Professional Documents
Culture Documents
Wageningen University
Wageningen, The Netherlands
ISBN: 0-8247-0686-2
Headquarters
Marcel Dekker, Inc.
270 Madison Avenue, New York, NY 10016
tel: 212-696-9000; fax: 212-685-4540
The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/
Professional Marketing at the headquarters address above.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, microlming, and recording, or by any information storage and retrieval system, without permission in
writing from the publisher.
Enzymes are the catalysts of life. They perform the majority of reactions in all living systemsall animals, plants,
and microorganisms. Our lives are absolutely dependent on them. Even inhaling in oxygen and exhaling carbon
dioxide are dependent on enzymes. Our plants take up minerals, water, and air, and combine them with carbon
dioxide and ammonia to provide our fruits and vegetables, via enzymes.
The Handbook of Food Enzymology is a single source that can answer questions about what enzymes are, how
they function catalytically, and how we depend on them and use them every minute of our life. This is not a book
that can be read in a single day. Rather, it is a handbook to be used to nd the answers to various questions: Who
are we biologically and chemically? How do we function biologically? How do our foods, most of our clothing, and
much of our shelter depend on enzymes? What are enzymes? What do they do? How can we manage them in the
most effective way to live a long and healthy life with an abundance of food?
The Handbook is divided into two major parts. The rst 26 chapters deal with general aspects of enzymology.
There are two chapters (1 and 2) on the history of enzymology, which discuss protein structure and the kinetics of
enzyme reactions and emphasize food production, and the trail of knowledge that led to this handbook. The next
two chapters (3 and 4) deal with how enzymes are highly specic and highly efcient catalysts and the environmental
factors that affect their activities. Chapter 5 describes how to inactivate enzymes when their action is no longer
desirable. Chapters 6 and 7 deal with regulatory issues pertaining to using enzymes in our foods to change the taste,
avor, aroma, color, texture, and nutritional quality. Chapter 8 deals with the evolution of enzymes under extreme
conditions of pH, temperature, salinity, toxicity, and pressure, and how these enzymes with special extremophilic
characteristics can be used effectively in foods. Chapters 9 through 16 deal with how enzymes produce the compo-
nents of our foodsthe proteins, lipids, carbohydrates, and nucleic acidsas well as the enzymes that make it
possible to synthesize our food components rapidly, efciently, nutritiously, and in the quantity needed to feed 6
billion persons each day. The last 10 chapters (1726) in Part I deal with management of enzymes postharvest and as
tools for increasing the quality and consumability of our foods.
Part II describes specic prototypic enzymes of the six chemical types of reactions catalyzed by enzymes:
oxidoreductases, transferases, hydrolases, lyases, isomerases, and ligases. The contributors of these 57 chapters
have followed a common format in which they discuss what the enzyme(s), does, its importance to feed and food
production, its chemical and biological properties, how to measure activity of the enzyme(s) and how to purify it
from raw animal, plant, or microbial sources. Being catalysts, the enzymes are produced in rather small amounts
(0.01 to 1% by weight of the total protein).
The contributors were carefully selected on the basis of their specic knowledge of the importance of enzymology
to food production, food preservation, and food quality. We are very proud and fortunate to have attracted rst-
John R. Whitaker
Alphons G. J. Voragen
Dominic W. S. Wong
Preface
INTRODUCTION
What Enzymes Do and Why They Are Highly Specic and Efcient Catalysts
John R. Whitaker
REGULATORY ISSUES
Regulatory Issues of Enzymes Used in Foods from the Perspective of the E.U. Market
Danielle P. Praaning
Properties of Extremophilic Enzymes and Their Importance in Food Science and Technology
Magnus M. Kristjansson and Bjarni Asgeirsson
Flavor Enhancement in Fruit Juices and Derived Beverages by Exogenous Glycosidases and
Consequences of the Use of Enzyme Preparations
Ziya Gunata
OXIDOREDUCTASES
Catalase
Dominic W. S. Wong and John R. Whitaker
Horseradish Peroxidase
Zhong Yi Yuan and Tai Jiao Jiang
Glutathione Peroxidase
Jun-qiu Liu and Gui-min Luo
Glucose Oxidase
A. J. Vroemen
Lactate Dehydrogenase
Hayao Taguchi
Alcohol Dehydrogenase
Sabato DAuria
Alcohol Dehydrogenase
Sandrine Dallet, Marie Trovaslet, and Marie Dominique Legoy
Amine Oxidase
Akio Ito and Jichun Ma
Prolyl 4-Hydroxylase
Robert B. Rucker, Ana Samimi, and Jerold A. Last
Lysyl Oxidase
Robert B. Rucker, Alyson E. Mitchell, Eskouhie Tchaparian, and Jerold A. Last
Superoxide Dismutase
Hirokazu Hara, Tetsuo Adachi, and Kazuyuki Hirano
Polyphenol Oxidase
Edna C. Ramrez, John R. Whitaker, and Victoria M. Virador
Laccase
William H. Flurkey
Xanthine Oxidase
John R. Whitaker
Sorbitol Oxidase
Kohei Oda and Kazumi Hiraga
TRANSFERASES
Starch Phosphorylases
Ying Yu
Amylosucrase
Volker Buttcher
Dextransucrase
Klaus Buchholz and Pierre F. Monsan
Levansucrase
Ki-Bang Song and Sang-Ki Rhee
Cyclodextrin Glycosyltransferase
Lubbert Dijkhuizen and Bart A. van der Veen
Limonoid Glucosyltransferase
Shin Hasegawa, Masayuki Kita, and Mitsuo Omura
Transglutaminase
Chang-Rak Ha and Ichiro Iuchi
Feruloyl Esterases
Craig B. Faulds and Gary Williamson
Lipase
Dominic W. S. Wong
Chlorophyllase
Roger F. McFeeters
Phytase
Onno Misset
-Amylases
Dominic W. S. Wong and George H. Robertson
-Amylases
Dominic W. S. Wong and George H. Robertson
Glucoamylase
Peter J. Reilly
Limit Dextrinase/Pullulanase
E. Ann MacGregor
Limit Dextrinase
James Hutchison Bryce
-Glucosidase
Asim Esen
-D-Fructofuranoside Fructohydrolase
Laura Cantarella, Francesco Alfani, and Maria Cantarella
-Galactosidase
Raymond R. Mahoney
Pectic Polysaccharides
H. A. Schols and Alphons G. J. Voragen
Pectic Enzymes
Jacques A. E. Benen, Alphons G. J. Voragen, and Jaap Visser
Pectic Esterases
Jacques A. E. Benen, Gert-Jan W. M. van Alebeek, Alphons G. J. Voragen, and Jaap Visser
Enzymes Releasing L-Arabinose and D-Galactose from the Side Chains of Pectin
Ronald P. de Vries and Jaap Visser
Xylanolytic Enzymes
Peter Biely
Enzymology of Endo-1,4--Mannanases
Henrik Stalbrand
Lysozyme
Akio Kato
Ribonucleases
Jaap J. Beintema and Wei Zhao
Proteolytic Enzymes
John R. Whitaker
Thermolysin
Kuniyo Inouye
LYASES
Alliinases
Edna C. Ramrez
ISOMERASES
Tetsuo Adachi Laboratory of Clinical Pharmaceutics, Gifu Pharmaceutical University, Gifu, Japan
Francesco Alfaniy Department of Chemistry, Chemical Engineering and Materials, University of LAquila, LAquila, Italy
Lucas C. Armstrong Division of Pulmonary and Critical Care Medicine, Toxic Substances Research and Training Program,
Department of Internal Medicine, University of California, Davis, Davis, California, U.S.A.
Bjarni Asgeirsson Department of Biochemistry, Science Institute, University of Iceland, Reykjavik, Iceland
Gerrit Beldman Laboratory of Food Chemistry, Department of Agrotechnology and Food Sciences, Wageningen University,
Wageningen, The Netherlands
Jacques A. E. Benen Laboratory of Microbiology, Department of Agrotechnology and Food Sciences, Wageningen University,
Wageningen, The Netherlands
James Hutchison Bryce International Center for Brewing and Distilling, Department of Biological Sciences, Heriot-Watt
University, Edinburgh, Scotland
y Deceased.
Maria Cantarella Department of Chemistry, Chemical Engineering and Materials, University of LAquila, LAquila, Italy
Sabato DAuria Institute of Protein Biochemistry and Enzymology, Naples, Italy, and Center for Fluorescence Spectroscopy,
University of Maryland, Baltimore, Maryland, U.S.A.
Sandrine Dallet Laboratory of Protein and Cellular Engineering, University of La Rochelle, La Rochelle, France
Ralph E. Dewey Department of Crop Science, North Carolina State University, Raleigh, North Carolina, U.S.A.
Asim Esen Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, U.S.A.
Craig B. Faulds Food Materials Science Division, Institute of Food Research, Colney, Norwich, England
Geoffrey Fincher Department of Plant Science, University of Adelaide, Glen Osmond, South Australia, Australia
William H. Flurkey Department of Chemistry, Indiana State University, Terre Haute, Indiana, U.S.A.
Harold W. Gardner* National Center for Agriculture Utilization Research, Agricultural Research Service, U.S. Department of
Agriculture, Peoria, Illinois, U.S.A.
Ziya Gunata Department of Bioengineering and Food Science, University of Montpellier II, Montpellier, France
Hirokazu Hara Laboratory of Clinical Pharmaceutics, Gifu Pharmaceutical University, Gifu, Japan
Shin Hasegawa Western Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture, Albany,
California, U.S.A.
Marc Hendrickx Laboratory of Food Technology, Katholieke Universiteit Leuven, Leuven, Belgium
Kazumi Hiraga Department of Applied Biology, Faculty of Textile Science, Kyoto Institute of Technology, Kyoto, Japan
Maria Hrmova Department of Plant Science, University of Adelaide, Glen Osmond, South Australia, Australia
*
Retired.
Kuniyo Inouye Laboratory of Enzyme Chemistry, Division of Food Science and Biotechnology, Graduate School of
Agriculture, Kyoto University, Kyoto, Japan
Akio Ito Department of Chemistry, Faculty of Science, Kyushu University, Fukuoka, Japan
Violeta G. Janolino Department of Food Science, North Carolina State University, Raleigh, North Carolina, U.S.A.
Tai Jiao Jiang Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, China
David Johnston Eastern Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture,
Wyndmoor, Pennsylvania, U.S.A.
Takuo Kawamoto Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto
University, Kyoto, Japan
Masayuki Kita Department of Plant, Cell, and Environment, National Institute of Fruit Tree Science, Tsukuba, Ibaraki, Japan
Magnus M. Kristjansson Department of Food Science, Science Institute, University of Iceland, Reykjavik, Iceland
Jerold A. Last Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, University of California,
Davis, Davis, California, U.S.A.
Marie Dominique Legoy Laboratory of Protein and Cellular Engineering, University of La Rochelle, La Rochelle, France
Jun-qiu Liu Key Lab of Molecular Enzymology and Engineering of Ministry of Education, Jilin University, Changchun, China
Gui-min Luo Key Lab of Molecular Enzymology and Engineering of Ministry of Education, Jilin University, Changchun,
China
Raymond R. Mahoney Food Science Department, University of Massachusetts, Amherst, Massachusetts, U.S.A.
Roger F. McFeeters Agricultural Research Service, U.S. Department of Agriculture, and Department of Food Science, North
Carolina State University, Raleigh, North Carolina, U.S.A.
Pierre F. Monsan Department of Biochemistry and Food Engineering, National Institute for Applied Sciences, Toulouse,
France
Toshiyuki Murai Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto
University, Kyoto, Japan
William E. Newton Department of Biochemistry, The Virginia Polytechnic Institute and State University, Blacksburg, Virginia,
U.S.A.
Philip OConnell Laboratory of Sensor Development, Department of Chemistry, National University of Ireland, Cork, Ireland
Kohei Oda Department of Applied Biology, Faculty of Textile Science, Kyoto Institute of Technology, Kyoto, Japan
Mitsuo Omura Department of Citrus Research, National Institute of Fruit Tree Science, Shimizu, Shizuoka, Japan
Peter J. Reilly Department of Chemical Engineering, Iowa State University, Ames, Iowa, U.S.A.
Sang-Ki Rhee Biomolecular Engineering Laboratory, Korea Research Institute of Bioscience and Biotechnology (KRIBB), and
Real Biotech Co., Ltd., KRIBB, Daejeon, Korea
George H. Robertson Western Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture,
Albany, California, U.S.A.
Robert B. Rucker Department of Nutrition, University of California, Davis, Davis, California, U.S.A.
Ana Samimi Pharmacology and Toxicology Graduate Program, University of California, Davis, Davis, California, U.S.A.
H. A. Schols Laboratory of Food Chemistry, Department of Agrotechnology and Food Sciences, Wageningen University,
Wageningen, The Netherlands
Chantal Smout Laboratory of Food Technology, Katholieke Universiteit Leuven, Leuven, Belgium
Ki-Bang Song Biomolecular Engineering Laboratory, Korea Research Institute of Bioscience and Biotechnology (KRIBB), and
Real Biotech Co., Ltd., KRIBB, Daejeon, Korea
Allan Keith Stobart School of Biological Sciences, University of Bristol, Bristol, England
Harold E. Swaisgood Department of Food Science, North Carolina State University, Raleigh, North Carolina, U.S.A.
Atsuo Tanaka Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto
University, Kyoto, Japan
Eskouhie Tchaparian Agricultural and Environmental Sciences, Department of Nutrition, University of California, Davis,
Davis, California, U.S.A.
Marie Trovaslet Laboratory of Protein and Cellular Engineering, University of La Rochelle, La Rochelle, France
Mitsuyoshi Ueda Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto
University, Kyoto, Japan
Gert-Jan W. M. van Alebeek Laboratory of Food Chemistry, Department of Agrotechnology and Food Sciences, Wageningen
University, Wageningen, The Netherlands
Bart A. van der Veen Department of Microbiology, University of Groningen, Haren, The Netherlands
Ann Van Loey Laboratory of Food Technology, Katholieke Universiteit Leuven, Leuven, Belgium
Jean-Paul Vincken Laboratory of Food Chemistry, Department of Agrotechnology and Food Sciences, Wageningen
University, Wageningen, The Netherlands
Jaap Visser Fungal Genetics and Technology Consultancy, Wageningen, The Netherlands
Alphons G. J. Voragen Laboratory of Food Chemistry, Department of Agrotechnology and Food Sciences, Wageningen
University, Wageningen, The Netherlands
Bruce P. Wassermany Department of Food Science, Rutgers University, New Brunswick, New Jersey, U.S.A.
John R. Whitaker Department of Food Science and Technology, University of California, Davis, Davis, California, U.S.A.
Chris Winkel Department of Enzyme Research, Quest International, Naarden, The Netherlands
Dominic W. S. Wong Western Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture,
Albany, California, U.S.A.
Ying Yu Department of Food Science, Rutgers University, New Brunswick, New Jersey, U.S.A.
Zhong Yi Yuan Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, China
y Deceased
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
I. INTRODUCTION tively) (1). This was during the time when the word
protein was coined to describe these compounds
This opening chapter focuses primarily on the history that have the compositional formula of carbon, hydro-
of applications of enzymes in food science and technol- gen, oxygen, and nitrogen 16%, with small
ogy. The evolution of knowledge on proteins and food amounts of sulfur and phosphorus. Methionine was
enzymology must also be understood from the funda- discovered in 1922 by Mueller (1) and in 1935 threo-
mental aspects. Enzymes are proteins. Therefore, the nine, the last amino acid, was discovered by Rose. It
history of food enzymology is a composite of advances was very difcult to determine the amino acid composi-
in both understanding the chemistry of protein struc- tion in the earlier days. With the discovery in 1952 by
ture and environmental behavior, as well as the physi- Moore and Stein (2) that the amino acids of protein
cochemical properties (kinetics and thermodynamics) hydrolyzates could be separated and identied readily
of enzyme behavior. within 24 h by ion exchange chromatography, the
This chapter will rst treat the history of food enzy- amino acid composition of many proteins became read-
mology from the evolution of protein chemistry and ily available. Today, the complete amino acid composi-
then the physicochemical behavior of enzymes in rela- tion of a protein is done by automated analysis in 224
tion to environmental conditions, including the effects h using nano- or picogram amounts of hydrolyzed pro-
of enzyme, substrate, cofactor and inhibitor concentra- tein, at a cost of $2535 per total analysis.
tions, and the effects of pH and temperature.
Active sites of proteins are the locations where the There is some overlap between this section and that of
biological function of a protein resides in the native Chapter 2. This is justied based on the need to present
protein. Study of active sites of proteins became a a complete picture in the book.
major eld of study for enzymologists and protein che-
mists beginning in the 1960s. But as early as 1902, 1. Catalase appears to be the rst enzyme discov-
Brown (28) and Henri (29) independently proposed ered. This occurred in 1812. In reality, the discovery
that enzymes must bind the substrate prior to conver- was that something in blood caused the evolution of
sion to product. Fischer in 1894 (30) proposed that the bubbles, shown to be O2 , when hydrogen peroxide was
active site must t the substrate as stereospecically as added. The reaction observed was
a key ts a lock (lock-and-key analogy) in order to be 2 H2 O2 ! 2 H2 O O 2
bound properly and be converted to product.
Koshland in 1965 (31) introduced the induced t 2. Diastase was discovered in wheat our by Payen
hypothesis to explain how apparently dissimilar mole- and Persoz in 1833 (34), when they reported that the
cules could sometimes serve as substrates for the same enzyme caused the solubilization of starch. In addi-
enzyme. tion, they found the enzyme to be thermolabile and
Initially, study of active sites of enzymes was precipitated by alcohol. They named the substance
approached by use of chemical modication of the diastase because it separated, by solubilization,
side chain groups of amino acid residues of the native the starch from the bran. They further used ase
protein, using loss of biological activity as the measure. in the name to designate it as an enzyme. The reac-
Also, the unfolding and refolding of proteins such as tion observed was
myoglobin and ribonuclease, especially around cofac- Starch H2 O ! dextrins
tors, hematin, etc., gave much information on active
3. Peroxidase in plants was reported by Schoenbein
sites. Much was learned about essential amino acids in
in 1855 (35, 36). Schoenbein used gum guaiac which
the active site. For example, in lysozyme chemical
changes in color from light tan to blue when H2 O2 was
modication established that Glu35, Asn37, Asn44,
added to an extract containing peroxidase. Several
Asp52, Asp59, Try62, Try63, Asp101, and Arg114
other compounds can be used to assay easily for per-
are a part of the active site (32). Asp52 and Glu35
oxidase as shown by the equation:
are the key catalytic general acid/general base residues
(COOH and COO are the catalytic groups). X-ray 4 Guaiacol 8 H2 O2 ! tetraguaiacol 8 H2 O
crystallography was key to determining that His69, colorless brown
Glu72, and His196 (chelates the essential cofactor
Zn(II)), Arg71, Arg145, Tyr198, Tyr248, and Phe279 4. Polyphenol oxidase was discovered by
are in the active site of carboxypeptidase A (33). Zn(II) Schoenbein in plant extracts in 1856 (36). It was
is responsible for catalysis. The other amino acid resi- named polyphenol oxidase because of its ability to
dues are key to binding the substrate properly in the oxidize phenols, cause browning of plant extracts and
active site. bruised plant tissues. The enzyme can use a number of
The best current method of determining the essenti- mono- and polyphenols plus O2 as substrates as shown
ality of amino acid residues in the active site is the use by:
of recombinant biotechnology to replace one amino Phenol 0:5 O2 ! o-dihydroxyphenol
acid at a time, thereby determining its effect on activ-
o-Dihydroxyphenol 0:5 O2 ! o-benzoquinone
ity, is much more certain than the two methods above.
Absence of binding and/or activity based on a single H2 O
Figure 1 Variation of initial velocity with substrate concen- Figure 2 Plot of reciprocal of initial velocity versus recipro-
tration for an enzyme-catalyzed reaction according to the cal of initial substrate concentration by the Lineweaver-Burk
Michaelis-Menten equation. (From Ref. 41.) method. (From Ref. 43.)
concentrated by heating with steam. The product Further enzymatic conversions (ferment actions)
rapidly found industrial application. observed in that early period summarized in
Payen and Persoz (12) identied numerous topics Franklands list of soluble ferments (14) and by
for further research, such as the existence of diastase Sumner and Somers (3) are summarized in Table 1.
in plants, its molecular weight and elementary compo- Alcoholic fermentation was a dominating topic of
sition, and the reaction products from different sub- the time in the eld called physiological chemistry
strates. Diastase was the dominating object of research which attracted a good deal of interest and activity
on enzymes throughout the century with 1020% of of leading scientists (5). Trying to identify the active
publications dealing with this topic owing to its eco- principle, Bethelot (15) showed, according to his own
nomic importance. Nevertheless the overall research interpretation, that a peculiar nitrogen-containing sub-
activities remained very weak in this eld, as compared stance, formed by yeast and precipitable by alcohol (he
to chemistry (see below). used casein) and comparable to diastase, could trans-
A few further activities were discovered in subse- form sugar into alcohol. He stated that this substance
quent years. Pectase was found in plants, both in solu- was different from yeast and that there was no produc-
ble and nonsoluble form. It was able to break down tion of yeast cells (8, 15). Thus, he claimed to have
pectose into pectinic acid, and was attributed to have observed the transformation of sugar into alcohol
some similarity to diastase. Neither material could be without the participation of any activity of living
crystallized (10: p. 30). Claude Bernard was the rst to organisms. He was convinced that the formation of
show lipolytic activity in pancreas in 1856 (4: p. 25). alcohol from sugar was a genuine chemical process.
Dobell (13) found that an extract from pancreas However, his experiments were not performed under
hydrolyzed both fat and starch; he gave a procedure sterile conditions, and obviously unknown in part,
for extraction and stabilization and named this pre- anaerobic bacteria (in tests where oxygen was
paration pancreatine. excluded) must have produced scarce amounts of alco-
Ferment
Enzyme source Catalyzed reaction Author
hol, carbon dioxide, but also hydrogen and acetic and there was about one article per year in the 1860s deal-
butyric acid (16). It was 40 years later when Buchner ing with soluble ferments, meaning enzyme activity,
succeeded in demonstrating cell-free alcohol formation and three or four dealing with fermentation. It was
unequivocally (see below). only in the 1880s that research and publication activ-
Pasteur presented a series of experimental results ities increased signicantly.
showing unequivocally that alcoholic fermentation
proceeded only in the presence of living organisms. B. Technology
And he showed that in all cases where fermentation
could be observed under precisely controlled condi- Work on technological issues was more important and
tions, it was microorganisms (of different kinds, also even dominating over that on basic aspects from the
present in the air) that must be present to initiate fer- beginning of the 19th century, especially for brewing,
mentation. So he presented strong evidence against the wine, bread, and acetic acid production. These topics
assumption of a generatio spontanea postulated by made up important parts of the chemical technology of
Gay-Lussac. He denitely changed, about two decades the time (7, 9, 10, 18). Knapp (9) considered scientic
later, the disciplinary context of fermentation away and practical industrial interests as equally important.
from chemistry, redening the subject to a new auton- Diastase application was a major issue from the
omous disciplinemicrobiology (1). 1830s onward. The diastase presents the best means
The overall research activity on soluble ferments, for production of dextrin and dextrin syrup. These
meaning enzymatic reactions, was scarce in this period. products not only become cheaper, but also more
Thus, in the German Annalen der Pharmacie in the pure and tasty. This latter argument was used in
period from 1833 to 1850, four articles were published favor of the application in food and beverage produc-
on ferments and fermentation, three of them on dia- tion. Thus, it was recommended for the manufacture of
stase. The key paper was essentially a translation from bread, pastries, chocolate, and soups; furthermore, it
Payen and Persoz (12, 17). In the German Journal fur was recommended for cider and fruit wine, but espe-
Praktische Chemie in the period from 1850 to 1860 no cially beer production, where a specic advantage was
paper dealt with soluble ferments (enzymatic activities; that the formation of haze at low temperature could be
eight papers were published on fermentation prevented (12, 17). Also, a sugar could be isolated after
[Gahrung], meaning microbial activity). In the extended action of diastase on starch. A mass balance
Bulletin de la Societe Chimique de Paris, one of the gave, in two experiments where the sugar was subse-
most important of the time for fermentation research, quently fermented by yeast and the amount of carbon
Figure 2 Surface culture with Aspergillus in a cabinet incubator. (From Ref. 4.)
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
a
The third and fourth levels of classication are given in Ref. 1:vxi.
Rate enhancementa
Factor (approximate)
Enzyme-Catalyzed Reactions
Experimental Factors that Affect Rates
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
Before proceeding further with discussion of rates, it is Figure 1 Typical enzyme-catalyzed reaction showing the
important to understand the pathway by which all pre-steady-state (msec), the constant rate, and the declining
enzyme-catalyzed reactions occur. This was rst appre- rate parts of the progress curve. (From Ref. 1)
k1 k2
Eo Ao Km EA EAAo EAKm Ao
E A EA E P 1 8
k1 k2
Rearranging Eq. (8):
Based on this premise, which has been shown now to
be universal for all enzyme-catalyzed reactions, EA Eo Ao Km Ao 9
Michaelis and Menten in 1913 (4) developed an equa- EA cannot be determined directly in most cases but it
tion to numerically describe the observed relationship can be substituted for since the rate of formation of
between velocity, v, and A. The derivation of the product, P, is
Michaelis-Menten equation is shown below, using the
dP=dt k2 EA vo 10
more modern concepts of initial velocity vo , and the
steady-state assumption. The EA complex has a non- Rearranging Eq. (10):
covalent bond. EA vo =k2 11
Let Eo total enzyme concentration
Ao total substrate concentration, which is Therefore
assumed to be much larger than Eo ; vo k2 Eo Ao =Km Ao 12
therefore, A Ao during initial velo-
city, vo , measurements < 5% A con- The maximum velocity, Vmax , is attained when all
verted to P during time of experiment) active sites of the enzyme are combined with substrate
EA enzymesubstrate concentration, which so that EA Eo . Therefore,
is a noncovalent complex k2 Eo equals Vmax and Eq. (12) gives Eq. (13):
P product concentration (one product is vo vmax Ao =Km Ao 13
formed in the general derivation; more
than one product may be formed and
released in a sequential manner in
C. Concept of Initial Velocity in Enzymology
many examples)
E free enzyme concentration
The importance of determining the initial velocity of
The rate of formation of EA is [see Eq. (1)]:
enzyme-catalyzed reactions was not fully appreciated
dEA=dt k1 EAo 2 and articulated until the early 1960s. Up to that time,
rates of enzyme-catalyzed reactions were frequently
While the rate of disappearance of EA is based on one-point assaysi.e., at a given time of
dEA=dt k1 EA k2 EA 3 the reaction. Also, reactions often were designed so
that many minutes or hours were used in the assays.
When steady-state conditions are reached (within a few Another reason is that the kinetics of non-enzyme-cat-
msec), dEA=dt dEA=dt so that alyzed reactions by chemists are designed to determine
k1 EAo k1 EA k2 EA k1 k2 EA 4 the order of reactions which must be done under con-
Rearranging: ditions of several half-lives of the reaction; i.e., > 90%
of the substrate must be converted to product to give
EAo =EA k1 k2 =k1 Km 5 four half-lives.
The order of chemical reactions is of major impor-
Km , the Michaelis constant, is given by either EAo = tance in determining the mechanisms of reactions.
ES or k1 k2 =k1 . Often the order is not zero, rst or second order, but
The free enzyme concentration, E, cannot be mea- is a fractional order (such as 1.5 for example). Not so
sured experimentally but it is related to Eo by the with enzyme-catalyzed reactions, which are rst, zero,
conservation equation or mixed order, depending on Ao in relation to Km for
E Eo EA 6 reasons explained later in this chapter.
Figure 2 shows the results of determining the velo-
Substitution of Eq. (6) for [E] in Eq (5) gives city of an enzyme-catalyzed reaction at the same pH,
20
21
These are very important concepts and equations
for the serious enzymologist. More information on
this subject is found in the three key articles by
Figure 4 Plot of the reciprocal of initial velocity, vo , versus Cleland (8). A detailed discussion is found in
reciprocal of initial substrate concentration, Ao , according Whitaker (1) on how to experimentally translate
to the Lineweaver-Burk method. (From Ref. 1.) these concepts into practice.
Table 2 Importance of Phosphate, Ribose, and Purine and Pyrimidine Bases in Cofactors
B. Unique Features of Cofactors These reactions [Eqs. (28) and (29)] follow a Ping-Pong
BiBi Mechanism, as described by the Cleland nomen-
Only limited examples of unique features of cofactors clature (Eq. 30).
can be given in this chapter. The reader is referred to
Whitaker (1) and specic references therein devoted to
detailed treatment of this very large topic.
1. Two Types of Cofactors
30
Mechanistically, the cofactors can be divided into two
groups, the coenzymes and the prosthetic groups. The Note that the prosthetic group FAD remains bound
coenzymes are loosely bound to the protein and gen- tightly to the enzyme E-FAD indicates covalent
erally act as a substrate in the catalyzed reaction. The bond), the E-FAD is reduced to E-FADH2 by removal
prosthetic groups are more tightly bound, some cova- of the two hydrogens from C1 of glucose and that the
lently, and they are regenerated to the starting form at E-FADH2 is oxidized back to E-FAD by O2 , permit-
the end of the reaction. An example of each type will ting it to be recycled catalytically in the process. The
clarify these differences. reaction takes place in two steps with only one sub-
There are many enzymes that require NAD or strate bound to the enzyme at a time.
NADH as coenzymes. The reaction catalyzed by alco-
hol dehydrogenase is shown in Eq. (27). 2. Same Cofactor, Different Reactions
k1 Earlier in this chapter, it was stated that Zn2 serves as
CH3 CH2 OH NAD CH3 CHO NADH H cofactor for > 150 different enzyme-catalyzed reac-
k1
tions. But the example chosen here to explain this con-
27
cept is the prosthetic group pyridoxal phosphate.
This is a reaction requiring the substrate ethanol and Pyridoxal phosphate is involved in amino acid conver-
the coenzyme NAD . NAD must bind to the enzyme sion to different products, as shown in Figure 7. The
rst, followed by binding of the ethanol. After cataly- prosthetic group binds to the amino acid in the same
sis, the product acetaldehyde dissociates from the way, forming a Schiff base intermediate, as shown by
enzyme, followed by NADH. This is an ordered BiBi the upper central formula. Which of the reactions is
Mechanism (see Sec. II.A.2). The reaction is a two- catalyzed depends on the specicity of the protein
substratethree-product (counting the H ) reaction. (apoenzyme) and the specic nature of the amino
It is a reversible reaction with the backward reaction acid. Therefore, pyridoxal phosphate and apoenzyme
favored. Coenzymes, being loosely bound to the together bind to the amino acid (substrate) to perform
enzyme, are generally lost during purication of the racemization, decarboxylation, deamination, , elim-
enzyme. The researcher must add the coenzyme back ination, , elimination, and addition reactions (to
to the inactive enzyme to restore activity. The concen- indole to form tryptophan) to give a variety of pro-
tration of both the apoenzyme and coenzyme will ducts.
V. KINETIC CONSEQUENCES OF
ness of fruits and vegetables. Many pharmaceutical
ENZYME INHIBITORS
compounds are designed to kill microorganisms by
A. Nature of Inhibitors selectively inhibiting their enzyme systems.
Mechanistically, there are two major groups of
Enzyme inhibitors are dened as any compound, except enzyme inhibitors. One group inhibits enzymes by
H ions (see pH effects), which decrease the activity reactions involving covalent bond formation (irrever-
when added to an enzyme reaction. Many enzyme inhi- sible inhibitors); the other group inhibits enzymes by
bitors are known. They are used to kill insects or reversible noncovalent bond formation (reversible inhi-
unwanted plants (herbicides) or animals (rotenone). bitors). Both types are important as they decrease or
They are used to prevent browning or preserve fresh- eliminate enzyme activity.
Figure 9 Linear competitive inhibition, plotted by the Figure 10 Simple linear noncompetitive inhibition, plotted
Lineweaver-Burk method. V Vmax in absence of inhibitor. by the Lineweaver-Burk method. See Figure 9 for additional
V 0 Vmax 1 Io =Ki in presence of inhibitor, K Km in explanation. (From Ref. 1.)
absence of inhibitor. K 0 1 Io =Ki in presence of inhibi-
tor. (Ref. 1.)
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
Figure 13 pH effect on activity of several enzymes. (a)
Pepsin acting on acetyl-L-phenylalanyl-L-diiodotyrosine at
36.78C with 5 104 M pepsin, reaction time 15 min (11).
(b) Ficus glabrata peroxidase acting on 0.03 M guaiacol and
Figure 11 Linear uncompetitive inhibition, plotted by the 0.005 M H2 O2 at 30.08C (12). (c) Trypsin acting on casein
Lineweaver-Burk method. See Figure 9 for additional expla- (13). (d) Hydrolysis of 5 104 M p-nitrophenyl phosphate
nation. (From Ref. 1.) by crude milk alkaline phosphatase at 358C in 0.1 M sodium
glycinate buffer, reaction time 15 min (J.R. Whitaker, unpub-
lished data). (From Ref. 1.)
a
Located at the end of polypeptide chain. Calories 4.186Joules.
Relative
Ea n0 =n ratesa
Substrate Catalyst (kcal/mol) (258C) (258C)
The overall theoretical rate enhancement shown in 7. EL King, C Altman. A schematic method of deriving
Table 6 is 1018 1036 . The best enzymes, such as catalase the rate laws for enzyme-catalyzed reactions. J Phys
and peroxidase, have rate enhancements 1020 . Chem 60:13751378, 1956.
Therefore the catalytic efciency of enzymes can be 8. WW Cleland. The kinetics of enzyme-catalyzed reac-
tions with two or more substrates or products. I.
explained by well-known chemical principles: enzymes
Nomenclature and rate questions. II. Inhibition:
are not magicians.
nomenclature and theory. III. Prediction of initial
velocity and inhibition by inspection. Biochim
Biophys Acta 67:104137, 173187, 188196, 1963.
REFERENCES 9. J Monod, J Wyman, JP Changeux. On the nature of
allosteric transitions: a plausible model. J Mol Biol
1. JR Whitaker. Principles of Enzymology for the Food 12:88118, 1965.
Sciences. 2nd ed. New York: Marcel Dekker, 1994. 10. DE Koshland Jr, G Nemethy, D Filmer. Comparison
2. AJ Brown. Enzyme action. J Chem Soc 81:373379, of experimental data and theoretical models in pro-
1902. teins containing subunits. Biochemistry 5:365385,
3. V Henri. General Laws of the Action of Diastases. 1966.
Paris: Hermann, 1903. 11. LE Baker. New synthetic substrates for pepsin. J Biol
4. L Michaelis, ML Menten. The kinetics of invertase Chem 193:809819, 1951.
action. Biochem Z 49:333369, 1913. 12. S Kon, JR Whitaker. Separation and partial charac-
5. M Eigen, GG Hammes. Elementary steps in enzyme terization of the peroxidases of Ficus glabrata latex. J
reactions. Adv Enzymol 25:138, 1963. Food Sci 30:977985, 1965.
6. H Lineweaver, D Burk. Determination of enzyme dis- 13. JH Northrop, M Kunitz, RM Herriott. Crystalline
sociation constants. J Am Chem Soc 56:658666, Enzymes. New York: Columbia University Press,
1934. 1948, p 12.
Inactivation of Enzymes
From Experimental Design to Kinetic Modeling
n1 A A0 expkt 6
n 6 1 A1n A01n n 1 kt 7
Figure 3 Thermal inactivation curves of tomato polygalac- Figure 4 Thermal inactivation curves of tomato polygalac-
turonase (PG) dissolved in 40 mM Na acetate buffer (pH 4.4) turonase (PG) dissolved in 40 mM Na acetate buffer (pH 4.4)
at 708C ( ), 758C ( ), 808C ( ), 858C (), and 908C ( ), at 588C ( ), 608C ( ), 628C ( ), and 658C ( ), modeled
modeled using a distinct isozyme model. Ao and A are initial using a fractional conversion model. Ao and A are initial
tomato PG activity and activity after thermal treatment, tomato PG activity and activity after thermal treatment,
respectively. (From Ref. 15.) respectively. (From Ref. 15.)
Figure 5 Arrhenius plot for thermal inactivation of orange Figure 6 Eyring plot for pressure inactivation at 458C of
pectinmethylesterase, dissolved in distilled water. (From Ref. avocado polyphenoloxidase, dissolved in phosphate buffer
17.) (pH 7, 0.1 M). (From Ref. 19.)
Danielle P. Praaning
DSM Food Specialties, Delft, The Netherlands
Enzymes were used for food production 2000 years In the European Union, there are so-called horizontal
before the birth of Christ. Good examples are bread, and vertical laws. Horizontal laws are those that cover
cheese, and beer production. Of course, it was all foods. Examples are the laws regarding additives
unknown at the time that it was the enzymes that and those regarding the labeling of foodstuffs.
were doing the job. Nowadays, the production of Vertical laws cover foods with a so-called standard of
food enzymes has become industrialized. European identity, also called compositional regulations or
enzyme manufacturers are market leaders and cater recipe laws. Examples of products for which such
70% of the worlds need for food enzymes, and this standards exist are cheese and bread.
market is still increasing with an annual growth of In both cases, vertical as well as horizontal, there
about 2%. are laws that have been harmonized throughout the
These developments have triggered the creation E.U. and laws that are still subject to national legisla-
of national laws and regulations to ensure both tion. This situation will probably remain so, especially
the safety of the consumer and fair trade. in view of the fact that in the treaty signed in
Although many laws in the European Union Maastricht in February 1992 (1), the European
have been harmonized over the last few decades, Union ofcially laid down the principle of subsidiar-
this has not (yet) happened in the case of ity, which states that the European Commission
enzymes. should not regulate what could be done just as well
The European Association of Manufacturers of or better at the national level.
Fermentation Enzyme Products (AMFEP), which For products regulated solely at the national level,
was founded in 1977, has set as one of its major the principle of mutual recognition applies. If a pro-
aims the achieving of the widest possible harmoniza- duct is legally produced in one E.U. country, it can
tion of regulatory requirements for enzymes in the also be sold in another E.U. country, unless this coun-
European Union (E.U.). To achieve this, AMFEP try has well-founded arguments to reject the product
has initiated discussions with the European on the grounds of protection of consumer health, or
Commission, national authorities, and the food indus- one of the other reasons listed in Article 30 of the
try. This chapter presents an overview of the tangle of Treaty establishing the European Community (2).
present regulations that are applicable for enzymes Other than in this situation, member states are not
when used in food. allowed to create unjustied trade barriers.
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
I. INTRODUCTION the Food Safety and Inspection Service FSIS), and the
Grain Inspection, Packers, and Stockyard
Thousands of enzymes are found in all of the raw Administration (GIPSA) or by a state department of
materials (plants, animals, microorganisms) produced agriculture. Increases, or decreases, of some enzyme
for foods. In those consumed raw, the enzymes are levels, such as polyphenol oxidase, can be done by stan-
active when ingested but are generally denatured by dard breeding methods. The insertion of a new gene
the acid pH of the stomach. They then become a and its expression to change the sensory, nutrition, cli-
(small) part of our total daily protein requirement. mate tolerance, insect resistance, or storage properties
When foods are processed prior to eating by heating, of the living organism is generally done by genetic engi-
the enzymes are denatured. In dried and freeze-dried neering. The inventor must demonstrate that there
products, the enzymes may be in active form. In fruits are no harmful effects to humans or the environment,
and vegetables, such as apricots, apples, dates, plums, including other types of activities (allergenic, for exam-
potatoes, mushrooms, teas, etc., the enzymes are still ple) of the food produced from such modications. The
active and must be inactivated by blanching ( 858C, federal and/or state departments of agriculture are
or controlled by adding enzyme inhibitors such as responsible for enforcing such modications to living
bisulte, ascorbic acid, or other specic inhibitors potential food sources.
before drying. But these naturally occurring enzymes Additions of enzymes to food materials during pro-
are not regulated. If an enzyme is added (by biotech- cessing are regulated by the Food and Drug
nology) to a plant to increase the level of enzyme, such Administration (FDA) unit of the Department of
as overproduction of 5-enolpyruvylshikimate-3-phos- Health and Human Services. Specically, the Center
phate synthase (EPSPS) or cloning a mutant gene for Food Safety and Applied Nutrition (CFSAN) of
encoding EPSPS that is insensitive to glyphosate to FDA is assigned this responsibility. In carrying out its
prevent killing of the plant by the herbicide glyphosate, responsibility, it may request advice from the National
while the unwanted grasses and weeds are killed, this is Academy of Sciences (NAS) or other expert bodies and
regulated. When an enzyme is added during the pro- scientists. The regulations of FDA become binding
cessing of a food, this is regulated. when published in the Federal Register (published by
Addition of enzymes, whether to increase the level of the Ofce of the Federal Register, National Archives
the enzyme or to include a new enzyme in a live plant or and Records Administration of the U.S. Government
animal or microbe to be used as food, is regulated by Printing Ofce).
the U.S. Department of Agriculture (USDA) via the The regulations are not only published in the
Animal and Plant Health Inspection Service (APHIS), Federal Register, but can be found also on several
Study Scope
Acute
Oral LD50 Two species (min) range nd for subacute studies.
Dermal
inhalation LD50 As needed (OSHA requirements, etc.).
Subacute
90-day feeding Two species: one nonrodent, dog preferred. Three levels (min) plus
control.
Complete biochemical and histopathologic workup on control,
highest level fed. Other levels if needed. Dene no-effect level.
High level to give effect.
Chronic
2 years Two speciesrat/dog. Carry rat to 25% survival of controls.
Perhaps combine with three-generation reproduction study.
Three levels plus control. Complete biochemical and
histopathologic workup.
Metabolic fate Search for animal model for man. Absorption, distribution kinetics,
organ/tissue concentration (if any) metabolic products dened.
Special Carcinogenicity, teratology, cytogenicity (host mediated, dominant
lethal, chick embryo).
Interaction Special dietary restrictions, drug interactions.
Source: Ref. 8.
Since the beginning of federal regulations involving 170Food Additives is a general section for all food
enzymes they have been, and continue to be, classied additives (pp 510). 171Food Additive Petitions is a
as food additives. As noted earlier in this chapter, they general section on how to petition FDA for approval
were originally classied as GRAS. But after 1958, of a food additive (pp 2126). Both must be consulted
when the Food Additive Amendment was passed, for the general aspects pertaining to denitions, classi-
safety of the enzyme-containing product had to be cation of foods and ingredients, and general policies
addressed, and in 1972, all GRAS status was revoked. on food additive regulations. The topics covered are
As pointed out in Section II above, these changes given in Table 2.
require extensive data in the petition asking for rein-
statement of GRAS status.
Regulation of food additives is covered in the Code
of Federal Regulation Title 21, 170199 (11). B. 173Secondary Direct Food Additives
Enzymes are covered collectively with other food addi-
tives in Part 170Food Additives, Chapter 1, Food 173Secondary Direct Food Additives Permitted in
and Drug Administration, Department of Health and Food for Human Consumption, Subpart BEnzyme
Human Services Subchapter B, Food for Human Preparations and for a Human Consumption, Subpart
Consumption as to regulation procedures, in 171, BEnzyme Preparations and Microorganisms lists
Food Additive Petitions; in 173, Secondary Direct those enzymes and microorganisms that have been
Food Additives Permitted in Food for Human afrmed by FDA and on what basis they were
Consumption; and in 184, Direct Food Substances approved (Table 3). Note that all the enzymes are
Afrmed as Generally Recognized as Safe. from microorganisms.
Table 3 Secondary Direct Enzymes and Microorganisms Permitted in Food for Human Consumptiona
Figure 4 A schematic representation of an ion pair network cluster in glutamate dehydrogenase from Pyrococcus furiosus. The
twofold axis relating to a CC# dimer is indicated by a diad. Potential ion pair interactions are indicated by dashed lines. The
backbone of the polypeptide chain is shown as a gray ribbon. The individual amino acids are indicated by their one letter code,
followed by their sequence number. (From Ref. 79. Copyright 1998 Blackwell Science Ltd.)
groups together in the folded state, it may not out- less unfavorable, so ion pair formation would become
weigh the unfavorable desolvation (free energy of increasingly favorable at higher temperatures and
hydration) cost. Elcock (92) has recently shown that could provide a signicant stabilization to hyperther-
at higher temperatures, the desolvation effect becomes mophilic proteins at elevated temperatures. It has also
William E. Newton
The Virginia Polytechnic Institute and State University, Blacksburg, Virginia, U.S.A.
E. Role of MgATP in Nitrogenase Catalysis formational changes in both nitrogenase proteins that
ensure delivery to and accumulation within the MoFe-
The overall reduction of N2 to yield 2NH3 is thermo- protein of multiple electrons prior to their delivery to
dynamically favorable, although the formation of two- substrate. In support of this concept, primary amino
electron-reduced intermediates, if they exist, is not. It acid sequence and structural comparisons show that
appears, then, that MgATP binding and hydrolysis the Fe-protein is a member of a large class of signal
during nitrogenase catalysis is not a thermodynamic transduction proteins that undergo conformational
requirement but rather it is used to drive electron changes upon MgATP binding and hydrolysis.
transfer toward substrate reduction and to prevent The following sequence of events, which occur dur-
the back ow of electrons to the Fe-protein. This situa- ing a single turn of the Fe-protein cycle and which
tion may be envisioned as the gating of electron ow account for the role of MgATP in nitrogenase cataly-
in which MgATP binding and hydrolysis causes con- sis, is widely accepted (77). First, the reduced Fe-pro-
y Deceased.
Abbreviations: ADP-Glc, adenosine diphosphoglucose; AGP, adenosine diphosphoglucose pyrophosphorylase; ATP, adeno-
sine triphosphate; ae, amylose extender mutant; BT2, brittle-2; GBSS, granule-bound starch synthase; Glc-1-P, glucose-1-phos-
phate; SBE, starch branching enzyme; SDBE, starch debranching enzyme; SH2, shunken-2; SS, starch synthase; su-1, sugary-1
mutant.
Chain elongation reactions use ADP-Glc as sub- III. STARCH BIOSYNTHETIC ENZYMES
strate. Starch synthases are glycosyltransferases that AND THEIR ISOFORMS
elongate oligosaccharide chains by successive addition
of (1,4)--linked glucose units to the nonreducing end Many isoforms of AGP, SS, and SBE have been puri-
of growing -glucan chains. There are at least two ed, characterized, and cloned from a variety of plants.
classes of starch synthase. The GBSSI, also known as The occurrence of having multiple isoforms encoded
the Waxy protein, is responsible for the synthesis of by distinct genes may be signicant for several reasons.
amylose. The soluble starch synthases are believed to First, each specic isoform may possess distinct cata-
be involved in chain elongation of amylopectins. lytic properties and therefore determine parameters
Starch chain length is thought to be controlled by spe- such as chain length or the spacing of branch points.
cic starch synthase isoforms. For example, GBSSI is specically responsible for the
The (1,6)--linked branch linkages are introduced synthesis of amylose, and waxy mutant maize which
by isoforms of SBE, which are believed to operate by lacks GBSSI is decient in amylose. In addition, the
a cut-and-past type mechanism (5). SBE is thought to isoforms of SBEI and SBEII transfer chains of differ-
bind to the sites where adjacent starch chains are jux- ent length. Second, the occurrence of multiple isoforms
taposed. The enzyme then cleaves at a (1,4)--linkage, enables differential expression with regard to both tis-
and the newly generated reducing end is then attached sue distribution, and plant growth and differentiation.
at another site to form a (1,6)--branch point. Spacing Therefore, the identication and characterization of
between branch points may be controlled by specic these isoforms is essential to the use of genetic engi-
SBE isoforms. neering to improve starch yield and starch quality.
Essential fatty acids are polyunsaturated fatty acids of the omega-6 (n-6) and omega-3 (n-3) families that are essential for life and good health,
and are not produced in man and must be obtained from plants as part of the diet.
is the high incidence of unsaturated and polyunsatu- plant sources for the synthesis of the other n6 and n3
rated fatty acids and particularly those of 18 carbon metabolites. Hence, these are the so-called essential
atoms. fatty acids (EFAs) and are required for the normal
Unsaturation generally describes a four-electron structure of all membranes, for cholesterol transport,
(double bond/olenic) linkage between carbon atoms. for the maintenance of normal permeability barriers
Polyunsaturated fatty acids (the so-called PUFAs) are and as precursors of eicosanoids (prostaglandins,
polyenoic acids having more than one olenic center, thromboxanes, leukotrienes).
the double bonds generally of a cis conguration and A further unsaturated fatty acid terminology, of
methylene interrupted. By convention PUFAs are particular value to biochemists, denes the position
described as being n3 or n6 fatty acids depending on of the double bonds from carbon atom 1 (i.e., the
whether the rst double bond is three or six carbon end of the molecule). The delta system is convenient
atoms from the methyl end (omega carbon), respec- for denoting all double bonds, their position in the
tively. Further double bonds are separated from each carbon chain, and the specic desaturase enzymes
other by a methylene group. Such fatty acids are also which catalyze their introduction into the molecule.
termed omega-3 and omega-6 acids. The omega-3 ser- Hence, linoleic acid and -linolenic acid become
ies are derived from -linolenic acid (C18:3 n3) by C18:29;12 and C18:39;12;15 , respectively. Fatty acid
chain elongation and further desaturation and include structures and trivial and systemic nomenclatures are
eicosapentaenoic acid (EPA; C20:5 n3) and docosa- given in Table 1. Desaturases, which abstract hydrogen
hexaenoic acid (DHA); C22:6 n3). On the other atoms from adjacent CH2 groups to produce double
hand, the omega-6 acids are derived from linoleic bonds, can be specically dened. For example, a 12
acid (C18:2 n6) and include -linolenic acid (GLA; desaturase or a 5 desaturase would introduce double
C18:3 n6) and arachidonic acid (AA; C20:4 n6). bonds at the 12 and 5 carbon atom in the acyl
Linoleic acid and linolenic acid are substances which chain, respectively.
cannot be synthesized de novo in the human body and Triacylglycerols, the major constituents of the oil,
so, like vitamins, must be provided in the diet from consist of a glycerol backbone to which a fatty acid
The reactions that channel oleate to phosphatidylcho- form R +H. The initiating species may be, for example,
line for its subsequent desaturation help to concentrate a transition metal ion, a radical generated by photolysis
substrate and give a greater chance(s) for desaturation. or high-energy irradiation, or a radical generated by
It is possible that desaturation of complex lipids, such decomposition of a hydroperoxide. In the propagative
as phosphatidylcholine, is not an efcient process and phase, further reactive species are generated as in the
that a number of mechanisms have evolved, including generation of a radical R from the alkene RH and its
transacylation, which enables the rapid formation of subsequent reaction with oxygen. In the termination
polyunsaturated fatty acids for storage and membrane phase there is an interaction of radicals to produce non-
lipid assembly. One might also speculate that the trans- initiating and nonpropagating products.
acylation may play some role in the assembly of tria- Autoxidation is facilitated by pro-oxidants and
cylglycerols containing > 70% of a particular fatty inhibited by antioxidants. Pro-oxidants, such as metals
acid and particularly those of an unusual nature per- or other radical initiators, operate by promoting the
haps by giving rise to species of diacylglycerols which initiation step or else they may inhibit the activity of
are immediately acylated directly with the required antioxidants. Antioxidants are frequently added to fats
fatty acid. Transacylation, however, could have a and fat-containing foodstuffs to prolong shelf life
more profound biochemical importance. It may help (108). These are often phenolic compounds which act
to regulate the size of an active diacylglycerol pool, a by interferring with the propagation sequence by con-
possible necessity which prevents the detrimental net serving propagating radicals into nonpropagating spe-
movement of phosphatidylcholine to diacylglycerols cies (109, 110). Their effectiveness is often increased by
(via cholinephosphotransferase) and the consumption compounds such as citric acid, ascorbic acid, or phos-
of endoplasmic reticulum structural lipid during the phoric acid (called synergists). In this regard, there is
period of extremely rapid oil deposition that is found considerable evidence for a contributory, antioxidant
in the developing seed. role for vitamins E and C and the carotenoidscon-
stituents of fruit, vegetables, beverages, and grainsin
the maintenance of health and the protection from
V. OXIDIZED FATTY ACIDS
coronary heart disease, cataracts, and certain cancers
A. Autoxidation and Antioxidants (111).
Recent work indicates possible important roles of
Historically, the oxidation of fatty acids has been per- polyphenolic components (the avonoids, phenylpro-
ceived as a deleterious process resulting in the genera- panoids, and phenolic acids) as contributors to antiox-
tion of off-avors in commercial produce. The process idant activities and also as agents of other mechanisms
of oxidation can be considered as either autoxidation, a contributing to cardioprotective and anticarcinogenic
chemical reaction that usually takes place at ambient actions. Epidemiological evidence suggests a negative
temperatures between atmospheric oxygen and an correlation between dietary antioxidants and coronary
organic compound, or as an enzyme-catalyzed reaction. arterial disease. Such an association may involve a
The complexity of the autoxidation reaction has been mechanism whereby antioxidants delay the onset or
well documented (106). The insertion of oxygen is also progression of the disease by reducing oxidative reac-
accompanied by the movement of double bonds result- tions, downregulating thrombic mechanisms, and
ing in the formation of a conjugated diene with a max- attenuating endothelial dysfunction.
imal absorbance at 234 nm. Spectral measurements at A large range of avonoids and phenylpropanoids
this wavelength have been used to monitor hydroper- found in fruit, vegetables, and beverages are more
Ralph E. Dewey
North Carolina State University, Raleigh, North Carolina, U.S.A.
