Cell and Molecular Biology: Nternational Eview of

VOLUME THREE HUNDRED AND NINETEEN
INTERNATIONAL REVIEW OF
CELL AND MOLECULAR
BIOLOGY
International Review of Cell
and Molecular Biology
Series Editors
GEOFFREY H. BOURNE 1949e1988
JAMES F. DANIELLI 1949e1984
KWANG W. JEON 1967e
MARTIN FRIEDLANDER 1984e1992
JONATHAN JARVIK 1993e1995
Editorial Advisory Board

PETER L. BEECH BRUCE D. MCKEE
ROBERT A. BLOODGOOD MICHAEL MELKONIAN
BARRY D. BRUCE KEITH E. MOSTOV
DAVID M. BRYANT ANDREAS OKSCHE
KEITH BURRIDGE MADDY PARSONS
HIROO FUKUDA TERUO SHIMMEN
MAY GRIFFITH ALEXEY TOMILIN
KEITH LATHAM GARY M. WESSEL
WALLACE F. MARSHALL
VOLUME THREE HUNDRED AND NINETEEN
INTERNATIONAL REVIEW OF
CELL AND MOLECULAR
BIOLOGY
Edited by
KWANG W. JEON
Department of Biochemistry
University of Tennessee
Knoxville, Tennessee
AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier
225 Wyman Street, Waltham, MA 02451, USA
525 B Street, Suite 1800, San Diego, CA 92101-4495, USA
125 London Wall, London EC2Y 5AS, UK
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
First edition 2015
Copyright © 2015 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by
the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety and
the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of
products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-802278-8
ISSN: 1937-6448
For information on all Academic Press publications

visit our website at http://store.elsevier.com/
CONTRIBUTORS
Sandra L. Accari
Professional and Continuing Education, Turitea Campus, Massey University, Palmerston
North, New Zealand
Martin Altvater
Institute of Biochemistry (IBC), Department of Biology (D-BIOL), ETH Zurich;
Molecular Life Science (MLS) Graduate School, Zurich, Switzerland
Simone Bergmann
Institute of Microbiology, Technische Universität Braunschweig, Braunschweig,
Lower Saxony, Germany
Sunil K. Chauhan
Schepens Eye Research Institute & Mass Eye and Ear, Department of Ophthalmology,
Harvard Medical School, Boston, MA, USA
Thomas G. Cotter
Tumour Biology Laboratory, School of Biochemistry and Cell Biology, Bioscience
Research Institute, University College Cork, Cork, Ireland
Ute Fischer
Institute of Biochemistry (IBC), Department of Biology (D-BIOL), ETH Zurich, Zurich,
Switzerland
Paul R. Fisher
Discipline of Microbiology, La Trobe University, Melbourne, VIC, Australia
Stefan Gerhardy
Biomolecular Structure and Mechanism (BSM) Graduate School, Zurich, Switzerland
William R. Jeffery
Eugene Bell Center for Regenerative Biology and Tissue Engineering, Marine Biological
Laboratory, Woods Hole, MA; Department of Biology, University of Maryland,
College Park, MD, USA
Kishore R. Katikireddy
Ikeda Lal
LV Prasad Eye Institute, Hyderabad, Telangana, India
Jung Weon Lee
Department of Pharmacy, Research Institute of Pharmaceutical Sciences, Tumor
Microenvironment Global Core Research Center, Medicinal Bioconvergence Research
Center, College of Pharmacy, Seoul National University, Seoul, Korea
ix j
x Contributors
Dilip Kumar Mishra

Department of Ocular Pathology, LV Prasad Eye Institute, Hyderabad, Telangana, India
Purnima Nerurkar
Vikram Govind Panse
Switzerland
Charanya Ramachandran
Sudhakar and Sreekanth Ravi Stem Cell Biology Laboratory, Prof. Brien Holden Eye
Research Centre, C-TRACER, LV Prasad Eye Institute, Hyderabad, Telangana, India
Eileen G. Russell
Tumour Biology Laboratory, School of Biochemistry and Cell Biology, Bioscience
Research Institute, University College Cork, Cork, Ireland
Virender S. Sangwan
Center for Ocular Regeneration, Dr Paul Dubord Chair in Cornea, LV Prasad Eye
Institute, Hyderabad, Telangana, India
Sabina Schütz
Sachin Shukla
Vivek Singh
Michael Steinert
Institute of Microbiology, Technische Universität Braunschweig, Braunschweig,
Lower Saxony, Germany
Christine Weirich
Switzerland
CHAPTER ONE
From Single Cells to Engineered

and Explanted Tissues: New
Perspectives in Bacterial
Infection Biology
Simone Bergmann* and Michael Steinert*
Institute of Microbiology, Technische Universit€at Braunschweig, Braunschweig, Lower Saxony, Germany
*Corresponding authors: E-mail: simone.bergmann@tu-bs.de; m.steinert@tu-braunschweig.de
Contents
1. Introduction 2
2. 2D Cell Culture 4
2.1 Culture of Immortalized Cell Lines versus Primary Cell Culture 7
2.2 Protozoa as Alternative Infection Models 9
2.3 Coculture Infection Models 10
2.3.1 Coculture-based generation of tissue barriers 11
2.3.2 Coculture of adherent cells and neutrophiles in suspension 13
3. 3D Cell Culture 14
3.1 Benefits and Limitations of 3D Scaffold 15
3.1.1 Coculture-based reconstruction of BBB with matrix scaffold 17
3.1.2 Requirements of 3D tissue models generating aireliquid surface 18
3.2 MicrogravitydVariations of 3D Cell Culture Models 19
4. Organ Equivalents and Tissue Explants 21
4.1 Organoids and Tissue Equivalents Providing Complex Cell Systems 22
“En Miniature”
4.2 Tissue ExplantsdPiece of Reality 24
4.3 Integration of Microfluidic Systems in 2D and 3D Cell Culture 26
5. Concluding Remarks and Future Perspectives 30
Acknowledgments 31
References 31
Abstract
Cell culture techniques are essential for studying hostepathogen interactions. In addi-
tion to the broad range of single cell type-based two-dimensional cell culture models,
an enormous amount of coculture systems, combining two or more different cell types,
has been developed. These systems enable microscopic visualization and molecular
analyses of bacterial adherence and internalization mechanisms and also provide a suit-
able setup for various biochemical, immunological, and pharmacological applications.
International Review of Cell and Molecular Biology, Volume 319
© 2015 Elsevier Inc.
j
ISSN 1937-6448
http://dx.doi.org/10.1016/bs.ircmb.2015.06.003 All rights reserved. 1
2 Simone Bergmann and Michael Steinert
The implementation of natural or synthetical scaffolds elevated the model complexity

to the level of three-dimensional cell culture. Additionally, several transwell-based cell
culture techniques are applied to study bacterial interaction with physiological tissue
barriers. For keeping highly differentiated phenotype of eukaryotic cells in ex vivo
culture conditions, different kinds of microgravity-simulating rotary-wall vessel systems
are employed. Furthermore, the implementation of microfluidic pumps enables con-
stant nutrient and gas exchange during cell cultivation and allows the investigation
of long-term infection processes. The highest level of cell culture complexity is reached
by engineered and explanted tissues which currently pave the way for a more compre-
hensive view on microbial pathogenicity mechanisms.
1. INTRODUCTION
Most basic studies on hostepathogen interaction have been focused
on cultured and, frequently, immortalized cell lines or animal experiments
(Mizgerd and Skerrett, 2008). The first reports highlighting the suitability
of in vitro cell culture models to study pathogenesis of microorganisms
were published in the early 1970s and focused on virusehost cell interac-
tions (Todaro et al., 1971). As early as in 1976, Taylor-Robinson (1976)
described the use of ciliated tracheal epithelium of animals to study
mycoplasma pneumonia infections. Since then, in vitro cell culture models
became increasingly popular in infection biology as they combine several
advantages. Compared to animal models they are cost-effective and acces-
sible, they allow experimental flexibility including high-throughput plat-
forms and they exhibit a high reproducibility. Moreover, the vast
innovations in cell biology such as microscopic imaging, genetic, biochem-
ical, and immunologic technologies allowed deep insights in host cell re-
sponses elicited by microbes. This includes the exploitation of host cell
components during adherence, invasion, replication, and evasion of patho-
gens. Meanwhile, several tissue culture collections and companies offer a
broad list of different immortalized cell lines and primary cells derived
from human and different animal species thereby allowing cell type-specific
investigations.
The central key point in cell culture-based infection biology is the level
of complexity which can be reached by an in vitro cell culture model in re-
gard to differentiation and reactivity, to adequately mimic the situation in a
complex host organism. In order to improve the value of data obtained from
cell culture models, the methods have been optimized and adapted to special
scientific questions. Many examples derived from different scientific
Cell Culture Techniques in Infection Biology 3
disciplines approved that simplified models enhance the probability to eluci-

date crucial or new specific interactions of individual components by
excluding the vast amount of overlaying interactions. A prerequisite for
this, however, is that conclusions drawn from model systems take into
account the fact that certain properties are not represented in the model.
Thus, it is widely accepted that the suitability of a specific cell culture-based
model to generate reliable data, which can be superimposed for the situation
in vivo, has to be validated for every single scientific question.
To overcome the limitation of isolated and often immortalized cells,
models of higher complexity have been generated during the last decades
of years. Figure 1 depicts the increase in model complexity and outlines
Figure 1 Schematic illustration of different cell culture models presented in this review.
The cell culture models are roughly categorized in two-dimensional (2D) cell culture,
three-dimensional (3D) cell culture with scaffolding materials, and organoids and tissue
explants. The cell culture models are distributed vertically according to the level of
complexity, reaching from cellular level to tissue level. The transwell models are
composed of a two chamber system separated by a porous membrane. In most of
the applied transwell cell culture systems, different cell types are cultivated on the up-
per and the lower site of the membrane. Several bloodebrain barrier (BBB) models also
implement scaffolding materials such as extracellular matrix (ECM) components for cell
culture. The rotary vessel culture enabled cell cultivation in microgravity environment.
the structure of this review. Beginning with monolayer cultures and proto-
zoa-based models as paradigms for simple two-dimensional (2D) cell culture,
we will discuss the benefits and limitations of established cell culture models
applied in infection biology. Following the line of increasing complexity,
several cocultivation techniques, transwell-based tissue models, and culture
systems with implementation of scaffolds will be presented in the second
part. The current highest level in complexity is reached in reconstructional
systems on the tissue level and includes the generation of tissue aggregates,
the cultivation of organoids and the use of organ-specific ex vivo tissue ex-
plants. Based on selected examples, the advantages and restrictions of these
complex three-dimensional (3D) systems will be commented within the
last chapter of this review.
2. 2D CELL CULTURE
In many experimental setups, the creation of a functional host cell sur-
face is sufficient to study initial interactions with bacterial pathogens such as
adherence, invasion, and induction of signal transduction processes. For de-
cades of years 2D cell monolayers grown on solid, impermeable plastic or
glass surfaces have been applied as simple and cost-effective strategy to
analyze principle mechanisms of bacterialehost cell interactions. Numerous
in vitro cell culture systems have been confirmed as suitable to provide in-
formation about specific bacterial virulence factors involved and also eluci-
dated the induction of many fold intracellular processes, such as signaling
cascades and cytoskeletal rearrangements on the host side (Table 1).
Depending on the physiological niche of the bacteria, pulmonary cells
are cultivated and infected with typical lung pathogens like Legionella pneu-
mophila and Streptococcus pneumoniae; gastrointestinal cells are used for infec-
tion studies with Helicobacter and Salmonella, and skin fibroblasts are chosen
for infection with causative agents of wound infections like staphylo-
coccidjust giving a few examples. These studies generated impressive trans-
mission electron microscopic pictures and opened up the field for more
detailed cell culture infection analyses. Several immunofluorescence staining
procedures have been developed, which can be applied after infection of
eukaryotic cell monolayers with pathogenic bacteria. These procedures
enable a microscopic visualization and also a differential quantification of
adhered and internalized bacteria (Bergmann et al., 2009, 2013; Elm
et al., 2004; Jensch et al., 2010; L€ uttge et al., 2012; Agarwal et al., 2013,
Table 1 Representative examples of bacterial pathogens analyzed in different cell

culture models as highlighted in this review
Model system/cell type Bacterial strain References
Single cell type monolayer
Macrophages Helicobacter pylori Wang et al., 2009
Macrophages, Legionella pneumophila Steinert et al. (1994), Allombert
Dictyostelium et al. (2014), Steinert et al.
discoideum (2000), Shevchuk and
Steinert (2009), and Skriwan
et al. (2002)
D. discoideum Mycobacterium marinum Meng et al. (2014)
Dendritic cells Streptococcus pneumoniae Rosendahl et al. (2013)
Epithelial and S. pneumoniae Steinford et al. (1989), Jensch
endothelial cells et al. (2010), Bergmann et al.
(2009, 2013), L€ uttge et al.
(2012), Agarwal et al. (2013,
2014), Pracht et al. (2005),
and Elm et al. (2004)
Endothelial cells Streptococcus pyogenes Amelung et al. (2011) and
Nerlich et al. (2009)
2D coculture model
Bilayer model Neisseria meningitides Birkness et al. (1995)
(epithelium and
endothelium)
Bloodebrain barrier Escherichia coli Huang et al. (1995, 2000) and
model Kim (2000)
Streptococcus agalactiae Nizet et al. (1997)
Listeria monocytogenes Greiffenberg et al. (1998)
Citrobacter feundii Badger et al. (1999)
S. pneumoniae Ring et al. (1998) and Untucht
et al. (2011)
Bloodecerebrospinal N. meningitides Steinmann et al. (2013)
fluid model
Lung coculture model Pseudomonas aeruginosa, Hurley et al. (2004)
Klebsiella pneumoniae,
E. coli
3D culture model with scaffolds
Transwell system with Chlamydia trachomatis Igietseme et al. (1994), Kane and
ECM Byrne (1998), Kane et al.
(1999), and Dessus-Babus
et al. (2002)
L. pneumophila Skriwan et al. (2002)
(Continued)
Table 1 Representative examples of bacterial pathogens analyzed in different cell

culture models as highlighted in this reviewdcont'd
Model system/cell type Bacterial strain References
L. monocytogenes Cossart and Lecuit (1998)
Neisseria gonoorhoeae Hopper et al. (2000)
S. pyogenes Ochel et al. (2014)
Epithelial airway tissue S. pneumoniae Fliegauf et al. (2013)
model P. aeruginosa Woodworth et al. (2008)
Rotary wall vessel culture
A549 lung epithelial Francisella tularensis David et al. (2014)
cells P. aeruginosa Carterson et al. (2005)
Cells on collagen- C. trachomatis Guseva et al. (2007)
coated microcarrier
beads
Human intestinal Salmonella enterica Nickerson et al. (2001)
Int-407 cells
Organoids
Human ileocecal E. coli (EHEC/EPEC) Carvalho et al. (2005)
colorectal
adenocarcinoma
HCT-8
Human skin equivalent Acinetobacter baumannii de Breij et al. (2010) and Breij
et al. (2014)
Tissue explants
Lung tissue explants Chlamydia pneumophila Rupp et al. (2004)
L. pneumophila J€ager et al. (2014) and Shevchuk
et al. (2014)
S. pneumoniae Szymanski et al. (2012)
Tonsil explants S. pyogenes Bell et al. (2012) and Abbot et al.
(2007)
Microfluidic perfusion
HUVEC Staphylococcus aureus Pappelbaum et al. (2013)
2014; Pracht et al., 2005; Amelung et al., 2011; Nerlich et al., 2009; Rohde
and Chhatwal, 2013). In addition, fluorescence staining of the actin cyto-
skeleton also visualized radical morphological changes of the eukaryotic
cells, e.g., induced by streptococcal adherence (Bergmann et al., 2009).
Complex responses of host immune systems are also studied using suspen-
sion cultures with prepared and differentiated macrophages or other cells
derived from the lymphoid tissues.
This chapter will provide an overview about the broad spectrum of 2D
cell culture models in infection biology. Beginning with the discussion of
key aspects in using immortalized versus primary nonimmortalized eukaryotic
monolayers in infection models, the second part is focused on protozoa-based
models in infection biology. Climbing to the next level of cell culture
complexity, the advantages of coculture models generated by simultaneous
cultivation of two or more different cell types will be described. The
cocultivation technique is used to regenerate tissue barriers and will be
demonstrated by the examples of a transwell-based reconstruction of a
bloodebrain barrier (BBB) and a bloodecerebrospinal fluid barrier (BCSFB).
2.1 Culture of Immortalized Cell Lines versus Primary Cell

