Structural re-
hibitors, in: Protease Inhibitors as Cancer Che- quirements for DNA topoisomerase I1 inhibi-
motherapeutic Agents (TROLL,w., KENNEDY, tion by anthracyclines, Jerusalem Symp. Quan-
A. R., Eds.), pp. 191-198. New York Plenum tum Chem. Biochem. 23 (Mol. Basis. Specif. Nu-
Press. cleic Acid Drug Interact.), 167-176.
BLUMBERG, P. M., PETTIT,G. R., WARREN,B. S., CARNEY,J. C., SCHEUER,P. J. (1993), Polohua-
SZALLASI,A., SCHUMAN, L. D., SHARKEY, N. none E, a topoisomerase-I1 inhibitor with selec-
A., NAKAMURA, H., DELL’AQUILA, M. L., DE tive lung tumor cytotoxicity from the Pohnpei
ORIES,D. J. (1989), The protein kinase C path- sponge Dysidea sp., Tetrahedron Lett. 34, 3727-
way in tumor promotion, in: Brain Carcinogene- 3730.
sis: Mechanisms and Human Relevance, pp. 201- CASSIDY,J. M., DOUROS,J. D. (1988), Anticancer
212. New York: Alan Liss. Agents Based on Natural Product Modes. New
BOGER, D. L., JOHNSON, D. S. (1996), CC-1065 York: Academic Press.
and the duocarmycins: understanding their bio-
CHAKRABORTY, A. K., MAJUMDER, HEMATA,K.,
logical function through mechanistic studies, An-
gew. Chem. (Int. Edn. Engl.) 35, 1438-1474. HODGSON,C. P. (1994), DNA topoisomerase in-
hibitors: potential uses in molecular medicine,
BOROS,C., HAMILTON, S. M., KATZ,B., KULAN- Expert Opin. Ther. Pat. 4, 655-668.
THAIVEL, P. (1994), Comparison of balanol from
Verticilliurn balanoides and ophidiocordin from CHEN,A. Y., LEROY,L. F. (1994), DNA topoiso-
Cordyceps ophioglassoides, J. Antibiot. 47,1010- merases: essential enzymes and lethal targets,
1016. Ann. Rev. Pharmacol. Toxicol. 34,191-218.
BORREL,M. N., PEREIRA,E., FIALLO,M. (1994), CHIMURA, H., ISHIZUKA, M., HAMADA,H., HORI,
Analysis of multidrug transporters in living cells, S., KIMURA,K., IWANAGA, J., TAKEUCHI, T.,
Met.-Based Drugs 1,175-182. UMEZAWA, H. (1968), A new antibiotic, macro-
BRADSHAW,R. A., PRENTIS,S. J. (1987), Onco- momycin, exhibiting antitumor and antimicro-
genes and Growth Factors. Amsterdam: Else- bial activity, J. Antibiot. 21, 4449.
vier. CHINO,M., NISHIKAWA, K., UMEKITA, M., HAYA-
BRAND,M. E., PUETT, D., GAROLA,R., FENOLL, SHI, C., YAMAZAKI, T., TSUCHIDA, T., SAWA,
K., COVEY,D. F., ZIMNISKI, D. J. (1988), Aro- T., HAMADA,M., TAKEUCHI, T. (1996), Heli-
matase and aromatase inhibitors: from enzymol- quinomycin, a new inhibitor of DNA helicase,
ogy to selective chemotherapy, in: Hormones, produced by Streptomyces sp. MJ929-SF2. I.
Cell Biology and Cancer: Perspectives and Poten- Taxonomy, production, isolation, physico-chem-
tials, pp. 65-84. New York Alan Liss. ical properties and biological activities, J. Anti-
BURCHELL, B., BAIRD,S., COUGHTRIE,M. W. H. biot. 49, 752-757.
(1991), The role of xenobiotic glucuronidating CHU,M. (1994), A novel class of antitumor metab-
enzymes in drug resistance of tumor cells and tis- olites from the fungus Nattrasia mangifera, Te-
sues, Proc. Int. Symp. Princess Takamatsu Can- trahedron Lett. 35, 1343-1346.
cer Res. Found., Volume Date 1990, 21 (Xeno- CHU, M., TRUUMEIS,I., ROTHOFSKY,M. L., PA-
biot. Cancer) 263-275. TEL, M. G., GENTILE,F., DAS, R. P., PUAR,S.
BURCK,K. B., LIN, E. T., LARRICK, J. W. (1988), M., UN, S. L. (1995), Inhibition of c-fos proto-
Oncogenes. New York, Heidelberg: Springer- oncogene induction by Sch 52900 and Sch 52901,
Verlag. novel diketopiperazines produced by Gliocla-
BUTA,J. G., KALINSKI,A. (1988), Camptothecin dium sp., J. Antibiot. 48, 1440-1445.
and other plant growth regulators in higher COLL,M., FREDERICH, C. A., WONG,A., RICH,A.
plants with antitumor activity, ACS Symp. Ser. (1987), A bifurcated hydrogen-bonded confor-
380,294-304. mation in the d(A,T) base pairs of the DNA
BUTLER,J., HAEY,B. M. (1987), Are reduced qui- dodecamer d (C6CA3T3GCG) and its complex
nones necessarily involved in the antitumor ac- with distamycin, Proc. Natl. Acad. Sci. USA 84,
tivity of quinone drugs? Br. J. Cancer (Suppl.) 8385-8389.
55, 53-59. CROW,R. T., CROTHERS,D. M. (1994), Inhibition
CAI, Y., FREDENHAGEN, A., HUG,P., MEYER,T., of topoisomerase I by anthracycline antbiotics.
PETER,H. H. (1996), Further minor metabolites Evidence for general inhibition of topoisomer-
of staurosporin produced by a Streptomyces ase I by DNA-binding agents, J. Med. Chem. 37,
longisporojlavus strain, J. Antibiot. 49, 519-526. 3191-3194.
Cur, C., UBUKATA, M., KAKEYA,H., ONOSE,R., tumor antibiotic with topoisomerase I mediated
OKADA,G., TAKAHASHI, I., ISONO,K., OSADA, DNA cleavage activity from Paecilomyces, J.
H. (1996a), Acetophtalidin, a novel inhibitor of Antibiot. 47, 949-951.
mammalian cell cycle, produced by a fungus iso- FISCHER,E. H., CHARBONNEAN, H., TACHS,N. K.
lated from a sea sediment, J. Antibiot. 49, 216- (1991), Protein tyrosine phosphatases: a diverse
219. family of intracellular and transmembrane en-
Cur, C., KAKEYA,H., OSATA,H. (1996b), Novel zymes, Science 253,401-406.
mammalian cell cycle inhibitors, tryprostatins A, FLEISCHMANN, F. F., CHAHWALA, S. B., COUNT-
B, and other diketopiperazines produced by As- LEY, L. (1986), Ras transformed cells: altered
pergillus fumigatus. 11. Physico-chemical proper- levels of phosphatidylinositol 4,s-bisphosphate
ties and structures, J. Antibiot. 49, 534-540. and catabolites, Science 232,407-410.
CURRAU,D. P. (1993), The camptothecins: a re- FOMICHOVA, E. V. (1992), Design of enzyme in-
born family of antitumor agents, J. Clin. Chem. hibitors as drugs, J. Antibiot. 45, 1185-1186.
SOC. (Tapei) 40(1), 1-6. FORNICA,J. (1977), The biogenesis of actinomy-
CURRAU,D. P., SISKO,J., YESKE,P. E., LIN, H. cins, Actinomycetes Relat. Org. l2(4), 12-20.
(1993), Recent applications of radical reactions FOSTEL,J., SHEN,L. L. (1994), DNA Topoisomer-
in natural product synthesis, Pure Appl. Chem.
ases, Design of Enzyme Inhibitors as Drugs, Vol.
65(6), 1153-1159.
2 (SANDLER,M., SMITH,H. J., Eds.), pp. 564-
DERVAN,P. D., SCOTT YOUNGQUIST, R., SLUKA, 624. Oxford: Oxford University Press.
J. P. (1987), in: Stereochemistry of Organic and
Bioorganic Transformations (BARTHMANN, W., FOX,K. R. (1990), Footprinting studies of the inter-
SHARPLESS,K. B., Eds.), pp. 221-234. Wein- action of quinomycin antibotic UK 63052 with
heim: VCH. DNA: comparison with echinomycin,J. Antibiot.
43, 1307-1315.
DIETEL,M. (1991), What’s new in cytostatic drug
resistance and pathology, Pathol. Res. Pract. 187, FOX, B. W. (1991), Natural products in cancer
892-905. treatment from bench to the clinic, Trans. R.
DUMITRIU, S. (1996), Polysaccharides in Medicinal SOC. Trop. Med. Hyg. 85,22-25.
Applications. New York, Basel, Hongkong: Mar- FRANZ,G. (1990), Biogene Cytostatica, PharmuZ
cel Dekker. 6,257-262.
ECKARDT,K., WAGNER,C. (1988), Biosynthesis of FREDENHAGEN, A., HUG,P., SAUTER,H., PETER,
anthracyclinones, J. Basic Microbiol. 28, 137- H. H. (1995), Paeciloquinones A, B, C, D, E and
144. F New potent inhibitors of protein tyrosine
EDO, K., SAITO,K., AKIYAMA-MURAI, Y., Mrzu- kinase produced by Paecilomyces carneus. 11.
GAKI,M.,KOIDE,Y., ~ S H I D AN. , (1988), An an- Characterization and structure determination, J.
titumor polypeptide antibiotic neocarzinostatin: Antibiot. 48, 199-204.
the mode of apo-protein - chromophore interac- FUNABASHI, Y., HORIGUCHI, T., IINUMA, S., TAN-
tion, J. Antibiot. 41, 554-562. IDA, S., HORADA, S. (1994), TAN-l496A, C and
EL KHADEM, H. S. (Ed.) (1982), Anthracycline An- E, diketopiperazine antibiotics with inhibitory
tibiotics. New York Academic Press. activity against mammalian DNA topoisomerase
END, D. W. (1987), Non-selective inhibition of I, J. Antibiot. 47, 1202-1218.
mammalian protein kinases by flavinoids in vi- FURK,J. L., ABDULHAMIED,T. A., PARKER,D.
tro, Res. Commun. Chem. Pathol. Pharmacol. 56, (1989), Potentiation of interleukin-2 release by
76-86. anticancer drugs, Agents Actions 26(1-2), 156-
FEIG, A. L., LIPPARD,S. J. (1994), Reactions of 157.
non-hemic iron (11) centers with dioxygen in bio- GAAL, J. C., PEARSON,C. K. (1986), Covalent
logy and chemistry, Chem. Rev. (Washington modification of proteins by ADP-ribosylation,
D.D.) 94,7594305. TIBS 11, 171-175.
FERDINAND, F. J. (1989), Mechanism of tumor in- GALE, E. F., CUNDLIFFE,E., REYNOLDS,P. E.,
duction by human retroviruses, Arzneim. RICHMOND, M. H., WARING,M. J. (1981), The
Forsch. 39(6), 735-740. Molecular Basis of Antibiotic Action. New York,
FIJII,N., YAMASHITA, Y., CHIBA,S., UOSAKI,Y., Sydney, Toronto: Wiley.
SAITOH,Y., TUJI,Y., NAKANO, H. (1993), UCE GIACCONE,G. (1994), DNA topoisomerases and
6, a new antitumor antibiotic with topoisomerase topoisomerase inhibitors, Pathol. Biol, 42, 346-
I medaited DNA cleavage activity from Actino- 352.
mycetes, J. Antibiot. 46, 1173-1174. GIRI,S., WANG,Q. (1989), Mechanism of bleomy-
FIJII,F., YAMASHITA,Y., CHIBATA,S., SAITOH, cin-induced lung injury, Comments Toxicol. 3,
Y., NAKANO,H. (1994), UCE 1022, a new anti- 145-179.
6 References 691
GOLDBERG, H. I. (1986), Molecular mechanisms of HARA,M., ASANO,K., KAWAMOTO, I., TAKIGU-
DNA sugar damage by antitumor antibiotics, CHI, T., KATSUMATA, S., TAKAHASHI, K., NA-
Ponti$ Acad. Sci. Scr. Varia 70, 425-462. KANO, H. (1990), Leinamycin, a new antitumor
GOLDBERG, H. I. (1991), Mechanism of neocarzi- antibiotic from Streptomyces. Producing organ-
nostatin action: role of DNA microstructure in ism, fermentation and isolation, J. Antibiot. 42,
determinaton of chemistry of bistranded oxida- 1768-1774.
tive damage, Acc. Chem. Res. 24, 192-196. HARADA, K., MAGUMI, T., SHIMADA, T., SUZUKI,
GOLDBERG, H. I., KAPPA,L. S., CHIN,D. H., LEE, M. (1993), Occurrence of four depsipeptides ae-
S. H. (1989), Bistranded oxidative DNA sugar rugicopeptins, together with microcystins from
damage by targeted antibiotic diradicals, New toxic cyanobacteria, Tetrahedron Lett. 34, 6091-
Methods Drug Res. 3, 27-42. 6094.
GOORMAGHTIGH, E., RUYSSCHAERT, J. M. (1984), HARPER,D. E., WELCH,D. R. (1992), Isolation,
Anthracycline glycoside-membrane interactions, purification, synthetics and antiinvasivelantime-
Biochim. Biophys. Acta 779,271-288. tastatic activity of U-77863 and U-77864 from
GREENSTEIN, M., MAIESE,W. M. (1984), Pre- Streptomyces griseoluteus strain WS6724, J. Anti-
screens for novel antineoplastic agents, Dev. Ind. biot. 45, 1827-1836.
Microbiol. 25, 267-275. HART, R. W., TURTURRO, A. (1988), Current
GREVER,M. R., SCHEPARTZ, S. A., CHAPNER, B. views of the biology of cancer, in: Carcinogen
A. (1992), The national cancer institute: cancer Risk Assessment (TRAVIS,C. C., Ed.), New
drug development and development program, York: Plenum Publishing Corp.
Semin. Oncol. 19,622-638. HARUYAMA, H., OHKUMA, Y., NAGAKI, H., OGI-
GUENGRICH, F. P. (1988), Roles of cytochrome TA,T., TANAKI, K., KINOSHITA,T. (1994), Mat-
P-450 enzymes in chemical carcinogenesis and lystatins, new inhibitors of type IV collagenases
cancer chemotherapy, Cancer Res. 48, 2946- from Actinomadura atramentaria. 111. Structure
2954. elucidation of matlystatins A to F, J. Antibiot.
GUERIITE-VOEGELEIN, F., GUENARD,D., Du- 47, 1473-1480.
BOIS,J., WAHL,A., POITIER,P. (1994), Chemi- HAYAKAWA, Y., NAKAGAWA, M., ANDO,T., SHI-
cal and biological studies in the taxol (pacitaxel) MAZU, A., SETO,H., OTAKE,N. (1982), Studies
and taxotere (docetaxel) series of new antitumor on the differentiation inducers of myceloid leu-
agents, J. Pharm. Belg. 49, 193-205. kemic cells. I. Latosillan, a new inducer of the
GUNDA,G. I., ALI, S. M., ZYGMUNT, J., JAGA- differentiation of M1 cells, J. Antibiot. 35, 1252-
SINGHE, L. R. (1994), Taxol: a novel antitumor 1254.
agent, Expert Opin. Ther. Pat. 4, 109-120. HAYAKAWA, Y., NAKAGAWA, M., KAWAI, H.,
GUNICKE, H., HOFMANN, J. (1992), Cytotoxic and TANABE,K., NAKAYAMA, H., SHIMAZU, A.,
cytostatic effects of antitumor agents induced at SETO,H., OSAKE,N. (1983), Studies on the dif-
the plasma membrane level, Pharmacol. Ther. ferentiation of inducers of myeloid leukemic
55, 1-30. cells. 111. Spicamycin, a new inducer of differen-
GUITERIDGE, J. M. C. (1992), Iron and oxygen ra- tiation of HL-60 human promyeolocytic leukem-
dicals in brain, Ann. Neurol. 32 (Suppl.), 516- ia cells, J. Antibiot. 36, 934-937.
521. HAYAKAWA, Y., HA, S., KIM,Y. J., FURIKATA, K.,
HAGIWARA, N., IRIE,K., FUNAKI, A., HAYASHI, SETO,H. (1991), Hamatorubicins A, B, C and D,
H., ARAI,M., KOSHIMIZU, K. (1988), Structure new isotetracenone antibiotics effective against
and tumor-promoting activity of new teleocidin- multidrug-resistant tumor cells, J. Antibiot. 44,
related metabolites (blastmycetins) from Strep- 1179-1 186.
toverticillium blastmyceticum, Agric. Biol. Chem. HAYAKAWA, Y., ISHIGAMI, K., SHIN-YA, K., SETO,
52, 641-648. H. (1994), Menoxymycins A and B, antitumor
HANADA, M., OKKUMA, H., YONEMOTO, T., TON- antibiotics generating active oxygen in tumor
ITA,T, OHBAYASHI, M., KAMEI,H., HIYAKI, T., cells, J. Antibiot. 47, 1344-1347.
KONISHI,M., KAWAGUCHI, H. (1991a), Madu- HECHT,S. M. (1994), RNA degradation by bleo-
ropeptin, a complex of new macromolecular an- mycin, a naturally occurring bioconjugate, Bio-
titumor antibiotic, J. Antibiot. 44,403-414. conjugate Chem. 5,513-526.
HANADA, M., KANEDA,K., NISHIYAMA,Y., Ho- HECHT,S . M., BERRY,D. E., MACKENZIE, L. J.,
SHINO,Y., KONISHI, M., OKI,T.(1991b), Hydra- DUSOY,K. W., NASUTI,C. A. (1992), A strategy
mycin, a new antitumor antbiotic. Taxonomy, for identifying novel mechanistically unique in-
isolation, physico-chemical properties, structure hibitors of topoisomerase I, J. Nut. Prod. 55,
and biological activity, J. Antibiot. 44,824-830. 401-413.
HEDGE,V. R., MILLER,J. R., PATEL,M. G., KING, diomycins A and B: non-steroidal estrogen re-
A. H., PUAR, M. S., HORAN,A., HART, R., ceptor antagonists produced by a Streptornyces,
YARBOROUGH,R., GULLO, V. (1993), Sch- J. Antibiot. 46, 1890-1893.
45752 - an inhibitor of calmodulin-sensitive cy- HOSOKAWA,N., IIMURA,H., NAGANAWA,H.,
clic nucleotide phosphodiesterase activity, J. An- HAMADA,M., TAKEUCHI,T. (1993), A new an-
tibiot. 46, 207-213. tibiotic, structurally related to leptomycin A,
HEINSTEIN,P. F., CHANG.C. J. (1994), Annu. Rev. flattens the morphology of r-ras" NRK cells, J.
Plant. Physiol. Plant. Mo1.-Biol. 45,663-674. Antibiot. 46, 676-679.
HENSENS,0.D., GOLDBERG,I. H. (1989), Mecha- HSIEH,T. S. (1992), DNA topoisomerases, Curr.
nism of activation of the antitumor antibiotic Opin. Cell Biol. 4, 396-400.
neocarzinostatin by mercaptan and sodium ICHIMURA, M., OGAWA, T., TAKAHASHI, K., Ko-
borohydride, J. Antibiot. 42, 761-767. BAYASHI,E., KAWAMOTA,I., YASUZAWA,T.,
HIDA,T., MERROI,M., TANIDA,S., HARADA,S. TAKAHASHI,I., NAKANO,H. (1990), Duocar-
(1994), Structures of duacin Al and B1 new mycin SA, a new antitumor antibiotic from
naphthyridinomycin-type antitumor antibiotics, Streptornyces sp., J. Antibiot. 43, 1037-1038.
J. Antibiot. 47, 917-921. ICHIMURA, M., OGAWA,T., KATSUMATA,S., TA-
HILL,B. T. (1986), Resistance of mammalians tu- KAHASHI, K., TAKAHASHI, I., NAKANO,H.
mor cells to anticancer drugs: mechanisms and (1991), Duocarmycin S, new antitumor antibiot-
concepts relating specificity to methotrexate and ics produced by streptomycetes: producing or-
vincristine, J. Antimicrob. Chernother. 18 (Suppl. ganism and improved production, J. Antibiot. 44,
B), 61-73. 1045-1053.
HOCHLOWSKI, J. E., MULLALLY,M. M., SPANTON, IIDA,T., KOBAYASHI,E., YOSHIDA,M., SANO,H.
S. G., WHITTERN,D. N., HILL, P., MCALPINE,J. (1989), Calphostins, novel and specific inhibitors
B. (1993), Reversion of multi-drug resistance to of protein kinase. 11. Chemical structures, J. An-
antitumor antibiotics was also reported for 5-N- tibiot. 42, 1475-1481.
acetylardeeminfrom Aspergillus fischeri, J. Anti- IIDA, K., ISHII, T., HIRAMA,M., OTANI,T., MI-
biot. 46,380-386. NAMI, Y., YOSHIDA,K. (1993), Synthesis and
HOFSTEAD, S . J., MATSON,J. A., MALACKO,A. absolute stereochemistry of the aminosugar
R., MARQUARDT,H. (1992), Kedarcidin, a new moiety of antibiotic C-1027 chromophore, Tetra-
chromoprotein antitumor antibiotic. 11. Isola- hedron Lett. 34,40794082.
tion, purification and physico-chemical proper- IKEKAWA, T., ASARI,T., MANABE,T., UMEJI,M.,
ties, J. Antibiot. 45, 1250-1292. YANOMA,S. (1980), Canacelunin, a cancer cell
HOLLIS-SHOEWALTER, H. D. (1988), Design, tu- agglutinin from Streptornyces spec., J. Antibiot.
mor biology, and biochemical pharmacology of 33,776-777.
anthrapyrazoles, in: Anthracycline and Anthra- IMOTO,M., UMEZAWA,K., ISSHIKI,K., KUNIMO-
cenedione-Based Anticancer Agents (LOWN, J. TO, S., SAWA,T., TACKEUCHI, T., UMEZAWA,
W., Ed.). Amsterdam, New York: Elsevier. H. (1987), Kinetic studies of tyrosine kinase inhi-
HOLLIS-SHOEWALTER, H. D., BUNGE, R. H., bition by erbstatin, J. Antibiot. 40,1471-1473.
FRENCH,J. C., HURLEY,T. R., LEEDS, R. L., IMOTO,M., YAMASHITA,T., SAWA,T., KURASA-
LEJA,B., MCDONNELL,P. D., EDMUNDS,C. R. WA, S., NAGANAWA, H., TAKEUCHI,T., BAO-
(1992), Improved production of pentostatin and QUAN,Z., UMEZAWA,K. (1988), Inhibition of
identification of fermentation metabolites, J. cellular phosphatidylinositol turnover by psi-tec-
Antibiot. 45, 1914-1918. torigen, FEBS Lett. 230, 43-46.
HORI,Y., ABE, Y., EZAKI,M., GOTO, T., OKUNA- IMOTO,M., UMEZAWA,K., TAKAHASHI,Y., NA-
MA, M., KOHSAKA,M. (1993a), R1128 sub- GANAWA, H., JITAKA, Y., NAKAMURA, H., KA-
stances, novel non-steroidal estrogen-receptor ZUMI,Y., SASAKI,Y., HAMADA,M., SAWA,T.,
antagonists produced by a Streptornyces. I. Tax- TAKEUCHI,T. (1990), Isolation and structure
onomy, fermentation, isolation and biological determination of inostamycin, a novel inhibitor
properties, J. Antibiot. 46, 1055-1062. of phosphatidylinositol turnover, J. Nat. Prod.
HORI, Y., ABE, Y., NISHIMURA,M., GOTO, T., 53,825-829.
OKUHARA,M., KOHSAKA,M. (1993b), R1128 IMOTO,M., SHIMURA,N., UMEZAWA,K. (1991),
substances, novel non-steroidal estrogen-recep- Inhibition of epidermal growth factor-induced
tor antagonists produced by a Streptomyces. 111. activation of phospholipase C by psi-tectorigen,
Pharmacological properties and antitumor activ- J. Antibiot. 44,915-917.
ities, J. Antibiot. 46, 1069-1075. ISHIDA,N., MIYAZAKI,K., KUMAGAI, K., RIKIMA-
HORI, Y., ABE, Y., SHIGEMATSU,N., GOTO, T., RU, K. (1965), Neocarzinostatin, an antitumor
OKUNARA,M., KOHSAKA,M. (1993c), Napyra- antibiotic of high molecular weight. Isolation,
6 References 693
physicochemical properties and biological activi- KAHN,P., GRAF,T. (1986), Oncogenes and Growth
ties, J. Antibiot. Ser. A 18, 68-76. Control. Berlin, Heidelberg: Springer-Verlag.
ISHIGAMI, K., HAYAKAWA, Y., SETO,H. (1994), KAKEYA,H., IMOTO,M., TAKAHASHI, Y., NAGA-
Cororubicin, a new anthracycline antibiotic gen- NAWA,H., TAKEUCHI, T., UMEZAWA,K.
erating active oxygen in tumor cells, J. Antibiot. (1993), Dephostatin, a novel protein tyrosine
47, 1219-1225. phosphatase inhibitor produced by Strepto-
ISONO,K. (1988), Nucleoside antibiotics: Structure, myces. 11. Structure determination, J. Antibiot.
biological activity and biosynthesis, J. Antibiot. 46, 17161719.
41, 1711-1752. KALKENVAN,C. K., HOEVENVAN DER, J. J. M.,
IWAI,Y., NAKAGAWA,A., SADAKANE, N., OMU- JONGDE, J. et al. (1991), Bepridil in combina-
RA, S . (1980), Herbimycin B, a new benzoquin- tion with anthracyclines to reverse anthracycline
oid ansamycin with anti-TMV and herbicidal ac- resistance in cancer patients, Eur. J. Cancer 27,
tivities, J. Antibiot. 38, 1114-1119. 739-744.
IWASAKI, S. (1989), Rhizoxin, an Inhibitor of Tu- KALMAN, T. I. (1989), Inhibition of thymidylate
bulin. Novel Microbial Products for Medicine synthase cycle of leukemia cells by antifolates:
and Agriculture (DEMAIN, A. L., SAMKUTI, G. the usefulness of cellular tritium assay, in: Chem.
A., HUNTER-CEVERA, J. C., ROSSMORE, H. W., Biol. Pteridines Proc. Int. Symp. Pteridines Folic
Eds.). Washington, D. C.: Society for Industrial Acid Deriv. 9th (1980) Meeting Date 1989 (CUR-
Microbiology. TINS, H., GHISLA, S., BLAN,N., Eds.), pp. 1192-
1197. Berlin: DeGruyter.
IZUMI,S., NAKAHARA, K., GOTOH,T., HASHIMO- KAMAI,F., ISSHIKI,K., UMEZAWA, Y., MORISHI-
TO, S., KINO,T., OKUHARA, M., ADRI,V., IMA- MA,H., NAGANAWA, H., TAKITA, T., TAKEU-
NAKA, H. (1983), Antitumor effects of novel im-
CHI, T., UMEZAWA, H. (1985), Vanoxonin, a
munoactive peptides, FK-156, and its synthetic new inhibitor of thymidylate synthetase. 11.
derivatives, J. Antibiot. 36, 566-574. Structure determination and total synthesis, J.
JACHOWSKI, S., REITENMEIR, C. W., SHERR,C. J., Antibiot. 38, 31-38.
ROCK,C. 0. (1986), A guanine nucleotide-de- KAMEI,H., OKA,M., HAMAGISHI, Y., TOMITA, K.,
pendent phosphatidylinositol 4d-diphosphate KONISHI, M., OKI, T. (1990), Piperafizines A
phospholipase C in cells transformed by the V- and B, potentiators of cytotoxocity of vincristine,
fms and V-fes oncogenes, J. Biol. Chem. 261, J. Antibiot. 43, 1918-1920.
4978-4985. KAMEI,H., NISHIYAMA, Y., TAKAHASHI, A., OBI,
JENSEN,P. R., FENICAL, W. (1994), Strategies for Y., OKI,T. (1991), Dynemicins, new antibiotics
the discovery of secondary metabolites from ma- with the 15-diyn-3-ene and anthraquinone sub-
rine bacteria: ecological perspectives, Annu. unit. 11. Antitumor activity of dynemicin A and
Rev. Microbiol. 48, 559-584. its triacetyl derivative, J. Antibiot. 44, 1306-
JOEL,P. S. (1994), Taxol and taxotere: from yew 1311.
tree to tumor cell, Chem. Ind. (London) ( S ) , KAMISHOHARA, M., KAWAI,H., ODAGAWA, A,,
172-175. ISOE,T., MOCHIZUKI, J.4, UCHIDA, T., HAYA-
JOHDO,O., TONE,H., OKAMOTO, R., YOSHIMO- KAWA, Y., SETO,H., TSURUO, T., OTAKE,N.
TO, A., NAKAGAWA, H., SAWA,T., TAKEUCHI, (1994), Antitumor activity of SPM VIII, a deri-
T. (1991a), Anthracycline metabolites from vative of the nucleoside antibiotic spicamycin,
Streptomyces violaceus A262. V. New anthracy- against human tumor xenografts, J. Antibiot. 47,
cline alldimycin A: a minor component isolated 1305-1312.
from obelmycin beer, J. Antibiot. 44, 1160- KANEDA, s., YOUNG, c., YAZAWA,K., TAKAHA-
1164. SHI, K., MIKAMI, Y., ARAI,T. (1987), Biological
JOHDO,O., WATANABE, Y., ISHIKURA,T., YOSHI- activities of newly prepared saframycins, J. Anti-
MOTO, A., NAGANAWA,H., SAWA, T., TAKEU- biot. 40,1640-1642.
CHI,T. (1991b), Anthracycline metabolites from KANEKO,T., WONG,H., UTRIG,J., SCHURIG, J.,
Streptomyces violaceus A262. 11. New anthracy- DOYLE,T. W. (1990), Water soluble derivatives
cline epelmycins produced by a blocked mutant of rebeccamycin, J. Antibiot. 43, 125-127.
strain SU2-730, J. Antibiot. 44,1121-11. KAO,J. W. Y., COLLINS, J. L. (1989), A rapid in
JOHDO,O., TONE,H., OKAMOTO, R., YOSHIMO- vitro screening system for the identification and
TO, A., NAGANAWA, H., SAWA, T., TAKEUCHI, evaluation of anticancer drugs, Cancer Invest. 7,
T. (1991c), Anthracycline metabolites from 303-31 1.
Streptomyces violaceus A262. IV. New anthracy- KARWOWSKI, J. P., JACKSON,M., SUNGA,G.,
cline yellamycins produced by a variant strain SHELDON,P., PODDIG,J. B., KOHL, W. L.,
SC-7, J. Antibiot. 44,1155-1159. KADAM, S. (1994), Dorrigocins: novel antifungal
antibiotics that change the morphology of ras- KELLEY,S. L., BASU,A., TEICHER,B. A., HACK-
transformed N/H/3T3 cells to that of nomal cells. ER, M. P., HAMER,P. H., LAZO, J. S. (1988),
I. Taxonomy of the producing organism, J. Anti- Overexpression of metallothionein confers re-
biot. 45, 862-689. sistance to anticancer drugs, Science 241(4874),
KASAI,M., KONO,M., SHIRAHATA, K. (1991), The 1813-1815.
derivation of a novel mitomycin skeleton: 3a-al- KHAN, S. H., MALTA, K. L. (1992), Recent Ad-
koxymitomycin, J. Antibiot. 44,301-308. vances in the Development of Potential Inhibitors
KATAGIRI,K. (1975), Quinoxaline, in: Antibiotics, of Glycosyltransferase. Glycoconjugates (ALLEN,
Vol. 3, Mechanism of Action of Antimicrobial H. J., KISAILUS,E. C., Eds.), pp. 361-378. New
and Antitumor Agents (CORCORAN, J. W., York: Marcel Dekker.
HAHN, F. E., Eds.). Berlin, Heidelberg, New KHOKHLOV,A. S., CHERCHES,B. Z., RESHETOV,
York: Springer-Verlag. P. D., SMIRNOVA, G. M., SOROKINA, I. B., PRO-
KATO, Y., OSAWA, Y., IMADA,T., IWASAKI,S., KOPTZEWA, T. A., KOLODITSKAYA,T. A.,
SHIMAZAKI,N., KOBAYASHI,T., KOMAI,T. SMIRNOV,V. V., NAVASHIN,S. M., FOMINA,I.
(1991), Studies on macrocyclic lactone antibiot- P. (1969), Physico-chemical and biological stud-
ics. XIII. Anti-tubulin activity and cytotoxicity of ies on actinoxanthin, an antibiotic from Actino-
rhizoxin derivates: synthesis of a photoaffinity myces globisporus 1131, J. Antibiot. 22, 541-
derivate, J. Antibiot. 44,66-75. 544.
KAWADA,S., YAMASHIDA,Y., UOSAKI,Y., GOMI, KHOSRAVI-FAR, R., Lox, A. D., KATO, K., DER,
K., IWASAKI,T., TAKIGUCHI, T., NAKANO,H. C. J. (1992), Protein prenylation: key to ras func-
(1992a), UCT4B, a new antitumor antibiotic tion and cancer intervention, Cell Growth Differ.
with topoisomerase I1 metiated DNA cleavage 3, 461469.
activity, from Streptomyces sp., J. Antibiot. 45, KIKUCHI,H., TETSUKA,T. (1992), On the mecha-
1182-1184. nism of lipoxygenase-like action of bleomycin-
KAWADA,M., SUMI,S., UMEZAWA,K., INONYE, iron complexes, J. Antibiot. 45, 548-555.
S., SAWA,T., SETO, H. (1992b), Circumvention KIM,Y. J., FURIKATA,K., SHIMAZU,A., FURIKA-
of multidrug resistance in human carcinoma KB TA, K., SETO,H. (1991), Isolation and structural
cells by polyether antibiotics, J. Antibiot. 45, elucidation of sekothrixide, a new macrolide ef-
556-562. fective to overcome drug-resistance of cancer
KAWASE, Y., TAKAHASHI,M., TAKATSU, T., cells, J. Antibiot. 44,1280-1282.
ARAI, M., NAKAJIMA,M., TANZAWA,K. KINGSTON,D. G. I. (1994), Taxol: the chemistry
(1996), A-72363 A-1, A-2, and C, novel hepara- and structure-activity relationship of a novel an-
nase inhibitors from Streptomyces nobilis SANK ticancer agent, Trends Biotechnol. 12,222-227.
60192. 11. Biological activities, J. Antibiot. 49, KITAHARA,M., ISHII, K., KUMADA,Y., SHIRAI-
61-64. SHI,T., FURUTA,T., MIWA,T., KAWAHANADA,
KAWASHIMA, A., NAKAGAWA,M., HAYAKAWA, H., WATANABE,K. (1985), 7-hydroxyguanine, a
Y., KAWAI,H., SETO, H., OTAKE,N. (1983), novel antimetabolite from a strain of Strepto-
Studies on differentiation inducers of myeloid myces purpurasceus. I. Taxonomy of the produc-
leukemia cells. 11. Citrinin, a new inducer of the ing organism, fermentation, isolation and biolog-
differentiation of M1 cells, J. Antibiot. 36, 173- ical activity, J. Antibiot. 38, 972-978.
174. KIWASHIMA, J., ITO, F., KATO, T., NIWANO,M.,
KAWATSU,M., YAMASHITA,T., OSONO,M., ISHI- KOSHINO,H., URAMOTO,M. (1994), Antitumor
ZUKA, M., TAKEUCHI, T. (1993), T cell activa- activity of heptelidic acid chlorohydrin, J. Anti-
tion by conagenin in mice, J. Antibiot. 46, 1687- biot. 47, 1562-1563.
1691. KLEGEMANN, M. E. (1993), Use of bacterial immu-
KAWATSU,M., YAMASHITA,T., ISHIZUKA,M., nomodulators for the treatment of cancer, in:
TAKEUCHI,T. (1994), Effect of conagenin of Recent Adv. Pharm. Ind. Biotechnol. Minutes
thrombocytopenia induced by antitumor agents Int, Pharm. Technol. Symp., 6th (1993), Meeting
in mice, J. Antibiot. 47, 1123-1129. Date 1992 (HINCAL,A. A., KAS, S. H., Eds.),
KEFRI, K., KONDO, H., YOSHINORI,T., ARAKA- pp. 179-195. Paris: Sante.
WA, H., NAKAGAMA, S., SATOH,F., KAWAMU- KNUGH,T. R., HOOK, J. W. III., BALAKRISHNAN,
RA, K., OKURA,A., SUDA,H., OKANISHI,M. M. S., CHEN.F. M. (1980), Spectroscopic studies
(1991), A new antitumor substance, BE-13793c of actinomycin and ethidium-complexes with
produced by a streptomycete. Taxonomy, fer- deoxyribonucleic acids, in: Nucleic Acids Geom-
mentation, isolation structure determination and etry and Dynamics (SARMA,R., Ed.), pp. 351-
biological activity, J. Antibiot. 44,723-728. 366. Elmsford, N Y Pergamon Press.
6 References 695
KOAMURA,M., YOSHIMURA, Y., MATSUMOTO, K., a solution from a solution of 7-amino substituted
TERUI,Y. (1988), New amedic acid antibiotics. mitomycins; mitomycin C and KW-2149, J. Anti-
11. Structure elucidation, J. Antibiof. 41, 68-72. biot. 48, 179-181.
KOBAYASHI, J., TSUDA,M., ISHIBASHI, M., SHIGE- KOPKA,M. L., YOON, C., GOODSELL, D., PIMA,P.,
MORI, H., YAMASHI, T., HIROTA,H., SAZAKI, DICKERSON, R. E. (1985), The molecular origin
T. (1993), Amphidinolide F, a new cytotoxic ma- of DNA-drug specificity in netropsin and dista-
crolide from the marine dinoflagellate Amphid- mycin, Proc. Natl. Acad. Sci. USA 82, 1 3 7 6
inum spec., J. Org. Chem. 58, 2645-2646. 1380.
KOGIRI,K., ARAKAWA, H., SATOH,F., KAWAMU- KOSHINO, H., TAKAHASHI, H., OSADA,H., ISONO,
RA, K., OKURA,A., SUDA,H., OKANISHI, M. K. (1992), Reveromycins, new inhibitors of eu-
(1991), New antitumor substances, BE-12406 A karyotic cell growth. 111. Structures of reveromy-
and BE-12406 B produced by a streptomycete. I. cins A, B, C and D. Inhibition of the mitogenic
Taxonomy, fermentation, isolation, physico- activity of epidermal growth factor, J. Anfibiot.
chemical and biological properties, J. Antibiot. 45, 1420-1427.
44,1054-1060. KOURANY-LEFOLL, E., CAPREROTE,o., SEVE-
KOMAGATA,D., SAWA, T., MURAOKA,Y., NET, T., MONTAGNAC, A.,PAIS,M. (1994), Pul-
IMADA,C., OKAMI,Y., TAKEUCHI, T. (1992a), veodictines Al-A7, and Cl-C2, antibiotic and
TA-3037A, a new inhibitor of glutathione S- cytotoxic guanidine alkaloids from the new cale-
transferase, produced by actinomycetes. I. Pro- donian sponge Sphleodictyon sp., Tetrahedron
duction, isolation, physio-chemical properties 11,3415-3426.
and biological activities, J. Antibiot. 45, 1117- KRUEGER,N. X., STREULI, M., SAITO,H. (1990),
1122. Structural diversity and evolution of human re-
KOMAGATA,D., SAWA, R., KINOSHITA,N., ceptor-like protein tyrosine phosphatases,
IMADA,C., SAWA,T., NAGANAWA, H., HA- E M B O J. 9,3241-3252.
MADA,M., OKAMI,Y., TAKEUCHI, T. (1992b), KRUGER,W. C., PSCHIGODA, L. M., MOSCOWITZ,
Isolation of glutathione transferase inhibitors, J. A. (1986), The binding of the antitumor anti-
Antibiot. 45, 1681-1683. biotic chartreusin to poly(dA-dT)-poly(dA-dT),
KOMIYAMA,K., EDANAMI, K., TANOH,A., YAMA- poly(dG-dC)-poly(dG-dC) calf thymus DNA,
MOTO,H., UMEZAWA, I. (1983), Studies on the transfer RNA and ribosomal RNA, J. Antibiot.
biological activity of stubomycin, J. Antibiot. 36, 39, 1298-1303.
301-311. KUBOHARA, Y., OKAMOTO, K., TANAKA, Y., As-
KOMIYAMA, K., FUNAYAMA, S., ANRAKU,Y., AHI,K., SAKURAI, A., TAKAHASHI, N. (1993),
MITO,A., TAKAHASHI, Y., OMURA, S., SHIMA- Differanisole A, an inducer of the differentiation
SAKI, H. (1989), Isolation of isoflavonoids pos- of Friend leukemic cells, induces stalk cell differ-
sessing antioxidant activity from the fermenta- entiation in Dictyostelium discoideum, FEBS
tion broth of Streptoymces sp., J. Antibiof. 42, Lett. 322, 73-75.
1344-1349, 1350-1355. KURAMOCHI, H., MOTEGI,A., TAKAHASHI, K.,
KOMIYAMA,K., TAKAMATSU, S., TAKAHASHI, Y., TAKEUCHI, T. (1988), DNA cleavage activity of
SHINOSE, M., HAYASHI, M., TANAKA, H., IWAI, ablomycin (NK313), a novel analog of bleomy-
Y., OMURA,S. (1993), New inhibitors of mela- cin, J. Antibiot. 41, 1846-1853.
nogenesis OH-3984 K1 and K2. I. Taxonomy, KURAMOCHI-MOTEGI, A., KURAMOCHI, H., TA-
fermentation, isolation, and biological activities. KAHASHI, K., TAKEUCHI, T. (1991), Cell killing
11. Physico-chemical properties and structural mode of liblomycin (NK313), a novel dose-survi-
elucidation, J. Anfibiof. 46, 1520-1525, 1526- val relationship different from bleomycins, J.
1529. Antibiot. 44, 429-434.
KONISHI, M., OHKUMA, H., MATSUMOTO, K., SAI- KURISAKI, T., MAGAE,J., ISONO,K., NAGAI,K.,
TOH, K., MIGAKI, T., O K I , T., KAWAGUCHI, H. YANASAKI, M. (1992), Effects of tautomycin, a
(1991). Dynemicins, new antibiotics with the 1,5- protein phosphatase inhibitor, on recycling of
diyn-3-ene and anthraquinone subunit. I. Pro- mammalian cell surface molecules, J. Antibiot.
duction, isolation and physico-chemical proper- 45,252-257.
ties, J. Antibiot. 44, 1300-1305. LAATSCH, H. (1992), A fungal polysaccharide exhi-
KONO,M., KASAI,M., SHIRAHATA, K., HIRAYA- biting antitumor activity, PharmuZ 21, 159-166.
MA, N. (1991), The configuration of mitiromycin LAM,Y. K. T., WICHMANN, C. F., HEINZ,M. S.,
and its derivation from mitomycin B, J. Antibiot. GUARIGLIA, L., GIACOBBE, R. A., MOCHALES,
44, 309-315. S., KONG, L., HONEYCUTT, S. S., ZINK, D.,
KONO,M., KASAI,M., SHIRAHATA, K., HIRAYA- BILLS, G. F., HUANG,L., BURG,R. W., MON-
MA, N. (1995), Isolation of albomitomycins from AGHAN,R. L., JACKSON, R., REID, G., MAC-
GUIRE, J. J., MCKNIGHT,A. T., RAGAN,C. I. MAEDA, J. (1993), Melastin, a novel product of
(1992), A novel phosphoinositol monophosphat- Streptomyces that selectively inhibits leukemia
ase inhibitor from Memnoniella echinata, J. Anti- cell growth, Biosci. Biotech. Biochem. 57, 969-
biot. 45, 1397-1404. 972.
LAM, K. S., VEITCH,J. A., GOLIK,J., KRISHNAN, MAEDA,J., YAZAWA,K., MIKAMI,Y., ISHIBASHI,
B., KLOHR, S. E., VOLK, K. J., FORENZA, S., M., KOBAYASHI,J. (1992), The producer and
DOYLE,T. W. (1993), Biosynthesis of esperami- biological activities of SO-075R1, a new mutacti-
cin A,, an enedyine antitumor antibiotic, J. Am. mycin group antibiotic, J. Antibiot. 45, 1448-
Chem. SOC. 115, 12340-12345. 1852.
LAN, K. H. W., FARLEY,J. R., BAYLINK,D. J. MAGAE,J., WATANABE,C., OSADA,H., CHENG,
(1989), Phosphotyrosyl protein phosphatase, X., ISONO,K. (1988), Induction of morphologi-
Biochem. J. 257,23-36. cal changes of human myeloid leukemia and ac-
LEE, M. D. (1992), Calicheamicins, a novel family tivation of protein kinase by a novel antibiotic,
of antitumor antbiotics. 4. Structure elucidation tautomycin, J. Antibiot. 41, 932-937.
of calicheamicins @Br, ylBr, J. Am. Chem. SOC. MAGAE, J., HINO, A., ISONO, K., NAGAI, K.
114,985-997. (1992), Respiratory burst induced by phorbol es-
LEE, M. D., MANNING,J. K., WILLIAMS,D. R., ter in the presence of tautomycin, a novel inhibi-
KUCK, N. A., TESTA, R. T., BORDERS,D. B. tor of protein phosphatases, J. Antibiot. 45,246-
(1989), Calicheamicins, a novel family of antitu- 251.
mor antibiotics. 3. Isolation, purification, and MAGAE,J., TSUJI,R. F., WANG,Z., KATAOKA,T.,
characterization of calicheamicins, J. Antibiot. LEE, M.-H., HANADA,T., KURISAKI,T., URA-
42,1070-1087. MOTO, M., YAMASAKI,M., NAGAI,K. (1993).
LEET, J. E. (1992), Kedarcidin, a new chromopro- Melastin, a novel product of Streptomyces that
tein antibiotic: structural elucidation of the ke- selectively inhibits leukemia cell growth, Biosci.
darcidin chromophore, J. Am. Chem. SOC. 114, Biotech. Biochem. 57, 969-972.
7946-7948. MAGNUS,P., BENNEIT,F. (1989), Synthetic studies
on the esperamicin/calicheamicin antitumor anti-
LEET, J. E., SCHROEDER,D. R., GOLIK,J., MAT-
biotics, Tetrahedron Lett. 30,3637-3641.
SON, J. A., DOYLE,T. W., LAM, K. S., HILL, S.
E., LEE, M. S., WHITNEY, J. L., KRISHNAN, B. S. MANGER,A. B. (1980), The actinomycins, Top.
Antibiot. Chem. 5,223-306.
(1996), Himastatin, a new antitumor antibiotic
MARTIN,D. G., CHICHESTER, C. G., DUCHAMP,
from Streptomyces hygroscopicus. 111. Structural
D. J., MISZAK,S. A. (1980), Structure of CC-
elucidation, J. Antibiot. 49, 299-311.
1065 (NSC-298223), a new antitumor antibiotic,
LEVITZKI,A. (1994), Protein tyrosine kinase inhib- J. Antibiot. 33, 902-903.
itors, in: New Molecular Targets in Cancer Che- MARTY, M., EXTRA,J. M., GIACCHETTI, S., Cu-
motherapy (KERR, D. J., WORKMAN,P., Eds.), VIER,C., ESPIE, M. (1994), Taxoids: a new class
pp. 67-79. Boca Raton, FL: CRC Press. of cytotoxic agents, Nouv. Rev. Fr. Hematol. 36
LIU, W. C., BARBACID,M., BULGAR,M., CLARK, (Suppl. l), 525-528.
J. M., ROSSWELL,A. R., DEAN,L., DOYLE, T. MATSHURA,T. (1988), Photochemical and oxida-
W., FERNANDES, P. B., HUANG, S., MANNE,V., tive strand cleavage of DNA, Stud. Org. Chem.
PIRNITZ,D. M., WELLS, J. S., MYERS,E. (1992), Amsterdam 88,353-366.
10’-Desmethoxystreptonigrin,a novel analog of MATSON,J. A., COLSON,K. L., BELOFSKY,G. N.,
streptonigrin, J. Antibiot. 45, 454-457. BLUMBERG,B. B. (1993), Sandramycin, a novel
LOOMIS,C. R., BELL, R. M. (1988), Sangivamycin, antitumor antibiotic produced by a Nocardioides
a nucleoside analogue, is a potent inhibitor of sp. 11. Structure determination, J. Antibiot. 46,
protein kinase C, J. Biol. Chem. 263, 1682- 162-166.
1692. MATSUMOTO,M., SETO, H. (1991), Stimulation of
LOWN,J. W. (Ed.) (1988), Anthracyclines and An- mammalian cell proliferation by lavanducyanin,
thracene-Dione-Based Anticancer Agents. Am- J. Antibiot. 44,1471-1473.
sterdam, New York Elsevier. MATSUMOTO,M., MATSUTANI,S., SUGITA, K.,
LOWN,J. W. (1993), Anthracycline and anthraqui- YOSHIDA,H., HAYASHI,F., TERN], Y., NAKAI,
none anticancer agents: current status and recent H., NOTANI,N., KAWAMURA, Y., MATSUMOTO,
developments, Pharmacol Ther. 60,185-214. K., SHOJI,J., YOSHIDA,T. (1992), Depudecin: a
LYBRAND, T. P. (1988), Interaction of peptide anti- novel inhibitor of NIH3T3 cells doubly trans-
biotics with DNA. Peptides. Chemistry and biol- formed by ras- and src-oncogene, produced by
ogy, Proc. Amer. Pept. Symp. loth, meeting date Alternaria brassicola, J. Antibiot. 45, 879-885.
1987 (MARSHALL,G. R., Ed.), pp. 416419. MATSUURA,N., ONOSE, R., OSADA, H. (1996),
Leiden: ESCOM Sci. Publ. Morphology reversion activity of phosmidosine
6 References 697
and phosmidoxine B, a newly isolated deriva- MUROI,M. (1990), Angiogenesis inhibitors TAN-
tive, on src-transformed NRK cells, J. Antibiot. 1323 C and D manufacture with Streptomyces,
49,361-365. Jpn. Patent J P 03290193.
MAYER,A. B., KATZ, E. (1978), Actinomycins, J. MURRAY,L., SIM,A. T. R., POSTAS,J. A. P., CA-
Chromatogr. Libr. 15 (Antibiotics: isolation, sep- PON, R. J. (1993), Isopalimurin: a mild protein
aration, purification), 1-33. phosphatase inhibitor from a Southern Austra-
MCALPINE,J. B., KARWOWSKI, J. P., JACKSON,M., lian marine sponge Dysidea sp., Aust. J. Chem.
MULLALY,M. M., HOCHLOWSKI, J. E., PREMA- 46,1291-1294.
CHANDRAN, U., BURRES,N. S. (1994), MLR-52, NAGAMURA,A., SATOH, M., IMURA,N. (1993),
(4' - demethylamino - 4 ',5'- dihydroxy - staurospo- Metallothionein 111 (International Conference of
rine), a new inhibitor of protein kinase C with Metallothionein, 3rd) Meeting Date 1992 (SUZU-
immunosuppressive activity, J. Antibiot. 47,281- KI, K. T., IMURA,N., KIMURA,M., Eds.). Basel:
286. Birkhauser.
MIYOSHI-SAITOH, M., MORISAKI,N., TOKIWA,Y., NAKANISHI, S. (1994), MS-282a and MS-282b, new
IWASAKI,S., KONISHI,M., SAITOH,K., OKI,T. inhibitors of calmodulin-activated myosin light
(1991), Dynemicins 0, P and Q: novel antibiotics chain kinase from Streptomyces tauricus ATCC
related to dynemicin A. Isolation, characteriza- 27470, J. Antibiot. 47, 855-861.
tion and biological activity, J. Antibiot. 44, 1037- NAKANISHI,S., MATSUDA, Y., IWAHASHI, K.
1044. KOSE, H. (1986), K-252b, c and d, potent inhibi-
MORI,K. (1989), Synthesis of differanisol A, an in- tors of protein kinase C from microbial origin, J.
ducer of differentiation, Liebigs Ann. Chem. 3, Antibiot. 39, 1067-1071.
303-305. NAKANO,H., KOBAYASHI, E., TAKAHASHI, I., TA-
MORRIS,S. J., THURSTON, D. E., NEVELL,T. G. MAOKI,T., K u z u u , Y., IBA, H. (1987), Stauro-
(1990), Evaluation of the electrophilicity of sporine inhibits tyrosine-specific protein kinase
DNA-binding pyrrolo[2,1-c][1,4]benzodioze- activity of Rous sarcoma virus transforming pro-
pines by HPLC, J. Antibiot. 43, 1286-1294. tein p60, J. Antibiot. 40,706-708.
MULLER,R. (1986), Proto-oncogenes and differen- NAKATA,M., SAITO, M., INOUYE,Y., NAKAMU-
tiaton, TZBS 11,129-132. RA, S., HAYAKAWA, Y., SETO,H. (1992), A new
MURADA,Y. (1988). Semisynthetic bleomycins, in: anthracycline antibiotic, cinerubin R. Taxono-
Horizons on Antibiotic Research (DAVIES,B. my, structural elucidation and biological activity,
D., ISHIKAWA,T., MAEDA, K., MITSCHER,L. J. Antibiot. 45, 1599-1608.
A., Eds.), pp. 88-101. Toyko: Japan Antibiotics NAKAYAMA,O., SHIGEMATSU,N., KATAYAMA,
Association. A., TAKASE,S., KIYOTO,S., HASHIMOTO,M.,
MURAMATSU,R., ABE, S., HAYASHI,H., YAMA- KOHSAKA,M. (1989), WS-9659A and B, novel
GUCHI,K., JINDA, K., SAKANO,K., NAKAMU- testosteron 5a reductase inhibitors isolated from
RA, S. (1991), Complete amino acid sequence of a Streptomyces. 11. Structural elucidation of WS-
phenomycin, an antitumor polypeptide antibiot- 9659 A and B, J. Antibiot. 42, 1230-1234.
ic, J. Antibiot. 44, 1222-1227. NATORI,T., MORITA,M., AKIMOTO,K., KOEZU-
MURAOKA,Y., SAITO, S., NOGAMI,T., UMEZA- KA,Y. (1994), Agelasphins, novel antitumor and
WA, K., TAKITA, T., TAKEUCHI, T., UMEZAWA, immunostimulatory cerebrosides from the ma-
H., SAKATA,N., HORI, M., TAKAHASHI,K., rine sponge Angelus induritianus, Tetrahedron
EKIMOTO, H., MINAMIDES, S., NISHIKAWA, K., 50,2771-2784.
KURAMOCHI,H., MOTEGI, A., FUKUOKA,T., NEER, E. J., CHAPHAM,D. E. (1988), Roles of G
NAKATANI,T., FUJII,A., MATSUDA,A. (1988), protein subunits in transmembrane signalling,
Bleomycin: structure-activity relationship in Nature 333, 129-134.
semisynthetic bleomycin, in: Horizons on Anti- NEIDLY,S., WARING,M. J. (Eds.) (1983), Molecu-
biotic Research (DAVIES,B. D., ICHIKAWA,T., lar Aspects of Anticancer Drug Action. Wein-
MAEDA, K., MITSCHER,L. A., Eds.), pp. 88- heim: Verlag Chemie.
101. Tokyo: Japan Antibiotics Research Asso- NISHIKAWA,K., SHIBASAKI, C., HIRATSUKA, M.,
ciation. ARAKAWA,M., TAKAHASHI, K., TAKEUCHI, T.
MURATA,M., MIYASAKA,T., TANAKA, H., OMU- (1991a), Antitumor spectrum of deoxyspergualin
RA, S. (1985), Diazoquinomycin A, a new anti- and its lack of cross-resistance to other antitu-
folate antibiotic inhibits thymidylate synthase, J. mor antibiotics, J. Antibiot. 44, 1201-1209.
Antibiot. 38, 1025-1033. NISHIKAWA, K., SHIBASAKI, C., UCHIDA,T., TA-
MURATA,M., TANAKA, H., OMURA,S. (1987), 7- KAHASHI,K., TAKEUCHI, T. (1991b), The na-
hydro-8-methyl-pteroylglutaminyl-glutamic acid, ture of in vivo cell-killing of deoxyspergualin
a new antifolate from an actinomycete, J. Anti- and its replication in combination with other an-
biot. 40,251-257. titumor agents, J. Antibiot. 44,1237-1246.
NISHIKORI,T., HIRUMA, S., KUROHAWA, T., SAI- OHSHIMA, S., YANAGISAWA, M., KATOH, A., Fu-
TO, S., SHIMADA, N. (1992), Microbial produc- JII, T., SANO,T., MATSUKAMA, S., FURUMAI,
tion of 4-thymidine, J. Antibiot. 45, 1376-1377. T., FUJIU,M., WATANABE, K., YAHOSE,K.,
NISHIMURA, C., NISHIMURA,T., TANAKA, N., Su- ARISAWA, M., OKUDA,T. (1994), Fusarium
ZUKI,H. (1987), Potentiation of the cytotoxicity merismoides Corola Nr. 6356, the source of the
of peplomycin against Ehrlich Ascites Carcino- protein kinase C inhibitor, azepinostatin. Taxon-
ma by bleomycin hydrolase inhibitor, J. Antibiot. omy, yield improvement, fermentation and bio-
40,1794-1795. logical activity, J. Antibiot. 47, 639-647.
NISHIOKA, H. (1994), Antitumor effect of piericid- OIKAWA, T., SHIMAMURA, M., ASHINO,H., NA-
in B, N-oxide inhibitors of phosphatidylinositol- KAMURA, 0. (1992), Inhibition of angiogenesis
phosphate turnover, J. Antibiot. 47, 441452. by staurosporine, a potent protein kinase inhibi-
NISHIOKA,H., SAWA,T., ISSHIKI, K., TAKAHASHI, tor, J. Antibiot. 45, 1155-1160.
Y., NAGANAWA, H., MATSUDA, N., HATTORI, OIKAWA, T., ASHINO, H., SHIMAMURA, M., HASE-
S., HAMADA, M., TAKEUCHI, T., UMEZAWA, H. GAWA,M., MORITA,I., MUROTA, S., ISHIZUN-
(1991), Isolation and structure determinaton of a KA,M., TAKEUCHI, T. (1993), Inhibition of an-
novel phosphatidylinositol turnover inhibitor, giogenesis by erbstatin, an inhibitor of tyrosine
piericidin B1 N-oxide, J. Antibiot. 44, 1283- kinase, J. Antibiot. 46, 785-790.
1285. OKA,M., KAMEI,H., HAMAGISHI, Y., TANITA, K.,
NOBLE,R.L. (1990), The discovery of the vinca al- MIGAKI, T., KONISHI, M., OKI, T. (1990), Chem-
kaloids - chemotherapeutic agents against can- ical and biological properties of rubiginone, a
cer, Biochem. Cell. Biol. 68, 1344-1351. complex of new antibiotics with vincristine-cyto-
NOZAKI,Y., HIDA,T., IINUMA, S., ISHII,T., SUDO, toxicity potentiating activity, J. Antibiot. 43,967-
K., MUROI,M., KANAMARU, T. (1993), TAN- 976.
1120, a new anthracycline with potent angiostat- OKA, H., YOSHINARI, T., MURAI,T., KAWAMURA,
ic activity, J. Antibiot. 46, 569-579. K., SATOH,F., FUNAISHI, K., OKURA,A.,
SUDA,H., OKANISHI, M., SHIZURI, Y. (1991), A
ODAI,H., SHINDO, K., ODAGAWA, A., MOCHIZU- new topoisomerase-I1 inhibitor, BE-10988,pro-
KI,J., HAMADA, M., TAKEUCHI, T. (1994), Inos- duced by a streptomycete. I. Taxonomy, fermen-
tamycins B and C, new polyether antibiotics, J. tation, isolation and characterization,J. Antibiot.
Antibiot. 47, 939-941.
44,486491.
OGAWARA, H., UCHINO,K., AKIYAMA, T., WA- OKADA, H., SUZUKI, H., YOSHINARI, T., ARAKA-
TANABE, S. (1985), A new 5’-nucleotidase in- WA,H., OKURA,A., SUDA,H., YAMADA, A.,
hibitor, nucleoticidin. I. Taxonomy, fermenta- UEMURA, D. (1994), A new topoisomerase I1 in-
tion, isolation and biological properties, J. Anti- hibitor, B E 22179, produced by a streptomycete
biot. 38, 153-156. I producing strain, fermentation, isolation and
OGAWARA, H., AKIYAMA, T., WATANABE, S., biological activity, J. Antibiot. 47, 129-135.
Iso, N., KOBORI, M., SEODA,Y. (1989a), Inhibi- OKURA, A., NAKADAIRA, M., NAITO,K., ISHIZU-
tion of tyrosine protein kinase activity by syn- KA,M., TAKEUCHI, T., UMEZAWA, H. (1986),
thetic isoflavones and flavones, J. Antibiot. 42, Effect of forphenicinol on y-interferon produc-
340-343. tion in mice sentisized with BCG, J. Antibiot. 39,
OGAWARA, H., HASUMI,Y., HIGASHI,K., ISHII, 569-580.
Y., SAITO,T., WATANABE, S., SUZUKI, K., Ko- OMURA, S., IMAMURA, N., KUGA,H., ISHIKAWA,
BORI, M., TANAKA, K. (1989b), Acetoxycyclo- H., YAMAZAKI, Y., OKANO,K., KIMURA,K.,
heximide and cycloheximide convert trans- TAKAHASHI, Y., TANAKA, H. (1985), Adechlo-
formed morphology of ras-transformed cells to rin, a new adenosine deaminase inhibitor con-
normal morphology, J. Antibiot. 42, 1530-1533. taining chlorine. Production, isolation, proper-
OGAWARA, H., HIGASHI, K., MANITA,S., TANA- ties, J. Antibiot. 38, 1008-1015.
KA, K., SHIMIZU, Y., SHUFANG, L. (1992), An OMURA, S., TANAKA, H., KUGA,H., IMAMURA, N.
inhibitor for inositol-specific phospholipase C (1986), Adecypenol, a unique adenosine de-
from Actinomadura sp., J. Antibiot. 45, 1365- aminase inhibitor containing homopurine and
1367. cyclopentene rings, J. Antibiot. 39, 309-310,
OGAWARA, H., HIGASHI, K., MACHIDA, T., TA- 1219-1224.
KASHIMA, J., CHIBA,N., MIKAWA, T. (1994), In- OMURA, S., VANDER PYL,D., INOKASHI, J., TA-
hibitors of diacylglycerol kinase from Drechslera KAKASHI, Y., TAKESHIMA, H. (1993), Pepticin-
sacchari, J. Antibiot. 47,499-501. namins, new farnesyl-protein transferase inhibi-
OHNO,M. (1989), From natural bleomycins to tors produced by an actinomycete. I. Producing
man-designed bleomycin, Pure Appl. Chem. 61, strain, fermentation, isolation and biological ac-
581-584. tivity, J. Antibiot. 46, 222-228.
6 References 699
OMURA,S., INOKOSHI, J., UCHIDA,R., SHIOMI,K., PETTIT,G. R., GAO, F., BLUMBERG, P. M., HER-
MASUMA,R., KAWAKUBO, T., TANAKA,H., ALD, c. L., COLL, J. O., KAMANO,Y., LEWIN,
IWAI, Y., KOSEMURA,S., YAMAMURA,S. N. E., SCHMIDT,J. M., CAAPIUS,J. C. (1996),
(1996), Andrastins A-C, new protein farnesyl- Antineoplastic agents 340. Isolation and struc-
transferase inhibitors produced by Penicillium tural elucidation of bryostatins 16-18, J. Nat.
sp. FO-3929. I. Producing strain, fermentation, Prod. 59,286-289.
isolation, and biological activities, J. Antibiot. 49, PIMENTEL, E. (1987), Oncogenes. Boca Raton, FL:
414-417. CRC Press.
OSADA,H., MAGAE,J., WATANABE, C., ISONO,K. PINEDO,H. M. (Ed.) (1980), Cancer Chemothera-
(1988), Rapid screening method for inhibitors of py: The EORT Cancer Chemotherapy. Amster-
protein kinase C, J. Antibiot. 41,925-931. dam, Oxford: Annual Excerpta Medica.
OSADA,H., TAKAHASHI, H., TSUNODA, K., KUSA- PIRRUNG, M. C., NAUHANS, S. K. (1994), Structure
KABE, H., ISONO,K. (1990), A new inhibitor of of curacin A, a novel antimitotic, antiprolifera-
protein kinase C, RK-286C (4 ‘-Demethylamino- tive, and bine shrimp toxic natural product from
4’-hydroxystaurosporin). I. Screening, taxono- the marine cyanobacterium Lyngbya majuscula,
my, fermentation and biological activity, J. Anti- Chemtracts. Org. Chem. 7, 128-129.
biot. 43, 163-167. POITIER,C. P., GUERITTE-VOEGELEIN, F., GUEN-
OTANI,T. (1993), Conformation studies and assess- ARD, D. (1994), Taxoids, a new class of antitu-
ment by spectral analysis of the protein-chromo- mor agents of plant origin: recent results, Nouv.
phore interaction of the macromolecular antitu- Rev. Fr. Hematol. 36 (Suppl. l), S21-S23.
mor antibiotic C-1027, J. Antibiot. 46, 791-802. POOT, M., HOEHN,H. (1993), DNA topoisomer-
OTSUKA,T., TAKASE,S., TERANO, H., OKUHARA, ases and the DNA lesion in human genetic insta-
M. (1992), New angiogenesis inhibitors WF- bility, Toxicol. Lett. 61, 297-308.
16775 A, and A2, J. Antibiot. 45, 1970-1973. POVIRK,C., AUSTIN,M. J. F. (1991), Genotoxicity
PAOLETTI,C. (1993), The localization of topoiso- of bleomycin, Mutat. Res. 257(2), 127-143.
merase I1 cleavage sites on DNA in the presence RAMSDELL, J. S., PEITIT, R. G. (1986), Three acti-
of antitumor drugs, Pharmacol. Ther. 60, 381- vators of protein kinase C, bryostatins, calyculin
387. and phorbol esters show differing specificities of
PAQUETTE, L. A. (1993), Synthetic progress toward action of CH4 pituitary cells, J. Biol. Chem. 261,
powerfully cytotoxic taxol and related taxane di- 17073-17 080.
terpenes, in: Organic Chemistry: Its State of Art RANCE, M. J., RUDDOCK,J. C., PACEY,M. S.,
(KISAHUEREK, M. V., Ed.), pp. 103-115. Basel: CULLEN,W. P., HUANG,L. H., JEFFERSON, M.
Helvetica Chimica Acta Publishers. T., MAEDA,H., TONE, J. (1989), UK-63052
PEREZ-SOLER,R., SUGARMAN,S., Zou, Y., complex, new quinomycin antibiotics from
PRIEBE,W. (1995), Use of drug carries to ame- Streptomyces bruegensis subsp. taponicus. Tax-
liorate the therapeutic index of anthracycline an- onomy, fermentation, isolation, characterization
tibiotics, ACS Symp. Ser. 574 (Anthracycline and antimicrobial activity, J. Antibiot. 42, 206-
Antibiotics), 300-319. 217.
PETTIT, G. R., CHICHACZ,Z. A., GAO, F., HER- RAO, A. N. (1991), Serine hydroxymethyl transfer-
ALD,c. L., BOYD,M. R. (1993a), Isolation and ase: a target for cancer chemotherapy, New
structure of the human cancer cell growth inhibi- Trends Biol. Chem. (OZAWA,T., Ed.), pp. 333-
tors. Spongistatins 2 and 3 from an Eastern 340. Tokyo: Japan Sci. SOC.Press.
Ocean Spongia sp., J. Chem. SOC. Chem. Com- RAUTH,M. A., RAYMOND, R. S. (1993), Cellular
mun. 1993, 1166-1193. approaches to bioreductive drug mechanisms,
PETTIT,G. R., HERALD,C. L., CHICHACZ,Z. A., Cancer Metastasis Rev. l2,153-164.
GAO,F., SCHMIDT, Y. M., BOYD,M. R. (1993b), REICHLE,H., DIDDENS, H., RASTAITER,J., BER-
Isolation and structure of the powerful human DEL,W. E. (1991), Resislenzmechanismen ma-
cancer cell growth inhibitors spongistatins 4 and ligner Zellen gegenuber Zytostatika, Dtsch.
5 from an African Spirastrella spinispinilifera Med. Wochenschr. 116,186-191.
(Porifera), J. Chem. SOC. Chem. Commun. 24, ROCKWELL,S., SARTOLINI, A. C., TOMASZ,M.,
1805-1807. KENNEDY, K. A. (1993), Cellular pharmacology
PETTIT,G. R., TAN,R., HERALD,D. L., CERNY,R. of quinone bioreductive alkylating agents, Can-
C., WILLIAMS,M. D. (1994), Antineoplastic cer Metastasis Rev. 12, 165-176.
agents 277. Isolation and structure of phakelli- ROSOWSKY,A,, FORSCH, R. A,, REICH, V. E.,
statin 3 and isophakellistatin 3 from a Republic FREISHEIM, J. H., MORAN,R. G. (1992), Side
of Comoras marine sponge, J. Org. Chem. 59, chain modified 5-deazafolate and 5-deazafolate
1593-1595. analogues as mammalian folylpolyglutamate
synthetase and glycinamide ribonucleotide for- SAWA,R., MATSUDA,N., UCHIDA,T., IKEDA,T.,