Anthony J. Kinney
DuPont Experimental Station, Wilmington, Delaware, U.S.A.
Colja Laane
DSM Food Specialties, Delft, The Netherlands
Yvonne Bruggeman
Unilever Research, Vlaardingen, The Netherlands
Chris Winkel
Quest International, Naarden, The Netherlands
Lipoxygenases (LOX) In situ and in vitro Dough extensibility Flour bleaching Destruction vitamin A
(off)-avor and strength
production
Alcohol oxidases and In situ and in vitro AO plus catalase, O2
dehydrogenases avor production removal
(AO, ADH) (ILOX)
Sulfhydryl oxidases Removal of cooked Dough strengthening in
(SOX) avor in UHT-milk combination with
POX
Peroxidases (POX, In vitro avor Crosslinking Assisting PPO in color LPO, antimicrobial agent
LPO) production and biopolymers formation
debittering
(Poly)phenol oxidases Debittering of coffee, (Dis)coloration; in vitro Laccase, O2 removal
(PPO) cacao, and olives production of colors
Carbohydrate oxidases GOX, mild acid Dough strengthening GOX plus catalase, O2
(GOX, HOX, production via H2 O2 ; thickening removal
PYROX) agent
Ascorbic acid oxidase Browning AAO plus catalase, O2 Destruction vitamin C
(AAO) removal
Xanthine oxidases (XO) Antimicrobial agent
Superoxide dismutases Removal reactive O
(SOD) species with catalase
Catalases Removal excess H2 O2
Cholesterol Cholesterol reduction
oxidoreductases
Anne S. Ponstein
Syngenta Mogen, Leiden, The Netherlands
Rob F. Beudeker
DSM Food Specialties, Delft, The Netherlands
Jan Pen
PlantZyme, Leiden, The Netherlands
Transgenic Transgenic
plants animals CHO cells Insect cells Fungi Yeast Bacteria
Homogenization of transgenic leaf material, followed by in vitro addition of the enzyme. In some cases as
by incubation at 95 C, resulted in conversion of endo- wood pulping, in vitro addition has limited success
genous starch by the transgene product (Table 3). In because the enzymes do not have sufcient access to
starch potatoes, the -amylase levels as obtained in the substrate. Finally, expression in the crop guaran-
tobacco (7, 66) would be sufcient to hydrolyze all tees an optimal dispersion of the enzyme throughout
the starch present in the tuber. Similar applications the biomass; i.e., an optimal homogeneous mixture of
for -amylase and other enzymes may be found in enzyme and substrate is obtained. This is less easily
other industrial processes, like for instance the applica- achieved by in vitro addition of enzymes.
tion of -amylase in the production of bioethanol,
heat-stable -glucanases for the degradation of cell C. Seed-Formulated Enzymes (Fig. 1c)
wall material in the brewing industry, and the produc-
tion of fatty acids in oil plants (23). Another concept for application is the direct use of
In this approach the aim is not producing enzymes engineering plant material in the processing, food, or
in crops per se, but creating either a higher-quality feed industry. The transgenic plant material serves as
input resource or an enabling process through expres- novel formulation for the expressed enzyme. Other
sion of the enzyme inside the plant. This application plant material can be used, depending on the process,
has unique advantages. Firstly, enzyme manufacturing but in general seeds may be the plant organ of choice,
costs will be low, because there is no need for purica- because these create a stable environment for long-
tion and formulation of the enzymes. Secondly, it obvi- term storage of the enzymes. The potential of plants
ates the necessity for separate production of enzymes. as production vehicles for enzymes is enormously
Thirdly, expression in the vicinity of the substrate increased by producing the enzyme in seeds and
assures better access to its substrate than is achieved directly applying these in the industrial process.
Sample 0 30 60
Control tobacco
Control tobacco 0:02 T.A.U. B. licheniformis -amylase/mg = =
Control tobacco 0:2 T.A.U. B. licheniformis -amylase/mg =
Control tobacco 2 T.A.U. B. licheniformis-amylase/mg
Transgenic tobacco line MOG227.3
Homogenization of transgenic leaf material containing Bacillus licheniformis -amylase, followed by incubation at 95 C for 30 min hydrolyzed
starch present in the leaves. Addition of 2 T.A.U./mg Bacillus licheniformis -amylase to the nontransgenic control leaf homogenate for 30 min or
of 0.2 T.A.U./mg for 60 min had the same effect, while incubation of nontransgenic control leaf homogenate had no visible effect on the
endogenous starch during the time of the experiment. The presence of starch, as demonstrated by I2 =KI-staining is indicated with , intermediate
staining with = and absence with .
Compared with the production followed by purica- This enzymes-in-seed approach is exemplied by the
tion, this concept has additional advantages over tra- direct application of milled transgenic -amylase seeds
ditional production systems. Firstly, the formulation in liquefaction of corn and potato starch (Fig. 2). Seeds
of the enzyme as obtained by genetic engineering of of a transgenic plant line were milled and used without
the crop is convenient; i.e., it does not require an any further purication in the liquefaction of potato
extra step in the manufacturing process, and it guar- and corn starch. From the nature and the quality of the
antees stable and safe storage of the enzyme. No aller- hydrolysis products obtained from corn and potato
genic problems due to the enzymes are to be expected, starch with the transgenic seeds it can be concluded
because these are contained in the seeds. Secondly, the that the enzyme as produced in tobacco is as suited
manufacturing costs will be lower, because purication for liquefaction of starch as the Bacillus licheniformis
and formulation of the enzyme are no longer required. amylase. As shown in Table 5, the dextrose equivalent
This concept is most easily applied in, but certainly not (DE) values, which were calculated from the HPLC
limited to, cases where enzymes are required for appli- patterns, were found to be similar to those obtained
cation in (processed) food or feed. The enzyme can with the commercial preparations and within the com-
then be expressed in the plants and plant organs that mercially acceptable range (DE-12, preferably DE16;
are the natural component in this animal feed and 67). Seeds containing -amylase maintained full
food. The transgenic plant material containing the enzyme stability for at least a year. When seeds are
enzyme is used as delivery vehicle for the protein the only organs used, seed-specic expression is pre-
with the same advantages as described above. ferred, because it will minimize the chance of effects
The transgenic seeds are a novel specialty enzyme on agronomic characteristics.
product to be added in small quantities into the indus-
trial process, provided the expression level is suf-
ciently high (additive route). This creates
exibility, because various enzymes can be engineered Table 4 Comparison of Bulk and Additive Application of
into plants in the same manner and used and combined Enzymes in Seed
according to the needs of the user. The total premium Additive Bulk
for production, transportation, etc., of transgenic plant
material will be lower than in case the transgene is built Expression level High Low
into the commodity crop. Production, transportation, Research costs Medium Low
and marketing can be strictly and simply controlled, Amount of plant material Low High
preventing the mingling of transgenic with nontrans- Importance host Low High
genic seeds. Alternatively, the enzyme can be expressed Agronomic performance Less critical Critical
Development costs Low High
at low levels in all (or a large part of the) seeds used in
IP production ?
the application (bulk route). A comparison between Level of control High Low
these two approaches in various aspects is given in Maintenance added value High Low
Table 4.
Potato Corn
DE values starch starch
VII. CONCLUSION
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
We shall not unravel the chemistry of life in mole- chromatography by Moore and Stein (4, 5) for separ-
cular detail without knowing at atomic or close to ating, identifying, and quantifying amino acids from
atomic resolution the structure of the biological protein hydrolysates, the Edman method of sequencing
macromolecules especially the proteins (1). proteins (6, 7), and the development of chromato-
graphic (4) and electrophoretic methods (8) for puri-
cation of amino acids and proteins all contributed to
I. INTRODUCTION
rapid advances in understanding the protein nature of The discovery of the chemical nature of the gene
enzymes. resulted from the merging of genetics and biochemistry
During that period, limited thought was given to between 1930 and 1950. In 1909, Archibald Garrod
protein biosynthesis. It was generally assumed that determined that the human disease alkaptonuria was
proteins were formed from activated derivatives of caused by a rare recessive mutation that was inherited
amino acids, probably by use of ATP for activation. according to Mendelian rules (8e). It was noticed that
But no details of the in vivo biosynthesis were known. patients with alkaptonuria had black urine. The dark
In the late 1950s and 1960s a few publications color was determined to be due to a breakdown pro-
appeared, based primarily on studies of microorgan- duct of phenylalanine. In his treatise on Inborn
isms, that began to give evidence that protein biosynth- Errors of Metabolism, Garrod wrote, We may
esis at the molecular level could be more complex than further conceive that the splitting of the benzene ring
originally thought. Especially, studies based on micro- in normal metabolism is the work of a special enzyme,
bial genetics indicated a precise correlation between [and] that in congential alkaptonuria the enzyme is
nucleotide sequences of DNAs and some RNAs and wanted.
the amino acid sequences of proteins. The research of Morgan and his colleagues in the
early 1900s on the inheritance of eye color in
Drosophila (8f), together with that of Beadle and
II. HISTORICAL BACKGROUND Tatum on the bread mold, Neurospora crassa, and its
rare mutant colony development, led Beadle and
There were many important contributions made in the Tatum to conclude that genes somehow control pro-
long journey of understanding protein biosynthesis tein (enzyme) structure (8g). These ndings led to
prior to the 1950s. Genes had been postulated before studies of the chemical structure of genes, as the rst
the turn of the 20th century based on the careful breed- step in elucidating the molecular basis for the genetic
ing studies of Gregor Mendel (8a). August Weismann control of protein synthesis. This occurred in 1944
in 1892 stated that the essence of heredity is the trans- when Avery et al., in studying Streptococcus pneumo-
mission of a nuclear substance of specic molecular niae from patients with pneumonia, showed that DNA
structure (8b). In 1903, Walter Sutton linked chromo- from smooth colonies can transform rough colo-
somes to Mendelian heredity (8c). He observed that all nies into the smooth form (9). They proved DNA was
cell divisions producing germ cells also formed thread- involved. Addition of several different proteases had
like chromosomes, and that a gamete received only one no effect on the transforming activity, eliminating
of each of the chromosomes of different morphologic involvement of proteins. However, small amounts of
types. Through detailed studies of the fruit y, puried deoxyribonuclease (DNAase) immediately
Drosophila melanogaster, Morgan and his colleagues inactivated the transformation.
showed that the genes segregated as four linkage In 1953, Watson and Crick deduced the double-heli-
groups corresponding to the four chromosomes (8d). cal structure of DNA, ushering in the modern era of
Similar observations followed for corn (10 chromo- molecular cell biology (10). At about this same time,
somes, 10 linkage groups) and peas (7 chromosomes, the electron microscope was invented. Light micro-
7 linkage groups). By 1920, the chromosome theory of scopes have a resolving power of 500 nm. Electron
heredity became an accepted fact in genetics. microscopes in the early stage had resolving power of
A. Differences Between Prokaryotic and There are ve steps in the biosynthesis of proteins:
Eukaryotic Protein Biosynthesis transcription phase in which mRNAs, tRNAs, and
rRNAs are produced; translation of mRNAs to give
Being unicellular, formation of bacterial mRNA is polypeptides; termination of translation; posttransla-
always accessible to ribosomes, as well as other com- tional modication of the polypeptides; and folding of
ponents of the protein biosynthesis pathway. Even as the polypeptides into their native structures.
the mRNA is being transcribed from DNA, the specic
tRNAs and tRNA synthetases begin translation of the 1. Transcription Phase
protein and transcription and translation are com-
pleted at almost the same rate and time. Therefore, The overall steps of transcription and translation in
the mRNA nucleotide bases are not modied (by eukaryotes are shown in Eq. (1).
methylation) as they are in eukaryotic cell mRNA. In RNA polymerase
eukaryotes, DNA transcription occurs in the nucleus, DNA 4 NTP ! mRNA, tRNA;
which contains ribosomal precursors undergoing for- transcription
mation (in the nucleolus), but there are no mature amino acids, tRNAs
ribosomes that can perform protein synthesis. In rRNA ! Proteins
ribosomes
eukaryotes, the mRNA is synthesized in the nucleus.
1
The mRNA must then be secreted through the nuclear
membrane into the cytoplasm for translation to occur. During the transcription phase, the mRNAs, tRNAs,
Therefore, transcription and translation are not and rRNAs are produced. The key enzyme is RNA
coupled in eukaryotic cells. polymerase II. The three steps are: initiation signalled
In eukaryotes, the RNA is not a functional mRNA, by the promoter recognized by RNA polymerase (in
tRNA, or rRNA until it is extensively modied in the the case of prokaryotes, it is a subunit of the RNA
nucleus before it is secreted into the cytoplasm where it polymerase (2 0 ; elongation involving a melting
associates with the ribosomes. Types of modication and unwinding of the double helix while sequentially
that occur in the nucleus are: (a) addition of chemical attaching ribonucleotides at the 3 0 end of the growing
groups at both ends of the mRNA; (b) cutting the RNA molecule; termination by the formation of stem-
mRNA into segments; and (c) recombining the seg- loop structures of Rho-dependent terminators.
ments by splicing of original noncontiguous segments The April 28, 2000, issue of Science published a
to produce a very different mRNA. Translation occurs detailed crystallographic study of the protein architec-
Structure 1
After synthesis of the three types of mRNAs in mRNA. At the end of the tRNA (rectangle labeled
eukaryotes, they must be secreted from the nucleus amino acid), there is an adeonsyl group to form a
via the cell membrane into the cytoplasm, where trans- covalent binding site for the specic amino acid (Ala
lation takes place. The time to produce the mRNA is shown here). The tRNA synthetase is responsible for
estimated to be about 12 sec, depending on the pro- the reaction leading to the bound aminoacyl-tRNA
moter clearance time, number of RNA polymerase [Eq. 2)].
molecules, and size of the gene.
enzyme
Amino acid ATP tRNA ! aminoacyl-
2
2. Translation Phase tRNA AMP PPi
The mRNA carries the genetic information for protein
biosynthesis that it has transcribed from a gene located Table 5 Degeneracy of the Genetic Code
on DNA via a three-letter code (codon), complemen-
tary to that of the gene. Table 4 lists the genetic codons Number of Amino acid Total number
for each of the 20 amino acids found in proteins. The synonymous of codons
codons
code contains some redundancies, such that 64 codons
(including the stop codons) may be used in protein 6 Leu, Ser, Arg 18
biosynthesis (Table 5). 4 Gly, Pro, Ala, Val, 20
There are tRNAs and tRNA synthetases for each Thr
one of the 20 amino acids found in nascent proteins. 3 Ile 3
The tRNAA1a for alanine is shown in Figure 4. The 2 Phe, Tyr, Cys, His, 18
tRNAs are small ribonucleotides of 7080 residues. Gln, Glu, Asn, Asp,
Some of the nucleotide bases are also modied follow- Lys
ing synthesis, as shown by the shaded circles in Figure 1 Met, Trp 2
Total number of codons for amino acids 61
4.
Number of codons for termination 3
The tRNAA1a contains an anticodon of IGC which Total number of codons in genetic code 64
recognizes and binds to the GCC codon on the
are illustrated with spaces separating them for conve- IV. POSTTRANSLATIONAL COVALENT
nience; in mRNA there are, of course, no spaces.) The MODIFICATION OF PROTEINS
translocation reaction is catalyzed in bacteria by the
elongation factor G, using energy from the hydrolysis Once the primary structure of a protein is completely
of GTP. With this movement, the empty tRNA is translated by RNA polymerase II, the protein is still
released from the P site, and the peptidyl-tRNA is biologically nonfunctional. It must undergo chemical
shifted to the P site. (In eukaryotes, too, there are modication involving nonhydrolytic and/or hydroly-
special proteins that serve in elongation.) The sequence tic modication of the protein, fold into the tertiary
of events is repeated for every amino acid added to the structure, form quaternary structure (if needed) and
growing chain. Note that two molecules of GTP are macromolecular structure (if needed), and combine
used in the addition of each amino acid. with cofactors (if needed) and often be transported to
the site where it performs its function. In this section,
c. Termination. When the ribosome arrives at the
we deal only with the posttranslational enzymatic
codon UAG, the translation is completed with the aid
modication.
of a terminator factor. Hydrolysis of the peptidyl-
tRNA on the ribosome releases the completed poly-
peptide and the last tRNA, and the two ribosomal A. Nonhydrolytic Posttranslational Enzymatic
subunits separate. At least three TFs are involved in Modication
E. coliRF1, RF2, and RF3. For eukaryotes, one
termination factor eRF recognizes all termination Table 6 lists examples of nonhydrolytic modication of
codons. 11 of the 20 amino acid residues, as well as the carboxy
and amino terminal ends of the proteins (1416). There 1 and one pro-2 chains align with each other with
is a highly specic enzyme for each of these modica- the N-terminal amino acid all at one end and the C-
tions. Some modications such as glycosylation and terminal amino acid at the other end. This triple helix
activation of the cascade of pro-proteins in blood clot- forms disulde bonds among the three molecules. At
ting involve a number of enzymes (see below). Most this point the procollagen (MW 115140 kDa for
likely, nonhydrolytic modications occur as soon as each molecule) is secreted through the membrane and
the N-terminal end of the growing protein chain transported to the location where needed. Collagen is
emerges from the ribosomal system. the major protein produced by animals. It forms a
A good example of posttranslational modication coating around all cells, brils, bundles of brils and,
of a protein is that for collagen (Fig. 8). As soon as higher structures, such as the sarcomeres and ber
a part of the nascent primary structure is formed on bundles of muscles.
the ribosome (polysome), hydroxylation of proline and At this point, highly specic and limited enzyme-
lysine (see Chapters 35 and 36) and glycosylation of catalyzed proteolysis occurs. The teleopeptides at the
hydroxylysine are known to occur (17, 18). Two pro-1 N-terminal end of the procollagen are removed by a
and one pro-2 chains are coded for by two different specic procollagen peptidase to give molecules of
mRNAs. While still in the unfolded state, the two pro- 95 kDa. The folded collagen molecules are aligned
Total Total
Amino acid Typical modications Amino acid Typical modications
(group) modications knowna (group) modications knowna
Arginine methyl-(mono- 6 Serine (hydroxyl) phospho-2 8
(guanidino) and di-)2 b glycosyl-2
ADP-ribosyl-2 methyl-2
citrulline phophopantetheine-2
ornithine ADP-ribosyl2
Lysine (
-NH2 ) glycosyl-2 33 Threonine phospho-2 6
phospho-2 (hydroxyl) glycosyl-2
pyridoxyl-2 methyl-2
biotinyl-2 Cysteine cystine1 7
lipoyl-2 (sulhydryl) glycosyl-2
acetyl-2 dehydroalanyl4
methyl-(mono-, heme
di-, tri-)-2 avin
-hydroxyl-1 seleno-
-glycosyl-2
crosslinks Aspartic acid/ -carboxyl-1 6
2 glutamic acid -phospho-2
Histidine methyl- (1- and 3-) 6 (carboxyl)a methyl-2
(imidazole) phospho- (1- and 3-)2
iodo- Asparagine/ glycosyl2 8
avin- glutamine R-NH-
(amide)a pyrrolidone
Proline 4-hydroxyl-1 5
3-hydroxyl-1 Carboxyl -amide6 11
3,4-dihydroxy-1 terminal residue -amino acid6
4-glycosyloxy-2 (carboxyl)
Phenylalanine -hydroxy-1 2 Amino terminal acetyl-2 19
(benzene ring) -glycosyloxy-2 residue (amino formyl-2
group) glucosyl-2
Tyrosine -hydroxy-1 18 amino acyl-2
(benzene ring/ -glycosyloxy-2 pyruvyl-2
hydroxyl) sulfono-1 -ketobutyryl-2
iodo- (mono-, di-)2 methyl-2
brom- (mono-, di-)2 glycuronyl-2
chloro- (mono-, di-)2 murein2
bis-ester1
adenylyl2 Total 135
uridylyl-2 modications
RNA
a
Including three each for aspartic acid and glutamic acid and four each for asparagine and glutamine.
b
General class of enzymatic reaction: 1 oxidation/reduction; 2 transfer; 4 formation of double bond or addition to double bond; 6 ligation.
Source: Refs. 14, 16.
both end to end in a staggered array and laterally to B. Hydrolytic Posttranslational Enzymatic
give cylindrical brils with diameter range from 50 Modication
2000 A, depending on the tissue and stage of develop-
ment. There are further crosslinkages, especially ins are probably modied by highly specic proteases,
through desmosine formation from four lysyl side since most proteins are secreted from the cell for trans-
chains by lysyl oxidase (see Chapter 37). port to other locations (organs and cells). Secretion
from the cell requires a signal peptide targeting the Preproinsulin (Fig. 9) is synthesized as a 100amino
nascent protein for transport. The signal peptide is acid polypeptide. Immediately on synthesis, a highly
required to translocate the protein out of the cell. specic protease in the luminal space of the endoplas-
Table 7 lists a number of physiological systems con- mic reticulum removes the N-terminal signal peptide
trolled by posttranslational proteolysis. Table 8 gives after amino acid 16 by hydrolysis. The remaining 84
some examples of highly specic proteases involved in amino acid proinsulin, in which two disulde bonds
physiological processes. have formed correctly (a posttranslational modica-
In addition to the posttranslational hydrolytic mod- tion), is transported to the Golgi body where a 33
ications to collagen (above), three other examples (of amino acid segment (the C segment in Fig. 9) is hydro-
many) will be given for preproinsulin, chymotrypsino- lyzed from proinsulin by two proteolytic events at two
gen A, and preproproteins in the blood-clotting cas- peptide bonds (Ala30 Arg31 and Arg63 Gly64 of proin-
cade. sulin) to give two peptides (A [21 amino acid residues]
Assembly bacteriophage
virus
membrane
procollagen ! collagen
fibrinogen ! fibrin
Defense reactions blood coagulation
brinolysis
complement reaction
Development maturation of spermatozoa,
release of ova, and fertilization
(proacrosin ! acrosin)
prochitin synthetase ! chitin
synthetase
prococoonase ! cocoonase
Digestion zymogen ! enzyme
Hormone production proinsulin ! insulin
angiotensinogen ! angiotensin
Oncogenic division, growth, migration, and
transformations adhesion
Tissue injury impairment of cell contact
inhibition
prekallikrein ! kallikrein
kininogen ! kinin
Translocation preprotein !protein
V. COMMERCIAL UTILIZATION OF
KNOWLEDGE FROM IN VIVO
PROTEIN BIOSYNTHESIS
Dominic W. S. Wong
U.S. Department of Agriculture, Albany, California, U.S.A.
A. Molecular Structure
Figure 3 Schematic illustration of Escherichia coli DNA polymerase I Klenow fragment, showing the N-terminal, central, and
C-terminal subdomains. Arrows indicate sheet; tubes represent helices (From PDB id: 1kfd).
REFERENCES
Figure 5 (A) Schematic diagram of the secondary structure 1. M Meselson, F Stahl. The replication of DNA in
of one domain of the subunit of Escherichia coli DNA Escherichia coli. Proc Natl Acad Sci USA 44:671
polymerase III holoenzyme. Arrows indicate sheets; tubes 682, 1958.
represent a helices. (B) Ribbon diagram of a domain with the 2. A Kornberg. Biologic synthesis of deoxyribonucleic
twofold symmetry axis indicated by an arrow. (C) Ribbon acid. Science 131:15031508, 1960.
diagram of one monomer, consisting of three structural 3. D Bramhill, A Kornberg. Cell 52:743-755, 1988.
domains of similar architecture. (Reprinted from Ref. 63, 4. ML DePamphilis. Eukaryotic DNA replication:
Copyright 1992, with permission from Elsevier Science.) anatomy of an origin. Annu Rev Biochem 62:2963,
1993.
D. Possessivity 5. AE Gorbalenya, EV Koonin. Helicases: amino acid
sequence comparison and structurefunction relation-
E. coli DNA polymerase III holoenzyme belongs to a ship. Curr Opin Struct Biol 3:419429, 1993.
group of replicative polymerases which also include 6. SS Patel, KM Picha. Structure and function of hex-
phage T4 DNA polymerase, and eukaryotic DNA americ helicases. Annu Rev Biochem 69:651697,
2000.
polymerase . These polymerases replicate DNA pro-
7. X Yu, MJ Jezewska, W Bujalowski, EH Egelman, 1996.
cessively along the entire length of the DNA template, The hexameric E. coli DnaB helicase can exist in
utilizing a sliding clamp. In E. coli DNA polymerase different quaternary states. J Mol Bio 259:714, 1996.
holoenzyme, the complex acts as a clamp loader in 8. LE Bird, K Hakansson, H Pan, DB Wigley.
an ATP activation to load a subunit (clamp) onto Characterization and crystallization of the helicase
Ton Kunst
Quest International, Naarden, The Netherlands
II. PROTEINS, PROTEIN HYDROLYSIS, Neither alkali nor acid hydrolysis is very specic;
AND PROTEIN HYDROLYSATES the rate of hydrolysis of the peptide bond is mainly
determined by the side chain of the amino acid residue
Proteins are linear polymers of amino acids that are of the peptide bond (9). In addition, a number of
covalently linked via a peptide bond. The hydrolysis of amino acids, e.g., cystine, serine, and threonine, are
proteins involves the hydrolytic degradation of this destroyed and undesired components such as lysino-
peptide bond using alkali, acid or enzyme [Eq. (1)]. alanine may be formed (10). Racemerization of
amino acid residues also occurs during alkali hydroly-
sis (4).
Enzymatic hydrolysis is much milder than acid or
alkali hydrolysis and does not lead to the destruction
1 of amino acids. It also is much more specic. By selec-
tion of the proper enzymes the cleavage of specic
Acid hydrolysis has traditionally been used to pro- peptide bonds is possible. The enzymatic hydrolysis
duce avor products (hydrolyzed vegetable protein or of proteins has extensively been studied by Adler-
HVP). Most of the literature on the production of Nissen (11).
protein hydrolysates with acids is published in the One of the key parameters in protein hydrolysis is
form of patents. A short summary and review is the degree of hydrolysis (DH). This is dened as the
given by Dave et al. (6). Today the use of acid hydro- percentage of peptide bonds cleaved. The DH is
lysis of proteins is limited owing to the risk of the usually approximated by the AN/TN ratio in which
formation of chloropropanols during the hydrolysis. AN is the amount of amino nitrogen and TN is the
Alkali hydrolysis of proteins has long been used to amount of total nitrogen as determined via the
produce foaming agents to substitute egg proteins or to Kjeldahl method. There are different methods to deter-
produce re extinguisher foams. Also, for the alkali mine the number of amino-nitrogen groups in the
hydrolysis most of the literature is published in the hydrolysate. Well-known and often applied methods
form of patents (7, 8). are the Formol titration (12), the determination of
After the actual hydrolysis process the enzyme is inac- The immobilization of enzymes, including proteolytic
tivated via e.g. a heat treatment. The hydrolysate can enzymes, has been extensively studied (see Chapters
be subjected to further processing steps such as separa- 2224 in this book). The advantage of the use of immo-
tion, ltration, or treatments with charcoal or ion bilized enzymes is the possibility to reuse the enzyme
exchange resins (16). Charcoal nonselectively removes (which saves costs) and the possibility to produce
both high- and low-molecular-weight organic com- hydrolysates in a continuous way. The application of
pounds and can be applied to remove off-color mate- immobilized proteolytic enzymes to produce protein
rial and to achieve slight reductions in off-odor and hydrolysates on an industrial scale up to now is lim-
bitterness. Products for patients suffering from phe- ited, however. When the immobilized enzyme is
nylketonuria (PKU) are treated with charcoal and/or applied in a batch reactor it must be separated from
ion exchange resins to remove phenylalanine from the hydrolysate via e.g. a crude ltration step. The
hydrolysates (29, 30). The treatments, however, also costs involved in this isolation can be higher than the
remove tyrosine and reduce the amounts of other savings obtained by reuse of the enzyme.
hydrophobic amino acids. In laboratory experiments immobilized enzymes
Filtration or separation of hydrolysates is generally have been used in columns. Packed columns easily
applied to remove insoluble components in order to get blocked by the insolubles in the raw material or
improve the clarity of the hydrolysate. Membrane by the insolubles formed during hydrolysis (see Sec.
ltration is often used to remove larger peptides and IV.B.3). Expanded bed columns, therefore, have to
undigested protein fragments that are potential anti- be used. With the recently developed beads very
genic materials. This results in a product with a good expanded bed columns can be made that can be
reduced allergenicity. Membrane ltration is also applied at a production scale for adsorption of pro-
applied to reduce the level of endotoxins in the hydro- teins from particulate-containing feedstocks (32).
lysate; this is important when the hydrolysate is These beads can be used also to immobilize enzymes
applied in parenteral formulae or in tissue culture on the surface of the beads. The costs involved are
media. The nal process step usually is drying of the considerable, however, and limit their application as
hydrolysate. carrier for immobilized enzymes.
In addition it is difcult to control the microbiolo-
gical quality of the hydrolysate because microorgan-
isms tend to build up on the carrier material. This
F. Membrane Bioreactors requires operation at elevated temperatures, which
increases the risk of protein denaturation and insolu-
Membrane bioreactors combine the hydrolysis step bilization. The application of immobilized enzymes in
and the membrane ltration step (31). The advantages expanded bed columns thus requires a delicate balance
of a membrane bioreactor are that the enzyme is between microbiological constraints on the one hand
retained (which saves costs), that the peptide size is and protein insolubilization at the other; further
controlled, and that peptides that inhibit the enzyme research is required to establish the best conditions.
are continuously removed after formation.
The disadvantage is that the insoluble fraction
in the hydrolysate is concentrated, which results in IV. APPLICATIONS
ux reduction and reduced productivity. With
prolonged operation times microbiological spoilage Intact, native proteins generally have well-dened
also may become a problem; not only the insoluble three-dimensional structures that affect their func-
protein material is retained, but also all micro- tional properties. Peptides are much smaller, the mole-
organisms. Especially, the more thermophillic cular structure is more random, and a tertiary structure
microorganisms may give rise to uncontrolled is rarely observed (33). For the secondary structure of
hydrolysis reactions. It is for these reasons that mem- proteins, the -helices and the -pleated sheets are the
brane bioreactors are currently not applied on an most important structures. Peptides in general do not
industrial scale for the production of protein hydro- have sufcient length to have enough H bonds to sta-
lysates. Further research is required to optimize the bilize the helix. Upon hydrolysis of proteins the hydro-
process. phobic areas, which are normally buried inside the
C. Organoleptic Properties
V. CONCLUDING REMARKS
Figure 10 Reversed-phase patterns of casein hydrolysates. Enzymatic hydrolysis is a valuable and exible tool to
(Top) Casein hydrolyzed with an endopeptidase (Neutrase, modify and improve the functional properties of pro-
NOVO). (Center) Casein hydrolyzed with an endopeptidase teins. To be able to produce a specic hydrolysate it is
(Neutrase, NOVO) and a carboxypeptidase (obtained from
necessary to have: (a) a detailed knowledge about the
autolysed yeast). (Bottom) Casein hydrolyzed with an endo-
relations between process conditions and the proper-
peptidase (Neutrase, NOVO) and an enzyme mixture con-
taining both carboxypeptidase and aminopeptidase ties of the hydrolysate obtained from a given protein
(BioProtease P23, Quest International). The casein hydroly- and enzyme in combination, and (b) a detailed know-
sates were analyzed on a C18 reversed-phase column using a ledge about the relations between the physicochemical
gradient of 0.1% TFA in water to 0.1% TFA in acetonitrile properties of the hydrolysate and their functional
as eluent. properties.
Rob M. M. Diks
Unilever Research and Development, Vlaardingen, The Netherlands
However, in the former case the synthesis reaction is It should be noted that for the Rhizomucor miehei
blocked by the fact that glycerol is strongly diluted in lipase a surprising phenomenon was reported in that
the aqueous phase and cannot effectively enter the the lipase showed a preference for the sn-1 position
active site of the lipase. when operating under very dry conditions (18). This
property could allow for stereospecic synthesis to be
B. Lipase Specicities carried out. As opposed to this regiospecicity,
lipases, such as that from Candida rugosa and the
Though the catalytic triad principle mentioned above one from Geotrichum candidum, are random and
is true for all lipases, the active site of lipases comes in hence can modify the fatty acids on all positions of
different shapes and sizes (16, 17). This restricts the the acylglycerols.
way in which acylglycerols can bind, or slows down The second type of specicity relates to the type of
and even prohibit certain acylglycerols from binding fatty acid being involved. For example, certain Cuphea
at all. The result is a preference of the lipase to react lipases show a distinct selectivity for short chain fatty
on certain acylglycerols or positions within, e.g., a tri- acids. Another important example is Geotrichum can-
glyceride molecule. This is often addressed as the selec- didum lipase B. Though being non regiospecic, it
tivity or specicity of a lipase. shows a strong selectivity for fatty acids with a cis--
Three major types of specicity can be distinguished 9 double bond (20, 21). Both this lipase and that from
(Table 1). The rst type involves positional or regio- Candida rugosa display a strong selectivity against
specicity, most lipases being either 1,3-specic or fully long-chain polyunsaturated fatty acids. This property
random (nonspecic). The 1,3-specicity is the most is probably related to the tunnel shape of the fatty acid
common and is displayed by, e.g., human gastric binding site of both lipases (22).
lipases, as well as microbial lipases such as from The easiest way of exploiting these fatty acid selec-
Rhizomucor miehei, Thermomyces sp., and various tivities in product purication or fatty acid enrich-
Rhizopus sp. As a result of a trough-shaped active ments is in single-stage reactions such as hydrolysis
site, triglycerides can bind only in specic positions or esterication, in combination with a physical
allowing lipases to modify acylglycerols on the sn-1 means of separating the desired end product and the
and sn-3 positions, both in hydrolysis and inter- or residual unwanted fatty acid fraction (see section on
transesterication. fatty acid enrichment).
Regiospecicity
sn-1, sn-3 specic Rhizomucor mieheia Triglycerides
Rhizopus oryzae interesterication/acidolysis
Rhizopus arrhizus synthesis by esterication
Rhizopus delemar Diglycerides
Rhizopus niveus triglyceride hydrolysis/glycerolysis (directed) fatty acid
Porcine pancreatic lipase esterication
Monoglycerides
triglyceride hydrolysis/glycerolysis fatty acid
esterication
Nonspecic Candida rugosa Fatty acid production by hydrolysis
Chromobacterium viscosum Mono- and diglycerides by directed glycerolysis
Pseudomonas uorescens Fatty acid production by hydrolysis
Pseudomonas cepacia Mono- and diglycerides by directed glycerolysis
Fatty acid specicity
Short-chain fatty acids Cuphea sp. Production of fatty acid concentrates by selective
hydrolysis
Long-chain poly- Geotrichum candidum Production of enriched glycerides by selective hydrolysis
unsaturated acids or enriched fatty acids by esterication
Candida rugosa Production of enriched glycerides by selective hydrolysis
or enriched fatty acids by esterication
Cis-9 unsaturated fatty Geotrichum candidum B Production of enriched glycerides by selective hydrolysis
acids or enriched fatty acids by esterication
Partial glyceride specic
Monoglycerides Potato acylhydrolase (patatin) Monoglycerides by fatty acid esterication
Mono- and diglycerides Penicillium camembertiib Mono- and diglycerides by fatty acid esterication
Penicillium cyclopiumb Mono- and diglycerides by fatty acid esterication
a
sn-1 specic under semi-dry operating conditions (18).
b
The selectivity for monoglycerides over diglycerides varies with the water activity (114, 115).
The third major lipase specicity is that for certain terms of the thermodynamic water activity maintained
types of acylglycerols. Penicillium camembertii lipase is in the reaction medium. It allows comparison of results
known to act upon mono- and diglycerides only, of enzymatic reactions in various organic solvents,
whereas other Penicillium lipases are reported to even though absolute water concentrations are signi-
hydrolyze only triglycerides. Patatin, a potato acyl cantly different (26).
hydrolase, is even more specic, almost exclusively It has been shown that the form of this relationship
producing monoglycerides in synthesis (23). is rather lipase specic. For example, Rhizomucor mie-
Unfortunately, details on its active site are not yet hei lipase has an optimum water activity 0:5 and still
available. retains 40% of its maximum activity at water activities
close to zero (27, 28). On the other hand, the lipases
C. Effect of the Thermodynamic Water Activity from Candida rugosa, Humicola, and Pseudomonas
cepacia not really showed a maximum, their activity
Apart from being a substrate, water also has a vital
function in maintaining the three-dimensional confor-
mation of the enzyme and hence its intrinsic activity
Humicola lanuginosa, currently known as Thermomyces
(24, 25). This effect is most conveniently expressed in lanuginosus.
Patrick F. Fox
University College, Cork, Ireland
Lipase Triglycerides H2 O ! fatty acids partial Off avor in milk; avor development in blue
glycerides glycerol cheese
Proteinase (plasmin) Hydrolysis of peptide bonds, particularly in Reduced storage stability of UHT products;
-casein cheese ripening
Alkaline Hydrolysis of phosphoric acid esters Index of pasteurization
phosphomonoesterase
Acid phosphomonoesterase Hydrolysis of phosphoric acid esters Cheese ripening; reduced heat stability of
milk;
Lysozyme Hydrolysis of mucopolysaccharides Bacteriocidal agent
-Glutamyl transpeptidase Transfer of -glutamyl residues Index of heat treatment
N-Acetylglucosaminidase Hydrolysis of glycoproteins Index of mastitis
Xanthine oxidase Aldehyde H2 O O2 ! Acid H2 O2 Pro-oxidant; cheese ripening
Sulfhydryl oxidase 2RSH O2 ! RSSR H2 O2 Amelioration of cooked avor
Superoxide dimutase 2O
2 2H ! H2 O2 O2 Antioxidant
Catalase 2H2 O2 ! O2 2H2 O Index of mastitis; pro-oxidant
Lactoperoxidase H2 O2 AH2 ! 2H2 O A Index of pasteurization; bactericidal agent;
index of mastitis; pro-oxidant
activities that have been detected in milk but which indices of animal health and of the mechanisms
have not been isolated and have no known signicance involved in the synthesis and secretion of milk. A
in milk are listed in Table 3; it is possible that some of very extensive literature has accumulated. The general
these enzymes are secreted by contaminating bacteria topic has been the subject of several general reviews
in milk. (110); in addition, the literature on the principal tech-
The indigenous enzymes in milk have attracted the nologically signicant enzymes has been reviewed
attention of researchers for > 100 years, mainly separately (see below).
because of their potential to cause defects in milk In this chapter the occurrence, distribution, isola-
and dairy products, especially lipase, and their useful- tion, and characterization of the principal indigenous
ness as indicators of the thermal treatment of milk. enzymes in bovine milk will be discussed, with empha-
More recently, they have assumed importance as sis on their commercial signicance in milk and dairy
Table 2 Other Enzymes That Have Been Isolated From Milk and Partially Characterized but Which Are of No Known
Signicance in Milk
Distribution
Enzyme Reaction catalyzed Source in milk
Patrick F. Fox
University College, Cork, Ireland
XII. -LACTAMASE
Ziya Gunata
University of Montpellier II, Montpellier, France
III. ENZYMES INVOLVED IN THE will be later discussed, the release of the aglycone moi-
RELEASE OF FLAVOR COMPOUNDS ety may occur through two-step (sequential hydrolysis)
FROM GLYCOSIDES or one-step (endoglycosidase) enzyme-catalyzed reac-
tions. The term endoglycosidase is used in this chap-
The elucidation of the enzymes responsible for the ter for glycosidases involved in avor release as a
release of avor compounds from glycosides is the function of their catalytic activities.
rst step in the development of technological applica-
tions. Glycosidases catalyze hydrolysis of the glycosidic A. Sequential Hydrolysis
bond of their carbohydrate substrates by two general
mechanisms, leading to either the retention or inversion Preliminary work carried out on the hydrolysis of grape
of the anomeric conguration at the cleavage point (44). glycosides, by using fungal enzyme preparations (pecti-
The mechanism of hydrolysis is presumed to proceed in nases, hemicellulases, and glucanases), has shown that
the same way as the acid-catalyzed cleavage of glycosicic the release of avor compounds depends on the pre-
bond, via a carbocation intermediate. Glycosidases dis- sence of glycosidases as side activities in the enzyme
cussed here refer to those enzymes which catalyze the preparations (47). With the aid of suitable analytical
hydrolysis of O-glycosyl compounds. techniques (TLC, HPLC, GC/MS), a study using pur-
As had been suggested as early as 1913 (45) and ied glycosidases and synthesized substrates was able to
1924 (46), volatile compounds (monoterpenes, vanillin) elucidate the sequential reaction involved in the hydro-
can be liberated from -D-glucosides through the lysis of disaccharidic avor precursors (Fig. 2) (48).
action of -glucosidase (EC 3.2.1.21). The involvement First, the inter-sugar linkage is cleaved by exoglyco-
of enzymes in the release of volatiles from disacchar- sidases; either an -arabinofuranosidase (EC 3.2.1.55),
ides, however, was shown only recently because the an -rhamnopyranosidase (EC 3.2.1.40), or a -apio-
relevant substrates were rst reported in 1982 (1). As furanosidase depending on the disaccharidic moiety.
This releases terminal sugars and -D-glucosides. demonstrated. With regard to microbial enzymes,
Subsequently, -glucosidase cleaves the aglyconic link- mainly those originating from GRAS (generally
age of -D-glucosides to liberate volatiles. This hydro- recommended as safe) will be discussed.
lysis mechanism has been patented (49). Such two-step
reactions, involving exoglycosidases (5, 6, 50, 51), are A. Plant-Originated Glycosidases
also possible in the hydrolysis of diglycosides of fruit
or plant volatiles, which possess -L-arabinopyrano- Most work has been devoted to grape glycosidases
syl, -D-xylopyranosyl or -D-glucopyranosyl as because the rst evidence of multiple forms of glycosi-
terminal sugars. dic avor precursors was found in this fruit. -
Glucosidase (55, 56, 69, 67, 68), -arabinofuranosidase
B. One-Step Hydrolysis of Disaccharide (56, 59), -arabinopyranosidase (19), -rhamnopyra-
Glycosides (Endoglycosidase) nosidase (56, 69), and -xylosidase (51) activities
were detected in grapes of various cultivars. The pre-
This one-step reaction occurs through the cleavage of sence of -apiosidase in grapes has not yet been con-
the aglyconic linkage which yields a disaccharide and rmed. On the basis of glycosidase activity
aglycone, the identity of which were conrmed by determinations with pNP-glycosylated substrates, -
HPLC and GC/MS analysis (Fig. 3) (5, 51). The glucosidase activity is considered the most important.
enzymes catalyzing this reaction were isolated from tea Multiple forms of -glucosidase exist in grape berries
leaves (5) and grapes (51). The enzyme from tea is called (51, 55, 59) and are also found in almond emulsin (69)
primeverosidase, after primeverosides, which are the but not in papaya fruit (54). The enzyme from papaya
most abundant glycoconjugated avors in this plant. fruit, unlike that from grapes, is readily soluble and
does not require the use of detergent for extraction.
Glycosidases are found predominantly in grape
IV. OCCURRENCE OF GLYCOSIDASES skins (67, 68), although contradictory results may
arise owing to the enzyme extraction method employed
Exoglycosidases involved in avor release are widely (59). -Glucosidase, -arabinofuranosidase, and -
distributed among plants (10, 5260) and microorgan- rhamnosidase activities have been shown to increase
isms (6166). The enzymes presented here mainly con- during the ripening of grape berries (56, 59, 67), unlike
cern those found in fruits or plants where the -glucosidase activity from papaya fruit which is unaf-
occurrence of glycoconjugated avors has been fected by the ripening stage of the fruit (54).
The substrates used may be composed either of a mix- The optimum pH activity of plant -glucosidase (55,
ture of glycosides isolated from fruits and plants by 59) and endoglycosidase (51) is similar (5.06.0) to that
selective retention on hydrophobic adsorbents (2, 12), of intracellular yeast -glucosidase (63, 87, 93, 105).
or of puried and synthesized natural glycosides (6, 35, This value is generally lower (4.05.0) for extracellular
101, 102). Alternatively, articial pNP-glycoconju- yeast (63, 92, 93) and Aspergillus enzymes (10, 61, 106).
gated substrates are often used. The optimum pH of an extracellular endo--glucosi-
Glycosidase activities can be determined by two dase from A. niger was found to be unusually low (3.4)
techniques: (100). Most -glucosidases exhibit only 515% of their
1. Direct analysis of glycosides by chromatographic maximum activity in the pH range of fruit juices (2.8
techniques (TLC, HPLC, GC) (2, 5, 6, 19, 35, 48, 51). 3.8).
Before GC analysis, a derivatization step, generally A. niger -arabinofuranosidase (62, 107) and -
involving the silylation and triuoroacetylation of gly- rhamnosidase (47, 108) possess similar pH optima to
cosides, is carried out in order to obtain their volatile that of extracellular -glucosidase from the same fun-
forms. gus, while the value for A. niger -apiosidase is higher
2. Analysis by colorimetric or chromatographic (5.06.0). (10, 99, 106). Both -arabinofuranosidase and
(TLC, HPLC, CPG) techniques of aglycone or sugar -rhamnosidase exhibit > 50% of their maximum
moieties (monosaccharides or disaccharides) released activities in the pH range of fruit juices.
from glycosides (2, 5, 48, 51, 103). In the case of agly- An important parameter that must be considered
cone liberated by enzymatic hydrolysis from natural for technological applications is the stability of glyco-
glycosides, the volatiles are recovered from the enzyme sidases in the acidic pH of fruit juices. Glycosidases
assay before GC analysis, by either liquidliquid differ notably in this respect (Fig. 4) (10).
extraction with a solvent or selective retention on -Glucosidases from grapes, S. cerevisiae, and C.
hydrophobic adsorbents (2, 6, 12). A colorimetric wickerhamii, are not very stable in these conditions.
assay of monoterpenes (aglycones) liberated from For example, C. wickerhamii -glucosidase was able
sugars has also been proposed (104). to hydrolyse -D-glucosides of monoterpenes in a
pNP-glycoconjugated substrates are widely used as wine only when the initial pH (3.0) was adjusted to
the standard for assaying glycosidase activities using 3.5 or higher (109). Only 10% of initial -glucosidase
mainly colorimetric techniques owing to their ready activity from S. cerevisiae was observed after 90 min
availability (except for certain disaccharidic substrates) incubation (208C) in a buffer at pH 3.0. (10). This may
and the simplicity of the analysis. Often, this analysis explain the large decrease in S. cerevisiae -glucosidase
does not reect the catalytic characteristics of enzymes, activity during grape juice fermentation. In fact the
owing to the presence of multiple forms of glycosidases optimum pH stability of the enzyme corresponds to
and, in some cases, the narrow aglycone specicity (see the pH of yeast cells (6.0). This decrease is also
Sec. VI.C). observed for -rhamnosidase and -arabinofuranosi-
dase activities from S. cerevisiae during juice
fermentation (10, 80). A -glucosiase from C. peltata
was fairly stable at pH 3.56.0 (95). The enzyme from a
mutant strain of C. molischiana was shown to be more
VI. FOOD-PROCESSING PROPERTIES OF stable at acidic pH than an enzyme from a wild-type
GLYCOSIDASES strain (110,111). This difference could be explained by
the difference in glycosylation of the relevant proteins.
This section will examine the properties of enzymes, In contrast, -glucosidase, -arabinofuranosidase,
focusing on the specic conditions encountered in the and -rhamnosidase from A. niger are remarkably
processing of fruit juice and derived beverages. In par- stable over a pH range of 2.57.0 (Fig. 4) (10, 62,
ticular, three important parameters inuencing avor 106). The enzymes exhibit > 80% of their maximum
release will be discussed: the effect of pH; aglycone and activity after 24 h of incubation in a buffer at pH
sugar specicity of glycosidases; and inhibition of 3.0. However, A. niger -apiosidase was less stable at
activity by sugar and sugar analogues. A summary of a pH < 4:0 (10, 106). The same trends were observed
the properties of -glucosidase from some plants and when the stability of A. niger glycosidases were mon-
microorganisms is given in Table 2. itored during grape juice fermentation at 208C (pH 3.0)
Yeast
S. cerevisiae (i) 6.06.8 5.07.0 45 45 6.7 13% at 100 mM n.d. 42% at 10 mM Primary 10, 80, 105
D. intermedia (i) 5.0 n.d. 55 50 3.0 n.d. n.d. n.d. n.d. 87
C. molischiana (e) 4.04.5 4.05.0 5060 60 7.0 n.d. n.d. 68% at 10 mM Primary tertiary 92, 116
C. wickerhamii (e) 4.5 4.05.0 50 50 230 18% at 100 mM n.d. 100% at 10 mM Primary tertiary 93, 94, 116
(i) 6.0 n.d. 50 50 9 n.d. n.d. 100% at 10 mM n.d. 93
C. peltata (e) 5.0 3.56.0 4550 50 1400 n.d. n.d. n.d. n.d. 95
D. hansenii (e) 4.05.0 n.d. 40 n.d. n.d. 10% at 100 mM n.d. n.d. Primary tertiary 89, 164
Fungus
A. niger (e) 4.06.0 2.57.0 6065 6065 3.03.8 90% at 100 mM n.d. 100% at 1 mM Primary tertiary 10, 61, 106, 113, 115
3.4 n.d. 65 n.d. 40 81% at 250 mM n.d. n.d. Primary tertiary 100, 161
4.06.0 4.07.0 55 60 543 n.d. n.d. n.d. n.d. 131
A. oryzae (e) 4.56.0 3.07.0 n.d. n.d. 958 10% at 100 mM n.d. 80% at 50 mM Primary tertiary 13, 129
B. cinerea (i) 6.57.0 n.d. 50 n.d. 5.5 n.d. 0.02 n.d. n.d. 64
(e) 3.0 4.010.0 60 n.d. n.d. n.d. n.d. n.d. n.d. 65
a
Temperature at which rapid inactivation occurs.
n.d., not determined; (i), intracellular; (e), extracellular.
(106). -Glucosidase and -arabinofuranosidase activ- not very surprising since 1 H and 13 C NMR spectra of
ities remained practically unchanged after 7 days, while disaccharide glycosides show that the orientation of
only 70% of initial -rhamnosidase and 50% of - the terminal sugar is unaltered whether the monoter-
apiosidase activities were observed. This remarkable penyl moiety is a primary or tertiary alcohol (35, 101).
stability of extracellular A. niger enzymes is interesting The hydrolysis of monoterpenyl -D-glucosides by
for fruit juice processing and could be explained by the plant (grape and almond; 2, 19, 55, 59, 115), fungal
glycosylation of proteins (110, 112). (yeast, A. niger; 10, 89, 115, 116), and bacterial (B.
polymyxa; 117) -glucosidases is, however, greatly
B. Effect of Temperature inuenced by the structure of the aglycone. Plant -
glucosidases showed a pronounced specicity and
The optimum temperature activities of plant (54, 55) acted on -D-glucosides of primary alcohols (e.g., ger-
and yeast -glucosidases (63, 89, 93, 95) is generally aniol, nerol, and citronellol) but not on those of ter-
lower (45508C) than that (50608C) of A. niger tiary alcohols (e.g., linalool, -terpineol, linalool
enzymes (61, 106, 113). At the normal temperature oxides) which, when hydrolyzed, can give rise to potent
(208C) of grape juice fermentation, for example, glyco- volatils. This shows the limitations in avor recovery
sidases exhibit only 1020% of their maximum activity. with plant endogenous -glucosidases in fruit juice
Rapid inactivation of glycosidases occurs at tem- processing.
peratures > 508C, (55, 63, 99, 113), although some la- The aglycone specicity of -glucosidases from S.
mentous fungal -glucosidases are only inactivated at cerevisiae and B. polymyxa resembled grape and
temperatures > 658C (61). The pH of the meidum plays almond -glucosidases more than microbial enzymes
an important role in the thermal denaturation of gly- from Aspergillus, Candida, Debaryomyces spp. The lat-
cosidases. In general, glycosidases from lamentous ter were able to catalyze the hydrolysis of both mono-
fungi are more heat stable than those from yeasts terpene primary and tertiary alcohol -D-glucosides.
and plants (61, 63). The rate of hydrolysis of -D-glucosides of mono-
terpene primary alcohols by various -glucosidases
C. Aglycone and Sugar Specicity of was found to be higher than that of -D-glucosides
Glycosidases of tertiary alcohols (115, 116). These results are in gen-
eral agreement with the catalytic activity of almond
The hydrolysis of monoterpenyl diglycosides by exo- emulsin -glucosidase on other primary and tertiary
glycosidases-arabinofuranosidase (48, 107), -ara- alkyl--D-glucosides (118). It is interesting to note
binopyranosidase (50, 114), -rhamnosidase (48), and that the opposite results were obtained when the
-apiofuranosidase (99, 106)is not signicantly inu- acid-catalyzed hydrolysis of these substrates was per-
enced by the structure of the aglycone moiety. This is formed (119).
aroma threshold in wines made from varieties, such as It is difcult to attribute the increase in concentra-
Muscat and Gewurztraminer, in which large amounts tion of each monoterpene directly to the enzymatic
of monoterpenes occur as avorless glycoconjugates release from glycoconjugate precursors. The acid-cata-
(2, 19, 109). These treated wines developed a more lyzed rearrangement of released monoterpenes occurs
intense oral note owing to the abundance of mono- in grape juice and wine (Fig. 6) (3638, 157), and some
terpenes. In nonoral varieties, such as Sauvignon, monoterpenes, as mentioned above, may be involved
Semillon, Shiraz, Greanche, Carignan, and Trajadura in yeast metabolism. The ideal way to determine the
(10, 109, 148152), the increase in monoterpenes is effect of enzymatic release from glycoconjugate precur-
below the aroma threshold of these compounds.