Culture
Many valuable bacterial pathogenicity mechanisms have been elucidated us-
ing 2D monolayer cell cultures indicating that certain scientific questions can
be answered and sometimes even require this kind of simplified cell culture
technique. A risk of using 2D monolayer cell culture models is the loss or
diminished expression of certain phenotypic characteristics that may mediate
bacterialehost cell interactions. This loss of phenotype results in progressive
alterations in biochemistry, function, and morphology and increases with
every passage of the culture as the cells diverge from the source tissue pheno-
type (Shaw, 1996). Of special importance are immortalized human cell lines,
which have been used extensively for the study of hostepathogen interac-
tions. These stable cell lines combine the advantage of an indefinite life span
allowing passaging of several hundred times with low or moderate culture
requirements. However, these lines exhibit aberrant properties attributable
to immortalization and artificial 2D growth conditions (Freshney, 2005).
Thus, several studies targeting the investigation of pathogenecell inter-
actions have been conducted with primary cells derived from different ani-
mals such as primary porcine endothelial and epithelial cells (Vanier et al.,
2007; Boekema et al., 2003). Nevertheless, in human-specific infection
biology, the manual cell preparation of specific cell types from organ mate-
rial remains a time-consuming and difficult process, retarded very often by a
limited access to donor material, the lack of cell type-specific selection
marker, a high risk of contamination, and suboptimal culture conditions.
Fortunately, several companies such as PromoCell GmbH, Germany, and
PAA laboratories GmbH, Germany (part of GE Healthcare, USA) are

meanwhile specialized on eukaryotic cell culture techniques and provide
an enormously extended catalog of different cell types from human lung,
skin, and gastrointestinal tract. These catalogs also include more sensitive
cell types such as primary epithelial, endothelial vascular cells, and smooth
muscle cells. Most of the primary monotypic cells are characterized by
certain tissue-specific properties such as unique morphology, proliferation
behavior, and receptor expression pattern and are considered as appropriate
representatives of a specific tissue. But this high level of specificity is based on
an expense of a higher sensibility toward culture conditions. In general, pri-
mary cells require more complex cell culture media and a more experienced
handling. The cells are more susceptible toward any kind of stress and in
contrast to immortalized cell lines, primary cells tend to rapidly lose their
high status of differentiation after being isolated from the native tissue
(Freshney, 2005). Fulminant changes in cell differentiation and functional
specificity are monitored especially for primary endothelial and epithelial
cells, whereas immortalized cell lines tend to keep their main characteristics
for longer proliferation rounds (Maqsood et al., 2013). A flow cytometry
analysis of human primary lung endothelial cells, which has been performed
exemplary for lung endothelial cells, clearly indicated significant changes in
endothelial marker expression like platelet endothelial cell adhesion mole-
cule 1 (PECAM1) within eight rounds of cell passaging and also depicted
an inversion of the integrin expression pattern (Bergmann et al., unpublished
data). This result is of highest interest, for example, with respect to the inves-
tigation of receptor-mediated bacterial adherence and invasion in infection
biology. In order to ensure that the cells in culture express the same type and
amount of surface receptors as in vivo, several cell culture infection analyses
are performed additionally with primary cells directly prepared from the
respective tissue. In addition to the receptor profile, a further critical aspect
in long-term cell culture is the risk to lose the ability of cell type-specific
functions. For example, endothelial cells generate secretory WeibelePalade
bodies, which are filled with coagulation factors, vasodilation activators, and
cytokines. The functional study of these WeibelePalade bodies in infection
processes requires the use of primary cells in low passage, since the endothe-
lial cells lose the ability of vesicle secretion after several in vitro passages
(L€uttge et al., 2012). Notably, some cell types allow a specific manipulation
of a special cell activity, which may be required to answer a scientific ques-
tion. A typical example is given by U937 cells, a human myelomonocytic
cell line arrested at a monoblastic stage. These cells have been extensively
used as a model to study cellular processes during different stages of mono-

cytic differentiation and are used as cell culture model for macrophage-
mediated uptake of pathogens like Helicobacter, Mycobacteria, Legionella, and
several more (Wang et al., 2009; Meng et al., 2014; Allombert et al.,
2014; Steinert et al., 1994). The low basal level of phagocytic activity of
U937 can be increased by the induction of terminal monocytic differentia-
tion through exposure to a combination of vitamin D3 and transforming
growth factor-b1 (Wright et al., 1999) and allowed complex analyses of bac-
terial uptake processes (Tacken and Batenburg, 2006). In general, immortal-
ized cell lines remain basic components of standardized and easy-to-handle
cell culture models. But in order to circumvent the functional limitations,
primary, differentiated, and functional active cells replace them in cell cul-
ture models, as soon as specific cell functions and receptors interactions
are major part of scientific interest.
2.2 Protozoa as Alternative Infection Models

Environmental protozoa are recognized as reservoirs and vehicles for several
important bacterial pathogens. Thus, it is likely that on an evolutionary time
scale, protozoaebacteria interactions have generated a pool of virulence
traits, which preadapted some bacterial species as human pathogens. The
protozoan mechanisms of phagocytosis, e.g., use signaling pathways and
cytoskeleton proteins closely related to those of macrophages, neutrophils,
and dendritic cells. Moreover, it has been shown that many of the virulence
factors required for pathogenicity in mammals are also important for path-
ogen survival during interactions with unicellular organisms such as Acantha-
moeba castellanii or Hartmannella vermiformis (Hilbi et al., 2011; Steinert et al.,
2003; Mody et al., 1993; Pearlman et al., 1988; Rowbotham, 1980). A
frequently utilized protozoan species, which allows highly relevant cross-
species comparisons and mutant screenings on both sides of the hoste
pathogen interaction and the analyses of fundamental cellular processes, is
Dictyostelium discoideum (Shevchuk and Steinert, 2009; Steinert et al.,
2003; Steinert et al., 2000; H€agele et al., 2000). The pathogens predomi-
nantly analyzed in D. discoideum are L. pneumophila, Mycobacterium spp., Pseu-
domonas aeruginosa, and Cryptococcus neoformans. Unlike other protozoa, D.
discoideum is amenable to genetic manipulation and combines many advan-
tages such as easy cultivation, availability of cellular markers, the knowledge
of cell signaling pathways, and an elaborated molecular tool box. Accord-
ingly, the social amoeba D. discoideum has become a prime model organism
in infection biology (Gerstenmaier et al., 2015; Weber et al., 2014; Bozzaro
et al., 2013; Clarke, 2010; Shevchuk and Steinert, 2009; Hilbi et al., 2007;
Steinert et al., 2003; Skriwan et al., 2002; H€agele et al., 2000).
2.3 Coculture Infection Models

Primary and immortalized cell culture models are not perfect representa-
tives of the complex cellular environment found in organisms. The devel-
opment of several cocultivation techniques derived from the need to
generate tissue barriers for the simulation of pathogen-driven traversal dur-
ing infection processes. In coculture infection models, two or more single
cell types are cultivated simultaneously as monolayer, in suspension, or as a
mixed coculture, which typically utilizes adherent monolayers in combina-
tion with a suspension cell type (Duell et al., 2011). An increasing amount
of cocultures is generated as adjacent bilayers with the aim to create a more
physiological and complex surrounding such as a tissue barrier (Lindén
et al., 2007).
In infection biology research, these models are commonly composed of
one epithelial cell type, which derived from a special tissue or an organ such
as the lung or the brain, and the corresponding endothelial or assisting cells.
These assisting cells provide nutrition, contact, and synthesize cofactors
required for the functionality of the overlaying epithelium. For this purpose,
many cocultivation cell culture systems are generated in a transwell two
chamber system allowing the separated culture of one cell type on top of
a porous membrane and the other cell type at the bottom. This type of
coculture is extensively used to elucidate the mechanism by which microor-
ganisms like Streptococci, Helicobacter, and Staphylococci adhere, invade, and
signal to the host. Moreover, these systems also enable an investigation of
immune responses stimulated by the pathogens themselves or their secreted
virulence factors including determination of cytokine release and moni-
toring of enhanced neutrophil transmigration (Lindén et al., 2007). An
important general factor that can affect experimental design and the success-
ful application of single culture as well as coculture models in infection
studies is the effect of microbial activities on host cell viability. Microbes
can rapidly influence the viability of host cells through necrosis, apoptosis,
and pyroptosis. As a consequence, survival rates of the cocultured cells can
decrease rapidly and alter the results (Wiegand et al., 2009). In addition to
a steady control of cell morphology and viability, negative side effects can
be reduced by decreasing infection time and bacterial load.
In the following paragraphs, representative examples for coculture-based
infection models will be described.
2.3.1 Coculture-based generation of tissue barriers

The BBB and the BCSFB, separating brain interstitial space from blood, are
anatomical and functionally unique barriers formed by the tight junctions
of brain microvascular endothelial cells (BMECs) (Pardridge, 1999; Rubin
and Staddon, 1999). Therefore, the reconstruction of such complex barrier
systems requires experienced knowledge about the detailed composition
and intercellular relationship between the involved cell types. The proper-
ties of the brain endothelium are supported and maintained by associated
cells, like astrocytes, pericytes, and microglia (Abbott, 2005). Astrocytes
are known to induce and regulate many BBB characteristics and functions,
namely the formation of tight junctions as well as the expression and asym-
metrical localization of special enzymes, e.g., within transport systems
(Abbott, 2005; Cecchelli et al., 2007). Viral, bacterial, fungal, and parasitic
pathogens have been reported to breach the BBB and enter the central ner-
vous system (CNS) through transcellular, paracellular, and the intracellular
“Trojan Horse” mechanisms. Microscopic visualization techniques clearly
demonstrated transcellular invasion of BMEC by bacterial and fungal path-
ogens including Escherichia coli (Huang et al., 1995, 2000; Kim, 2000),
Group B Streptococcus (Nizet et al., 1997), Listeria monocytogenes (Greiffen-
berg et al., 1998), Citrobacter freundii (Badger et al., 1999), S. pneumoniae
(Ring et al., 1998), Candida albicans (Jong et al., 2001), and several more.
These studies revealed that the development of CNS inflammation is
accompanied by the release of cytokines (De Vries et al., 1997; Sun
et al., 1998). Moreover, barrier-forming coculture models identified mi-
crobial toxins and various microbial surface components as potent inducer
of inflammation response of the brain endothelium, which leads to collat-
eral damage of brain tissue and loss of barrier function (Stamatovic et al.,
2008). As typical example for toxin-mediated damage of lung and brain tis-
sues, the pore-forming cytotoxin pneumolysin of S. pneumoniae directly
activates proinflammatory cellular cascades and induces apoptotic cell death
(Hirst et al., 2004). Further destructive effects on tissue are also induced by
active cleavage of junctional proteins, which enables pathogen transmigra-
tion of cellular barriers (Attali et al., 2008a). This kind of penetration
mechanism was reported for S. pneumoniae bacteria recruiting the serine
protease activity of plasmin for degradation of catherines (Attali et al.,
2008b). Paracellular penetration of the BBB has also been suggested for
the Lyme disease pathogen Borrelia burgdorferi (Garcia-Monco et al.,
1990; Comstock and Thomas, 1991) and the syphilis-causing agent Trepo-
nema pallidum (Haake and Lovett, 1994).
The complexity of BBB models ranges from transwell-based models

with one or more cell types (astrocytes, pericytes, and endothelial cells)
(Nakagawa et al., 2007) up to 3D model systems (Cucullo et al., 2007;
Parkinson et al., 2003; Stanness et al., 1997). Thereby, the central compo-
nent of most in vitro models of the BBB is BMEC, which exhibited a typical
cobblestone-like pattern at confluence, reconstitute in vitro their tight junc-
tions and maintain specific cell properties up to passage 14 (Banks, 1999;
Persidsky, 1999; Huang et al., 2000). BMEC-based in vitro models of the
BBB have been used in various studies on cellular and molecular mecha-
nisms of CNS infections caused by bacteria, virus, fungus, and parasites
(Banks, 1999; Fusai et al., 2000; Huang et al., 2000; Jong et al., 2001).
Initially, because of difficulties in isolating BMECs and growing them in cul-
ture, most of the in vitro studies in pathogenesis of the CNS infection had
been carried out using large vessel endothelial cells such as human umbilical
vein endothelial cells (HUVEC) (Townsend and Scheld, 1995). Such kind
of systemic endothelial cells is still widely accepted and used as model cells
for endothelial functions, although any kind of brain-specific property is
missing. Thus, in disregard of species-specific interactions, bovine, murine,
and human BMECs have been successfully used to dissect many pathogenic
mechanisms of the CNS infection in vitro (Stins et al., 1997; Banks et al.,
1998; Huang et al., 2000).
In order to study the complexity of the cell-to-cell interaction associated
with the pathogenesis of bacterial pathogens such as Neisseria meningitidis, a
bilayer model with endothelial and epithelial cells (EC/EP bilayer) had been
established (Birkness et al., 1995). In this model system, endothelial and
epithelial cells are separated by a microporous membrane, which allows
analysis of bacterial transmigration through the multiple layers via micro-
scopic visualization techniques. Transwell-based transmigration models
also enable biological impedance determination by measurements of the
transepithelial electrical resistance (TEER), thereby monitoring the tightness
of the barrier and any loss of barrier function. Interestingly, such coculture
models revealed that many bacteria are efficiently penetrating from the api-
cal surface through the epithelial layer and through the membrane to the
bottom layer of endothelial cells without causing significant damage to
the host cells (Birkness et al., 1995).
A further typical bilayer cell model simulating lung tissue was developed
by Mul and colleagues and was composed of primary human endothelial
(human papilloma virus-immortalized HUVEC cell line or primary
HUVECs) and lung epithelial cells (H292 or primary bronchial epithelial
cells). The cells were simultaneously cultured on opposite sides of transwell

culture inserts (Mul et al., 2000). This model was applied for transmigration
analysis of polymorphnuclear neutrophils (PMNs) (Parkos et al., 1991; Zeil-
lemaker et al., 1996). PMN was labeled with calcein-AM prior to the start of
the transmigration assay and after lysis, the amount of fluorescence in each of
culture insert compartments or attached cells was measured in a spectroflu-
orometer at the end of the experiment. The above-mentioned examples
indicate the high level of flexibility of bilayer cell culture models to generate
and maintain tissue barriers and to investigate specific cellular events corre-
lating with the infection process at specific tissue sites.
2.3.2 Coculture of adherent cells and neutrophiles in suspension