myltransferase inhibitors. Synthesis and in vitro SAWA,T., NAGANAWA, H., HAMADA, M., TA-
biological evaluation, J. Med. Chem. 35, 1578- KEUCHI, T. (1991), Dioxamycin, a new benz-
1588. [alanthraquinone antibiotic, J. Antibiot. 44,396-
ROTHENBERG, M. L. (1993), Taxol, taxotere and 402.
other new taxanes, Curr. Opin. Invest. Drugs 2, SCAHILL, T. A., JENSEN,R. M., SWENSON, D. H.,
1269-1277. HAKENBUHLER, N. T., PEKOLD,G., WIEREN-
SAE, H., NEMOTO,T., ISHIWATA, H., OJIKA,M., GA,W., BRAHME, N. D. (1990), An NMR study
YAMADA, K. (1993), Isolation structure and syn- of the covalent and noncovalent interactions of
thesis of dolastatin D, a cytotoxic cyclic depsi- CC-1065 and DNA, Biochemistry 29, 2852-
peptide from the sea hore Dolabella auricularia, 2860.
Tetrahedron Lett. 34, 8449-8452. SCHEID,W., WEBER,J., ROETTGERS,U., TRAUT,
SAKAI,M., KAWAI,Y., AONO,M., TAKAHASHI, K. H. (1991), Enhancement of the mutagenicity of
(1989), A new activity of herbimycin A: inhibi- anticancer drugs by the calcium agonists vera-
tion of angiogenesis, J. Antibiot. 42, 1015-1017. pamil and fendilihe, Arzneim. Forsch. 41, 901-
SAKATA, N., KARUBE,T., TANAKA, M., HAYASHI, 904.
H., HORI,M., HOTTA,K., HAMADA,M. (1989), SCHWOB,E., NASMYTH, K. (1993), CLBS and
Nucleotide sequence of the macromomycin apo- CLB6, a new pair of B cyclins involved in DNA
protein gene and its expression in Streptomyces replication in Saccharomyces cerevisiae, Genes
macromomyceticus, J. Antibiot. 42, 1704-1712. Dev. 7, 1160-1175.
SAKATA,N., SUZUKAKE-TSUCHIYA, K., MORIYA, SEBTI,S. M., LAZO,J. S. (1988), Metabolic inacti-
Y., HAYASHI, H., HORI,M., OTANI,T., NAKAI, vation of bleomycin analogs by bleomycin hy-
M., AOYAGI, T. (1992), Aminopeptidase activity drolase, Pharmacol. Ther. 38,321-329.
of an antitumor antibiotic, C-1027, J. Antibiot. SEKI-ASANO, M., OKAZAKI, T., YAMAGICHI, M.,
45, 113-117. SAKAI,N., HAMADA,K., MIZORIE,K. (1994),
SALUZ,H. P., WIEBAUER, K. (Eds.) (1995), Natu- Isolation and characterization of new 18-mem-
ral products for studying the structure of nucleic bered macrolides FD-891 and FD-892, J. Anti-
acids and DNA biomolecular complexes, Chap- biot. 47, 1226-1233.
ter 3 in: DNA and Nucleoprotein Structure in SEKIZAWA, R., IINUMA, H., NAGANAWA, H., HA-
vivo. Investigations on ProreidDNA Interactions MADA, M., TAKEUCHI, T., YAMAZUMI, J., UME-
and Nucleic Acid Structures. Austin, T X Landes ZAWA,K. (1996), Isolation of novel saquayami-
Company. cins as inhibitors of farnesyl-protein transferase,
SANCELME,M., FABRE, S., PRUDHOMME,M. J. Antibiot. 49,487-490.
(1994), Antimicrobial activities of indolocarba- SHIBATA,K., SATSUMABAYASHI, S., SANO, H.,
zole and bis-indole protein kinase C inhibitors, KOMIYAMA, K., NAKAGAWA, A., OMURA,S.
J. Antibiot. 47, 792-798. (1988), Chemical modification of hitachimycin.
SARTORELLI, A. C., TOMASZ,M., ROCKWELL, S. Synthesis, antibacterial, cytocidal, and in vivo
(1993), Studies on the mechanism of the cyto- antitumor activities of hitachimycin derivatives,
toxic action of the mitomycin antibiotics in hy- J. Antibiot. 41, 614-623.
poxic and oxygenated EMT6 cells, Adv. Enzyme SHINDO,K., MATSUOKA, M., KAWAI,H. (1996a),
Regul. 33, 3-17. Studies on cochleamycins, novel antitumor anti-
SASSE,F., STEINMETZ, H., H ~ L L EG.,
, REICHEN- biotics. I. Taxonomy, production, isolation and
BACH, H. (1993), Rhizopodin, a new compound biological activities, J. Antibiot. 49, 241-244.
from Myxococcus stipitatus (Myxobacteria) SHIOMI,K., YANG, Y., INOKOSHI, J., VAN DER
causes formation of rhizopodia-like structures in PYL, D., NAKAGAWA, A., TAKESHIMA, H.,
animal cell cultures, J. Antibiot. 46, 741-748. OMURA,S. (1993), Pepticinnamins, new farne-
SATO,K. (1989), Glutathione transferases as mark- syl-protein transferase inhibitors produced by an
ers of preneoplasia and neoplasia, Adv. Cancer actinomycete. 11. Structural elucidation of pepti-
Res. 52, 205-255. cinnamin E, J. Antibiot. 46, 229-234.
SATO,Y., MORIMOTO, A., KINE,A., OKAMURA, SINGH,S. B., ZINK,D. L., LIESCH,J. M., GOETZ,
K., HAMANAKA, R., KOHNO,K., KUWANO,M., M. A., JENKINS, R. J., NALLIN-OMSTEAD, M.,
SAKATA, T. (1993), Irsogladine is a potent inhib- SILVERMAN, K. C., BILLS,G. F., MOSLEY,R. T.,
itor of angiogenesis, FEBS Lett. 322, 155-158. GIBBS,J. B., ALBERS-SCH~NBERG, G., LING-
SATOH,T., NISHIMURA, Y., KONDO,S., TAKEU- HAM,R. B. (1993), Isolation and structure of
CHI,T. (1996), Synthesis and antimetastatic ac- chaetomellic acids A and B from Chaetomella
tivity of 6-trichloroacetamido and 6-guanidino acutiseta: farnesyl pyrophosphate mimic inhibi-
analogues of siastatin B (1996), J. Antibiot. 49, tors of ras farnesyl-protein transferase, Tetrahe-
321-325. dron 49,5917-5926.
6 References 701
SINGH,S. B., JONES,E. T, GOETZ,M. A., BILES, thracycline producing Streptomycetes, in: Ge-
G. F., NALLIN-OUSTEAD, M., JENKINS,R. G., netics and Molecular Biology of Industrial Mi-
LINGHAM, R. B., SILVERMAN, K. C., GIBBS,J. croorganisms (HERSHBERGER, C. L., QUEENER,
B. (1994a), Fusidienol: a new inhibitor of ras far- S. W., HEGEMAN, G., Eds.), pp. 68-84. Washing-
nesyl-protein transferase from Fusidium gri- ton: American Society for Microbiology.
seum, Tetrahedron. Lett. 35, 46934696. STUBBE,J., KOZARICH,J. W. (1987), Mechanism
SINGH,S. B., NESCH, J. M., LINGHAM,R. B., of bleomycin-induced DNA degradation, Chem.
GOETZ,M. A., GIBBS,J. B. (1994b), Actino- Rev. 87,1107-1136.
planic acid A a macrocyclic polycarboxylic acid SUGANUMA, M., YOSHIZAWA,S., HIROTA,M.,
which is a potent inhibitor of ras farnesyl-protein NAKOYASU, M., OJIKA,M., FUJUKI, H., SUGO-
transferase, J. Am. Chem. SOC. 116, 11606- RI, H., WAKAMATSU, K., YAMADA,K., SIGU-
11607. MURA,T. (1988), Okadaic acid an additional
SINGH,S. B., ZINK,D. C., NESCH,J. M., BALL,R. non-phorbol-12-tetradecanoiete-13-acetate-type
G., GOETZ,M. A., BOLESSA, E. A., GIACOBBE, tumor pormoter, Proc. Natl. Acad. Sci. USA 85,
R. A., SILVERMAN, K. C., BILLS,G. F., PELAEZ, 1768-1771.
F., CASCALES, C., GIBBS,J. B., LINGHAM, R. B. SUZUKAKE-TSUCHIYA, K., MORIGA,Y., HORI,M.
(1994c), Preussomerins and deoxypreussomer- (1989), Induction by herbimycin A of contact in-
ins: novel inhibitors of ras farnesyl-protein trans- hibition in v-src-expressed cells, J. Antibiot. 42,
ferase, J. Org. Chem. 59, 6296-6302. 1831-1834.
SLICHENMYER, W. J., ROWINSKY, E. K., DONE- SUZUKAKE-TSUCHIYA, K., MORIYA, Y., YAMAZA-
KOWER, R. c., KAUFMANN, s. H. (1993), The KI, K., HORI, M., HOSOKAWA, N., SAWA,T.,
current status of camptothecin analogs as antitu- IINUMA, H., NAGANAWA, H., IMADA,C., HA-
mor agents, J. Nut. Cancer Inst. 85(4), 271-291. MADA,M. (1990), Screening for antibiotics pre-
SOLAIMAN, D. (1988), Bleomycin and its Metal ferentially active against ras oncogene-expressed
Complexes. Metal-Based Anti-Tumour Drugs cells, J. Antibiot. 43, 1489-1495.
(GIELEN,M. F., Ed.), pp. 235-256. London: SUZUKAKE-TSUCHIYA, K., MORIYA,Y., KAWAI,
Publisher Freund. H., HORI,M., UEKARA,Y., IINUMA, H., NAGA-
NAWA,H., TAKEUCHI, T. (1991), Inhibition of
SONODA,T., OSADA,H., URAMOTO,M., UZAWA,
J., ISONO,K. (1989), Epiderstatin, a new inhibi- pinocytosis by hygrolidin family antibiotics: Pos-
sible correlation with their selective effects on
tor of the mitogenic activity induced by epider-
oncogene-expressed cells, J. Antibiot. 44, 344-
mal growth factor. 11. Structure elucidation, J.
348.
Antibiot. 42, 1607-1609.
SUZUKI,H., MENEGAZZI,M., CARCERERI, D.,
STEFANELLI, s., SPONGA,F., FERRARI,P., SOT- PRATI,A., GEROSA,F., TAMMASI, M., SCARPA,
TANI,C., CORTI,E., BRUNATI, C., ISLAM,K. A., SORRENTINO, S., DE PRISCO,R., LIBONATI,
(1996), Inhibition of myo-inositol monophos- M. (1990), Studies on potential markers in hu-
phatase ATCC20928 factors A and C. Isolation, man prostatic carcinoma and on a possible role
physico-chemical characterization and biological of poly(ADP4bose) polymerase in carcinogen-
properties, J. Antibiot. 49, 611-616. esis, in: Pathology of Gene Expression (FRATI,
STEINHERZ,L., STEINHERZ,P. (1991), Delayed L., AARONSON, S. A., Eds.). New York: Raven
cardiac toxicity from anthracycline therapy, Pe- Press.
diatrician 18, 49-52. TAKAHASHI, A. (1993), Russuphelins B, C, D. E
STIERLE,A., STIERLE,D., STROBEL, G., BIGNAMI, and F, new cytotoxic substances from the
G., GROTHANS, P. (1994), Endophytic fungi of mushroom Russula subniglicans HONGO,
Pacific yew (Taxus brevifolia) as a source of tax- Chem. Pharm. Bull. 41, 17261729.
01, taxanes and other pharmacophores, ACS TAKAHASHI, M. (1994), Adenophostins A and B:
Symp. Ser. 557 (Bioregulators for Crop Protec- potent agonists of inositol1,4,5-trisphosphatere-
tion and Pest Control), 6477. ceptor produced by Penicillium brevicompac-
STREKOWSKI, L. (1992), Molecular basis for the en- cum. Taxonomy, fermentation, isolation, physi-
hancement and inhibition of bleomycin-me- co-chemical and biological properties, J. Anti-
diated degradation of DNA and DNA-binding biot. 47, 1643-1647.
compounds,. Adv. Detailed React. Mech. 2 TAKAHASHI, A., NAKAMURA, H., KAMEYAMA, T.,
(Mech. Biol. Importance), 61-109. KURASAWA, S., NAGANAWA, H., OKAMI,Y.,
STROHL,W. R., BARTEL,P. L., CONNORS, N. C., TAKEUCHI, T., UMEZAWA, H. (1987a), Bisuca-
ZHU, C.-B., DOSCH,D. C., BEALE,J. M., JR., berin, a new siderophore sensitizing tumor cells
FLOSS,H. G., STUTZMAN-ENGWALL, K., OT- to macrophage-mediated cytolysis. 11. Physico-
TEN, s. L., HUTCHINSON, c. R. (1989), Biosyn- chemical properties and structure elucidation, J.
thesis of natural and hybrid polyketides by an- Antibiot. 40,1671-1676.
TAKAHASHI, I., KOBAYASHI, E., ASANO,K., Yo- TANAKA, N., OKABE,T., ISONO,F., KASHIWAGI,
SHIDA,M., NAKANO,H. (1987b), UCN-01, a se- M.,NOMOTO,K., TAKAHASHI, M., SHIMAZU,
lective inhibitor of protein kinase C from Strep- A., NISHIMURA, T. (1989), Lactoquinomycin, a
tomyces, J. Antibiot. 40,1782-1784. novel anticancer antibiotic. I. Taxonomy, isola-
TAKAHASHI, K., KATSUTOSHI, E., HISAO,M., SEI- tion and biological activity, J. Antibiot. 38, 1327-
KI, N. (1987c), Liblomycin, a new analogue of 1332.
bleomycin, Cancer Treat. Rev. 14, 169-177. TANAKA,S., OHKUBO,M., KOJIRI,K., SUDA,H.
TAKAHASHI, I., ASONO,K., KAWAMOTO, I., TA- (1992), A new indolocarbazole antitumor sub-
MAOKI,T., NAKANO,H. (1989), UCN-01 and stance ED-110, a derivative of BE-l3793c, J. An-
UCN-02, new selective inhibitors of protein kin- tibior. 45, 1797-1798.
ase C. I. Screening, producing organism and fer- TANAKA, S., OKABE,T., NAKAJIMA, S., YOSHIDA,
mentation, J. Antibiot. 42, 564-570. E., SUDA,H. (1994), BE-23372M, a novel pro-
TAKAHASHI, C., NUMATA,A., MATSUMURA, E., tein tyrosine kinase inhibitor. 1. Producing or-
MJNOURA,K., ETO, H., SHINGU,T., ITO, T., ganism, fermentation, isolation, and biological
HASEGAWA, T. (1994), Leptosins I and J, cyto- activities, J. Antibiot. 47,289-293.
toxic substances produced by a Leprosphaeria TATSUDA,K., YASUDA,S. (1996), Synthesis and
sp. Physico-chemicalproperties and structures, J. biological evaluation of caloporoside analogs, J.
Antibiot. 47, 1242-1249. Antibiot. 49, 713-715.
TAKAHASHI, S., HASHIMOTO,P., HAMANO,K., TATSUDA,K., IKEDA,Y., MIURA,S. (1996), Syn-
SUZUKI,T., NAKAGAWA, A. (1996), Melanoxa- thesis and glycosidase inhibitory activities of
zal, new melanin biosynthesis inhibitors discov- nagstatin triazole analogs, J. Antibiot. 49, 836-
ered by using the larval haemolymph of the silk- 837.
worm Bombyx mori. Production, isolation, TAYLOR,G. S. (1989), CAMP-dependent protein
structural elucidation and biological properties, kinases, J. Biol. Chem. 264, 8443-8446.
J. Antibiot. 49, 513-518. THORNALLEY, P. J. (1990), The glyoxylase system:
TAKAMIYA, K., YOSHIDA,E., TAKAHASHI, T., new developments towards functional character-
OKURA,A., OKANISHI,M., KOMIYAMA, K., ization of a metabolic pathway fundamental to
UMEZAWA, I. (1988), The effect of kazusamycin biological life, Biochem. J. 269, 1-11.
B on the cell cycle and morphology of cultured TOLSTIKOV, V. V., HOLPNEKOZLOVA,N. Y., OR-
L1210 cells, J. Antibiot. 41, 1854-1861. ESHKINA, T. D., OSIPOVA,T. V., PREOBRAZ-
TAKARA,M., TOMIDA,A., NISHIMORA, T., YA- HENSKAYA, M. N., SZTARICSKAI, F., BALZAR-
MAGUCHI, H., SUZUKI, H. (1990), Resorthiomy- INI,J., DE CLERCQ, E. (1992), Amides of anti-
cin, a novel antitumor antibiotic. 111. Potentia- biotic streptonigrin and amino dicarboxylic acids
tion of antitumor drugs and its mechanism of ac- or amino-sugar. Synthesis and biological evalua-
tion, J. Antibiot. 43, 138-142. tion, J. Antibiot. 45, 1020-1025.
TAKE,Y., KUBO,T., TAKEMORI, E., INOYE,Y., TOMASZ,M. (1992), Mitomycin C: DNA sequence
NAKAMURA, S., NISHIMURA, T., SUZUKI,H., specificity of a natural DNA cross-linking agent,
YAMASUCHI, H. (1989), Biological properties of Adv. DNA sequence specific Agents 1,247-261.
streptonigrin derivatives. 111. In vitro and in vivo TOMASZ,M. (1994), DNA-adducts of mitomycin.
antitumor activities, J. Antibiot. 42, 968-1976. ZARC Sci. Publ. 125 (DNA-Adducts: Identifica-
TAKEUCHI, T. (1992), Novel physiologically active tion and Biological Significance), 349-357.
substance, Jpn. Patent JP 04217681. TROLL,W., WIESNER,R., FRENKEL,K. (1987),
TAKITA,T. (1984), The bleomycins: properties, Anticarcinogenic action of protease inhibitors,
biosynthesis and fermentation, in: Biotechnology Adv. Cancer Res. 49,265-283.
of Industrial Antibiotics (VANDAMME, E. J., TROWBRIDGE, I. S . (1991), CD45, J. Biol. Chem.
Ed.), pp. 595-621. New York Marcel Dekker. 266,23 517-23 520.
TAKITA,T., MURAOKA,Y. (1990), Biosynthesis TSCHAGOSHI, S., TAKENDO,T. UMEZAWA,H.
and Chemical Synthesis of Bleomycin. Biochem- (1986), Antitumor substances, in: Biotechnology,
istry of Peptide Antibiotics (KLEINKAUF,H., Vol. 4 (REHM,H.-J., REED,G., Eds.), pp. 509-
VON DOEHREN, H., Eds.), pp. 289-309. Berlin: 530. Weinheim: VCH.
DeGruyter. TSUCHIYA, K. S., MORIYA,Y., HON,M., ITOH,O.,
TAKITA,T., OGINO,T. (1987), Peplomycin and li- TAKEUCHI, T., EKIMOTO,H., HIRATSUKA, M.
blomycin, new analogs of bleomycin, Biomed. (1992), Synergism between 5-fluorouracil and
Pharmacother. 41,219-226. oxanosine in inhibition of growth of ras-ex-
TAKITA,T., MURAOKA,Y., TAKAHASHI,K. pressed cells in vitro and in vivo, J. Antibiot. 45,
(1989), Bleomycins: basic research, Gann Mo- 283-285.
nogr. Cancer Res. 36 (Antitumor. Nat. Prod.), TSUKAMOTO, S. (1993), Microcystilide A. A novel
59-70. cell-differentiation-promotingdepsipeptide from
6 References 703
Microcystis seruginosa NO-15-1840, J. Amer. UMEZAWA,K., HARESAKU, M., MURAMATSU, M.,