However, there is no information available about the
effect of this increase on the avor nuances that are
sometimes detected in treated wines.
Among the monoterpenes, an increase was observed
for geraniol, nerol, citronellol, linalool, furan, and
pyran forms of linalool oxides, -terpineol, and mono-
terpene polyols (e.g., 8-hydroxylinalools, p-menthene-
7,8-diol, exo-2-hydroxy-8-cineol). The highest increase
often occurred for geraniol and nerol, which impart
oral attributes to the wines (10, 146, 147, 151, 152).
The concentration of linalool, and linalool oxides,
and some polyols often increased to a lesser extent
because of the low activity of fungal -glucosidase
toward the relevant glycoconjugates. However,
mono- and diglycosides of these tertiary alcohols are
more prone to acidic hydrolysis in fruit juice than those
of primary alcohols (35).
Citronellol concentration increased greatly in some
enzymatically treated wines (Muscat). This could not
be the consequence of the enzymatic hydrolysis of
related glycosidic precursors since the glycosidic con-
jugates of this terpenol have not been detected in
grapes (153). The increase was explained by the action
of yeast (S. cerevisiae) reductase on geraniol and nerol,
which were more abundant in the enzymatically trea-
ted wines (154156). The production of citronellol was
inuenced by the yeast strain (155). The high amounts
of this compound resulted in the perception of citrus Figure 6 Acid-catalyzed rearrangement of monoterpenes.
attributes in the wines. (From Ref. 157.)
of grape glycosidic extracts. These molecules are odor- cally treated white wine from a nonoral grape
less and are considered not to generate odor-active variety (150) contained high levels of vanillin, which
compounds by further reactions (42, 158). In contrast, is a potent avorant usually derived from wood used in
3-hydroxy-7,8-dehydro--ionol, an acetylenic alcohol, wine aging.
detected in treated wines (10, 109) can generate a
highly esteemed, avorant, -damascenone, along
with a major product, 3-hydroxy--damascone, at
wine pH during storage through acid catalyzed reac-
tions (Fig. 9) (42, 70, 158). A high concentration of -
damascenone, exceeding several fold its odor thresh-
old, has been detected in an enzymatically treated wine
1 month after the end of fermentation (147). This can
be explained by the fact that glycosidase treatment
may have increased the concentration of an allenic
triol which is known to rearrange much faster than
aforementioned acetylenic alcohol to generate -
damascenone (158, 160).
-Damascenone, like other norisoprenoid avor-
ants, is generally produced in tiny amounts (in the
range of ppt) (42, 158, 160) that are difcult to quan-
tify by conventional GC/MS techniques. Nowadays,
stable isotope dilution techniques are being used
more and more because of their sensitivity and accu-
racy in quantitative analysis of trace avorants.
Among the shikimate-derived volatiles, benzyl alco-
hol, 2-phenyethanol, vanillin, eugenol, and zingerone Figure 9 Products issued from 3-hydroxy-7,8-dehydro--
were often detected at higher concentrations in enzy- ionol by acid-catalyzed reactions in fruit juice and wine.
matically treated wines (10, 109, 150). An enzymati- (From Ref. 158.)
was obtained. ENTE-1 is useful when swelling of the maximum intensity at 360 nm) for several minutes.
gels is to be avoided (10). Photocrosslinkable resin The gel lm formed is cut into small pieces and
prepolymers can be autoclaved at 1208C and stored used as the entrapped biocatalyst.
in the dark. Bead-type biocatalysts entrapped with photo-
Among these photocrosslinkable resin prepolymers, crosslinkable resin prepolymers can also be obtained
ENT and ENTP prepolymers are commercially avail- (11). Ten parts (by weight) of prepolymer are mixed
able from Kansai Paint Co., Tokyo, Japan. with 0.08 parts of the initiator, 2 parts of 3% sodium
alginate, 2 parts of distilled water, and 2 parts of the
biocatalyst. The mixture obtained is dropped into
3. Procedures 1.5% aqueous calcium chloride solution to form the
Typical procedures for entrapment with photocross- beads. The beads thus formed are irradiated by near-
linkable resin prepolymers are as follows: 1 g of ultraviolet light. After irradiating for 5 min, the
ENT-4000 or mixture of ENT-4000 and ENTP-4000 entrapped biocatalyst beads obtained are washed
(ENT-4000 content is above 70%) is mixed with with sterile water. For practical applications, the
10 mg of a photosensitizer such as benzoin ethyl bead-type is often preferable to the cube-type or lm-
ether. The mixture is melted by warming at 608C. type.
An adequate buffer, such as 50 mM potassium phos- Tube-type biocatalysts can be easily prepared by
phate buffer (0.5 mL) is added to the molten mixture using ENTE-1. The inner surface of glass tubes
and mixed at 608C. The molten mixture is cooled to (ID 2 mm) is silanized with vinyltriethoxysilane (10).
48C, and 2 mL of solution or suspension of biocata- ENTE-1 is mixed with photosensitizer and the
lysts is added to the mixture. The obtained mixture is solution or suspension of biocatalyst. This viscous
layered on a sheet of transparent polyester lm mixture is injected at the top of each silanized glass
framed (70 cm2) with a plastic tape (thickness tube to form a thin layer of biocatalyst-containing
0.5 mm). The layer is covered with a transparent prepolymer on the inner surface of the tube. The
thin lm to eliminate air and then illuminated with tube is illuminated with near ultraviolet light for
near-ultraviolet light (wavelength range 300400 nm; several minutes under an atmosphere of nitrogen to
Polyethylene glycol
MW of NCO content in content in Property of gel
Prepolymer polyether diol prepolymer (%) polyether diol (%) formed
glycol dimethacrylate (18, 19), are also useful gel mate- developed for the entrapment of different kinds of bio-
rials for entrapment by radiation. In the case of 10% catalysts.
polyvinyl alcohol, 57 Mrad of radiation is sufcient
to obtain rigid gels (17). The glass-forming prepoly- A. Enzymes
mers, such as polyethylene glycol diacrylate and poly-
ethylene glycol dimethacrylate, are used for Examples of enzymes entrapped with prepolymers are
entrapment through radiation polymerization at low listed in Table 3. Some enzymes listed in Table 3 have
temperature. Entrapment can be carried out at -248C been employed for development or appraisal of new
with 10% prepolymer and a radiation of 1 Mrad (19). prepolymer methods. Not many reports are available
A mixture of polyvinyl alcohol and polyethylene glycol concerning the applications of enzymes entrapped by
dimethacrylate is also applicable (18). This radiation prepolymer methods.
method has several advantages such as possible appli- Lipases entrapped with hydrophobic photocross-
cation of a wide variety of prepolymers as gel materials linkable resin prepolymers (ENTP-2000 and ENTP-
and entrapment at low temperature. However, utiliza- 4000) have been applied to the hydrolysis and ester
tion of this method is limited because radiation appa- exchange reaction of triacylglycerides. Production of
ratus is required. cacao butter-like fat from olive oil and stearic acid or
Water-soluble linear polyacrylamide (MW 15104 palmitic acid by enzymatic ester exchange reaction has
to 17 104 partially substituted with the acyl- been successfully achieved with ENTP-entrapped
hydrazide group was synthesized (20). This pre- Rhizopus delemar lipase in an organic solvent system
polymer can be gelled and used for entrapment of (21, 22). Entrapment markedly enhanced the opera-
biocatalysts in the presence of a crosslinking agent tional stability of the enzyme. In the case of the hydro-
such as glyoxal, glutaraldehyde, or periodate-oxidized lysis of olive oil to glycerol and fatty acids, Candida
polyvinyl alcohol. cylindracea lipase entrapped with ENTP showed high
activity (23). Entrapment also made it possible to use
the enzyme repeatedly or continuously. PU-entrapped
Candida cylindracea lipase was used for optical resolu-
III. BIOCATALYSTS ENTRAPPED BY tion of dl-menthol through enantioselective esterica-
PREPOLYMER METHODS tion in an organic solvent (24).
Enzymes entrapped with photocrosslinkable resin
As described above, various kinds of prepolymers hav- prepolymers were applied to the analyses of various
ing different physicochemical properties have been compounds. Tubes entrapping glutamate decarboxyl-
ase (10) and lysine decarboxylase (25) were prepared 2. The cell wall and/or cell membrane of intact cells
with the emulsion-type photocrosslinkable resin prepo- often prevents substrates, products, and other reaction
lymer ENTE-1. These tubes were used in continuous- components from permeating into and out of cells. In
ow systems to measure the concentration of L-gluta- this case, appropriate treatment of cells before or after
mate and L-lysine, respectively, in fermentation broths. immobilization is required to eliminate the barriers to
Thin enzyme lms, in which the enzymes were permeability.
entrapped with photocrosslinkable resin prepolymers, The immobilized cells can be kept in a growing state
were mounted on an oxygen electrode and used as by appropriate supply of suitable nutrients. This type
enzyme electrodes (13). Several oxygen-consuming oxi- of immobilized cells (immobilized growing cells) is very
dases, such as choline oxidase and glucose oxidase, advantageous. Immobilized growing cells serve as
were employed for this purpose. These enzyme electro- self-proliferating and self-regenerating biocatalysts.
des were found to be stable. However, immobilized growing cells also have some
disadvantages as follows:
1. Various nutrients and energy sources are required
B. Cells to maintain the cells in living or growing state(s), and
yield of wanted products may be lowered by the con-
Applications of immobilized cells are very attractive sumption of substrates and/or products as nutrients or
and important because cells possess multienzyme sys- energy sources.
tems which mediate complex reactions (3). In the case 2. Products are liable to be contaminated by leakage
of immobilization of cells, it is not necessary to extract of cells from carriers.
the enzymes from the cells. This avoids inactivation of In spite of these disadvantages, cell immobilization
the enzymes during the time-consuming and expensive is a key technology to construct an efcient bioprocess
purication procedures, and makes it possible to use because immobilized cells have many attractive char-
the enzymes under more stable conditions. However, acteristics, as described above. For immobilization of
immobilized cells have some disadvantages that should cells, entrapment methods have been mainly used.
be considered for practical application: Especially, other immobilization methods, such as car-
1. Byproducts may be synthesized because the cells rier binding and crosslinking methods, are difcult to
contain many enzymes, which may catalyze undesir- use for immobilization of living or growing cells. For
able reactions. Selection of a strain, mutation, or entrapment of cells, prepolymer methods have been
proper treatment of cells, or genetic improvement of also widely used. Various types of cells have been
a cell line helps to overcome this problem. entrapped by prepolymer methods.
Bioreactions with hydrophilic substrates are generally 1. K Mosbach, ed. Methods Enzymol, Vol 44. New
carried out in aqueous solutions. Hydrophilic proper- York: Academic Press, 1976.
ties of gels are favorable for such reaction systems. On 2. I Chibata, ed. Immobilized Enzymes. Research and
the other hand, proper organic solvents are suitable for Development. New York: John Wiley and Sons, 1978.
3. S Fukui, A Tanaka. Immobilized microbial cells.
bioconversion of hydrophobic compounds in terms of
Annu Rev Microbiol 36:145172, 1982.
the solubility of the compounds and of shifting the
4. S Fukui, A Tanaka, T Iida, E Hasegawa. Application
reaction equilibrium to the desired direction. In such of photo-crosslinkable resin to immobilization of an
cases, hydrophobic gels favor the transformation of enzyme. FEBS Lett 66:179182, 1976.
hydrophobic substances owing to high afnity of the 5. A Tanaka, S Yasuhara, S Fukui, T Iida, E Hasegawa.
compounds to the gels. One of the most representative Immobilization of invertase by the use of photo-cross-
examples is the successful ester exchange of triglycer- linkable resin oligomers and properties of the immo-
ides with Rhizopus delemar lipase in water-saturated n- bilized enzyme. J Ferment Technol 55:7175, 1977.
hexane, indicating the need of organic solvent and the 6. YY Chun, M Iida, H Iizuka. Studies on microbial
advantage of hydrophobic gels (21). In this case, transformations. XIX. Use of immobilized cells of
hydrolysis of triglycerides was preferential when the Streptomyces roseochromogenes for the 16/-hydroxyl-
ation of dehydroepiandrosterone. J Gen Appl
concentration of water was high, and the hydrophilic
Microbiol 27:505509, 1981.
gel-entrapped enzyme preparation exhibited only low
7. A Tanaka, S Yasuhara, M Osumi, S Fukui.
activity. It was also found that the activities of the Immobilization of yeast microbodies by inclusion
entrapped biocatalysts corresponded closely to the par- with photo-crosslinkable resins. Eur J Biochem
tition of the substrates between the gels and the exter- 80:193197, 1977.
nal solvents (35). 8. A Tanaka, N Hagi, S Yasuhara, S Fukui.
Prepolymer methods, especially those with photo- Immobilization of catalase with photo-crosslinkable
crosslinkable resin prepolymers and urethane prepoly- resin prepolymers. J Ferment Technol 56:511-
mers, again offer easy selection of gels having an 515,1978.
optional hydrophilicityhydrophobicity balance by 9. K Sonomoto, A Tanaka, T Omata, T Yamane, S
using different types of prepolymers or by mixing Fukui. Application of photo-crosslinkable resin pre-
polymers to entrap microbial cells. Effects of increased
these prepolymers.
cell-entrapping gel hydrophobicity on the hydrocorti-
sone 1 -dehydrogenation. Eur J Appl Microbiol
C. Ionic Properties Biotechnol 6:325-334, 1979.
10. N Itoh, N Hagi, T Iida, A Tanaka, S Fukui. Photo-
Anionic or cationic property of gels has a signicant crosslinkable prepolymer method for immobilization
of biocatalysts. Preparation of immobilized glutamate
inuence on the apparent activities of gel-entrapped
decarboxylase tubes and their application for auto-
biocatalysts through an electrostatic interaction with
mated analysis of L-glutamate. J Appl Biochem
the charged substrates. Furthermore, the optimum 1:291-300, 1979.
reaction pH may be shifted by using gels with proper 11. T Iida, M Sakamoto, H Izumida, Y Akagi.
ionic properties. Characteristics of Zymomonas mobilis immobilized
When choline oxidase entrapped with photocross- by photo-crosslinkable resin in ethanol fermentation.
linkable resin prepolymers was used as an enzyme elec- J Ferment Bioeng 75:28-31, 1993.
trode for assay of choline, the enzyme entrapped with 12. S Fukushima, T Nagai, K Fujita, A Tanaka, S Fukui.
an anionic prepolymer exhibited signicantly higher Hydrophilic urethane prepolymers: convenient mate-
activity. On the other hand, the enzyme entrapped rials for entrapment of enzymes. Biotechnol Bioeng
with a cationic prepolymer showed markedly lower 20:1465-1469, 1978.
13. S Fukui, K Sonomoto, N Itoh, A Tanaka. Several
activity (13). The results can be explained by different
novel methods for immobilization of enzymes, micro-
afnity of the cationic substrate choline for the gels
bial cells and organelles. Biochimie 62:381-386,
with different ionic properties. 1980.
Shift of the optimum pH was also observed in the 14. K Sonomoto, IN Jin, A Tanaka, S Fukui. Application
case of the production of adenine arabinoside through of urethane prepolymer method to immobilization of
transglycosylation by Enterobacter aerogenes cells biocatalysts. 1 -Dehydrogenation of hydrocortisone
entrapped with the photocrosslinkable resin prepoly- by Arthrobacter simplex cells entrapped with urethane
mers of different ionic characters (36). prepolymers. Agric Biol Chem 44:1119-1126, 1980.
signal of the secreted enzymes using genetic engineering S. cerevisiae itself is unable to utilize starch. A
techniques. Figure 2 shows the general structure of the number of strategies have been adopted for the
gene for cell surface immobilization of an enzyme (2). 30 - construction of starch-utilizing systems, which include
Half of -agglutinin contains a glycosylphosphatidyl- the addition of amylolytic enzymes in culture broth
inositol (GPI) anchor attachment signal at the C-term- and the introduction of heterologous genes encoding
inal end, like other cell surface proteins. amylolytic enzymes into yeast cells for secretive
production of the enzymes. As one of strategies, gluco-
amylase (EC 3.2.1.3) from Rhizopus oryzae (3), an
III. GENETIC IMMOBILIZATION OF exo-type amylolytic enzyme cleaving -1,4-linked
AMYLOLYTIC ENZYMES ON YEAST and -1,6-linked glucose effectively from the non-
CELL SURFACE reducing end and/or -amylase (EC 3.2.1.1) from
Bacillus stearothermophilus CU21 (BSTA) (4), an
Although starchy materials are available in abun- endo-type amylolytic enzyme cleaving -1,4-glycosidic
dance as carbon sources for cultivation, wild-type linkage, were displayed on the yeast cell surface
(Fig. 3). The constructed strains immobilizing gluco- mid for expression of the glucoamylase/-agglutinin
amylase or glucoamylase and -amylase should be fusion gene containing the secretion signal sequence
able to saccharify starch on its cell wall and assimilate of the glucoamylase under the control of the
the released glucose to proliferate and ferment. GAPDH promoter (2). The plasmid pGA11 was con-
structed as follows (Fig. 4): an XhoI site was generated
A. Immobilization of Glucoamylase on Yeast at the end of the glucoamylase coding region on the
Cell Surface Using a Multicopy Plasmid plasmid pYGA2270 (5) by site-directed mutagenesis
(U.S.E. Mutagenesis Kit, Pharmacia Biotech. Co.,
1. Construction of a Multicopy Plasmid Uppsala, Sweden) with the primers 50 -GCACCTG
A plasmid pGA11 to immobilize glucoamylase on the CCGCTGGCTCGAGAAATTTAAATGC-30 and 50 -
yeast cell surface was constructed as a multicopy plas- CTGTGACTGGTGACGCGTCAACCAAGTC-30 as
mutation and selection primers, respectively. A DNA
fragment containing glucoamylase coding region was
isolated from the mutated plasmid by EcoRI-XhoI
digestion. A DNA fragment of the -agglutinin gene
(AG1), containing 30 -half of the coding region encod-
ing 320 amino acids of the -agglutinin and 446 bp of
the 30 -anking region, was prepared by PCR (primers,
50 -GTACCTCGAGCGCCAAAAGCTCTTTTATC-
30 and 50 -GCGGTACCTTTGATTATGTTCTTCTTT
CTAT-30 ) with genomic DNA from S. cerevisiae MT8-
1 (MATa ade his3 leu2 trp1 ura3). (6) as a template,
followed by digestion by XhoI and KpnI. These two
fragments were substituted for the EcoRI-KpnI section
between the GAPDH promoter and the GAPDH ter-
minator of the yeast expression cassette vector
pYE22m. The resulting plasmid was named pGA11.
2% glucose). After cells were cultivated aerobically at The cell wall of S. cerevisiae mainly consists of glu-
308C in modied Burkholders medium (containing can and mannoproteins. Glucan is composed of -1,3-
0.002% adenine sulfate, 0.002% L-histidineHCl, and -1,6-linked glucose. Some mannoproteins seemed
0.003% L-leucine, 0.002% uracil, and 1% casamino to be covalently linked with glucan, because they can
acids) to which 2% glucose was added as a carbon be extracted by -1,3- or -1,6-glucanase. The localiza-
source, culture medium and cell pellets were isolated tion of the glucoamylase protein and its association
by centrifugation to measure the glucoamylase activ- with cell wall was determined. Cells were harvested
ities in both fractions. Cell growth in the culture broth by centrifugation at 3000g and washed in cooled
was measured by absorbance at 600 nm. For measure- (08C buffer (10 mM Tris-HCl [pH 7.8], 1 mM phenyl-
ment of glucoamylase activity (2), the substrate for methylsulfonyl uoride [PMSF]). The cells, the buffer,
glucoamylase reaction was prepared by adding soluble and glass beads (diameter 0.450.50 mm) were mixed
starch to boiling 20 mM sodium acetate buffer (pH in a ratio of 1:2:1 wet-w/v/w in a glass tube and agitated
4.6) to give a concentration of 0.5%. After keeping vigorously, using a benchtop vortex mixer, for 5 min at
0.9 mL of the solution at 308C for 5 min, 0.1 mL of 08C. The cell wall fraction was recovered by centrifu-
enzyme solution was added and the mixture was incu- gation of the homogenate at 1000g for 5 min and was
bated at the same temperature for 15 min. The reaction washed with the same buffer. Glucoamylase was
was stopped by boiling the mixture for 10 min and the extracted from this fraction in a two-step procedure.
concentration of glucose was determined using F-kit First, noncovalently bound proteins, and proteins
for glucose (Boehringer Mannheim, Mannheim, bound through disulde bridges, could be extracted
Germany). One unit of glucoamylase was dened as from cell wall with hot sodium dodecyl sulfate
the amount of enzyme required to release 1 mmol glu- (SDS). Following this, the SDS-treated cell wall was
cose/min from starch. As a result, the cells harboring further digested with laminarinase (-1,3-glucanase).
the plasmid pGA11 had only the cell-associated glu- To quantify the amount of the SDS-extracted and glu-
coamylase activity without secretion of the active canase-extracted glucoamylase protein, intensities of
enzyme (2). signals from each fraction on the lter were measured
4. Assimilation of Starch
When cells were cultivated aerobically with 1%
Figure 5 Plate assay for detection of amylolytic activity. 1, soluble starch as the sole carbon source, the cells
S. cerevisiae with cell surface-immobilized glucoamylase harboring the plasmid pGA11 utilized starch and
(S. cerevisiae MT8-1/pGA11); 2, S. cerevisiae secreting glu- proliferated to reach an absorbance of 10, which
coamylase; 3, S. cerevisiae MT8-1 as the control (S. cerevisiae was the same level as in the case of the culture on
MT8-1/pYE22m). 1% glucose.
paring the number of colonies (cells corresponding to bp of the 30 -anking region, was prepared by the same
A600 10 grown on a nonselective YPD plate with method as the case of glucoamylase. These two frag-
that grown on a selective SD (synthetic [S] medium ments and BamHI-XhoI fragment of the GAPDH pro-
[0.67% yeast nitrogen base without amino acid with moter were inserted into the plasmid pRS404 at its
appropriate supplements] containing 2% glucose) BamHI-KpnI site.
plate using the replica-plate method. As for IGA On the constructed plasmid, introduction of the
strain, the number of colonies on SD plate without XbaI site changed its corresponding amino acid
uracil was 97% of that on YPD plate. When cultiva- sequence Lys-Arg to Ser-Arg, where Lys-Arg is the
tion was repeated three times, the number of colonies recognition sequence of the Kex2 endopeptidase to
of uracil nonauxotrophs was still 90%. cleave -factor prosequence. Therefore, the site-direc-
ted mutagenesis was performed by PCR-based techni-
C. Immobilization of -Amylase on Yeast Cell que to change Ser-Arg to Lys-Arg. The resulting
Surface Using an Integrative Plasmid plasmid was named PIAA11. When the IAA strain
harboring pIAA11 was cultivated in modied
1. Construction of an Integrative Plasmid
Burkholders medium containing 2% glucose at 308C
To express the -amylase gene from B. stearothermo- for 24 h, 60% of the total -amylase activity was
philus CU21 (BSTA in S. crevisiae and display the detected in the culture supernatant. This solubilization
synthesized protein on its cell surface (8), the following seemed to be caused by the proteolytic processing of
DNA fragments were prepared (Fig. 8b). The prepro the fusion protein.
leader sequence of the -factor precursor (MF1) was The deduced amino acid sequence of the BSTA gene
rst fused to the -amylase gene. The prepro secretion contains one possible processing site for the Kex2
signal sequence of the MF1 was amplied from the endopeptidase, Val-Pro-Arg sequence, at amino acid
plasmid pLS01 by PCR, using the primers 50 - residues No. 481483. The Kex2 enzyme exhibits the
GCAGCTCGAGATGAGATTTCCTTCAATTTTT- substrate specicity toward each carboxyl site of
ACTGC-30 and 50 -TCTCTAGAATCCAAAGATAC- Lys-Arg, Arg-Arg, and Pro-Arg sequences (11). Then,
CCCTTC-30 . pLS01 has a 1.7-kbp EcoRI-EcoRI frag- a Kex2-resistant BSTA/-agglutinin fusion protein was
ment of the MF1 inserted at EcoRI site of pUC13 constructed by eliminating 33 residues from the
(10). The amplied fragment was cut with XhoI and C-terminus of the BSTA. The C-terminal truncated
XbaI, then ligated at the XbaI site with the -amylase BSTA/-agglutinin fusion gene was introduced into
gene fragment, which was amplied from pBR2.0A (4) S. cerevisiae (Fig. 9), using the integration plasmid
by PCR using the primers 50 -CCTCTAGAGCCGCA- pIAA12. For construction of the plasmid pIAA12,
CCGTTTAACGGCAC-30 and 50 -GCCTCGAGCCA- PCR was performed with primers 50 -GCAGCT-
GGCCATGCCACCAACCG-30 and digested with CGAGATGAGAT-TTCCTTCAATTTTTACTGC-30
XbaI and XhoI. A DNA fragment of the -agglutinin and 50 -TCCTCGA-GCCAGGAACCCAAACCGAA-
gene (AG1), containing 30 -half of the coding region ACCG-30 and pIAA11 as a template, followed by
encoding 320 amino acids of the -agglutinin and 446 XhoI digestion. The XhoI fragment of the plasmid
pIAA11 was substituted for this newly synthesized IAA strain on a starch-containing plate, unlike that
fragment to produce the plasmid pIAA12 to be immo- of the IGA/IAA strain, formed a small halo strictly
bilized without cleavage by proteases. Southern blot around the colony as observed in the case of the
hybridization analysis was carried out to conrm the IAA and IGA strains, showing the expression of
integration into chromosomal DNA. The obtained both glucoamylase and -amylase on the cell surface.
strain was named IAA strain. The IAA strain These results demonstrated that glucoamylase and -
exhibited -amylase activity only in the cell pellet frac- amylase were coexpressed and coimmobilized on the
tion. This indicated that the genetic immobilization of cell surface of the same cell.
-amylase without leaking any detectable activity to the
culture medium was performed by the elimination of 2. Assimilation of Starch
the possible Kex2 enzyme processing site.
The transformants were cultivated aerobically at 308C
in modied Burkholders medium containing 1% solu-
D. Coimmobilization of Glucoamylase and - ble starch as the sole carbon source. The IGA and
Amylase on Yeast Cell Surface IGA/IAA strains grew on starch and reached an
absorbance of 10 monitored at 600 nm, which was
Coexpression of glucoamylase and -amylase on the the same level as in the case of the culture on 1%
cell surface was expected to be effective for efcient glucose, although the growth on glucose of these
assimilation of starch (Fig. 10). An S. cerevisiae strains was more rapid. No growth on starch was
strain was constructed in which glucoamylase/- observed with the MT8-1 and IAA strains. The strain
agglutinin gene and C-terminal truncated -amylase/ codisplaying glucoamylase and -amylase, IGA/
-agglutinin gene were integrated into its chromo- IAA, grew faster than the glucoamylase-displaying
somes (8). The plasmid pIAA12 was cut by strain, IGA (Fig. 11). This result demonstrated that
HindIII and introduced into the IGA strain harbor- the yeast strain armed with -amylase in addition to
ing pIGA11. The cell of uracil and tryptophan non- glucoamylase had the enhanced ability to degrade
auxotrophy was picked up and named IGA/IAA starch by sequential hydrolytic reaction on the cell sur-
strain. Integration of the fusion genes was conrmed face.
by Southern blot hybridization analysis using URA3
and TRP1 as probes.
IV. GENETIC IMMOBILIZATION OF
CELLULOLYTIC ENZYMES ON
1. Detection of Glucoamylase and -Amylase
YEAST CELL SURFACE
The strain IGA/IAA exhibited both glucoamylase
and -amylase activities only in the cell pellet fraction. Cellulose, consisting of glucose units linked together
Immunouorescent labeling and immunoelectron by -1,4-glycosidic bonds, is the most abundant carbo-
microscopy conrmed the location of both enzymes hydrate in the biosphere. An estimated synthesis rate
on the yeast cell surface. The colony of the IGA/ of cellulose is 4 107 tons/year. For a long-range
Figure 13 Plate assay for detection of CMCase activity. 1, S. cerevisiae with cell surfaceimmobilized CMCase (S. cerevisiae
MT8-1/pCMC11); 2, S. cerevisiae secreting CMCase; 3, S. cerevisiae MT8-1 as the control (S. cerevisiae MT8-1/pYE22m).
Figure 14 Construction of the multicopy plasmid pBG211 for cell surface immobilization of -glucosidase.
3. Assimilation of Cello-oligosaccharides
The transformants were precultivated in YPD me-
dium and cultivated aerobically in S-medium (Sec.
III.B.2) containing cellobiose as the sole carbon
Harold E. Swaisgood
North Carolina State University, Raleigh, North Carolina, U.S.A.
Philip OConnell
National University of Ireland, Cork, Ireland
George G. Guilbault
University College, Cork, Ireland
I. ADVANTAGES AND DISADVANTAGES and enhances the thermal stability of the enzymes.
Limited dynamic ranges can be extended through the
The advantages of enzyme analyses are numerous, as use of diffusion limiting membranes (5). Gibson et al.
are the disadvantages. The advantages of using (6) have employed polyelectrolyte backbones along
enzymes must therefore be weighed against the disad- with sugar alcohols to stabilize enzymes for extended
vantages for each application. periods of time, even at room temperature. Glucose
The benets of using enzyme analyses include rapid oxidase was shown to have an operational stability
assay times (relative to most analytical assays), low for 5 months (7) when modied. With so much effort
cost, inherent selectivity, and simplistic systems. The being focused on improving enzymatic analyses, the
disadvantages include pH and temperature depen- applications of the technology are likely to expand
dence, a limitation to primarily aqueous analysis, lim- throughout the coming decade.
ited dynamic ranges, and relatively low stability. Thus,
enzymatic analyses are conned to those areas where
the advantages outweigh the disadvantages.
In recent years much work has been focused on II. METHODOLOGIES AND TOOLS
overcoming the shortcomings mentioned above and AVAILABLE
hence widening the applicability of enzymes in ana-
lyses. It has been shown that pH and temperature In science, one of the main methodologies is standar-
dependence of enzymes is signicantly reduced when dizationto be able to compare different techniques
immobilized; this is probably due to the enforced struc- or products to a standard. However, in a relatively new
ture induced by immobilization (1). Thus, the use of area such as enzymatic analyses, such standards are
immobilization can increase the ruggedness of most rare. McAteer et al. (8) have proposed a model for
enzymatic assays. Several groups (24) have focused shelf life prediction of stabilized commercial enzyme
on the use of enzyme sensors in organic media. They assays. Being able to place a well-dened shelf life on
found that when using organic solvents their sensors a product would make it much more viable in the
could still work well, but that the enzyme response was commercial marketplace. Boujtita et al. (9) have con-
dependent on both the concentration and type of sol- centrated on making biosensors applicable to indus-
vent. Such technology allows the detection of enzy- trial needs. They compared different modes of
matic substrates sparingly soluble in aqueous media production of renewable carbon paste electrodes and
Figure 1 Schematics of direct, indirect, and inhibition enzyme assays. Example system 1 (a) is a direct enzyme assay; system 2
(b) is an ELISA, an indirect assay; and example 3 (c) is an inhibition assay.
Figure 2 Schematic of a simple enzymatic amplication system for malate using an oxygen sensor.
B. Indirect Enzyme Analysis Is Composed of limit of detection (LOD) of 100 cells/mL. Ivnitski et al.
Many Approaches (63) have reviewed the different techniques used for
bacteriological detection using immunosensors.
1. Immunoassays
It can be difcult to detect the presence of pesticides
The most commonly used indirect technique is ELISA, using immunoassays. Owing to their small size, the
or enzyme-linked immunosorbant assay. An ELISA antibodies are raised against pesticides conjugated to
system can be designed to detect any molecule that immunogenic carriers and may thus be of low afnity/
can produce an immunogenic response, and many selectivity. Keay and McNeil (64) used an ampero-
molecules that dont produce one on their own can metric immunosensor to detect atrazine.
be modied to generate a response. This makes the Polychlorinated biphenyls (PCB) are found through-
technique very versatile. While other biorecognition out the food chain (sh extracts, clams, milk), and
molecules are available, such as DNA, they are not Bender and Sadik (65) produced an electrochemical
generally coupled to enzymes. The enzyme labels are immunosensor to detect for them. Dankwardt and
selected very carefully to fulll dened parameters. The Hock (66) reviewed the use of immunoassays to detect
enzyme must be cheap, highly puried, easy to conju- pesticides in foods and discussed the use of immuno-
gate to antibodies, and have colorless substrates with a sensors.
highly colored product. The main enzymes employed Marine toxins are an area of increasing interest, as
are horseradish peroxidase and alkaline phosphatase their incidence of outbreaks has been rising in recent
(55). They both meet the above requirements, but to years. Kreuzer et al. (67) developed an assay for oka-
different degrees. Given the versatility of ELISA, it is daic acid in shellsh. This diarrhetic toxin was detect-
not surprising that a large number of food assays have able down to a concentration of 6 1013 M in a rapid
been based on it. Primarily these have been focused time of 90 min, far better than any HPLC system for
upon the pathogenic bacteria and pesticides in foods, okadaic acid. Also, a rising issue in recent times has
which can be difcult to detect using traditional tech- been the use of genetically modied (GM) foods. Brett
niques. et al. (68) have concluded that antibody-based assays
Bacteria analyzed by ELISA include Listeria mono- are the natural detector for the presence of novel pro-
cytogenes (56), Salmonella sp. (57), Escherichia coli teins from GMs. Van Duijin et al. (69) have used
(58), Staphylococcus aureus (59), and Bacillus cereus immunoblotting to detect the presence of CP4 synthase
(60). The popularity of the method is due to the sensi- in GM soya. Thus, ELISA and immunosensors could
tivity of the technique, which can yield results much become powerful tools in the conrmation of GM pro-
faster than conventional techniques. Crowley et al. (61) ducts in our foods.
developed a rapid immunosensor based on ampero-
metric detection for Listeria monocytogenes in milk
2. Induced Bioluminescence
samples. The assay was performed in 3.5 h, which is
an improvement over the 24 h needed to complete even Another type of indirect enzyme analysis is the use of
the most rapid of traditional analysis. Abdel-Hamid et luciferase-encoded bacteriophage. The bacteriophage
al. (62) developed a sensor for E. coli O157:H7 with a infect specic bacteria present in the sample and
the case. Having remote laboratories means that the of protein is of primary importance in determining
results are not instantaneous. It also means that nutritional quality. Hsu et al. (88) developed a multi-
unstable samples may have changed by the time they enzyme technique for estimating protein digestibility.
are analyzed. Enzyme-based systems, especially those
incorporated within ow injection analyzers, overcome
this problem. As the systems are small and of low cost
REFERENCES
relative to other automated analyzers, they can easily
be tted on-site. With the sensors tted within the pro- 1. AM Klibanov. Enzyme stabilization by immobiliza-
duction process, a near instant reading can be tion. Anal Biochem 93:125, 1979.
achieved. Miertus et al. (47) developed a multianalyte 2. L Campanella, G Favero, MP Sammartino, M
sensor for quality control of red wine. By measuring a Tomassetti. Analysis of several real matrices using
range of analytes simultaneously a picture of the true new mono-, bi-enzymatic, or inhibition organic
state of the product can be obtained. Since wine is phase enzyme electrodes. Anal Chim Acta
liquid, there are few problems with sample pretreat- 393:109120, 1999.
ment; however, there are problems with the matrix, 3. J Wang, N Naser, D Lopez. Organic-phase biosensing
which contains many potentially electroactive organic of secondary alcohols with a Ta. Brockii alcohol dehy-
acids. Thus, any sensor hoping to measure in this drogenase electrode. Biosens Bioelectron 9:225230,
1994.
matrix must address this problem. Minimal pretreat-
4. S Kroger, S J Setford, APF Turner. Assessment of
ment was also required for a sh freshness sensor glucose oxidase behaviour in alcoholic solutions
developed by Carsol et al. (86). The sh exudates using disposable electrodes. Anal Chim Acta
were merely diluted in buffer and analyzed for the 368:219231, 1998.
presence of inosine, inosine monophosphate, and 5. L Gorton, E Csoregi, E Dom nguez, J Emneus, G
hypoxanthine (K-value). Okuma et al. (87) developed Jonsson-Pettersson, G Marko-Varga, B Persson.
a system to determine the initial stages of spoilage in Selective detection in ow analysis based on the com-
chicken; as in the sh freshness assay, the system relies bination of immobilised enzymes and chemically mod-
on the response of three analytes to show the spoilage. ied electrodes. Anal Chim 250:203248, 1991.
The markers selected, putrescine, cadaverine, and sper- 6. TD Gibson, BLJ Pierce, JN Hulbert, S Gillespie.
midine, allow an earlier assessment of the state of Improvements in the stability characteristics of bio-
sensors using proteinpolyelectrolyte complexes.
putrication of meat than the K-value. Kelly et al.
Sensors Actuators B33:1318, 1996.
(53) developed a sensor for the determination of L- 7. VG Gavalas, NA Chaniotakis, TD Gibson. Improved
lysine in heat-treated foods, but the breakdown of operational stability of biosensors based on enzyme
food proteins requires a hydrolysis step, which can polyelectrolyte complex adsorbed into a porous
be performed in 15 min. Thus, the pretreatment of carbon electrode. Biosens Bioelectron 13:12051211,
the sample is the time-limiting step. The digestibility 1998.
Dominic W. S. Wong
U.S. Department of Agriculture, Albany, California, U.S.A.
-Amylase
Conversion of starch to dextrins (liquefaction) in the production of corn syrup
Supplement to our types low in -amylase to ensure a continuous supply of fermentable sugar for yeast growth and gas
production in dough making
Solubilization of adjuncts (nonmalt carbohydrate materials from barley and other cereal grains) used in brewing
Glucoamylase
Conversion of dextrins to glucose (saccharication) in the production of corn syrup
Conversion of residual dextrins to fermentable sugar in brewing for the production of light beer
-Amylase
Production of high-maltose syrup
Xylose (glucose) isomerase
Isomerization of glucose to fructose in the production of high-fructose corn syrup
-Glucanase
Breakdown of -glucans in malt and other raw materials to aid ltration of wort after mashing in brewing
Lipase
Enhancing avor development and shortening the time for cheese ripening
Production of specialty fats with improved qualities
Production of enzyme-modied cheese/butter from cheese curd or butterfat
Papain
Used as meat tenderizer
Used in brewing to prevent chill-haze formation by digesting proteins that can otherwise react with tannic substances to
form insoluble colloid particles.
Chymosin
Curding of milk by specic proteolytic action on -caseins in cheese making
Microbial proteases
Processing of raw plant and animal proteins
Production of sh meals, meat extracts, texturized proteins, and meat extenders
Pectinase
Treatment of fruit pulp to facilitate juice extraction and for clarication and ltration of juice
Lactase
Additive for dairy products for individuals lacking lactase
Breakdown of lactose in whey products for manufacturing polyactide
Acetolactate decarboxylase
Reduction of maturation time in wine making by converting acetolactate to acetoin
(In the absence of the enzyme, acetolactate is oxidized to diacetyl that requires secondary fermentation to reduce it to
acetoin)
Lysozyme
Antimicrobial preservative
Glucose oxidase
Conversion of glucose to gluconic acid to prevent Maillard reaction in products caused by high heat used in dehydration
Potential use to remove O2 in food packaging for protection against oxidative deterioration
Cellulase
Conversion of cellulose wastes to fermentable feedstock for ethanol or single-cell protein production
Protein stability can be expressed in terms of the the folded state. Macrounfolding represents the
free energy, associated with the micro- and macro- stability of the macroscopic state of the protein
unfolding of the protein. Microunfolding represents molecules and the energy required for the protein
qualitatively characteristics of the rigidity of the changing from the native to the unfolded macroscopic
protein structure, and is associated with local, state (2). Adjustment of conformational exibility is a
reversible, noncooperative unfolding reactions within key step in the thermal adaptation of proteins.
Catalase
Dominic W. S. Wong
U.S. Department of Agriculture, Albany, California, U.S.A.
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
The k10 for pure catalase from human erythrocytes is iron despite his partial purication of catalase.
3:4 107 M1 sec1 . This value permits estimation of Willstatter (13) also doubted the function of iron in
the concentration of catalase in blood and tissues. peroxidase.
There are some problems in estimating the concen- In 1930, Zeile and Hellstrom (14) prepared purer
tration of catalase because of molecular heterogeneity catalase from liver and demonstrated the hematin nat-
due to genetic origin (not all isozymes have the same ure of the prosthetic group. The analogies between
specicity), conformational alterations due to -SH catalase and other hematin compounds were soon
group oxidation or dissociation of the active tetramer recognized (1517). Bray and Gorin (18) postulated
(catalatic activity) into the dimer which has peroxidatic the reaction with iron of the hematin prosthetic
activity but not catalatic activity (68). group was probably that shown in Eq. (5). This was
the rst formulation of a catalytic role for quadrivalent
(ferryl) iron.
B. Historical Aspects
Fe2 H2 O2 ! FeO2 H2 O 5
Catalase appears to be the rst potential enzyme
Earlier, Manchot and Lehmann (19) reported that a
recorded (about 1812), based on the appearance of
pentavalent iron Fe2 O5 could perform a catalatic
rapid bubble evolution when biological extracts were
reaction [Eq. (6)].
mixed with H2 O2 . The catalatic activity was regarded
originally as an attribute of protoplasm, a term Fe2 O5 2H2 O2 2H ! 2FeOH2 2O2 H2 O
shared with other general catalysts such as platinum. 6
In 1901, Loew introduced the specic name catalase
(9). But the true nature of catalase was overlooked by But in enzymes, the theory of the enzyme-substrate
the more prevalent discussion of the actions of metals complex formation was an alternate explanation to
in biological oxidation. Wieland (10) considered the the free-radical hypothesis required by Eq. (6). Von
initial oxidation reaction to be the transfer of hydrogen Euler and Josephsons data (20) supported the idea
to oxygen to form peroxide, and the role of the metal that catalase formed a Michaelis complex with H2 O2 ,
was to remove the peroxide so formed. Warburg (11) with Km of 0.025 M. Bonnichsen et al. (21) proved that
considered that hydrogen peroxide was not formed the reaction velocity could be explained by the collision
during biological oxidation, but that the iron of the theory of reaction rates and strongly argued against
cells activated O2 to react directly with substrates. the assumptions of chain mechanisms postulated ear-
Therefore, he did not consider special catalatic or per- lier. At the same time Chance (22) discovered the pri-
oxidative roles for the heavy metals in contrast to mary compound (FeOOH2 ) formed between catalase
Wieland (10). Hennichs in 1926 (12) also failed to and H2 O2 (compound I; [Eq. (7)], which then reacts
recognize that the prosthetic group of catalase contains with a second H2 O2 [Eq. (7)].
Figure 2 Rates of inactivation of catalase, peroxidase, lipoxygenase, and lipase in English green pea homogenates incubated at
60 C. (From Ref. 27.)
mostly van der Waals contacts between the heme vinyl from the heme iron, respectively. The ligand 6 site on
and methyl groups and nonpolar side chains, and a few the distal side of the heme is unoccupied.
charge interactions between the heme propionyl The tetrameric enzyme from human and bovine
groups and Arg and carboxylic acid residues. At the contains four molecules of tightly bound NADPH
proximal side of the heme, Tyr357 provides a protein located toward the carboxyl ends of 5 and 10 helices
ligand (as ligand 5) with the Fe-phenolic oxygen dis- (42, 43). However, the binding of NADPH is not
tance of 1.9 A. Ionization of the Tyr-OH is assisted by essential for activity, but may function to prevent the
interactions with the guanidinium group of Arg353. formation of compound II, an inactive intermediate in
On the distal side of the heme are the essential His74 the catalatic cycle (see below). The binding site shows
and Asn147, with their N atoms at 4.3 A and 6.2 A relative afnities in the order of: NADPH > NADH >
positive charge of the heme group, allowing the nal C. Formation of Compound II
transfer of the O to the heme iron.
Catalase can slowly form compound II by a two-step
one-electron reduction of compound I under a steady
B. Reduction of Compound I low concentration of H2 O2 or with other special sub-
strates, such as ferrocyanide and ascorbate. Catalase
In the second half of the cycle, catalase compound I compound II is an inactive form in the catalatic mode
reacts with a second molecules of H2 O2 to yield the of the enzyme. This is in contrast to HRP compound
native enzyme with generation of O2 and H2 O. II, which is an active species produced by two sequen-
However, compound I reacts poorly with other hydro- tial one-electron reductions of compound I with most
gen donors, such as alcohol, hydroxylamine, or formic of its substrates. The difference in the reactivities of
acid. Reduction of compound I occurs via a hydride compound II species of catalase and HRP is probably
transfer by abstracting the -hydrogen by the heme due to protein control of the accessibility of substrates
ferryl oxygen (55). Deprotonation of the substrates to the heme group (57). In addition, the phenoxylate
-proton is unlikely to occur because the His74 is oxygen of the proximal Tyr residue provides more elec-
less nucleophilic as the heme in compound I is posi- tron density to the heme iron than the His residues of
tively charged. The -hydrogen abstraction is stereo- HRP, and favors a two-electron transfer instead of a
specic for the pro-R hydrogen in the oxidation of sequential addition of electrons (58).
ethanol substrate (56). A hydride transfer facilitates The slow inactivation of catalase by its own sub-
the development of a double-bond character at O strate can be prevented by experimentally limiting the
O , and the nal transfer of the proton to the exposure to H2 O2 , or reducing the concentration of
heme ferryl oxygen (Fig. 2). compound I. However, the primary inhibition of the
Horseradish Peroxidase
Plant (a) Peroxidases of prokaryotic origin: Determination of the complete amino acid sequence of
peroxidase yeast cytochrome C peroxidase, HRP-C was largely contributed by Welinder (21).
chloroplast and cytosolic ascorbate HRP-C consists of 308 amino acid residues, a hemin
peroxidases, gene duplicated bacterial group, and eight neutral carbohydrate side chains
peroxidases. attached to asparagine residues. The amino acid
(b) Secreted fungal peroxidases: sequence and carbohydrate attachment points are
lignin peroxidase (Phanechaete shown in Figure 1. The molecular weight was reported
chryosporium), manganase peroxidase
to be between 40,000 and 45,000, and the carbohydrate
(P. chrysoporium), ink cap mushroom
moiety constitutes 18%. HRP-C is the most exten-
peroxidase (Coprinus sinereus),
chloroperoxidase (Caldariomyces sively investigated isoenzyme in the plant peroxidase
fumago), cytochrome C peroxidase superfamily. Within the superfamily, HRP-C belongs
(Pseudomonas aeruginosa). to Class III of the plant extracellular peroxidases (22).
(c) Classical plant peroxidases: Peroxidases from owering plants have considerable
HRP isoenzyme C is the most identity of amino acid sequences (4064%). In particu-
intensively studied. lar, those sequences in the regions of the two His resi-
Animal Myeloperoxidase, verdoperoxidase, dues (the distal His42 and proximal His170 in HRP)
peroxidases eosinophil peroxidase, prostaglandin H and Cys residues forming disulde bonds are con-
synthase, glutathione peroxidase, served (12, 22).
lactoperoxidase, thyroid peroxidase.
Gajhede et al. (23) published a 2.15 A resolution
Catalases They are found in plants and animals.
structure of HRP-C, which gave complete information
These are tetrameric proteins.
on overall structure and its comparison to other per-
oxidases. HRP-C has the same overall fold typical of
the peroxidase superfamily, with the secondary struc-
ture elements (especially the 10 prominent helices AJ),
IV. PROPERTIES AS PROTEINS
heme group, and calcium-binding sites in similar posi-
A. HRP Isoenzymes tions. The secondary and the tertiary structures of
HRP-C are summarized in Figure 2.
Over 40 isoenzymes of peroxidase have been found in More recently, the synthetic and native HRP-C
the horseradish plant. By cloning techniques, some of have been cloned in various microorganismsE. coli,
the isoenzymes come from different genes and others Baculovirus, and Pichia pastoris, etc. (2426).
may be produced by posttranslational modication,
especially by degree of glycosylation. There are differ-
V. PROPERTIES AS AN ENZYME
ences in amino acid composition and sequences among
the isoenzymes (19). Based on the isoelectric points, the A. Specic Mechanism of Action
isoenzymes can be classied as shown in Table 2 (20).
The normal peroxidase cycle is shown in Figure 3. The
reaction of HRP with H2 O2 produces a two-electron
oxidized species known as Compound I in which the
ferric iron is oxidized to a ferryl (FeIV
O) species and
Table 2 Horseradish Peroxidase Isoenzymes
the porphyrin to a radical cation. Stepwise reduction of
FORM pI pHopt Carbohydrate (%) Compound I by two substrate-derived electrons pro-
duces Compound II, in which the porphyrin radical
HRP-A 4 Acid Higher cation has been quenched, and subsequently the resting
HRP-B, -C 5.75, 9.63 Neutrala, Low ferric state form. The second-order rate constants
slightly for the reaction HRP and H2 O2 are: k1 , (1:58 0:05
alkaline
107 M1 s1 ; k2 , (9:97 0:28 106 M1 sec1 ; and k3 ,
HRP-D, -E 10.6, > 12 Strongly Low
alkaline (3:64 0:01 102 M1 sec1 (1, 27, 28).
Generally, k3 is always 1020 times slower than k2 ,
a
Neutral HRP-C is the main isoenzyme in the plant. and under most steady-state conditions is rate limiting.
The values of k2 and k3 depend on the nature of the play a role in mediating the reactions with a general
substrate. acid/base mechanism. The aromatic donor molecule
Both competitive binding (1) and spectroscopic interacts with a substrate accessible channel contribu-
studies (29) have revealed that the peroxide- and ted by Phe residues (Phe68, Phe142, and Phe179) and
substrate-binding sites are not the same. Specic-site the heme C18 methyl group.
mutation analyses (30, 31) and the high-resolution
crystallographic structure of HRP-C (23) have B. Substrate Specicity
provided detailed information on structural basis of
HRP catalysis. Peroxide binding, activation, and Peroxidase is specic to the hydrogen acceptors. Only
Compound I formation occur in the distal heme hydrogen peroxide, methylhydroperoxide, and ethyl-
pocket, where the key residues (His42 and Arg38) hydroperoxide can be used as hydrogen acceptors.
Specic activity
Substrate Enzyme (nM) Km M kcat (s1 ) kcat =Km s1 m1 (units/mg)
ABTSa 0.2 8:0 102 4:1 103 5:1 0:2 1:15 103
Guaiacolb 2 5:8 103 4:2 102 7:2 1:1 102 c
Table 5 Peroxidase Purication Using the Traditional Biochemical Methods in Combination with
Reverse-Micelle Extraction (Horseradish Peel only; Method C)a
Glutathione Peroxidase
Chemical
Substrate structure Examples
Glutathione g-Glu-Cys-Gly
Hydroperoxides R-O-O-H H2O2
Organic hydroperoxidase: ethyl, cumene, tert-butyl, 5-phenyl-4-pentenyl,
linoleic, linoleic acid hydroperoxides, and their methyl esters: 13-hydroperoxy-
9, 11-octacadienoic acid, cholesterol 7b-hydroperoxide; allopregnanolone 17a-
hydroperoxide; thymine hydroperoxide.
Lipid hydroperoxide: phosphatidylethanolamine, phosphatidyl-choline,
phosphatidylserine, cardiolipin and phosphatidic acid hydroperoxides: mono-
and dihydroperoxydilinoleoylphosphatidyl cholines
respiratory tissues (blood vessels, lung, heart), diges- III. UTILITY OF GPX
tive tissues (liver, kidneys), and brain tissues where
oxygen is required. In rats, a high level of the enzyme The biological damage such as peroxidation of lipids
is found in the liver with lower concentrations detected and nucleic acid damage can be caused by reactive
in red cells, heart, lung, and kidney. In hepatocytes oxygen species. Antioxidant defenses consist of a num-
70% of the GPX activity is from the cytosol and ber of antioxidant enzymes protecting cells and mem-
30% is localized within the mitochondrial matrix (20). branes against the reactive oxygen. The rst line of
Classical GPX is present abundantly in erythro- defense seems to be governed by superoxide dismutase
cytes, kidney, and liver (16). The lens of different ani- which acts by converting the superoxide ion O2: into
mal species contain remarkably high GPX activity hydrogen peroxide. Secondly, catalase, and GPX share
(21, 22). PHGPX has been identied in all tissues in the task of metabolizing H2 O2 to water. In the last line
which it was searched for, namely, rat kidney, heart, of defense, the GPX is the sole enzymatic system cap-
lung, muscle, and brain; bull spermatozoa; and sh able of preventing lipid peroxidation. The enzyme can
liver (23). A high level of GPX activity has been reduce a wide variety of hydroperoxides, and acts as an
observed in rat testis (24). pGPX is detected in plasma, important part of the antioxidant defense of biology
milk, and lung (6, 7). giGPX has been found in the body.
cytosol of human and rat gastrointestinal tracts (9). Many studies have shown that low selenium in the
It was detected in human liver and colon, but not diet can result in low levels of GPX activity leading to
other human tissues including kidney, heart, lung, oxidative damage and several medical disorders such
placenta, and uterus. In rodent tissues, giGPX is only as Keshan disease, Kashin-Beck disease, cancer, heart
detected in the gastrointestinal tract, not in other disease, premature aging, and diabetes. The impor-
tissues. In rat, the highest activity of the enzyme is tance of this enzyme as an oxidant defense enzyme
detected in stomach followed by decreasing activities has led to its proposed strong possibilities to be used
in esophagus, colon, and intestine. in treatment for these diseases (2932).