CNS infection caused by microbial pathogens can lead to devastating neuro-
logical disability and death. Moreover, a critical point during the course of
central nervous system infection is the influx of leukocytes from the blood
into the brain across the BBB but also across the BCSFB. The pathogens
have developed a variety of strategies to breach the endothelium of the
BBB or the choroid plexus epithelium of the BCSFB, which normally pre-
vents entry of toxic substances into the brain interstitium. For many impor-
tant meningitis-causing pathogens such as Neisseria meningitides (Pron et al.,
1997), Haemophilus influenzae (Smith, 1987), E. coli (Parkkinen et al., 1988),
L. monocytogenes (Prats et al., 1992), Streptococcus suis (Sanford, 1987), and
some enteroviruses (Tabor-Godwin et al., 2010), experimental data suggest
involvement of the choroid plexus during pathogen entry into the brain.
Microbial infection of the CNS induces an increased transmigration rate
of polymorphnuclear neutrophils (PMNs) into the subcellular spaces as
the first line of defense, promoted by an IL-8 release of epithelial or endo-
thelial cells (Wittchen, 2009; Chin and Parkos, 2007). Various human
models of the BBB employing immortalized cell lines have been developed
(Stins et al., 2001; Weksler et al., 2005; Muruganandam et al., 1997),
whereas in vitro systems mimicking the BCSFB are limited to animal-
derived cell material, including rat cell lines and primary porcine choroid
plexus epithelial cells (PCPEC) (Haselbach et al., 2001; Shi and Zheng,
2005; Gath et al., 1997; Kitazawa et al., 2001; Zheng and Zhao, 2002). Tis-
sue barriers generated by PCPEC were used for leukocyte migration studies
and elucidated cell migration via the paracellular and the transcellular route
(Wewer et al., 2011; Steinmann et al., 2013). The establishment of an
inverted transwell filter system with porcine or human malignant choroid
plexus papilloma cells (HIBCPP) comprising high barrier characteristics
enabled basolateral infection of host cells as well as investigation of transmi-

gration of leukocytes from the pathophysiologically relevant blood side to
the apical cerebrospinal fluid side (Tenenbaum et a., 2013). This model
was successfully applied to study the influence of N. meningitidis infection
of HIBCPP on the transmigration of PMNs and monocytes (Steinmann
et al., 2013). Another infection model consisting of alveolar epithelial cells
(A549) and human PMN grown on inverted culture inserts was used to
determine whether also bacteria such as P. aeruginosa, Klebsiella pneumoniae,
and E. coli are capable of inducing PMN migration across these epithelial
barriers (Hurley et al., 2004). Calu-3 cells as well as primary human ATII
cells were also applied in inverted culture insert systems for similar objectives
(Zemans et al., 2009). The afore-mentioned examples indicate the suitability
and of transwell-based cell culture models for investigation of bacterial path-
ogenicity mechanisms. The very sophisticated requirements of reconstructed
tissue barriers have been complied by several variations in supplementation
of scaffolding materials. This kind of cell culture technique initialized the
field of 3D cell culture technique.
3. 3D CELL CULTURE
In many cases, the cocultivation of different cell types requires the
assembly of a special kind of scaffold providing docking points and stimula-
tion for adequate cell morphology, growth behavior, and survival. These
scaffolds are formed by nanometer-sized fibers and pores, which are essential
to ensure a true 3D environment for the cell. Studies spanning over two
decades of research provide several evidences that growing cells within
3D scaffolds reduce the gap between cell cultures and physiological tissues.
Thus, in order to maintain at most the physiological properties of cell tissues,
great emphasis was laid on using 3D cell culture models that display
functional and phenotypic features of in vivo tissues (Fraley et al., 2010;
Dhimolea et al., 2010). Meanwhile, an increasing amount of new technol-
ogies, such as nanotechnology engineering, provides further evidence that
the 3D in vitro cultivation of epithelial cells is crucial for such cells to sense
and respond properly to receptor complex presentation (Discher et al., 2005;
Nickerson et al., 2004). In fact, key events in the life cycle of a cell, such as
proliferation, migration, and apoptosis, are regulated by organizing princi-
ples that are determined by the cellular context (Bissel et al., 2002). These
organizing principles are maintained by cellecell and celleECM
(extracellular matrix) interactions, involve cytoskeletal orientation and

signaling, i.e., tyrosine phosphorylation and Rho/Ras/Rac activation, and
establish a 3D communication network that maintains the specificity and
homeostasis of the tissue (Kleinman et al., 2003). It is widely accepted
that 3D cell cultures that reestablish such physiological cellecell and celle
ECM interactions can mimic the specificity of real tissues better than con-
ventional 2D cultures. 3D cultures are currently used in a broad range of
cell biology studies including tumor biology, cell adhesion, cell migration,
and epithelial morphogenesis. Example methodologies have also entered
the field of infection biology and include the application of collagen-coated
polycarbonate transwell filter chamber inserts (Costar) and/or ECM-coated
filter invasion chambers (BioCoat Matrigel; Becton Dickinson) to examine
the mechanisms of pathogen entry and intracellular fate of actin-recruiting
Listeria (Cossart and Lecuit, 1998), for studying interaction of Neisseria, Chla-
mydia, and other pathogens with polarized and nonpolarized cell layers
(Hopper et al., 2000; Igietseme et al., 1994; Kane and Byrne, 1998; Kane
et al., 1999; Dessus-Babus et al., 2002; Kazmierczak et al., 2001) and the
reactivity to antibiotics and innate inflammatory response (Kenny et al.,
1997; Linzmeier and Ganz, 2005; Sansonetti, 2001).
3.1 Benefits and Limitations of 3D Scaffold

Tissue architecture is better represented by 3D cell culture than 2D cell cul-
ture. Moreover, it has been reported that mechanical and biochemical cues
and cellecell communication are lost under the simplified and highly biased
conditions of 2D cell cultures. This opinion is based on different cell culture
observations, which will be discussed in the following paragraph. For
example, fibroblasts that migrate on a 2D substrate exhibit a different shape
and a different distribution of transmembrane adhesion proteins compared
with fibroblasts within a 3D collagen scaffolding matrix (Walpita and Hay,
2002; Cukierman et al., 2001; Meshel et al., 2005). Moreover, cells cultured
in 3D reveal different gene expression levels compared with their 2D coun-
terparts. Melanoma cells cultured on flat substrates upregulate and downregu-
late other genes compared with melanoma cells cultured in 3D as spheroids. A
further example is given by the observation that mammary epithelial cells that
are cultured on flat plastic surfaces dramatically upregulate the expression of
mRNA that codes for b1-integrins. By contrast, culturing on a recombinant
basement membrane induces expression levels of mRNA that are comparable
with those in the breast tissue (Delcommenne and Streuli, 1995). The impact
of the mode of cell cultivation system on expression of cell surface receptors,
which are specifically required for bacterial interaction, is also highlighted in

an infection study with pathogenic Streptococci (Ochel et al., 2014). Here, a
gelatin-coated transwell system was used to analyze uptake of M1-protein
expressing Streptococcus pyogenes bacteria by differentiated, polarized endothe-
lial cells. In addition to microscopic visualization of the bacterial uptake pro-
cess, this cell culture infection model was also used to analyze the contribution
of the cytoskeleton and of the intracellular trafficking system to streptococcal
uptake (Ochel et al., 2014). In addition to expression profiles of differentia-
tion markers and tissue-specific receptors, it is often documented that the
topography and cytoskeletal organization of nonpolarized epithelial cells
cultured in a 2D fashion in vitro on impermeable plastic or glass surfaces re-
sembles that of simple fibroblasts. In general, 2D cell cultures are characterized
by the following morphological properties: plasma membrane proteins are
evenly distributed circumferentially; the perinuclear region is broad with
the Golgi complex juxtaposed to the nucleus; and microtubules extend hor-
izontally out to the periphery in an orientation parallel to the basement mem-
brane. In contrast, epithelial cells cultured in vitro in a 3D polarized
orientation on suitable collagen or natural extracellular scaffold matrices are
three to five times taller and have a different topography: plasma membrane
proteins are separated into distinct apical and basolateral membranes by tight
junctions and junctional complexes, and are functionally compartmentalized
therein; the Golgi complex assumes a supranuclear position in the apical cyto-
plasm domain and microtubules are arranged vertically in an apical-to-basal
axis parallel to the lateral membrane (Guseva et al., 2007; Yeaman et al.,
1999). Biophysical experiments have highlighted further differences between
2D and 3D cell migration: 2D culture situation allows only limited multishape
migration behavior, whereas, depending on the specific biological situation,
cell migration can be of mesenchymal or amoeboid type, individual or collec-
tive in clusters and multicellular sheets (Friedl, 2004). The migration speed of
cells also depends on the sterical and mechanical properties of the employed
3D matrix (Zaman et al., 2006). On flat surfaces, the speed with which cells
migrate is related to the strength of the cell-surface adhesion, as determined by
integrin-dependent anchorage. Whereas the maximum migration speed on
2D substrates is reached in regimes of intermediate adhesiveness (DiMilla
et al., 1991). Many substances are used in the production of scaffolding bio-
materials including ceramics, synthetic polymers (polyurethanes, silicones,
polyglycolic acid, polylactic acid, polyanhydrides, polyorthoesters), and natu-
ral polymers (chitosan, glycosaminoglycans) and collagens. Substantial varia-
tions in matrix stability and lack in standardized reproducibility have to be
mentioned as current bottle necks, when using scaffold matrices in cell culture
systems. This kind of difficulties can be partially overcome by using precoated
culture material provided by some companies. During the last decades of
years, collagen has become a favorable component of cell culture scaffolds.
Several properties hallmark the suitability of collagens as cell culture embed-
ding scaffold. Depending on the collagen subtype, either robust protein fibrils
or smaller net forming can be generated thereby providing a perfusable, gelat-
inous hydrogel. Collagen is a natural substrate for cells, and collagen gels
encourage cellular growth and have an impact on morphology, migration,
and adhesion of cells (Kleinman et al., 1982). Nevertheless, because most of
the currently available ECM gels are extracted from animals or cultured cells,
quality control is difficult. For example, the amount of undesired soluble
components varies between batches, which reduces the reliability and repro-
ducibility of the assay. Progress is achieved with fully synthetic fibrous
biopolymer scaffolds, and gels of self-assembling synthetic oligopeptides are
now available for 3D cell cultures (for example, the commercially available
PuraMatrix). At pH and temperature conditions that are compatible with
that of tissue culture, the oligopeptide building blocks form a well-defined
scaffold made of nanometer-sized fibers. These fibers and pores are essential
to ensure a true 3D environment for the cell (Gelain et al., 2006; Horii
et al., 2007). A further advantage is that such gels can be custom-tailored
with specific amino acid sequences that are recognized by the cell’s adhesion
receptors (Zhang et al., 2004). With regard to infection biology, it has to be
mentioned that the choice of applied scaffold might affect the interaction be-
tween pathogens and the tissue. Some bacterial infections require the interac-
tion with ECM proteins such as collagen by inducing an initial contact, which
is subsequently mediating adherence to cell surfaces. On the other hand, un-
specific interaction of bacteria with some scaffolding material might interrupt
any further interacting process with the tissue. The following chapter will
discuss typical examples for 3D scaffold-based cell culture models, which
are widely applied to analyze the interaction of pathogens with the human
and porcine BBB and the respiratory tract. In addition, the beneficial effects
of microgravity on 3D cell culture will be discussed.
3.1.1 Coculture-based reconstruction of BBB with matrix scaffold

The BBB is a dynamic system and needs continuous induction processes, as
evidenced by the fact that cells from brain microvessels can lose BBB fea-
tures in monocultures (Reichel et al., 2003). EC/EP-bilayers might
improve the cell layer complexity but several studies have shown that
astrocytes or related neuroepithelial cells contribute to the induction of

barrier properties in BMEC (Kniesel and Wolburg, 2000). Based on this
observation, an in vitro model consisting of both human BMEC and astro-
cytes has been developed for analysis of penetration of HIV-1 and blood-
derived monocytes into the CNS (Persidsky et al., 1997; Persidsky, 1999).
Like the EC/EP bilayer system, human BMEC and astrocytes are cultured
in this model on opposite sides of a collagen-coated porous membrane in
tissue culture inserts. This leads to a direct contact of the astrocyte end-feet
network with the BMEC monolayer. This system has been successfully
used to study the mechanisms of virus-infected or activated leukocyte
migration across the BBB (Persidsky, 1999). In another model, C6 astro-
cytes were implemented as key element in an optimized in vitro BBB
model (Hurst and Fritz, 1996; Untucht et al., 2011). C6 astrocytes are a
cell line derived from a rat glioma and secrete soluble factors inducing
BBB-specific gene expression in brain endothelial cells (Hurst and Fritz,
1996; Hurst et al., 1998; Kuchler-Bopp et al., 1999). In this model, human
brain endothelial cells were embedded in a semisynthetic basement mem-
brane (MatrigelÔ ). The cocultivation with the C6 astrocytes allowed a
more detailed characterization of BBB penetration mechanisms by try-
panosomes and was approved to be suitable for analyses of drug penetration
properties (Untucht et al., 2011; K€ uhne et al., 2012).
3.1.2 Requirements of 3D tissue models generating aireliquid

surface
The commercially available epithelial airway tissue model (EpiAirwayTM)
of respiratory tract tissue enables cultivation on a microporous membrane
at an aireliquid surface. It consists of normal human tracheal or bronchial
epithelial cells and has been previously used for in vitro tests of nasal
bioavailability (Agu et al., 2004; Chen et al., 2006). In order to confirm
functionality of cells in culture and to circumvent the limitation of avail-
ability of human tissue material for preparation of primary cells, animal tis-
sue sources are used for generation of respiratory tract cell culture models.
For this purpose, construction of an aireliquid interface culture of murine
respiratory epithelial was performed by preparation of respiratory epithelial
cells from dissected mouse trachea. The cells were cultured on collagen-
coated transwell cell culture inserts and ciliogenesis was induced with an
insulin, transferrin, and selenous acid containing (ITS) premix, retinoic
acid and by exposure of the apical cell surface to air (Fliegauf et al.,
2013). This kind of upper respiratory cell culture model effectively
simulates cell type-specific function including vigorous ciliary beating, al-

lows long-term analyses and is widely used to study airway clearance mech-
anisms of the host against respiratory pathogens like S. pneumoniae and P.
aeruginosa (Fliegauf et al., 2013; Woodworth et al., 2008). The EpiAir-
wayTM has been shown to have a pseudostratified ciliated epithelium
(Wengst and Reichl, 2010), which closely resembles the conditions in
the tracheobronchial epithelium (Wadell et al., 1999; Cotton et al., 1987).
Using this principal model, a reconstruction of nasal mucosa was re-
ported employing isolated human nasal fibroblasts in collagen matrix
covered by RPMI 2650 epithelial cells, which derived from squamous
cell carcinoma of the nasal septum (Wengst and Reichl, 2010). These con-
structs show a differentiated nonrespiratory-like epithelium and create
permeation barrier properties comparable to excised nasal mucosa (Wengst
and Reichl, 2010). RPMI 2650 cells are able to form a confluent cell layer
and develop sufficient TEER, as well as an appropriate permeation barrier,
when cultured at the aireliquid interface (Bai et al., 2008). Under these cul-
ture conditions, the cells also seemed to express tight junction proteins,
although no pseudostratified or ciliated morphology as in vivo could be
achieved. Furthermore, this method has been shown to induce the differen-
tiation of human nasal epithelial cells in primary and serial cultures better
than in liquid-covered cultures (Lee et al., 2005; Yeh et al., 2007). A disad-
vantage of EpiAirwayTM compared to an RPMI epithelial monolayer
model is the more complex handling of the constructs. Nevertheless, this
model is considered as a suitable permeation model for drug development,
and as suitable infection model to study pathogenehost interaction,
although no organotypic differentiation is achieved. In general, the use of
3D cell culture models enhances the repertoire of in vitro models to solve
specific scientific requirements, although they are not able to completely de-
pict the whole in vivo situation.
3.2 MicrogravitydVariations of 3D Cell Culture Models