Chem. SOC. 115, 11046. MATSUSHIMA,T. (1987), Mutagenicity of an-
TSUNAKAWA, M., KAMEI,H., KONISHI,M., MIGA- thracycline glycosides and bleomycins in a Sal-
KI, T.,OKI,T., KAWAGUCHI, H. (1988), Poro- monella assay system, Biomed. Pharmacother.
thramycin, a new antibiotic of the anthramycin 41,214-218.
group: production, isolation, structure and bio- UOSAKI,Y., YASUZAWA,T., HARA,M., SAITOH,
logical activity, J. Antibiot. 41, 1366-1373. Y., SANO,H. (1991), Sapurimycin, new antitu-
TSURUOKA, T., FUKUYASHI, H., ISHI,M., USUI,T., mor antibiotic produced by streptomyces struc-
SHIBAHARA, S., INOUYE,S. (1996), Inhibition of ture determinaton, J. Antibiot. 44, 40-44.
mouse tumor metastasis with nojirimycin-related UOSAKI,Y., KAWADA,S., NAKANO,H., SAITOH,
compounds, J. Antibiot. 49, 155-161. Y., SANO,H. (1993), UCT4B, a new antitumor
UBUKATA,M., UZAWA,J., OSADA,H., ISONO,K. antibiotic with topoisomerase I1 mediated DNA
(1993), Respinomycins A l , A2, B, C, and D, a cleavage activity. Structure determination, J. An-
novel group of anthracycline antibiotics. 11. Phy- tibiot. 46, 235-240.
sicochemical properties and structure elucida-
VAN DER PYL, D., INOKOSHI,J., SHIOMI, K.,
tion, J. Antibiot. 46, 942-951. YANG,H., TAKESHIMA, H., OMURA,S. (1992),
UBUKATA, M., MORITA, T., URAMOTO, M.,
Inhibition of farnesyl protein transferase by glio-
OSADA, H. (1996), Sparoxomycins A1 and A2,
toxin and acetylglyotoxin, J. Antibiot. 45, 1803-
new inducers of the flat reversion of NRK cells
1805.
transformed by temperature sensitive Rons sar-
VAN DER WILT, C. L., PETERS,G. J. (1994), New
koma virus. 11. Isolation, physicochemical prop-
erties and structure elucidation, J. Antibiot. 49, targets for pyrimidine antimetabolites in the
65-70. treatment of solid tumors. I. Thymidylate
UCHIDA,T., IMOTO,M., WATANABE,Y., MIURA, synthase, Pharm. World Sci. 16, 84-103.
K., DOBASHI,T., MATSUDA,N., SAWA,T., NA- VANDENBOSCHE,H., MOLREELS,H., KOYMANS,
GANAWA, H., HAMADA,M., TAKEUCHI,T., L. M. H. (1994), Aromatase inhibitors - mecha-
UMEZAWA, H. (1985), Saquayamycins, new nisms for nonsteroidal inhibitors, Breast Cancer
aquayamycin-group antibiotics, J. Antibiot. 38, Res. Treat. 30,43-55.
1171-1181. VANEK,Z., TAX,J., KOMERSOVA,I., SEDMERA,
UCHIDA,R., SHIOMI,K., INOKOSHI,J., MASUMA, P., VOKOUN,J. (1977), Anthracyclines, Folia
R., KAWAKUBO,T., TANAKA,H., IWAI, Y., Microbiol. 22, 139-159.
OMURA, S. (1996), Kurasoins A and B, new pro- VYAS, D. M. (1993), Recent advances in medical
tein farnesyltransferase inhibitors produced by chemistry of taxol, Pharmacochem. Libr. 20
Paecilomyces sp. FO-3684, J. Antibiot. 49, 932- (Trends in Drug Research), 261-270.
934. WAGNER,C., ECKARDT,K., IHN,W., SCHUMANN,
UEDA, H., NAKAJAMA,H., HORI,Y., FUJITA,T., G., STENGEL,C., FLECK,W. F., TRESSELT, D.
NISHIMURA,M., GOTO, T., OKUHARA, M. (1991), Biosynthese der Anthracycline: eine
(1994), FR 901228, a novel antitumor bicyclic Neuinterpretation der Ergebnisse zur Daunomy-
depsipeptide produced by Chromobacterium cin-Biosynthese, J. Basic Microbiol. 31, 223-
violaceum, No. 968, I. Taxonomy, fermentation, 240.
isolation, physico-chemical and biological prop- WAKUSAWA, S., INOKO, K., MIGAMOTO, K.
erties, and antitumor activities, J. Antibiot. 47, (1993), Staurosporine derivatives reverse multi
310.
drug resistance without correlation with their
UEHARA, Y., MURAKAMI,Y., SUZUKAKE-TSU-
protein kinase inhibitory activities, J. Antibiot.
CHIYA, K., MORIYA,Y., SANO, H., SHIKATA,
46,353-355.
K., OMURA,S. (1988), Effects of herbimycin
derivatives on src oncogene function in relation WALL,M. E., WANI,M. C. (1993), Camptothecin
to antitumor activity, J. Antibio?. 41, 831-834. and analogs: synthesis, biological in vitro and in
UMEZAWA,K. (1989), Screening of oncogene func- vivo activities, and clinical possibilities, ACS
tion inhibitors, possible antitumor agents. Anti- Symp. Ser. 534 (Human Medical Agents from
cancer Drugs, Colloques INSERM, Vol. 191, Plants), 146-169.
(TAPIERC,H., Ed.), pp. 13-21. WANG,H. H. J. (1992), Intercalative drug binding
UMEZAWA,H., IMOTO,M., SAWA,T., ISSHIKI,K., to DNA, Curr. Opin. Struc?. Biol. 2, 361-368.
MATSUDA,N., IINUMA,H., HAMADA, M., TA- WANG,A. H. J., LEGHETTO, G., QUIGLEY,G. J.,
KEUCHI, T. (1986), Studies on a new epidermal RICH, A. (1987), Interactions between an an-
growth factor-receptor kinase inhibitor, erbsta- thracycline antibiotic and DNA: molecular struc-
tin, produced by MH435-hF3, J. Antibiot. 39, ture of daunomycin complexed to d( CGTACG)
170-1 73. at 1,2A-resolution, Biochemistry 26, 1152-1163.
WARDMAN,P. (1990), Bioreductive activation of YAMASHITA,T., NAOI,N., HIDAKA,T., HORI, S.,