The study by Godin and Ganett (25) showed that
the distribution of GPX in mammalian tissues is
IV. PROPERTIES AS PROTEIN
strictly functionally related to other antioxidant
enzymes such as superoxide dismutase (SOD; EC A. Molecular Weight and Physical Properties
1.15.1.1) and catalase (EC 1.11.1.6). It should be
kept in mind that the level of GPX is subject to numer- cGPX, pGPX, and giGPX are tetrameric proteins of
ous inuences, including dietary selenium, age and sex, four identical subunits, with each subunit containing
and estrous cycle, and other environmental factors one atom of selenium in the form of selenocysteine
(2628).It has been documented that GPX activity is residue. The average molecular weight is 1925 kDa
subject to large species differences in all organs (20). (33, 34, 42). PHGPX is a monomeric protein of 23
Subunits
Subcellular molecular Activiation energy
Enzyme location weight (kDa) Subunits Isoelectric (kcal/mol1 ) Ref.
GSH. A proposed model of glutathione binding was exons; exons 3 and 5 show homology with exons of
given by Epp et al. (45). cGPX, indicating that this gene, although related to
Compared with cGPX, the three residues for cGPX, is distinct from it (40, 46). Multiple isoforms
binding GSH are mutated or deleted in pGPX. The of PHGPX have been detected in human, rat, pig, and
surface of pGPX is more negatively charged, especially bovine tissues (5, 23, 24, 4951). Another intercellular
close to the active site. These differences may affect GPX is giGPX. The gene expression was found in
substrate binding and lead to different enzyme human liver and colon and also detected in mouse
activities (34, 41). and rat gastrointestinal tract (9). giGPX has been
Although there are no data on the crystal structure regarded as the most closely related offspring of the
of PHGPX, a model has been developed by molecular cGPX family (43).
dynamics calculation (43). According to this model, An articial GPX mimic, selenosubtilisin, was cre-
the shape of the core of these enzyme families compris- ated by converting the active serine residue of the cat-
ing the pleated sheets and three of the helices is largely alytic site of subtilisin into selenocysteine. It displays
identical to that of cGPX. However, the deletion of GPX activity but not as efcient as cGPX (52, 53).
several amino acids in the PHGPX sequence causes a Another articial selenium-containing enzyme has
substantial short-cut of the subunit interaction loop been generated by monoclonal technique and chemical
and an elimination of a small helix. In the PHGPX modication (54). The abzyme displays a higher GPX
model, the active site appears more freely accessible. activity compared with rabbit liver cGPX.
This allows the enzyme to attack more complex hydro- All the cDNA sequences of these GPX contain a
peroxy lipids. UGA codon coding for selenocysteine (9, 41, 38).
The UGA codon, usually functioning as a stop
codon, was demonstrated to encode the amino acid
C. Isoforms residue selenocysteine, which could be incorporated
directly during protein synthesis (46, 38).
The existence of multiple forms of GPX is due to the
expression of different genes (41, 46). Many mamma-
lian cGPX genes display highly conserved sequences
V. ENZYMETIC PROPERTIES
around the site of selenium (47). The gene for human
pGPX has been found at chromosome 5q32, and con- A. Kinetics of GPX
tains ve exons rather than two for cGPX. Rat and
bovine pGPX genes have been identied similar to The steady-state kinetics of bovine erythrocyte GPX
human pGPX (48). The gene for PHGPX has seven was studied by Flohe et al. (56). The enzyme system
cGPX PHGPX
Substrate f1 108 msec f2 106 msec f1 106 msec f2 104 msec
H2O2 0.56 1.27 5.5 1.3
tert-Butyl hydroperoxide 13.5 2.24 1.4 31
Cumene hydroperoxide 7.8 2.24 9.1 2.5
Linoleic acid hydroperoxide 5.5 2.3
Phosphatidyl choline hydroperoxide 1.2 1.6
Source: Refs. 5, 20, 56.
Enzyme Hydroperoxide Km mM 1mM GSH kcat 103 min1 kcat =Km 107 M1 min1
Glucose Oxidase
A. J. Vroemen
DSM Food Specialties, Delft, The Netherlands
Lactate Dehydrogenase
Hayao Taguchi
Tokyo University of Science, Noda, Chiba, Japan
crab, arachnids, and gastropods, but insects or crusta- are available commercially (such as F-kits for L- and
ceans only possess L-LDHs. On the other hand, some D/L-lactate from Boehringer Mannheim).
lactic bacteria contain D- or L-LDHs, or both, and L-LDHs from many sources can be purchased from
therefore ferment D-, or L-, or both lactic acids (3, 6). commercial source, such as the muscle, heart, and liver
Activity staining with a tetrazolium salt (see below) isozymes from rabbit, cow, and pig, and the enzymes
is often used for the histological analysis of the iso- of lobster and some bacteria. The D-LDHs of lactic
zymes, or phylogenic or taxonomic analysis of LDHs bacteria, for example, Lactobacillus leichmannii and
in lactic bacteria. LDHs are usually separated by elec- Leuconostoc mesenteroides enzymes, are also available
trophoresis of a native protein sample, and distinct from commercial sources.
bands on the electrophoresis gel are visualized by stain-
ing with D- or L-lactate, or both.
V. STRUCTURES AND STRUCTURE
FUNCTION RELATIONSHIP
IV. UTILIZATION OF LDH IN FOODS
The three-dimensional structures of LDHs were rst
LDHs are used for the quantitative or qualitative determined for vertebrate L-LDHs in apo and ligand
determination of pyruvate, lactate, and related com- complex forms (2, 4), and then extensively analyzed for
Alcohol Dehydrogenase
Sabato DAuria
Institute of Protein Biochemistry and Enzymology, Naples, Italy, and University of Maryland, Baltimore,
Maryland, U.S.A.
II. PROPERTIES AS PROTEIN alcohol formation is favored. ADHs can be very effec-
tive for the conversion of the readily available alde-
Alcohol dehydrogenase (ADH) (EC 1.1.1.1) is an oxi- hydes citronellal and citral for the production of ()-
doreductase enzyme that catalyzes the oxidation of citronellol and geraniol, compounds with a much
alcohols and the reduction of carbonyl compound higher avor impact than the corresponding aldehydes
such as aldehydes and ketones. The enzyme displays or ketones (8). Thus, the potential applications of
a broad substrate specicity, being active on a variety ADHs in industry are numerous. With mesophilic
of primary, secondary, branched, and cyclic alcohols. enzymes, however, limitations due to narrow speci-
Alcohol dehydrogenases are widely distributed city, instability to heat and organic solvents, and loss
enzymes and have been found in many microorgan- of activity on immobilization have been incurred. The
isms, fungi, and animal cells (5). Most bacterial use of these enzymes from thermophilic sources in
ADHs are so-called soluble enzymes and are located immobilized and continuous reactor systems has also
in the cytoplasmic, but some have been shown to be been proposed for the regeneration of NAD(P)H
membrane bound. The intracellular compartmentation needed for such reactors, the lifetime of which would
of eukaryotes gives rise to greater possibilities of var- be extended with stable enzymes. The ADH from
iation in location. Thus, Saccharomyces cerevisiae con- Thermoanaerobium brockii has been shown to reduce
tains three ADHs of which two are cytoplasmic and aliphatic acyclic ketones asymmetrically with an opti-
the third is mitochondrial (6). Some of these enzymes mal purity. Furthermore, this enzyme has been used to
(e.g., ADHs from horse liver and yeast) have been obtain optically pure cyclic ethers, which are con-
extensively studied, because they can provide new stituents of civet used as a xative in the perfume
data on the relationship between structure and func- industry (9).
tion, being oligomeric, coenzyme dependent, and con- ADHs have been subdivided into different families
taining structural and functional metal atoms (5). according to the polypeptide chain lengths: the short-
ADH requires a nicotinamide cofactor for the cat- chain family consisting of non-metalloenzymes with
alytic activity, and it is capable of distinguishing the subunits of 250 residues (10); the medium-chain
diastereotopic methylene hydrogens at the dihydropyr- family with subunits of 350375 residues, often
idine C-4 position of NADH or NADPH, transferring containing zinc (11); and the long-chain family with
the hydride to the substrate stereospecically (7). This subunits of > 700 residues (12). Moreover, ADHs
interesting enzymatic feature suggests a potential use lacking zinc as well as alcohol dehydrogenases requir-
of ADHs for the production of alcohols with a 100% ing iron for activation have been described (13). A
optical yield (7). monomeric ADH was also isolated from
ADHs catalyze the interconversion of alcohols to Saccharomyces cerevisiae, but its metal content is yet
aldehydes/ketones: unknown (14).
ADHs have been classied as = proteins with high
Alcohol NADP Aldehyde=ketone amounts of -helices and -structures. Figure 2 shows
NADPH H the far-UV circular dichroism spectrum of B. acidocal-
darius ADH. The deconvolution of the spectrum
This reaction is highly reversible and the direction of resulted in 22% -helix and 45% -structure, sug-
the reaction is inuenced by pH of the reaction gesting that ADHs from thermophilic microorganisms
medium. At pH 9 the equilibrium of this reaction is could also have a secondary structural organization
directed toward aldehyde formation, while at pH 7 similar to that described for mesophilic ADHs (15).
Figure 4 shows the 22 amino acid residues strictly isolated recently from thermophilic bacteria utilize
conserved in 16 ADHs from different sources (11). It is NADP as coenzyme (29, 30), while the ADHs from
worth noting that His ! Arg substitution at the posi- the eubacteria Bacillus acidicaldarius and Bacillus
tion 47 (HLADH numbering) has been reported to stearothermophilus require NAD as coenzyme (15,
affect both the afnity for the coenzyme and pH opti- 31). Moreover, ADHs containing pyrroloquinolin qui-
mum for the activity in human ADH isoenzymes, the none (Fig. 5) as a cofactor have been isolated from
native enzyme displaying a lower optimum and higher various Gram-negative bacteria. Based on the cofactor
Km and Vmax values than the mutant (18, 25). requirement two classes of ADHs can be distinguished:
Table 1 shows some biochemical features for some quinoprotein alcohol dehydrogenase (containing PQR
ADHs with His or Arg at position 47. and Ca2 ) and quinohemoprotein alcohol dehydrogen-
ase (containing PQQ, Ca2 , and heme c). The rst class
includes the methanol dehydrogenases from methylo-
III. PROPERTIES AS ENZYMES
tropic bacteria and alcohol dehdyrogenases occurring
in Pseudomonas (32).
According to their source ADHs differ in substrate and
HLADH is a well-characterized zinc-containing
coenzyme specicity (5). It is commonly believed that
ADH. Each subunit of the dimeric enzyme contains
ADHs from eukaryotes are mainly NAD dependent,
one catalytic and one structural zinc atom (5) and
while those from prokaryotic organisms are either
binds one ligand of NAD(P)H.
NAD or NADP dependent. However, two ADHs
Figure 4 The 22 amino acid residues strictly conserved in several ADHs (11). The amino acids are numbered according to
HLADH sequence.
The most widely accepted reaction mechanism for The stereospecicity of the ADH isolated from the
alcohol dehydrogenase is indicated in Figure 6 (33) and thermoacidophilic archaeon Sulfolobus solfataricus
includes: binding of NAD ; binding of alcohol sub- (36) toward an asymmetric ketone was examined by
strate by replacing a zinc-bound water molecule; using 3-methyl-butan-2-one as a model compound.
deprotonation of the alcohol; hydride transfer from The hydride attack, using the enzyme immobilized on
the alkoxide ion to NAD , yielding NADH and a Eupergit C was on re-face of the ketone, producing the
zinc-bound aldehyde; dissociation of NADH. In corresponding (s)-3-methyl-butan-2-ol with a 100%
applicable steps (79), the zinc-bound water molecule optical yield. Moreover, the potentiality of the use of
equilibrates with a zinc-bound hydroxyl ion, which this enzyme for the preparation of chiral compounds
gives rise to a dead-end complex. The catalytic zinc was also pointed out by the use of resting cells of S.
in HLADH is bound to Cys46, His67, and Cys174. solfataricus to reduce a series of acyclic, cyclic, and
The fourth ligand is a water molecule that is linked aromatic ketones (37). Another interesting aspect of
to a hydrogen-bonding network involving Ser48 and stereospecicity of ADH from S. solfataricus was dis-
His51 (34). The four cysteine residues coordinating the covered by studying the oxidation of secondary alco-
noncatalytic zinc (Cys97, 100, 103, and 111) and the hol enantiomers (38): (s)-enantiomers are far better
binding amino acids of the catalytic zinc are arranged substrates than (r)-enantiomers. For example, (s)-2-
in a distorted tetrahedral geometry (35). butanol is oxidized about 60 times more efciently
The nicotinamide ring system is redox active, than (r)-2-butanol, and (r)-sec-phenethyl alcohol is
accepting a hydride of two electrons and a proton to not oxidized at all. In a detailed study Keinan and
form 1,4-dihydronicotinamide derivatives of NADH coworkers showed that the ADH isolated from
or NADPH. The reversible hydride transfer from a Thermoanaerobium brockii exhibited a high stereospe-
reduced substrate to NAD , and that from NADH cicity depending on the enzyme assay temperature
to an oxidized substrate, is stereoselective and charac- (39). However, the nicotinamide cofactors are too
teristic of individual enzyme. Each enzyme is able to expensive to be used as stoichiometric reagents in
transfer stereoselectively one of the diastereotopic large-scale synthesis. Thus, recycling of the cofactors
methylene hydrogens at C-4 of NADH to a substrate is needed if enzymes requiring nicotinamide cofactors
carbonyl group or an equivalent sp2 center with high are to be used in a preparative scale. In spite of the use
enantiofacial or diastereofacial selectivity. of HLADH with its broad substrate acceptance, nar-
row stereoselectivity and bidirectional functionality,
there is continuing research for novel reductases, in
particular for alcohol dehydrogenases puried from
thermophilic microorganisms, that are stable and
active at high temperature and in the presence of
organic solvents. Enzymes isolated from thermophilic
sources display high levels of enzyme activity at high
temperatures, being barely active at room temperature.
Figure 7 shows the thermophilic behavior of two
ADHs: B. acidocaldarius ADH (moderate thermophilic
microorganism), and S. solfataricus ADH (hyperther-
mophilic microorganism). Several research groups
Figure 5 Structure of pyrroloquinolin quinone. have related these functional features to a high struc-
tural rigidity at room temperature (40) that could dehydrogenase activity by monitoring the decrease in
prevent the utilization of these enzymes in biotechno- absorbance at 340 nm, as a consequence of the oxida-
logical applications. tion of the NADH (NADPH) molecules.
The alcohol dehydrogenase from yeast exhibits the
maximal enzymatic activity in the range of pH 8.69.0
IV. ENZYMATIC ASSAY at 25 C. The typical assay for measuring it is the fol-
lowing: 50 mM Tris-HCl (pH 8.6), 10 mM ethanol, 1.0
Usually alcohol dehydrogenase activity is assayed by mM NAD , and 1:0 g enzyme/mL nal volume, at
following the increase in absorbance at 340 nm owing 20 C. The ADH activity is determined by monitoring
to the reduction of the NAD NADP molecules. On the increase of absorbance at 340 nm. One unit of
the contrary, it is also possible to assay alcohol enzyme activity is dened as the amount of the enzyme
that catalyzes the reduction of 1:0 M of NAD at
25 C.
For assaying enzymatic activity of thermophilic
alcohol dehydrogenases it is of fundamental impor-
tance to use a sealed cuvette, in order to avoid the
evaporation of the assay mixture. The alcohol dehy-
drogenase from the thermophilic microorganism
Bacillus acidocaldarius shows the maximal enzyme
activity at 80 C. However, Bacillus acidocaldarius
ADH activity is usually assayed spectrophotometri-
cally at 65 C by measuring the increase in the absor-
bance at 340 nm in a sealed quartz cuvette. The
reaction mixture contains 50 mM glycine-NaOH (pH
10.0) and 1:0 g protein/1.0 mL nal volume. The con-
centrations of cofactors and substrate are 5.0 mM
NAD and 3.0 mM ethanol for alcohol oxidation,
and 0.12 mM NADH and 3.0 mM p-anisaldehyde
for aldehyde reduction, respectively. One unit of
enzyme activity is dened as the amount of the enzyme
that catalyzes the reduction of 1:0 M of NAD at
65 C, using ethanol as substrate.
It is also possible to monitor ADH activity by uor-
escence spectroscopy. In this case, we follow the
increase of the uorescence emission at 450 nm
(upon excitation at 340 nm) due to NADH formation.
The uorescence assay is more sensitive than the spec-
Figure 7 Thermophilic behavior of Bacillus acidocaldarius trophotometer one, and it is utilized when it needs the
ADH(A) and Sulfolobus solfataricus ADH (B). detection of low ADH activity levels.
Alcohol Dehydrogenase
complex methodologies have been studied for recycling tion. Pasteur showed that living yeast cells caused fer-
the redox coenzyme, as for instance the coupling of mentation in the absence of air, converting sugar into
ADH with another enzyme together with a coupled ethanol and carbon dioxide. Research later in the 19th
substrate in order to regenerate the coenzyme (12). century showed that fermentation resulted from the
An alternative approach consists in using a coenzyme action of substances contained within the yeast cells.
mimic, bearing functional similarity to NAD (13). Later, one of the major discoveries of fermentation
Independently of the cofactor recycling, ADHs microbiology was made by Hansen at the Carlsberg
obviously have been used for the production of alco- center in Copenhagen while working on wild yeast.
hols during the alcoholic fermentation or the produc- These wild yeast were known to give problems during
tion of aldehydes which represent a major source of beer fermentation and, by isolating pure cultures of
numerous avors. yeast, which he then used in the brewing process,
In the case of alcoholic beverage production, it is Hansen initiated the use of pure cultures in beer pro-
not the enzyme that is used but the cell. The produc- duction. Alcoholic beverages are produced by the alco-
tion of beverages by alcoholic fermentation is one of holic fermentation of the sugar-containing material to
the oldest fermentations known. Beer and wine were ethanol and carbon dioxide (see Chapter 2 for other
probably the two earliest alcoholic drinks produced. A details).
brief history of the knowledge in alcoholic fermenta- Fermentation is carried out by species of the yeast
tion could be summarized as follows. Until Pasteurs Saccharomyces. In some cases, sugar is present natu-
work in the late 19th century, little was known of the rally, such as in grapes used in wine making; in others,
actual processes and mechanism of alcohol produc- sugars are produced from starches in cereals, as in beer
site is then in a more hydrophobic environment in the 3, and 4); 35% are in pleated-sheet regions (I,
holoenzyme than in the apoenzyme. All water mole- II, III), and 14% are in reverse bends. Thus, a
cules that are located in the active site in the apoen- large number of residues, 32%, have no regular sec-
zyme are displaced in the ternary complex, and the ondary structure. An important region in the domain,
catalytic reaction of the enzyme takes place in a com- necessary for the structural stability of the enzyme, is
pletely water-free environment (15). The increase of the lobe comprising residues 95113 that binds the
hydrophobicity around the nicotinamide group in the structural zinc atom by four sulfur atoms from cysteine
ternary enzyme-NAD -alcohol complex facilitates the residues 97, 100, 103, and 111. Although it is bound
reaction (8). near the surface of the molecule, this zinc atom is com-
pletely surrounded by the protein and is not accessible
3. Catalytic Domain in the native conformation of the enzyme (6).
The catalytic domain comprises residues 1175 and
319374 plus two zinc atoms. Both ends of the poly-
4. Substrate-Binding Site
peptide chain are within this region. The two zinc
atoms of the subunit are bound to ligands from the The substrate-binding site, lined almost exclusively by
domain; these bindings are illustrated in Fig. 2. Only hydrophobic residues, is located in a cleft between the
the zinc atom participates in the catalytic activity; the coenzyme-binding core and the catalytic domain. This
second zinc atom is remote from the active site and cleft is open in the apoenzyme form, which is well
may help in stabilizing protein folding (a structural suited for the initial coenzyme binding and does not
zinc). Only 19% of the residues are helical (1; 2; hinder its entrance.
Primary alcohols
Ethanol (a) 1.14 16.3 0.460 2,480
Propanol (b) 8.20 0.390 21,000
Butanol (b) 5.20 0.270 19,000
Pentanol (b) 4.30 0.150 29,000
Hexanol (b) 2.80 0.076 37,000
Heptanol (b) 2.50 0.020 125,000
Octanol (b) 2.30 0.040 58,000
Secondary and branched alcohols
Isopropanol (c) 0.58 9.0 64
2-methyl-propanol (b) 5.30 0.4 13,000
(R)-2-butanol (c) 2.00 7.5 270
(S)-2-butanol (c) 1.00 1.35 740
2-methyl-1-butanol (S) (b) 5.70 0.31 18,000
2-methyl-1-butanol (RS) (b) 5.30 0.52 10,000
3-methyl-1-butanol (b) 2.80 0.15 19,000
(S)-2-pentanol (c) 0.87 0.73 1,200
(R)-2-pentanol (c) 1.01 31.7 32
3-pentanol (c) 2.26 1.6 1,400
(S)-2-octanol (c) 1.52 0.18 8,400
(R)-2-octanol (c) 0.053 1.92 29
Table 4 Kinetics Parameters for Oxidation of Alcohol and Reduction of Aldehydes Catalyzed by
YADH
Oxidation of ethanola
Reduction of aldehyde
pH kcat sec1 ) KM NADH (M) KM substrate (M) kcat =KM (M1 sec1 b
6 125.0 14.4 463 270,000
7.1 125.0 12.5 413 303,000
8 47.6 6.3 186 256,000
9 7.6 1.6 97.7 78,000
Amine Oxidase
the highest absorption at 315 nm and uorescences at length of 380 nm with an excitation wavelength of
380 nm. According to the absorbance or the intensity 315 nm.
of uorescence, the concentration of the product can
be quantitatively determined. The following protocol
2. Radiochemical Method
is a modication of Krajls method (15). In a test
tube, mix 0.25 mL of 200 mM potassium phosphate This method provides more sensitive determination
buffer, pH 7.5, and suitable volume of the enzyme but needs special devices and laboratory for using
sample. Add distilled water to a nal volume of radioisotopes. Many 14 C-labeled substrates for MAO
0.90 mL. Incubate the test tube at 30 C for 5 min, are commercially available. For example, DuPont
then add 0.10 mL 1 mM kynuramine preincubated at NEN(r) provides hydroxytrypamine binoxalate,
30 C, mix immediately, and incubate at 30 C with 5-[2-14 C[tryptamine bisuccinate, [side chain-2-
14
shaking for 20 min. Terminate the reaction by adding C]tyramine hydrochloride, [1-14 C]phenylethylamine
0.4 mL of 20% TCA. Precipitate the protein by hydrochloride, etc. MAO can oxidize these compounds
centrifugation at 10,000 g for 10 min. Take 0.5 mL and produce radioactive products. The enzyme activity
of the supernatant and add 1.5 mL of 1 N NaOH, can be estimated by measuring the radioactivity of the
and measure the intensity of uorescence at a wave- products.
IC50 (M)
Serotonin Deprenyl 1:5 106 5:2 105 human MAO expressed in Cos cell 18
Clorgyline 7:0 1010 human MAO expressed in Cos cell 18
Ro41-1049 3:6 108 human MAO expressed in HEK293 cell 11
Ro41-2064 2:1 107 human MAO expressed in HEK293 cell 11
Phenylethylamine Deprenyl 2:7 109 human MAO expressed in Cos cell 18
6
5:0 10 1:3 107 rat MAO expressed in yeast cell 10
Clorgyline 4:0 107 human MAO expressed in Cos cell 18
2:5 108 7:9 105 rat MAO expressed in yeast cell 10
Lazabemide 1:2 104 1:1 108 human MAO expressed in HEK293 cell 11
Ro41-2064 1:8 105 human MAO expressed in HEK293 cell 11
Subsequently TPQ has been identied biochemically Crystal structures have recently been reported for
and its presence inferred from amino acid sequence several amine oxidases (5862).
homology in copper-containing amine oxidases from
various sources (3440). B. Properties as Enzyme
Amine oxidase cDNA clones have been obtained
from various sources, and the primary structure of The enzyme catalyzes the oxidation by molecular oxy-
the enzyme was deduced from the cDNAs (4152). gen of the primary amino group of di- and polyamines
Sequence alignment of representative amino acid to corresponding amino aldehydes, hydrogen peroxide,
sequences of bacterial, yeast, plant, and mammalian and ammonia. The amino aldehyde products from
amine oxidases suggests separate protein domains putrescine, cadaverine, and spermidine spontaneously
with a carboxyl-terminal domain encoding the cofactor cyclize to 1-pyrroline, 1-piperidine, and 1,5-diaza-
consensus sequence and putative copper-binding sites, bicyclononane, respectively.
and a more divergent amino-terminal domain accom- Substrate specicity varies widely among the dif-
modating differing physiological functions and sub- ferent enzymes (Table 4) (2032, 63, 64). The best
strate specicities of these proteins. There are 33 substrates for all plant enzymes are putrescine and
completely conserved residues mainly within two cadaverine. The relative reaction rates with some sub-
regions: residues 286459 in the pea enzyme, approxi- strates are shown in Table 4. Km values for putrescine
mately centered on Tyr387, and the C-terminal end. seem to be of the same order of magnitude for all
In this enzyme the amino acid sequence of the phe- enzymes investigated (0.1 mM). The pH optimum is
nylhydrazine-labeled peptide was determined to be found around pH 7.0 with putrescine or cadaverine as
VGNXDNVID, in which the X corresponds to substrate. Plasma and tissue amine oxidases also
Tyr387. TOPA quinone is posttranslationally gener- show considerable species-related variations in sub-
ated by oxidation of a specic tyrosine precursor strate specicity. Some endogenously occurring aro-
occurring in the consensus sequence of NYD/E in matic amines such as tyramine and tryptamine are
the active site of amine oxidases (5357). The oxida- metabolized well by copper-containing amine oxidase
tion to TPQ is a spontaneous reaction mediated by in homogenates of rat blood vessels, and also in vitro
the bound copper ion. Numerous studies point inhibition of the enzyme can potentiate vasoconstric-
toward the presence of three histidines as ligands to tor actions of these amines in rat vascular prepara-
copper in the enzyme. One HXH motif 4050 resi- tions. These amines are poor substrates for human
dues toward the carboxyl terminus from the cofactor amine oxidase, thus complicating attempts to general-
consensus sequence is conserved in all alignments. ize possible physiological roles for these enzymes. The
Kinetic parameters
endogenously occurring aliphatic amines methylamine preparations of the enzyme (2833). The extinction
and aminoacetone are metabolized in vitro to formal- coefcient at 500 nm is 4100 M1 cm1 and at 278
dehyde and methyglyoxal, respectively, by amine oxi- nm is 246,000 M1 cm1 for highly puried Lens
dase in some animal (including human) tissues enzyme. The aerobic addition of phenylhydrazine
suggesting the possibility of toxicological conse- and semicarbazide to the Lens enzyme is followed by
quences of cellular function (2123, 2527). the formation of a strong absorption band at 445 nm
Inhibitors are common to all these enzymes. Metal (e 64,000 M1 cm), 492 nm (e 8,600 M1 cm1 ),
chelators such as diethyldithiocarbamate, cuprizone, o- and 360 nm (e 34,800 M1 cm1 ), respectively, con-
phenanthroline, 2,2 0 -bipyridil, 8-hydroxyquinoline, comitant with the disappearance of the absorption
and CN compounds inhibit Cu-containing amine oxi- band at 500 nm. In anaerobiosis a stable yellow inter-
dases noncompetitively with different sensitivities. mediate absorbing at 346, 432, and 462 nm is observed
Carbonyl group reagents irreversibly form adducts in the presence of substrates. Under this condition a
with the enzymes with concomitant loss of catalytic Cu(I)-semiquinone state is generated by substrate
activity. The most effective agents are hydrazines reduction, and this intermediate may react directly
(phenylhydrazine and benzylhydrazine), hydrazides with oxygen. The interaction of the Lens enzyme
(semicarbazide, thiosemicarbazide, and phenylsemicar- with primary amines gives rise to a covalent amino
bazide), hydroxylamine, isoniazide, and iproniazid group quinone compound, a quinoketimine, which is
(2027). converted to quinoaldimine. This releases the aldehyde
In addition to the protein absorbance maximum at product and forms a species containing Cu(II) and a
278 nm, the visible spectrum of the enzyme is charac- reduced form of TPQ, aminocatechol, which coexists
terized by a broad absorption peak centered around with the radical species containing Cu(I) and TPQ
500 nm which confers a typical pink color to puried semiquinone. Finally, the semiquinone reacts with
Prolyl 4-Hydroxylase
(1)
The enzyme from human placenta and chick embryos As noted above, the peptides (Pro-Pro-Gly)5 or
exists as a heterotetramer (an 2 2 tetramer) with a (Pro-Pro-Gly)10 are excellent substrates for detection
Phase I Step 1. Chick embryo calvaria (15 days old) are incu-
bated in the presence of [2,3,4,5-3 H]proline (16 Ci/
E Fe2 ! E Fe2 2
mmol [New England Nuclear, MA] in the proline-
E Fe 2
2-OG ! E Fe 2
2-OG 3 free basal medium (Dulbeccos modied Eagles med-
ium [DME], GIBCO, Long Island, NY) with - 0 -
E Fe 2
2-OG O2 ! E Fe 2
2-OG O2 4 dipyridyl to inhibit endogenous hydroxylation of pro-
collagen chains. First, a basal medium is prepared con-
E Fe 2
2-OG O2 Peptide !
taining DME with NaHCO3 (3.7 g/L), penicillin (100
E Fe2 2-OG O2 Peptide 5 units/mL), streptomycin (100 g/mL), and sodium
ascorbate (50 g/mL) added. This medium is adjusted
Phase II to pH 7.6.
E Fe2 2-OG O2 Peptide ! Step 2. From 100 to 200 calvaria should be used.
The calvaria (one per 0.5 mL) are rst incubated in
E Fe2 Succ CO2 Peptide-OH 6 the basal DME medium without isotope addition at
3940 C for 1 h.
E Fe2 Succ CO2 ! EFe2 Succ CO2 7 Step 3. Next, the calvaria are transferred to basal
E Fe2 ! E Fe2 8 DME, but with the following additions: sodium pyru-
vate (1.0 mM), ; 0 -dipyridyl (0:2 M), and L-
[2,3,4,5-3 H]proline at 5 Ci/mL. Culture asks are
Ascorbic acid is essential in keeping both iron and the
ushed with 95% oxygen 5% CO2 , capped, and the
enzyme in a reduced state. Enzyme activity is lost after incubation is continued for 1224 h.
four to six catalytic turns if a strong reducing agent, such Step 4. The calvaria are harvested by decanting
as ascorbic acid, is not present. The Km values for Fe2 , 2- the medium. The decanted medium is saved, centri-
OG, and ascorbate are 2, 5, and 100 M, respectively. 2-OG, fuged (2000 g for 10 min) to remove particulate mate-
2-oxoglutarate; Succ, succinate. rial, and dialyzed against distilled water to remove
unincorporated [3 H]-L-proline. Following dialysis,
the medium is stored (20 C) and saved for Step 5.
Step 5. The decanted calvaria are homogenized in
D. Inhibitors 0.5 M acetic acid (1:5 w/v) using a Polytron (Brinkman
Instruments) and extracted twice (12 h each with agita-
Several classes of inhibitors for prolyl hydroxylase tion at 4 C). Since the dialyzed medium contains col-
have been described, including proline analogs, agents lagen, it is added to the acetic acid extracts. The
that interfere with 2-oxoglutarate or cofactor binding, combination is exhaustively dialyzed against 0.05 M
and peptides containing 5-oxoproline (13). Table 1 Tris, pH 7.4, until < 100 dpm of [3 H]-L-proline is
lists various inhibitors and their modes of action. detected per mL dialysate.
Active-site modication
Diethyl pyrocarbonate Modication of active site histidine residues, Ki 10 M 7
Substrate binding
Mimosine IC50 100 M 16, 17
S4682 Ki 150 nM 18
Iron chelation
Pirfenidone (methyl-1-phenyl-2-[1 H]- Can be used in vivo (0.5% of diet, w/w) 19
pyridone)
HOE 077 (2,4-pyridine dicarboxylic Can be used in vivo (200 g/g dry food) 20
acid bis[2-methoxyethyl] amide)
; 0 -dipyridyl and related compounds Classical inhibitor of prolyl and other iron-requiring 21, 22
hydroxylases,
Ki 1 M
2,2 0 -bisbipyridine-5,5 0 -dicarboxylic acid Most potent inhibitor of its type, Ki 50100 nM. Note: 23
2,2 0 -bipyridines lacking a 5-carboxylate are poor inhibitors
2-Oxo-glutarate binding
2,7,8-trihydroxyanthraquinone Competes with 2-oxo-glutarate, Ki 40 nM 24
Oxalylalanine and oxalylglycine Structural analogs of 2-oxoglutarate in which the CH2 25, 26
moiety is replaced by -NH-, Ki 100200 nM
Step 6. Following dialysis, the material is ali- 13 100 mm glass culture tubes, and the reaction is
quoted and frozen until use. Note, however, that the stopped by freezing.
storage of tritiated samples in the frozen state can Step 2. The tritiated water that is released (Step 1)
result in some tritium exchange with water. is collected by vacuum distillation (60 C) and radio-
Additional dialysis prior to use or passage through a activity determined (1). Hughes (27) has described a
Sephadex G-10 column equilibrated in sample buffer simple method for measuring release of tritium that
to remove tritiated water is often necessary. As a qual- may also be considered. A useful variant of this
ity control measure, the collagenous nature of the sub- assay is to use ion-exchange chromatography to sepa-
strate may be assessed by digestion overnight with rate the tritiated substrate from titriated water (28),
puried bacterial collagenase (1). The digested material allowing one to omit the vacuum distillation step and
is divided into two portions: one that was dialyzed the need for specialized equipment (see chapter on
against 0.05 M Tris buffer at pH 7.8 and one that Lysyl Oxidase).
was not dialyzed. Loss of > 80% of the tritium follow- Step 3. Data are expressed as tritium released per
ing collagenase digestion and dialysis indicates that mg protein and tissue weight of the extract or DNA in
most of the labeled material is collagen. the sample. This assay is reliable when at least 10 ng of
prolyl 4-hydroxylase is present in samples. The assay is
also useful for semiquantication, if standardized by
B. Assay for Prolyl Hydroxylase using puried or partially puried prolyl 4-hydroxylase
as a reference. The assay is linear over the range of 5-50
ng of enzyme.
Step 1. The assay reaction mixture (0.5 mL) should
contain 80 M FeSO4 , 0.5 mM 2-oxo-glutarate, 0.1 The same procedures may be used to prepare [14 C]-
mM dithiothreitol, 2.0 mM sodium ascorbate, 1 mg L-proline-labeled collagen as substrate. About 40,000
bovine serum albumin, and 300 units of catalase dis- 50,000 dpm of [14 C]-L-proline-labeled collagen is
solved in 0.05 M Tris-HCl buffer, pH 7.8. The sub- needed per assay. This substrate may be stored for
strate should contain at least 100,000 dpm of tritium. long periods (1016 months), in contrast to the tri-
Prior to its addition, the substrate should be boiled for tiated substrate. [14 C]-L-hydroxyproline formed during
810 min to denature the collagen. Reaction mixtures the enzymatic reaction can be measured following
are incubated at 37 C for 1 h on a rotating platform in hydrolysis (6 N HCl, 1224 h, 98 C) by any procedure
Lysyl Hydroxylase
Km Specic activity
Recombinant
lysyl hydroxylase
(rat) 122 106 ND ND 100 60
Lysyl hydroxylase
(chick embryo) 220 70 ND 2 10 20.2
Inhibitor Description
by injection of an equal volume of KH2 PO4 , pH 5.0, buffer in preparation for collagenagarose afnity
buffer. The reaction is allowed to incubate at room chromatography. The collagenagarose afnity col-
temperature for 30 min, after which the lters are umn is prepared by coupling citrate-soluble calf skin
placed in a toluene-based scintillation cocktail (20). collagen to cyanogen bromide-activated agarose (20).
The entire dialysate is applied to the collagenagarose
column. The enzyme is eluted from the column with
VIII. LYSYL HYDROXYLASE enzyme buffer containing 60% (v/v) ethylene glycol.
PURIFICATION As a nal step, the enzyme fractions containing
activity are concentrated to 23 mL by ultraltration
All steps are carried out at 4 C. Chick embryos (14 using an Amicon PM30 membrane. The concentrated
days old; 50100 embryos) are homogenized (1:4) in sample is centrifuged at 20,000 g for 10 min and the
an enzyme buffer (0.2 M NaCl, 0.1 M glycine, 10 M supernatant fraction is loaded onto an 8% agarose gel
dithiothreitol, 0.02 M Tris-HCl buffer, pH 7.5, con- column (Bio-Gel A 1.5 m, 2004000 mesh, Bio-Rad,
taining 0.1% Triton X 100 and 60% (v/v) glycerol). 1 90 cm). The nal puried enzyme is subsequently
This homogenate is centrifuged at 15,000g for 30 min eluted with enzyme buffer. This process usually results
and then subjected to ammonium sulfate precipitation. in a 5000 to 10,000-fold increase in specic activity
The supernatant is saturated to 17% with solid with a recovery of 26%. The enzyme fractions in ethy-
NH4 2 SO4 and centrifuged at 15,000 g for 20 min, lene glycol are the most stable. There is often a large
the pellet is discarded, and the supernatant is once loss of activity during the gel ltration step (20).
again saturated to 55% with NH4 2 SO4 and centri- The puried enzyme has a molecular weight of 180
fuged at 15,000 g for 20 min. This pellet is dissolved kDa determined by gel ltration column chromatogra-
in the enzyme buffer and dialyzed against two 16-L phy under nondenaturing conditions, while it migrates
volumes of the same buffer supplemented with 0.003 at 85 kDa in polyacrylamide gels in the presence of
M MnCl2 overnight. The dialyzed fraction is adjusted SDS. As noted above, the enzyme is a homodimer
to 20 mg protein/mL and next passed (ow rate, 40 (16). Since the protein expressed by insect cells infected
50 mL/h) through a concanavalin A-agarose afnity with baculovirus containing the lysyl hydroxylase
column (4050 mL of packed gel) equilibrated with cDNA sequence from rat is active with the addition
the dialysis buffer. Lysyl hydroxylase is then competi- of cofactors such as ascorbate, -ketoglutarate, and
tively eluted with methyl -D-glucoside (1 M) dis- ferrous iron (17), it is likely that no other cofactors
solved in the enzyme buffer and ethylene glycol remain to be discovered. That insect cells infected
(40%:60%, v/v). with baculovirus containing the lysyl hydroxylase
The efuent from chromatography on concanavalin cDNA express lysyl hydroxylase with a similar specic
A-agarose is next dialyzed overnight against enzyme activity to that of the chick embryo lysyl hydroxylase
Lysyl Oxidase
1 NH2 -Met-Arg-Phe-Ala-Trp-Thr-Val-Leu-Phe-Leu-Gly-Gln-Leu-Gln-Phe-Cys-Pro-Leu-Leu-Arg-
21 Cys-Als-Pro-Gln-Ala-Pro-Arg-Glu-Pro-Pro-Ala-Ala-Pro-Gly-Ala-Trp-Arg-Gln-Thr-Ile-
41 Gln-Trp-Glu-Asn-Asn-Gly-Gln-Val-Phe-Ser-Leu-Leu-Ser-Leu-Gly-Ala-Gln-Tyr-Gln-Pro-
61 Gln-Arg-Arg-Arg-Asp-Ser-Ser-Ala-Thr-Ala-Pro-Arg-Ala-Asp-Gly-Asn-Ala-Ala-Ala-Gln-
81 Pro-Arg-Thr-Pro-Ile-Leu-Leu-Leu-Arg-Asp-Asn-Arg-Thr-Ala-Ser-Ala-Arg-Ala-Arg-Thr-
101 Pro-Ser-Pro-Ser-Gly-Val-Ala-Ala-Gly-Arg-Pro-Arg-Pro-Ala-Ala-Arg-His-Trp-Phe-Gln-
121 Val-Gly-Phe-Ser-Pro-Ser-Gly-Ala-Gly-Asp-Gly-Ala-Ser-Arg-Arg-Ala-Ala-Asn-Arg-Thr-
141 Ala-Ser-Pro-Gln-Pro-Pro-Gln-Leu-Ser-Asn-Leu-Arg-Pro-Pro-Ser-His-Val-Asp-Arg-Met-
161 Val-Gly-Asp-Asp-Pro-Try-Asn-Pro-Tyr-Lys-Try-Ser-Asp-Asp-Asn-Pro-Tyr-Tyr-Asn-
181 Tyr-Asp-Thr-Tyr-Glu-Arg-Pro-Arg-Ser-Gly-Ser-Arg-His-Arg-Pro-Gly-Tyr-Gly-Thr-Gly-
201 Tyr-Phe-Gln-Tyr-Gly-Leu-Pro-Asp-Leu-Val-Pro-Asp-Pro-Tyr-Tyr-Ile-Gln-Ala-Ser-Thr-
221 Tyr-Val-Gln-Lys-Met-Ser-Met-Tyr-Asn-Leu-Arg-Cys-Ala-Ala-Glu-Glu-Asn-Cys-Leu-Ala-
241 Ser-Ser-Ala-Tyr-Arg-Ala-Asp-Val-Arg-Asp-Tyr-Asp-His-Arg-Val-Leu-Leu-Arg-Phe-Pro-
261 Gln-Arg-Val-Lys-Asn-Gln-Gly-Thr-Ser-Asp-Phe-Leu-Pro-Ser-Arg-Pro-Arg-Tyr-Ser-Trp-
281 Glu-Trp-His-Ser-Cys-His-Gln-His-Tyr-His-Ser-Met-Asp-Glu-Phe-Ser-His-Tyr-Asp-Leu-
301 Leu-Asp-Ala-Ser-Thr-Gln-Arg-Arg-Val-Ala-Glu-Gly-His-Lys-Ala-Ser-Phe-Cys-Leu-Glu-
321 Asp-Thr-Ser-Cys-Asp-Tyr-Gly-Tyr-His-Arg-Arg-Phe-Ala-Cys-Thr-Ala-His-Thr-Gln-Gly-
341 Leu-Ser-Pro-Gly-Cys-Tyr-Asp-Thr-Tyr-Ala-Ala-Asp-Ile-Asp-Cys-Gln-Trp-Ile-Asp-Ile-
361 Thr-Asp-Val-Gln-Pro-Gly-Asn-Tyr-Ile-Leu-Lys-Val-Ser-Val-Asn-Pro-Ser-Tyr-Leu-Val-
381 Pro-Glu-Ser-Asp-Tyr-Ser-Asn-Asn-Van-Val-Arg-Cys-Glu-Ile-Arg-Tyr-Thr-Gly-His-His-
401 Ala-Tyr-Ala-Scr-Gly-Cys-Thr-Ilc-Scr-Pro-Tyr-COOH
Underlined and highlighted are sites associated with: leader sequence cleavage (Cys-Ala), glycosylations
(Ans-Arg-Thr), prolysyl oxidase peptide cleavage to lysyl oxidase (Gly-Asp-Asp), potential trypsin-like
proteinase cleavage sites (Arg-Arg and Arg-Arg-Arg), the proposed copper binding region
(Trp-His-Ser-Cys-His-Gln-His-Tyr-His-Ser), and the location of the Lys and Tyr associated with lysine
tyrosylquinone formation (39, 40, and Refs. cited).
kcat =Km increases with increasing peptide length. recombinant tropoelastin (MW 70,000, 56 residues
Insertion of a Glu residue immediately preceding the of Lys per 1000 total residues) is 5 M (29).
Lys residue causes an increase in the kcat =Km 10-fold
over that for -Lys-Glu- and about vefold over B. Effect of Environmental Factors
sequences containing only Lys. Replacement of Glu
with Gln increases the Km and lowers kcat =Km . Lysyl oxidase is optimally active at pH 7.88.6 and 50
Changes in kcat =Km also respond to the carbon chain 60 C. No externally added cofactors are required, since
length of the vicinal amino acid at this position. In copper and TPQ are tightly associated with the
particular, decreasing the carbon chain length enzyme.
decreases kcat =Km (26, 27). For simple aliphatic
mono- and particularly diamines, as the carbon chain
C. Proposed Mechanism
length decreases, such compounds become poorer sub-
strates; e.g., ethylenediamine is an irreversible inhibitor
Lysyl oxidase carries out oxidative deaminations by a
of lysyl oxidase (28). The Km s for simple amine sub-
classical Ping-Pong mechanism [Eqs. (2) and (3)]
strates are in the 100 M to mM range (see Ref. 2 and
Refs. cited within). The Km for the oxidation of A RCH2 NH2 BO2 !P RCHO
2
Q H2 O2 and NH3
(3)
Shah et al. (30) have studied selected aspects of this al. (30) noted, such stereospecicity and proton
mechanism. To iterate, substrate efciency often exchange uniquely differentiates lysyl oxidase from
decreases with decreased carbon chain length. most of the other semicarbazide-sensitive amine oxi-
Regarding the two half reactions, enzyme reoxidiza- dases.
tion is the most rate-limiting step. 1 H NMR spectro-
scopy of the alcohol that is reductively derived D. Inhibitors
(nonenzymatically) from the aldehyde product of the
lysyl oxidasecatalyzed oxidation of deuterated tyra- The most potent inhibitors are derivatives of semicar-
mine indicates that the pro-S, but not the pro-R, bazides and aminoallylnitriles, e.g., -aminopropioni-
alpha-deuteron is catalytically abstracted. As Shah et trile (BAPN). BAPN administration in vivo promotes
Superoxide Dismutase
Figure 1 Comparison of amino acid sequences of human, rat, bovine, horse, yeast, and spinach Cu,Zn-SODs. Asterisks indicate
identical amino acid residues.
that EC-SOD is a secreted protein. The N-terminal Since the central core active site portions of Cu,
region (amino acid residues 195) of the mature Zn-SOD and EC-SOD are conserved, the enzymatic
enzyme shows no homology with other proteins, but properties of the two enzymes are very similar.
the region corresponding to amino acid residues 96 However, the properties of Mn-SOD, which has no
193 shows 50% homology with the carboxyl-term- structural homology of active site with that of Cu,Zn-
inal two-thirds of the sequences of eukaryotic Cu,Zn- SOD and EC-SOD, are signicantly different from
SODs. The amino acid residues binding the Cu and Zn those of the other two enzymes. Both Cu,Zn-SOD
atoms (Cu atom ligands His96, -98, -113, and His163; and EC-SOD are very sensitive to cyanide. The values
Zn atom ligands His113, -121, -124, and Asp127) and of the IC50 of cyanide for human Cu,Zn-SOD and
the cysteines forming the intrasubunit disulde bridge human EC-SOD are 10 M and 3 M, respectively
(Cys107-Cys189) in EC-SOD can all be identied with (10). In contrast, Mn-SOD is insensitive to cyanide
corresponding residues in Cu,Zn-SOD (Fig. 2). The and the enzymatic activity of Mn-SOD is not inhib-
carboxyl-terminal region (194222), including nine ited at 10 mM cyanide (11). All SOD isoforms are
amino acids with a positive charge, confers the afnity inhibited by azide. The values of the IC50 of azide for
of EC-SOD for heparin and heparan sulfate. There is Cu,Zn-SOD, Mn-SOD, and EC-SOD are 21, 20, and
one deduced N-glycosylation site (Asn89). 6.5 mM, respectively (10, 12). Cu,Zn-SOD and EC-
The primary structure and three-dimensional struc- SOD are sensitive to H2 O2 , while Mn-SOD is resis-
ture of Mn-SOD show no homologies with those of tant to it (10, 11). Cu,Zn-SOD and EC-SOD are inac-
Cu,Zn-SOD or EC-SOD (7). tivated at 1 mM H2 O2 with half-times of 9.3 and 6.2
7.1 min, respectively.
Polyphenol Oxidase
I. INTRODUCTION
Grenache Sultana
Tomato Tobacco grape grape Apple Bean Pokeweed Spinach Sugarcane
Tomato 100 90.48 59.52 60.03 56.97 56.8 55.2 53.74 43.2
Tobacco 100 58.01 58.52 56.49 56.83 53.83 51.43 43.68
Grenache grape 100 99.18 67.17 64.42 64.22 57.83 43.16
Sultana grape 100 67.34 64.42 64.4 57.73 43.09
Apple 100 69.7 63.03 54.88 45.29
Bean 100 62.69 56.84 43
Pokeweed 100 57.58 43.44
Spinach 100 40.42
Sugarcane 100
The GCG program OldDistances was used to calculate percent similarity among the plant PPOs from a complete PileUp alignment; part of this
alignment is depicted in Figure 1. Genbank accession numbers are as follows: tomato, Z12837 S61013; tobacco, Y12501; Grenache grape,
U83274; Sultana grape, Z27411; apple, D87670; bean, Z11702 S45506; pokeweed, D45385; spinach, X90869; sugarcane, U46014.
ure of BH2 is not known. Lerch and Ettlinger (3) Kinetically, the mechanism of peach polyphenol
reported lag periods of 0.3216.0 min for S. glauces- oxidase (5) and S. glaucescens polyphenol oxidase (3)
cens polyphenol oxidase acting on N-acetyl-L-tyrosine followed an ordered BiBi mechanism (Fig. 5). The O2
hydrazide and p-hydroxyphenylacetic acid, respec- must bind rst to the deoxy form (Cu(I) state) of the
tively. Whitaker (unpublished data) showed that 1 enzyme followed by phenol. The ordered BiBi mechan-
107 M 4-methylcatechol added to mushroom poly- ism assumes that the products come off in the order of
phenol oxidase, along with p-cresol, was able to elim- the benzoquinone followed by the hydrogen peroxide.
inate the 9-min lag period entirely. Equation (4) shows This author does not know of published experimental
that the BH2 acts on the Cu(II) met form to reduce the data to support this assumption.
two coppers to Cu(I) (deoxy form), which can then The overall proposed mechanism of action of poly-
bind O2 and oxidize the monophenol substrate to the phenol oxidase is shown in Figure 6. The top part
o-diphenol. Would similar experiments with other shows the pathway for oxidation of o-diphenols. The
polyphenol oxidases show that all polyphenol oxidases met form (at the No. 1 position in the A cycle) is
can do both types of reactions? thought to be the resting form of the enzyme when
Broad bean
Substrate Potatoa Peachb leafc Peard
absorbance after 2 min as the benzoquinone is con- In the measurement of monophenol activities, the
verted to other products that do not absorb maximally rate of conversion of the intermediate product (an o-
at the wavelength of benzoquinones. Therefore, it is diphenol) to the benzoquinone is generally measured.
critical to obtain initial velocity (v0 ) early on in the But in some cases, kcat values for the monophenol and
reaction. As shown by Eqs. (1) and (2), the oxidation product o-diphenol can be very similar (3). Examples
of monophenols to benzoquinone plus H2 O requires are shown in Table 7.
two equivalents of oxygen, while oxidation of o-di- The monophenol oxidation can be measured
phenols to benzoquinone requires one equivalent of uniquely by use of 18 O-labeled O2 or tritium-labeled
oxygen. substrate.
In reactions involving monophenols, a lag phase is
generally observed (see Fig. 4). As shown in Eq. (1) A. Measurement of Monophenolase Activity
and Figure 6, BH2 is required to reduce the met form
1. Use of Tritium-Labeled Substrate
of polyphenol oxidase to the deoxy form that can bind
O2 . Addition of as little as 1 107 M catechol to the As shown in Eq. (1), one atom of oxygen (from O2 ) is
reaction can eliminate the lag period. covalently attached to the 2 position of p-cresol, along
Table 7 Ratios of kcat =Km and kcat =kcat , in Parentheses, for Some Monophenol/o-Diphenol
Substrate Pairs
Substrate pairs
2. Polarographic Method
The polarographic method, also called the O2 -elec-
trode method, determines the rate of O2 uptake. As
shown in Figures 7 and 8 and Table 6, the O2 electrode
and spectrophotometric methods give linear relation-
ships over a wide range of enzyme concentrations. The
chronometric method measures the lag time before
browning occurs or ascorbic acid is exhausted, and
the manometric method (O2 uptake) does not give
linear relationships between enzyme concentration Figure 8 Relationship between enzyme concentration and
and O2 uptake. L O2 consumed by oxidation of 4-methyl catechol by
The concentration of enzyme, substrate, buffers, and apple catechol oxidase using spectrophotometric, chrono-
pH used may be the same as described for the spectro- metric, and Warburg techniques. The reaction was per-
photometric method (above). Specic conditions are formed under the experimental conditions given in Figure
also given in the legends of Figures 7 and 8 (47). 7. (From Ref. 47.)