Studies conducted during space flight by the United States National Aeronau-
tics and Space Agency showed that cell lines grown in suspension culture in low
shear microgravity tend to aggregate and exhibit morphologies more typical of
native tissues (Unsworth and Lelkes, 1998; Hammond and Hammond, 2001).
The low shear microgravity environment is also thought to be representative
of in vivo conditions in sites such as the uterus or within the brush border
microvilli, as predicted by mathematical modeling (Guo et al., 2000; Stock
and Vacanti, 2001). Based on these findings, current attempts to improve
the phenotypic expression of eukaryotic immortalized host cells have

involved the use of low shear stress in 3D culture conditions (Barrila et al.,
2010; Hjelm et al., 2010). To simulate low shear microgravity conditions
in the laboratory, rotating wall vessels (RWVs) were developed (Schwarz,
R.P., Wolf, D.A. Patent no. 5.026.650, 1991) (Hammond and Hammond,
2001). These sterile cylindrical chambers are rotated on a specialized motor-
ized stand within an incubator at a velocity that allows the cells to float freely
in suspension culture. Gas exchange is provided through a gas-permeable sil-
icon rubber membrane. ECM material such as collagen beads or sheets can
also be added to the bioreactor to support cell aggregation. By growing cells
in RWV bioreactors, tissue-derived cell cultures form 3D aggregate and
express many of the in vivo phenotypic characteristics (Barrila et al., 2010;
Carterson et al., 2005). Such an RWV-culture was generated with A549,
an immortalized human lung epithelial cell line, in order to study the infec-
tion process of Francisella tularensis (David et al., 2014). Data derived from un-
infected probes indicate that monolayer-cultured A549 cells display a high
level of cell-cycle activity and increased expression of oncogenes. This expres-
sion profile shows significant differences to normal lung epithelial cells.
Conversely, the RWV-cultured A549 cells appeared to be differentiated,
polarized, mucus producing cells, with a complex ECM, and are therefore
better representatives of the in vivo lung epithelial cells (Carterson et al.,
2005). A variation of this RWV-based 3D cell culture model is achieved
by providing microcarrier-bead surfaces for generation of polarized cellular
monolayers. Such a model is established to study the intracellular infection
process of Chlamydia trachomatis (Guseva et al., 2007). Hereby, the in vitro
cell culture system is based on cytodex collagen-coated microcarrier beads,
rotating in a spinner bottle together with different cells such as McCoy
(derived from knee joint synovial fluid from an arthritis patient), HEC-1B
(endometrial carcinoma cells), or HeLa cells (derived from cervix carcinoma).
The beads were kept slowly in suspension while a cell culture monolayer
grows on their surface. This system can be routinely monitored by phase mi-
croscopy (Guseva et al., 2007) and has been confirmed as more suitable to
study C. trachomatis infection since higher infection rates were achieved in
this 3D system than in 2D in vitro cultures. Nevertheless, it has to be kept
in mind that infection parameters also differ between various cell types used
in the 3D model system and are assumed to be a reflection of the fundamen-
tally different physiological functions between, for example, endometrial
versus endocervical epithelial cells (Guseva et al., 2007). Moreover, compar-
ison of 3D cell culture models using polarized cells with those using
nonpolarized cells revealed specific differences in the infection process

depending on the C. trachomatis serovar subtype. The luminal serovar E escape
from polarized endometrial epithelial cells via their apical domains to spread
canalicularly to the upper genital tract, whereas the invasive, lymphotrophic
serovar L2 exit via the basal domains (Wyrick et al., 1989). It is assumed
that this latter result occurs due to a different microtubule orientation. In
polarized cells, microtubules are oriented parallel to the lateral membrane
with an apical-to-basal axis, on which serovar L2 are known to traffic
(Clausen et al., 1997). In another model system, human intestinal Int-407 cells
and A549 lung epithelial cells were grown in microgravity and have been
used as tissue models of infection with Salmonella enterica serovar Typhimu-
rium and P. aeruginosa infections, respectively (Nickerson et al., 2001;
Carterson et al., 2005). In those studies, the tissue was propagated in micro-
gravity, but the bacterial challenge was conducted in normal gravity. It has to
be mentioned that microgravity is not only affecting eukaryotic host cells but
also alters gene expression of bacteria and also bacterial growth (Nickerson
et al., 2003, 2004). This is of importance since microarray analysis indicated
that S. enterica serovar Typhimurium grown in microgravity express different
subsets of genes than those grown in a normal gravitational field (Wilson et al.,
2002a,b).
4. ORGAN EQUIVALENTS AND TISSUE EXPLANTS

Regarding the cellular monolayer as simplest reductionist cell culture
model, a higher complexity is achieved by coculture of different cell types
and 3D culture models using a scaffold material and rotating culture condi-
tions. The next level of complexity is reached by cell aggregates connected
in tissue-like manner called “cellular spheroids” or “organoids.” Cellular
spheroids are simple 3D systems, which take advantage of the natural ten-
dency of many cell types to aggregate without the requirement of external
scaffolding material. The tissue counterparts of these suborgan structures are
common to most of the epithelial organs and are known as “acini” in mam-
mary tissue as well as lungs and “tubules” in the kidney (Carvalho et al.,
2005). These models offer several evaluation possibilities such as immune
fluorescence microscopic visualization of bacteriaecell interaction. Further-
more, the organoids also allow histological analyses after formalin fixation
and paraffin embedding. And subsequent to trypsin-based cell dissociation
from the organoids, single cell analysis is enabled by flow cytometry.
Moreover, after homogenization of organoid tissues, enzyme activities such

as function of alkaline phosphatase can be determined using commercially
available enzyme test kits (Carvalho et al., 2005). The simple spherical ge-
ometry allows for relatively easy modeling of dynamic processes, such as
growth and invasiveness of solid tumors (Stein et al., 2007; Jiang et al.,
2010). Cellular spheroids have become the system of choice for therapeuti-
cally orientated biomedical studies (M€ uller-Klieser, 1997; Sutherland, 1988;
Sutherland et al., 1971), are applied in biotechnology (Kale et al., 2000), and
are straightforward to apply in high-throughput screens (Ivascu and Kubbies,
2006; Zhang et al., 2005). Spheroids can be obtained from single cultures or
cocultures from a broad range of cell types (mono- or multicellular spher-
oids). They are produced either by the hanging drop technique (Timmins
et al., 2005; Kelm et al., 2003) or by using RWV cultures or other commer-
cially available rotating cell culture systems (RCCS) (Synthecon, Houston,
TX; Castaneda and Kinne, 2000; Unsworth and Lelkes, 1998; Nickerson
et al., 2004). Thereby, the above-mentioned RWV-systems provide suitable
conditions for the stable generation of tissue aggregates, such that growth
can be supported for several weeks. The highest level of cell culture
complexity is represented by reconstructed organoids and tissue equivalents
based on special cell types cultured under organ or tissue-specific conditions
and by tissue explants. The following examples are selected to elucidate
advantages and limitations by employing these highly sophisticated tissue
culture systems in infection biology.
4.1 Organoids and Tissue Equivalents Providing Complex

Cell Systems “En Miniature”
The human ileocecal colorectal adenocarcinoma-HCT-8 cell line was selected
to establish a 3D intestinal organoid model for evaluation of the interactions of
prototypic strains of the enterohemorrhagic Escherichia coli (EHEC) and
enteropathogenic Escherichia coli (EPEC) with the apical borders of such cells
under conditions of microgravity (Carvalho et al., 2005). This model was
based on hydrated cell culture sheets of small intestine submucosa (Cook
Biotech, West Lafayette, IN) providing a scaffold layer for HCT-8 cells.
The HCT-8 cells derived from enterocytes at the junction of the large and
small bowel (Tompkins et al., 1974) rapidly form 3D structures in a low shear
microgravity culture generated by the RCCS (from Synthecon, Houston,
TX). The size and surface area of the apical organoid cells appeared more
typical of normal intestinal epithelium than HCT-8 cells grown in mono-
layers. Similar to results obtained from simple 3D models, certain tissue
markers are better represented in 3D aggregates formed under microgravita-

tional conditions than in the same tissue types grown in 2D cultures
(Nickerson et al., 2001; Carterson et al., 2005). This also includes E-cadherin,
ZO-1, symplekin, and villin expression in HCT-8 organoids (Carvalho et al.,
2005). In addition, expression of disaccharidases and alkaline phosphatase was
significantly greater in the organoid-grown HCT-8 cells compared with cells
in 2D tissue culture. From these observations, it has been concluded that the
organoid tissue was more differentiated and a better representative than, for
example, HCT-8 cells cultured in monolayers (Carvalho et al., 2005). Note-
worthy, in contrast to polarized Caco-2 cells that grow in a single layer, the
HCT-8 organoid cells formed multilayered structures. As such, the interface
between the single layer of gut epithelial cells and the mesenchymal layer was
not recapitulated in the organoid model. Nonetheless, the morphology of the
assembled cells varied according to their positions within the aggregate, with
the surface layer of cells most resembling normal gut epithelium. Similarly,
villin expression was concentrated at the tissue surface compared to that
seen in underlying cells. This differential expression of villin is also seen in
vivo where epithelial cells express increasingly more villin as they mature
and move to the tips of the villi (Maunoury et al., 1992).
It has been postulated that the cells at the surface of the organoid receive
signals for differentiation from the low-shear microgravity fluid environ-
ment and thus provide a suitable model for bacterial interactions such as
EHEC-infection at the lumenal surface of gut tissue (Carvalho et al.,
2005). As mentioned above, the microgravity also influences bacterial
gene expression. This was reported for Salmonella grown under micro-
gravity, and also for the level of intimin production by EHEC (Carvalho
et al., 2005). A comparison of the data obtained from the 3D RCCW-
model with other infection models revealed some similarities such as
intimin-mediated pedestal formation. Nevertheless, results from other cell
culture models also elucidated differences with regard to the adherence
mechanisms depending on the expression of bundle forming pili and intimin
(Carvalho et al., 2005; Donnenberg and Kaper, 1992; Hicks et al., 1998).
Therefore, microgravity has an influence on bacteriaehost interaction,
which has to be taken into consideration in comparison with data obtained
from different infection models.
In infection biology, several bacteriaehost interactions, such as adher-
ence, replication, and invasion, are strongly influenced by environmental
conditions, which in turn depends on the physicochemical barrier properties
of the tissue surface and its nutrient availability. For instance, in contrast to
the wet environment required for 3D models of the nasal epithelium, asso-
ciation of bacteria with the human skin is a process that takes place under
relatively dry conditions (Shepherd et al., 2009). Some tissue-engineered
air - exposed human skin models are 3D systems, which to a high degree
mimic the native skin (El Ghalbzouri et al., 2008, 2009). Such epidermal
skin equivalents are generated by culturing primary keratinocytes at the
aireliquid interface on cell-free matrices (e.g., inert filters or de-epidermized
dermis). The keratinocytes proliferate, migrate, and differentiate during
epidermal development, resulting in skin equivalents that contain all layers
of the native epidermis and elicit barrier properties with many similarities
with the human skin (El Ghalbzouri et al., 2008; Thakoersing et al.,
2012). A 3D human skin equivalent has been applied to study skin coloni-
zation by Acinetobacter strains and to evaluate the effects of disinfectants and
other antimicrobial agents (Breij et al., 2014). Acinetobacter baumannii is able
to colonize the skin of hospitalized patients (Borer et al., 2007; Dijkshoorn
et al., 1987; Marchaim et al., 2007; Zeana et al., 2003), which can be a
source of infection and spread to other patients and the environment. As
expected, the skin equivalent model revealed several differences compared
to other in vitro cell culture models using monolayer cell culture on plastic
petri dishes with respect to biofilm formation and induction of inflammatory
responses (de Breij et al., 2010; Breij et al., 2014) and provided further
evidence for the importance of the stratum corneum as a protective barrier
against infections. As a specific type of 3D cell culture model, tissue equiv-
alents may provide a better environment to study pathogenehost interac-
tions, although this latter example clearly indicates that the quality of the
model also depends on the presence of cell structures indispensable for the
replique of a specific in vivo situation.
4.2 Tissue ExplantsdPiece of Reality

The ex vivo cocultivation of different cell types already represents a major
step toward more system complexity and provides valuable data of the
cell conglomerates in response to infective agents. The cultivation of specific
tissue explants grown in in vitro organ cultures (IVOC) provides a complex
multicellular and rather physiological environment for the study of hoste
pathogen interactions. The benefits and limitations of tissue explants will
be discussed based on three examples: explants of human lung, human ton-
sils, and porcine skin.
Tissue explants play an important role in in vitro analysis of bacterial path-
ogens showing restrictive species-specificities such as the human pathogen of
Legionnaires’ disease L. pneumophila for the human lung. Until recently,

mammalian models such as guinea pigs, mice, rhesus monkeys, and marmo-
sets were employed to address immunological, pathological, and pharmaco-
logical questions (Baskerville et al., 1981; Blanchard et al., 1987; Fitzgeorge
et al., 1983), although data from these animal models cannot be easily gener-
alized because of important interspecies differences in the expression, func-
tion, and localization of immune molecules (e.g., receptors, signaling
intermediates, response molecules). Cell culture assays on the other hand
lack the complex interaction networks between the specialized cell types
and extracellular components in the human lung. Therefore, tumor-free pul-
monary tissue samples were developed as novel infection model for L. pneu-
mophila. Histopathological analyses revealed that this approach narrows the
gap between current infection models and actual human infections. It allows
to characterize tissue damage, bacterial dissemination, and the host’s molecu-
lar response after an infection with L. pneumophila including divers proteome
and transcriptome-based analyses (J€ager et al., 2014; Shevchuk et al., in press).
A similar ex vivo infection model of human lung tissue was also estab-
lished for the analysis of pneumonia infections induced by pneumococci
with the focus on detection of prostaglandin production in infected lung
(Szymanski et al., 2012). This model was based on former description of a
resolving infection model of mice lung tissue (Dockrell et al., 2003) and
on an ex vivo model of acute chlamydial infection, using human lung tissue
explants for interaction studies with the obligate intracellular pathogen Chla-
mydia pneumophila (Rupp et al., 2004). The lung explant infection enabled
the determination of the prostaglandin expression profile, the cellular local-
ization of prostaglandin production, and the decipherment of underlying
signaling pathways (Szymanski et al., 2012).
As advantage of the lung explant model, the lung cell types are still orga-
nized into the unique lung architecture. Therefore, cell-specific behavior
can be studied. Resident cells including alveolar macrophages are still pre-
sent in the tissue and are capable of contributing to the observed innate im-
mune response as shown previously by Xu and coworker (2008). However,
the lung explant infection does not allow for investigations of aspects of im-
munity such as the recruitment of immune cells from the blood circulation
(Szymanski et al., 2012).
Tonsil explants represent a common model to mimic, for instance
S. pyogenes-induced tonsillitis and pharyngitis, which are rarely life-threat-
ening, but among the most frequent of human infectious diseases affecting
both adults and children (Bell et al., 2012). With the aim to further
investigate bacterial adherence following the infection progress, immortal-

ized human keratinocytes (HaCaT cells) were employed as a model for
the tonsil epithelium. Additionally, tonsil explants of the control group
and of patients suffering from recurrent acute tonsillitis were challenged in
vitro with an M1-type S. pyogenes (Abbot et al., 2007). A comparison be-
tween the streptococcal effects on expression of host defense peptides in ton-
sils derived from different patients revealed a marked variability related
specifically to individual tonsil samples within each of the groups. In these
experiments, this kind of variability tended to mask any significant trends.
A further application of tissue explants has been described in biofilm
research. While in vitro biofilm models using nonbiological surfaces suffi-
ciently enable the generation of robust biofilms under certain physical and
chemical environmental conditions, biological tissues such as dermal sub-
strates contribute to a high degree to biofilm formation by providing attach-
ment points and nutrition. These tissues strongly influence size and
conformation of the biofilm and alter the reactivity of the agents being
tested. As example, porcine skin was applied in a microbial biofilm model
as both, the substrate for attachment and the primary source of nutrition
(Yang et al., 2013). The produced biofilm resembled more closely the char-
acteristics of biofilms that are found in human wounds. In addition, this
model can be used to assess the direct efficacy of antimicrobial dressings
against mature biofilms.
In general, the use of tissue explants enables pathogenicity research in a
very physiological environment ensuring species-specific unique character-
istics. Nevertheless, the increased complexity of the explanted tissue also ac-
cumulates individual differences, which interfere with the cooperative
responses of single cell types to the infection challenge. In order to generate
principal and reproducible statements, a defined selection of donors is an
important precaution for such type of infection model. Unfortunately, these
culture systems are currently available only to facilities equipped to collect
and rapidly convey biopsy materials to the research laboratory. In addition,
IVOC studies are limited to bacterialehost interactions that occur during
the short life span of the organ biopsy material.
4.3 Integration of Microfluidic Systems in 2D and 3D Cell