quinones: redox properties and thiol reactivity, KIMURA,K., TAKEUCHI,T., UMEZAWA,H.
Free Radicals Res. Commun. 8,219-229. (1979), Studies on auromomycin, J. Antibiot. 32,
WARING,M. E. (1975), Ethidium and propidium, 330-339.
in: Antibiotics, Vol. 3, Mechanisms of Action of YAMASHITA,T., SAKAI,M., KAWAI,Y., AONO,
Antimicrobial and Antitumor Agents (CORCO- M., TAKAHASHI,K. (1989), A new activity of
RAN, J . W., HAHN, F. E., Eds.), pp. 141-166. herbimycin A: inhibition of angiogenesis, J. An-
Berlin, Heidelberg, New York: Springer-Ver- tibiot. 42, 1015-1017.
lag. YAMASHITA, Y., SAITOH,Y., ANDO,K., TAKAHA-
WARING,M. J. (1979), Echinomycin, triostin and SHI,K., OHNO, H., NAKANO,H. (1990a), Sainto-
related antibiotics, in: Antibiotics, Vol. 512, pin, a new antitumor antibiotic with topoisomer-
Mechanisms of Action of Antibacterial Agents ase I1 dependent DNA cleavage activity, from
(HAHN,F. E., Ed.), pp. 173-194. Berlin, Heidel- Paecilomyces, J. Antibiot. 43, 1344-1346.
berg, New York: Springer-Verlag. YAMASHITA,Y., KAWADA,S., FUJII,N., NAKANO,
WARING,M. J. (1990), The molecular basis of spe- H. (1990b), Induction of mammalian DNA to-
cific recognition between echinomycin and poisomerase I1 dependent DNA cleavage by an-
DNA. Jerusalem Symp. Quantum Chem. Biol. 23 titumor antibiotic streptonigrin, Cancer Res. 50,
(1990) (Molecular Basis of Specific Nucleic 5841-5844.
Acid-Drug Interactions) 225-24. YAMASHITA, T., OSONO, M., MASUDA,T., ISHIZU-
WARING,M. J. (1992), Echinomycin, Puthof. Biof. KA, M., TAKEUCHI, T., KAWATSU,M. (1993),
40,1022-1024. Effect of conagenin in tumor bearing mice. Anti-
WAIT, P. M., HICKSON,I. D. (1994), Structure and tumor activity, generation of effector cells and
function of type I1 DNA topoisomerases, Bio- cytokine production, J. Antibiot. 46, 1692-1698.
chem. J. 303,681-695. YAMAZAKI, M., YAMASHITA, T., HARADA,T., NI-
WEBER, H., SCHU, P., ANKE, T., VELTEN, R., SHIKORI,T., SAITO,S., SHIMADA, N., FUGII,A.
STEGLICH, W. (1994), Caloporoside, a new (1992), 44-Homooligomycins A and B, new anti-
inhibitor of phospholipases C from Caloporus tumor antibiotics from Streptomyces bottropen-
sis, J. Antibiot. 45, 171-179.
dichrous (Fr.) Ryr., J. Antibiot. 47, 1188-1194.
YAMAZAKI, T., TATEE,T., HOYAMA,T., KOJINA,
WEINBERG,R. A. (1985), The action of oncogenes F., TAKEUCHI,T. (1993), Bequinostatins C and
in the cytoplasm and nucleus, Science 230, 770- D, new inhibitors of glutathione S-transferase
776. produced by a Streptomyes spec. MB84-DF12, J.
WEISS, R. B. (1992), The anthracyclines will ever Antibiot. 46, 1309-1311.
find a better doxorubicin? Semin Oncol. 19,670- YAMAZAKI, K., AMEMIAY,M., ISHIZUKA, M., TA-
686. KEUCHI, T. (1995), Screening for apoptosis in-
WHITE,J. R., STROSHANE,R. M. (1984), Daunoru- ducers in microbial products and induction of
bicin and adriamycin: properties, biosynthesis, apoptosis by cytostatin, J. Antibiot. 48, 1138-
and fermentation, Chapter 19, in: Biotechnology 1140.
of Industrial Antbiotics (VANDAME, E. J., Ed.), YANAGISAWA,M. (1994a), Hispidospermidin, a
pp. 569-594. New York, Basel: Marcel Dekker. novel phospholipase C inhibitor produced by
WILMAN,D. E. V. (Ed.) (1990), The chemistry of Chaetomia hispidulum (Cda) Moesz NR 7127. I.
antitumor agents. New York: Chapman and Screening, taxonomy, and fermentation, J. Anti-
Hall. biot. 47, 1-5; 11. Isolation, characterization and
WOLF, M., BAGGIOLINI,M. (1988), The protein structural elucidation, J. Antibiot. 47, 6-17.
kinase inhibitor staurosporin, like phorbol es- YANAGISAWA, M. (1994b), Hispidospermidin, a
ters, induces the association of protein kinase C novel phospholipase C inhibitor produced by
with membranes, Biochem. Biophys. Res. Com- Chaetomia hispidulum (Cda) Moesz NR 7127.
mun. 154,1273-1279. 11. Isolation, characterization and structural elu-
YAGINUMA, S., MOTO, N., TSUJINO,M., SUDATE, cidation, J. Antibiot. 47, 6-17.
Y., HAGASHI,M., OTANI,M. (1981), Neplano- YARDIN,Y., ULLRICH,A. (1988), Molecular tar-
cin A, a new antitumor antibiotic. I. Producing gets of signal transduction by growth factors,
organism, isolation and characterization, J. Anti- Biochemistry 27, 3113-31 19.
biot. 34,359-366. YOSHIDA,E., KOMIYAMA, K., NORITO,K., WATA-
YAMAGUCHI, T., FURUMAI, T., SATO,M., OKUDA, NABE, Y., TAKAMIYA, K., OKURA, A., FUNAI-
T., ISHIDA,N. (1970), Studies on a new antitu- SHI, K., KAWAMURA, K., FUNAYAMA, S., UME-
mor antibiotic, largomycin. I. Taxonomy of the ZAWA, I. (1987), Antitumor effect of kazusamy-
largomycin-producing strain and production of cin B on experimental tumors, J. Antibiot. 40,
the antbiotic, J. Antibiot. 23, 369-372. 1596-1604.
6 References 705
YOSHIDA,S., NAGAWARA,H., AOYAGI,T., TA- YUAN,B. D., Wu, R. T., SATO, I., OKABE, T., Su-
KEUCHI, T., UMEZAWA, A. (1988), Benadrostin, ZUKI,H., NISHIMURA, T., TANAKA,N. (1985),
new inhibitor of poly(ADP-ribose) synthetase, Biological activity of cadeguomycin. Inhibition
produced by actinomycetes. 11. Structure eluci- of tumor growth and metastasis. Immunostimu-
dation, J. Antibiot. 41, 1015-1018. lation, and potentiation of 1-P-D-arabino-fura-
YOSHIDA,M., HOSHIKAWA, Y., KOSEKI,K., HORI, nosylcytosine, J. Antibiot. 38, 642-648.
K., BEPPU,T. (1990), Structural specificity for ZHOU, N., JAMES,T. L., SHAFER, R. H. (1989),
biological activity of trichostatin A, a specific in- Binding of actinomycin D to (d(ATCGAT)),:
hibitor of mammalian cell cycle with potent dif- NMR evidence of multiple complexes, Biochern-
ferentiation-inducing activity in Friend leukemia istry 28, 5231-5239.
cells, J. Antibiot. 43, 1101-1106. ZIMMER,C., STORL, K., STORL, J. (1990), Micro-
YOSHIMOTO,A., JOHDO,O., WATANABE,Y., NI- bial DNA topoisomerases and their inhibition
SHIDA,H., OKAMOTO,R. (1992), Production of by antibiotics, J. Basic Microbiol. 30,209-224.
a new anthracycline antibiotic betaclamycin B by ZUNINO,F., CAPRANICO, G. (1990), DNA topoiso-
microbial conversion with a specific aclacinomy- merase I1 as the primary target of antitumor an-
cin-negative mutant, J. antibiot. 45, 1005-1007. thracyclines, Anti-Cancer Drug Res. 5, 307-317.
Biotechnology
Index
A Agaricus bisporus, endogenous siderophore 228