Figure 9 Mono Q [R-CH2 -N CH3 3 HR 5/5 column chromatography of grape polyphenol oxidase. Protein was eluted
initially with a sodium chloride linear gradient (00.3 M), held for 2 mL at 0.3 M sodium chloride, and then stepped up to 1
M (dashed line). Activity of polyphenol oxidase is indicated by closed circles (and shaded area). Protein is indicated by open
squares. The small peak of activity at Fraction No. 30 is that designated as Mono Q II in the text.
the supernatant was concentrated by ultraltration substrate and measuring the initial change in absor-
(Amicon microconcentrator, Diao PM10 membrane). bance at 420 mM (pH 6.5, 25 C). One unit of activity
The concentrated solution was brought to 40% is dened as an increase in absorbance of 0.001/min.
saturation with ammonium sulfate, let stand for 30 The Mr of the mature PPO was determined on gel
min, centrifuged, and the precipitate discarded. The ltration columns. Mr was 41:2 2 kDa (replications
supernatant was further brought to 95% saturation 4) on Sephadex G-75 (ne), 34.6 kDa on Superose
with ammonium sulfate, let stand for 30 min, centri- 12 column (50), 28.0 kDa on TSKgel G3000 SW, and
fuged, and the supernatant separated. The supernatant 22.9 kDa on TSKgel 2000 SW. Superose 12 columns
still contained 50% of total activity; owing to very low are known to give Mr values that are too low (51).
protein concentration, some PPO remained in solution. Note that the Mr values on TSK gels are even lower
The precipitate was stored at 4 C (or frozen) until all than those on Superose 12. All standard proteins were
extractions were done. the same (bovine serum albumin, ovalbumin, -lacto-
The precipitates from all extractions were com- globulin [dimer], chymotrypsinogen, and ribonuclease
bined, dissolved in a minimal amount of buffer (5 A). Based on our use of Sephadex columns for Mr
mM sodium phosphate buffer, pH 8.0), and dialyzed determination for 37 years, we suggest the correct Mr
with 200 volumes of the same buffer. The sample was is near 41.2 kDa.
applied to a Mono Q [(R-CH2 -N CH3 3 HR 5/5 col-
umn linked to an FPLC system and the polyphenol
oxidase was eluted with 0.3 M NaCl in the 5 mM REFERENCES
sodium phosphate buffer, pH 8.0, as shown in Figure
9 (Mono Q I peak; at Fraction No. 26). A second, 1. CF Schoenbein. On ozone and oronic actions in
smaller peak (Mono Q II) with polyphenol oxidase mushrooms. Phil Mag 11:137141, 1856.
2. Enzyme Nomenclature. Recommendations of the
activity and brown color was eluted with 1 M NaCl
Nomenclature Committee of the International Union
at Fraction No. 31. Native PAGE of Mono Q I peak
of Biochemistry. San Diego: Academic Press, 1992.
showed only one protein band (Coomassie Blue R-250 3. K Lerch, L Ettlinger. Purication and characteriza-
stain) and one concomitant activity band (stained with tion of a tyrosinase from Streptomyces glaucescens.
8 mM catechol, 0.06% 3-methyl-2-benzothiazolinone Eur J Biochem 31:427437, 1972.
hydrazone [MBTH]). Native PAGE of the Mono Q 4. TC Wong, BS Luh, JR Whitaker. Isolation and char-
II peak gave ve protein bands, all of them active. acterization of polyphenol oxidases of clingstone
Most likely, these are modied PPO owing to reaction peach. Plant Physiol 48:1923, 1971.
with PPO-produced benzoquinones. The purication 5. NJ Rivas, JR Whitaker. Purication and some prop-
of Grenache PPO is summarized in Table 8. We used erties of two polyphenol oxidases from Bartlett pears.
the rst peak (Mono Q I) for crystallization, which Plant Physiol 52:501507, 1973.
6. AC Allan, JRL Walker. The selective inhibition of
corresponded to unmodied polyphenol oxidase as
catechol oxidases by salicylhydroxamic acid.
indicated by no brown color.
Phytochemistry 27:30753076, 1988.
Protein was determined using the 1976 Bradford 7. JRL Walker, RF McCallion. The selective inhibition
assay (48) or by following the absorbance at 214 nm of ortho- and para-diphenol oxidases. Phytochemistry
for column chromatography. Activity was determined 19:373377, 1980.
by a modication of the colorimetric method described 8. WH Flurkey, B Ratcliff, L Lopez, J Kuglin, RM
by Ponting and Joslyn (49) using 5 mM catechol as Dawley. Differentiation of fungal tyrosinases and lac-
Laccase
William H. Flurkey
Indiana State University, Terre Haute, Indiana, U.S.A.
ments during chestnut blight pathogenic infections. products formed during enzymatic oxidations. Some of
Although laccases are different from lignin peroxi- these tests have been used to distinguish white rot fungi
dases, Mn peroxidases, and other peroxidases, they from brown rot fungi and have been used for taxo-
can utilize some of the same substrates, but they are nomic classication. Immunocytochemical location of
proposed to be a less strong oxidant than these perox- the fungal laccase from Rigidoporous lignosus grown
idases. Because laccases can oxidize a variety of phe- on wood was carried out using antilaccase polyclonal
nolic and phenolic-like compounds, they can also sera and immunogold labeling (17). These studies
cause browning-like reactions in fruits and vegetables. found laccase to be located in fungal cytoplasm, in
The browning reactions decrease consumer product vesicle-like structures at the plasmalemma, in the cell
appeal, nutritional value, and marketability. The asso- wall, and in the extracellular slime layer connecting
ciation of fungal laccases with many plant and fruit fungal cells to wood. In liquid-grown cultures of the
products poses other potential problems related to fungus Coprinus congregatus, laccase located in the
browning and crosslinking of polymeric material, and hyphal tips was suggested to be involved with sclero-
could affect product characteristics and properties dur- tia/primordia formation (18).
ing storage and processing. In plants, cytochemical localization of laccase has
relied upon the use of syringaldazine, 2,7-diamino-
uorene (DAF), 2,2 0 -azinobis(3-ethylbenzthiazoline-
III. LOCATION OF LACCASE
6-sulfonate; ABTS), and 4-methylcatechol as sub-
A. Tissue Localization strates. Laccase was localized to active lignifying
zones of xylem cells in loblolly pine sections using
In plants, laccase has been located in lignifying xylem DAF as a substrate (19). Similarly, cytochemical loca-
cells where it is associated with cell walls. Extracellular lization of laccase in tobacco stem cells was found in
laccase is secreted by plant cell cultures. In these cul- the outermost lignifying zones in xylem tissue (20).
ture conditions, laccase was not found in the cyto- Driouch et al. (21) found laccase excreted in the
plasm, but it was associated with cell walls. An extracellular medium and associated with cell walls in
epidermal location of laccase was indicated in stem sycamore cell cultures by cytochemical and immuno-
tissue. cytochemical methods. In contrast, De Marco and
Roubelakis-Angelakis (22) observed laccase in areas
B. Methods for Determining Location where lignication was thought to occur in tobacco
protoplasts.
Spot tests, either of liquid culture samples or cultures Tissue printing of stem cross sections using 4-methyl
grown on solid media, have relied on the use of colored catechol and syringaldazine as cytochemical substrates
177 178 179 238 239 240 718 719 720 721 722 723 803 804 805 806 807 808 809 810 811 812 813 814
Laccase
Aspergillus nidulans H W H H S H H P I H K H H C H I A S H Q M G G M
Phebia radiata H W H H S H H P F H L H H C H I D W H L E A A L
Coriolus hirsutus H W H H S H H P F H L H H C H I D F H L E A G F
Neurospora crassa H W H H S H H P I H L H H C H I A W H V S G G L
Cryphonectria parasitica H W H H S H H P I H L H H C H I A W H V S A G L
Filobasidiella neoformans H W H H S H H P Y H L H H C H I G W H L T E G K
Pleurotus ostreatus H W H H S H H P F H L H H C H I D W H L E I G L
Agaricus bisporus H W H H A H H P F H L H H C H I D W H L E A G L
Basidiomycete PM1 H W H H S H H P F H L H H C H I D F H L E A G F
Acer pseudoplatanus H W H H A H H P M H L H H C H F E R H T T W G M
Ascorbate oxidase
Cucumber H W H H G H H P W H L H H C H I E P H L H M G M
Pumpkin H W H H G H H P W H L H H C H I E P H L H M G M
Zucchini H W H H G H H P W H L H H C H I E P H L H M G M
2 3 3 3 1 2 3 3 1 3 1 1
Numbers at the top of the columns represent alignment of amino acid residue numbers in ascorbate oxidase. Numbers at the bottom of the columns represent amino acid residues
associated with the type 1, 2, or 3 copper centers of both enzymes.
Temperature Temperature
Laccase source pH optimum pH stability optimum ( C) stability ( C)
Fungal
Coriolus hirsutus 3.54.5
Schizophyllum commune 5.46.0
Rhizoctonia practicola 67
Coriolus versicolor 45
Agaricus bisporus 3.6, 5.6 t1=2 40 min @ 60
Pleurotus eryngii I 4 57 65 t1=2 30 min @ 50
Pleurotus eryngii II 3.5 812 55 t1=2 25 min @ 55
Basidiomycete PM1 4.5 39 80 60 min @ 60
Trametes villosa 1,2,3 55.5
Pleurotus ostreatus POXC t1=2 30 min @ pH 3 5060 t1=2 30 min @ 60
POXA1 t1=2 24 h @ pH 3 4565 t1=2 200 min @ 60
POXA2 t1=2 2 h @ pH 3 2535 t1=2 10 min @ 60
Monocillium indicum 3.0, 5.0 26 60 t1=2 10 min @ 70
Cerrena unicolor 5.05.6 68 60
Plant
Schinus molle 6.2
Acer pseudoplatanus 6, 6.8 410
forms varies considerably. Some are stable over a wide D. Mechanism of Action
pH range (pH 37) while others are more pH sensitive.
Many laccases are relatively heat stable, but specic Laccases contain three copper centers. The type 1 cop-
thermal stability properties are associated with each per center (T1) is responsible for the blue color of the
laccase isoform (Table 3). Half-lives of 30 min at 50 protein because of the intense absorption at 600 nm.
65 C for laccase isoforms are not uncommon, while T1 centers also show small hyperne splitting in an
temperature optima are generally > 50 C. electron paramagnetic spectrum (EPR). Type 1 centers
are proposed to involve the 1e oxidation of the sub-
C. Inhibitors strate. Type 2 (T2), or normal copper centers, have no
absorption to the visible spectrum and show EPR spec-
Laccases are inhibited by a variety of compounds trum characteristic of tetragonal Cu complexes. Type 3
that can be classied into groups such as Cu centers (T3) contain two copper ions and are termed
chelators, reducing agents, detergents, and free coupled binuclear copper centers. They have no EPR
radical scavengers. Cu chelators include EDTA, signals but do absorb in the UV spectrum (330 nm). In
CN , sodium azide, and diethyldithiocarbamic acid. laccases, T2 and T3 centers form a trinuclear copper
Cetyltrimethylammonium bromide (CTAB), a catio- cluster. This cluster is responsible for the 4e reduction
nic detergent, seems to inhibit some laccases in a com- of molecular oxygen to water.
petitive or noncompetitive manner (36). Short-chain The exact mechanism for laccase catalysis has not
carboxylic acids have also been reported to inhibit been elucidated. A proposed mechanism appears in the
some laccases (37). Desferal inhibits laccase by deacti- work by Solomon et al. (6). In this mechanism an elec-
vating phenoxy radicals (38). Tiron, a superoxide ion tron from the substrate is transferred to the T1 copper
and hydroxyl ion quinone scavenger, also decreased site. The site transfers an e to the T2T3 sites.
the activity of laccase-like oxidases in tobacco xylem Reduction of the trinuclear cluster is thought to
(39). Halides, especially F , are potent inhibitors of occur by either sequential reduction of each copper
laccase (40). N-Hydroxyglycine has been reported to in the trinuclear cluster or by reduction of the T3 cop-
be a specic inhibitor of laccase by Murao et al. (41) per pair by T1 and T2. Because there are a total of 4
and later by others (4244). However, recent work in Cu ions, a variety of intermediates with different Cu
our laboratory suggests this may be an artifact (data oxidation states in the copper sites have been postu-
not shown). lated. A somewhat different mechanistic scheme is
Source: Dr. D. Yaver and Dr. F. Xu at Novo Nordisk Biotech Inc., Davis, CA.
Because fungal and plant laccases are glycoproteins, activity and levels. The distribution of expressed lac-
some purication methods have employed lectin af- case isoforms is often dependent on the type of culture
nity chromatography on concanavalin A (Con A) conditions (shake vs. stationary), culture medium, and
Sepharose. Following is a summary of the purication phenolic inducer used. Many of the puried laccases
used by Drouich et al. (21) to isolate laccase from have small fold purication factors (based on the spe-
sycamore cell cultures. Proteins in the extracellular cic activity of starting material compared to puried
uid were precipitated with 5 M ammonium sulfate samples), indicating the laccase is a major protein in
at 4 C. After centrifugation of the precipitated protein, the extracellular uid or that there is a signicant loss
the pellet was dissolved in 10 mM Tris buffer (pH 7.4) in activity during purication. Puried laccases are
containing 1 mM MnCl2 and 1 mM CaCl2 . The sample typically blue in color when concentrated, but the
was applied to a Con A Sepharose column Agaricus bisporus laccase was reported to be yellow
(1:5 30 cm) equilibrated in the above buffer. in color (34) and a white laccase (50) has also been
Laccase was eluted from the column with buffer con- reported. Yellow laccases, modied by reactive pro-
taining 50 mM methylmannoside. Laccase was preci- ducts of lignin degradation, may be generated from
pitated from the solution with 5 M ammonium sulfate, blue laccases during the isolation of laccases (35).
washed with cold acetone, and dissolved in a small Puried laccases are usually stable for several years
volume of 10 mM Tris buffer (pH 8.8), containing when stored frozen. Rhus vernicifera laccase is com-
10% glycerol. The sample was subjected to preparative mercially available from Sigma Chemical Co.; a
native gel electrophoresis using the Laemmli system. Coriolus versicolor laccase is available from Mercian
After locating the laccase activity by staining the gel Corp., Tokyo.
briey with catechol, areas of active enzyme were
excised from the gel and subjected to electroelution.
Further purication of laccase for antibody production REFERENCES
used preparative SDS-PAGE, followed by electro-
elution. 1. H. Yoshida. Chemistry of lacquer (Urushi), part 1. J
Chem Soc 43:472486, 1883.
B. Other Aspects of Purication 2. G Bertrand. Sur la presence simultanee de la laccase et
de la tyrosinase dans le sue quelues champignons. C R
Hebd Seances Acad Sci 123:463465, 1896.
Laccases are generally expressed constitutively in cell
3. J Laborde. Sur lacasse des vins. C R Hebd Seances
culture, but higher levels can be produced using phe- Acad Sci 123:10741075, 1896.
nolic compounds as inducers. Some of these inducers 4. A Messerschmidt. Multi-Copper Oxidases. New
include phenolic acids, anthroquinones, or 2,5-xyli- Jersey: World Scientic, 1975.
dine. Chelators, such as EDTA and EGTA, in the 5. CF Thurston. The structure and function of fungal
culture medium have been found to repress laccase laccases. Microbiol. 140:1926, 1994.
I. INTRODUCTION but does not exhibit much activity with proteins such
as reduced RNase, whereas the mammalian enzyme
Most of the characteristics of sulfhydryl oxidase (EC exhibits no activity with DTT, contains iron, and
1.8.3.) reported in this chapter are those of the enzyme rapidly oxidizes reduced proteins (7). A mammalian
isolated from the skim milk membrane vesicles avoprotein sulfhydryl oxidase that oxidizes DTT
obtained from bovine milk (13). The enzyme is an has been isolated from the male reproductive tract
integral membrane iron-containing glycoprotein (810).
found in mammary tissue, kidney, and pancreas (4).
Immunouorescent staining also indicated its location
in the endothelial cells lining the capillaries of the II. IMPORTANCE TO FOOD QUALITY
kidney, heart, and small intestine (4). The enzyme cat- AND UTILIZATION IN FOOD
alyzes oxidation of sulfhydryl groups of cysteine, PROCESSING
cysteine-containing peptides, and proteins to disuldes
with molecular oxygen as the electron acceptor accord- Heating of milk causes protein denaturation resulting
ing to the stoichiometry given below (1, 5): in exposure of the sulfhydryl group, Cys121, in -
lactoglobulin and other chemical reactions of the sul-
2RSH O2 ! RSSR H2 O2
fhydryl and disulde residues of proteins yielding vola-
This enzyme differs from thiol oxidase (EC 1.8.3.2) tile sulfur compounds resulting in formation of a
not only by its substrate specicity and protein char- cooked avor. Ultra-high-temperature (UHT) steri-
acteristics but also in the fact that water is a product of lized and aseptically packaged milk, typically processed
the thiol oxidasecatalyzed oxidation (6), whereas, at 140150 C for 24 sec, initially has strong cooked
hydrogen peroxide is the product of sulfhydryl oxi- avor that slowly dissipates with storage (11). The
dasecatalyzed reactions. Also, sulfhydryl oxidase dif- intensity of the avor is objectionable to most North
fers from glutathione oxidase (EC 1.8.3.3) because of Americans and northern Europeans who are accus-
its broader substrate specicity and it does not contain tomed to high-quality refrigerated pasteurized milk.
FAD (6). Hence, the enzyme is a sulfhydryl:oxygen UHT treatment causes extensive denaturation of the
oxidoreductase (EC 1.8.3. ), but it has not been globular whey proteins and complete loss of sulfhydryl
assigned a serial number as yet. oxidase activity. Dissipation of the cooked avor
The mammalian enzyme is also distinct from micro- occurs concomitantly with the oxidation of sulfhydryl
bial sulfhydryl oxidase. The microbial enzyme from groups. Furthermore, sulfhydryl oxidasecatalyzed
Aspergillus niger is a soluble avoprotein that oxidizes oxidation of the exposed sulfhydryls also results in
small thiol compounds such as dithiothreitol (DTT) elimination of cooked avor (1116).
Relative pseudo-rst-order
Substrates Km (mM) rate constant (Vmax =KM )
As indicated by the kinetic parameters in Table 1, activity is restored within minutes in the presence of
reduced proteins are excellent substrates. Fully sulfhydryl oxidase, whereas > 15 h is required in its
unfolded and reduced RNase A (1, 28, 30) and chymo- absence. Restoration of biological activity proves
trypsinogen A (31) are rapidly oxidized with nearly that native disulde bonds are formed. Comparison
complete restoration of biological activity (Figs. 3, of the intermediates formed during enzyme-catalyzed
4). Thus, sulfhydryl groups are oxidized and biological oxidation and air or GSH/GSSG-catalyzed oxidation
Figure 3 Comparison of the rates of sulfhydryl oxidase Figure 4 Comparison of the rates of sulfhydryl oxidase
catalyzed oxidation (*) and activity regeneration (*) with catalyzed oxidation (*) and restoration of trypsin-activata-
that observed in air (~, ~) for reduced ribonuclease. Also ble chymotryptic activity *) with that observed in air (~,
included are the data (&) obtained with the microbial ~) for immobilized reduced chymotrypsinogen A. Activity
enzyme from A. niger (7). Activity data are plotted as data are plotted as (Y1 Y)/(Y1 Y0 ) where Y1 and Y0 are
(Y1 Y)/(Y1 Y0 ) where Y1 and Y0 are the activities at in- the activities at innite and 0 time, respectively. Nearly 50%
nite and 0 time, respectively. Nearly 100% of the activity was of the original biological activity was restored. (From Ref.
restored. (From Refs. 1 and 7.) 31.)
Table 2 Three Methods for Preparing Skim Milk Membrane Vesicles from Whey
1. Adjust the whey to 50% 1. Apply to a CPG-3000 column 1. Dialter the whey against six volumes
saturation with NH4 2 SO4 at 4 C (300 nm pore diameter) equilibrated of 0.047 mM phosphate buffer,
2. Incubate overnight (with 0.047 mM phosphate buffer, pH 7.0, using a 100,000-dalton
3. Collect precipitate by centrifuging pH 7.0. exclusion limit membrane.
at 16,500 g for 60 min at 4 C. 2. Membrane vesicles are obtained in 2. Concentrate the retentate containing
Represents crude vesicles. the void volume fraction. the vesicles 10-fold.
1. Recirculate 150 mL solubilized vesicles prepared from 1. The enzyme is biotinylated using 0.25 mL of biotin-
1 L whey through 60 mL of the adsorbent at 30 C HPDPa /mL of solubilized membrane vesicles. Incubate for
for 4 h. 90 min at room temperature.
2. Remove adsorbed materials by washing with 1 L 2. Recirculate the biotinylated enzyme in 47 mM phosphate
phosphate buffer, pH 7, containing the detergent. buffer, pH 7, through monomeric avidin beads for 20 min
3. Reductively release the enzyme by recirculating 100 mL at room temperature. Add 0:5 mg total protein/mL of
of 2 mM GSH until no thiol groups can be detected adsorbent.
with DTNB. 3. Wash the matrix with ve volumes of 47 mM phosphate
4. Regenerate the adsorbent by recirculating 50 mL buffer, pH 7, containing 1% C12 E9 .
DTT in 8 M urea/1 M NaCl, followed by 0.1 M 4. Release the enzyme by recirculating 0.30.5 mL 50 mM
acetic acid. DTT/mL matrix in 47 mM phosphate buffer for 20 min at
room temperature
a
Biotin-HPDP is N-[6-(biotinamido)hexyl]-3 0 (2 0 -pyridyldithio)propionamide.
conditions, yielding sulfhydryl oxidase activity in the prepared by reaction of cysteine with 1-ethyl-3-(3-
pellet. Finally, the pellet is solubilized in twice the dimethylaminopropyl)carbodiimide-activated succina-
volume of the previous solution and again centrifuged midopropyl glass as described by Sliwkowski et al.
under the same conditions, yielding the puried (25). A typical purication using this procedure results
enzyme in the pellet. in a 4800-fold increase in the specic activity compared
The other two methods of purication are based on to that in whey (Table 4).
the catalytic mechanism or the active-site sulfhydryl Bioselective adsorption on monomeric avidin is
group (Table 3). First, the membrane vesicles are solu- based on selective biotinylation of the active site
bilized by adjustment to 1% C12 E9 in phosphate buf- sulfhydryl group of the enzyme with N-[6-(biotin-
fer, pH 7, and centrifuged at 100,000g to remove any amido)hexyl]-3 0 (2 0 -pyridyldithio)propionamide (bio-
insoluble material. Transient covalent chromatogra- tin-HPDP) (35). The resulting mixed disulde,
phy on cysteinylsuccinanidopropyl glass based on the containing a biotinyl residue, strongly binds to avidin
fact that sulfhydryl oxidase follows a substituted but can be released by reduction of the disulde bond
enzyme mechanism. Thus, a mixed disulde is formed (35). The puried enzyme obtained by this procedure
between the enzyme and the immobilized cysteinyl resi- exhibits a 3100-fold increase in specic activity com-
due which cannot be reduced without the addition of pared to that of whey (Table 5).
free thiols (34). Cysteinylsuccinamidopropyl glass is
Xanthine Oxidase
John R. Whitaker
University of California, Davis, Davis, California, U.S.A.
large number of electron acceptors are known, includ- kinetic studies on the rate of reduction of Methylene
ing O2 , NAD , methylene blue, 2,6-dichlorophenol- Blue. Dixon (14) showed that aldehydes are substrates
indophenol, triphenyltetrazolium chloride, phenazine for XO. In 1939, Ball (15) and Corran et al. (16)
methosulfate, cytochrome c, and ferricyanide. obtained nearly pure XO, with a golden brown color
due to intrinsic avin (FAD), plus another unknown
chromophore (now known to be molybdenum) (17,
B. Historical Aspects 18). In 1954, Richert and Westerfeld (19) reported
that addition of ferric chloride in the diet of rats sub-
In 1902, Schardinger (11) determined that fresh milk stantially increased the activity of XO in rat liver; they
decolorized Methylene Blue. In 1922, Morgan et al. found iron, FAD, and molybdenum to be present in
(12) reported that bovine milk converted hypoxanthine the ratio of 8 : 2 : 1 in XO. In 1955, Avis et al. (20)
and xanthine into uric acid under both aerobic and developed an isolation procedure for XO from cows
anaerobic (when Methylene Blue was included) condi- milk which gave crystalline XO with A280 =A450 of 5.0
tions. They found the same activity in liver, spleen, and 5.2, a level of purity that has not been increased to the
lungs of rats and cows. The activity, which they named current time. The ratio of iron, molybdenum, and
xanthine oxidase, was destroyed by boiling. In 1924, FAD was 8.1 : 1.4 : 2.0. Later, the iron, labile sulde,
Dixon and Thurlow (13) partially puried XO and did molybdenum, and FAD were found to be in the ratio
Relative Relative
Substrate rates Substrate rates
Source: Ref. 6.
two iron/sulfur cofactors, the FAD cofactor and the Cleland Nomenclature (35) in Eq. (4) where X is
Mo-pterin cofactor. Each subunit functions indepen- xanthine and U is uric acid.
dently of the other. The reader is referred to the pub-
X U O2 H2 O2
lication by Enroth et al. (9) for a color-coded diagram # " # "
of the XO molecule. E-FAD E-FADX E-FADH2 E-FAD2 O2 E-FAD
E-FADH2 O E-FADH2 O2
The tertiary structure of aldehyde oxidoreductase, a
member of the xanthine oxidase family, is also known 4
for Desulfouibrio gigas (32, 33); it is very similar to that
In 1964, Bray et al.(36) used electron paramagnetic
of XO.
resonance spectroscopy to investigate the rates of elec-
tron transfer to and decay from the several cofactors of
B. Mechanism of Action of Xanthine Oxidase the enzyme-substrate system. The XO and xanthine
were mixed very rapidly (< 1 msec) and frozen very
When xanthine and O2 are added to XO, the reaction rapidly (< 1 msec) in liquid N2 . The Mo- signal,
pathway was shown by Gutfreund and Sturtevant (34)
to be that described by Fig. 2. The rate constants k1 ,
k2 , k3 , and k4 for the reactions were determined by
rapid reaction techniques to be 5:0 105 M1 sec1 ,
10:5 sec1 , > 4:0 105 M1 sec1 , and 21:5 sec1 ,
respectively. The rate constants k1 and k3 are for for-
mation of the Michaelis-Menten complexes of
xanthine E FAD and O2 E-FADH2 , respectively,
while k2 and k4 are the rate constants for oxidation of
xanthine to uric acid and reduction of O2 to H2 O2 ,
respectively.
When the 1/(turnover numbers) are plotted versus
1/[xanthine] at three different concentrations of O2 : air
saturated (0.24 mM), 60% O2 saturated enzyme and
the results of the two are extrapolated to 1O2 concen-
tration, all three lines are parallel (Fig. 3) (21), indicat-
ing that the reaction follows a Ping-Pong Bi-Bi Figure 2 Mechanism of action of xanthine oxidase as deter-
mechanism. This is diagrammed according to the mined by fast reaction kinetics. (From Ref. 34.)
7
While Eqs. (6) and (7) give the same nal products, the
Figure 3 Effect of xanthine and O2 concentrations on the mechanisms of the reactions are fundamentally differ-
rate of xanthine oxidation by xanthine oxidase. T.N. is the ent.
turnover number measured at pH 8.3 and 25 C. (From Ref. In the rst proposed mechanism [Eq. (6)], the pro-
21.) duct of hydroxylation of the substrate 2-hydroxy-6-
methylpurine (B; in the enolate tautomeric form) is
suggested to be coordinated to the molybdenum via
believed to be conversion of Mo(VI) to Mo(V), the induced oxygen atom from the OH group of the
reached a maximum in 15 msec. The Mo-, molybdenum cofactor (A) of the enzyme to form the
FADH FAD ! FADH, and Fe(II) FeIII ! bracketed intermediate/transition state compound (C).
FeII signals were maximum at 40, 45, and 100 The intermediate/transition state compound is pro-
msec, respectively (Fig. 4). Therefore, Bray et al. (36) posed to be formed in a base-assisted (B: group in
suggested that the sequential steps in the reductive the active site of the enzyme) nucleophilic attack of
reaction are: xanthine ! Mo ! FAD ! Fe ! O2 . the MoVI OH group on the C-8 of the substrate with
More recently, the reaction order has been shown to a concomitant hydride transfer to the Mo S group
IV
be: xanthine ! Mo ! 2Fe=2S ! FAD ! O2 (21). (bracketed structure) to give Mo SH (D). The very
Therefore, the overall reaction involving the ow of rapid species is formed by a one-electron oxidation
electrons is best described by Eq. (5). and deprotonation to give the EPR-detectable
MoV OS(OR) species (D).
In the second proposed mechanism [Eq. (7)], the
purine substrate (B0 ) is complexed (noncovalently) to
the molybdenum cofactor (A 0 ) of the enzyme as the
keto tautomer permitting insertion of the C8H-bond
See Section IV.B.1. The cloudy ltrate was stored for 16 h at 5 C, then
reltered through PF 40 and 11 sterimats of grade
GS to give 20 L of clear yellowish brown liquid
IV. PURIFICATION OF XANTHINE (M4).
OXIDASE
4. M4
A. Xanthine Oxidase from Cows Milk
M4 was chromatographed on a 6 in.12 in. Pyrex
An abbreviated description of the method of Avis et al. column packed with calcium phosphate gel/Celite 545
(20) for the purication of xanthine oxidase from cows (1 : 1) (see Ref. 38 for preparation). M4 was cooled to
milk, published in 1955, is given below. This method is 5 C, adjusted to pH 6.2 with 0.1 N acetic acid, and
chosen for several reasons: passed through the column at 4 lb/in.2 pressure. The
Since XO/XDH have two avins per mol, the turnover 6. Titrate the liquid portion from above using 30
number is per active site. The XDH/XO ratio is calcu- mL enzyme preparation/mL calcium phosphate gel
lated on the basis of these specic assays. (38). Typically, 1 L of calcium phosphate gel is stirred
well with 470 mL of enzyme solution. Centrifuge for 10
2. Enzyme Purication min at 4,000g. Discard supernatant.
7. Resuspend the precipitate in phosphate buffer
Purication of XO in the reduced XDH form (8).
(see 5 above) and centrifuge to collect insoluble mate-
1. To 15 L fresh unpasteurized cows milk, add 50 rial.
mL of 0.75 M dithiothreitol (DTT), 15 mL of 0.1 M 8. Resuspend the precipitate in 500 mL of 10%
phenylmethanesulfonyl uoride (to inhibit serine pro- ammonium sulfate. Centrifuge. Repeat the washing
teases), 15 mL of 1.0 M sodium salicylate, 240 g of of the precipitate three more times, saving the washes.
sodium bicarbonate, and 2840 g of ammonium sulfate. 9. To the combined washes, add ammonium sul-
Stir rapidly for 1 h at 4 C (all steps below are done at fate to increase its concentration to 50%. Centrifuge at
4 C). 16,000g for 30 min. Save the precipitate.
2. Add 2.5 L chilled 1-butanol and stir for 30 min. 10. Resuspend the precipitate in a mixture of 80%
3. Centrifuge the suspension in 800-mL aliquots of 0.05 M Tris, 0.3 mM EDTA, pH 7.8, buffer and
for 15 min at 4000g. Remove the top butanol layer 20% of 0.1 M sodium pyrophosphate, 0.3 mM EDTA,
by suction. Remove the solid lipid layer. Discard both. pH 8.5, with 2.5 mM DTT (Buffer A). Dialyze against
4. To the 15 L of liquid, add 160 g/L of ammo- 2 L of the same buffer overnight.
nium sulfate and stir for 30 min. Then let stand for 2 h 11. A 100-mL portion is removed from the dialysate
to permit the precipitate to rise to the top. Skim off the and loaded onto a folate afnity column, prepared
brown precipitate (in 1:7 L); centrifuge for 1 h at according to Nishino et al. (40), and equilibrated in
16,000g. Remove the liquid and discard. Buffer A. The column is washed with 500 mL of
5. Resuspend the solid precipitate in 500 mL of Buffer A until the 280 nm absorbance reaches a mini-
0.05 M potassium phosphate buffer containing 0.3 mum, then elution is begun with Buffer A containing
mM EDTA, 1 mM salicylate, 2.5 mM DTT, pH 6.0. 10 mM salicylate. XDH moves gradually down the
Dialyze twice, using 6 L of the same buffer solution column and elutes as a single dark band.
each time. Centrifuge the dialysate ( 470 mL) at 12. The dark band fractions from step 11 above are
16,000g for 1 h. Decant the liquid portion and save. pooled, concentrated, and dialyzed in 0.1 M sodium
Discard precipitate. pyrophosphate, 0.3 mM EDTA, pH 7.5, buffer. The
XDH/XO
specic XDH/XO Dehydrogenase:
Purication Volume XDH/XO Protein activity purication oxidase activity
step (mL) (total units) (mg/mL) (units/mg) (fold) (ratio)
Source: Ref. 8.
liquid fractions from step 1 are stored at 4 C in the 7. CM Harris, V Massey. The reaction of reduced
buffer above with 1 mM salicylate and 2.5 mM DDT xanthine dehydrogenase with molecular oxygen. J
added. During storage there is some conversion of Biol Chem 272:83708379, 1997.
XDH to XO. XO can be converted back to XDH by 8. J Hunt, V Massey. Purication and properties of milk
xanthine dehydrogenase. J Biol Chem 267:21479
adding fresh DTT-containing buffer.
21485, 1992.
Table 3 summarizes the results of the purication of 9. C Enroth, BT Eger, K Okamoto, T Nishino, T
Nishino, EF Pai. Crystal structures of bovine milk
milk XDH.
xanthine dehydrogenase and xanthine oxidase: struc-
ture-based mechanism of conversion. Proc Natl Acad
REFERENCES Sci USA 97:1072310728, 2000.
10. A Soto, T Nihino, K Noda, Y Amaya, T Nishino. The
1. T Nishino, T Nishino. The conversion from the dehy- structure of chicken liver xanthine dehydrogenase:
drogenase type of rat liver xanthine dehydrogenase by cDNA and the domain structure. J Biol Chem
modication of cysteine residues with uorodinitrobe- 270:28182826, 1995.
zene. J Biol Chem 272:2985929864, 1997. 11. F Schardinger. Ueber das Verhalten der Kuhmilch
2. T Nishino, K Okamoto, S Nakanishi, H Hori, T gegen Methylenblau und seine Verwendung zur
Nishino. The mechanism of conversion of xanthine Unterscheidung von ungekochter und gekochter
dehydrogenase to xanthine oxidase. Keio Univ Symp Milch. Z Unters Nahr Genussm 5:11131121, 1902.
Life Sci Med (Oxygen Homeostasis and Its 12. EJ Morgan, CP Stewart, FG Hopkins. Anaerobic and
Dynamics), 1:333339, 1998. aerobic oxidation of xanthine and hypoxanthine by
3. J Hunt, V Massey. Milk xanthine dehydrogenase. In: tissues and by milk. Proc R Soc Lond Ser B94:109
B Curti, S Ronchi, G Zanetti, eds. Flavins 131, 1922.
Flavoproteins Proc Int Symp, Vol 10. Berlin: de 13. M Dixon, S Thurlow. Studies of xanthine oxidase. I.
Gruyter, 1991, pp 695698. Preparation and properties of active materials.
4. T Nishino, Y Amaya, K Noda. Cysteine residues Biochem J 18:971975, 1924.
responsible for dehydrogenase-oxidase conversion of 14. M Dixon. Studies on xanthine oxidase. VII. The spe-
rat liver xanthine dehydrogenase. In: B Curti, S cicity of the system. Biochem J 20:703718, 1926.
Ronchi, G Zanetti. Flavins Flavoproteins Proc Int 15. EG Ball. Xanthine oxidase: purication and proper-
Symp, Vol. 10. Berlin: de Gruyter, 1991, pp 703706. ties. J Biol Chem 128:5167, 1939.
5. T Nishino, S Nakanishi, K Okamoto, J Mizushima, H 16. HS Corran, JG Dewan, AH Gordon, DE Green.
Hori, T Iwasaki, T Nishino, K Ichimori, H Xanthine oxidase and avoprotein. Biochem J
Nakazawa. Conversion of xanthine dehydrogenase 33:16941706, 1939.
into oxidase and its role in reperfusion injury. 17. EC DeRenzo, E Kaleita, P Heytler, JJ Oleson, BL
Biochem Soc Trans 25:783786, 1997. Hutchings, JH Williams. The nature of the xanthine
6. VH Booth. LXV. The specicity of xanthine oxidase. oxidase factor. J Am Chem Soc 75:753, 1953.
Biochem J 32:494502, 1938; VH Booth. LXVI. The 18. DA Richert, WW Westerfeld. Isolation and identica-
xanthine oxidasealdehyde system. Biochem J 32:503 tion of the xanthine oxidase factor as molybdenum. J
507, 1938. Biol Chem 203:915923, 1953.
Harold W. Gardner
U.S. Department of Agriculture, Peoria, Illinois, U.S.A.
Total Specic
Volume Protein activity activity Fold
Purication step (mL) (mg) (units)a (units/mg) purication
above can be utilized, except the substrate is linoleic Triton X-100 was removed from the supernatant
acid dissolved in a minimum of methanol to aid dis- after centrifugation with Amberlite XDA-2. An initial
persal. This avoids the use of inhibitory surfactants, purication was achieved by a DEAE-Toyopearl col-
but alcohols are also inhibitory at high concentrations umn, and nal separation was accomplished with
(96). Methanol can be kept to a minimum to give neg- Mono Q. One isozyme that did not bind to DEAE-
ligible inhibition (e.g., 25 L substrate dissolved in Toyopearl required a different purication with CM-
methanol injected into 2.5 mL buffered enzyme solu- Toyopearl and Mono S columns.
tion). With the advent of cDNA technology, LOX is
increasingly puried from E. coli; however, it has
B. Assays for Screening Purposes been found necessary to cultivate E. coli at low tem-
peratures in order to obtain active LOX (105, 106).
In instances where seed or other material is being
screened for absence or presence of LOX activity,
there is need for rapid assay methods. A number of REFERENCES
colorimetric tests have been developed that rely on
the detection of either hydroperoxides or free radical 1. C Su, M Sahlin, EH Oliw. A protein radical and ferryl
cooxidation reactions (100103). Researchers have also intermediates are generated by linoleate diol synthase,
developed a continuous-ow amperometric detection a ferric hemeprotein with dioxygenase and hydroper-
of hydroperoxides to assay LOX activity (104). oxide isomerase activities. J Biol Chem 273:20744
20751, 1998.
2. M Hamberg, C Su, E Oliw. Manganese lipoxygenase:
discovery of a bis-allylic hydroperoxide as product
VII. PURIFICATION
and intermediate in a lipoxygenase reaction. J Biol
Chem 273:1308013088, 1998.
The method of Axelrod et al. (84) has been used suc- 3. GJ Piazza, ed. Lipoxygenase and Lipoxygenase
cessfully by our laboratory to isolate large quantities of Pathway Enzymes. Champaign, IL: AOCS Press,
soybean LOX-1. The method employs preparation of a 1996.
crude extract from hexane-defatted soybean our with 4. GA Veldink, MP Hilbers, WF Nieuwenhuizen, JFG
0.2 M Na acetate buffer (pH 4.5), followed by Vliegenthart. Plant lipoxygenase: structure and
NH4 2 SO4 fractionation and then two sequential mechanism. In: AF Rowley, H Kuhn, T Schewe,
chromatographic separations on DEAE-Sephadex eds. Eicosanoids and Related Compounds in Plants
(Table 1). The isolate can be stored as a concentrated and Animals. London: Portland Press, 1998, pp 69
96.
suspension in 2.3 M (NH4 2 SO4 at 4 C (refrigerator)
5. A Grechkin. Recent developments in biochemistry of
for many years. The authors (84) also outlined proce-
the plant lipoxygenase pathway. Prog Lipid Res
dures for the isolation of LOX-2 and LOX-3. 37:317352, 1998.
A more recent method chromatographically sepa- 6. HW Gardner. Analysis of plant lipoxygenase metabo-
rated all six isozymes found in soybean seedlings lites. In: WW Christie, ed. Advances in Lipid
(70). Seedlings were homogenized in 50 mM Na phos- MethodologyFour. Dundee, Scotland: Oily Press,
phate (pH 6.8) containing 1.5% Triton X-100. The 1997, pp 143.
Sorbitol Oxidase
Figure 3 Comparison of the deduced amino acid sequence of SOX with that of the rat L-gulonolactone oxidase. The deduced
amino acid sequence of the SOX gene (upper, I) is compared with that of rat L-gulonolactone oxidase (lower, II). Asterisks and
dots indicate identical and related amino acids, respectively. Closed circles indicate the histidine residue that is possible for
covalent attachment of FAD.
Figure 4 Optimum pH (A) and heat stability (B) of SOX. (A) SOX activity was measured at various pHs for 10 min at 37 C.
Buffers used were 40 mM citrate-Na2 HPO4 (pH 4.07.5, open circles), 40 mM Tris-HCl (pH 7.58.5, closed circles), 40 mM
glycine-NaOH (pH 8.510.0, open triangles). (B) SOX was incubated at various temperatures for 15 min in 40 mM Tris-HCl
buffer, pH 7.5, and the remaining activity was measured at 37 C.
Relative
activity Relative activity
Compound (%) Compound (%)
D-Sorbitol 100 (Km 0.26 mM) D-Fructose 0
D-Xylitol 93.5 (Km 0.38 mM) Saccharose 0.30
D-Mannitol 55.0 (Km 7.38 mM) Maltose 1.30
D-Arabitol 39.0 (Km 7.73 mM) Lactose 0.10
Ribitol 4.10 Methanol 0
meso-Erythritol 3.80 Ethanol 0
D-Threitol 0 1-Butanol 0
Inositol 3.30 1-Propanol 0
Glycerol 3.70 2-Propanol 0
D-Glyceraldehyde 0 1,2-Propanediol 0.80
Dihydroxyacetone 0 1,3-Propanediol 2.70
Dihydroxyacetone phosphate 0 1,2-Butanediol 0.80
D-Arabinose 0 1,3-Butanediol 3.20
D-Glucose 2,3-Butanediol 0
D-Galactose 3.50 1,4-Butanediol 7.40
D-Mannose 1.00 Ethylene glycol 0.20
L-Sorbose 0.70 PVAb 0.10
a
The enzyme assay was carried out at 50 mM concentration of each compound.
b
PVA; Polyvinyl alcohol 2000.
Starch Phosphorylases
Ying Yu
Rutgers University, New Brunswick, New Jersey, U.S.A.
B. Starch Phosphorylase Isoforms tend to exhibit high afnity for large-highly branched
glucans, such as glycogen. Pho2-type starch phos-
Higher-plant starch phosphorylases are classied into phorylases have an apparent monomer size of 90
two groups based on their subcellular localization kDa (13, 14). The second distinct starch phosphory-
and afnity for various glucans (13). Members of lase group, originally classied as the L-type, but
each group can be readily distinguished by the use now known as the Pho1-type group represents starch
of zymograms (Fig. 2). The cytosolic class of starch phosphorylases localized in plastids such as chloro-
phosphorylases has frequently been referred to as H- plasts and amyloplasts, and are generally character-
type phosphorylases, but more recently the term ized by a monomer size of 100 kDa. Pho1-type
Pho2-type starch phosphorylase, based primarily starch phosphorylases prefer oligoglucans such as
upon the migration behavior on zymograms, has maltodextrins.
replaced the earlier nomenclature. In addition to In spinach leaf or pea seed, one form of each starch
their cytosolic localization, Pho2-type phosphorylases phosphorylase type is present. However, the relative
ratio of the two starch phosphorylase types varies dur- C. Starch Phosphorylase Gene Classications:
ing plant development (15, 16). During seed germina- Signicance of the Species Specic Pho 1
tion, the amount of Pho2-type starch phosphorylase is Insertion Sequences
increased while Pho1-type starch phosphorylase
remains at a constant, lower level. The developing cDNA clones representing both Pho1- and Pho2-type
seeds, however, contain much higher amounts of isoforms of starch phosphorylase have been obtained
Pho1-type starch phosphorylase than Pho2-type starch from potato and several other plant species. In potato
phosphorylase. tuber, cDNA clones for one cytosolic (Pho2-type) (22)
In maize, characterization of four starch phosphor- isoform and two plastidial (Pho1-type) isoforms (6, 23,
ylase isoforms has been reported (17, 18). The maize 24) have been characterized. cDNAs for Pho1-type
endosperm is rich in starch phosphorylase activity, starch phosphorylases have also been obtained from
and starch phosphorylase expression patterns corre- sweet potato (25), broad bean (26), and spinach leaf
late closely with starch deposition (19, 20). The four (8). Predicted amino acid sequences are well conserved.
isoforms differ in their pH optima, primer depen- For example, spinach Pho1-type starch phosphorylase
dence, chromatographic properties, development has an overall amino acid sequence identity of 75%
expression patterns, and their localization. Isoforms with potato (6, 8) and sweet potato (27) Pho1-type
I, II, and III occur mainly in the endosperm, while starch phosphorylases. The sequence of spinach
isoform IV is localized in the embryo. The activities Pho1-type starch phosphorylase has 62% identity
of isoforms I, II, and III were each reduced in the with Pho2-type starch phosphorylases from potato
mutant sh4, and their enzymatic properties such as (22) and broad bean (26).
pH optima and sensitivity to activators and inhibitors Starch phosphorylase cDNAs are typied by four
were altered (18). However, the properties of isoform distinct regions (Fig. 3). Region 1, the transit peptide,
IV did not seem to be affected by the sh4 mutation exhibits low homology among species. Regions 2 and
(21). These studies were mainly based upon separa- 4, at the N- and C-termini of the mature protein,
tion of isoforms by ion exchange chromatography. At respectively, are highly conserved. Region 3, distin-
that time, it was not possible to prepare antibodies or guished by a 50- to 80-residue stretch of additional
to clone cDNAs. Therefore, it is not known if these amino acids, is commonly referred to as an insertion
starch phosphorylase isoforms represent distinct gene sequence. Insertion sequences are present only in
products. Pho1-type starch phosphorylases. As with the transit
Amylosucrase
Volker Buttcher
Aventis CropScience GmbH, Frankfurt, Germany
I. INTRODUCTION rhoeae and the N. meningtidis are the only two species
that are pathogenic. All other Neisseria sp. are non-
The enzyme amylosucrase was discovered by Hehre pathogenic and are known to be commensals.
and Hamilton (1). The reaction demonstrated an in
vitro glycogen synthesis without any activated nucleo- B. Molecular Genetics
tide sugars for the rst time. The reaction of amylosu-
crase can be described as follows: The GC content of Neisseria sp. ranges from 49.3 to
Sucrose ! Amylose Fructose 55.6 mol% (3). The size of the chromosome is 1:5
106 bp (4). The gene for amylosucrase was cloned (5, 6)
The amylosucrase splits off the glucosyl part from the and sequenced. Up to now no sequences of other amy-
substrate and links it covalently to an acceptor mole- losucrases have been published, but it is assumed that
cule. During this reaction fructose is released. The the DNA sequences among the Neisseria spp. are very
reaction mechanism is still not proven, but it is similar. The amylosucrase gene of N. polysaccharea
assumed that the glycosyl residue is transferred to the could be strongly expressed in various E. coli strains
nonreducing end of an alpha-1,4-glucan. For the (6). Because of its own secretion-signal-like leader
enzyme reaction no additional energy is needed but a sequence in recombinant E. coli the enzyme is secreted
primer seems to be necessary to trigger the reaction. into the medium, but the bulk of the protein is retained
The best primer is glycogen which shows enormous in the cytosol.
stimulation of the amylosucrase reaction (2).
A. Microbiology
II. PROPERTIES OF THE
AMYLOSUCRASE
Amylosucrase occurs only in prokaryotes and only in
some genera of the genus Neisseria (Table 1). The amy- A. Protein Data
losucrase is located in the cytoplasm of the cells, except
in N. polysaccharea, which secretes the amylosucrase in The puried amylosucrase is a very stable enzyme. It
the medium. remained active for several days in the reaction buffer
The genus Neisseria is the type genus of the family (100 mM Na citrate, pH 6.7) at 37 C. The enzyme
Neisseriaceae which contains 18 species that can be could be stored over 1 year at 20 C without signi-
isolated from humans and animals. The Neisseriaceae cant loss of activity. The MW of the protein is around
are gram-negative bacteria. The morphology in this 72 kDa and the calculated isoelectric point is at pH
family can be diplococci, cocci, or rods. The N. gonor- 5.59.
Dextransucrase
Klaus Buchholz
Braunschweig Technical University, Braunschweig, Germany
Pierre F. Monsan
National Institute for Applied Sciences, Toulouse, France
Source: Refs. 7, 8.
Activation energy
KM (mM) Optimal pH Optimal temperature ( C) (kCal mol1 ) Reference
It must be pointed out that Michaelis kinetics only product formation follow the same scheme with an
apply in the absence of acceptors. Constants like Vmax , even more pronounced tendency; dextran formation
KM , and KI are dependent on acceptor concentrations may be nearly suppressed at high acceptor concentra-
(cf. next paragraph). Thus with good acceptors, appar- tion (Fig. 2) (65).
ent Vmax values increase with increasing acceptor con- The nal product concentrations depend signi-
centration, from 5.8 (U/mL) with substrate only up to cantly on the ratio of substrate and acceptor. The
19.1 (U/mL) at 600 mM maltose concentration (64). rst acceptor product (panose in the presence of mal-
They decrease with increasing concentrations of weak tose) is favored at high maltose excess, whereas tetra-
acceptors like fructose, from 1.3 (U/mL) to 0.38 (U/ and pentasaccharides are formed in comparable
mL) at 2.75 M fructose concentration (54). Apparent amounts, as well as some higher oligosaccharides, at
KM values increase as well with increasing acceptor about equimolar substrate and acceptor concentra-
concentration, from 12 mM at zero to 163 mM at tions.
600 mM maltose (64). A KM value of 27 mM was In the presence of weak acceptors, like fructose,
obtained in the absence of an acceptor whereas this initial overall reaction rates decrease signicantly
value was 40 mM at a fructose concentration of 1.39 with increasing acceptor concentration, whereas
M (54). It is obvious that in the presence of acceptors, they increase with the substrate concentration.
kinetics are complex and Michaelis kinetics do not Initial rates of acceptor product formation increase
apply. with both substrate and acceptor concentration, the
In the presence of strong acceptors, like maltose or dextran formation being suppressed to a major
isomaltose, several general effects may be summar- extent only at very high acceptor concentration
ized: initial overall reaction rates (sucrose consump- (range of 10% at 3 M fructose concentration)
tion or sum of acceptor product and dextran (Fig. 3) (66). The acceptor product of fructose,
formation) signicantly increase both with substrate leucrose, does not act as an acceptor, so high yields
and acceptor concentration. Initial rates of acceptor can be obtained (54).
tion degree up to 6 are easily separated within 30 min ity in the range of 510 U/g wet weight) were used.
elution time. Products are detected and quantied by Samples from the reaction digests were taken after
measuring the refractive index of the eluent solution. 15, 30, 45, 60, and 90 min; diluted; and heated at 100
C for 5 min to inactivate the enzyme. After cooling to
ambient temperature, the samples were membrane-l-
C. Immobilized Dextransucrase
tered through a 0.2-m cellulose nitrate lter and ana-
lyzed by HPLC. [Note that prolonged storage of
For immobilized dextransucrase it is obvious that
frozen samples prior to analysis will give wrong results
systematic errors in the activity measurement can be
due to a slowly continuing reaction (72)]. The enzyme
avoided only when dextran formation is suppressed as
activity is calculated from the fructose concentrations
far as possible. This can be achieved by adding an excess
by linear regression. In case of immobilized dextransu-
of good acceptor like maltose (71). A test in which
crase, corrections must be made for the decrease in
maltose was applied in vefold excess was developed
reaction volume by sampling whereas the amount of
for this purpose (56, 68). However, it must be taken
immobilized enzyme remains constant.
into account that in the presence of excess maltose the
maximal activity of dextransucrase is higher by a factor
of 3 at low concentrations (sucrose 28, maltose 50 D. Analytical Methods: HPLC
mmol/L) and a factor of 2 at high concentrations
(sucrose 280, maltose 150 mmol/L) (65). The test solu- Samples were analyzed by HPLC with the
tion was composed of 730 mM maltose and 146 mM LiChrospher 100 NH2 column (Merck, Darmstadt,
sucrose in a 0.025 M calcium acetate buffer at pH 5.4. Germany). Acetonitrile-water eluent mixtures from
To avoid microbial growth the solution furthermore 80:20 to 70:30 (v/v) were used at 0.8 mL/min. A
contained 200 mg/L sodium azide and a droplet of refractometer (Knauer, Berlin, Germany) served as
toluene. detector, while data were acquired with Gynkosoft
With native dextransucrase, 4.8 mL of this solution (Gynkotek, Munich, Germany). The analytical error
was incubated with 0.2 mL dextransucrase solution is 5% at low sugar concentration (< 8 g/L) and 3
(activity in the range of 515 U/mL) at 25 C. When % at higher concentrations. For complex mixtures of
dextransucrasecalcium alginate beads were tested, 20 oligosaccharides high-performance anion chromato-
mL of the maltose test solution and 1.0 g beads (activ- graphy is the optimal method (73).
Levansucrase
Levansucrase is a kind of transferase which catalyzes a The enzyme catalyzes hydrolysis and polymerization
fructosyl transfer from sucrose to various acceptor reactions concomitantly (Reaction 1), resulting in fruc-
molecules. The product, levan, consists of D-fructofur- tose homopolymer (levan) and free glucose. This reac-
anosyl residues linked predominantly by -(2,6) link- tion occurs when sucrose exists as the sole fructosyl
age as a main chain with some -(2,1) branching donor and acceptor, and involves three steps: initia-
points. Both linkage types are formed by the single tion, propagation, and termination (1). Chains of
enzyme levansucrase. The enzyme catalyzes the follow- levan grow in a stepwise fashion by repeated transfer
ing reactions: of a hexosyl group from the donor to growing acceptor
molecules. The enzyme primarily catalyzes a coupled
1. Polymerization reaction
reaction by a Ping-Pong mechanism, i.e., sucrose
Sucrosen !Glucosen hydrolysis followed by transfructosylation involving a
fructosyl-enzyme intermediate (2).