Culture
3D scaffolds provide tissue-like connectivity, although they are quite
limited in controlling the cell culture conditions, in nutrient and drug de-
livery, and in running simultaneous assays during cell culture (Beebe,
2013). However, building 3D vascularized organs remains the main

technological barrier to be overcome. One of the major challenges is the
inclusion of a vascular network to support cell viability in terms of nutrients
and oxygen perfusion. Irrespective of cultivation in two or three dimen-
sions, cell growth in vivo is significantly affected by the diffusion-limited
distribution of oxygen, nutrients, and other molecules (Minchinton and
Tannock, 2006). Oxygen, nutrients, and other molecules are continuously
consumed and produced by cells. Such dynamic distributions are not
mimicked in conventional 2D or 3D cell culture (Martinez et al., 2008;
Yamada and Cukierman, 2007). For an ideal 3D cell culture system,
continuous nutrition and oxygen supply, and waste removal through the
culture medium, must be ensured. The microenvironment provided by
microfluidic systems should be able to mimic this in vivo. Thus, integration
of microfluidics with such 3D scaffolding systems allows dynamic manip-
ulation of culture conditions biochemically and biomechanically and pro-
vides a microenvironment that allows formation of artificial tissues from
cultured cells (Trietsch et al., 2013; Choi et al., 2007). An exemplary
technical setup, which combines tissue explant culture with microfluidic
medium supply, is depicted in Figure 2. In the 1990s, the development
of microfluidic technology created a platform for highly complex and
dynamic microenvironments that are controllable, reproducible, and
adaptable to specific cell culture situations. Microfluidic cell culture allows
controlling fluid flow in the micrometer and nanometer scale in precisely
defined geometries and facilitates simultaneous manipulation and analysis
starting from a single cell level to larger populations and up to tissues
cultured on fully integrated and automated chips (Mehling and Tay,
2014). Microfluidic systems add several benefits to in vitro cell culture.
For example, the microscale dimensions of such microfluidic systems are
compatible with those of many microstructures and environments native
to in vivo systems. For example, the distance between adjacent capillaries
in many in vivo animal tissue models is in the microscale region. Moreover,
some substrates like polydimethylsiloxane (PDMS) used in microfluidic
devices are permeable to oxygen, an important factor influencing cell pro-
liferation. A microfluidic system using PDMS chips also enables the culture
of blood vessel cells on the inner surface of microchannels where flow and
shear stresses of the blood circulation can be controlled (Fiddes et al., 2010).
Thus, this system mimics functional aspects of the vasculature (Schimek
et al., 2013). Different technical solutions are commercially available and
most of them integrate multiple steps such as cell culture, cell sampling,
Figure 2 A trendsetting tissue culture model combines the use of tissue explants with
microfluidic perfusion systems (Photograph used with kind permission from ibidi GmbH,
Germany.), thereby generating an autonomous ex vivo system. This system acts as a
vascular tubing, which supports the metabolism requirements and gas exchange of
the tissue explant for longer cultivation periods.
fluid control, cell capture, cell lysis, mixing, and detection on a single
device. The different microfluidic systems have been categorized as
glass/silicon-based, polymer-based, and paper-based platforms, based on
the substrates used for microdevice fabrication (Li et al., 2012). A detailed
overview about the benefits of each single system is provided in a compre-
hensive review by Li and colleagues (Li et al., 2012).
The use of microfluidic system provides new insights into infection
biology. For example, in order to study the interaction of pathogens with
the endothelial layer of blood vessels, human endothelial cells were effec-
tively cultured to confluence in gel-free microslides, followed by infection
with pathogenic Staphylococcus aureus bacteria (Pappelbaum et al., 2013).
The application of a continuous and defined shear force induces significant
changes in cell morphology and replication rates resulting in cell shapes
typical for the functional endothelium of blood vessels. The polymer-based
microslides allowed a microscopic visualization of long vWF (von Willebrand

factor)-fibers, which were secreted by the endothelium in response to an
infection stimulus. The vWF was bound by S. aureus bacteria, thereby
demonstrating the use of host-derived vWF as bacterial adhesion cofactor
(Pappelbaum et al., 2013). The employment of microfluidic systems in gen-
eration of synthetic capillaries also enabled investigation of pathogenicity
mechanisms of the malaria causing agent Plasmodium falciparum. Significant
results were obtained regarding the adhesion of infected red blood cells to
host cell ligands, the rheological responses to changing dimensions of capil-
laries with shapes and sizes similar to small blood vessels, and the phagocytosis
of infected erythrocytes by macrophages (Antia et al., 2007). Moreover, a
microfluidic 3D bone tissue model was established for high-throughput eval-
uation of wound-healing and infection-preventing biomaterials (Lee et al.,
2012). Osteoblasts are not able to form any 3D structure beyond a confluent
layer during conventional 2D culture. However, 3D bone tissue-like struc-
tures can be formed by long-term dynamic culture in microfluidic chambers
as a result of the proliferation of murine preosteoblasts, thereby forming a
confluent layer on the bottom chamber surface. Based on a 3D aggregation
with produced collagen and calcium, the mineralized 3D tissue is formed by
the self-organization of osteoblasts in the microfluidic chambers. The dy-
namic flow of the microfluidic device emulates the nutrient and waste trans-
port function of the microcirculation. Moreover, the microscale geometrical
form confines the culture chambers analogous to the size of micrometer-scale
pores present in 3D scaffolds in tissue engineering. The tissue morphology,
which is generated by the microfluidic device, is composed of randomly ori-
ented collagen fibers and contains calcified materials and osteocytes. This tis-
sue strongly resembles those of primary bone tissue and may provide a basis
model for further biochemical and mechanical processes involved in bacteria-
induced bone infection (Lee et al., 2012).
Other examples are renal and hepatic cells that have also been suc-
cessfully cultured in close correspondence to the microarchitecture of
the respective tissues (Lee et al., 2007). In addition to these homotypic tis-
sue culture models, heterotypic tissue culture models that mimic the
respective tissue closely both from a histologic as well as from a physiolog-
ical and functional point of view have been achieved in microfluidic cell
culture devices (Ho et al., 2013; Huh et al., 2010). This allows high-
throughput pharmacological studies and might ultimately result in using
microfluidic cell culture systems also for regenerative purposes (Harink
et al., 2013).
5. CONCLUDING REMARKS AND FUTURE

PERSPECTIVES
The use of eukaryotic cells or protozoa in 2D monolayer culture is
irreplaceable in many areas of infection research. Nevertheless, the physio-
logical relevance of the information retrieved from in vitro studies is often
quite limited and requires additional confirmation. Dynamic distributions
of oxygen and nutrients are not mimicked in conventional 2D cell culture,
and stimulatory effects of a highly complex 3D environment on cell growth
and cell function are not recapitulated by the 2D cell culture. In order to
improve the semiphysiological environment with single cell type culture,
some trendsetting technical developments are currently in progress. With
the aim to analyze tissue-specific hostepathogen interactions, the current
development in monolayer-based 2D cell culture techniques is focused on
the preparation of highly specialized cell types from all relevant tissues,
which maintain the high differentiation level in optimized cell culture me-
dia. This task is complemented by various projects aiming at optimization of
immortalization techniques for the generation of cell lines with long-term
replication activities retaining cell type-specific differentiation and function-
ality (Schmedt et al., 2012; Robin et al., 2015). The ability to cultivate more
sensitive endothelial cells for longer periods of time in in vitro systems will
offer the possibility to expand the scientific knowledge about tissue destruc-
tive inflammation responses like cytokine release, procoagulative reactions,
and key signaling patterns. In this regard, some recent developments in stem
cell research might also invent new options to refine and facilitate the gen-
eration of tissue models independent of the availability of specific cell mate-
rial or tissues (McCracken et al., 2014). Primary keratinocytes have been
successfully used for generation of human epidermal equivalents (HEEs),
but only a limited number of HEEs can be generated from one sample of
epidermis (G€ otz et al., 2012). In order to develop an HEE model that can
be produced in an unlimited number of genetically identical units, human
embryonic stem cells (hESCs) were used to induce pluripotent stem cells
(iPSCs) (Petrova et al., 2014). The stem cells are primary cells that are
capable of infinite proliferation and whose genetic footprint can be fully
characterized. The hESCs/iPSCs can be stimulated to differentiate into ker-
atinocytes with gene expression profiles similar to those of normal human
keratinocytes (Petrova et al., 2014). These hESC/iPSC-derived keratino-
cytes were used to generate HEEs in an aireliquid interface culture exposed
to a sequential high-to-low humidity environment. HEEs generated from
hESC/iPSC-derived keratinocytes expose a functional permeability barrier