acarbose 438f Agrobacterium rhizogenes, initiation of plant hairy
acarbose family, structure of - 472 root cultures 619
Acinetobacter baumannii, acetinobactin 219 Agrobacterium tumefaciens, agrobactin 204
Acinetobacter huemolyticus, acetinoferrin 221f agrobactin, catecholate siderophore of Agrobac-
acinetobactin, siderophore of Acinetobacter bau- terium tumefaciens 204
mannii 219 - structure 204
- structure 219 ajmalicine 607ff
acetylcholine receptor agonist 123 - from plant cell cultures 608f
acetylenes, antifungal activity 503 - - yields 608
acinetoferrin, dihydroxamate siderophore of Aci- - structure 607
netobacter haemolyticus 221f D-alanine, biosynthesis 330
actagardine, action on cell wall of susceptible bac- alanine racemase, characterization 580
teria 357 - role in cyclosporin biosynthesis 580f
- primary structure 337f albomycins 211f
actinoidin A 375 - hydroxamate siderophores of Streptomyces, fer-
- structure 373 rioxamine type 211f
Actinomadura madurae, maduraferrin 219f - structure 211
actinomycetes, as sources of nonribosomal pep- Alcaligenes eutrophus, poly-phydroxybutyrate
tides 304ff 178
- lateral gene transfer to Bacillaceae 457 alcaligin, hydroxamate siderophore of Bordetella
- oligosaccharide structures 35 pertussis 210f
actinomycin D 650ff - structure 211
- antitumor agent 651 aleurodiscal 502f
- structure 651 - antifungal activity 503
actinomycins, antitumor activity 650ff - structure 502
ACVS see &(L-a-aminoadipy1)-L-cysteinyl-D-Val- algae see oleaginous microalgae
ine synthetase alkaloids, from plant cell cultures 603ff
acyl-coenzyme A :isopenicillin N acyltransferase alldimycin, antitumor agent 659
(AT) 257ff - structure 659
adechlorin, inhibition of adenosine deaminase alliacols, cytotoxic activity 507
648 - structure 505
adenomycin, cyclitol moieties 454 allosamidin, chitinase inhibitor 440
adenosine deaminase inhibitors, treatment of leu- allosamidins, structure 472
kemia 648 alterobactin, peptide siderophore of Alteromonas
Aerobacter aerogenes, aerobactin 222f luteoviolacea 212
aerobactin, siderophore of Aerobacter aerogenes - structure 212
222f Alteromonas haloplanctis, bisucaberin 210f
Aeromonas hydrophila, amonabactins 206 Alteromonas luteoviolacea, alterobactin 212
A-factor 27 ametantron, antitumor agent, structure 651
- of Streptomyces griseus, cascade 87f amikacin, semisynthetic aminoglycosides 461
- structure 80 amino acids, found in lantibiotics 327ff. 347
108 Index
- - D-alanine 330 - semisynthetic 461

- - didehydroalanine 328 - sorbistins 439f
- - didehydrobutyrine 328 - streptomycin-related - 427ff
- - lanthionine 328 - - boholmycin 428
- - methyllanthionine 328 - - hygromycin A 428f
- - structures 327 - - kasugamycins 427
- uncommon -, in glycopeptide antibiotics 384ff - - minosaminomycin 428
amino hexitol components, structures 406 - - myomycin 428
amino pentitol components, structures 406 - - spectinomycins 427
&(L-cY-aminoadipyl)-L-cysteinyl-D-valinesynthet- - streptomycins 415ff
ase (ACVS) 254ff - target site interactions 449
aminochelin, catecholate siderophore of Azoto- - trehalosamines 437f
bacter vinelandii 203, 203 - trehazolin 439
- structure 203 - validamycins 438f
aminocyclitol aminoglycosides, producing bacterial aminopeptidase inhibitors, from basidiomycetes
genera 401 518ff
aminoglycoside-like compounds, structures 402 - - structure 519
aminoglycoside pathways, overview 448f amonabactins, catecholate siderophore of Aero-
aminoglycoside producers, resistance in - 44Off monas hydrophila 206
- - genes 440 - structure 206
aminoglycoside production, common gene pool ampicillin, effect in mouse septicemia 380
462 Amycolatopsis sp., production of dalbaheptides
aminoglycosides 397ff 380
- acarbose 438f amylostatin family, structure of - 472
- adverse effects in humans 451 anabasine, biosynthesis 625
- allosamidin 440 - structure 624
- as translation inhibitors 450 Anchusu officinalis, accumulation of rosmarinic
- bactericidal effects 450 acid 599f
- biochemical classification 400 ancovenin, primary structure 336f
- biosynthesis of - 409ff androgen receptor antagonists 116
- biosynthetic genes, in Streptomyces 447 androgens, inhibition of - 685
- bluensomycin 415f angiogenesis inhibitors 673ff
- carbohydrate precursors, from primary metabol- - structures 674
ism 459 anguibactin, catecholate siderophore of Vibrio un-
- compilation of 400 guillarum 206f
- decoding site of bacterial ribosomes 450 - structure 207
- deoxystreptamine-containing - 432ff animal fats, commercial markets 135
- - apramycins 436 - fatty acid composition 136
- - biosynthetic pathways 433 ansamitocin, inhibition of the microtubular sys-
- - destomycins 436 tem 668
- - 4,5-disubstituted - 434 - structure 668
- - 4,6-disubstituted - 434f anthocyanin, structure 610
- ecology 401 anthocyanins, from plant cell cultures 609f
- evolutionary aspects 457f anthracycline analogs 660
- fortimicins 429 anthracyclines, antitumor agents 652, 657ff
- functions of - 4opff - - structures 657ff
- genetics of 409ff - formation of oxygen radicals 661f
- glycosidase inhibitors 436f - pigmented antibiotics, as antitumor agents 660
- hexitol components, structures 406 - structures 34
- istamycins 429ff anthramycins, antitumor agents 656
- oligosaccharidic - 438f anthraquinones, anticancer drugs 652
- pathway engineering 459ff - biosynthetic pathway 602f
- pathway formula 400 - from plant cell cultures 601ff
- pentitol components, structures 406 - - yields 601
- producing bacteria 401 - structure 601
- regulation of production, in streptomycetes antibacterial agents, from basidiomycetes 501ff
443ff antibiotic export, regulatory cassettes, in strepto-
- resistance mechanisms in 44Off mycetes 92
Index 709
antibiotic formation, in strepomycetes, factors de- inhibitors of, cellular mitogenic signal transduc-
termining onset of - 86 tion pathway 675f
- pathway-specific activator genes 89f - glutathione S transferase 672
antibiotic production, effect of a two-component - mitosis 665
regulatory system 83 - protein kinases 676ff
- effect of metabolic imbalances 74 - protein phosphatases 676ff
- gene clusters 81 - ras-farnesyltransferase 683
- genes affecting also morphological differentia- - sexual hormone production 685
tion 84f - the microtubular system 665
- genetics 79ff - tyrosine protein kinases 679
- in streptomycetes, metabolite interference 72ff intercalating antibiotics, actinomycins 65Off
- multicopy inhibition of - 84 - anthracyclines 652
- onset of -, role of protein phosphorylation 81f - ellipticin 650
- organization of antibiotic biosynthetic genes - p-quinones 652
79ff - quinoxaline antibiotics 652
- regulation in Streptomyces, physiological as- - - structures 651f
pects 71ff - non-classical approaches 670
- regulation of - 57ff - nucleobase analogs 647f
- - plactams 65ff - nucleoside antibiotics 647f
- - effect of growth conditions 59f - radical generation 667