Levan or Oligosaccharides
When water acts as an acceptor, a free fructose is
2. Hydrolysis reaction generated from both sucrose and levan (Reaction 2).
This reaction occurs in all the levansucrase-catalyzed
Sucrose H2 O ! Fructose Glucose reactions mentioned above, but the rate is much slower
Levann H2 O ! Levann1 Fructose when compared to a sugar acceptor. Reaction 3 occurs
in the presence of an acceptor in the environment. The
3. Acceptor reaction enzyme transfers the fructosyl residue of sucrose spe-
cically to the C-1 OH of aldose in the acceptor.
Sucrose Acceptor Molecules ! Compounds containing hydroxyl groups, such as
Fructosyl-Acceptor Glucose methanol, glycerol, and oligosaccharides, can act as
fructosyl acceptors. The reaction mechanism yields a
4. Exchange reaction nonreducing sugar compound and a series of oligosac-
charides, in which the sugar molecule with one more
Sucrose 14 CGlucose ! fructose moiety remains as a major reaction product.
Fructose-14 CGlucose Glucose The reaction occurs predominantly in the presence of a
high concentration of fructosyl donors, such as sucrose
5. Disproportionation reaction or rafnose. Reaction 4 might be considered analogous
noma in Swiss albino mice, and hypothesized that the fructosyl derivatives, alkyl fructosides, which are
effect might be associated with the polymer of a spe- expected to be applicable in similar or improved uses
cic molecular size. of alkyl glucoside and used as a nonionic surfactant
In addition to the above, a novel aqueous two-phase and as raw materials for the production of sugar esters
system has been developed by mixing polyethylene gly- of fatty acids (17).
col (PEG) and levan, which offers a potential applica-
tion of levan in separation and purication of
biological materials, as an alternative to PEG/dextran IV. GENETIC PROPERTIES OF
and PEG/salt two-phase systems (13). The enzymatic LEVANSUCRASE
or chemical hydrolytic products of levan may be used
in the food industry as sweeteners or dietary ber: The molecular weight of bacterial levansucrases is
ultra-high-fructose syrups (UHFS) and -(2,6)-linked reported to be in the range of 4564 kDa. Several ana-
fructofuranosyl oligosaccharides (7). Several microor- logies in their primary structures have conrmed that a
ganisms producing various levan-assimilating enzymes close relationship exists among levansucrases from
such as exo- or endo-type levanase, fructose dimer- or Gram-negative bacteria. The molecular weight of
heptamer-producing levanase, and levan fructotrans- levansucrases from Gram-negative bacteria is nearly
ferase were isolated and characterized (14). Together the same, about 4547 kDa, while those from Gram-
with the difructose anhydride compounds (DFA I and positive bacteria are slightly larger (5464 kDa).
III) from inulin, which are -(2,1)-linked fructans Eight levansucrase genes from microorganisms have
found in plants, DFA IV from levan is an attractive been cloned and characterized to date (18, 19). The
material as low-calorie sugar, non or antitooth- deduced amino acid sequences are aligned in Figure
decaying sweetener, stabilizer, and antianemic agent 2. All levansucrases share several conserved regions,
in the food and pharmaceutical industries (15). which are thought to be important for the enzyme
Another reason for the increasing interest in levan- activity. Although conservation is observed, dissimilar-
sucrase is that the enzyme can catalyze the transfer of ity exists depending on the source of the enzyme.
fructosyl residue to a variety of sugar molecules. Some Levansucrases from Gram-negative origin show rela-
of the resulting hetero-oligosaccharides have potential tively high similarity (> 50%), when compared with
applications as sweeteners and probiotics (16). the Gram-positive bacterial enzymes. However, very
Currently fructo-oligosaccharides and lactosucrose, little similarity (< 30%) exists among enzymes from
which are formed using a fungal fructosyltransferase, the two sources. In the case of protein secretion
are available in the market as a functional food. mechanisms, they are also thought to be different on
Levansucrase enables the formation of a variety of the basis of the presence or absence of a signal peptide.
A. pH Optimum
E. Kinetic Constants
The enzyme has a high activity at pH 56, and is stable
at pH 47. No activity is observed below pH 3 and The Km values of levansucrase from B. subtilis are
above pH 9. Thus, the enzyme has a very narrow spec- 20 mM for sucrose and 50 mM for the analog glu-
trum of pH with a trapezoid-shaped curve. cosido-sorboside at pH 6.0 and 37 C. The afnity con-
stant for the short-chain levan (DP 40) acting as an
B. Optimum Temperature acceptor-activator is 5 mM under the same conditions
(21).
The range of optimal temperature for levan formation
is rather uniform (1840 C) with the exception of the
psychrophilic enzyme of Z. mobilis (0 C), B. subtilis F. Storage Property
(> 10 C), and the thermostable enzyme R. aquatilis
(50 C). One striking feature is that the synthesis of The enzyme solution is stable at 4 C for several
levan by levansucrase occurs far more effectively at months without loss of activity. The enzyme was rela-
low temperatures than at room temperature or at tively resistant to freezing, thawing, and protease treat-
higher temperature, though the apparent activity deter- ment.
summary of the purity and recovery of the enzyme is the puried enzyme is subjected to SDS-PAGE. The
given in Table 1. As a result of chromatography on enzyme can be puried from culture broth 18.3-fold to
DEAE-Toyopearl 650 M column, two distinct peaks a specic activity of 3:75 U mg1 protein, with a yield
of sucrose-hydrolyzing activity can be seen (Fig. 4). of 16.5%. The molecular weight of the enzyme is
The rst peak is responsible for levansucrase; the sec- 91,000 by Superose 12 gel ltration, and 46,000 by
ond one, for sucrase. After the successive purication SDS-PAGE, indicating that levansucrase is a dimer.
steps of the protein by hydroxyapatite chromatogra- The optimum pH for the enzyme activity is around
phy and second ammonium sulfate purication, the pH 4.0 for sucrose hydrolysis, and is around pH 5.0
small amount of contaminated proteins can be elimi- for levan formation (24).
nated from the enzyme solution by gel ltration chro- A method monitoring the presence of sucrose-split-
matography (FPLC). A single band is observed when ting enzymes has been developed using electrophoresis
and zymogram staining with 2,3,5-triphenyltetrazo-
lium chloride (TTC) reagent prior to purication
(27). Levansucrase activity can be easily distinguished
by the formation of a white and opalescent band. The
use of periodic acid Schiff (PAS) reagent to stain levan
formed by levansucrase on polyacrylamide gel has also
been developed (28). Some enzymes form large aggre-
gates in nature or during purication, causing the loss
of enzymes and difculties in purication. A small
amount of unknown carbohydrates may be responsible
for the formation of aggregates. This problem can be
overcome by the treatment with Triton X-100 (29). In
the case of the Z. mobilis enzyme, such aggregation is
not observed during purication. To eliminate the
copurication of other enzymes with sucrose-splitting
activity, the levan formation activity is monitored
using rafnose as a substrate, which does not normally
serve as a substrate for glucosyltransferase or dextran-
Figure 4 DEAE-anion exchange chromatogram of the sucrase.
extracellular saccharolytic enzymes of Z. mobilis ZM1. Cell
washed solution from Z. mobilis ZM1 is put on the column
(2:5 30 cm) pre-equilibrated with 20 mM phosphate buffer
(pH 6.8), and eluted with NaCl gradient (00.5 M). REFERENCES
Levansucrase activity is monitored by sucrose hydrolysis
activity and the type of enzymes contained in fractions is 1. RG Chambert, G Gonzy-Treboul, R Dedonder.
monitored by electrophoresis and zymogram staining with Kinetic studies of levansucrase of Bacillus subtilis.
TTC reagent. (From Ref. 24.) Eur J Biochem 41:285, 1974.
Cyclodextrin Glycosyltransferase
sources for growth. This involves a cell-associated CGTase enzymes generally display only a relatively
cyclomaltodextrinase (EC 3.2.1.54), yielding glucose, low starch hydrolytic activity (Fig. 2A). They mainly
maltose, and maltotriose (8). Glucose is metabolized catalyze transglucosylation reactions that can be
intracellularly via glycolysis. described as: Gn Gm ! Gn x Gm x,
in which G(n) is the donor and G(m) the acceptor
oligosaccharide consisting of n and m residues, respec-
tively. Disproportionation (Fig. 2B) can be regarded as
the default reaction, and is also catalyzed by several
other members of the -amylase family (e.g., 4--glu-
canotransferase, EC 2.4.1.25). The specic CGTase
catalyzed reaction is the cyclization reaction (Fig.
2C) in which the part of the donor that has been
cleaved off also acts as the acceptor, resulting in for-
mation of a cyclodextrin. This reaction can be
described as: Gn ! cyclic Gx Gn x. The
reverse reaction, coupling, is also catalyzed by the
enzyme (Fig. 2D).
Figure 3 Structure and properties of cyclodextrins. (a) -, -, and -cyclodextrins; (b) 3D form and properties of cyclodextrins;
a: 14.6, 15.4, and 17.5 A for -, -, and -cyclodextrins, respectively; b: 4.75.3, 6.06.5, 7.58.3 A for -, -, and -cyclodextrins,
respectively; (c) formation of inclusion complexes of cyclodextrins and hydrophobic molecules. (From Ref. 71.)
-amylases (33) and CGTases (31, 3437) are quite domain; in glucoamylases (family 15 of glycoside
similar. -Amylases generally consist of three struc- hydrolases) it is attached to the C- or N-terminus of
tural domains, A, B, and C, while CGTases show a the catalytic domain via a glycosylated linker (Fig. 5).
similar domain organization with two additional The E-domain consists of 110 amino acids and is
domains, D and E (Fig. 5). Domain A contains a responsible for the adsorption of the enzyme onto
highly symmetrical fold of eight parallel -strands granular starch (see below).
arranged in a barrel encircled by eight -helices. This All CGTases studied not only catalyze the
so-called (=8 -or TIM-barrel catalytic domain of three transglycosylation reactions but also display
300400 residues is present in all enzymes of the - hydrolytic activity (Fig. 2). Incubation of CGTase
amylase family (38). The =8 barrel was rst with starch thus may not only yield cyclodextrins
found in the structure of chicken muscle triose phos- but also linear products. In some cases this has
phate isomerase (39), but it has been shown to be wide- resulted in misidentication of CGTase enzymes.
spread in functionally diverse enzymes (40). Several The Thermoanaerobacterium thermosulfurigenes EM1
prolines and glycines anking loops connecting the - CGTase was initially identied as an -amylase
strands and -helices have been found to be highly because of its relatively high hydrolytic activity (44).
conserved in these enzymes (41). The catalytic and sub- Further studies showed that it possesses a clear cycli-
strate-binding residues conserved in the -amylase zation activity, as in other CGTases; amino acid align-
family are located in loops at the C-terminal of - ments with other CGTases also show a high overall
strands in domain A. The loop between -strand 3 sequence similarity, with all ve domains present
and -helix 3 of the catalytic domain is rather large (25). The CGTase converts starch not only into cyclo-
and is regarded as a separate structural domain. This dextrins (35%) but also into linear sugars (11%) (45).
B-domain consists of 44133 amino acid residues and In contrast, the CGTase from B. circulans strain 251
contributes to substrate binding. The C-domain is has minor hydrolytic activity; this enzyme exclusively
100 amino acids long and has an antiparallel -sand- forms cyclodextrins (46).
wich fold. Domain C of the CGTase from B. circulans Recently the 3D structure of the Bacillus stearother-
strain 251 contains one of the maltose-binding sites mophilus maltogenic -amylase Novamyl has been
(MBS) observed in the structure derived from mal- determined (47). Unlike the conventional -amylases
tose-dependent crystals (see below) (31). This MBS is of family 13, Novamyl exhibits the ve-domain struc-
involved in raw starch binding (42); the C-domain thus ture associated with CGTase enzymes (Fig. 5, G2A).
contributes in substrate binding. This C-domain may Its overall sequence similarity with CGTase is high,
determine bond specicity, since in enzymes hydrolyz- and the 3D structures are very similar. Most interest-
ing or forming 1,6) bonds (e.g., pullulanase, isoamy- ingly, Novamyl contains a 5 amino acid residue inser-
lase, branching enzyme) the A-domain is followed by a tion in the B-domain, partially enclosing the active site
different domain (Fig. 5) (43). The D-domain is 90 at the 3 subsite. The absence of this insertion in
amino acids in size and is almost exclusively found in CGTases allows for additional subsites, with up to 7
CGTases; it carries an immunoglobulin fold but its characterized crystallographically (32). Novamyl thus
function remains unknown. The E-domain is more is either an -amylase or a CGTase with unusual
widespread in starch degrading enzymes. In enzymes properties. Studies of evolutionary relationships within
of the -amylase family it is present as the C-terminal the -amylase family have provided evidence that
Limonoid Glucosyltransferase
Shin Hasegawa
U.S. Department of Agriculture, Albany, California, U.S.A.
Masayuki Kita
National Institute of Fruit Tree Science, Tsukuba, Ibaraki, Japan
Mitsuo Omura
National Institute of Fruit Tree Science, Shimizu, Shizuoka, Japan
Transglutaminase
Transglutaminase (TGase, EC 2.3.2.13, -glutamyl- Classication and nomenclature of TGases have been
peptide, amine--glutamyl-transferase) is a class of traditionally based on the function of the TGase (blood
enzymes that catalyze the acyl transfer between the coagulation Factor XIII), on cell, tissue, or organ con-
-carboxyamine group of a peptide-bound glutamine taining the TGase (keratinocyte TGase, liver TGase, or
residue and the primary amino group of various sub- prostate gland TGase), or subcellular distribution of
strates in a calcium-dependent reaction. the TGase (extracellular matrix TGase, cytosolic
TGase was rst discovered in guinea pig liver and TGase, or membrane-bound TGase). Recently, some
characterized by Waelsch and coworkers (1, 2), and TGases have been frequently called by abbreviations
identied as an enzyme catalyzing incorporation of such as TGase K (18) or C (19), and TGase 1 (3), 2
polyamines into glutamine residues of proteins or (34) or 3 (19, 20). However, these recent names are
peptides. Since then, the reaction of blood-clotting somewhat inadequate to classify the many TGases
Factor XIIIcatalyzing polymerization of brin has from various sources including bacteria and plants.
been found to be identical to that of the liver Therefore, the traditional names will be used here.
TGase. The reaction is also involved in keratiniza-
tion of epidermis (3), hair formation (4), or copula-
B. Reaction Catalyzed by TGase
tory plug formation (5). More recently, TGase has
been implicated as a widely distributed enzyme that
TGase catalyzes the following reactions.
participates in various biological phenomena such as
apoptosis (6), cell growth and differentiation (7), 1. Formation of
-(-glutamyl) lysine isopeptide
embryonic cell differentiation (8), retinoic acid crosslinks between proteins or peptides, resulting in
induced abnormalities of sh embryo during the polymerization (Fig. 1).
early development (9), and fertilization envelope for- 2. Incorporation of amines into proteins or pep-
mation (1015, 81). tides. This reaction plays a role in deamination or
There are some studies on TGase from an aspect of amine xation. (Fig. 2).
food science (16, 17). The TGase reaction must be 3. Formation of N,N-bis-(-glutamyl) derivative
closely related to the visco-elasticity of food materials crosslinks (Glu-CONH-R-NHCO-Glu) between pro-
such as meat and sh eggs. teins or peptides. (Fig. 3) (21).
Molecular
Source (cell, tissue, weight
organ, etc.) Organisms (daltons) Natural substrate Remarks References
(Continued)
Molecular
Source (cell, tissue, weight
organ, etc.) Organisms (daltons) Natural substrate Remarks References
The molecular weights of TGases were calculated from the amino acid sequences predicted from cDNAs or genomic DNAs. The molecular sizes
(kDa) were those of the TGases isolated by ordinary biochemical methods. The complete sequences of them have not been determined. The
asterisk in the column of References indicates the literature where only natural substrates for TGase are described.
TGase (29), erythrocyte band 4.2 [pallidin; human 60 amino acids longer than those of other vertebrate
(30); mouse (31)], chicken erythrocyte tissue TGase TGases, respectively. It may be predicted that they
(32), guinea pig liver TGase (33), mouse macrophage have a large insertion at the N-terminal regions, or at
TGase (34), bovine aorticendothelial cell TGase (35), the C-terminal regions. In addition, the multiple align-
chicken limb TGase (36), mesenchyme TGase of Ciona ment reveals that the sequence of a novel human
intestinalis embryo (37), Limulus hemocyte TGase (38), TGase X or 5 is especially characteristic in a large
and TGase of grasshopper embryo [annulin (8)]. The deletion of about 90 amino acids at the N-terminal
amino acid sequences of liver TGases of shes [red sea region and an insertion of about 50 amino acids near
bream (39); a salmon, Oncorhynchus keta (40)] are the central region.
available. We tentatively analyzed molecular phylogenetic
As shown in Figure 4, amino acid sequences of var- relationships of various TGases of vertebrates. The
ious TGases were aligned by the program CLUSTAL human TGase X or 5 was removed from the analysis.
W ver 1.7 (41) in DNA Data Base of Japan (DDJB). In addition, the amino acids corresponding to gaps
The overall sequences of keratinocyte TGases, human were removed from the analysis. A tree was con-
blood clotting Factor XIIIA, and mesenchyme TGase structed by the neighbor-joining method (42) using
of Ciona intestinalis embryo were 130150, 50, and the program MEGA (43). The tree was rooted with
Inhibitors or cofactors were searched for mainly by ATP or GTP has been well elucidated for various tis-
an assay method utilizing amine incorporation. Metal sue TGases [guinea pig liver TGase (5); erythrocyte
ion chelating reagents such as EDTA and EGTA inhi- TGase (59); Factor XIII (60); rat brain TGase (52)].
bit the activity of various TGases, which can be Moreover, it has been demonstrated that a human
restored by the addition of Ca2 . In addition, iodoa- tissue TGase has nucleotide hydrolysis activity similar
cetic acid, iodoacetamide, and p-chloromercuribenzo- to ATPase or GTPase (61). This raises the possibility
ate inhibit TGase activity, while 2-mercaptoethanol that the enzyme regulates cell receptor signaling (62).
and dithiothreitol (DTT) stabilize the activity. Studies show that the GTP and ATP hydrolysis
Therefore, TGase is well known as a Ca2 -dependent sites are localized within the core domain of the
SH enzyme. tissue TGase (63). Recently, a 77-kDa GTP-binding
On the other hand, some synthetic compounds such protein, Gh5, was isolated from pig heart membranes
as 2-[3-(diallylamino)-propionyl]benzothiophene (55), with a binding afnity of nucleotides in order:
and 1,3,4,5-tetramethyl-2-[(2-oxopropyl)thio]imidazo- GTP > GDP > ITP ATP, which is similar to that
nium chloride or L-68277 (56) are well-known non- observed for other G-proteins involved in receptor sig-
competitive inhibitors of TGase. naling. The Gh5 also exhibits TGase activity (64).
Tissue TGase is inhibited by nucleotides. ATP inhi- The optimum pH of the TGase activity was highly
bits the activity of rat liver TGase in a concentration- variable as follows: 7.5 for 101-kDa Physarum polyce-
dependent way. Complete inhibition was obtained with phalum TGase (65); 8.0 for Caenorhabditis elegans
3 mM ATP. ADP inhibited the TGase activity simi- TGase (66); 9.09.5 for red sea bream TGase (67),
larly to ATP, but AMP had much less inhibition. and 56 for egg envelope TGases of rainbow trout
There was no signicant inhibition by adenosine and (1315).
adenine. CTP possessed the same inhibitory activity as For guinea pig liver TGase, the optimal incorpora-
that of ATP, while GTP and UTP had 50% of the tion of amines into proteins occurs at pH 7.28.5 (2).
ATP-induced inhibition (57). Such inhibitory effect of However, the pH of optimal ammonia liberation (see
Fig. 1) varies with the substrate: 7.07.5 for unmodi- A. TGase, Reinoic Acid, and Induction of
ed insulin; 6.57.0 for acetylated insulin (68). On the Apoptosis
other hand, there is a report for the same enzyme that
the optimal incorporation of hydroxylamine into N- Clarication of the compounds involved in natural cell
carbobenzoxy (CBZ)-L-glutaminylglycine occurs near death including apoptosis (programmed cell death) has
pH 6.0 (69). become one of the most interesting subjects in biochem-
istry and molecular biology. One of the effective com-
pounds in the death pathway is considered to be tissue
TGase, because of a parallel induction of apoptosis and
IV. BIOLOGICAL FUNCTION OF TGase activity [see the review of Fesus and coworkers
TRANSGLUTAMINASE (6)]. Tissue TGase-mediated crosslinking of some intra-
cellular proteins occurs in cells undergoing apoptotic or
As described in the Introduction, TGase participates in programmed cell death (70, 71). Such protein crosslink-
various biological phenomena such as blood clotting, ing may stabilize the apoptotic bodies and prevent the
keratinization, hair formation, and copulatory plug leakage of cytosolic proteins into the extracellular space
formation. In the present article, we focus our atten- during fragmentation of cell (72). The induction and
tion on the following subjects to briey review the activation of TGase-mediated apoptosis are also sup-
recent development of TGase studies; apoptotic cell ported by a recent study that TGase is involved in pro-
death, and egg envelope hardening in sh. grammed cell death of C. elegans (66).
(Continued)
a possibility that the TGase-mediated crosslink forma- Sports, and Culture of Japan to I.I. Thanks are
tion of nuclear proteins is closely related to some also due Irikawa Trout Hatchery, Tokyo, and Fuji
important cellular events. Trout Farm, Shizuoka, for supplying rainbow trout
2. Antibodies specic to several TGases are com- eggs.
mercially available: antitransglutaminase II (Mono)
(Quartett GmbH, Berlin, Germany); and antitransglu-
taminase (type II) (Upstate Biotechnology Inc., REFERENCES
Waltham, MA., US). These antibodies may be cross-
reactive to various TGases. For example, a polyclonal 1. NK Sarkar, DD Clarke, H Waelsch. Biochim Biophys
antibody against guinea pig liver TGase was crossreac- Acta 25:451452, 1957.
tive to human keratinocyte TGase (104), and a mono- 2. H Waelsch, MJ Mycek. Methods Enzymol V:833843,
clonal antibody against guinea pig liver TGase 1962.
recognized a TGase present in mouse dermal broblast 3. SM Thatcher, RH Rice. Cell 40:685695, 1985.
(105). 4. N Martinet, HC Kim, JE Girard, TP Nigra, DH
3. Recently, a distinct enzyme cleaving
-(-gluta- Strong. SI Chung, JE Folk. J Biol Chem 263:4236
myl) lysine isopeptide crosslinks between proteins or 4241, 1988.
5. J Seitz, C Keppler, S Huntemann, U Raush, G.
peptides was found and termed isopeptidase (106). In
Aumuller. Biochim Biophys Acta 1078:139146, 1991.
the medicinal leech, Hirudo medicinalis, this same
6. L Fesus, A Madi, Z Balajthy, Z Nemes, Z Szondy.
enzyme is called destabilase (107). Experientia 52:942-949, 1996.
7. PJ Davies, JP Basilion, EA Chiocca, J Johnson, S
Poddar, JP Stein. Am J Med Sci 296:164170, 1988.
ACKNOWLEDGMENTS 8. MA Singer, M Hortsch, CS Goodman, D Bentley.
Dev Biol 154:143159, 1992.
9. T Sato. Masters thesis, Sophia University, 1999.
The authors wish to thank Dr. D. Wong U.S. 10. DE Battaglia, BM Shapiro. Cell Biol 107:24472454,
Department of Agriculture, Agricultural Research 1988.
Service, Western Regional Research Center, for cri- 11. I Iuchi, C-R Ha, H Sugiyama, K Nomura. Dev
tical reading of the present article. Our studies cited Growth Differ 38:299306, 1996.
in the present article were supported in part by 12. C-R Ha, I Iuchi. Comp Biochem Physiol 118B:293
Grant-in-Aid from Ministry of Education, Science, 301, 1997.
Feruloyl Esterases
Craig B. Faulds
Institute of Food Research, Colney, Norwich, England
Gary Williamson
Nestle Research Centre, Lausanne, Switzerland
The commercial production of vanillin from agro- Other enzymes that have been puried are listed in
industrial wastes such as cereal brans and sugar beet Table 1. FAEA from A. niger and from A. tubingesis
pulp requires treatment of the starting materials with show 92% homology in amino acid sequence (12), but
enzyme mixtures containing xylanase and feruloyl there is very little similarity between the feruloyl
esterase. The released ferulic acid is bioconverted by esterases sequenced to date. The primary protein
microbes (Aspergillus niger, Pycnoporus cinnabarinus, sequence of FAEA shows some homology to lipases.
Pseudomonas spp.) into vanillin (6, 7). Other possible Based on this homology, FAEA possesses conserved
uses are in baking or any food process involving plant residues at the active site: the nucleophilic serine (resi-
cell walls. due 133), the catalytic acid (Asp174), and the catalytic
base (His247). Based on modeling studies using the
lipase template and on chemical modication using
III. PROPERTIES OF THE PROTEIN dithiobisnitrobenzoic acid treatment of denatured
enzyme, FAEA has three intramolecular disulde
In recent years, the number of microbial cinnamoyl bonds and one free cysteine. CinnAE has not yet
esterases identied has reached > 30, with 10 genes been cloned, but BLAST searches of short protein
sequenced. Some esterases are part of a modular com- sequences show no homology with any other protein.
plex (89), contain cellulose-binding modules (10, 11), There is limited glycosylation of FAEA, while
or exist as only a catalytic domain (12). No structural CinnAE is extensively glycosylated, possibly up to
data have been published, although recently a crystal 40% (ww). A. tubingensis FAEA is more susceptible
of the cinnamoyl esterase domain of XynZ from the to proteolysis than A. niger FAEA, and this can give
anaerobic bacteria Clostridium thermocellum was rise to multiple bands on Western blots in culture
obtained (13). Two feruloyl esterases will be considered supernatants. This susceptibility may be due to residue
here: feruloyl esterase A (FAE-III as puried from differences at locations on the protein surface (16).
culture supernatants of A. niger CBS120.49 or recom- Although no more than one gene encoding feruloyl
binant FAEA; Mr 29,700) (14), and CinnAE esterases have been sequenced from a single source,
(Mr 150,000 [dimer]), also from A. niger (15). there is some evidence of enzymatic isoforms existing.
Figure 3 The most abundant diferulic acids found in plant cell walls: 8-0-4 0 , 8-5 0 benzofuran and 5-5 0 diferulic acids linked to
sugars.
11
07
a
Comparison, the rate of deesteriation with NaOH is also shown.
Source: Ref. 18.
Table 3 Inuence of Length of the Aliphatic Chain on Activity of FAEA and CinnAE on Methyl Esters
8 0.79 56
4.4
a
For comparison, the rate of deesteriation with NaOH is also shown.
Source: Ref. 18.
uloyl esterases has been proposed: type A (e.g., A. niger Like many esterases, feruloyl esterases catalyze the
FAEA) and type B (e.g., A. niger CinnAE, P. funicu- reverse reactioni.e., synthesis of an ester bond (21).
losum FAEB), although there are exceptions (e.g., P. Although this reaction has only been reported with
uorescens XYLD), and not all enzymes displaying pentanol, the potential is there to make feruloylated
feruloyl esterase activity have been tested for activity esters with novel biological activities and applications.
using all the substrates used for FAEA and CinnAE.
There is clearly a dependence of the rate of reaction on
the number and nature of sugars in the substrate
V. MEASUREMENT OF FERULOYL
(Tables 6, 7) (19). However, it is not known if this is
ESTERASE ACTIVITY
due to the presence of subsites on the esterases. The pH
optimum of FAEA is 5.0, and that of CinnAE is 6.0.
To measure feruloyl esterase activity, a range of low-
The stability of FAEA to thermal and chemical
molecular-weight synthetic methyl esters can be used
denaturation has been studied. Thermodynamic para-
as substrates, as described above (Fig. 6). Suitable nat-
meters are shown in Table 8. FAEA is most stable at
ural substrates, which are feruloylated, are wheat ara-
pH 56, where it retains 50% of its catalytic activity
binoxylan or sugar beet pulp (Figs. 1, 2). The rate of
after 1 h incubation at 60 C (20).
8 0.8 55
2 9.9 61
1.4 61 33
1.6 880 45
a
For comparison, the rate of deesteriation with NaOH is also shown
Source: Ref. 18.
Table 5 Inuence of Position of a Single Methoxy Group on Activity of FAEA and CinnAE on Methyl
Esters
2 9.9 61
22 0.31 99
41 0.85 65
a
For comparison, the rate of deesteriation with NaOH is also shown.
Source: Ref. 18.
reaction is increased in the presence of a xylanase or total assay volume is 0.6 mL in disposable semimicro
arabinofuranosidase/arabinanase respectively. cuvettes with a 1-cm path length. For higher substrate
Table 9 shows the activity of native FAEA and concentrations, 0.3 mL in 0.1-cm path length cuvettes
recombinant FAEA on methyl ferulate at pH 6.0, 37 can be used. Dissolve methyl ferulate in methanol and
8C (12). The assay for feruloyl esterases can be per- add a small amount to 100 mM MOPS, pH 6.0.
formed in several ways, but the most convenient is to Determine the A335 of a solution of 100 M
use methyl ferulate as a substrate and monitor the (A335 1:4; E335 for MFA is 1400 M1 cm1 ). Add
reaction in a spectrophotometer (19). The limitation 0100 L of the enzyme solution to appropriate
of this method is that the absorbance of methyl fer- volumes of assay mixture so that nal volume is 0.6
ulate makes it difcult to use the substrate at concen- mL (1-cm path length) or 0.3 mL (0.1 cm) and incu-
trations about the Km . As a standard method, the bate at 378C. Calculate the rate using E of
8 0.79 55
6 0.014 54
2 0.22 85
a
For comparison, the rate of deesteriation with NaOH is also shown.
Source: Ref. 18.
Table 7 Inuence of Sugars on Activity of FAEA and Table 8 Inuence of Sugars on Activity of FAEA and
CinnAE on Feruloylated Oligosaccharides; All Ferulic CinnAE on Feruloylated Oligosaccharides; All Ferulic
AcidSugar Linkages are 1 ! 5 (substrates derived from AcidSugar Linkages are 1 ! 2 (substrates derived from
limited hydrolysis of wheat or maize bran) limited hydrolysis of sugarbeet pulp)
1728 3.5 0 14
2448 30 0 204
7600 34
0 109
1086 17
3270 nd
The most accurate and time-consuming assay for
the measurement of the release of ferulic acid is by
HPLC. This is the only assay that can be used when
measuring the release of phenolic acids (ferulic acid
and dimers) from insoluble plant cell wall-derived
material (26). Samples of plant material (e.g., 10
mg) and enzyme preparation diluted in buffer (in a
9300 M1 cm1 at 335 nm (19). Other assays reported total volume of 0.5 mL) are incubated at a specic
include, using a microtiter plate containing AraF (22), temperature for selected times. It is important to
ethyl cinnamate in agar with pH sensitive dye (23), mix the reaction tube during this step to ensure
separation of substrate and product by capillary complete interaction between the enzyme and the
zone electrophoresis (24) or MUTMAC (4-methylum- substrate. Reactions are stopped by the addition of
belliferoyl [p-trimethyl ammonium cinnamate chlor- glacial acetic acid (0.2 mL) and centrifuged, and the
ide]) with detection by uorescence (25). supernatant is ltered through a 0.22-m lter. The
G (H2 O) (kJ/mol) Slope (m) GuHCl1=2 (M) kunfolding (sec1 ) kunfolding sec1 )
with GuHCl as (kJ/mol.M) with (30 C, pH 6.0) (60 C, pH 6.0) (60 C, pH 6.0)
denaturant GuHCl as with 0.1 mM
(30 C, pH 6.0) denaturant sinapic acid
30 C, pH 6.0)
The rate of irreversible thermal denaturation is given by the rst-order rate constant, kunfolding , and the stability is
slightly increased in the presence of sinapic acid, which is a product of the hydrolysis of methyl sinapate.
Source: Ref. 20.
Lipase
Dominic W. S. Wong
U.S. Department of Agriculture, Albany, California, U.S.A.
active site, and does not show interfacial activation water interface. The inhibitory action of bile salts is
(30). A lipase isolated from guinea pig was found to observed on both microbial and pancreatic lipases,
carry a deletion in the lid sequence, and shows no but the activity of the pancreatic enzyme can only be
interfacial activation (31). In addition to providing restored by the addition of pancreatic colipase, indicat-
accessibility of the active site for the substrate, the ing that specic interactions occur between the pan-
opening of the lid causes a concomitant conforma- creatic lipase and colipase (35).
tional change for proper positioning of the residues Porcine pancreatic colipase is a single polypeptide
forming the oxygen hole during catalysis (22, 25). chain containing 96 amino acid residues with ve dis-
ulde bridges, and has an isoelectric point of 5.0 (36).
Its primary structure has been determined by protein
V. COLIPASE sequencing (37, 38). The protein is synthesized as a
procolipase of 101 amino acid residues, and the active
Adsorption of protein molecules at the lipid-water form, colipase96 , is produced by proteolytic (tryptic)
interface often leads to unfolding and denaturation. removal of the N-terminal pentapeptide (Val-Pro-
For pancreatic lipases, adsorption at the interface Asp-Pro-Arg) (39, 40).
results in decreased activity and consequently inactiva- Colipase consists of a small N-terminal region and a
tion. In the physiological environment, the activity of three-ngers region held together by disulde bridges
pancreatic lipases is regulated by the presence of bile (41, 42). Colipase exhibits an overall amphipathic char-
salts (cholic and chenodeoxycholic acids conjugated acter with most hydrophilic residues found in the N-
with glycine or taurine, such as taurodeoxycholate, terminal region, whereas most of the hydrophobic resi-
glycodeoxycholate, or taurocholate) and other amphi- dues are located at the tip of the ngers in the opposite
philic compounds, such as phospholipids and proteins. side that constitute the lipid-binding site (4346).
The presence of bile salts creates a negatively charged Binding of colipase to lipase occurs between two
lm on the triacylglycerol micelle that prevents lipase hairpin loops (residues 4446, 6567) in the N-terminal
adsorption (32, 33). The cooperative formation of an region of colipase, and the -sheet of the C-terminal
enzymebile salt complex with a dissociation constant domain of the lipase molecule (41, 46). The interac-
of 1:4 1015 M prevents the adsorption to the inter- tions are mostly hydrophilic with the formation of
face (34). For a pancreatic lipase to act at the interface, two salt bridges and six hydrogen bonds. Porcine pan-
the enzyme requires the binding of colipase, a 10-kDa creatic lipase binds to colipase to form a 1:1 complex
protein also secreted by the exocrine pancreas, to facil- with a Kd of 5 107 M in the absence of an interface
itate adsorption of the enzyme at bile salt-coated lipid- (34, 47, 48). In the presence of 2 mM sodium tauro-
chain into a position for its main chain nitrogen to and the desorption of the bound enzyme at the
stabilize the negatively charged oxyanion (Ser153-O) interface occur after each catalytic turnover cycle.
developed in the tetrahedral intermediate (8). In the The Pseudomonas glumae lipase exhibits a kinetic
Rhizomucor miehei lipase, the movement of the lid in behavior which is explained by the hopping model
the active conformation results in the positioning of (74). Lipoprotein lipase shows a similar mode of action
the hydroxyl and amide bond nitrogen of Ser82 for (75).
its interactions with and stabilization of the ester car-
bonyl oxygen of the substrate complex (22). In the
Candida rugosa lipase, the open state involves a change VII. ACYL TRANSFER REACTIONS
in the conformation of the lid, leading to a movement
of the NH of Ala132 into a proper position for the In the reaction mechanism described above for lipase-
oxyanion hole (25). catalyzed hydrolysis of lipids, deacylation involves
The kinetic behavior of lipases at the lipid-water nucleophilic attack of the acylenzyme by H2 O. This
interface is far more complicated than what is step can also be accomplished by other nucleophiles,
described above for the hydrolysis of dissolved sub- such as alcohols and amines, resulting in an acyl trans-
strates. The lipase in the aqueous phase must rst fer instead of hydrolysis. In an aqueous environment,
bind to the interface (E ! E ) before it can bind to hydrolysis is the favorable route, whereas in organic
the substrate. Physically, it involves movement of the media, the acylenzyme is exposed to nucleophiles with-
lid, and the enzyme assumes an open conformation. In out competition from H2 O, and acyl transfer prevails.
addition, the lipase regenerated in the deacylation step It has been shown that Rhizomucor miehei lipase
can either diffuse from the interface or stay bound retains high rates in esterication and transesterica-
(Fig. 7). The pathway through which E is recycled tion at water activity < 0:0001 (76).
can be described by two different models (72, 73). In The acyl transfer reaction provides a novel use of
the scooting model, E stays in the interface between lipases as catalysts in organic synthesis. Using amines
catalytic turnovers. In the extreme situation, it remains as nucleophiles, esters can be converted to the corre-
bound to the interface until all substrate has been sponding amides in organic media at ambient tempera-
depleted. In the hopping model, only a small ture. For example, Candida antarctica lipase-catalyzed
amount of the substrate will be hydrolyzed in each aminolysis of ethyl octanoate gives a 95% yield of
binding event. In the extreme situation, the binding octanamide (77). Peptide bonds can be formed
Enzyme U 1 mmol of butyric acid released per min. Volume, activity, and protein are expressed as average SD from three preparations
subsequently combined for DEAE chromatography.
Source: Ref. 119.
Chlorophyllase
Roger F. McFeeters
U.S. Department of Agriculture and North Carolina State University, Raleigh, North Carolina, U.S.A.
Mg2 7, 8 Position
Compound () reduced () R R1 R2 R3 R4
Figure 1 Ring structure for compounds that are substrates IV. PROPERTIES
or inhibitors of chlorophyllase
The molecular weight for chlorophyllase has been
reported to range from 27 to 65 kDa. Chlorophyllase
(6) and Shioi et al. (7). The reaction is stopped, and the from citrus fruit peel was found to be 27 kDa in one
product is separated from the remaining substrate by study (13) and 35 kDa in another (11). Phaeodactylum
partitioning between an organic phase and an aqueous was reported to have two enzymes that were 43 and 46
acetone phase after KOH solution is added to raise the kDa in size (14). Isozymes from Chenopodium album
pH. For measurement of pheophytin or chlorophyll were 41.3 and 40.2 kDa (12). The largest size chloro-
hydrolysis, McFeeters et al. (4) added 60% hex- phyllase reported to date was 65 kDa in Chlorella reg-
ane:40% acetone and KOH solution to raise the reac- ularis (15). Most frequently, however, molecular size
tion mixture pH to 8.5. The nal proportion of has been found to be near 38 kDa. This includes the
solvents was 50% hexane, 33.3% acetone, and 16.7% chlorophyllase from Chlorella protothecoides (7), tea
water. After separation of the two phases, substrate leaf sprouts (10), sugar beet leaves (16), and rye seed-
depletion or, preferably, product formation is mea- lings (17).
sured spectrophotometrically. To measure hydrolysis Most often, chlorophyllases have been found to
of methyl and ethyl chlorophyllides or pheophorbides, have pH optima near pH 7. However, McFeeters et
McFeeters (8) used a mixture of hexane, acetone, and al. (4) and Ogura (18) found chlorophyllases from
2-butanone to separate reaction products from the Ailanthus altissima and tea, respectively, had acidic
substrates. optima. McFeeters et al. (4) showed a pH optimum
near 7 could be obtained erroneously if the pH of the
aqueous phase in the usual organic solvent/aqueous
III. PURIFICATION partitioning mixtures was not raised by the addition
of KOH. Therefore, earlier reports of pH optima for
Chlorophyllase was rst puried to homogeneity by chlorophyllases may be in error. This work indicated
Moll and Stegwee (9). Subsequently, Shioi et al. (7) the absence of ionizable groups in the enzyme that
developed a somewhat simplied method. A homoge- affected substrate binding, but there did appear to be
neous preparation of the enzyme was prepared from a pKa 3.4 group in the active site of the enzyme
Chlorella protothecoides with over a 50% yield using a involved in substrate hydrolysis. Use of acetone in
butanol extraction, ammonium sulfate precipitation, the reaction mixture when determining the optimum
and chromatography on size exclusion columns pH of chlorophyllase could raise the apparent pKa of
(Table 2). The same procedure, except that the butanol this group and give a higher pH optimum in the pre-
solubilization step was unnecessary, resulted in puri- sence of acetone than in the absence of this solvent.
cation of the enzyme from tea leaf sprouts but with The effect of different substituents of the chloro-
only a 9.6% yield (10). Trebitsch et al. (11) puried phyll a molecule on the ability of chlorophyllase to
chlorophyllase from the avedo of mature green bind and hydrolyze potential substrates has been inves-
oranges, determined the N-terminal sequence of the tigated. The picture that emerges is that of a rather
puried enzyme, and used it for generation of antibo- high specicity esterase. First, chlorophyllase does
dies to chlorophyllase. Tauchiya et al. (12) puried two not require the presence of a metal ion in the center
of the tetrapyrrole ring. This is demonstrated by the (19) found that chlorophylls a 0 and b 0 were not hydro-
fact that chlorophyllase has been found to hydrolyze lyzed by chlorophyllase from two different plants and
chlorophylls and pheophytins with similar Km and that the presence of these isomers did not affect the
Vmax value (4). Kinetic data suggested that the Km is hydrolysis rates of the corresponding chlorophylls.
a binding constant for those substrates. Data have not These results suggested that a carbomethoxy group
been obtained to determine the effect of different metal on the opposite side of ring V, as occurs in chlorophyll,
ions in the porphyrin ring on chlorophyllase hydroly- prevents both binding and hydrolysis of chlorophylls
sis. Transesterication experiments have shown that a 0 and b 0 . Comparison of the kinetics of hydrolysis of
chlorophyllase can use a variety of primary and sec- methyl pheophorbide a with 9-hydroxy methyl pheo-
ondary alcohols in place of phytol alcohols as sub- phorbide a showed that introduction of a hydroxyl
strates (2). This implies that the enzyme can also group at C9 in ring V reduced the binding afnity of
hydrolyze chlorophyllide esters of the same alcohols. substrate dramatically while, at the same time, having
Protochlorophyll and 1-vinyl protochlorophyll are not little effect on substrate hydrolysis rate (8).
hydrolyzed by chlorophyllases, but they bind with af- There is little information available on the stability
nities similar to pheophytin a (4). This showed that a of chlorophyllase from different plants. In general,
double bond between C7 and C8 does not affect bind- assays for activity have been performed at 30 C at a
ing to chlorophyllase but that it prevents hydrolysis of pH near 7. Recently, a 30 C optimal temperature was
phytol. Similarly, removal of the carbomethoxy group reported for chlorophyll hydrolysis by crude chloro-
from C10 of pheophytin greatly reduces hydrolysis phyllase extracts from artichoke (20). Chlorophyllase
rates with little effect upon binding (4). Fiedor et al. activity was nearly inactivated during a 1-h incubation
Figure 2 N-terminal sequences of chlorophyllases from orange (Citrus sinensis L.) (A) and Chenopodium album (B and C).
(From Refs. 11, 12.)
Phytase
Onno Misset
DSM Patents & Trademarks, Delft, The Netherlands
Intra- or
Organism extracellular Reference
Fungi
Agrocybe pediades Extra 6
Aspergillus spp. Extra 7
Aspergillus amstelodami Extra 8
Aspergillus awamori Extra 9
Aspergillus chevalieri Extra 8
Aspergillus candidus Extra 8
Aspergillus cuum Extra 8
Aspergillus fumigatis Extra 10, 11
Aspergillus avus Extra 7, 8
Figure 1 Structure of phytic acid. Carbon 1 of the myo- Aspergillus niger Extra 7, 8, 14
inositol ring has been indicated. Aspergillus repens Extra 8
Aspergillus sydowi Extra 8
Aspergillus terreus Extra 7, 12, 15
Aspergillus vesicolor Extra 7, 8
ester bonds (subgroup 1), in particular phosphomo- Aspergillus wentii Extra 8
noesters (subsubgroup 3). Botrytis cinerea Extra 8
EC 3.1.3.8 is a 3-phytase (recommended name) and Emericella nidulans Extra 13
hydrolyzes rst the ester bond at the 3-position of the Geotrichum candidum Extra 8
substrate. The systematic name is myo-inositol hexaki- Mucor species Extra 7
sphosphate 3-phosphohydrolase. EC 3.1.3.26 is 6-phy- Mucor proformis Extra 8
tase (recommended name) and hydrolyzes rst the Mucor racemosus Extra 8
Myceliophtora thermophila Extra 10, 12
ester bond at the 6-position of the substrate. The sys-
Neurospora crassa Extra 16
tematic name is myo-inositol hexakisphosphate 6- Paxillus involtus Extra 6
phosphohydrolase. Penicillium species Extra 7
Penicillium caseicolum Extra 17
Peniophora lycii Extra 6, 18
II. NATURAL OCCURRENCE OF
Rhizopus oryzae Extra 8
PHYTASE Rhizopus oligosporus Extra 8
Rhizopus stolonifer Extra 8
Phytases are found in both the plant and microbial Talaromyces fumigatus Extra 10, 13
kingdoms. While most plant phytases belong to the Thermomyces lanugosis Extra 19, 20
class of 6-phytases (EC 3.1.3.26), most microbial phy- Trametes pubescens Extra 6
tases are considered to be 3-phytases (EC 3.1.3.8). Ruminant microorganisms Extra 21
However, not all publications mention whether the Selenomonas
phytase involved is a 3- or 6-phytase; the reason for Prevotella
this is that it is rather complicated to determine the Treponema
Megasphaera
nature of the myo-inositol pentaphosphate (1,2,4,5,6
Yeast
or 1,2,3,4,5) generated by the enzymes.
Saccharomyces cerevisiae Extra 8
The occurrence of phytase in germinating plants has Schwanniomyces occidentalis Extra 2224
been reviewed recently (2, 5). The rst phytase pre- Bacteria
parations were of plant origin; Suzuki et al. in 1907 Aerobacter aerogenes Intra 25
isolated them from rice and wheat bran (3). All pre- Bacillus subtilis Extra 26, 27
sent-day commercial phytase products are derived Bacillus natto N-77 Extra 28, 29
from lamentous fungi, in particular from Aspergillus Bacillus DS11 Extra 30, 31
spp. For this reason this chapter will further focus on Escherichia coli Intra 32
the microbial, in particular fungal, phytases. Escherichia coli B Intra 33
Extensive screening programs have been carried out Klebsiella aerogenes Intra 34
Pseudomona species Intra 35
in the past and are still carried out today. These pro-
Produced by Production
Company Trademark GMOa organism Phytase from Reference
case of the PhyA phytase of Aspergillus cum (strain B. Primary, Secondary, Tertiary, and
NRRL 3135), the molecular weight of the precursor Quaternary Structures
phytase (51,086 daltons; Table 4) could be calculated
from the gene-derived sequence (43, 44). In addition, The primary sequences of some 2025 phytases are
the mature protein was also sequenced completely by known. Table 2 shows that many of these sequences
Edman degradation (45). A comparison of the two are available at the Web from sequence databases such
sequences allowed identifying the position of the as SWISS-PROT, TrEMBL, and PIR. Yet, many
maturation site; it was found that the signal peptide (new) sequences are available from the patent literature
consists of 23 amino acids, resulting in a mature pro- only.
tein of 444 amino acids with a calculated molecular Pairwise alignment of the amino acid sequences and
weight of 48,846 daltons. The chemically and gene- calculation of the homology revealed that the phytase
derived sequences of the mature protein were found sequences can be derived into three groups: fungal
to be identical. phyA sequences, fungal phyB sequences, and others
In a recent study, Wyss et al. (49) determined the such as the sequences from Bacillus. The homology
molecular weights of several fungal phytases by SDS- within a group can be as high as 4597% as is illu-
PAGE, analytical ultracentrifugation, and gel ltra- strated in Table 5 for the fungal PhyA phytases.
tion analysis (see Table 4 for a summary). The experi- Between the groups the homology is much lower; i.e.,
mentally determined molecular weights are much between phyA and phyB the homology is 2327%
larger than the values calculated on the basis of the whereas no signicant homology exists between the
amino acid sequences. The difference in molecular Bacillus phytases and the fungal phyA and/or phyB
weight must be attributed to the glycosyl chains phytases (not shown). Figure 2 shows a multiple align-
that are attached to the enzymes. The authors con- ment of fungal PhyA sequences. The N-terminal resi-
cluded that all phytases are monomers and that the dues with a gray background have been shown or are
contribution of the sugar chains to the molecular suggested to belong to the signal peptide. Residues in
weight of the glycosylated mature enzyme could white with a black background are conserved in at
amount to 2065%. least ve sequences.
Precursor Mature
A.nig01 A.nig02 A.awa A.fum A.nid A.ter01 A.ter02 T.the T.lan M.the P.lyc P.inv 1 P.inv 2 T.pub A.ped
A.nig01
A.nig02 96
A.awa 97 95
A.fum 66 65 66
A.nid 62 62 62 67
A.ter01 62 62 63 60 59
A.ter02 60 60 61 59 57 87
T.the 61 61 61 60 59 59 87
T.lan 54 53 54 52 54 49 48 59
M.the 46 46 46 49 48 45 47 45 46
P.lyc 40 41 41 40 40 43 43 43 38 43
P.inv 1 41 41 40 40 40 41 41 37 40 41 55
P.inv 2 38 39 38 39 39 39 39 36 39 40 55 85
T.pub 41 39 41 40 39 39 39 39 40 41 51 62 63
A.ped 38 38 38 37 42 38 38 37 37 39 51 58 58 61
The secondary and tertiary structures of phytase are 3). The homology between these two types of phytases
known in great detail because the rst 3-D structure of is relatively low, and the enzymological properties are
phytase from A. niger (cuum) was recently determined very different (51).
by x-ray crystallography at a resolution of 2.5 A by In relation to phytases, several publications also
researchers of HoffmanLa Roche (50). To be success- refer to acid phosphatases (EC 3.1.3.2). However,
ful in crystallizing the phytase, the enzymes had to be these enzymes show no amino acid sequence homol-
deglycosylated completely. ogy with phytases whatsoever, and also their sub-
The structure has an =-domain similar to that of strate spectrum differs from phytase since they
rat acid phosphatase and an -domain with a new fold cannot use phytate as a substrate (51, 53).
(Fig. 3). The high homology that exists between fungal Therefore, fungal PhyA phytases can be regarded as
PhyA phytases suggests that all these phytases (Table 5 very specialized acid phosphatases because of their
and Fig. 2) will have the same secondary structural unique property to efciently dephosphorylate phytic
elements and tertiary fold and topology. acid.
Not all the 444 amino acids could be detected in the
electron density map: residues 16 and 249252 (num-
2. From Posttranslational Modication
bering of the mature protein; add 23 to obtain the
precursor protein numbering) were not visible, pre- Extracellular phytases undergo two posttranslational
sumably because of too high a mobility. All 10 cysteine modications: glycosylation and maturation.
residues are involved in a disulde bridges: Cys817, Glycosylation takes place in the Golgi apparatus of
Cys48391, Cys192442, Cys241259, and Cys413 eukaryotic cells like fungi and yeast (and not in pro-
421. The amino acid alignment in Figure 2 shows karyotic cells such as bacteria), and it involves only
that, with the exception of the cysteines involved in those proteins that are secreted by the cell.
bridge Cys817, the eight other cysteines and therefore Glycosylation is known to generate a heterogeneous
the four disulde bridges are completely conserved in population of glycoproteins. Since the glycosyl chains
the fungal PhyA phytases. Amino acids involved in contribute substantially to the molecular weight of
catalysis and substrate binding will be discussed later. phytase, the molecular weight will show a distribution
around an average value instead of having one single
C. Isoforms value.
Glycosylation of different phytases depends on the
1. From Different Genes
microbial expression system (49). In Aspergillus, glyco-
In Aspergillus cuum and A. awamori, two phytase sylation was found to be moderate, whereas it was
genes have been identied: PhyA and PhyB (Table excessive and highly variable in Humicola polymorpha
Spec. act.
kcat KM kcat =KM at Vmax
Substrate (sec1 ) (mM) (mM1 :sec1 ) (U/mg) Conditions Reference
Phytate:
Dodeca-sodium phytate (IP6) 216 0.040 5400 126 58 C, pH 5:0 52
Dodeca-sodium phytate (IP6) 348 0.027 12888 58 C, pH 5:0 51
Dodeca-sodium phytate (IP6) 171 0.250 684 100 37 C, pH 5:5 44
Dodeca-sodium phytate (IP6) < 0:005 103 37 C, pH 5:0 53
Mono-potassium phytate (IP6) 138 0.025 5520 81 58 C, pH 5:0 52
Dipotassium-tetramagnesium 222 0.070 3171 130 58 C, pH 5:0 52
phytate (IP6)
Myo-inositol-1,2,4,5,6- 477 0.16 2980 58 C, pH 5:0 51
pentakisphopate (IP5)
Myo-inositol-1,2,5,6- 1554 1.00 1554 58 C, pH 5:0 51
tetrakisphophate (IP4)
Myo-inositol-triphophate (IP3) 164 0.20 820 58 C, pH 5:0 51
Myo-inositol 2-monophosphate 12 4.50 3 7 58 C, pH 5:0 52
(IP1)
Para-nitrophenylphosphate 104 0.27 385 60 58 C, pH 5:0 52
Beta-glycerophosphate 17 1.66 10 10 58 C, pH 5:0 52
Na2 H2 pyrophosphate 42 0.29 145 24 58 C, pH 5:0 52
ATP 104 0.35 300 60 58 C, pH 5:0 52
Beta-NADP 17 0.50 34 10 58 C, pH 5:0 52
Glucose-6-phosphate 28 8.30 3 2 58 C, pH 5:0 52
Phenylphosphate 103 0.53 194 6 58 C, pH 5:0 52
4-nitrophenylphosphate bis 200 0.72 278 117 58 C, pH 5:0 52
D-fructose-1,6-bisphosphate 59 1.60 37 34 58 C, pH 5:0 52
Adenosine 5 0 -diphosphate 22 0.39 56 13 58 C, pH 5:0 52
Beta-naphthylphosphate 22 0.53 42 13 58 C, pH 5:0 52
The structure of phytase was compared with the active-site residues (Arg81, His82, Arg85, Arg165,
structure of rat acid phosphatase (50). The substrate His361, and Asp362) are involved in electrostatic
prole of phytase differs from the rat acid phosphatase interactions with the scissile 3-phosphate group of
in that phytase is able to hydrolyze the large phytate the substrate. Furthermore, His82 is in a favorable
substrate compared to the smaller substrates of the rat position to make the nucleophilic attack at the
enzyme. Inspection of the active site of the two phosphorus and Asp362 is in a position to proto-
enzymes reveals on one side of the substrate-binding nate the leaving alcohol [for a further discussion see
pocket the following conserved residues (numbering of (50)].
the A. niger precursor protein): Arg81 (Arg11 in the rat
acid phosphatase), His82 (His12), Gly83 (Gly13),
Arg85 (Arg15), Pro87 (Pro17) of the RHGxRxP pep- VI. PHYTASE ASSAYS
tide as well as Arg165 (Arg79), His361 (His257), and
Asp362 (Asp258) which all belong to the =-domain The (specic) activity of phytase is usually determined
of the enzyme. On the other side of the substrate-bind- with its natural substrate phytate or the chromogenic
ing pocket (the -domain) no similarities were found. substitute para-nitrophenylphosphate. The rst assay
Here, the active site of the phytase is much larger than involves the detection of the liberated inorganic phos-
that of the rat acid phosphatase, thus allowing the phate while the second one detects the yellow para-
accommodation of the larger phytate substrate. nitrophenol spectrophotometrically.