similar to native human skin and are indistinguishable from HEEs generated
from tissue-derived human keratinocytes under the same condition. This
model has been considered as suitable alternative for elucidating the molec-
ular mechanisms of barrier development, perturbation, and recovery, as well
as how mutations of genes involved in cornification and lipid metabolism
affect permeability barrier homeostasis (Mildner et al., 2010; Simpson
et al., 2010). The use of pluripotent stem cells for cell culture models may
provide new options for generation of various tissue models in infection
biology.
The step from 2D cell culture toward 3D culture systems is hallmarked
by providing scaffolds for generation of tissue-like structures. This higher
level of complexity is represented by a huge and extending variety of tech-
nical systems including coculture, microgravity, and various perfusable
setups. The other side of the coin is that handling of 3D cell culture requires
a higher level of technical knowledge and experience in order to generate
stable systems allowing reproducible infection analyses with respect to iden-
tical matrix content and texture. Organoids and tissue equivalents are
currently the best in vivo-simulating systems, although the lack of a contin-
uous nutrient supply and oxygen exchange limit possible applications.
Recent reports describe successful attempts to combine engineered and
explanted tissues with a microfluidic device providing long-term perfusion.
This trendsetting technique holds enormous potential for basic and applied
research in infection biology.
ACKNOWLEDGMENTS
To provide a focused and clear review, we were forced to select representative examples and
citations, and apologize for not mentioning all groups working with cell culture models in
infection biology. We thank Janine Rasch for providing photographs and the Deutsche For-
schungsgemeinschaft (DFG) for financial support. Authors’ own research was funded by DFG
grants (BE 4570/4-1 and STE 838/8-1).
REFERENCES
Abbot, E.L., Smith, W.D., Siou, G.P., Chiriboga, C., Smith, R.J., Wilson, J.A., Hirst, B.H.,
Kehoe, M.A., 2007. Pili mediate specific adhesion of Streptococcus pyogenes to human
tonsil and skin. Cell. Microbiol. 9, 1822e1833.
Abbott, N.J., 2005. Dynamics of CNS barriers: evolution, differentiation, and modulation.
Cell. Mol. Neurobiol. 25, 5e23.
Agarwal, V., Sroka, M., Fulde, M., Bergmann, S., Riesbeck, K., Blom, A.M., 2014. Binding
of Streptococcus pneumoniae endopeptidase O (PepO) to complement component C1q
modulates the complement attack and promotes host cell adherence. J. Biol. Chem.
289 (22), 15833e15844.
Agarwal, V., Kuchipudi, A., Fulde, M., Riesbeck, K., Bergmann, S., Bloom, A., 2013. Strep-
tococcus pneumoniae endopeptidase O (PepO): a multifunctional plasminogen and fibro-
nectin binding protein, facilitating evasion of innate immunity and invasion of host
cells. J. Biol. Chem. 288 (10), 6849e6863.
Agu, R.U., Valiveti, S., Earles, D.C., Klausner, M., Hayden, P.J., Wermeling, D.P.,
Stinchcomb, A.L., 2004. Intranasal delivery of recombinant human parathyroid hor-
mone [hPTH (1-34)], teriparatide in rats. Endocr. Res. 30, 455e467.
Allombert, J., Lazzaroni, J.C., Bailo, N., Gilbert, C., Charpentier, X., Doublet, P.,
Vianney, A., 2014. Three antagonistic cyclic di-GMP-catabolizing enzymes promote
differential Dot/Icm effector delivery and intracellular survival at the early steps of Legion-
ella pneumophila infection. Infect. Immun. 82, 1222e1233.
Amelung, S., Nerlich, A., Rohde, M., Spellerberg, B., Cole, J.N., Nizet, V.,
Chhatwal, G.S., Talay, S.R., 2011. The FbaB-type fibronectin-binding protein of
Streptococcus pyogenes promotes specific invasion into endothelial cells. Cell. Microbiol.
13 (8), 1200e1211.
Antia, M., Herricks, T., Rathod, P.K., 2007. Microfluidic modeling of cell-cell interactions
in malaria pathogenesis. PLoS Pathog. 3, e99.
Attali, C., Durmort, C., Vernet, T., Di Guilmi, A.M., 2008a. The interaction of Streptococcus
pneumoniae with plasmin mediates transmigration across endothelial and epithelial mono-
layers by intercellular junction cleavage. Infect. Immun. 76, 5350e5356.
Attali, C., Frolet, C., Durmort, C., Offant, J., Vernet, T., Di Guilmi, A.M., 2008b. Strepto-
coccus pneumoniae choline-binding protein E interaction with plasminogen/plasmin stim-
ulates migration across the extracellular matrix. Infect. Immun. 76, 466e476.
Badger, J.L., Stins, M.F., Kim, K.S., 1999. Citrobacter freundii invades and replicates in human
brain microvascular endothelial cells. Infect. Immun. 67, 4208e4215.
Bai, S., Yang, T., Abbruscato, T.J., Ahsan, F., 2008. Evaluation of human nasal RPMI 2650
cells grown at an air-liquid interface as a model for nasal drug transport studies. J. Pharm.
Sci. 97, 1165e1178.
Banks, W.A., 1999. Physiology and pathology of the bloodebrain barrier: implications for
microbial pathogenesis, drug delivery and neurodegenerative disorders. J. Neurovirol.
5, 538e555.
Banks, W.A., Kastin, A.J., Arimura, A., 1998. Effect of spinal cord injury on the permeability
of the bloodebrain and blood-spinal cord barriers to the neurotropin PACAP. Exp.
Neurol. 151, 116e123.
Barrila, J., Radtke, A.L., Crabbe, A., Sarker, S.F., Herbst-Kralovetz, M.M., Ott, C.M.,
Nickerson, C.A., 2010. Organotypic 3D cell culture models: using the rotating wall
vessel to study host-pathogen interactions. Nat. Rev. Microbiol. 8, 791e801.
Baskerville, A., Fitzgeorge, R.B., Broster, M., Hambleton, P., Dennis, P.J., 1981. Experi-
mental transmission of legionaires’ disease by exposure to aerosols of Legionella
pneumophila. Lancet 2, 1389e1390.
Beebe, D.C., 2013. The use of cell lines to “model” ocular tissues: cautionary tales. Invest
Ophthalmol. Vis. Sci. 54.
Bell, S., Howard, A., Wilson, J.A., Abbot, E.L., Smith, W.D., Townes, C.L., Hirst, B.H.,
Hall, J., 2012. Streptococcus pyogenes infection of tonsil explants is associated with a human
beta-defensin 1 response from control but not recurrent acute tonsillitis patients. Mol.
Oral Microbiol. 27, 160e171.
Bergmann, S., Lang, A., Rohde, M., Rennemeier, C., Grashoff, C., Preissner, K.T.,
Hammerschmidt, S., 2009. Integrin-linked kinase is required for vitronectin-mediated
internalization of Streptococcus pneumoniae by host cells. J. Cell Sci. 122, 256e267.
Bergmann, S., Schoenen, H., Hammerschmidt, S., 2013. The interaction between bacterial
enolase and plasminogen promotes adherence of Streptococcus pneumoniae to epithelial and
endothelial cells. Int. J. Med. Microbiol. 303, 452e462.
Birkness, K.A., Swisher, B.L., White, E.H., Long, E.G., Ewing Jr., E.P., Quinn, F.D., 1995.
A tissue culture bilayer model to study the passage of Neisseria meningitidis. Infect. Immun.
63, 402e409.
Bissell, M.J., Radisky, D.C., Rizki, A., Weaver, V.M., Petersen, O.W., 2002. The organizing
principle: microenvironmental influences in the normal and malignant breast. Differen-
tiation 70, 537e546.
Blanchard, D.K., Stewart 2nd, W.E., Klein, T.W., Friedman, H., Djeu, J.Y., 1987. Cytolytic
activity of human peripheral blood leukocytes against Legionella pneumophila-infected
monocytes: characterization of the effector cell and augmentation by interleukin 2.
J. Immunol. 139, 551e556.
Boekema, B.K., Stockhofe-Zurwieden, N., Smith, H.E., Kamp, E.M., van Putten, J.P.,
Verheijden, J.H., 2003. Adherence of Actinobacillus pleuropneumoniae to primary cultures
of porcine lung epithelial cells. Vet. Microbiol. 93 (2), 133e144.
Borer, A., Gilad, J., Porat, N., Megrelesvilli, R., Saidel-Odes, L., Peled, N., Eskira, S.,
Schlaeffer, F., Almog, Y., 2007. Impact of 4% chlorhexidine whole-body washing on
multidrug-resistant Acinetobacter baumannii skin colonisation among patients in a medical
intensive care unit. J. Hosp. Infect. 67, 149e155.
Bozzaro, S., Buracco, S., Peracino, B., 2013. Iron metabolism and resistance to infection by
invasive bacteria in the social amoeba Dictyostelium discoideum. Front. Cell. Infect. Micro-
biol. 3, 50.
Breij, E.C., de Goeij, B.E., Verploegen, S., Schuurhuis, D.H., Amirkhosravi, A., Francis, J.,
Miller, V.B., Houtkamp, M., Bleeker, W.K., Satijn, D., Parren, P.W., 2014. An anti-
body-drug conjugate that targets tissue factor exhibits potent therapeutic activity against
a broad range of solid tumors. Cancer Res. 74, 1214e1226.
Carterson, A.J., Honer zu Bentrup, K., Ott, C.M., Clarke, M.S., Pierson, D.L.,
Vanderburg, C.R., Buchanan, K.L., Nickerson, C.A., Schurr, M.J., 2005. A549 lung
epithelial cells grown as three-dimensional aggregates: alternative tissue culture model
for Pseudomonas aeruginosa pathogenesis. Infect. Immun. 73, 1129e1140.
Carvalho, H.M., Teel, L.D., Goping, G., O’Brien, A.D., 2005. A three-dimensional tissue
culture model for the study of attach and efface lesion formation by enteropathogenic
and enterohaemorrhagic Escherichia coli. Cell. Microbiol. 7, 1771e1781.
Castaneda, F., Kinne, R.K., 2000. Short exposure to millimolar concentrations of ethanol
induces apoptotic cell death in multicellular HepG2 spheroids. J. Cancer Res. Clin.
Oncol. 126, 305e310.
Cecchelli, R., Berezowski, V., Lundquist, S., Culot, M., Renftel, M., Dehouck, M.P.,
Fenart, L., 2007. Modelling of the bloodebrain barrier in drug discovery and
development. Nat. Rev. Drug Discov. 6, 650e661.
Chen, S.C., Eiting, K., Cui, K., Leonard, A.K., Morris, D., Li, C.Y., Farber, K., Sileno, A.P.,
Houston Jr., M.E., Johnson, P.H., Quay, S.C., Costantino, H.R., 2006. Therapeutic
utility of a novel tight junction modulating peptide for enhancing intranasal drug
delivery. J. Pharm. Sci. 95, 1364e1371.
Chin, A.C., Parkos, C.A., 2007. Pathobiology of neutrophil transepithelial migration: impli-
cations in mediating epithelial injury. Annu. Rev. Pathol. 2, 111e143.
Choi, N.W., Cabodi, M., Held, B., Gleghorn, J.P., Bonassar, L.J., Stroock, A.D., 2007.
Microfluidic scaffolds for tissue engineering. Nat. Mater. 6, 908e915.
Clarke, M., 2010. Recent insights into host-pathogen interactions from Dictyostelium. Cell.
Microbiol. 12, 283e291.
Clausen, J.D., Christiansen, G., Holst, H.U., Birkelund, S., 1997. Chlamydia trachomatis uti-
lizes the host cell microtubule network during early events of infection. Mol. Microbiol.
25, 441e449.
Comstock, L.E., Thomas, D.D., 1991. Characterization of Borrelia burgdorferi invasion of
cultured endothelial cells. Microb. Pathog. 10, 137e148.
Cossart, P., Lecuit, M., 1998. Interactions of Listeria monocytogenes with mammalian cells dur-
ing entry and actin-based movement: bacterial factors, cellular ligands and signaling.
EMBO J. 17, 3797e3806.
Cotton, C.U., Stutts, M.J., Knowles, M.R., Gatzy, J.T., Boucher, R.C., 1987. Abnormal
apical cell membrane in cystic fibrosis respiratory epithelium. An in vitro electrophysio-
logic analysis. J. Clin. Invest. 79, 80e85.
Cucullo, L., Hossain, M., Rapp, E., Manders, T., Marchi, N., Janigro, D., 2007. Develop-
ment of a humanized in vitro bloodebrain barrier model to screen for brain penetration
of antiepileptic drugs. Epilepsia 48, 505e516.
Cukierman, E., Pankov, R., Stevens, D.R., Yamada, K.M., 2001. Taking cell-matrix adhe-
sions to the third dimension. Science 294, 1708e1712.
David, J., Sayer, N.M., Sarkar-Tyson, M., 2014. The use of a three-dimensional cell culture
model to investigate host-pathogen interactions of Francisella tularensis in human lung
epithelial cells. Microbes Infect. 16, 735e745.
de Breij, A., Dijkshoorn, L., Lagendijk, E., van der Meer, J., Koster, A., Bloemberg, G.,
Wolterbeek, R., van den Broek, P., Nibbering, P., 2010. Do biofilm formation and in-
teractions with human cells explain the clinical success of Acinetobacter baumannii? PLoS
One 5, e10732.
De Vries, G., McLaughlin, A., Rhodes, J., 1997. The immunomodulatory actions of E-type
prostaglandins. Expert Opin. Investig. Drugs 6, 7e16.
Delcommenne, M., Streuli, C.H., 1995. Control of integrin expression by extracellular
matrix. J. Biol. Chem. 270, 26794e26801.
Dessus-Babus, S., Darville, T.L., Cuozzo, F.P., Ferguson, K., Wyrick, P.B., 2002. Differ-
ences in innate immune responses (in vitro) to HeLa cells infected with nondisseminating
serovar E and disseminating serovar L2 of Chlamydia trachomatis. Infect. Immun. 70,
3234e3248.
Dhimolea, E., Maffini, M.V., Soto, A.M., Sonnenschein, C., 2010. The role of collagen
reorganization on mammary epithelial morphogenesis in a 3D culture model. Biomate-
rials 31, 3622e3630.
Dijkshoorn, L., Michel, M.F., Degener, J.E., 1987. Cell envelope protein profiles of Acine-
tobacter calcoaceticus strains isolated in hospitals. J. Med. Microbiol. 23, 313e319.
DiMilla, P.A., Barbee, K., Lauffenburger, D.A., 1991. Mathematical model for the effects of
adhesion and mechanics on cell migration speed. Biophys. J. 60, 15e37.
Discher, D.E., Janmey, P., Wang, Y.L., 2005. Tissue cells feel and respond to the stiffness of
their substrate. Science 310, 1139e1143.
Dockrell, D.H., Marriott, H.M., Prince, L.R., Ridger, V.C., Ince, P.G., Hellewell, P.G.,
Whyte, M.K., 2003. Alveolar macrophage apoptosis contributes to pneumococcal clear-
ance in a resolving model of pulmonary infection. J. Immunol. 171, 5380e5388.
Donnenberg, M.S., Kaper, J.B., 1992. Enteropathogenic Escherichia coli. Infect. Immun. 60,
3953e3961.
Duell, B.L., Cripps, A.W., Schembri, M.A., Ulett, G.C., 2011. Epithelial cell coculture
models for studying infectious diseases: benefits and limitations. J. Biomed. Biotechnol.
2011, 852419.
El Ghalbzouri, A., Commandeur, S., Rietveld, M.H., Mulder, A.A., Willemze, R., 2009.
Replacement of animal-derived collagen matrix by human fibroblast-derived dermal
matrix for human skin equivalent products. Biomaterials 30, 71e78.
El Ghalbzouri, A., Siamari, R., Willemze, R., Ponec, M., 2008. Leiden reconstructed human
epidermal model as a tool for the evaluation of the skin corrosion and irritation potential
according to the ECVAM guidelines. Toxicol. Vitro 22, 1311e1320.
Elm, C., Braathen, R., Bergmann, S., Frank, R., Vaerman, J.P., Kaetzel, C.S.,
Chhatwal, G.S., Johansen, F.E., Hammerschmidt, S., 2004. Ectodomains 3 and 4 of
human polymeric Immunoglobulin receptor (hpIgR) mediate invasion of Streptococcus