- screening assays, regulation of the cell cycle
- - in streptomycetes 70ff
- - in unicellular bacteria 61ff 647
- - peptide antibiotics 67ff - - use of transformed cells 647
- role of small diffusible signaling compounds - side effects, reduction of - 669
78f antifolates 648f
- stimulation by ppGpp 61ff, 75ff - structures 648
antibiotic regulatory genes, pleiotropic -, interac- antifungal agents, from basidiomycestes 50lff
tions 85ff antihelminthic compounds, from secondary
antibiotic resistance, in streptomycetes, induction metabolites 10
of - 92ff anti-HIV activity, of cyclosporins 540ff, 550f
antibiotics see also p-lactam antibiotics, see also antimetastasis drugs 673f
dalbaheptides, see also lantibiotics, see also pep- antimicrobial agents, compilation of 4ff
tide antibiotics antineoplastic agents see antitumor agents
- aminoglycosides, target site interactions 449 antitumor agents 641ff, see also anticancer drugs
- biotechnical drugs 643
- anthracycline type - 658
- classical - 647ff
- binding non-covalently to the DNA, structures - compilation of - 6ff
649
- forming covalent bonds with DNA 656ff
- evolutionary significance 60
- - anthracyclines 657
- intercalating - 650ff - - aziridines 655f
- plactam biosynthesis 247ff - - bleomycin 661, 663
- macromolecular antitumor - 663ff - - cyclopropane 655
- resistance mechanisms against -, producing mi- - - epoxide compounds 655f
crobes 442 - - streptonigrin 656
anticancer drugs see also antitumor agents - - structures 657
- angiogenesis inhibitors 673ff - from basidiomycetes 503ff
- - structures 674 - from plant cell cultures 614ff
- antifolates 648 - non-classical action on tumors 646
- antimetastatic inhibitors 673f - of marine organisms 685
- - structures 674 - plant-derived drugs 643
- baccatins, from the European yew tree 669 - potentiators of cytotoxicity 670ff
- binding non-covalently to the DNA 649f - - structures 671
- - structures 649 - prescreening for - 645ff
- DNA topoisomerase inhibitors, camptothecins - radical generation, by anthracyclines 661
655 - - by bleomycin 663
- - podophyllotoxins 655 - - by enedyines 664
- - structures 653f - showing selective toxicity 682ff
- immunomodulators 675f antitumor polysaccharides 508
710 Index
antiviral agents, from basidiomycetes 509ff, 520 Bacillus spp., peptide antibiotic production 67ff
- - structures 510f - - surfactin 67ff
antramycin, antitumor agent 662 Bacillus subtilis, stationary-phase functions 67
- structure 662 - subtilin production 331
apramycin family, structure of - 469 - surfactin production 67ff
apramycins 436 - surfactin synthesis, scheme 68
arachidonic acid (ARA), formation in, algae 161 bacteria, accumulation of PHB 178
- - microalgae 167f - affection by dalbaheptides 377ff
- - phycomycetes 161 - affection by lantibiotics 354ff
arbekacin, semisynthetic aminoglycosides 461 - aminocyclitol aminoglycoside-producing - 401
arbutin, production by biotransformation 622 - as producers of lantibiotics 326
Archaea, aminoglycoside-likeglycolipid structure - as sources of nonribosomal peptides 304ff, 308f
402 - as sources of ribosomal peptides 299
arginine-vasopressin receptor antagonists 121 - Plactam producing -, actinomycetes 250
Arthrobacter sp., arthrobactin 222 - - eubacteria 250
arthrobactin, siderophore of Arthrobacter sp. 222 - production of PUFA 147ff
ascidians, as sources of nonribosomal peptides - - eicosapentaenoic acid 147f
313 - regulation of antibiotic production 57ff
aspergillic acid, iron-binding compounds 233 - - in unicellular - 61ff
- structure 233 - resistance mechanisms in - 440ff
Aspergillus nidulans 25Off - signaling molecules, structures 14
- clustering of p-lactam biosynthetic genes 259ff bacterial siderophores, acinetobactin 219
- gene expression, regulation of - 266 - carboxylate siderophores 223ff
- penicillin biosynthetic cluster, regulation of 21 - - cepabactin 224
- penicillin biosynthetic enzymes 255ff - - rhizobactin DM4 223
asperlicin 117 - - staphyloferrin A 223
- activities 122 - - staphyloferrin B 224
- structure 119, 122 - - vibrioferrin 224
astaxanthin, structure 174 - catecholate siderophores 201ff
- use as feed for pen-reared salmonids 175 - - agrobactin 204
AT see acyl-coenzyme A: isopenicillin N acyltrans- - - aminochelin 203
ferase - - amonabactins 206
ATPase inhibitors 519f - - anguibactin 206f
atramycin, antitumor agent, structure 662 - - azotochelin 203
atrial natriuretic peptide receptor antagonists - - chrysobactin 202f
120f - - desferrithiocin 207
atromentic acid, iron-binding compounds 232 - - enterobactin 201f
- structure 232 - - ferrorosamine A 208
atromentin, iron-binding pigments 231 - - fulvibactin 205f
- structure 231 - - myxochelin A 203
Atropa belladonna, transgenic -, enhanced - - parabactin 204f
scopolamine production 625f - - protochelin 203
autoimmune diseases, treatment with cyclosporin - - pyochelin 207f
A 540 - - siderochelin A 208
aziridines, antitumor agents 6561 - - vibriobactin 205f
Azotobacter vinelandii, catecholate siderophores - - vulnibactin 205f
203 - - yersiniabactin 207f
Azotobacteriaceae, azotobactin 216 - citrate hydroxamates 220ff
azotobactin, chromophore 216 - - actinoferrin 221f
- peptide siderophore of Azotobacteriaceae 216 - - aerobactin 222f
azotochelin, catecholate siderophore of Azotobact- - - arthrobactin 222
er vinelandii 203 - - rhizobactin 1021 221
- structure 203 - - schizokinen 220f
- frankobactin 219
- hydroxamate siderophores 209ff
B - - albomycins 211f
baccatin 615 - - alcaligin 210f
baccatins, antitumor agents 667 - - bisucaberin 210f
Index 711
- - ferrimycins 211 bicyclic plactams see plactam antibiotics
- - ferrioxamines 209f biopreservatives, nisin 358
- keto hydroxy bidentates 225 biosurfactants 180f
- maduraferrin 219f biosynthetic pathways, of secondary metabolites
- mycobactins 217ff 31ff
- peptide siderophores 212ff biosynthetic strategies, formation of nucleoside
- - alterobacin 212 antibiotics 32
- - ferribactin 214 biotechnical drugs, as antitumor agents 641ff
- - ferrocins 212f biotransformations, of dalbaheptides, deacylation
- - pseudobactins 213ff 389
- - pyoverdins 213ff - of dalbaheptides 389f
bacteriocinkytolysin, structural gene 344 - - deglycosylation 389
basidiomycetes, aminopeptidase inhibitors 518ff - - glycosylation 389
- antibacterial metabolites 501 - with cultured plant cells 621ff
- antitumor compounds 503ff bisanthrene, antitumor agent, structure 651
- antitumor polysaccharides 508 bisucaberin, hydroxamate siderophore of Altero-
- antiviral compounds 509ff monas haloplanctis 21Of
- ATPase inhibiting metabolites 520f - structure 211
- cultivation of - 490f blennin, leukotriene biosynthesis inhibitor, struc-
- cytotoxic compounds 503ff ture 523
- herbicidal compounds 513ff bleomycin 661
- immunosuppressive compounds 509 - mode of antitumor action 663
- inhibitors of cholesterol biosynthesis 521f - structure 663
- inhibitors of leukotriene biosynthesis 522f blood group markers 403
- insecticidal compounds 515f bluensidine, biosynthetic pathway 417,421
- life cycle 490 bluensomycin 415f
- metabolic inhibitors of signal transduction 524 - biosynthetic pathway 416
- metabolites from - 489ff, 496ff boholmycin, streptomycin-related aminoglyco-
- - history 490 sides 428
- - oudemansis 496ff - structure 464
- - pleuromutilin 496 Bordetella pertussis, alcaligin 210f
- - screening of - 496 botryococcene 174ff
- - strobilurins 496ff - structure 174
- metabolites inducing differentiation of leukemia Botryococcus braunii, hydrocarbon production
cells 524 175f
- nematicidal compounds 515f bryostatins, inhibition of protein kinase C 677
- phospholipase inhibiting metabolites 518f butanolides, of Streptomyces virginiae, structure
- platelet aggregation inhibitors 512f 80
- primary and secondary metabolism, interrela- butirosin family, structure of - 467
tionship 492 butirosins 434
- producing, ondemansins 499 butterfat, fatty acid composition 136
- - strobilurins 499 butyl-methyl threonine, biosynthetic pathway 581
- reverse transcriptase inhibitors 509ff - polyketide synthase 581
- uses 490 - role of transaminases 581
beef tallow, fatty acid composition 136 y-butyrolactone autoregulators, synthesis by Strep-
benastatins, glutathione S transferase inhibitors tomyces sp. 446
672 ybutyrolactones, as regulatory signals 27ff
benzophenanthridine alkaloids, from plant cell - role in antibiotic production 78ff
cultures 603ff - structures 80f
- - yields 605
- structures 604
berberine, structure 604 C
betacyanin, structure 610 C-1027 chromophore, antitumor agent, structure
betacyanins, from plant cell cultures 611 666
betalains, from plant cell cultures 611 cadeguomycin, antitumor activity 647
betaxanthins, from plant cell cultures 611 caffeic acid, from plant cell cultures 598ff
bialaphos production, in Streptomyces hygroscopi- caffeoyl putrescine, from plant cell cultures,
cus, regulatory genes 91 yields 600
712 Index
- structure 599 - parabactin, of Micrococcus denitrijicans 204f

calcineurin 558f, 561f - protochelin 203
- phosphatase activity 5492 - pyochelin 207f
- role in T cell activation 549 - siderochelin A 208
calicheamicin PIBr, antitumor agent, structure - vibriobactin 205f
666 - vulnibactin 205f
caloporoside 518f - yersiniabactin 207f
- inhibition of, phosphoinositol metabolism 680 Catharanthus roseus, accumulation of monoter-
- phospholipase, inhibition of - 518 pene indole alkaloids 607ff
- structure 520 - - vinblastine 614f
calphostin, antitumor activity 679 - - vincristine 614
calyculin, inhibition of protein kinase C 677 - callus cultures 607
camptothecin-type compounds, as inhibitors of to- - indole alkaloids 665
poisomerase ll 655 - suspension cultures 608
cancer cells see also tumor cells CBE see cocoa butter equivalent
- cell cyle 644 cell cultures see plant cell cultures
cancer genes see oncogenes cell divison, phases 644
candicidin production, in Streptomyces griseus 71f cellular growth regulation, signal transduction
- - growth-phase dependency 71 pathway 644
- - phosphate repression 74 cepabactin, carboxylate-type siderophore of Pseu-
Candida bombicola, sophorolipid 180 domonas cepacia 224f
cantharanthine 607f - structure 224
- structure 607 cephaloridine, effect in mouse septicemia 380
carbapenem synthesis, by Erwinia carotovora, reg- cephalosporin biosynthesis 251f
ulation of - 65f - genes 251
- quorum regulation of - 66 - - cloning of - 259
carbon catabolite repression, global regulation 59 - - clustering of - 259ff
carboxylate siderophores 223ff - pathway 252
- cepabactin 224 cephalosporins, semi-synthetic - 264
- rhizobactin DM4 223 Cephalosporium acremonium 249ff
- staphyloferrin A 223f - gene expression, regulation of - 267
- staphyloferrin B 224 - penicillin biosynthetic enzymes 254ff
- vibrioferrin 224 - use of recombinant DNA technology, in p-lac-
cardiac glycosides 617 tam biosynthesis 2641
- production by biotransformation 622 cephamycin biosynthesis 252f
carnocin UI49, structure 336 - genes 251
pcarotene, formation in halophilic algae 175 - - cloning of - 259
- in fungi 174f - - clustering of - 259ff
- in microalgae 172ff ceramides, use in skin care formulations 184
- structure 173 chalciporon, antimicrobial activity 495
carotenoids, occurrence in microalgae 172ff chartreusin, structure 649
- structures 173f Chlorella minutissima, formation of eicosapenta-
caryophyllanes, structure 506 enoic acid 169
catabolite regulation, in microbial drug synthesis chloropeptin I, inhibition of HIV replication 125
11 chloropeptins 124
catabolite repression 59 - structures 126
catecholate siderophores, agrobactin, of Agrobac- cholecalciferol, structure 171
terium tumefaciens 204 cholecystokinin receptor antagonists 116
- aminochelin 203 cholesterol, structure 173
- anguibactin 206f cholesterol biosynthesis, inhibitors of -, from ba-
- azotochelin 203 sidiomycetes 518f
- chrysobactin 202f chrysobactin 202f
- desferrithiocin 207 - catechol-type siderophore 202f
- enterobactin 201f - structure 202
- ferrosamine A 208 cinnamic acid derivatives, from plant cell cultures
- fluvibactin 205f 598ff
- myxochelin A 203 cinnamycin, antibiotic activity 357f
- of Azotobacter vinelandii 203 - prepeptide structure 345
Index 713
- primary structure 336f cyclophilin B, complexed with a cyclosporin deri-
- solution structure 340 vative 570f
- structural gene 344 - - X-ray structure 570
citrate hydroxamates, acetinoferrin 221f cyclophilin C, complexed with cyclosporin A, X-
- aerobactin 222f ray structure 571
- arthrobactin 222 cyclophilins 5475 560ff
- rhizobactin 1021 221 - amino acid residues 560f
- schizokinen 220f cyclopropane, antitumor agent 655
- structure 220 cyclosporin A 538ff
clavicoronic acid 509f - 3D-structure, in apolar environment 561
- structure 510 - - in polar environment 562
clitocine 509f - effector site 548
- structure 510 - immunosuppressive properties 539ff
cocoa butter, fatty acid composition 136 - indications 539
cocoa butter equivalent (CBE) 150 - mode of immunosuppressive action 546ff
- costs of yeast processs 155 - - dual domain concept 546,548
- formation, from Cryprococcus mutants 153 - structure 538
- growth of Cryptococcus, oxygen limitation 154 - T cell selective immunosuppressant 539
- price 148 - X-ray structure 553
Coleus blumei, accumulation of rosmarinic acid cyclosporin A analogs, non-immunosuppressive -,
598ff antiviral activity 541, 551
collybial, antiviral activity 520 - - effect on multidrug resistant tumor cells 540,
- structure 522 551
complement C5a receptor antagonists 121 - SDZ NIM 811, structure 541
coprogens, structure 227 - SDZ PSC 833,541
- trihydroxamate siderophores 227 - SDZ214-103 580
Coptis japonica, accumulation of protoberberines - - structure 538
604f - - X-ray structure 562
coriolin, ATPase inhibiting metabolite 521 cyclosporin biosynthesis 572ff
- structure 521 - building units 573
cosmomycins, redifferentiation of cancer cells 682 - cofactors 573
crinipellins, cytotoxic activity 508 - cyclosporin synthetase 572ff
- structure 506 - mechanistic aspects 577ff
Cryptococcus curvatus 152ff - modified thiotemplate mechanism 578
- cocoa butter fatty acids 153 cyclosporin C, cleavage of peptide bonds 558
- fatty acyl composition 153 cyclosporin<yclophilin complexes 549f
CsA see cyclosporin A - crystal structures 565ff
cyanobacteria, as sources of nonribosomal pro- cyclosporin receptors 547f
teins 307f cyclosporin synthetase, characterization 573f
cyclitols 454f - enzymatic activities 572f
- basic pathways 402ff - gene 573f
- benzoic acid derivatives, structures 407 - - manipulation by integrative transformation
- biogenesis 402ff 574ff
- components of nucleoside-type antibiotics 454 - - structure 575
- dehydroquinate pathway 403ff - theoretical reprogramming 576
- myo-inositol pathway 403ff cyclosporin variants, biosynthesis 580
- sugar components, attachment of - 405ff cyclosporins 535ff
cycloleucomelone, iron-binding compounds 232 - activity against tumor multidrug resistance 540,
- structure 232 551
cyclophilin A, complexed with cyclosporin A, 3D- - analogs 294f
structure 566 - anti-HIV activity 540ff, 550f
- complexed with cyclosporin derivatives 568ff - antiinflammatory effects 550
- - X-ray structures 568 - antimalarial activity 551
- 3D-structure, complexed with a tetrapeptide - binding to cyclophilins 547ff, 558f
563f - - effects of valine for leucine substitutions 560
- peptidyl-prolyl isomerase active site 564ff - chemical synthesis 552ff
- target of cyclosporin A 560 - constituent amino acids 551f
714 Index
- cyclic peptide backbone, chemical transforma- - lipoglycopeptides, naturally occurring - 375

tions of - 555 - physicochemical properties 376f
- fermentation 552ff - - isoelectric points 376
- immunosuppressive activity 539ff - - reverse-phase HPLC retention times 377
- - mechanism 546ff - producing organisms 372,374f, 380
- - relation to structure 556ff - resistance to - 379
- incorporating non-proteinogenic amino acids - ristocetin type, naturally occurring - 372
556 - screening methods 381
- macrocycle, selective ring opening 555 - uncommon amino acids in - 384ff
- mode of immunosuppressive action 542ff - vancomycin type, naturally occurring - 374
- resistance to systemic proteases 551 daunomycin, antitumor agents 652
- structural investigations 561ff daunorubicin, antitumor agent, structure 658
- structure-activity relationships 557ff daunorubicin production, in Streptomyces peuceti-
- structures in polar and non-polar solutions 552 cus, regulatory genes 91
- treatment of autoimmune diseases 540 deacylation, of dalbaheptides 389
- use in transplant rejection therapy 539f deglycosylation, of dalbaheptides 389
- variety of conformations 571 7-dehydrocholesterol, structure 171
cyclovariegatin, iron-binding compounds 232 dehydroquinate pathway 403ff
- structure 232 dehydroquinate synthase, substrates 405
cytodifferentiation, in Streptomyces, A-factor 28 6-deoxyhexoses 410ff
- - signal cascade 27 - as moieties of antibiotics 456
- role of secondary metabolism 16 - biosynthesis, general pathway 408
cytokine receptors 545 - biosynthetic pathways, gene clusters 455
cytokines, release upon T cell activation 545f - structural variants, in secondary metabolites
cytolysinlbacteriocin 335f 410ff
- primary structure 335 - - Occurrence in natural products 414
cytotoxic agents, from basidiomycetes 503ff deoxyilludin M,antitumor activity 503
cytotoxic metabolites, marine organisms 687 - structure 504
desferrithiocin, orally available iron chelator 207
destomycin family, structure of - 469
D destomycins 436
dalbaheptide fermentation, feedback inhibition of devazepide, activities 122
- 382f - structure 122
- media 381f devazepide receptor antagonist 121
- - carbon sources 381f DHA see docosahexaenoic acid
- - effect of phosphates 382 DHGLA see dihomo-ylinolenic acid
- - nitrogen sources 382 diazaquinomycin B, thymidylate synthase inhibi-
- product recovery 383 tion 648
- purification 383 2,3-didehydroalanine 328ff
dalbaheptides 369ff 2,3 didehydrobutyrine 328ff
- actinoidin type, naturally occurring - 375 3,6-dideoxyhexosepathway, of gram-negative bac-
- biological activity 377ff teria 414
- - in vitro antibacterial activity 377f Digitalis lanata 621ff
- - in vivo efficacy 379f - tissue culture process 621
- - mechanism of action 378f digitoxin, production by biotransformation 622
- - pharmacology 379f digoxin, production by biotransformation 622
- biosynthesis 384ff dihomo-ylinolenic acid (DHGLA), as precursor
- biotransformation of - 389f of prostaglandins 160
- - deacylation 389 - formation by deletion of desaturase 160
- - deglycosylation 389 dihydrostreptose, biosynthetic pathway 422f
- - glycosylation 389 dioxamycin, antitumor agent, structure 662
- chemical modifications of - 387f Dipodascopsis uninucleata, formation of prosta-
- definition 371 noid-type lipids 184f
- descriptive chemistry 371ff distamycin, antitumor agent 650
- effect in mouse septicemia 380 - structure 649
- effect on pathogens 378 diterpenoids, Occurrence in basidiomycetes 493
- glycolysation of - 387 DNA replication, inhibition of topoisomerases
- glycopeptide skeleton 372 653
Index 715
DNA synthesis, inhibition by antitumor agents epidermin, C-terminus, aminovinyl cysteine resi-
655 due 349
DNA topoisomerase inhibitors 654ff - medical uses 359
- camptothecins 655 - prepeptide structure 345
- podophyllotoxins 655 - primary structure 332ff
- structures 654f epidermin biosynthesis, from pre-epidermin, gene
DNA topoisomerases, as targets of antitumor cluster 342
drugs 653ff - - scheme 342
docetaxel, inhibition of the microtubular system - gene cluster 348
668 - regulation of - 351
- structure 668 epilancin K7, primary structure 332ff
docosahexaenoic acid (DHA), in bacteria 147f epirubicin, antitumor agent, structure 658
- in microalgae 16% epoxide compounds, antitumor agents 656f
- in phytoplankton species 169f ergocalciferol, structure 173
- nutritional role 162 ergosterol 171f
- occurrence in marine fungi 163 - structure 171
dopamine receptor antagonists 114 ergotamine, as receptor antagonist 112
doxorubicin 651f Erwinia carotovora, carbapenem production, regu-
- antitumor agent 651f, 658f lation of - 65
- structure 651, 658 Erwinia rhapontici, ferrorosamine A 208
drugs see microbial drugs erythromycin, effect in mouse septicemia 380
Dunaliella salina, &carotene content 173 erythromycin A 122ff
duocarmycins, antitumor agents 657f - activities 124
duramycins, antibiotic activity 357f - structure 123
- primary structure 336f Escherichia coli, carbon catabolite-repressible ope-
- solution structure, stereorepresentation of - rons 60
340 - microcin C7 synthesis 61ff
dynemicin A 664ff - ppGpp 61ff
- antitumor agent 665 - - synthesis 63
- structure 666 - RNA polymerase sigma factor 61ff
- antitumor activity 664 - - regulation of - 64
esperamicin A, 664ff
- antitumor agent 666
E - enrichment pattern 667
echinomycin 651f - structure 666
- antitumor agent 651f estrogen receptor agonists 112, 123
- structure 651 estrogen receptor antagonists llSf, 686
eicosapentaenoic acid (EPA), in bacteria 147f estrogens, inhibition of -, treatment of hormone-
- in microalgae 168f dependent tumors 685
- physiological effects 162 ETA see eicosatrienoic acid
- precursor of prostaglandins 162 ethanol, conversion to lipid 142
eicosatrienoic acid (ETA), occurrence in marine ether lipids, archaebacterial - 181ff
fungi 163 - - structures 182
ellipticin 650 eukaryotes, DNA transfer to prokaryotes 263
endothelin receptor antagonists 116ff Euphorbia milli, anthocyanin production 610
enedyines, cytotoxic antitumor agents 664ff exochelins 217f
- macromolecular antitumor antibiotics 663ff - structure 218
- mode of action 664ff
- structures 666
enterobactin 201f F
- catecholate siderophore 201f u factors 61ff
- of enteric bacteria 202 - initiation of transcription 62
- structure 201 - regulation of - 64
enzyme inhibitors, of the growth-regulating path- - structure 62
way 646 fasciculols, plant growth inhibitors, structures 515
EPA see eicosapentaenoic acid fatty acids 146ff
epelmycin A, antitumor agent 659 - esterification 146
- structure 659 - in bacteria, polyunsaturated fatty acids 147f
716 Index
- in fungi, polyunsaturated fatty acids 151 - ferrichromes 225f