A molecular modeling study of the phytasephy- Many conditions have been reported in the litera-
tate complex showed that all conserved charged ture for the phytase assay using phytate as a sub-
pH optimum Temperature
Phytase from at 37 C Ref. optimum (pH) Ref.
A. niger (cuum) NRRL 3135 2.56 54 58 (5.0) 54
2.56 44 50 (5.5) 44
2.56 53 55 (5.0) 37
26 19 55 (5.5) 19
A. fumigatis 38 53 55 (5.0) 37
37.5 11, 36
A. terreus 9A1 47 53 50 (5.0) 37
A. terreus CBS 47 53 45 (5.0) 37
E. nidulans 5.57 53 45 (5.0) 37
M. thermophila 3.56.5 53
T. lanuginosis 38 19 65 (5.5) 19
A. pediades 36 6 50 (5.5) 6
P. lycii 36 18 50 (5.5) 18
B. subtilis 69 27
E. coli 2.56 53
a
In all cases, phytate was used as substrate.
strate. The basic principle, however, is the same. phate, pH 4.0, and 100 L 10 mM PNPP. The absor-
Variations exist with respect to the incubation tem- bance of the liberated para-nitrophenol is measured
perature, pH, and choice of buffer, substrate concen- spectrophotometrically at 30 nm as a function of
tration, and the like. A suitable assay is carried out as time. The absorbance increase (A/min) can be con-
follows (44): 100 L enzyme solution or distilled verted to a rate expressed in M/min using an extinc-
water as a blank is added to 900 L of a solution tion coefcient of 3690 M1 cm1 . One unit of phytase
composed of 0.25 M sodium phosphate buffer, pH is dened as the formation of 1 mole of para-nitro-
5.5, and 1 mM phytic acid (sodium salt). The result- phenol per minute.
ing mixture is incubated for 30 min at 37 C. The
reaction is stopped by the addition of 1 mL of 10%
trichloroacetic acid. After the reaction has termi- VII. PURIFICATION
nated, 2 mL of a reagent composed of 3.66 g of
FeSO4 7 H2 O in 50 mL of ammonium molybdate Numerous purication methods for the various phy-
solution (2.5 g (NH4 )6 Mo7 O24 4 H2O and 8 mL tases discussed here have been published in the litera-
H2 SO4 , diluted to 250 mL with demineralized ture; for a summary see Table 9. In general, the
water), is added. The intensity of the blue color is complexity of the method and number of steps
measured spectrophotometrically at 750 nm. The involved depend on the expression level of the enzyme.
measurements are indicative of the quantity of phos- This means that for the purication of phytase from
phate released in relation to a calibration curve of nonrecombinant strains, more steps are required to
inorganic phosphate in the range of 01 mM. One obtain a homogeneous phytase preparation than pure
unit of phytase is dened as the amount of enzyme phytase from overexpressing strains.
that is capable of forming 1 mole of phosphate per
minute. Specic activities of various phytases are
summarized in Table 8. ACKNOWLEDGMENTS
The activity of phytase can also be determined con-
veniently by using para-nitrophenylphosphate (PNPP) I would like to acknowledge my colleagues Dr. Rob
as a substrate. The conditions for this determination Beudeker and Dr. Jan-Metske van der Laan for
are as follows: 10 L of an enzyme solution is added to helpful discussions and critically reading the manu-
a solution containing 890 L of 0.25 M sodium phos- script.
Original strain or
Phytase from overexpressed in Method Remarks Reference
continued
Original strain or
Phytase from overexpressed in Method Remarks Reference
Misset
4. Anion exchange chromatography Q-Sepharose Fast Flow
5. Anion exchange chromatography Source 30Q
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
Phytase
P. lycii A. oryzae IFO4177 1. Anion exchange chromatography Q-Sepharose Fast Flow 18
2. Hydrophobic interaction chromatography Phenyl
Toyopearl 650S
3. Desalting Sephadex G25
4. Anion exchange chromatography Q-Sepharose Fast Flow
5. Anion exchange chromatography Source 30Q
T. pubescens A. oryzae IFO4177 1. Anion exchange chromatography Q-Sepharose Fast Flow 6
2. Hydrophobic interaction chromatography Butyl
Toyopearl 650S
3. Desalting Sephadex G25
4. Anion exchange chromatography Q-Sepharose Fast Flow
5. Desalting Sephadex G25
6. Anion exchange chromatography Source 30Q
T. lanuginosis Fusarium venenatum 1. Anion exchange chromatography Q-Sepharose Big Beads 19
2. Ultraltration
3. Anion exchange chromatography MonoQ HR10/16 column
4. Ultraltration
5. Cation exchange chromatography MonoS HR 5/5 column
T. thermophilus S. cerevisaea 1. Hydrophobic interaction chromatography Butyl Sepharose 49
Fast Flow
2. Gel permeation chromatography Sephacryl S-300
A. niger NW205 or 1. Desalting Fast-Desaltin HR 10/10 or Sephadex G-25 49
H. polymorpha superne
2. Anion exchange chromatography Poros HQ/M
a
Most methods comprise as a rst step the removal of the cells by centrifugation or ltration. The clear, occasionally ultraltrated, culture supernatant is then subjected to the further
stages as indicated.
703
REFERENCES organisms. I. Isolation and identication of phytase
producing mold. J Ferment Technol 46:858862,
1. E Graf. Chemistry and application of phytic acid: an 1968.
overview. In: E Graf, ed. Phytic Acid Chemistry and 16. N Atsuhisa et al. Phytase and its production.
Applications. Minneapolis: Pilatus Press, 1986. Japanese patent application JP7059562 to Ichibiki
2. DM Gibson, ABJ Ullah. Phytases and their action on KK. Priority date: 19 August 1993. Publication
phytic acid. In: Inositol Metabolism in Plants. New date: 7 March 1995.
York: Wiley-Liss, 1990, pp 7792. 17. H Matsuoka. New phytase and its production.
3. RJ Wodzinski, AHJ Ullah. Phytase. Adv Appl Japanese patent application JP7067635 to Amano
Microbiol 42:263302, 1996. Pharmaceutical Co Ltd. Publication date: 14 March
4. W Aehle, O Misset. Enzymes for industrial applica- 1995.
tions. In: HJ Rehm, G Reed, eds. Biotechnology. 18. SF Lassen, L Bech, CC Fuglsang, J Breinholt, A
Weinheim: Wiley-VCH, 1999, pp 191216. Ohmann, PR Ostergaard. Peniophora phytase.
5. NR Reddy, SK Sathe, DK Salunkhe. Phytates in International patent application WO 98/28408 to
legumes and cereals. Adv Food Res 28:192, 1982. Novo-Nordisk. Priority date: 20 December 1996.
6. SF Lassen, L Bech, A Ohmann, J Breinholt, CC Publication date: 2 July 1998.
Fuglsang. Phytase polypeptides. International patent 19. RM Berka, MW Rey, KM Brown, T Byun, AV Klotz.
application WO 98/28409 to Novo-Nordisk. Priority Molecular characterization and expression of a phy-
date: 20 December 1996. Publication date: 2 July 1998. tase gene from the thermophilic fungus Thermomyces
7. TR Shieh, JH Ware. Survey of microorganisms for the lanuginosus. Appl Environ Microbiol 64:44234427,
production of extracellular phytase. Appl Microbiol 1998.
16:13481351, 1968. 20. RM Berka, K Rey, AV Klotz. Polypeptides having
8. SJ Howson, RP Davis. Production of phytate-hydro- phytase activity and nucleic acids encoding same.
lysing enzyme by some fungi. Enzyme Microb International patent application WO97/35017 to
Technol 5:377382, 1983. Novo-Nordisk. Priority date: 18 March 1996.
9. CS Piddington, CS Houston, M Paloheimo, M Publication date: 25 September 1997.
Cantrell, A Miettinen-Oinonen, H Nevalainen, J 21. LJ Yanke, KJ Cheng, CW Forsberg, LB Selinger, HD
Rambosek. The cloning and sequencing of the genes Bae, L Zhou. DNA-sequences encoding phytases of
encoding phytase (phy) and pH 2.5-optimum acid ruminal microorganisms. International patent appli-
phosphatase (aph) from Aspergillus niger var. awa- cation WO 97/48812 to United Kingdom
mori. Gene 133:5562, 1993. Government (Canada). Priority date: 14 June 1996;
10. APGM van Loon, D Mitchell. Polypeptides with phy- publication date 24 December 1997.
tase activity. European patent application EP0684313 22. L Sequilha, C Lambrechts, H Boze, G Moulin, P
to HoffmanLa Roche. Priority date 15 April 1994. Galzy. J Ferment Bioeng 74:711, 1992.
Publication date 29 November 1995. 23. T Suzuki, D Mochizuki, SI Tawaki, M Shimada, J
11. L Pasamontes, M Haiker, M Wyss, M Tessier, APGM Tokuda. Novel phytase. European patent application
van Loon. Gene cloning, purication and characteri- EP0699762 to Mitsui Toatsu Chemicals. Priority date:
zation of a heat-stable phytase from the fungus 5 May 1994; publication date: 6 March 1996.
Aspergillus fumigatus. Appl Environ Microbiol 24. D Mochizuki, J Tokuda, T Suzuki, S Masao, S
63:16961700, 1997. Tawaki. New phytase. Japanese patent application
12. DB Mitchell, K Vogel, BJ Weimann, L Pasamontes, JP8289782 to Mitsui Toatsu Chemicals. Publication
APGM van Loon. The phytase subfamily of histidine date: 5 November 1996.
acid phosphatases: isolation of genes for two novel 25. MP Greaves, G Anderson, DM Webley. The hydro-
phytases from the fungi Aspergillus terreus and lysis of inositol phosphates by Aerobacter aerogenes.
Myceliophthora thermophila. Microbiol 143:245252, Biochim Biophys Acta 132:412418, 1967.
1997. 26. VK Powar, V Jagannathan. Purication and proper-
13. L Pasamontes, M Haiker, M Henriquez-Huecas, DB ties of phytate-specic phosphatase from Bacillus sub-
Mitchell, AP van Loon. Cloning of the phytases from tilis. J Bacteriol 151:11021108, 1982.
Emericella nidulans and the thermophilic fungus 27. J Apajalahti, P Heikkinen, M Lauraeus, A Morgan,
Talaromyces thermophilus. Biochim Biophys Acta P Nurminen, O Siikanen. Phytase. UK patent
1353:217223, 1997. application GB2316082 to Finnfeeds Int. Priority
14. T Skowronski. Some properties of partially puried date: 13 August 1996; publication date: 18
phytase from Aspergillus niger. Acta Microbiol Pol February 1998.
27:4148, 1978. 28. M Shimizu. Purication and characterization of phy-
15. K Yamada, Y Minoda, T Kobayashi, Y Hidaka, H tase from Bacillus subtilis (natto) N-77. Biosci Biotech
Matuo, M Kobayashi. Studies on phytase of micro- Biochem 56:12661269, 1992.
-Amylases
Figure 1 Comparison of regions of sequence similarities among: Bacillus stearothermophilus (5), Bacillus amyloliquefaciens (6),
Bacillus licheniformis (7), Aspergillus oryzae (8), Micrococcus sp. 207 (9), Saccharomycopsis buligera (10), barley isozyme 1 (11),
Drosophila melanogaster (fruit y) (12), yellow mealworm (13), porcine pancreatic (14), human pancreatic (15). The three residues
involved in catalysis are indicated by asterisks. Numbering of the sequences starts with N-terminus of the mature protein.
Figure 2 A stereo view of the barley -amylase (isoform 2). The structure consists of three domainsthe central (=8 -barrel
(domain A), a ve-stranded antiparallel -sheet domain C at the bottom, and a protruding loop at the top (domain B). (From
Ref. 29, PDBid: 1 amy.)
glucoside oxygen side), with the product maintaining lent intermediate if nucleophilic attack occurs prior to
the -conguration. In inverting glycosidases, such as the complete formation of the cation.
-amylase and glucoamylase, a back-side attack results
in an inversion of the conguration. The appealing C. Ring-Opening Reaction
aspect of this model is that the catalytic reactions of
both inverting and retaining glycosidases can be An alternative mechanism that is not as well received
explained by a single unied mechanism (50). as the above two models involves endocyclic carbon
The main difference between this model and the oxygen cleavage as the rate-determining step, leading
displacement mechanism lies in whether a transient to the formation of an acyclic oxocarbonium ion inter-
or stable oxocarbonium ion intermediate exists (51, mediate (44, 53). In this model, there is no distortion of
52). The formation of a stable oxocarbonium ion inter- the substrate into a twist-boat conformation as shown
mediate is preferred if bond breaking precedes the by molecular dynamics simulation. In the chair con-
attack of water facilitated by general base catalysis. formation of both and -anomers, the antiperiplanar
The oxocarbonium ion is readily collapsed into a cova- orientation of the exocyclic O4 lone pair fullls the
stereoelectronic requirement in the fragmentation of proteinaceous inhibitors (61, 62). Analysis of the crys-
the ring CO bond (54). It has been also postulated tal structure indicates that the head of Tendamistat
that protonation occurs at the cyclic oxygen instead of containing the highly conserved triplet residues binds
the glucosidic oxygen, and the water nucleophilic directly to the active site of the porcine pancreatic -
attack is followed by a rotation of the hemiacetal cen- amylase, blocking the ve binding subsites 3 to 2,
ter to position for cleavage (55). with Arg19 forming a salt bridge with the catalytic
residue, Glu233 (for subsite nomenclature, refer to
Ref. 62). In the formation of the enzymeinhibitor
V. -AMYLASE INHIBITORS complex, the conformation of the enzyme active site
is induced to accommodate tight binding of
Two families of -amylases inhibitors (-AI), including Tendamistat and related inhibitors.
carbohydrate- and protein-based inhibitors, have been The seeds of common beans contain a family of
studied in detail. The knowledge of the inhibitory proteins involved in a defense mechanism, composed
action of these compounds is of great interest in med- of phytohemaglutinin, and -amylase inhibitor. The
icine in the attempt to develop therapeutics for dia- -AIs inhibit the activity of some mammalian and
betes, obesity, and hyperlipemia. Proteinaceous -AIs insect -amylases, but not of the endogenous plant
are widespread in microorganisms and higher plants. enzyme. Red kidney bean -AI has a KD of 3:5
Tendamistat, one of the most potent inhibitors of 1011 M at pH 6.9 and 30 C (63), and black bean
mammalian -amylases, is an acidic protein of 74 -AI 4:4 109 M at pH 6.9 at 37 C (64). There
amino acids isolated from the culture medium of are at least ve known sequences derived from -
Streptomyces tendae (56, 57). This class of inhibitors AIs of common domesticated beans, wild common
also includes HAIM (S. grieseosporeus) (58), AI-3688 bean, white kidney bean, and black bean (65). The
(S. aureofaciens) (59), and PAIM (S. griseosporeus) inhibitors are glycoproteins synthesized as prepropro-
(60). Tendamistat binds tightly with porcine pancreatic teins which require proteolytic cleavage for activation
-amylase with a very high inhibition constant of 9 (66). Translational processes produce the active inhi-
1012 M (56). The molecular structure assumes a - bitor formed by the association of two polypeptide
barrel of two -sheets, each consisting of three antipar- chains, and -subunits (67). In the crystal structure
allel strands. The -strands are connected by loops and of the enzymeinhibitor complex, -AI-I exists as a
turns which contain the active-site triplet Trp18- dimer that binds two porcine pancreatic enzyme
Arg19-Tyr20 characteristics of all Tendamistat-type molecules (68). Each monomer consists of a concave
b-Amylases
dues with a molecular weight of 56 kDa. The enzyme Waals contacts between Val99 and the substrate. The
consists of a large =8 -barrel core; a smaller lobe L3 lid serves to shield the active site from being
formed by three long loops (L3, L4, and L5) extend- exposed to the solvent, and also positions the sub-
ing from 3, 4, and 5 strands; and a C-terminal tail strate in proximity to the two catalytic residues
loop of 50 residues extending from 8 and covering Glu186 (acting as a general acid) and Glu380 (acting
the entrance of the barrel. Between the surface of the as a general base). Glu186 has been identied as the
lobe region and the core is a cleft that opens into a catalytic residue by afnity labeling with 2,3-epoxy-
pocket 18 A deep that contains the catalytic resi- propyl--D-glucopyranoside (13). The binding energy
dues Glu186 and Glu380, and an essential residue upon closing the lid for -maltose estimated in an
Asp101. This architecture contrasts to that of -amy- automated docking experiment is 22 kcal/mol, indi-
lases which have the reactive center located in a long cating considerable enhancement of interactions
cleft open at both ends (11). The deep pocket con- between the substrate and the subsites (14, 15). The
formation found in -amylase is also present in glu- active site of soybean -amylase can accommodate
coamylase, although the latter has an =8 -core two maltose molecules, corresponding to a maltote-
domain (12). That the catalytic site is situated in a traose. The two maltose molecules bind in tandem in
cavity allows endwise hydrolysis of the starch chain the active site cleft occupying subsites 1, 2 and 1,
with the length of glucose units cleaved in a con- 2, respectively, with the nonreducing end at the 1
trolled manner. Substrate binding causes a movement subsite and cleavage occurring between subsite 1
of some segments in L3 (residues 96103) to close and 1 (for subsite nomenclature, refer to 16).
over the bound substrate like a hinged lid. In the Substrates with longer chain length may have the
crystal structure of soybean -amylase complexed outer glucose units anchored by hydrophobic interac-
with maltose, L3 in the closed conformation interacts tion with the methyl groups of Leu383 at the cleft
with maltose forming hydrogen bonding between entrance, as revealed in the crystal structure of soy-
Asp101 and O-2 of the glucose unit 1, and Van der bean -amylase complex with -cyclodextrin (6).
Figure 2 A stereo view of barley -amylase (isoform 2). The structure consists of a large =8 -barrel core, a smaller lobe
formed by three long loops (L3, L4, and L5), and a C-terminal tail loop residues extending from 8 and covering the entrance of
the barrel (Ref. 9, PDBid: 1bly).
Glucoamylase
Peter J. Reilly
Iowa State University, Ames, Iowa, U.S.A.
(16). These values can vary slightly because of different occur. To overcome this handicap, many lamentous
degrees of O-glycosylation produced under different fungi have evolved the ability to produce very large
culture conditions. Isoelectric focusing also indicates amounts of extracellular GA. The placement of the
multiple forms of GA I and GA II (17), presumably active site in a well explains both the exo-acting nat-
also caused by glycosylation differences. GAs from ure of GA and its multichain attack pattern.
other organisms have widely varying molecular A. niger and A. awamori GAs were the rst GAs
weights, depending on their primary structures and sequenced, in 1983 and 1984, respectively (25, 26). Of
degrees of glycosylation, and more specically on the 21 archaeal, eubacterial, yeast, and lamentous
whether they possess linkers and starch-binding fungal GAs whose full catalytic domain primary struc-
domains. tures are known at present, computational analysis
shows that all except those from Schwanniomyces occi-
B. Catalytic Domain dentalis and Schizosaccharomyces pombe are substan-
tially homologous, and have essentially the same
The crystal structure of the catalytic domain of GA secondary and tertiary structures as the A. awamori
from a strain of A. awamori var. X100, a slightly var. X100 GA catalytic domain (18, 27, 28). -
different strain from the A. awamori/A. niger form Helices 2 and 3 are shortened and fused in the GA
used industrially, was the rst GA structure to be from the lamentous fungus Humicola grisea var. ther-
elucidated (12). The domain is an (; 6 barrel, moidea, and the peripheral -helix 11 is missing from
with six interior -helices surrounded by six exterior the GAs from the eubacterium Clostridium var. G0005
-helices (Fig. 2), a structure found elsewhere among and the archaeon Methanococcus jannaschii (18). In
glycohydrolases only in Clostridium thermocellum addition, a number of loops vary in length in different
endo-1,4--glucanases A (22) and D (23) (CelA and GA catalytic domains.
CelD, respectively) and in Thermomonospora endo/ GA is a member of family 15 of glycosyl hydrolases
exo-cellulase E4 (24). There are 13 helices in all, (2932), the families being classied by primary
with helix 11, counting from the N-terminus of the sequence homology. Although nearly all the members
domain, peripheral to the rest. The helices form a bed of this family are GAs, a glucodextranase (EC 3.2.1.70)
supporting a network of loops, in the center of which from Arthrobacter globiformis is also present (32).
is found the active site. This is a well 10 A deep Of the other three glycosidases with ; 6 barrel
and 15 A wide at its mouth, a very primitive struc- structures, C. thermocellum endoglucanases CelA
ture ensuring that the enzyme will have a low turn- and CelD are in families 8 and 9, respectively, and
over number, since the substrate chain must penetrate T. fusca endo/exocellulase E4 is in family 9, so
the well deeply and undergo cleavage, followed by the their primary structures are not closely related to
remaining chain and then the liberated glucose mole- that of GA. However, they may well share a distant
cule leaving the well before the next reaction can ancestor.
C. O-Glycosylated Linker
Figure 3 Phylogenetic tree (18) showing the seven subfamilies of GAs from eubacteria (Clostridium), archaea (Methanococcus),
yeast (Saccharomyces, Saccharomycopsis, and Arxula), and lamentous fungi (Rhizopus and Aspergillus). Dark circles: catalytic
domains; medium-light ovals: starch-binding domains; light rods: O-glycosylated regions or linkers; black ovals: other domains;
vertical bar: cell wall. (From Ref. 18.)
All but one of the many GAs that have been sub-
jected to subsite mapping with various substrates
appear to have six or seven subsites, with the rst,
binding the nonreducing glucosyl residue of the sub-
strate, having a small negative or positive binding free
energy, and the ve or six subsites to the other side of
the cleavage point having progressively smaller but
always negative binding free energies (17, 5064).
This causes substrates with more glucosyl residues to
be bound more rmly and cleaved at higher rates than
those with fewer residues. With all substrates and all
but one GA, values of KM decrease essentially mono-
tonically with increasing substrate chain length, while
values of kcat increase until reaching a maximum at
substrates with four or ve glucosyl residues (Figs. 6,
7).
The reason for the uniformity of subsite structure
with different GAs has now become clear, in that all
have essentially the same tertiary structure, with speci-
cally the same active-site structure and general acidic
and basic catalytic residues (18). The one exception to
this uniformity is the Swanniomyces castelli GA, which
has different kinetics (65) and which is not homolo-
gous with the other GAs.
The number of substrate-binding subsites in GA,
obtained by mathematical analysis of the kinetics of
the hydrolysis of oligosaccharides of different lengths,
does not agree with either the crystal structures of GA
Figure 5 Schematic of GA mechanism. with various ligands or with simulated docking studies,
Limit Dextrinase/Pullulanase
E. Ann MacGregor
University of Manitoba, Winnipeg, Manitoba, Canada
Values of Km for pullulan for Klebsiella pullulanase active between pH 5.5 and 7, and is stable in the
are given as 2.5 mM (10), 78 M, or 21 M (43) in pH range 512 (9); the B. acidopullulyticus and B.
glucose equivalents, and 70 M (44) in terms of con- deramicans enzymes have maximum activity at pH
centration of hydrolyzable -1,6-glucosidic bonds. For 46 and pH 3.754.9, respectively, and are stable at
the B. acidopullulyticus enzyme, Km for pullulan is said pH 49 (10) and 3.56 (11), respectively. The limit
to be 2.5 mM glucose equivalents (10), while barley dextrinase of barley, on the other hand, has optimum
limit dextrinase has a Km for amylopectin of 26 mM activity at pH 5.56.5 in the presence of bovine serum
glucose equivalents (3). albumin, which seems to stabilize the active enzyme
The turnover number (kcat ) for Klebsiella pullula- (30).
nase has been measured as 99 sec1 or 220 sec1 at 30 Temperature optima for Klebsiella and B. acidopul-
C (43), 53 sec1 at 25 C, 132 sec1 at 40 C (44), and lulyticus pullulanases are 50 C and 65 C, respectively,
290 sec1 at 60 C, while measured values for the B. and both enzymes are stable at temperatures up to 55
acidopullulyticus enzyme are 2:9 103 or 3:6 103 C at pH 5 (9, 10). The enzyme from B. deramicans is
sec1 at 60 C, all on pullulan (10). For barley limit said to retain 75% of its activity when held at 60 C for
dextrinase acting on a well-dened dextrin, 62 -0-- 24 h at pH 4.5, and has optimum activity at 60 C (11).
maltotriosyl maltotriose, at 40 C, the number is The barley limit dextrinase shows maximum activity at
90 sec1 (42). 4050 C (45, 46) and is stable under the mashing con-
The response to pH and temperature varies with ditions in a brewery for at least 1 h at 60 C, but loses
the enzyme source. Klebsiella pullulanase is most activity rapidly at 65 C (47).
Enzyme source K. pneumoniae (9) B. acidopullulyticus (10) B. deramicans (11) Barley (46)
Limit Dextrinase
mented by yeast, except in the rare cases of yeast cular weight of 103105 kDa with a pH optimum of
strains that secrete enzymes capable of hydrolyzing 5.05.5 (5.56.5 in the presence of bovine serum albu-
-1,6-glucosidic bonds. Normally, therefore, if min [BSA]). The pI values reported in the literature
branched dextrins are present in brewers wort, they vary from 4.2 to 4.6.
will be found in the nal beer. In beer, dextrins con- The gene-encoding limit dextrinase has been iso-
tribute to caloric value and may also impart fullness lated (11). In an independent study, the cDNA isolated
or body (8). In contrast, during fermentation of distil- by Burton et al. (11) has been used to show that there is
lers wort, dextrin degradation may continue because only one limit dextrinase gene in barley and that the
the wort is not boiled and therefore malt enzymes that gene is located on the long arm of chromosome 4H,
survive mashing temperatures can remain active. where it has now been mapped [C-D Li, X-Q Zhang,
However, dextrins remaining after fermentation will RCM Lance, LC MacLeod, GB Fincher, P Langridge,
contribute nothing to the new make distilled spirit. unpublished data; cited by Burton et al. (11)].
Therefore, the requirements for limit dextrinase activ- Molecular techniques have been used to establish a
ity differ between brewers and distillers. A brewer may clear distinction between strictly -1,6-specic de-
desire only limited activity of limit dextrinase to ensure branching enzymes, such as limit dextrinase, and
that adequate fermentable sugars are produced during dual-bond type enzymes that attack -1,4-glucosidic
mashing (but sufcient branched dextrins remain), and -1,6-glucosidic bonds (12).
whereas a distiller would wish for complete hydrolysis When amino acid sequences of debranching
of -1,6-glucosidic bonds so that the greatest possible enzymes were examined for phylogenic relatedness
amount of fermentable sugar is available to yeast (9). using the PileUp program (13), two distinct groups
were distinguished (11). Debranching enzymes that
attacked pullulan were grouped together, whereas the
III. PROPERTIES OF LIMIT DEXTRINASE isoamylase type fell into a second group.
PROTEIN Partial amino acid sequence analysis provides evi-
dence that barley limit dextrinase belongs to the -
The properties of limit dextrinase have been thor- amylase family of multidomain amylolytic enzymes
oughly reviewed by Stenholm (10). In summary, the with a characteristic =8 -barrel catalytic domain
enzyme from malted barley is reported to have a mole- (12, 14, 15).
David Johnston
U.S. Department of Agriculture, Wyndmoor, Pennsylvania, U.S.A.
Action Systematic name Trivial name(s) EC No. Substrate specicity Linkage hydrolyzed Main products formed
Endo 1,4--D-glucan-4- Cellulase, endoglucanase 3.2.1.4 Cellulose, 1,3-1,4-- 1,4- 1,4--dextrins, mixed 1,3-1,4--
glucanohydrolase (EG) glucans dextrins
1,3-1,4--D-glucan- Lichenase, -glucanase 3.2.1.73 Lichenin, 1,3-1,4-- Next 1,4- after 1,3- Mixed 1,3-1,4--dextrins
4-glucanohydrolase glucans -linkage
1,3--D-glucan-3/4- Laminarinase 3.2.1.6 Laminarin, 1,3-1,4- Next 1,4-- or 1,3-- Mixed 1,3-1,4--dextrins
glucanohydrolase -glucans after 1,3--linkage
1,3--D-glucan-3- Laminarinase 3.2.1.39 Laminarin, pachyman, 1,3- 1,3--dextrins
glucanohydrolase (1,3-1,4--glucans) (1,4--dextrins)
Exo 1,4--D-glucan- Cellobiohydrolase 3.2.1.91 Cellulose, 1,3-1,4-- 1,4- Cellobiose
cellobiohydrolase (CBH) glucans
1,4--D-glucan- Exoglucanase 3.2.1.74 1,4--glucans 1,4- Glucose
glucohydrolase
-Glucosidase Cellobiase 3.2.1.21 Wide variety of -D- 1,4- Glucose
glucosides 1,3-
1,6-
been presented which would show that CBDs have one this enzyme a much more open active site (Fig. 5). The
active role in hydrolysis of the crystalline cellulose, e.g., tunnel shape of the active site in the CBH I, as well as
breaking down the hydrogen bonds. CBDs have been that found in other CBHs, gives a structural explana-
reported to lower the apparent Km , but also to reduce tion for the action of the enzyme. It is thought that the
the apparent kcat probably by slowing down the mobi- long tunnel keeps the enzyme bound to the same cel-
lity of the enzyme (24). lulose chain by multiple interactions as it progressively
Owing to several three-dimensional protein struc- moves along the chain hydrolyzing cellobiose units.
tures of cellulases that are available together with The open active site groove found in EGs allows
detailed biochemical characterization, many special them to hydrolyze in the middle of cellulose chains
structural features of cellulases have been revealed. (2527). It has been speculated that cellulases can dis-
For example the EG I and the CBH I of T. reesei play more or less a continuum of structures with active
have a sequence identity of 45%. Comparison of sites covered to different degrees, giving more endo-
their structures shows that the active site of CBH I is like or exo-like enzymes, making the division
located in a tunnel formed by surface loops, while unclear as mentioned earlier. The degree of endo- or
many of these loops are missing in EG I, which gives exo-character can be determined by either the pre-
Figure 5 Comparison of the structure of endo- and exoglucanases. (A) Core domain from CBH I; (B) core domain of EG I from
T. reesei.
b-Glucosidase
Asim Esen
Virginia Polytechnic Institute and State University, Blacksburg, Virginia, U.S.A.
In the glycosylation step, the nucleophilic glutamate the nucleophilic glutamate residue. In all structures
residue in the motif YI/VTENG attacks at the anome- except myrosinase, a -S-glucosidase, the residues of
ric carbon (C1 ) of the substrate and forms a covalent the TFNEP and I/VTENG motifs are involved in gly-
glycosyl-enzyme intermediate with concomitant release cone binding and catalysis within a crater-shaped
of the aglycone after protonation of the glycosidic oxy- active site (31). The two catalytic glutamic acid resi-
gen by the acid catalyst glutamic acid residue in the dues (i.e., the nucleophile and the acid/base catalyst)
motif (T(F/L/M)NEP (29, 30). In the deglycosylation are positioned within the active site at expected dis-
step, the second catalytic glutamate residue, now an tances ( 5:5 A) (Fig. 2). In myrosinases, the motif
anion and a base catalyst, removes a proton from that is homologous to TFNEP of -O-glucosidases is
water, and the resulting OH group performs a nucleo- TINQL, in which the acid/base catalyst glutamic acid
philic attack on the covalent bond between the glycone residue is replaced by a glutamine residue. There is no
and the enzyme, releasing the glycone and regenerating need for protonic assistance for aglycone departure in
Table 1 Puricationa of the Sorghum -Glucosidase Isozyme Dhr2 after Expression in E. coli
b-D-Fructofuranoside Fructohydrolase
Laura Cantarella
University of Cassino, Cassino, Italy
Francesco Alfaniy and Maria Cantarella
University of LAquila, LAquila, Italy
y
Deceased.
F puried MW MW Protein
Source [pl] Carbohydrate Form (kDa) (kDa) band Other isoforms [pI]
Arabidopsis thaliana L., mature green acid [4.75] positive monomer 58 52 4 acid [4.65; 4.70; 4.85;
leaves (9) 4.95] and 1 cw [9.0]
Avena sativa, oat seedling (5) acid [8.6] 59 single
acid [4.4] 108 single
Cichorium intybus L., roots (10) neutral tetramer 260 65 single multiple peaks
Daucus carota L., cell culture (12) neutral octamer 456 57 single multiple peaks
alkaline tetramer 504 126 single
Seed seedling (23, 24) acid [5.7] positivea monomer 68 68, 43, 25 3 bands acid [3.8] cw bound
Glycine max, sprouting hypocotyl (6) alkaline negativeb tetramer 240 58 single acid
Hordeum vulgare L., leaves (34) acid positiveb monomer 64 64 single
multimer 116 multi
multimer 155 multi
Lycopersicon esculentum, pericarp acid monomer 52 52 single 2
tissue (17)
Prunus avium L., mesocarp (35) acid positiveb multimeric 400 63 single 1
Solanum tuberum, tuber (36) acid 10.9%b dimer 60 30 single
Sugar cane juice (37) neutral 22 monomer 15
dimer 35
tetramer 66
decamer 160
Vicia faba L., cotyledon (8) alkaline [5.2] homotetramer 238 53.4 acid
a
N-glycosylated with high-mannose-containing and high-xylose-containing complex glycans.
b
Is a glycoprotein as indicated by positive reaction with periodic acidSchiff reagent and its ability to bind ConA.
Protein Other
Source Location Carbohydrate Form MW (kDa) MW (kDa) band isoforms
Avena sativa cv. (5) soluble acid 2.4 4.5 2.9 Sta Inu 4.5
soluble acid 6.7 5.0 17 Inu 5.0
Beta vulgaris L. (11) soluble acid 3.8 5.0
alkaline 33.3
acid NaClreleased 0.56
acid EDTAreleased 4.2
extracellular acid I 12.3
extracellular acid II 4.4 4.8a 7.0
Cicer arietinum L. (51) neutral 14.2
Cichorium intybus (10) neutral 1020 Raf Cel, Inu, Kes, Mal, Nys, 7.07.5
Tre
Cucumis melo L. (7) soluble acid 2.2 Raf, Sta
cw acid EDTAreleased 3.2
cw acid NaClreleased 2.2
soluble neutral 10 6.57.0
Daucus carota L. (12) neutral 18.7 7.5 Raf
14.3 6.8 Sta
alkaline 20.7 7.5 Raf, Sta
15.1 8
Daucus carota L. (24) soluble acid 5 Raf Inu 5
Fragaria ananassa (14) soluble acid 3.5 Raf, Sta 4.6
cw NaClreleased 3.7 5.0
cw EDTAreleased 4.4 4.2
Glycine max L. (52) alkaline 10 Sta, Raf Cel, Gen, Mal, Tur, Lac,
Mele, Tre, -CH3 -D-Glc,
-CH3 -D-Glc
(6) alkaline 10 Cel, Lac, Mal, Raf 7.07.6
Lycopersicon soluble acid 7.9 9.23b 4.5
Esculentum (16) soluble acid 7.7 9.08b
soluble acid 7.4 8.42b
Oryza sativa (26) alkaline 70.1 Raf, Mal 7.0
soluble acid 0.9 4.7 Raf Mal 3.54.0
soluble acid 12.1 30.2 Raf Mal 5.0
acid cw- bound 4.3 Mal 4.5
Phaseolus vulgaris (53) alkaline 8.9 7.7
soluble acid 2.4 5 14 [5.0] Sta, Kes Mele
Prunus avium L. (35) soluble acid 4 Sta Tre, Mal, Lac, Mel, Mele
Ricinus communis (25) soluble acid 8.7 3.5 Raf Mele, -Gal, - and -
Glc
Solanum tuberum (36) soluble acid 16 4.7 5.0
Sugar cane (37) soluble acid 2.8 2.7c
neutral 0.32 2.8c
Vicia faba L. (8) alkaline 10.1 Lac, Mal, Raf, Sta, Tre 0.58b 7.4
Wheat (27) soluble acid 3.5 Raf Mel, Mele
cw acid 1.7
a
mole/h.
b
mole/min.
c
mole sucrose hydrolyzed/h/mg protein.
d
Cellobiose (cel), -galactosides (-Gal), gentiobiose (Gen), - and -glucosides (--Glc), -methyl-D-glucopyranoside (-CH3 -D-Glc), -
methyl-D-glucopyranoside (-CH3 -D-Glc), inulin (Inu), 1-kestose (Kes), lactose (Lac), maltose (Mal), melibiose (Mel), melezitose (Mele), 1, 1-nystose
(Nys), rafnose (Raf); stachyose (Sta), turanose (Tur), , -trehalose (Tre).
Optimal Optimum pH
KmSuc KmRaf d
Vmax temperature
Source Form (mM) (mM) Nonsubstratesc (mM min1 ) ( C) Activity Stability
Arthrobacter sp. K-1a (38) 9.1 15.1 Pal, Inul, Cel, Mal 100% (Suc) 55 6.56.8 5.510
49% (Raf)
Aspergillus niger (40) Specic F 30160b Mal, Inu, Mele 6.0
Nonspecic F 40 Mal, Mele 5.0
Candida utilis (41) 11 150 333 (Suc) 5.5 3.06.0
215 (Raf)
Kluyveromyces fragilis (54) -Mercaptoethanolreleased 13.6 46.1 5.0
Pichia anomala (42) 16 82 Inu, Mele 38 4.5 4.56.5
Pseudomonas uorescens (19) Internal 90 6.5
Saccharomyces rouxii (55) Internal 83 5.5 4.57.5
Saccharomyces cerevisiae (2) Internal 25 150 3.55.5 6.09.0
External 26 15 3.55.5 3.07.5
a
Other substrates: erlose, fructosylxyloside, neokestose, lactosylfructoside, stachyose.
b
Non-Michaelis-Menten kinetics.
c
Cellobiose (Cel), inulin (Inu), inulobiose (Inul), maltose (Mal), melezitose (Mele), palatinose (Pal).
d
Rafnose (Raf), sucrose (Suc).
Source F puried AgNO3 CaCl2 CoCl2 CuSO4 HgCl2 I2 KCl MgCl2 MnCl2 NaCl NH4 Cl ZnCl2 MgSO4 EDTA Aniline Fructose Glucose Tris
Arthrobacter sp.K-1
(38)
Aspergillus oryzae (58)
Kluyveromyces fragilis
(54)
Avena sativa cv., oat soluble acid
seedlings (5) soluble acid
Cichorium intybus, neutral
chicory roots (10)
Cucumis melo L., soluble acid
mesocarp (7) cw EDTAreleased
cw NaClreleased
soluble neutral
Daucus carota L., neutral
cell culture (12) alkaline
Fragaria ananassa, soluble acid
fruit (14) cw NaClreleased
cw EDTAreleased
Glycine max L., alkaline
nodule (52)
Prunus avium L., soluble acid
mesocarp (35)
Wheat, coleoptiles soluble acid
(27)
Assay
conditons
ery, R, are listed in Table 8 for F from some micro- formed and is discarded. The supernatant, containing
organisms and plants. F, is adjusted to pH 7.3 with a Trizma base solution.
The procedure of Chu et al. (43) for internal F 5. The supernatant is loaded onto a Whatman
from S. cerevisiae, leading to a 19,390-fold purication microgranular DE 52 column (2 18 cm). This is
and 19% activity recovery, is hereafter outlined. All developed with 1 L linear gradient of 50350 mM
operations are carried out at 05 C. NaCl in buffer 1. External F is eluted at 0:1 M
1. Yeast paste is suspended in equal weight of 50 NaCl, while the internal form is eluted at 0.2 M
mM Tris-HCl pH 7.5, with 5 mM MgCl2 , 40 mM NaCl (R 54; P 1:5).
MSH, 1 mM phenylmethylsulfonyluoride (PMSF, 6. Fractions exhibiting internal F activity are
a protease inhibitor), and broken in a Bead-Beater. pooled and concentrated by precipitation with solid
Cell lysate is centrifuged 20 min at 30,000g (R 100; (NH4 2 SO4 up to 80% saturation (R 35;
P 1). P 5024).
2. 1% streptomycin sulfate is added to the super- 7. The precipitate is dissolved in 12 mL buffer 1
natant and after 1 h in an ice bath; nucleic acids and with 0.1 M NaCl and loaded onto a Ultrogel AcA 44
ribosomes are removed by centrifugation (R 98; column (1:5 98 cm) equilibrated with the same buf-
P 1:04). fer. The F fractions are pooled and added with
3. A precipitate is obtained from streptomycin NH4 2 SO4 to a nal concentration of 20%, then chro-
supernatant at 4085% (NH4 2 SO4 saturation and is matographed on a 1 4 cm column of phenyl-
then dissolved in 10 mM Tris-HCl buffer, pH 7.3 (buf- Sepharose CL-4B equilibrated with buffer 1, plus
fer 1) containing 1 mM MSH and 0.5 mM PMSF, 20% (NH4 2 SO4 . A linear gradient, 200%
and dialyzed thoroughly against this buffer. (NH4 2 SO4 in buffer 1, is applied for eluting internal
4. 2 M acetic acid solution is added to the dialy- F in a pure form at 5% (NH4 2 SO4 (R 22;
sate to bring the pH to 4.9. A proteic precipitate is P 19,610).
b-Galactosidase
Raymond R. Mahoney
University of Massachusetts, Amherst, Massachusetts, U.S.A.
used much compared to chemical estimation of the analyzed for glucose, preferably by the use of glucose
products. oxidase and peroxidase in conjunction with a dye such
Synthetic substrates such as o-nitrophenyl -D- as o-dianisidine [Eqs. (2) and (3)].
galactopyranoside (ONPG) and p-nitrophenyl -D-
galactopyranoside (PNPG) are usually hydrolyzed glucose
D-glucose+O2 ! H2 O2 D-gluconolactone
faster than lactose and provide an easy, convenient, oxidase
colorimetric assay since the nitrophenol released
absorbs in the range 400420 nm when in alkaline 2
solution.
peroxidase
For detection of very low levels of activity, uori- H2 O2 o-dianisidine ! oxidized o-dianisdine
metric methods of assay can be used with substrates
such as 4-methylumbelliferone -D-galactoside (8). max 436 nm 3
The conditions for assay vary with the source of the
enzyme and the substrate; representative examples are Alternatively, the galactose produced can be esti-
given below. mated indirectly by use of galactose dehydrogenase
[Eq. (4)].
A. Assay of b-Galactosidase Activity on Lactose
with Neutral-pH Enzyme galactose
D-galactose NAD !
dehydrogenase
4
Enzyme 100 L ( 30 U/mL) is added to 4 mL of 5%
lactose in 0.1 M potassium phosphate buffer, pH 6.6, D-galactonolactone NADH H
containing 3.2 mM MgCl2 , at 30 C, or to 4 mL milk at
the same temperature. The reaction is stopped after 10 The increase in absorbance at 340 nm is a measure
min by adding 0.1 mL of 4 N HCl and then neutralized of the NADH produced and of the lactose hydro-
by adding 0.1 mL of 6.4 N NaOH. The solution is then lyzed.
Source: Ref. 26
Pectic Polysaccharides
soybeans). This is also the case for the amount and apiose, aceric acid, KDO, and DHA) in a dened and
distribution of methyl esters present, since for apples rather conserved structure (Fig. 1). This structural ele-
various populations of xylogalacturonans with a rather ment is involved in the crosslinking of two pectin mole-
distinctive methyl ester level have been reported (7). cules within the cell wall through a borate diester (11,
12). No enzymes able to degrade the complex HBG
structure have been described in detail.
IV. HIGHLY BRANCHED
GALACTURONAN (HBG) OR
RHAMNOGALACTURONAN II V. RHAMNOGALACTURONAN
HBG or RG-II consists of a backbone of about nine - Rhamnogalacturonan has a backbone of up to 300
(1 ! 4)-linked galacturonosyl residues carrying four repeating units of alternating -(1 ! 2)-linked rham-
side chains containing a number of rare sugars (e.g., nosyl and -(1 ! 4)-linked galacturonosyluronic acid
Source: Ref. 8.
residues, which may be methyl esteried and/or acety- RG segments. A common feature of these RG struc-
lated (Fig. 4) (7, 13, 14). Depending on the origin of the tures is that they are also (highly) substituted with
cell walls examined (type of cell wall, tissue, plant seg- acetyl groups, either attached to O-2 or to O-3 or to
ment, etc.), the length of the side chains attached to both O-2 and O-3 of the galacturonic acid moieties (7,
2080% of the rhamnose residues can vary from one 14) which inhibits RG-hydrolase and -lyase (see Chap.
single galactose residue up to chains of 50 residues or 73).
more being composed of arabinose (e.g., sugar beet),
galactose (e.g., ax, garlic, onion) or both (e.g., soy,
potato) (7, 13). Taking all possible variations into VI. ARABINANS
account, it can be stated that rhamnogalacturonans
are a family of quite complex structures with only Arabinans are branched homoglycans mainly com-
the rhamnogalacturonan backbone in common. posed of a backbone of -L-(1 ! 5)-linked arabinosyl
Recently, a whole family of rhamnogalacturonan- residues and in native form substituted with -arabi-
specic carbohydrases has been recognized which is nofuranosyl residues at O-2 and/or O-3 position. Such
discussed in Chapter 73. Using a specic rhamnogalac- a substitution can be a single arabinosyl residue, but
turonan hydrolase, it was shown that rhamnogalactur- longer (branched) chains can also be present mostly
onan contains blocks highly ramied with long neutral one to three arabinosyl residues (see Fig. 5), resulting
sugar side chains, and in addition segments of alternat- in complex structures (see the reviews 1517 and refer-
ing rhamnose and galacturonic acid sequences with or ences herein).
without single-unit galactose substitution on O-4 of the Arabinans have been described to be present in cell
rhamnosyl residues up to 30 residues in length. Some walls from apples, sugar beet, rapeseed, carrot, cow-
rhamnogalacturonan preparations (e.g., from syca- pea, azuki- and soybean, rape, mustard, lemon, grape,
more cells) were resistant to degradation by RG hydro- cabbage, onion, potato, and others (17). Arabinans
lases (7), demonstrating the absence of low-substituted can have molecular weights up to 10 kDa and are
Figure 2 Homogalacturonan fragment of pectin. Acetyl groups may be present at O-2, O-3 or both at O-2 and O-3 for some
pectins from some sources. (From Ref. 2.)
often covalently linked to galacturonic acidrich poly- arabinofuranosyl residues attached through the O-3
mers. Although arabinans have been isolated in a pure position (20). O-6 substitution of the galactan back-
form, the extraction conditions used may have resulted bone with -galactose is also found. Figure 6 shows
in chemical (-elimination, chain peeling) or enzymatic the schematic structure of type I arabinogalactans.
degradation of any attached pectin structures (17 and Depending on the way of extraction, remnants of
references herein). the pectic backbone may still be present.
Galactans and arabinogalactans in which the galactan Arabinogalactans of type II are highly branched poly-
backbone is substituted with various amounts of ara- saccharides with ramied chains of -D-galactopyra-
binose or galactose have been found in a broad range nose residues joined by 1,3 and 1,6 linkages. They are
of higher plants (1820). Aspinall (21) classied the more common in plants than type I and have been
arabinogalactans as type I and type II arabinogalac- reported in leaves, stems, roots, oral parts, seeds,
tans, where both types may be present as side chains of and media of suspension cultured cells (20).
pectin or as a single polymer. Type II arabinogalactans In general, the interior backbone of a type II galac-
may also be present as side chains of a protein segment tan consists mainly of -(1 ! 3)-linked galactose moi-
and as such be referred to as arabinogalactan proteins eties; -(1 ! 6)-linked galactopyranosyl residues occur
(AGPs). mainly in the exterior chains which may be terminated
with an L-arabinopyranosyl residue. The (1 ! 3)-
A. Arabinogalactan Type I galactan may be branched through O-6 with arabino-
furanose residues or (to a lesser extent) with arabino-
Arabinogalactans of type I are quite common for var- pyranose units, while also longer -(1 ! 3)-linked
ious types of plant tissues (20) and consist of a arabinose chains may be present. In general, galactose
(1 ! 4)-linked linear chain of -D-glactopyranosyl is more abundantly present than arabinose, although
residues. Pure galactans have been isolated from the ratio of arabinose to galactose may differ signi-
lupine, potato, and tobacco, although type I galactans cantly. Arabinogalactans of type II may be present
usually are substituted with short chains of (1 ! 5)- within the ramied regions of pectins but are also
Figure 4 The alternating structure of rhamnogalacturonan segments of pectin. The number, length, and structure of the side
chains may vary signicantly depending on the origin of the pectin. The acetyl groups may be substituted to O-2, O-3 or both O-2
and O-3 of the galacturonic acid residues. (From Refs. 7, 13, 14.)
often reported to be linked to a hydroxyproline-rich followed by precipitation with alcohol and possibly
protein (AGP), of which gum arabic from Acacia sene- treatment with acidic alcohol to remove salts like
gal is a well-known example. The hypothetical struc- potassium, sodium, and calcium. The precise condi-
ture of a type II arabinogalactan is shown in Figure 7. tions (pH, time, temperature) will determine the mole-
It should be emphasized that many variations with cular weight, the amount of neutral sugars still present,
respect to the model shown have been reported includ- and the degree of methyl esterication (2, 5).
ing substitution with glucuronic acid (22, 23). Arabinans and (arabino)galactans are not yet being
used routinely by the food industry. However, for
research purposes, these polysaccharides are commer-
VIII. ISOLATION OF POLYSACCHARIDE
cially available and are usually obtained by alkaline
STRUCTURES
treatment at increased temperatures to enhance degra-
dation of the pectic backbone and lowering the
It should be realized that all commercially available
strength of hydrogen bridges between the neutral poly-
polymeric substrates to be used in enzyme screening
saccharides and the cellulose matrix. Further purica-
and characterization have been isolated from the com-
tion of neutral polysaccharides includes preparative
plex plant material. As mentioned above, pectins with
anion exchange chromatography to remove charged
a high galacturonic acid content (> 70%) are extracted
impurities.
under acidic conditions from suitable raw materials,
Dened polysaccharides with various degrees of given rather than extensive descriptions of the method
methyl esterication or branching are usually not com- mentioned.
mercially available and should be prepared in the
laboratory. Purication of polysaccharides is often a
A. Substrates
compromise: extensive handling in order to obtain
pure polysaccharides often changes the chemical com-
Pectic substrates can be purchased from several (spe-
position (as compared to the polysaccharide in its
cialized) suppliers (e.g., Megazyme, Ireland; Sigma-
native form), whereas mild procedures may result in
Aldrich Fine Chemicals, USA), but the origin, purity,
the presence of contaminating polysaccharides, ori-
and specications are not always clear and straightfor-
ginally linked to the polysaccharide of interest. This
ward. When monitoring crude enzyme mixtures, rela-
latter phenomenon may result in the presence of galac-
tively small amounts of other polysaccharides
turonic acids in an arabinan or galactan preparation.
accompanying the polysaccharide indicated on the
More universal assays to monitor enzyme action as
label may give erroneous results.
based on the release of reducing end groups may so
A broad range of synthetic substrates like p-nitro-
produce unreliable results because they do not given
phenyl- and p-methylumbelliferylglycosides can be
information as to which glycosidic bond in the sub-
obtained (e.g., Sigma-Aldrich) to assay glycosidase
strate is split. Analysis of the degradation product by
activity. Specic oligosaccharides may also be pur-
high-performance liquid chromatography (HPLC) or
chased from the same companies, although the choice
mass spectrometry (MS) will avoid such a confusion
and purity are limited. Alternatively, oligosaccharides
in most cases.
can be produced and puried in the laboratory (24).