pneumoniae into the epithelium. J. Biol. Chem. 279 (8), 6296e6304.
Fiddes, L.K., Raz, N., Srigunapalan, S., Tumarkan, E., Simmons, C.A., Wheeler, A.R.,
Kumacheva, E., 2010. A circular cross-section PDMS microfluidics system for replication
of cardiovascular flow conditions. Biomaterials 31, 3459e3464.
Fitzgeorge, R.B., Baskerville, A., Broster, M., Hambleton, P., Dennis, P.J., 1983. Aerosol
infection of animals with strains of Legionella pneumophila of different virulence: compar-
ison with intraperitoneal and intranasal routes of infection. J. Hyg. (Lond.) 90, 81e89.
Fliegauf, M., Sonnen, A.F., Kremer, B., Henneke, P., 2013. Mucociliary clearance defects in
a murine in vitro model of pneumococcal airway infection. PLoS One 8, e59925.
Fraley, S.I., Feng, Y., Krishnamurthy, R., Kim, D.H., Celedon, A., Longmore, G.D.,
Wirtz, D., 2010. A distinctive role for focal adhesion proteins in three-dimensional
cell motility. Nat. Cell Biol. 12, 598e604.
Freshney, R.I., 2005. Culture of Animal Cells: A Manual of Basic Technique, fifth ed.
Wiley-Liss Publication, Wiley-Blackwell.
Friedl, P., 2004. Prespecification and plasticity: shifting mechanisms of cell migration. Curr.
Opin. Cell Biol. 16, 14e23.
Fusai, T., Parzy, D., Spillmann, D., Eustacchio, F., Pouvelle, B., Lepolard, C., Scherf, A.,
Gysin, J., 2000. Characterisation of the chondroitin sulphate of Saimiri brain microvas-
cular endothelial cells involved in Plasmodium falciparum cytoadhesion. Mol. Biochem.
Parasitol. 108, 25e37.
Garcia-Monco, J.C., Villar, B.F., Alen, J.C., Benach, J.L., 1990. Borrelia burgdorferi in the
central nervous system: experimental and clinical evidence for early invasion. J. Infect.
Dis. 161, 1187e1193.
Gath, U., Hakvoort, A., Wegener, J., Decker, S., Galla, H.J., 1997. Porcine choroid plexus
cells in culture: expression of polarized phenotype, maintenance of barrier properties and
apical secretion of CSF-components. Eur. J. Cell Biol. 74, 68e78.
Gelain, F., Bottai, D., Vescovi, A., Zhang, S., 2006. Designer self-assembling peptide nano-
fiber scaffolds for adult mouse neural stem cell 3-dimensional cultures. PLoS One 1,
e119.
Gerstenmaier, L., Pilla, R., Herrmann, L., Hermann, H., Prado, M., Villafano, G.J.,
Kolonko, M., Reimer, R., Soldati, T., King, J.S., Hagedorn, M., 2015. The autophagy
machinery ensures nonlytic transmission of mycobacteria. Proc. Natl. Acad. Sci. U.S.A.
112 (7), E687eE692.
Gotz, C., Pfeiffer, R., Tigges, J., Blatz, V., Jackh, C., Freytag, E.M., Fabian, E.,
Landsiedel, R., Merk, H.F., Krutmann, J., Edwards, R.J., Pease, C., Goebel, C.,
Hewitt, N., Fritsche, E., 2012. Xenobiotic metabolism capacities of human skin in
comparison with a 3D epidermis model and keratinocyte-based cell culture as in vitro
alternatives for chemical testing: activating enzymes (Phase I). Exp. Dermatol. 21,
358e363.
Greiffenberg, L., Goebel, W., Kim, K.S., Weiglein, I., Bubert, A., Engelbrecht, F., Stins, M.,
Kuhn, M., 1998. Interaction of Listeria monocytogenes with human brain microvascular
endothelial cells: InlB-dependent invasion, long-term intracellular growth, and spread
from macrophages to endothelial cells. Infect. Immun. 66, 5260e5267.
Guo, P., Weinstein, A.M., Weinbaum, S., 2000. A hydrodynamic mechanosensory hypoth-
esis for brush border microvilli. Am. J. Physiol. Ren. Physiol. 279, F698eF712.
Guseva, N.V., Dessus-Babus, S., Moore, C.G., Whittimore, J.D., Wyrick, P.B., 2007. Dif-
ferences in Chlamydia trachomatis serovar E growth rate in polarized endometrial and
endocervical epithelial cells grown in three-dimensional culture. Infect. Immun. 75,
553e564.
Haake, D.A., Lovett, M.A., 1994. Interjunctional invasion of endothelial cell monolayers.
Methods Enzymol. 236, 447e463.
Hagele, S., Kohler, R., Merkert, H., Schleicher, M., Hacker, J., Steinert, M., 2000. Dictyos-
telium discoideum: a new host model system for intracellular pathogens of the genus
Legionella. Cell. Microbiol. 2, 165e171.
Hammond, T.G., Hammond, J.M., 2001. Optimized suspension culture: the rotating-wall
vessel. Am. J. Physiol. Ren. Physiol. 281, F12eF25.
Harink, B., Le Gac, S., Truckenmuller, R., van Blitterswijk, C., Habibovic, P., 2013.
Regeneration-on-a-chip? The perspectives on use of microfluidics in regenerative
medicine. Lab. Chip 13, 3512e3528.
Haselbach, M., Wegener, J., Decker, S., Engelbertz, C., Galla, H.J., 2001. Porcine Choroid
plexus epithelial cells in culture: regulation of barrier properties and transport processes.
Microsc. Res. Tech. 52, 137e152.
Hicks, S., Frankel, G., Kaper, J.B., Dougan, G., Phillips, A.D., 1998. Role of intimin and
bundle-forming pili in enteropathogenic Escherichia coli adhesion to pediatric intestinal
tissue in vitro. Infect. Immun. 66, 1570e1578.
Hilbi, H., Hoffmann, C., Harrison, C.F., 2011. Legionella spp. Outdoors: colonization,
communication and persistence. Environ. Microbiol. 3, 286e296.
Hilbi, H., Weber, S.S., Ragaz, C., Nyfeler, Y., Urwyler, S., 2007. Environmental predators
as models for bacterial pathogenesis. Environ. Microbiol. 9, 563e575.
Hjelm, B.E., Berta, A.N., Nickerson, C.A., Arntzen, C.J., Herbst-Kralovetz, M.M., 2010.
Development and characterization of a three-dimensional organotypic human vaginal
epithelial cell model. Biol. Reprod. 82, 617e627.
Hirst, R.A., Kadioglu, A., O’Callaghan, C., Andrew, P.W., 2004. The role of pneumolysin
in pneumococcal pneumonia and meningitis. Clin. Exp. Immunol. 138 (2), 195e201.
Ho, C.T., Lin, R.Z., Chen, R.J., Chin, C.K., Gong, S.E., Chang, H.Y., Peng, H.L., Hsu, L.,
Yew, T.R., Chang, S.F., Liu, C.H., 2013. Liver-cell patterning lab chip: mimicking the
morphology of liver lobule tissue. Lab. Chip 13, 3578e3587.
Hopper, S., Wilbur, J.S., Vasquez, B.L., Larson, J., Clary, S., Mehr, I.J., Seifert, H.S., So, M.,
2000. Isolation of Neisseria gonorrhoeae mutants that show enhanced trafficking across
polarized T84 epithelial monolayers. Infect. Immun. 68, 896e905.
Horii, A., Wang, X., Gelain, F., Zhang, S., 2007. Biological designer self-assembling peptide
nanofiber scaffolds significantly enhance osteoblast proliferation, differentiation and 3-D
migration. PLoS One 2, e190.
Huang, S.H., Stins, M.F., Kim, K.S., 2000. Bacterial penetration across the bloodebrain bar-
rier during the development of neonatal meningitis. Microbes Infect. 2, 1237e1244.
Huang, S.H., Wass, C., Fu, Q., Prasadarao, N.V., Stins, M., Kim, K.S., 1995. Escherichia coli
invasion of brain microvascular endothelial cells in vitro and in vivo: molecular cloning and
characterization of invasion gene ibe10. Infect. Immun. 63, 4470e4475.
Huh, D., Matthews, B.D., Mammoto, A., Montoya-Zavala, M., Hsin, H.Y., Ingber, D.E.,
2010. Reconstituting organ-level lung functions on a chip. Science 328, 1662e1668.
Hurley, B.P., Siccardi, D., Mrsny, R.J., McCormick, B.A., 2004. Polymorphonuclear cell
transmigration induced by Pseudomonas aeruginosa requires the eicosanoid hepoxilin
A3. J. Immunol. 173, 5712e5720.
Hurst, R.D., Fritz, I.B., 1996. Properties of an immortalised vascular endothelial/glioma cell
co-culture model of the bloodebrain barrier. J. Cell. Physiol. 167, 81e88.
Hurst, R.D., Heales, S.J., Dobbie, M.S., Barker, J.E., Clark, J.B., 1998. Decreased endothe-
lial cell glutathione and increased sensitivity to oxidative stress in an in vitro bloodebrain
barrier model system. Brain Res. 802, 232e240.
Igietseme, J.U., Wyrick, P.B., Goyeau, D., Rank, R.G., 1994. An in vitro model for im-
mune control of chlamydial growth in polarized epithelial cells. Infect. Immun. 62,
3528e3535.
Ivascu, A., Kubbies, M., 2006. Rapid generation of single-tumor spheroids for high-
throughput cell function and toxicity analysis. J. Biomol. Screen 11, 922e932.
J€ager, J., Marwitz, S., Tiefenau, J., Rasch, J., Shevchuk, O., Kugler, C., Goldmann, T.,
Steinert, M., 2014. Human lung tissue explants reveal novel interactions during Legionella
pneumophila infections. Infect. Immun. 82, 275e285.
Jensch, I., Bergmann, S., Fulde, M., Hammerschmidt, S., 2010. PavB is a surface-exposed
adhesin of Streptococcus pneumoniae contributing to nasopharyngeal colonization and air-
ways infections. Mol. Microbiol. 77 (1), 22e43.
Jiang, H.-L., Kim, Y.-K., Cho, K.-H., Jang, Y.-C., Choi, Y.-J., Chung, J.-H., Cho, C.-S.,
2010. Roles of spheroid formation of hepatocytes in liver tissue engineering. Int. J. Stem
Cells 3 (2), 69e73.
Jong, A.Y., Stins, M.F., Huang, S.H., Chen, S.H., Kim, K.S., 2001. Traversal of Candida
albicans across human bloodebrain barrier in vitro. Infect. Immun. 69, 4536e4544.
Kale, S., Biermann, S., Edwards, C., Tarnowski, C., Morris, M., Long, M.W., 2000. Three-
dimensional cellular development is essential for ex vivo formation of human bone. Nat.
Biotechnol. 18, 954e958.
Kane, C.D., Byrne, G.I., 1998. Differential effects of gamma interferon on Chlamydia tracho-
matis growth in polarized and nonpolarized human epithelial cells in culture. Infect.
Immun. 66, 2349e2351.
Kane, C.D., Vena, R.M., Ouellette, S.P., Byrne, G.I., 1999. Intracellular tryptophan pool sizes
may account for differences in gamma interferon-mediated inhibition and persistence of
chlamydial growth in polarized and nonpolarized cells. Infect. Immun. 67, 1666e1671.
Kazmierczak, B.I., Mostov, K., Engel, J.N., 2001. Interaction of bacterial pathogens with
polarized epithelium. Annu. Rev. Microbiol. 55, 407e435.
Kelm, J.M., Timmins, N.E., Brown, C.J., Fussenegger, M., Nielsen, L.K., 2003. Method for
generation of homogeneous multicellular tumor spheroids applicable to a wide variety of
cell types. Biotechnol. Bioeng. 83, 173e180.
Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.J., Frey, E.A., Finlay, B.B., 1997.
Enteropathogenic E. coli (EPEC) Transfers Its Receptor for Intimate Adherence into
Mammalian Cells. Cell. 91 (4), 511e520.
Kim, K.S., 2000. E. coli invasion of brain microvascular endothelial cells as a pathogenetic
basis of meningitis. Subcell. Biochem. 33, 47e59.
Kitazawa, T., Hosoya, K., Watanabe, M., Takashima, T., Ohtsuki, S., Takanaga, H.,
Ueda, M., Yanai, N., Obinata, M., Terasaki, T., 2001. Characterization of the amino
acid transport of new immortalized choroid plexus epithelial cell lines: a novel in vitro
system for investigating transport functions at the blood-cerebrospinal fluid barrier.
Pharm. Res. 18, 16e22.
Kleinman, H.K., McGarvey, M.L., Martin, G.R., 1982. Role of endogenous collagen syn-
thesis in the adhesion of human skin fibroblasts. Cell Biol. Int. Rep. 6, 591e599.
Kleinman, H.K., Philp, D., Hoffman, M.P., 2003. Role of the extracellular matrix in
morphogenesis. Curr. Opin. Biotechnol. 14, 526e532.
Kniesel, U., Wolburg, H., 2000. Tight junctions of the bloodebrain barrier. Cell. Mol. Neu-
robiol. 20, 57e76.
Kuchler-Bopp, S., Delaunoy, J.P., Artault, J.C., Zaepfel, M., Dietrich, J.B., 1999. Astrocytes
induce several bloodebrain barrier properties in non-neural endothelial cells. Neurore-
port 10, 1347e1353.
Kuhne, S., Untucht, C., Steinert, M., Watzig, H., 2012. Fast investigations from biological
matrices using CEetest of a bloodebrain barrier model. Electrophoresis 33, 395e401.
Lee, J.H., Gu, Y., Wang, H., Lee, W.Y., 2012. Microfluidic 3D bone tissue model for high-
throughput evaluation of wound-healing and infection-preventing biomaterials. Bioma-
terials 33, 999e1006.
Lee, M.K., Yoo, J.W., Lin, H., Kim, Y.S., Kim, D.D., Choi, Y.M., Park, S.K., Lee, C.H.,
Roh, H.J., 2005. Air-liquid interface culture of serially passaged human nasal epithelial
cell monolayer for in vitro drug transport studies. Drug Deliv. 12, 305e311.
Lee, P.J., Hung, P.J., Lee, L.P., 2007. An artificial liver sinusoid with a microfluidic endothe-
lial-like barrier for primary hepatocyte culture. Biotechnol. Bioeng. 97, 1340e1346.
Li, X.J., Valadez, A.V., Zuo, P., Nie, Z., 2012. Microfluidic 3D cell culture: potential appli-
cation for tissue-based bioassays. Bioanalysis 4, 1509e1525.
Linden, S.K., Driessen, K.M., McGuckin, M.A., 2007. Improved in vitro model systems for
gastrointestinal infection by choice of cell line, pH, microaerobic conditions, and opti-
mization of culture conditions. Helicobacter 12, 341e353.
Linzmeier, R.M., Ganz, T., 2005. Human defensin gene copy number polymorphisms:
comprehensive analysis of independent variation in alpha- and beta-defensin regions at
8p22-p23. Genomics 86, 423e430.
Luttge, M., Fulde, M., Talay, S.R., Nerlich, A., Rohde, M., Preissner, K.T.,
Hammerschmidt, S., Steinert, M., Mitchell, T.J., Chhatwal, G.S., Bergmann, S.,
2012. Streptococcus pneumoniae induces exocytosis of Weibel-Palade bodies in pulmonary
endothelial cells. Cell. Microbiol. 14, 210e225.
Maqsood, M.I., Matin, M.M., Bahrami, A.R., Ghasroldasht, M., 2013. Cell Biol. Int. 37,
1038e1045 (International Federation of Cell Biology).
Marchaim, D., Navon-Venezia, S., Schwartz, D., Tarabeia, J., Fefer, I., Schwaber, M.J.,
Carmeli, Y., 2007. Surveillance cultures and duration of carriage of multidrug-resistant
Acinetobacter baumannii. J. Clin. Microbiol. 45, 1551e1555.
Martinez, I., Nedredal, G.I., Oie, C.I., Warren, A., Johansen, O., Le Couteur, D.G.,
Smedsrod, B., 2008. The influence of oxygen tension on the structure and function
of isolated liver sinusoidal endothelial cells. Comp. Hepatol. 7, 4.
Maunoury, R., Robine, S., Pringault, E., Leonard, N., Gaillard, J.A., Louvard, D., 1992.
Developmental regulation of villin gene expression in the epithelial cell lineages of
mouse digestive and urogenital tracts. Development 115, 717e728.
McCracken, K.W., Cata, E.M., Crawford, C.M., Sinagoga, K.L., Schumacher, M.,
Rockich, B.E., Tsai, Y.H., Mayhew, C.N., Spence, J.R., Zavros, Y., Wells, J.M.,
2014. Modelling human development and disease in pluripotent stem-cell-derived
gastric organoids. Nature 516 (7531), 400e404.
Mehling, M., Tay, S., 2014. Microfluidic cell culture. Curr. Opin. Biotechnol. 25, 95e102.
Meng, Q.L., Liu, F., Yang, X.Y., Liu, X.M., Zhang, X., Zhang, C., Zhang, Z.D., 2014.
Identification of latent tuberculosis infection-related microRNAs in human U937 mac-
rophages expressing Mycobacterium tuberculosis Hsp16.3. BMC Microbiol. 14, 37.
Meshel, A.S., Wei, Q., Adelstein, R.S., Sheetz, M.P., 2005. Basic mechanism of three-
dimensional collagen fibre transport by fibroblasts. Nat. Cell Biol. 7, 157e164.
Mildner, M., Jin, J., Eckhart, L., Kezic, S., Gruber, F., Barresi, C., Stremnitzer, C.,
Buchberger, M., Mlitz, V., Ballaun, C., Sterniczky, B., Fodinger, D., Tschachler, E.,
2010. Knockdown of filaggrin impairs diffusion barrier function and increases UV sensi-