- in lipoglycopeptides 386f - fusarinines 228f
- in microalgae, arachidonic acid (ARA) 167f - rhizoferrins 229f
- - docosahexaenoic acid (DHA) 169f - rhodotorulic adid 228
- - eicosapentaenoic acid (EPA) 168f fungi see also oleaginous molds
- - ylinolenic acid (GLA) 167 - as sources of nonribosomal peptides 309
- in molds, arachidonic acid (ARA) 161f - p-lactam producing -, actinomycetes 250
- - dihomo-y-linolenic acid (DHGLA) 160 - - eubacteria 250
- - docosahexaenoic acid (DHA) 162f - formation of PUFA 151,158
- - eicosapentaenoic acid (EPA) 162 - method of resistance to their own fungicide
- - eicosatrienoic acid (ETA) 163 501
- - ylinolenic acid (GLA) 159f - occurrence of $-carotene 174f
- - polyunsaturated fatty acids 158 - oil product containing y-linolenic acid 158
- structures 139 - producing, oudemansins 499
ferrosamine A, catecholate siderophore of Envinia - - strobilurins 499
rhapontici 208 - signaling molecules 14
ferribactin, peptide siderophore of Pseudomonas - transformation systems, for $-lactam produc-
fluorescens, non-fluorescent - 214 tion 253
ferrichromes, fungal siderophores 225f fursarinines 228f
- structure 225 - structure 228
ferric uptake regulation proteins 200
ferrimycins 211
ferrioxamines, hydroxamate siderophores of Strep- G
tomyces pilosus 209f gallidermin, medical uses 359
- structures 209 - primary structure 333
ferrocins 212f - solution structure, stereorepresentation of -
- peptide siderophore of Pseudomonas fluores- 339
cens 213 gearbox promoters, expression at low growth
- structure 212 rates 64
fibrinogen receptor antagonists 115 gene clusters, biosynthetic - 17ff
fimicolon 502f - - analysis of - 20ff
- cytotoxic activity 503 - - listing 18f
- structure 502 - - organization of - 21
FK506 546ff - in streptomycin production 420
- immunosuppressive properties 546f gene products, in streptomycin production 418f
- structure 546 genes, biosynthetic -, identification of - 20
FK506 binding proteins 547f genetic exchanges, between microbial species 15
flavidulols, structures 509 gentamicin family, structure of - 468
fluvibactin 205f gentamicins, biosynthetic pathway 433
- catecholate siderophore of Vibrio fluvialis 206 ginseng, from plant cell cultures 611
- structure 205 GLA see y-linolenic acid
folic acid metabolism, inhibition by thimidylate glutarimde antibiotics, redifferentiation of cancer
synthase 649 cells 683
fomannosin, plant growth inhibitor 514 glutathione S transferase inhibitors, structures
- structure 514 673
formycin production, by Streptomyces lavendulae, glycopeptide antibiotics see dalbaheptides
effect of ppGpp 77 glycosyl phophatidylinositols, resemblance to ami-
fortimicin family, structure of - 465 noglycosides 402
fortimicins 429ff - structure 402
- biosynthetic pathway 429 glycosylation, of dalbaheptides 389
- enzymes, involved in production of - 431 glyoxylate cycle, key enzymes, inhibitors 513
- gene products, involved in production of - 431 gp120-CD4 binding inhibitors, structures 125ff
frankobactin, structure 219 groundnut oil, genetic engineering of - 138
frogs, as sources of ribosomal peptides 301
fucoxanthin, structure 173
fulvoferruginin, structure 505 H
fungal siderophores 225ff hairy root cultures, fermentation techniques 620
- coprogens 227 - production of tropane alkaloids 619
index 717
- scopolamine secretion 620 - antitumor activity 503
- secondary product formation 619f - biosynthesis 492
- transformation with Agrobacterium rhizogenes - toxic principle 493
719 immunomodulators, antitumor effects of - 673f
Hansenula ciferri, formation of ceramides 184 - structures 674
hatomarubigins, inhibitors of multi-drug resistant immunophilins 547%559ff
tumor cells 672 - 3D-structure 561f
hemimycin 502f immunosuppressant macrolides, model of the
- cytotoxic activity 503 mode of action 548
- structure 502 immunosuppressants, of microbial origin, dual do-
heptapeptide structure, of dalbaheptides 372 main concept 546
herbicidal compounds, from basidiomycetes 513ff immunosuppression, by cyclosporins 539ff
- - structures 514f - - mode of action 542ff
herbimycin, immunostimulation 675 - - side effects 540
- structure 686 - - use in transplantation surgery 539f
herbimycin A, angiogenesis inhibitor 674 - by FK506 546ff
- contact inhibition of tumor cells 680 - by rapamycin 546ff
- phosphoinositol kinase inhibiton 681 - drug-immunophilin complexes 549f
hexosamine pathway 423 immunosuppressive agents, from basidiomycetes,
hexosamines, as moieties of antibiotics 457 structures 509
hexoses, amination of - 415 immunosuppressors, from secondary metabolites
hispidospermidins 680 7
HIV see human immunodeficiency virus indolocarbazole compounds, antitumor activity
homoserine lactones 62f 679
horizonal transfer hypothesis, @-lactam biosynthet- - bactericidal activity 679
ic pathway 262f - fungicidal activity 679
hormone-dependent tumors, treatment of - 685 insecticidal compounds, from basidiomycetes 515f
human immunodeficiency virus (HIV), anti-HIV - - structures 516
activity assay 124 insects, as sources of ribosomal peptides 300
- blocking of HIV entry 124 intercalating drugs 650ff
- HIV disease, use of cyclosporins 54Off, 55Of - actinomycins 65Off
- inhibition of replication 125 - ellipticin 650
hydromycin, antitumor agent, structure 662 - structures 651
hydroxamate siderophores 209ff interspecific communication 13, 15
- albomycins 211f IPNS see isopenicillin N synthase
- alcaligin 21Of iron nutrition, role of siderophores 235
- bisucaberin 210f iron transport, siderophore-mediated systems
- ferrimycins 211 233ff
- ferrioxamines 209f - - in enterobacteria 234
hydroxycinnamoyl putrescines 600f iron uptake 2OOff
7-hydroxyguanine, antitumor activity 647 isepamicin, semisynthetic aminoglycosides 461
p-hydroxyphenylglycine, biosynthesis 385 isochalciporon, antimicrobial activity 495
hygromycin, structure 464 isochromophilones 124ff
hygromycin A, streptomycin-related aminoglyco- - inhibition of HIV replication 125
sides 428 - structures 125ff
hygromycin family, structure of - 469 isopenicillin N synthase (IPNS) 256f
hyoscyamine, from plant cell cultures 613 isoprenoid lipids, structures 182
hyphodontal 511f isovelleral, antimicrobial activity 495
- structure 511 isovellerol, antimicrobial activity 495
hypnophilin, cytotoxic activity 507 istamycin family, structure of - 465
- structure 505 istamycins 429ff
- biosynthetic pathway 429
I
ibotenic acid, insecticidal compound 515f K
- structure 516 kanamycin family, structure of - 467
idarubicin, antitumor agent, structure 658 kanamycins, biosynthetic pathway 433
illudins 491ff kasugamycin, structure 464
718 Index
kasugamycins, streptomycin-related aminoglyco- - - veterinary - 359

sides 427 - biological activities 353ff
keto hydroxy bidentates 225 - carnocin U149 336
krestin, antitumor polysaccharide 509 - chemical characteristics 326
kuehneromycins 511f - chemistry of - 325ff
- structure 511 - cinnamycin 336f, 340, 357
- compilation of - 326f
- cytolysinlbacteriocin 335f
L - definition 325
Plactam antibiotics 247ff - disruption of energy transduction 355
- biosynthesis of carbapenem 65 - duramycins 336f, 340, 357
- biosynthetic clusters 21 - effect on cytoplasmic membrane of susceptible
- biosynthetic enzymes 253ff bacteria 354
- - compartmentalization of - 261f - epidermin 332ff
- biosynthetic genes 251 - epilancin K7 332ff
- - cloning of - 259 - gallidermin 333, 339
- - clustering of - 259ff - lacticin 481 332, 335
- - DNA sequences 260 - lactocin S 332, 335
- - orientation of transcription 261 - leader peptidases 348ff
- - regulation of gene expression 265ff - mersacidin 3371 357
- biosynthetic pathway, cephalosporin C 252 - modified amino acids in - 327ff, 347f
- - evolution of - 262f - - D-alanine 330
- - penicillin G 252 - - didehydroalanine 328
- genetic transformation systems, for fungi 253 - - didehydrobutyrine 328
- - for prokaryotes 253 - - lanthionine 328
- history 249f - - methyllantionine 328
- penicillin biosynthesis, enzymes 253ff - - structures 327
- producing organisms 250,253 - mutacin 336
- use of recombinant DNA technology 264f - nisin 330ff, 339
lacticin 481, primary structure 332 - non-disulfide cyclic peptides 281
- structure 335 - Pep 5 332ff, 340
lactocin S, prepeptide structure 345 - physical characteristics 326
- primary structure 332,335 - presursor peptide 328
- structural gene 344 - producing organisms 326
lactococcin, prepeptide structure 345 - ribosomal biosynthesis 325
- structural gene 344 - salivaricin A 332, 335
Lactococcus lactis, nisin production 330 - streptococcin A-FF22 334
lanosterol, structure 171 - subtilin 331ff, 339
lanthionine, biosynthesis 328f - transport proteins, genes 348
- biosynthetic pathway 329 - type A, induced pore formation 356
- non-protein amino acid 347 - - induction of autolysis 356
- structure 327 - - modes of action 353ff
lantibiotic biosynthesis, epidermin 342 - - primary structures 330ff
- genetic regulation 341ff - - solution structures 339f
- leader peptide, role of - 346f - type B, mode of action 357f
- modification events, sequence of - 353 - - primary structures 336ff
- nisin, gene clusters 343 - - role in human immune and blood pressure
- regulation, by proteins 350f regulatory systems 336
- structural genes 341ff - - solution structures 340f
lantibiotic prepeptides, leader peptide 345ff lantibiotic synthesizing cells, self-protection mech-
- primary structure 345 anisms 351ff
- structural genes, transcription of - 341ff leader peptidases 348ff
]antibiotics 323ff see also antibiotics leader peptides, in lantibiotic biosynthesis 345ff
- actagardine 337f, 357 - - conserved motifs 346f
- ancovenin 336f - - roles 346f
- applications of - 358f leaianafulvene, cytotoxic activity 508
- - as foodlbeverage preservatives 358 - structure 506
- - medicaUparamedica1- 359 lentinan 508
Index 719
lentinellic acid, antibacterial activity 501 - oligopeptide siderophore of Actinomadura ma-
- structure 502 durae 21%
leucomelone, iron-binding pigments 231 - structure 220
- structure 231 Magnus test, screening of receptor-active com-
leukotriene B4 receptor antagonists 114 pounds 109f
leukotriene biosynthesis, inhibitors of -, from ba- mammals, as sources of ribosomal peptides 302f
sidiomycetes 522f marasmic acid 503f
ligand receptors, antagonists 117ff - antitumor activity 503
- - producers 117 - structure 504
ligands, radio-labeled, radioimmunoassay 110 maytansine, inhibition of the microtubular system
- - screening of receptor-active compounds 668
110 - structure 668
light emission, by the autoinducer of Vi6rio Mead acid see eicosatrienoic acid
jischeri 66 medermycin, antitumor agent, structure 662
lincomycin A, biosynthetic gene, “sugar” subclus- menoxymycin, antitumor agent, structure 662
ter 453 mersacidin, action on cell wall of susceptible bac-
- hypothetical biosynthetic pathway 452 teria 357
lincosamides 451ff - prepeptide structure 345
- structure 473 - primary structure 337f
- sugar containing moieties 451 - structural gene 344
ylinolenic acid (GLA), in microalgae 167 merulidial 504f
- medicinal uses 159 - structure 504
- production by phycomyceters 159 metabolic engineering, of secondary pathways, ni-
linseed oil, genetic engineering of - 138 cotine biosynthesis 624f
lipid accumulation, biochemistry 143ff - - scopolamine production 625f
- culture conditions 140ff - - serotonin biosynthesis 624
- effect of carbon-to-nitrogen ratio 141 - - use of plant cell cultures 623ff
- efficiency 142ff metabolism, secondary see secondary metabolism
- - in molds 142f metastasis, prevention of - 673f
- - in yeasts 142f methylenomycin biosynthesis, by Streptomyces
- in continuous culture 141 coelicolor 74f
- patterns 139ff methylenomycin production, in Streptomyces coeli-
- - in molds 140 color, negative regulation of - 91f
- - in yeasts 140ff N-methyl-L-glucosamine, biosynthetic pathway
- yields, on carbohydrates 142 424
- - on ethanol 142 methyllantionine 328
lipids 133ff microalgae see also oleaginous microalgae
- biosurfactants 180f - carotenoid contents 172ff
- classification 138f microbial drugs 2ff
- ether lipids l8lff - catabolite regulation 11
- fatty acid profile, in yeasts 148f - phase-dependency of biosynthesis 11
- - in molds 157f microbial lipids see lipids
- nomenclature 138f microbial siderophores see siderophores
- phospholipids 183f microcin C7, synthesis in E. coli 61ff
- prostanoid-type - 184f microcins 61
- sphingolipids 183 microcystin, inhibition of protein kinase C 677
lipoglycopeptides, fatty acids, origin 386 minosaminomycin, streptomycin-related aminogly-
- naturally occurring - 375 cosides 428
- producing organisms 380 - structure 464
Lithospermum erythrorhizon, accumulation of shi- mitogenic signal 644
konins 601f mitomycin C, antitumor agent 655
mitosis, cellular signal transduction pathway 677
mitoxantron, antitumor agent, structure 651
M mniopetals 510ff
macrolides, immunosuppressant -, model of the - structures 511
mode of action 548 molds see oleaginous molds
- structures 34 molluscs, as sources of nonribosomal peptides 314
maduraferrin 219f - as sources of ribosomal peptides 301
720 Index
monoterpene indole alkaloids, biosynthetic path- Nicotiana tabacum, production of caffeoyl putres-
way 609 cine 599ff
- from plant cell cultures 606ff nicotine, biosynthesis 624f
Morinda citrifolia, accumulation of anthraqui- nisin, prepeptide structure 345
nones 603 - primary structure 33Off
morphinan alkaloids, from plant cell cultures 613 - solution structure, in aqueous solution 339
morphological differentiation, genes affecting also - - in lipophilic solvents 339
antibiotic production 84f - structural gene 342
Mortierella alpina, formation of arachidonic acid - use in foods and beverages 358
161 nisin biosynthesis, gene clusters 343
- formation of eicosapentaenoic acid 162 NMDA receptor antagonists 114
motilides, receptor agonists 122f Nocardia lactamdurans, p-lactam biosynthesis 251
mucidin, biosynthesis 500 - clustering of p-lactam biosynthetic genes 259ff
- structure 501 Nocardia sp., siderochelin A 208
Mucor circinelloides, fatty acid profiles 158 nogalamycin, antitumor agent, structure 658
multidrug resistance, of tumor cells 540, 551, 672 nojirimycin, biosynthetic pathway 437
muscarine 112, 123 novobiocin resistance, in streptomycetes, induction
muscarinic acetylcholine receptor, agonists 112 of - 93
- antagonists 112ff nucleobase analogs 647f
muscimol, insecticidal compound 515f - structure 648
- structure 516 nucleoside antibiotics 647f
mushrooms see basidiomycetes - biosynthetic strategies of formation 32
mutacin, structure 336 - structural class 40
mycelianamide, iron-binding compounds 233 - structure 648
- structure 233
Mycobacterium sp., mycobactins 217
mycobactins 217ff 0
- function as iron shuttle 217 OHHL, of Vibrio ficheri 65f
- structure 218 - - structure 65
myo-inositol pathway 403ff - regulators 65f
myo-inositol-~-3-phosphatesynthase, substrates okadaic acid, inhibition of protein kinase C 677
405 oleaginous microalgae 164ff
myornycin, streptomycin-related aminoglycosides - as sources of eicosapentaenoic acid 169
428 - cultivation of - 164
myomycins, structure 464 - fatty acid profiles 165f
myxochelin A, structure 203 - fatty acids, arachidonic acid (ARA) 167f
- - docosahexaenoic acid (DHA) 169f
- - eicosapentaenoic acid (EPA) 168f
N - - y-linolenic acid (GLA) 167
nannochelins 223 - lipid contents 165
naphthoquinones, from plant cell cultures 601f oleaginous microorganisms 140ff
- - yields 601 - pathway of triacylglycerol formation 143ff
neamine 434 - - scheme 144
nematicidal compounds, from basidiomycetes oleaginous molds 155ff
515f. 521 - fatty acid profiles 156ff
- - structures 516 - fatty acids, arachidonic acid 161f
neomycin, biogenesis of components 435 - - dihomo-y-linolenic acid (DHGLA) 160
neomycin family, structure of - 466 - - docosahexaenoic acid (DHA) 162f
neomycins, biosynthetic pathway 433 - - eicosapentaenoic acid (EPA) 162
neoplanocin, antitumor activity 647 - - eicosatrienoic acid (ETA) 163
neoplasm inhibitors see anticancer drugs - - y-linolenic acid (GLA) 159f
neoplastic cell lines, use in screening for antican- - formation of PUFA 158
cer drugs 645ff - lipid accumulation 140
netropsin, antitumor agents 650 - - efficiency 142f
- structure 649 - lipid analyses 147
neurokinin A, receptor antagonists 120 - lipid contents 156
neuropeptides, receptor angatonists 120 - triacylglycerol storage 146
Index 721
oleaginous yeasts, cocoa butter equivalent (CBE) penicillin biosynthesis 119f
148. 150 - enzymes 253ff
- - process costs 155 - - ACVS 254
- conversions of fatty acids 150 - - AT 257ff
- fatty acid profile of lipids 148f - - IPNS 256
- increase of stearic acid content 150ff - gene dosage 264
- - by direct feeding 150 - genes 251
- - by inhibition of stearoyl desaturase 151f - - cloning of - 259
- - by metabolic manipulation 154 - - clustering of - 259ff
- - by mutation 152ff - pathway 252
- lipid accumulation 14Off - - compartmentalization of - 262
- - efficiency 142f - regulation of gene expression 264
- lipid analyses 147 - side chain exchange 257
- lipid contents 149 penicillin fermentation 292
- triacylglycerol storage 146 Penicillium chrysogenum 249ff
oleandomycin formation, by Streptomyces antibio- - clustering of p-lactam biosynthetic genes 259ff
ticus, growth-phase dependency 72 - gene expression, regulation of - 266
oligoglycosides, structural class 40 - penicillin biosynthetic enzymes, AT 257ff
oligopeptides, biosynthesis of - 36 pentosamines, as moieties of antibiotics 456
omphalotin, nematicidal activity 522 pentoses, amination of - 415
- structure 522 pentostatin, inhibition of adenosine deaminase
oncogenes 645 648
- classification 676 peptide antibiotics 67ff, 277ff
oncogene-transformed cell lines, availability 682 - biosynthesis 284ff
- substances changing the morphology of - 682ff - biotechnology of - 284ff
ornibactins, structure 216 - directed biosynthesis 296
- tetrapeptide siderophores of Pseudomonas cepa- - future prospects 297f
cia-like strains 216f - nonribosomal origin 2791
oudemansins 496ff - production by Bacillus sp. 67
- blockage of respiration in eukaryotes 497 - reprogramming, genetic approaches 297
- fungistatic activity 497 - ribosomal origin 278f
- mode of action 497ff - side chain modifications 282
- possible functions 501 - structures 278ff
- producing organisms 499 peptide biosynthesis 284ff
- structures 498 - biosynthetic modules 286ff
oxanosin, antitumor activity 647 - fermentation procedures 292,294
oxytocin receptor antagonists 121 - gene clusters 286f
- - compilation of - 287f
- mechanisms 285
P - multienzyme systems 292
paclitaxel 665,667 - - domains 290
- inhibition of the microtubular system 668 - nonribosomal system 285
- structure 668 - ribosomal system 285
Panax ginseng, saponin aglycone, structure 611 - structural alterations 294
panepoxydone, inhibition of signal transduction - tripeptide model system 291
524 peptide families, analogs 294f
panudial, antithrombotic compounds, structure - - aureobasidin 295
512 - - cyclosporin 295
Papaver somniferum, accumulation of benzophen- - - gramicidin 295
anthridine alkaloids 605f - - gramicidin S 295
parabactin 204f peptide ligand receptors, antagonists of - 116ff
- catecholate siderophore of Micrococcus denitri- - - structures 118ff
ficans 204f peptide siderophores 212ff
- structure 204 - alterobactin 212
Paracoccus denitrificans, parabactin 204f - azotobactin 216
paromomycins, biosynthetic pathway 433 - ferribactin 214
Peganum harmala, serotonin biosynthesis 624 - ferrocins 212f
722 Index
- ornibactins 216f peptolide SDZ, 3D-structure, in apolar environ-

- pseudobactins 213ff ment 562
- pyoverdins 213f peptolide SDZ 214-103 synthetase 580
peptide synthetases 286ff Pep$ biosynthetic pathway 334
- adenylate forming domains 286 - prepeptide structure 345
- branched cyclopeptides 293 - primary structure 332ff
- carrier domains 289 - structural gene 344
- compilation of - 293 Phafia rhodozyma, astaxanthin production 175
- condensation domains 290 PHB see poly-/3-hydroxybutyrate
- consensus sequences 289 phellodonic acid, structure 505
- cyclodepsipeptides 293 phenoxazinone synthase, effect of glucose on ac-
- cyclopeptides 293 tivity 73
- domain construction 288 phopholipids, yeast “lecithin”, extraction of spent
- epimerization domains 290 brewer’s yeast 183
- lactones 293 phorbol, structure 678
- linear peptides 293 phorbol esters, activation of protein kinase C 677
- N-methylation domains 29Of - tumor promoters 677
- thioesterase domains 291 phosphilipids 183f
peptides, branched cyclic - 293 phosphoinositol kinase inhibitors 681
- compilation of - 299ff, 304ff phosphoinositol metabolism, inhibitors of - 680ff
- cyclic - 279, 293 - - structures 681
- cyclodepsipeptides 293 phospholipase inhibitors, from basidiomycetes
- formed by in vitro synthesis 296f 518f
- gene-encoded - 283f phospholipids, medicinal uses 184
- lactones 293 - structures 139
- linear 283,293 Phycomyces blakesleeanus, p-carotene production
- monocyclic- 279f 175
- multicyclic - 282f pilatin 504f
- of nonribosomal origin 279ff, 284 - structure 504
- - from actinomycetes 304ff pinicoloform, inhibition of differentiation of leu-
- - from ascidians 313 kemia cells 524
- - from bacteria 304, 308f plant cell cultures 604
- - from cyanobacteria 307f - biotechnological importance 596ff
- biotransformations 621ff
- - from fungi 309f
- - from molluscs 314 - commercial interest in - 595
- expression of pathways 603
- - from plants 311f
- metabolic engineering 623ff
- - from scorpions 314 - of transgenic plants 623f
- - from sea hares 314
- - from spiders 314
- production of secondary metabolites 593ff
- regulation of secondary pathways 623
- - from sponges 312f
- - future prospects 298
- technology 596
plant cell suspension cultures, secondary product
- of ribosomal origin 278ff formation 598ff
- - D-amino acids 281 anthocyanins 609f
- - features 284 anthraquinones 601ff
- - fromfrogs 301f antitumor compounds 614ff
- - from insects 300 benzophenanthridine alkaloids 603ff
- - from mammals 302f betalains 611
- - from microbial sources 299 cardiac glycosides 617
- - from molluscs 301 cinnamic acid derivatives 598ff
- - from plants 299 immunologically active polysaccharides 612
- - from scorpions 301 monoterpene indole alkaloids 606ff
- - from snakes 302 morphinan alkaloids 613
- - from spiders 301 naphthoquinones 601f
- - future prospects 297 protoberberine alkaloids 603ff
- - production of - 283 quinoline alkaloids 614
- - structures 278ff tropane alkaloids 613f
peptidic drugs, structural class 39 vanillin 617f
Index 723
plant growth regulators, from secondary metabol- - in fungi 151, 158
ites 9 - in microalgae 165ff
plant oils, commercial markets 135 - nutritional roles 164
- fatty acid composition 136 Porphyridium curenturn, formation of polyunsatu-
- fatty acid profiles 158 rated fatty acids 168
- modification by genetic engineering 138 ppGpp 61ff, 75ff
plant tissue cultures, expression of pathways 597 ppGpp synthesis, activation of - 63
- novel pharmacologically active compounds 621 pre-epidermin 342
- secondary metabolites 595 pre-lantibiotics see lantibiotic prepeptides
- technology 596 pre-nisin 341
plants, as sources of nonribosomal peptides 311f proferrorosamine A, structure 208
- as sources of ribosomal peptides 299 prokaryotes, DNA transfer to eukarvotes 263
- secondary metabolites, from cell cultures 595ff 1 transformation systems, for plactam produc-
- - uses 595 tion 253
platelet activating factor receptor antagonists protease inhibitors, microbial -, antitumor effects
114f 675
platelet aggregation, screening of receptor-active protein kinase inhibitors, antitumor agents 677ff
compounds 110 - structures 678
platelet aggregation inhibitors, from basidiomy- protein phosphatase inhibitors, antitumor agents
cetes 512f 677ff
- - structures 512 - structures 678
pleuromutilin, antibiotic action 496 protein phosphorylation, role in antibiotic produc-
- production by fermentation of Pleurotus 496 tion 81f
- structure 495 protoberberine alkaloids, from plant cell cultures
pleurotellic acid, structure 505 603ff
pleurotellol, cytotoxic activity 507 - structures 604
- structure 505 protochelin, catecholate siderophore of Azotobact-
Pleurotus sp., pleuromutilin production 496 er vinelandii 203
podophyllotoxins 617
- as inhibitors of topoisomerase 11 655 - structure 203
protooncogene induction, inhibition of - 682
podoscyphic acid, structure 510
poly-b-hydroxy alkanoates 177ff pseudobactins 213ff
- peptide siderophore of Pseudomonas jluores-
- Occurrence in Pseudomonas oleovorans 180 cens 213
poly-phydroxybutyrate (PHB), accumulation in
bacteria 178 - structures 213, 215
- biosynthetic pathway 178 Pseudomonas aeruginosa, pyochelin 207f
- structure 177 - rhamnolipids 180f
- synthesis by condensation of acetyl-CoA units Pseudomonas cepacia, cepabactin 224
179 Pseudomonas cepaciu-like strains, ornibactins
polyenes, structures 34 216f
polyesters 177ff Pseudomonas jluorescens, ferribactin 214
- poly-P-hydroxyalkanoates 177ff - ferrocins 213
polyethers, structures 34 - pseudobactins 213ff
polyketide drugs, structural class 39 - pyoverdins 213f
polyketide synthases 32ff Pseudomonas oleovorans, poly-phydroxyalka-
- scheme 33 noates 180
polyketides 32ff PUFA see polyunsaturated fatty acids
- structures 34 pulcherrimic acid, iron-binding compounds 233
polyprene hydrocarbons, as components of the - structure 233
alga Botryococcus braunii 175f putrebactin, structure 211
polyprenoids 175f pyochelin, catecholate siderophore of Pseudomo-
polysaccharides, immunologically active -, from nus aeruginosa 207f
plant cell cultures 612 - structure 207
polysporic acid, iron-binding pigments 231 pyoverdins 213ff
- structure 231 - defective mutants 214
polyunsaturated fatty acids (PUFA) 138f - peptide siderophore of Pseudomonas jluores-
- dietary uses 158 cens 213
- in bacteria 147ff - structures 213, 215
724 Index
Q - of peptide ligands, screening of antagonists