Branched polysaccharides like arabinan or arabino-
IX. METHODS TO MONITOR PECTIC galactans can be rather easily de-branched by mild acid
ENZYMES treatment (17, 20, 25, 26) to obtain a more suitable
substrate for endoacting enzymes. A wide variety of
It should be realized that the summary of frequently both soluble azopolysaccharides and insoluble azur-
used methods given below is only a short list selected ine-crosslinked polysaccharides which can be used to
from the literature and from our own laboratory and is screen for specic classes of carbohydrases are com-
not complete. In most cases, adequate references are mercially available from Megazyme.
mass accuracy, is easy to operate, and requires rela- of galacturonic acid oligomers having methyl or ethyl
tively little sample preparation. For oligosaccharides, esterication, glycosidation, or amidation, and oligo-
mass determinations in the range of 4004000 are mers as present in an enzyme digest have been studied
rather easy to perform. by Maldi Tof MS and nanoelectrospray MS in detail
Figure 10 shows a Maldi Tof mass spectrum for the (61, 63, 6770).
pentamer/hexamer fraction of enzymatically degraded
soybean arabinogalactan, obtained after size exclusion
chromatography of the whole digest, as shown in C. Screening Using Plate Assays and
Figure 9. It can be seen that the different oligomers Chromogenic Substrates
consisting of arabinose and galactose can be easily
recognized. Obviously, isomers having the same mole- When huge amounts of enzyme samples have to be
cular mass cannot be distinguished from each other. If screened for specic activities (e.g., cDNA libraries),
necessary, electrospray tandem MS analysis following commonly used assays are based on the property of
HPAEC analysis may be used to verify the precise the substrate to specically bind to a coloring agent
structure of different isomers present (26). Kabel et like Congo Red (7173) or Ruthenium Red (74), or
al. (66) described the off-line coupling of HPAEC by using polysaccharides where a chromogen is cova-
and Maldi Tof MS where on-line desalting by mem- lently attached (Megazyme catalog). In specic cases,
brane suppressors, microtiterplate fraction collection, enzyme action may result in a modied substrate which
and automated sample handling followed by Maldi can specically be precipitated by metal complexation
Tof MS analysis allow determination of the molecular in a plate assay (75). Also the use of synthetic sub-
mass of individual peaks within a complex HPAEC strates in plates like p-methylumbelliferylglycosides
elution pattern rather routinely. The precise structure and p-nitrophenylglycosides to screen for glycosidases
Figure 10 MALDI-Tof MS of a size exclusion chromatography fraction containing mainly pentamers and hexamers released
from soybean pectic arabinogalactan type I by arabinogalactan-degrading enzymes. (From Ref. 26.)
Pectic Enzymes
helical structure suggests that these pectinases have grinding of intact tissues into a semiuid system in
evolved from one ancestral gene (13). The unique which cells and cell wall fragments are suspended in
structural features of each class will be described in cell liquid. From this pulp a crude juice can be
the corresponding chapters. obtained by mechanical separation methodse.g.,
Coutinho and Henrissat (14) have classied the pressing, sieving, or centrifugation. Clear juices can
pectic enzymes based on amino acid sequence be obtained from the crude juice by additional clari-
similarities with the intention that such similarities cation treatments: cloudy or pulpy juices by adjust-
reect structural features of these enzymes. The ment of the amount of neness of the pulp particles
classication can be found at the following URL: through mechanical treatments in which large and
http://afmb.cnrs-mrs.fr/CAZY/CAZY/db.html insoluble particles are removed. Clarication is
obtained through reduction of the viscosity of the
cloudy and pectin-rich crude juice by treatment with
II. ENDOGENOUS AND EXOGENOUS enzymes, which causes coagulation and precipitation
PECTIC ENZYMES IN FRUIT AND of the cloud particles. The viscosity reduction also
VEGETABLE PROCESSING allows the production of fruit juice concentrates
which, for economic and technological reasons, can
Pectic enzymes occur naturally in many fruits and be stored and transported.
vegetables (endogenous enzymes), but they are also From the pulp of apples, grapes, and pears, juice
added as processing aids (exogenous enzymes). The can be obtained quite easily by pressing. In the manu-
latter are mostly derived from food-grade fungi, e.g., facture of juices from berries or bananas, grinding
Aspergillus niger, A. aculeatus. In higher plants, parti- results in a highly viscous juice which sticks as a gel
cularly pectin methyl esterase and polygalacturonase, to the pulp particles. A mechanical separation of juice
both endo- and exoacting, are present. Pectin acetyl and pulp particles is here impossible. To facilitate this
esterase has been found in citrus fruit and more separation the pulps are treated with enzymes, which
recently also pectate lyase and rhamnogalacturonan are able to degrade the polysaccharides forming the
hydrolase have been found in plants. The endogenous gel. This enzyme treatment results in higher juice yields
enzymes are assumed to play important roles in plant and improves the color of the juice. This treatment is
development and during ripening. The modern geno- also successfully used for grapes and for stored over-
mic approach will enable us to better understand their ripe apples.
presence in plants in numerous isoforms and their roles Industrial enzymes used in these applications con-
in plant developments. Pectic enzymes are also pro- tain pectin methyl esterases, polygalacturonases, pectin
duced by many microorganisms, and here also numer- lyases, rhamnogalacturonan hydrolases, rhamno-
ous isoforms have been described. The impact of galacturonan lyases, and rhamnogalacturonan acetyl
endogenous enzymes on fruit and vegetable processing esterases in varying amounts, as well as arabinanases,
will be discussed in the chapters devoted to the various galactanases, xylanases, -1,4-glucanases, amylases,
pectic enzymes (1517). many glycosidases, and proteases.
As processing aids, pectic enzymes have predomi- The enzymes of technological relevance in the var-
nantly found applications in the fruit juice industry. ious steps of fruit juice manufacture have been identi-
Traditional processes to manufacture fruit juices in ed and we are now more and more able to formulate
essence consist of mechanical destruction of cells by enzyme cocktails tailored to specic applications. It
Pectic Esterases
Currently, 3D structures are only available for the In the premolecular biology era, numerous pectin
Erwinia chrysanthemi PME (PDB code: 1QJV) (28) methylesterases, primarily of plant origin, were char-
and the Aspergillus aculatus rhamnogalacturonan acet- acterized. In the past few years the pectin acetylesterase
ylesterase (RGAE) (PDB codes: 1DEO and 1DEX) and rhamnogalacturonan acetylesterase activities were
(29). The structure of the PME reveals a similar topol- discovered and studied, but the number of reports on
ogy as found for the pectate and pectin lyases, and these enzymes is limited (3, 1418).
polygalacturonases/rhamnogalacturonases (3033). Pectin methylesterases hydrolyze the ester bond
This topology is characterized by a right-handed par- between methanol and the carboxylic function of
allel beta-helix structure with loops protruding from galacturonic acid. The action of the enzyme thus
the helix core that form the substrate-binding cleft. results in the formation of a free carboxylic function,
The conservation of the unique topology among methanol, and H3 O . In general, the plant and bacter-
these enzymes strongly indicates that a complete set ial PMEs have pH optima between pH 6 and 8 whereas
of different pectinases have evolved from the same some fungal PMEs have pH optima between pH 4 and
ancestral gene. 6. The plant enzymes often require the addition of 0.1
The structure adopted by RGAE is completely dif- 0.2 mM NaCl for the optimal catalysis. Action of any
ferent. This enzyme folds into an == structure (29), PME on (highly) methylated pectins does not result in
which is different from the normally observed = full demethylation. Generally, the residual DM is
structure common to esterases and lipases. As for the between 20% and 30%. In part this is due to the pre-
= enzymes in the RGAE, the active-site residues are sence of acetyl groups in certain pectins (24).
Ser, Asp, and His, in the same sequential order, but the Pretreatment or simultaneous treatment with PAE
Asp and His residues are only three residues apart increases the amount of methyl groups removed, but
instead of being located on different loops (29). still not all groups will be hydrolyzed (2).
Based on the strict sequence conservation in the ester- Plant and microbial PMEs differ in their mode of
ase family 12 of Ser and His and two additional resi- action. Whereas plant enzymes generally remove
dues, Gly and Asn, which both act as hydrogen bond blocks of methyl groups on a single chain, fungal
donors to the oxyanion hole, it has been proposed that PMEs attack the methyl groups on the pectin ran-
this group of esterases be named the SGNH hydrolase domly, resulting in a random distribution of the
family (29). unmethylated galacturonic acid groups. Recently,
three different isoforms from mung bean hypoco-
tylsPME, PME, and PMEwere subjected to
B. Isoforms of Pectic Esterases a detailed analysis of their mode of action (35). This
elaborate study in which PME activities were measured
The Arabidopsis genome sequencing project has at pH 5.6 and 7.6 revealed that the three isoforms have
resulted in > 50 putative pectin methylesterases and different substrate specicities and rates. Strikingly, for
numerous pectin acetylesterases. The databases also PME at pH 5.6, a sharp drop in activity was observed
contain multiple (putative) PME sequences from when the DM decreased below 70% whereas at pH 7.6
other plants. The presence of isoforms at the genetic the reaction rate remained constant down to 40% DM.
level therefore seems a common theme among plants. For PME and PME this effect was less pronounced.
Such a large variety of pectic esterases is not present in However, PME showed at both pH values the stron-
microorganisms. At the genetic level, two PME iso- gest preference for highly methylesteried pectin. Not
forms have been identied in E. chrysanthemi (27, 34) only in this respect did the enzymes differ; in their
whereas all other microorganisms studied so far have mode of action a clear difference was also observed.
only one pme, pae, or rgae gene. The mode of action of the enzymes was simulated and
Eukaryotic enzymes and proteins are prone to N- tted to the actual distribution of methylesters after
and O-linked glycosylation. Heterogeneous glycosyla- partial hydrolysis of various DM pectins as recorded
tion can often lead to multiple isoforms of the same by NMR (35).
enzyme. There are no reports describing the analysis of For PME at both pH values the reaction was best
(heterogeneous) glycosylation patterns of pectic ester- described by a single-chain mechanism. The enzyme
ase isoforms. remains attached to one substrate molecule and con-
Polygalacturonases
Jacques A. E. Benen
Wageningen University, Wageningen, The Netherlands
Jaap Visser
Fungal Genetics and Technology Consultancy, Wageningen, The Netherlands
Ronald P. de Vries
Utrecht University, Utrecht, The Netherlands
Jaap Visser
Fungal Genetics and Technology Consultancy, Wageningen, The Netherlands
Bacterial
Bacillus subtilus A 61 7.0 66
Bacillus subtilus B 65 5.3 6.5 67
Streptomyces massasporeus B 54 (gf) 5.0 1.67 68
Streptomyces purpurascens A 62 3.9 6.5 16
495 (gf)
Thermomonospora fusca A 46 6.0 0.1 69
92 (gf)
Fungal
Aspergillus niger A 128 66.5 4.1 0.6 3
Aspergillus niger B 60 5.56 3.7 0.48 3
Aspergillus niger A 83 3.3 3.4 0.68 70
Aspergillus niger B 67 3.3 3.4 0.52 70
Aurobasidium pullulans B 105 4.04.5 71
210 (gf)
Trichodema reesei B 53 7.5 4.0 1.2 72
Plant
Daucus carota (carrot) B 94 4.7 4.2 1.33 73
a
MW is determined by SDS-PAGE, unless otherwise indicated (gf gelfiltration chromatography:
b
Determined using p-nitrophenyl--L-arabinofuranoside as a substrate.
MW
Organism Active against Action (kDa) pI pHopt Topt Km (mM) Ref.
compared to the Aspergillus endoarabinases. The and EC 3.2.1.90, respectively. The exogalactanases and
1 ! 3-exogalactanase from A. niger (11) and Irpex the 1 ! 6)-endogalactanases should also be included
lacteus (12) differ signicantly in their molecular mass in this class (EC 3.2.1.).
(66 and 51 kDa, respectively). They are both inhibited For only one type of galactanases, the 1 ! 4-
in their activity by branching of the 1 ! 3-galactan endogalactanases, have genes been characterized. On
chain. the basis of the sequence the encoded enzymes are all
For some galactanases transglycosylation activity assigned to family 53 of the glycosyl hydrolases (17).
has been observed. The 1 ! 4-endogalactanase These enzymes have a retaining mechanism, and for
from Penicillium citrinum produced several oligosac- the Pseudomonas uorescens subsp. cellulosa
charides using soybean arabinogalactan as donor, 1 ! 4-endogalactanase the catalytic residues have
and mono- or disaccharide derivatives containing - been determined (45). For the 1 ! 4-endogalacta-
galactosyl residues as acceptors (42). The 1 ! 4- nase for A. aculeatus crystals have been obtained (46),
exogalactanase from A. niger synthesized oligosacchar- but a three-dimensional structure has not yet been
ides with a higher degree of polymerization when incu- reported.
bated with short galacto-oligosaccharides (43). A
comparison was made of the transfer products of gly-
cerol formed by Bacillus subtilis 1 ! 4-exogalacta- VII. b-D-GALACTOSIDASES
nase, E. coli -galactosidase, and P. citrinum 1 ! 4-
endogalactanase (44). The -galactosidase transferred Endogalactanases enhance the release of monomeric
90% of the galactose residues to the primary hydroxyl galactose by -D-galactosidases, which remove sin-
group of glycerol, whereas the 1 ! 4-endogalacta- gle-terminal galactose residues. -Galactosidases have
nase transferred 80% of the galactose residues to the been detected in prokaryotes and lower and higher
secondary hydroxyl group. The 1 ! 4-exogalacta- eukaryotes and are active on a wide range of substrates
nase was less specic. Both products were formed, as discussed in other chapters of this book. This chap-
although the amount of transfer to the secondary ter will therefore only briey mention the function of
hydroxyl group was approximately twice as high. -galactosidases in pectin hydrolysis.
Plant -galactosidases are believed to be involved in
fruit ripening. -Galactosidase was demonstrated to
C. Classication and Mechanism increase in parallel with an increase in tissue softness
(4749). This is most likely caused by removal of the
So far, only 1 ! 4- and 1 ! 3)-endogalactanases galactose residues from pectin, thus altering the cell
have been included in the classication and nomencla- wall structure. The -galactosidase from A. niger was
ture of the Commission on Enzymes of the shown to be highly active on the side chains of sugar
International Union of Biochemistry. Like all other beet pectin (15). It enhanced the activity of the A. niger
enzymes discussed in this chapter, they belong to the 1 ! 4-endogalactanase, resulting in an efcient
group of O-glycosylases and are numbered EC 3.2.1.89 hydrolysis of the galactan chains.
Xylanolytic Enzymes
Peter Biely
Slovak Academy of Sciences, Bratislava, Slovakia
Approximate ratio
Structural type Source Nature of side chains Xyl:side chain subst.
the ease of migration of acetyl groups among the E. Xylanolytic System and Its Components
hydroxyl groups, it is difcult to determine their actual
distribution between positions 2 and 3. Softwood Depending on its complexity, the complete breakdown
xylans are similar to hardwood xylans, but contain of xylan requires the action of several enzyme compo-
-1,3-linked L-arabinofuranosyl residues instead of nents which can be considered as a xylanolytic system
acetyl groups. Xylans from grasses contain a smaller (3). As depicted in Figure 1, the more complex the
proportion of uronic acids, but are more highly xylan structure, the more components of the xylanoly-
branched with L-arabinofuranosyl residues. As indi- tic system are required. The crucial enzyme for xylan
cated in Table 1, the arabinofuranosyl residues also depolymerization is endo--1,4-xylanase (-1,4-xylan
can be linked to the main chain by -1,2-linkages, xylanohydrolase; EC 3.2.1.8, abbreviated EX). EX
often to D-xylopyranosyl residues already substituted attacks the main chain most easily and therefore
by -1,3-linked L-arabinofuranosyl residues. In cer- most rapidly at non-substituted regions, generating
eals, the ratio of Xyl : Ara is low, ranging from 1 to nonsubstituted and branched or esteried oligosac-
2.2 (44). Xylans from graminaceous plants also contain charides. The main chain substituents are liberated
25% by weight of acetyl groups and 12% of pheno- by corresponding glycosidases or esterases: -L-arabi-
lic (cinnamic) acids esterifying the C-5 position of ara- nosyl residues by -L-arabinofuranosidases (-L-ara-
binose side chains (45). In sugar beet pectins, ferulic binofuranoside arabinofuranosidase; EC 3.2.1.55), D-
acid esteries the OH-2 of L-arabinofuranose or the glucuronosyl and 4-O-methylglucuronosyl residues by
OH-6 of D-galactose (46). Ferulic acid [3(3-methoxy- -glucuronidase (EC 3.2.1.139), acetic acid, and ferulic
4-hydroxy phenyl)-2-propanoic] is the major cinnamic acid and p-coumaric acid residues by corresponding
acid found in cereals and in the Gramineae (47). In esterases (EC numbers not assigned). -Xylosidase
wheat bran, one of 150 of xylose residues is substituted (xylobiase or exo--1,4-xylanase) (-1,4-xylan xylohy-
with L-arabinose esteried with ferulic acid (45). drolase; EC 3.2.1.37) attacks xylo-oligosaccharides
Ferulic acid is synthesized in plants via the shikimic from the nonreducing end, liberating D-xylose.
pathway and is an intermediate in lignin biosynthesis The hydrolases removing xylan side chains have
(48). In some plants it has been shown to be involved in been referred to as accessory or auxiliary xylano-
the chemical linkages between xylan and lignin, con- lytic enzymes. Two categories of accessory enzymes
tributing to integrity of the plant cell walls. can be recognized: those that operate on both poly-
Intermolecular crosslinking may result from the dimer- meric and oligomeric substrates, and those that are
ization of feruoyl and p-coumaroyl residues or from active only on branched or substituted oligosacchar-
the formation of ester linkages between polysacchar- ides generated by EXs. There is usually no sharp
ides and lignin (49, 50). boundary between these groups. The enzymes operat-
EXs 10 # # # # #
-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-
EXs 11 " 2 " 2 2
j j j
1 1 1
MeGlcA MeGlcA MeGlcA
Scheme 1
glycosidic linkages in the xylan main chain that are The degradation of a cereal L-arabino-D-xylan with
near substituents, such as MeGlcA and acetic acid. two EXs of Aspergillus awamori (EXI, 39 kDa, pI 5.7
The alteration of the xylan main chain by replacing 6.7, most probably a member of family 10, and EXIII,
-1,4-linkages by -1,3-linkages, as it is in rhodyme- 26 kDa, pI 3.33.5, most probably a member of family
nan, represents a more serious steric barrier for EXs of 11) (90) is shown in Scheme 2. The cleavage site speci-
family 11 than for EXs of family 10. Consequently, cities L-arabinosyl-D-xylan resemble those on 4-O-
EXs of family 10 liberate smaller products from 4-O- methyl-D-glucurono-D-xylan. The L-arabinosyl sub-
methyl-D-glucurono-D-xylan, rhodymenan, and, with stituents represent a more serious steric hindrance for
Araf
1
j
EXI # # # # 2 # #
-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-4Xyl1-
EXIII " 3 " 3 3
j j j
1 1 1
Araf Araf Araf
Scheme 2
At sufciently high concentrations of xylotriose, xylo- cellulose and xylan (106, 107). The best-known enzyme
biose initially is generated as the only product of xylo- in this regard is endoglucanase I from Trichoderma
triose degradation. Analogous reaction pathways are reesei, a member of family 5 glycosyl hydrolases. The
also observed with longer oligosaccharides. The main enzyme catalyzes the hydrolysis of both polysacchar-
molecular and catalytic properties of EXs of families ides in the same active site (108). Although the enzyme
10 and 11 are summarized in Table 2. belongs to the same clan as EXs of family 10, the
products the enzyme generates from glucuronoxylan
and rhodymenan are more similar to those liberated
3. Xylan-Degrading Endo--1,4-Glucanases
by EXs of family 11 (87). In T. reesei the enzyme is
It has been unambiguously established that at least one coinduced as a component of the cellulolytic and not
type of endo--1,4-glucanase exhibits activity on both of the xylanolytic system (3).
Source: http://expasy.hcuge.ch/cgi-bin/list/glycosid.txt.
Scheme 4
An interesting question related to the substrate spe- a related polymeric substrate. Glucuronoxylan can
cicity of -glucuronidase is whether the enzyme is be used for the -glucuronidase assay only in the
specic for GlcA or MeGlcA. The enzyme from presence of xylan-depolymerizing enzymes which
Aspergillus niger hydrolyzed GlcA-Xyl3 about two generate aldouronic acids of the required structures
times faster than MeGlcXyl3 (172). This example sug- (190). The most common substrate for -glucuroni-
gests that the 4-O-methyl group in the acid is not deci- dase assays are linear aldouronic acids obtained by
sive for the action of the enzymes. enzymic or acid hydrolysis of glucuronoxylan. From
A surprising property of microbial -glucuronidases such aldouronic acids, liberation of MeGlcA is
is their inability to hydrolyze 4-nitrophenyl--D-glu- usually followed by determination of the reducing
curonide, a compound that could otherwise serve as power of the free uronic acid by a modication of
a convenient chromogenic substrate. The only -glu- the Somogyi copper reagent and the Nelson phos-
curonidase capable of hydrolyzing this compound was phomolybdenic acid reagent according to Milner
the snail enzyme (185). Microbial -glucuronidases and Avigad (191). The method is quite specic for
require uronic acid to be linked to at least one xylo- uronic acids and suffers only slightly from the pre-
pyranosyl residue, which can be in the form of an aryl sence of neutral reducing sugars. The assay could be
glycoside as in 4-nitrophenyl-2-O-(methyl-4-O-methyl- affected by some salts (citrate, phosphate), however.
-D-glucuronosyl)--D-xylopyranoside (188). Alternative methods include HPLC determination of
Optimal pH values for bacterial -glucuronidases free MeGlcA (173, 174) or the equivalent amount of
are in the range of 5.46.5 and in the range of 3.5 xylo-oligosaccharides (169), or GLC of trimethylsilyl
ous enzymes have modular architecture. Some of AcXEs existing as single-domain enzymes have a
AcXEs, such as those from Pseudomonas uorescens molecular mass in the range of 2348 kDa and acidic
(209) or Trichoderma reesei (221), contain cellulose- or neutral pI values. There does not seem to be a spe-
binding domains (CBDs) separated from the catalytic cial relationship between these two parameters and
domain by linker regions. Others, like AcXE from their assignment into carbohydrate esterase families
Streptomyces lividans (217), contain a xylan-binding (Table 4).
domain (77). AcXE of Cellulomonas mi is a part of Three-dimensional structures are known only for
a bifunctional enzyme which contains both CBD and AcXEs belonging to family 5. Crystallization has
XBD (82). The two-catalytic module architecture, been reported for AcXEII from P. purpurogenum
represented by various combinations of EXs linked (227) and AcXE from T. reesei (228). The protein
to AcXEs, is very common particularly among acetyl- from P. purpurogenum shows an = hydrolase fold,
xylan-degrading bifunctional enzymes of ruminal similar to that of cutinase (227). Cysteine residues form
anaerobic bacteria (212). AcXE domains of such ve S-S bridges which makes the structure very rigid
bifunctional enzymes can be found in carbohydrate and tightly packed. The substrate binding site appears
esterase families 1, 2, 3, and 4 (Table 4). With the to be small and able to accommodate only an acetyl
bifunctional EX-AcXE enzyme from Clostridium group, thus giving a relatively high substrate specicity
thermocellum (83) it has been demonstrated that the to the enzyme.
two catalytic domains act synergistically in the degra-
dation of acetylxylan. This observation again suggests E. Properties as Enzymes
that the production of an enzyme having both deb-
1. Substrate Specicity
ranching and depolymerizing activity can be a great
advantage for a microorganism competing for a Substrate specicity of AcXEs is one of the least-clar-
carbon source with microorganisms using separate ied aspects of this xylanolytic component. There is
enzyme components. very limited knowledge on the structure-function rela-
Scheme 6
B. Properties as Proteins
Xyl1-4Xyl1-4Xyl-R
3 2 Fungal -L-arabinofuranosidases occur mostly as
j ! Xyl1-4Xyl1-4Xyl-R Ara monomeric enzymes with molecular masses between
32 and 84 kDa and pI values between 3.2 and 7.5
1 (44). The majority of described fungal enzymes are
Araf products of Aspergillus spp. (146). and have pI values
3:5. Bacterial and Streptomyces -L-arabinofurano-
R 0 -Xyl1-4Xyl1-4Xyl-R sidases frequently occur as oligomers, form dimers up
32 to octamers, with subunits of molecular mass between
j ! R 0 -Xyl1-4Xyl1-4Xyl-R Ara 33 and 75 kDa. pI values were reported in the range of
3.89.0.
1 Some -L-arabinofuranosidases show modular
Araf architecture similar to other xylanolytic enzymes.
Interesting in this regard is -L-arabinofuranosidase
from T. reesei (244, 245). Two different forms of the
-L-Arabinofuranosidase is a component of hemi- enzyme have been isolated from a culture uid of the
cellulolytic systems that participates in hydrolysis of fungus. The 25-kDa form corresponds to the catalytic
plant polysaccharides built from or containing nonre- domain only (245). The 53-kDa form, the rst form of
ducing terminal -L-arabinofuranosyl residues (44, the enzyme described (244), also contains a noncataly-
238). Such polysaccharides have been recognized as tic xylan-binding domain (245). The large 53-kDa form
part of pectic substances, gum exudates, and wood can be converted to the shorter one by proteolytic
and cereal hemicelluloses (239). The pectic substances digestion (245). -L-Arabinofuranosidase from P.
mainly include L-arabinan and various types of L-ara- uorescens was reported to contain a cellulose-binding
Scheme 7
Jean-Paul Vincken
Wageningen University, Wageningen, The Netherlands
Figure 1 (A) Structure of the cellobiosyl building unit of cellulose, the xylobiosyl building unit of xylan, and a trisaccharide XG
subunit of xyloglucan. Note that glucose and xylose are very similar except for C-6. (B) Two general branching patterns
of xyloglucans, XXXG and XXGG. Small shaded circles represent acetyl groups. (C) Hypothetical (nonexisting) molecule
showing the different structural elements encountered in xyloglucans. G -D-Glcp; X -D-Xylp-1 ! 6--D-Glcp;
L -D-Galp-1 ! 2--D-Xylp-1 ! 6--D-Glcp; F -L-Fucp-1 ! 2--D-Galp-1 ! 2--D-Xylp-1 ! 6--D-Glcp; S
-L-Araf -1 ! 2--D-Xylp-1 ! 6--D-Glcp; A -D-Xylp-1 ! 6--D-Glcp-2 1--L-Araf ; B -D-Xylp-1 ! 6-
-D-G l cp-2 1--D-X ylp; C -D-X y lp-1 ! 6--D-G l cp-2 1--D-Xylp-3 1--L-Araf ; J -L-Galp-1 ! 2-
-D-Galp-1 ! 2--D-Xylp-1 ! 6--D-Glcp; T -L-Ar af -1 ! 3--L-Ar af -1 ! 2--D-X y lp-1 ! 6--D-G l c p. ( D )
Flat backbone conformation of the xyloglucan heptadecamer GXXFGXXXG [adapted from (4)]. NR, nonreducing end of
the polymer; R, reducing end. (E) Schematic representation of a part of the cellulose-xyloglucan network in the primary cell of
plants [adapted from Roberts and McCann (9)].
Enzyme Organism EC No. XG CMC Xylan MWcore Family Cat. mech. Cat. res. CBD? Anchors?
determine whether these enzymes have xyloglucanase means that xyloglucanase activity may not be reserved
activity. Probably, it is important that the substrate- to retaining enzymes.
binding groove is lined with the appropriate amino In the previous paragraphs it was shown that the
acid residues at certain positions; this may determine structure of xyloglucan determines the properties of
whether the xylosyl side chains on the glucan backbone these molecules, such as binding to cellulose surfaces.
are tolerated in the groove. Also, xylosyl groups may Next to this, the structure also greatly inuences the
be involved in the molecular recognition of the sub- enzymic digestibility; i.e., the nature of the sidechains
strate by xyloglucanases. From the examples above it determines how fast xyloglucan is degraded by endo-
is clear that the family classication does not have a glucanases. It is difcult to relate the tolerance to par-
predictive value with respect to the substrate specicity ticular side chains to individual endoglucanases,
of endoglucanases. The xyloglucanase activity of each because most studies have been carried out with rather
individual endoglucanase should be veried experi- crude mixtures of enzymes from Trichoderma.
mentally. There are a number of reports showing In Figure 2 a number of rules that seem to have a
that there are enzymes in plant extracts that have xylo- general validity are summarized. The presence of fuco-
glucanase activity. Possibly, some of these xylogluca- sylated side chains can hinder cleavage by the endoglu-
nases will be classied in family 9. At this moment, canase (17). Two adjacent fucosylated side chains as in
there is no experimental evidence supporting this XFFGXXXG can give partial resistance to endogluca-
remark. It is interesting to note that the enzymes of nase, meaning that prolonged incubations are needed
family 9 act with an inverting mechanism. This to generate the constituent oligosaccharides XFFG
Figure 2 Overview of xyloglucan structures, which (a) are easily cleaved by endoglucanases (solid arrow), (b) are partially
resistant to endoglucanase (thin arrow), and (c) are completely resistant to endoglucanases (thin arrow with cross). For details on
xyloglucan structure, see Figure 1C. Small shaded circle on open box (bottom right structure) represents an acetyl group attached
to a glucosyl residue.
Figure 3 Schematic illustration of mode of action of endoglucanases (EGI and EGIII) from Trichoderma reesei on two different
types of xyloglucan (XXXG and XXGG). For details on xyloglucan structure, see Figure 1C.
Figure 4 Schematic illustration of three reactions that can be catalyzed by XET. NR, nonreducing end of the polymer; R,
reducing end. Vertical dotted line represents site of cleavage by XET.
branched galacturonans are shown in boldface. lyases (L1 and L4). Again, at present the plant
Glycosyl hydrolase family 28 (GH28) is a particularly enzymes have not been characterized. Despite the
large family, comprising enzymes with a very diver- large differences in substrate specicity (homogalac-
gent substrate specicity. Hydrolases with activity turonan, XGA, and RG) and catalytic mechanism
toward homogalacturonan, XGA, and RG are pre- (hydrolase versus lyase), it turns out that the three-
sent within this family. Further, GH28 contains dimensional structures of pectin-degrading enzymes
many putative proteins from plant origin. An inter- are very similar (2). All the enzymes investigated so
esting observation is that Arabidopsis members of far possess the characteristic right-handed parallel -
GH28 outnumber those derived from one fungus helical structure, which has been discussed for homo-
(e.g., Aspergillus niger). It remains to be established galacturonan-degrading enzymes elsewhere in this
whether these Arabidopsis proteins represent an array book. In this chapter, we will focus on microbial
of different pectinases with activity toward the var- enzymes, because there are currently no biochemical
ious galacturonan-containing polymers in plants. data available for the plant enzymes. XGA-degrading
Surprisingly, galacturonan lyases are distributed enzymes will be discussed rst, followed by RG-
over as many as seven different families (L1L4 and degrading enzymes. Subsequently, a few words will
L9L11). Also in this case the Arabidopsis genome- be said on the detection of enzyme activity toward
sequencing effort has provided clues on the possible heterogalacturonans. Finally, applications with rele-
existence of an abundance of plant galacturonan vance to the food industry will be discussed.
Figure 1 Schematic illustration of xylogalacturonan degradation by exopolygalacturonase (closed arrowheads) and endoxylo-
galacturonan hydrolase (open arrowheads). Both enzymes are hindered by the presence of methoxyl groups. Numbers indicate
1st, 2nd, 3rd, etc., cleavage of the substrate by the enzyme. Open circles represent -D-GalpA residues, shaded circles -D-Xylp
residues, and small closed triangles methoxyl groups. NR indicates the nonreducing end; R, the reducing end.
Figure 2 The three-dimensional structure of A. aculeatus endoRGH. (A1) Cartoon indicating the four parallel sheets; the
putative catalytic amino acids, Asp156 and Asp180, are indicated as black sticks. (A2) View into the superhelix of the molecule
at a 90 rotation to A1. (B1) Space-lling model of endoRGH (similar view as with cartoon A1) clearly showing the substrate-
binding groove of the enzyme in which the catalytic residues (in black, only one visible) are located. The illustration also shows
that endoRGH is heavily glycosylated. Amino acids with O-linked glycosylation (Thr, Ser) and those with N-linked glycosylation
(Asn) are shown in black. (B2) View of endoRGH at a 180 rotation to B1 to show that glycosylation is conned to one side of
the molecule.
Figure 3 Schematic illustration of enzymes involved in the degradation of rhamnogalacturonan. Both endoRGH and endoRGL
are hindered by the presence of acetyl groups. Numbers indicate 1st, 2nd, 3rd, etc., cleavage of the substrate by the enzyme. Open
circles represent -D-GalpA residues, closed boxes -L-Rhap residues, shaded triangles -D-Galp, and small shaded circles acetyl
groups. NR indicates the nonreducing end; R, the reducing end.
Figure 4 Mode of action of Aspergillus aculeatus and Botryotinia fuckeliana endoRGH and Aspergillus aculeatus endoRGL on
degalactosylated rhamnogalacturonan oligosaccharides of various length. Thick arrows indicate preferred cleavage sites; less
preferred sites are indicated with thinner arrows. Closed arrowheads marked with 2 indicate the second cleavage site of the
enzyme. Open circles represent -D-GalpA residues; and closed boxes -L-Rhap residues.
Table 1 Relative Abundance of 1 ! 3; 1 ! 4)--Glucans in Cereal Grains and in Cell Walls from the Starchy Endosperm
barley 1 ! 3; 1 ! 4)--glucans have weight average grain, where a signicant proportion of total endo-
molecular weights in the range 200,000300,000 (12, sperm glucosyl residues are embodied in cell wall
13). These values correspond to polysaccharides con- 1 ! 3; 1 ! 4)--glucan (15), it might be expected
taining 12001800 glucosyl residues. that a battery of enzymes would catalyze the complete
The high viscosities of cereal 1 ! 3; 1 ! 4)--glu- conversion of the polysaccharide to glucose, which
can solutions can be attributed not only to the high could be translocated as an energy source to the devel-
molecular masses of the polysaccharides, but also to oping seedling. In elongating coleoptiles, only partial
their high degree of molecular asymmetry (Fig. 2). hydrolysis of the 1 ! 3; 1 ! 4)--glucan might be
Woodward et al. (9) showed that the water-soluble required to loosen crosslinking polysaccharides in
barley 1 ! 3; 1 ! 4)--glucan has an axial ratio the wall during turgor-driven cell expansion.
(average length:average width) of 100. It might be
anticipated that such long, extended molecules would A. Enzymes That Release Large Fragments of
aggregate and precipitate from aqueous solution. 1 ! 3; 1 ! 4)-b-Glucans
However, the irregular spacing of the (1 ! 3)-linkages
along the polysaccharide chain introduces irregularly The rst step involves a hypothetical endoacting
spaced links that preclude extensive intermolecular enzyme (designated Endo-X in Fig. 3) that releases
associations and prevent precipitation from solution. polymeric wall 1 ! 3; 1 ! 4)--glucan into solution,
Thus, the arrangement of the (1 ! 3)- and (1 ! 4)-- without hydrolyzing it to low-molecular-mass oligo-
glucosyl residues along the polysaccharide chain saccharides. An enzyme with this action pattern in
accounts for the solubility of large 1 ! 3; 1 ! 4)-- germinated barley grain has been the subject of persis-
glucan molecules in water, their extended exible chain tent reports in the malting and brewing literature, and
conformations, and their tendency to form aqueous has been given the name -glucan solubilase (16). The
solutions of high viscosity (6). It is precisely these enzyme has not yet been puried for detailed charac-
properties that enable relatively low abundance terization, and in one report it was suggested that the
1 ! 3; 1 ! 4)--glucans to exert such a large inu- enzyme originated from microorganisms that reside on
ence on cereal grain utilization in food and related the surface of the barley grain (17). However, another
industries. 1 ! 3; 1 ! 4)--glucan endohydrolase extracted
from maize coleoptiles also releases relatively high-
III. OVERVIEW OF ENZYMIC molecular-mass 1 ! 3; 1 ! 4)--glucan from cell
HYDROLYSIS OF 1 ! 3; 1 ! 4)-b- walls (18). It is unlikely that the maize coleoptiles
GLUCANS would be heavily contaminated with microorganisms.
Furthermore, the NH2 -terminal sequence of the maize
Enzymes that mediate the depolymerization of wall- endohydrolase (19) matches the sequences of numer-
bound 1 ! 3; 1 ! 4)--glucans and their degradation ous cDNAs in the rice and maize EST databases, and
products have been mostly extracted from germinated is clearly of plant origin.
barley grain, young barley seedlings, or maize or barley
coleoptiles. Several different types of enzymes would be B. 1 ! 3; 1 ! 4)-b-Glucan Endohydrolases
required to completely depolymerize 1 ! 3; 1 ! 4)--
glucans to glucose. Candidate enzymes and their action Much more precise information is available for the
patterns are summarized in Figure 3. In germinated 1 ! 3; 1 ! 4)--glucan endohydrolases of the EC
as does their ability to very rapidly reduce the viscosity enzyme-substrate interactions by diffusing nonhydro-
of (1 ! 3; 1 ! 4--glucan solutions (20). Chen et al. lyzable S-glycoside substrate analogs (44) into barley
(33) used proton NMR to show that the anomeric (1 ! 3; 1 ! 4--glucan endohydrolase crystals, in
conguration of released products is retained during the expectation that the substrate analogs will be
hydrolysis. bound but that their S-glycosidic linkages will not
be cleaved by the enzyme. Thus, the substrate analogs
B. Three-Dimensional Structure should remain associated with the enzyme during the
generation of x-ray diffraction data.
The 3D structure of barley (1 ! 3; 1 ! 4--glucan Measurements of the substrate binding cleft that
endohydrolase isoenzyme EII has been determined at runs across the surface of the barley (1 ! 3; 1 ! 4-
2:22.3 A resolution by x-ray crystallography (34). -glucan endohydrolase indicate that it is long enough
The enzyme adopts a (=8 TIM barrel fold. The most to accommodate up to eight glucosyl residues of the
striking feature of the enzyme is a deep substrate-bind- (1 ! 3; 1 ! 4--glucan substrate (34).
ing cleft that extends across its surface (Fig. 4). The
open cleft is consistent with the enzymes endohydro-
VI. CEREAL b-GLUCOSIDASES
lytic pattern, because it would allow the enzyme to
bind its substrate at essentially any position along the A. Substrate Specicity and Action Pattern
polysaccharide backbone for the hydrolysis of internal
glycosidic linkages. Two -glucosidases puried from extracts of germi-
nated barley grain have been designated isoenzymes
C. Substrate Binding I and II (26, 27, 45), and their properties are
shown in Table 3. The enzymes are members of the
The precise details of substrate binding and the iden- family 1 group of glycoside hydrolases (31), although
tity of amino acid residues involved have not been it is difcult to classify them into existing Enzyme
dened. When relatively long (1 ! 3; 1 ! 4--oligo- Commission groups. The enzymes hydrolyze 4NPG
glucoside substrates are allowed to diffuse into and are therefore referred to as -glucosidases of the
crystals, they clearly bind to the enzyme. However, EC 3.2.1.21 group (26, 27, 37). However, more detailed
they are subsequently hydrolyzed and products studies on their substrate specicities show that the
diffuse away, even at low temperatures and sub- enzymes hydrolyse a range of di- and oligosacharides.
optimal pHs, and it has therefore not been possible Highest activity is observed for cellodextrin substrates,
to generate diffraction data for the enzyme-substrate for which the rate of hydrolysis increases as the length
complex (M Hrmova, JN Varghese, GB Fincher, of the substrate increases from cellotriose to cellohex-
unpublished). We are now attempting to dene aose (Fig. 5) (26, 27). In all cases, the barley enzymes
remove single glucose units from the nonreducing ends glucosidases are more typical of polysaccharide exo-
of the substrates, and anomeric conguration is hydrolases of the (1 ! 4)--glucan glucohydrolase
retained during hydrolysis (27). The barley -glucosi- group (EC 3.2.1.74). The uncertainties associated
dases also catalyze glycosyl transfer reactions, through with the classication of the barley enzymes are
which higher oligosaccharides containing (1 ! 3)-, acknowledged by Hrmova et al. (37), who propose
1 ! 4)-, and (1 ! 6)--linkages are generated during that the enzyme be referred to as a -glucosidase/-glu-
hydrolysis of 4NPG (37). The increased activity of the can exohydrolase until its native substrate in germi-
enzymes as chain length of cello-oligosaccharides nated barley grains has been identied unequivocally.
increase (Fig. 5) is suggestive of a polysaccharide exo- The preference of barley grain -glucosidases for
hydrolase rather than an enzyme that is specic for (1 ! 4)--oligoglucosides of increasing chain length
low-molecular-mass -oligoglucosides (27). Indeed, suggests that it might have an extended substrate-bind-
the action patterns and specicities of the barley - ing region.
Figure 6 Three-dimensional modeled structure of the barley -glucosidase. Stereoview space-lling representation of (=8
TIM barrel fold. The aromatic amino acid residues lining the surface of the active site pocket are colored in black. The drawing
was generated using RasMol (43). (From Ref. 37.)
Figure 8 Three-dimensional structure of the barley -glucan exohydrolase. Stereoview space-lling representation of the overall
structure. The (=8 TIM barrel (colored in white) and the (=6 sandwich (colored in light gray) represent domains 1 and 2,
respectively. The linker region connecting the two domains is on the left and is marked in black. Active-site region of the enzyme,
which is bounded by Trp286 and Trp434 (colored in dark gray), contains a trapped glucose moiety (in a dark gray wireframe
representation). The drawing was generated using RasMol (43). (From Ref. 35.)
Enzymology of Endo-1,4-b-Mannanases
Henrik Stalbrand
Lund University, Lund, Sweden
-Mannanases also hydrolyze heteromannans. Glycoside hydrolases like cellulases and hemicellulases
However, their activity is generally restricted by the commonly have a modular structure. The catalytic
galactosyl sidegroups and the main-chain glucose resi- modules may be attached to other module(s), such as
dues of the substrate (22). The subsite-binding proper- carbohydrate-binding modules (10). The modules are
ties may vary among enzymes of different origin. The often connected by linker peptides. Furthermore, as
major hydrolysis products from incubation of fungal reviewed (10, 45), anaerobic cellulolytic organisms
as well as bacterial -mannanase with heteromannans like Clostridium spp. frequently form large cell-asso-
are mixed oligosaccharides, mannotriose, and manno- ciated multienzyme complexes. Although several -
biose, sometimes with the production also of mannose mannanases are sole catalytic modules of 3045
(49, 54, 55), as reviewed by Viikari et al. (56). These kDa (13, 15, 18), some -mannanases are modular
products are hydrolyzed further by exohydrolases, also proteins with molecular masses from 50 kDa up to
secreted by the microorganisms. Thus, the cooperation 100 kDa (12, 14, 16, 19, 20, 23, 25, 65). Examples of
of several enzymatic activities is needed for the degra- modular -mannanases from anaerobes are enzymes
dation of heteromannans. In addition to -mannanase which carry an additional endoglucanase catalytic
(EC 3.2.1.78), -galactosidase (EC 3.2.1.22) and - module and cellulose binding modules (12) or putative
mannosidase (EC 3.2.1.25) are also needed for the protein binding modules which may be involved in the
complete hydrolysis of galactoglucomannan into formation of multienzyme complexes (19, 65).
monomeric sugars (57, 58). -Mannosidases hydrolyze Also, some aerobic organisms produce modular -
terminal nonreducing mannose residues, and -galac- mannanases (14, 16, 20). The T. reesei family 5 -man-
tosidases hydrolyze -1,6-linked galactosyl side groups nanase has a C-terminal cellulose-binding module (16,
from oligomeric and sometimes polymeric substrate 38, 66). Biely and Tenkanen (58) discussed the possi-
(5962). -Glucosidase (EC 3.2.1.21) may be needed bility that the presence of a cellulose-binding module
for the hydrolysis of non-reducing-end glucose resi- on the T. reesei -mannanase is part of the reason for
dues. For the complete hydrolysis of acetylated galac- the superior performance in bleaching experiments for
toglucomannan, acetyl esterase is also needed (63, 64). this enzyme (67). The Cellulomonas mi enzyme is the
Lysozyme
Akio Kato
Yamaguchi University, Yamaguchi, Japan
Figure 1 Effects of the temperature and pH on the relative Lysozyme consists of 129 amino acids and the molecu-
activity of lysozyme in egg white during storage. **, 5 C lar weight is 14,307 daltons. The protein contains many
(nal pH 9.0); **, 20 C (nal pH 9.3); *- - -*, 30 C more basic amino acids (11 arginines and six lysines)
(nal pH 9.4); *- - -*, 40 C (nal pH 9.5); * - *, than acidic amino acids (six aspartic acids and two glu-
stored in an atmosphere of CO2 at 30 C (nal pH 6.5). (From tamic acids), and the isoelectric point is 11.2. The
Ref. 4.) amino acid and nucleotide sequences of prelysozyme
are shown in Figure 3. The sequence 1 to 18 is the
signal peptide, and the N-terminus of mature lysozyme
cient on Gram-negative bacteria. Some attempts to is the position 1 (lysine). Lysozyme has four disulde
enhance the antimicrobial action for Gram-negative bonds: Cys6127, Cys30115, Cys6480, and Cys76
bacteria have been done by our colleagues. Ibrahim 94. Human lysozyme, which consists of 130 amino
et al. in 1991 found that the covalent attachment of acids, is very similar to hen egg white lysozyme in its
palmitic acid residues to the lysyl residues of lysozyme primary and tertiary structures and the positions of
Figure 2 Antimicrobial activity of lysozyme-dextran conjugate for ve Gram-negative bacteria. (a) V. parahaemolyticus IFO
12713; (b) E. coli IFO 12713; (c) A. hydrophila IFO 13286; (d) P. mirabilis IFO 12668; (e) K. pneumoniae IFO 14438. *, Control
(medium without lysozyme or conjugate); *, addition of 0.05% lysozyme; | addition of 0.05% lysozyme-dextran conjugate.
The abscissa is the heating time at 50 C. (From Ref. 7.)
disulde bonds. -Lactalbumin, which consists of 123 mined by x-ray crystallography (8). The ribbon draw-
amino acids, is also homologous to lysozyme in its pri- ing model is shown in Figure 4. As shown in Figure 4, it
mary and tertiary structure and in the position of di- divides the molecule into two partsa predominantly
sulde bonds. The x-ray analysis of hen egg white helical core, and an irregular -strand core. The helical
lysozyme was carried out by Blake et al. in 1965 as core consists of four -helices (515, 2535, 8899, and
the rst enzyme to have its tertiary structure deter- 108115), and the -strand core consists of three -
V. PROPERTIES AS ENZYME
Figure 5 Space-lling CPK model of hen egg white lysozyme. Left: Enzyme without substrate, showing active site crevice.
Right: Enzyme-substrate complex, with hexamer NAG substrate. (From Ref. 10.)
Ribonucleases
sis) reaction is much slower than the rst (transferase), could be isolated in relatively large amounts from
and several members of the ribonuclease A superfamily bovine pancreas, a major product of the beef industry
do not even hydrolyze the intermediate cyclic phos- (14). This enzyme has played a key role in the devel-
phates (6). This example shows that the classication opment of the study of proteins, with four Nobel
of these ribonucleases as hydrolases (EC 3.1.27) is less Prizes in chemistry (in 1972 and 1984) awarded for
correct and that the previous classication as trans- research on it. However, it was later found that ribo-
ferases (EC 2.7.7) is more appropriate. nuclease is only present in high amount in the pancreas
of ruminants and species that have a ruminantlike
digestion, and in a number of species with cecal diges-
II. IMPORTANCE OF RIBONUCLEASES tion (15, 16). For instance, the ribonuclease content in
TO QUALITY OF FOOD AND IN FOOD human pancreas is < 1% of that in bovine pancreas,
DIGESTION and in the 1970s ribonuclease disappeared again as an
important digestive pancreatic enzyme in textbooks for
Enzymes to be included as topics in a Handbook of medical students.
Food Enzymology should have either a role (or poten- Ribonuclease was considered around 1980 as a dull
tial role) in food production or processing, or be enzyme, without a relevant biological function and not
important in digestion of food components in the very interesting for chemical investigations. However,
human species, in its domesticated animals, or in bio- since then the enzyme underwent a renaissance by the
technology. In this respect the history of ribonuclease discovery of proteins with novel biological actions,
research is a disappointing one. One of the rst which after structural studies proved to be homologs
enzymes to be isolated and puried in crystalline of ribonucleases with their ribonuclease activities being
form has been ribonuclease, a digestive enzyme which a condition of these special actions (17, 18).
space-lling model of carbohydrate-free bovine ribo- (40). This would be analogous to the function of ribo-
nuclease is compared with the same model with three nucleases in the digestion of ribonucleic acid from the
biantennary complex-type oligosaccharide chains stomach microora in ruminants and herbivores with
attached at positions 22, 34, and 62, representing the ruminantlike digestion postulated by Barnard (15).
minor glycosylated component of horse ribonuclease. Ruminants and species with ruminantlike digestion
Probably the carbohydrate moieties do not occupy (camel, hippopotamus, sloth, kangaroo) have less car-
xed positions because of the presence of a number bohydrate attached to their ribonucleases. Therefore, it
of exible bonds, but the latter model shows the seems that glycosylation of pancreatic ribonuclease
increase in volume as a result of the addition of carbo- may not be advantageous for species with stomach
hydrate. fermentation. These species escaped glycosylation of
The glycosylation states of the ribonucleases inves- their ribonucleases not only by amino acid replace-
tigated were correlated with the type of fermentation ments in the protein (camel, reindeer), but probably
(stomach or cecal) used by the species involved (Fig. also by developing a much less effective carbohy-
4). It was found that species with cecal digestion, such drate-attaching enzyme system in the pancreas than
as pig, horse, and the hystricognath rodents (guinea encountered in other species. Hippopotamus ribonu-
pig and relatives), produce ribonucleases with large clease is an example of the second solution. This
carbohydrate chains attached to several positions at enzyme has four carbohydrate attachment sites, with
the surface. Therefore, it was suggested that the pre- carbohydrate attached to only one of them in only a
sence of carbohydrate protects ribonuclease from fraction of the molecules.
absorption in the gut, enabling it to be transported If glycosylation of a ribonuclease is avoided by
to the large intestine where it should hydrolyze the amino acid replacements, the enzyme will also not be
ribonucleic acid derived from the cecal microora glycosylated if its gene is expressed at other sites of the
Man Langur
polypeptide sequence of 1061 amino acids, much Moiseyev et al. (55, 56) have isolated, characterized
longer than the extracellular enzymes discussed in the and sequenced two plant ribonucleases isolated from
previous section. A similar activity is exhibited by the ginseng calluses. These enzymes, with polypeptide
already mentioned ribonuclease P (EC 3.1.26.5), of lengths of 150 residues, are homologous with cyto-
which the ribonucleic acid component is the catalyti- toxic intracellular pathogenesis-related proteins (IPR
cally active one. or PR-10) from several plants such as parsley (57)
More specialized E. coli ribonucleases are ribonu- and bean (58), and with pollen allergens from tree spe-
cleases III (EC 3.1.26.3) and H1 and H3 (EC 3.1.26.4), cies such as birch (59) and alder (60). This suggests that
which cleave double-stranded (ds) RNA molecules and the cytotoxic properties of these latter proteins are
the RNA strand of RNA-DNA hybrids, respectively. caused by their ribonuclease activities. However, this
Other, less well characterized ribonucleases in E. coli hypothesis is still being met with skepticism. In addi-
are ribonucleases IV and F (producing 3 0 -phosphate tion, this protein family does not contain any con-
termini; EC 3.1.27.7), P2, O, PIV, PC and N. served histidine residues, a universal active-site
Few intracellularly synthesized eukaryotic ribonu- residue in all other investigated ribonuclease families
cleases have been identied and characterized. Two so far investigated.
of the few examples are the 2-5A (=oligoadenylate So, although much knowledge has already been col-
with 2 0 ! 5 0 internucleotide linkages)dependent lected about intracellular RNA processing and degrad-
human and murine ribonucleases L (52) which cleave ing activities in eukaryotes (61, 62), little is known
RNA with the production of 3 0 -phosphoryl and 5 0 - about the responsible enzymes. Part of these enzymes
hydroxyl groups. The polypeptide chains of these will be proteins, but many others may be enzymatically
enzymes have chain lengths of 750 residues, and active RNA molecules (ribozymes). Penny and cowor-
each includes a ribonuclease domain of 200 residues, kers recently published a model for the evolution of life
with some sequence similarity with ribonuclease E. from an RNA world to the emergence of eukaryotes
This suggested homology, however, is not in agreement and prokaryotes (63), and summarized the presence of
with the different cleavage specicities of these mam- RNAs in modern organisms that are relics from the
malian and E. coli enzymes. Another ribonuclease E ancient RNA world. Proteins have gradually replaced
related activity in mammalian cells is the product of RNAs in catalysis by virtue of their superior catalytic
the human ard-1 gene, which encodes a basic, proline- properties. However, in the case of large substrates
rich 13.3-kDa molecular-weight protein (53). such as RNA the rate of diffusion is the slowest reac-
Expression of this gene in E. coli can complement a tion step, and catalytic perfection will not help much in
mutant of the ribonuclease E gene, but there are no speeding up the reaction rate. Therefore, it is not sur-