quinoline alkaloids, from plant cell cultures 614 116ff
p-quinones, antitumor agents 652 resistance genes, in aminoglycoside producers
quinoxaline antibiotics, antitumor agents 652 440
quorum sensors 65 resistance mechanisms, in bacteria 44Off
reveromycins, interruption of eukaryotic cell
growth 682
R reverse transcriptase inhibitors, from basidiomy-
ranalexin, structural gene 280 cetes 509ff
rapamycin 546ff - - structures 51Of
- biosynthetic clusters 21 rhamnolipids 180f
- immunosuppressive properties 546f - structure 181
- structure 546 - surfactant formation by Candida spp. 180
rapeseed oil, genetic engineering of - 138 rhizobactin 1021, dihydroxamate siderophore of
ras-farnesyltransferaseinhibitors 685 Rhizobium meliloti 221
- structures 686 rhizobactin DM4, carboxylate-type siderophore of
receptor-active compounds 107ff see also recep- Rhizobium meliloti 223
tor agonists, receptor antagonists, - structure 223
- new screening methods 107ff Rhizobium meliloti, rhizobactin 1021 221ff
- screening assays 109ff - rhizobactin DM4 223
- - Magnus test 109f rhizoferrin 229f
- - using radio-labeled ligands 110 - fungal siderophores 229
- - using recombinant cells l l O f - structure 229
- screening of, agonists 122ff Rhodosporidium toruloides, fatty acyl composi-
- - antagonists l l l f f tion 152
receptor agonists see also receptor-active com- Rhodotorula glutinis, lipid accumulation patterns
pounds 140f
- erythromycin A 122ff rhodotorulic acid, fungal siderophores 228
- motilides 122f - structure 228
- muscarine 123 ribostamin family, structure of - 467
- zearalenone 123 ribostamycins 434
receptor antagonists l l l f f see also receptor-active ristocetin A 372f
compounds, - structure 373
- androgen receptor 116 RNA polymerase, of E. coli, d subunit 64
- ANP receptor 120f rosmarinic acid, from plant cell cultures 598ff
- arginine-vasopressin receptor 121 - - yields 599
- C5a receptor 121 - structure 599
- CCK receptors 116, 121 rotamases 547f
- devazepide receptor 121
- dopamine receptor 114
- endothelin receptor 116ff S
- estrogen receptor 115f Saccharomyces cerevisiae, accumulation of sterols
- fibrinogen receptor 115 172
- ligand receptors 112 - formation of ergosterol 172
- LTB4 receptor 114 - transformed -, screening of steroid receptors
- muscarinic acetylcholine receptor 112f 111
- neurokinin receptors 120 Saccharopolyspora erythraea, polyketide synthase
- NMDA receptor 114 33
- oxytocin receptor 121 saframycin A, antitumor agent, structure 662
- PAF receptor 114f salivaricin A, prepeptide structure 345
- peptide ligands 117ff - primary structure 332
- structures 113 - structure 335
- substance P receptor 120 sanguinarine, from plant cell cultures, yields 606
receptor binding, of viruses, inhibitors 124ff saponin aglycone, from plant cultures 611f
receptor genes, of recombinant cells, use in saponins, from plant tissue cultures 611f
screens of receptor-active compounds 110 schizokinen 220f
receptors, of low molecular weight ligands, screen- schizokinen A, structure 221
ing of antagonists 112ff schizonellins, structure 504
Index 725
scopolamine, from hairy root cultures 620 - incorporation into cellular structures 15
- from plant cell cultures 613 - main biosynthetic pathways 31ff
- production in transgenic Atropa belladonna - overproduction of - 29f
plants 625f - pharmacological activities 7ff
- structure 624 - precursor pools of - 29f
scorodonin 502f - producing organisms 4ff
- antifungal activity 503 - production in plant cell cultures 593ff
- structure 502 - properties 4ff
scorpions, as sources of nonribosomal peptides - regulation in streptomycetes 443ff
314 - roles in producing organisms l l f f
- as sources of ribosomal peptides 301 - structural classification 38ff
screening, of natural products, compilation of - - structural types 4ff
4ff - sugar components in - 397ff
- of bioactive metabolites, from basidiomycetes secondary pathways, single site manipulations
496 626
- of dalbaheptides 381 seldomycin family, structure of - 469
- of natural products 3ff seldomycins, biosynthetic pathway 433
- of plant secondary metabolites 595 septicemia, in mice, effect of dalbaheptides 380
- of receptor-active compounds 107ff serotonin, biosynthesis in plant cell cultures 624
sea hares, as sources of nonribosomal peptides - structure 624
314 serpentine, from plant cell cultures 608f
secondary metabolic aminotransferases 422 - - yields 608
secondary metabolism, evolution of - 16 sexual hormone production, inhibitors of -, treat-
- general aspects Iff ment of hormone-dependent tumors 685
- negative feedback control 30 shikimic acid, from plant cell cultures 603
- overproduction of enzymes, in transgenic plant shikonins, from plant cell cultures 601f
cell cultures 627 - - yields 601
- peptide bond formation 36 sideramines, ferric ion-chelating compounds 493
- regulation of - 17ff siderochelin A, catecholate siderophore of Nocar-
- - in eukaryotes 22f dia sp. 208f
- regulatory molecules, induction of cytodifferen- - structure 208
tiation 16 siderophore-mediated iron transport 233ff
secondary metabolite formation, effect, of nutrient siderophores 14, 199ff see also bacterial sidero-
depletion 24ff phores, fungal siderophores
- - of stress conditions 24ff - biological properties 226
secondary metabolites, adaptation to metabolic - carboxylate-type 223ff
imbalances 13 - catecholate-type 201ff
- agricultural uses 9ff - citrate hydroxamates 22Off
- - plant growth regulators 9 - ferric uptake regulation proteins 200
- as a signal of interspecific communication 13, - history 201
15 - hydroxamate-type 209ff
- as drugs 2ff - iron(II1) binding ligands 201
- - antimicrobial agents 4ff - metal-to-ligand charge transfer bands, colora-
- - antitumor agents 6f tion 200
- as endogenous signals 13 - mycobactins 217ff
- as metabolic reserves 13, 15 - peptide-type 212ff
- biological role 493 - role in iron nutrition 235
- biosynthesis, biosynthetic modifications 37f signal cascades, in Bacillus subtilis, surfactin syn-
- - use of cell-free enzymes 37 thesis 68
- biosynthetic genes clusters 17ff signaling molecules, from bacteria and fungi 14f
- enhancement by metabolic engineering 623ff - - structures 14
- - nicotine 624f - PPGPP 61
- - scopolamine 625f single cell oils 135ff
- - serotonin 624 - advantage over single cell protein 185
- expression of regulatory mechanisms 23ff snakes, as sources of ribosomal peptides 302
- formation from a single precuror 31f solution structures, of lantibiotics 338ff
- genetic instability 26f sophorolipids, structure 180
- immunosuppressors 7 - surfactant formation by Pseudomonas spp. 180
726 Index
sorbistins 439ff - oleandomycin production, growth-phase de-

- biogenesis of - 439 pendency 72
- structure 472 - phenoxazinone synthase 73
soybean oil, fatty acid composition 136 Streptomyces clavuligerus, plactam biosynthesis
- genetic engineering of - 138 251
sp&tinomychs, streptomycin-related aminoglyco- - clustering of p-lactam biosynthetic genes 259ff
sides 427 - penicillin biosynthetic enzymes 254ff
- structure 464 Streptomyces coelicolor 71ff, 81, 83
spenolimycin, structure 464 - antibiotic regulatory genes 85ff
sphingolipids, of yeasts, structure 183 - biosynthetic genes 447
spicamycin, antitumor activity 647 - methylenomycin production 74f, 91f
spiculisporic acid, structure 181 - - on minimal medium 75
spiders, as sources of nonribosomal peptides 314 - ppGpp production 75ff
- as sources of ribosomal peptides 301 - regulatory cassettes for antibiotic export 92
spiramycin formation, regulatory gene of in Strep- Streptomyces fradiae, tylosin resistance 93
tomyces ambofaciens 90 Streptomyces glaucescens, biosynthetic genes 447
Spirulina maxima, as source of polyunsaturated - regulatory cassettes for antibiotic export 92
fatty acids 167 Streptomyces griseus, A-factor cascade 87f
Spirulina platensis, as source of polyunsaturated - A-factor-induced regulation 446
fatty acids 167 - candicidin production 71f
- pcarotene content 172 - - phosphate repression 74
sponges, as sources of nonribosomal peptides - decision phase 445
312f - decision-making growth phase 447
Staphylococcus hyicus, staphyloferrin A 223 - exponential growth phase 445
staphyloferrin, carboxylate-type siderophore of - stationary growth phase 445
Staphylococcus hyicus 223f - streptomycin biosynthetic genes 91
staphyloferrin A 2231 - streptomycin production, regulation of - 445
- structure 224 Streptomyces hygroscopicus, bialaphos produc-
staphyloferrin B, structure 224 tion 91
stationary growth phase, of Bacillus subtilis 67 - rapamycin production 546f
staurosporin, inhibition of protein kinase C 677 Streptomyces lavendulae, formycin production 77
stearic acid, increase in yeasts 15Off Streptomyces peuceticus, daunorubicin production
- - by direct feeding 15Of 91
- - by inhibition of stearoyl desaturase 151f Streptomyces pilosus, ferrioxamines 209f
- - by metabolic manipulation 154 Streptomyces sp., formation of anticancer drugs
- - by mutation 152ff 660
stearoyl desaturase, inhibition of - 151f Streptomyces tsukubaensis, FK506 production 546
stearoylphytosphingosine, structure 184 Streptomyces virginiae 78f
stearoylvelutinal, antimicrobial activity 495 streptomycetes, antibiotic formation, factors deter-
sterigmatocystin, biosynthetic clusters 21 mining onset of - 86
steroid receptors, screening with transformed Sac- - antibiotic production 7Off
charomyces 111 - - regulation of - 70
steroidal compounds, from plant cell cultures - ybutyrolactones, A-factor 78
611f - induction of antibiotic resistance 92ff
sterols 170ff - regulation of antibiotic production, interfering
- accumulation in yeasts 171f metabolites 73
- commercial extraction of eukaryotes 170ff - regulation of secondary metabolite production
- structures 171 443ff
streptidine, biosynthetic pathway 417,421 - regulatory genes 27
streptococcin A-FF22, prepeptide structure 345 streptomycin, biosynthetic genes 417
- primary structure 334f - - in Streptomyces griseus 90
Streptomyces, cytodifferentiation, A-factor 27 - pathway for its release 426
- - signal cascade 27 streptomycin biosynthesis, regulation of - 88
- hydroxamate siderophores, ferrioxamine type streptomycin production, enzymes, in biosynthetic
211f pathway 418f
Streptomyces ambofaciens 92 - gene clusters 420
- spiramycin production in - 90 - gene products 418
Streptomyces antibioticus 72ff - in Streptomyces griseus, regulation of - 445
Index 727
streptomycins 415ff Taxus brevifolia 615
- basic pathway formula 463 teicoplanin 375f
- biosynthetic pathway 416 - effect in mouse septicemia 380
- condensation of subunits 425 - heptapeptide, amino acid composition 384
- export of - 425 - structure 376
- processing of subunits 425 teicoplanin aglycone, stereo model 377
streptonigrin, cytotoxic antitumor agent 656 terpenes, biosynthesis 35
striatals 493f terpenoids, structural class 39
- biosynthesis 494 testosterone 5a-reductase inhibitors 686
- structure 504 tetracyclines, structures 34
striatins, cytotoxic activity 507 tetrafibricin, fibrinogen binding antagonists
- structure 504 115
strobilurins 496ff Thalictrum sp., accumulation of protoberberines
- blockage of respiration in eukaryotes 497 604f
- fungistatic activity 497 thelephoric acid, iron-binding compounds 232
- mode of action - 497ff - structure 232
- possible functions 501 thioguanine, antitumor activity 647
- producing organisms 499 thiouridine, antitumor activity 647
- structures 498 tiamulin, antibiotic action 496
substance P receptor antagonists 120 - structure 495
subtilin, prepeptide structure 345 tissue culture technology see plant cell cultures,
- primary structure 331ff plant tissue cultures
- solution structure, in aqueous solution 339 tolypocin, iron-binding compounds 232
- - in lipophilic solvents 339 - structure 232
- structural gene 343f topoisomerases, inhibition by anticancer drugs
sugar components, attachment to cyclitols 405ff 653ff
- basic pathways 402ff torulene 175
- structure 174
- biogenesis 402ff
- in secondary metabolites 397ff
transduction of an extracellular signal, protein
sugar derivatives, monomeric - 436f kinases, effect on neoplastic cell growth 675
transgenic plant cell cultures 623ff
sunflower oil, genetic engineering of - 138 - overexpression of enzymes 627
surfactin synthesis 67f transplant rejection therapy, use of cyclosporin A
- by Bacillus subtilis 67ff 539f
- - signal cascade 68
trehalosamine family, structure of - 470
suspension cultures see plant cell suspension cul- trehalosamines 437f
tures - as glycosidase inhibitors 437
Syringa vulgaris, production of verbascoside 599f trehazolin, structure 472
- trehalase inhibitor 439
triacylglycerols 146ff
T - biosynthetic pathway 144
T cell activation, cytokine receptor expression - glucose conversion, overall stoichiometry
545 145
- gene 545 - pathway of synthesis 143ff
- inhibition of -, by rapamycin 547 trichostatins, phosphoinositol kinase inhibition
- - role of immunophilins 548f 681
- release of cytokines 545f triglycerides see triacylglycerols
- role of calcineurin 549 triptolide, structure 615
- signal recognition 542ff tropane alkaloids, from plant cell cultures 613f
- - binding of antigen 542f - from plant hairy root cultures 619
- - costimulatory signals 542ff tumor cells see also cancer cells
- signal transduction 544f - efflux of cytotoxic drugs 672
tautomycin, inhibition of protein kinase C 677 - multidrug resistance 540,551
taxol 665ff - redifferentiation of - 683
- biosynthetic pathway 617 tylosin resistance, in streptomycetes, induction of
- from plant tissue cultures 614ff - 93
- - yields 616 tyromycin, structure 519
- structure 615, 668 tyrosine protein kinase inhibitors 680
728 Index
U violaxanthin, structure 173

UK 63.052, antitumor agent, structure 651 virus receptor binding, inhibition, by chloropep-
tins 124ff
- - by isochromophilones 124f
V - - of HIV entry 124ff
validamycin family, biogenesis of - 438 vitamin A, structure 174
- structure of - 471 vitamin DZ, formation in yeasts 172
validamycins 438f vulnibactin 205f
vancomycin 373f - catecholate siderophore of Vibrio vulnificus
- biotransformation of - 390 206
- effect in mouse septicemia 380 - structure 205
- structure 373 vulpinic acid, iron-binding compounds 232
vanilla flavor, from plant cell cultures 617f - structure 232
Vanilla planifolia, plant cell cultures 618
vanillin 617
variegatic acid, iron-binding compounds 232 W
- structure 232 wax esters 176f
variegatin, iron-binding pigments 231 - pathways to a diunsaturated wax ester 176
- structure 231
verbascoside, from plant cell cultures, yields 600
- structure 599 X
Vibrio anguillarum, anguibactin 206f xerocomic acid, iron-binding compounds 232
vibriobactin 205f - structure 232
- catecholate siderophore of Vibrio cholerae 205 xerulin, inhibitor of cholesterol biosynthesis 518
- structure 205 - structure 519
Vibrio cholerae, vibriobactin 205f
vibrioferrin, carboxylate-type siderophore of Vi-
brio parahaemolyticus 224 Y
- structure 224 yeasts see also oleaginous yeasts
Vibrio fischeri, light emission 66 - sphingolipids 183
- OHHL 65f - sterol contents 171f
- - structure 65 - yeast “lecithin” 183
Vibrio fluvialis, fluvibactin 206 yellamycin B, antitumor agent 659
Vibrio parahaemolyticus, vibroferrin 224 - structure 659
Vibrio vulnificus, vulnibactin 206 yersiniabactin 207f
vinblastin, from plant cell cultures 614 - catecholate siderophore of Yersinia enterocoliti-
- inhibition of the microtubular system 668 ca 208
- structure 615, 668 - structure 207
vinca alkaloids 665 Yersinia enterocolitica, yersiniabactin 208
vincristin, from plant cell cultures 614
- inhibition of the microtubular system 668
- structure 668 8 2
vindesin, inhibition of the microtubular system zealarenone, as estrogen receptor agonist 112
668 zearalenone 123
- structure 668 zymosterol, structure 171

Biotechnology Products of Secondary Metabolism

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Biotechnology Products of Secondary Metabolism

Uploaded by

Copyright:

Available Formats

Biotechnology

Executive Editor: Dr. Hans-Joachim Kraus

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication Data:

Die Deutsche Bibliothek - CIP-Einheitsaufnahme

NE: Rehm, Hans J. [Hrsg.]

OVCH Verlagsgesellschaft mbH, D-69451 Weinheim (Federal Republic of Germany), 1997

Printed on acid-free and chlorine-free paper.

Pro$ Dr. M. J. Beker Pro$ Dr. T. K. Ghose

Pro$ Dr. J. D. Bu’Lock Pro$ Dr. I. Goldberg

Pro$ Dr. C. L. Cooney Pro$ Dr. G. Goma

Pro$ Dr. H. W. Doelle Sir D. A. Hopwood

Prof Dr. J. Drews Pro$ Dr. E. H. Houwink

Pro$ Dr. A. Fiechter Pro$ Dr. A. E. Humphrey

Prof. Dr. I. Karube Prof. Dr. K. Schiigerl

Prof. Dr. M . A. Lachance Prof. Dr. P. Sensi

Prof. Dr. Y. Liu Prof. Dr. Y. H. Tan

Prof. Dr. J. F. Martin Prof. Dr. D. Thomas

Prof. Dr. B. Mattiasson Prof. Dr. W . Verstraete

Prof. Dr. M . Roehr Prof. Dr. E.-L. Winnacker

Prof. Dr. H. Sahm

Introduction 9 Glycopeptide Antibiotics

Prof. Dr. Timm Anke Prof. Keith Chater

Dr. Jochen Berlin Jurgen Distler

Dr. Mervin Bibb Dr. Hans von Dohren

Dr. Bruno Cavalleri Dr. Klausjiirgen Dornberger

Dr. Hartmut Drechsel Prof. Dr. Gunter Jung

Dr. Gerhard Erkel Dr. Jorg Kallen

Dr. Hans Fliri Prof. Dr. Horst Kleinkauf

Dr. Friedrich Gotz Dr. Giancarlo Lancini

Prof. Dr. Udo Grafe Dr. Henk van Liempt

Dr. Ralph Jack Dr. Vincent Mikol

Prof. Dr. Satoshi Omura Dr. Kurt Schorgendorfer

Prof. Dr. Wolfgang Piepersberg Dr. Torsten Schwecke

Dr. Valkrie F.J. Quesniaux Dr. Paul L. Skatrud

Prof. Colin Ratledge Dr. Haruo Tanaka

Prof. Dr. habil. Hans-Peter Saluz Dr. Matthew B. Tobin

Dr. Elisabeth Schneider-Scherzer Dr. Malcolm D. Walkinshaw

Dr. Gerhard Weber Prof. Dr. Giinther Winkelmann

1 General Aspects of Secondary

1 Introduction: The Importance of Secondary Metabolites as Drugs 2

1 Introduction: The crease in resistant nosocomial and opportu-

Tab. 1. Selected Natural Products Detected by Screening Efforts Published in 1995196

Compound Producing Organisms Structural Selected Research Group

Compound Producing Organisms Structural Selected Research Group

Compound Producing Organisms Structural Selected Research Group

Benzastatins Streptomyces ALK free radical KRIBB

Compound Producing Organisms Structural Selected Research Group

Thiazinotrieno- Streptomyces sp. PK-PEP cytostatic Institute of Microbial

Compound Producing Organisms Structural Selected Research Group

Phenopyrazin , Penicillium sp. PK-AA radical Kitasato

Compound Producing Organisms Structural Selected Research Group

Fluvirucin Streptomcyces sp. PK-GLYC phospholipase Univ. Keio

Compound Producing Organisms Structural Selected Research Group

Methylstrept- Streptomyces sp. PK-mod. herbicidal Hoechst India

Tab. 2. Presumed “Roles” of Secondary Metabolites in Their Producer Organism

the formation of large amounts of antibiotics lated by secondary metabolites in hetero-

be preserved (secondary metabolism as a tochromes, chlorophylls, sexual pheromones

(LANDANet al. 1990 MILLERand INGOLIA, tors of Streptomyces differentiation (Fig. 2)

discoideum, suggesting the similarity of mam- 2.2 Regulation of Microbial

Tab. 3. Biosynthetic Clusters Identified

Compound Type 0rganism Selected References'

Compound Type Organism Selected References'

Me1anin Aspergillus nidulans TAKANO