tivity in a human skin model. J. Invest. Dermatol. 130, 2286e2294.
Minchinton, A.I., Tannock, I.F., 2006. Drug penetration in solid tumours. Nat. Rev. Cancer
6, 583e592.
Mizgerd, J.P., Skerrett, S.J., 2008. Animal models of human pneumonia. Am. J. Physiol.
Lung Cell. Mol. Physiol. 294, L387eL398.
Mody, C.H., Paine 3rd, R., Shahrabadi, M.S., Simon, R.H., Pearlman, E., Eisenstein, B.I.,
Toews, G.B., 1993. Legionella pneumophila replicates within rat alveolar epithelial cells.
J. Infect. Dis. 167, 1138e1145.
Mueller-Klieser, W., 1997. Three-dimensional cell cultures: from molecular mechanisms to
clinical applications. Am. J. Physiol. 273, C1109eC1123.
Mul, F.P., Zuurbier, A.E., Janssen, H., Calafat, J., van Wetering, S., Hiemstra, P.S.,
Roos, D., Hordijk, P.L., 2000. Sequential migration of neutrophils across monolayers
of endothelial and epithelial cells. J. Leukoc. Biol. 68, 529e537.
Muruganandam, A., Herx, L.M., Monette, R., Durkin, J.P., Stanimirovic, D.B., 1997.
Development of immortalized human cerebromicrovascular endothelial cell line as an
in vitro model of the human bloodebrain barrier. FASEB J. 11, 1187e1197.
Nakagawa, S., Deli, M.A., Nakao, S., Honda, M., Hayashi, K., Nakaoke, R., Kataoka, Y.,
Niwa, M., 2007. Pericytes from brain microvessels strengthen the barrier integrity in
primary cultures of rat brain endothelial cells. Cell. Mol. Neurobiol. 27, 687e694.
Nerlich, A., Rohde, M., Talay, S.R., Genth, H., Just, I., Chhatwal, G.S., 2009. Invasion of
endothelial cells by tissue-invasive M3 type group A streptococci requires Src kinase and
activation of Rac1 by a phosphatidylinositol 3-kinase-independent mechanism. J. Biol.
Chem. 284 (30), 20319e20328.
Nickerson, C.A., Goodwin, T.J., Terlonge, J., Ott, C.M., Buchanan, K.L., Uicker, W.C.,
Emami, K., LeBlanc, C.L., Ramamurthy, R., Clarke, M.S., Vanderburg, C.R.,
Hammond, T., Pierson, D.L., 2001. Three-dimensional tissue assemblies: novel models
for the study of Salmonella enterica serovar Typhimurium pathogenesis. Infect. Immun. 69,
7106e7120.
Nickerson, C.A., Ott, C.M., Wilson, J.W., Ramamurthy, R., LeBlanc, C.L., Honer zu
Bentrup, K., Hammond, T., Pierson, D.L., 2003. Low-shear modeled microgravity: a
global environmental regulatory signal affecting bacterial gene expression, physiology,
and pathogenesis. J. Microbiol. Methods 54, 1e11.
Nickerson, C.A., Ott, C.M., Wilson, J.W., Ramamurthy, R., Pierson, D.L., 2004. Microbial
responses to microgravity and other low-shear environments. Microbiol. Mol. Biol.
Rev. 68, 345e361.
Nizet, V., Kim, K.S., Stins, M., Jonas, M., Chi, E.Y., Nguyen, D., Rubens, C.E., 1997. In-
vasion of brain microvascular endothelial cells by group B streptococci. Infect. Immun.
65, 5074e5081.
Ochel, A., Rohde, M., Chhatwal, G.S., Talay, S.R., 2014. The M1 protein of Streptococcus
pyogenes triggers an innate uptake mechanism into polarized human endothelial cells.
J. Innate Immun. 6 (5), 585e596.
Pappelbaum, K.I., Gorzelanny, C., Grassle, S., Suckau, J., Laschke, M.W., Bischoff, M.,
Bauer, C., Schorpp-Kistner, M., Weidenmaier, C., Schneppenheim, R., Obser, T.,
Sinha, B., Schneider, S.W., 2013. Ultralarge von Willebrand factor fibers mediate
luminal Staphylococcus aureus adhesion to an intact endothelial cell layer under shear
stress. Circulation 128, 50e59.
Pardridge, W.M., 1999. Bloodebrain barrier biology and methodology. J. Neurovirol. 5,
556e569.
Parkinson, F.E., Friesen, J., Krizanac-Bengez, L., Janigro, D., 2003. Use of a three-dimen-
sional in vitro model of the rat bloodebrain barrier to assay nucleoside efflux from
brain. Brain Res. 980, 233e241.
Parkkinen, J., Korhonen, T.K., Pere, A., Hacker, J., Soinila, S., 1988. Binding sites in the rat
brain for Escherichia coli S fimbriae associated with neonatal meningitis. J. Clin. Invest. 81,
860e865.
Parkos, C.A., Delp, C., Arnaout, M.A., Madara, J.L., 1991. Neutrophil migration across a
cultured intestinal epithelium. Dependence on a CD11b/CD18-mediated event and
enhanced efficiency in physiological direction. J. Clin. Invest 88 (5), 1605e1612.
Pearlman, E., Jiwa, A.H., Engleberg, N.C., Eisenstein, B.I., 1988. Growth of Legionella pneu-
mophila in a human macrophage-like (U937) cell line. Microb. Pathog. 5, 87e95.
Persidsky, Y., 1999. Model systems for studies of leukocyte migration across the bloodebrain
barrier. J. Neurovirol. 5, 579e590.
Persidsky, Y., Stins, M., Way, D., Witte, M.H., Weinand, M., Kim, K.S., Bock, P.,
Gendelman, H.E., Fiala, M., 1997. A model for monocyte migration through the
bloodebrain barrier during HIV-1 encephalitis. J. Immunol. 158, 3499e3510.
Petrova, A., Celli, A., Jacquet, L., Dafou, D., Crumrine, D., Hupe, M., Arno, M., Hobbs, C.,
Cvoro, A., Karagiannis, P., Devito, L., Sun, R., Adame, L.C., Vaughan, R.,
McGrath, J.A., Mauro, T.M., Ilic, D., 2014. 3D in vitro model of a functional epidermal
permeability barrier from human embryonic stem cells and induced pluripotent stem
cells. Stem Cell Rep. 2, 675e689.
Pracht, D., Elm, C., Gerber, J., Bergmann, S., Rohde, M., Seiler, M., Kim, K.-S.,
Jenkinson, H.F., Nau, R., Hammerschmidt, S., 2005. PavA of Streptococcus pneumoniae
modulates adherence, invasion and meningeal inflammation. Inf. Immun. 73 (5),
2680e2689.
Prats, N., Briones, V., Blanco, M.M., Altimira, J., Ramos, J.A., Dominguez, L., Marco, A.,
1992. Choroiditis and meningitis in experimental murine infection with Listeria
monocytogenes. Eur. J. Clin. Microbiol. Infect. Dis. 11, 744e747.
Pron, B., Taha, M.K., Rambaud, C., Fournet, J.C., Pattey, N., Monnet, J.P., Musilek, M.,
Beretti, J.L., Nassif, X., 1997. Interaction of Neisseria meningitidis with the components of
the bloodebrain barrier correlates with an increased expression of PilC. J. Infect. Dis.
176, 1285e1292.
Reichel, A., Begley, D.J., Abbott, N.J., 2003. An overview of in vitro techniques for bloode
brain barrier studies. Methods Mol. Med. 89, 307e324.
Ring, A., Weiser, J.N., Tuomanen, E.I., 1998. Pneumococcal trafficking across the bloode
brain barrier. Molecular analysis of a novel bidirectional pathway. J. Clin. Invest. 102,
347e360.
Robin, J.D., Wright, W.E., Zou, Y., Cossette, S.C., Lawlor, M.W., Gussoni, E., 2015. Isola-
tion and immortalization of patient-derived cell lines from muscle biopsy for disease
modeling. J. Vis. Exp. 95.
Rohde, M., Chhatwal, G.S., 2013. Adherence and invasion of streptococci to
eukaryotic cells and their role in disease pathogenesis. Curr. Top. Microbiol. Immunol.
368, 83e110.
Rosendahl, A., Bergmann, S., Hammerschmidt, S., Golmann, O., Medina, E., 2013. Lung
dendritic cells facilitate extrapulmonary bacterial dissemination during pneumococcal
pneumonia. Front. Cell. Infect. Microbiol. 3, 21.
Rowbotham, T.J., 1980. Preliminary report on the pathogenicity of Legionella pneumophila
for freshwater and soil amoebae. J. Clin. Pathol. 33, 1179e1183.
Rubin, L.L., Staddon, J.M., 1999. The cell biology of the bloodebrain barrier. Annu. Rev.
Neurosci. 22, 11e28.
Rupp, J., Droemann, D., Goldmann, T., Zabel, P., Solbach, W., Vollmer, E.,
Branscheid, D., Dalhoff, K., Maass, M., 2004. Alveolar epithelial cells type II are major
target cells for C. pneumoniae in chronic but not in acute respiratory infection. FEMS
Immunol. Med. Microbiol. 41, 197e203.
Sanford, S.E., 1987. Gross and histopathological findings in unusual lesions caused by Strep-
tococcus suis in pigs. II. Central nervous system lesions. Can. J. Vet. Res. 51, 486e489.
Sansonetti, P., 2001. Phagocytosis of bacterial pathogens: implications in the host response.
Semin. Immunol. 13, 381e390.
Schimek, K., Busek, M., Brincker, S., Groth, B., Hoffmann, S., Lauster, R., Lindner, G.,
Lorenz, A., Menzel, U., Sonntag, F., Walles, H., Marx, U., Horland, R., 2013. Inte-
grating biological vasculature into a multi-organ-chip microsystem. Lab. Chip 13,
3588e3598.
Schmedt, T., Chen, Y., Nguyen, T.T., Li, S., Bonanno, J.A., Jurkunas, U.V., 2012. Telo-
merase immortalization of human corneal endothelial cells yields functional hexagonal
monolayers. PLoS One 7 (12), e51427.
Shaw, J.A., 1996. In: Shaw, A.J. (Ed.), Epithelial Cell Culture e A Practical Approach. IRL
Press at Oxford University Press, p. 218.
Shepherd, J., Douglas, I., Rimmer, S., Swanson, L., MacNeil, S., 2009. Development of
three-dimensional tissue-engineered models of bacterial infected human skin wounds.
Tissue Eng. Part C Methods 15, 475e484.
Shevchuk, O., Steinert, M., 2009. Screening of virulence traits in Legionella pneumophila and
analysis of the host susceptibility to infection by using the Dictyostelium host model
system. Methods Mol. Biol. 470, 47e56.
Shevchuk, O., Abidi, N., Klawonn, F., Wissing, J., Nimtz, M., Kugler, C., Steinert, M.,
Goldmann, T., J€ansch, L., 2014. HOPE-fixation of lung tissue allows retrospective pro-
teome and phosphoproteome studies. J. Proteome Res. 13 (11), 5230e5239.
Shi, L.Z., Zheng, W., 2005. Establishment of an in vitro brain barrier epithelial transport sys-
tem for pharmacological and toxicological study. Brain Res. 1057, 37e48.
Simpson, C.L., Kojima, S., Getsios, S., 2010. RNA interference in keratinocytes and an orga-
notypic model of human epidermis. Methods Mol. Biol. 585, 127e146.
Skriwan, C., Fajardo, M., Hagele, S., Horn, M., Wagner, M., Michel, R., Krohne, G.,
Schleicher, M., Hacker, J., Steinert, M., 2002. Various bacterial pathogens and symbionts
infect the amoeba Dictyostelium discoideum. Int. J. Med. Microbiol. 291, 615e624.
Smith, A.L., 1987. Pathogenesis of Haemophilus influenzae meningitis. Pediatr. Infect. Dis. J. 6,
783e786.
Stamatovic, S., Keep, R.F., Andjelkovic, A.V., 2008. Barin endothelial cell-cell junctions:
how to “open” the bloodebrain barrier. Curr. Neuropharmacol. 6 (3), 179e192.
Stanness, K.A., Westrum, L.E., Fornaciari, E., Mascagni, P., Nelson, J.A., Stenglein, S.G.,
Myers, T., Janigro, D., 1997. Morphological and functional characterization of an in vitro
bloodebrain barrier model. Brain Res. 771, 329e342.
Stein, A.M., Demuth, T., Mobley, D., Berens, M., Sander, L.M., 2007. A mathematical
model of glioblastoma tumor spheroid invasion in a three-dimensional in vitro
experiment. Biophys. J. 92 (1), 356e365.
Steinert, M., Ott, M., L€ uck, P.C., Tannich, E., Hacker, J., 1994. Studies on the uptake and
intracellular replication of Legionella pneumophila in protozoa and in macrophage e like
cells. FEMS Microbiol. Ecol. 15 (3e4), 299e308.
Steinert, M., Hentschel, U., Hacker, J., 2000. Symbiosis and pathogenesis: evolution of the
microbe-host interaction. Naturwissenschaften 87, 1e11.
Steinert, M., Leippe, M., Roeder, T., 2003. Surrogate hosts: protozoa and invertebrates as
models for studying pathogen-host interactions. Int. J. Med. Microbiol. 293, 321e332.
Steinmann, U., Borkowski, J., Wolburg, H., Schroppel, B., Findeisen, P., Weiss, C.,
Ishikawa, H., Schwerk, C., Schroten, H., Tenenbaum, T., 2013. Transmigration of pol-
ymorphnuclear neutrophils and monocytes through the human blood-cerebrospinal
fluid barrier after bacterial infection in vitro. J. Neuroinflammation 10, 31.
Stins, M.F., Badger, J., Sik Kim, K., 2001. Bacterial invasion and transcytosis in transfected
human brain microvascular endothelial cells. Microb. Pathog. 30, 19e28.
Stins, M.F., Prasadarao, N.V., Zhou, J., Arditi, M., Kim, K.S., 1997. Bovine brain microvascular
endothelial cells transfected with SV40-large T antigen: development of an immortalized cell
line to study pathophysiology of CNS disease. Vitro Cell. Dev. Biol. Anim. 33, 243e247.
Stock, U.A., Vacanti, J.P., 2001. Tissue engineering: current state and prospects. Annu. Rev.
Med. 52, 443e451.
Sun, D., Coleclough, C., Cao, L., Hu, X., Sun, S., Whitaker, J.N., 1998. Reciprocal stim-
ulation between TNF-alpha and nitric oxide may exacerbate CNS inflammation in
experimental autoimmune encephalomyelitis. J. Neuroimmunol. 89, 122e130.
Sutherland, R.M., 1988. Cell and environment interactions in tumor microregions: the
multicell spheroid model. Science 240, 177e184.
Sutherland, R.M., McCredie, J.A., Inch, W.R., 1971. Growth of multicell spheroids in tissue
culture as a model of nodular carcinomas. J. Natl. Cancer Inst. 46, 113e120.
Szymanski, K.V., Toennies, M., Becher, A., Fatykhova, D., N’Guessan, P.D., Gutbier, B.,
Klauschen, F., Neuschaefer-Rube, F., Schneider, P., Rueckert, J., Neudecker, J.,
Bauer, T.T., Dalhoff, K., Dromann, D., Gruber, A.D., Kershaw, O., Temmesfeld-
Wollbrueck, B., Suttorp, N., Hippenstiel, S., Hocke, A.C., 2012. Streptococcus pneumo-
niae-induced regulation of cyclooxygenase-2 in human lung tissue. Eur. Respir. J. 40,
1458e1467.
Tabor-Godwin, J.M., Ruller, C.M., Bagalso, N., An, N., Pagarigan, R.R., Harkins, S.,
Gilbert, P.E., Kiosses, W.B., Gude, N.A., Cornell, C.T., Doran, K.S., Sussman, M.A.,
Whitton, J.L., Feuer, R., 2010. A novel population of myeloid cells responding to cox-
sackievirus infection assists in the dissemination of virus within the neonatal CNS.
J. Neurosci. 30, 8676e8691.
Tacken, P.J., Batenburg, J.J., 2006. Monocyte CD64 or CD89 targeting by surfactant protein
D/anti-Fc receptor mediates bacterial uptake. Immunology 117, 494e501.
Taylor-Robinson, D., 1976. The use of organ cultures and animal models in the study of
Mycoplasma pneumoniae infections. Infection 4, 4e8.
Tenenbaum, T., Steinmann, U., Friedrich, C., Berger, J., Schwerk, C., Schroten, H., 2013.
Culture models to study leukocyte trafficking across the choroid plexus. Fluids Barriers
CNS 10, 1.
Thakoersing, V.S., Gooris, G.S., Mulder, A., Rietveld, M., El Ghalbzouri, A.,
Bouwstra, J.A., 2012. Unraveling barrier properties of three different in-house human
skin equivalents. Tissue Eng. Part C Methods 18, 1e11.
Timmins, N.E., Harding, F.J., Smart, C., Brown, M.A., Nielsen, L.K., 2005. Method for the
generation and cultivation of functional three-dimensional mammary constructs without
exogenous extracellular matrix. Cell. Tissue Res. 320, 207e210.
Todaro, G.J., Zeve, V., Aaronson, S.A., 1971. Cell culture techniques in the search for cancer
viruses of man. Vitro 6, 355e361.
Tompkins, W.A., Watrach, A.M., Schmale, J.D., Schultz, R.M., Harris, J.A., 1974. Cultural
and antigenic properties of newly established cell strains derived from adenocarcinomas
of the human colon and rectum. J. Natl. Cancer Inst. 52, 1101e1110.
Townsend, G.C., Scheld, W.M., 1995. In vitro models of the bloodebrain barrier to study
bacterial meningitis. Trends Microbiol. 3, 441e445.
Trietsch, S.J., Israels, G.D., Joore, J., Hankemeier, T., Vulto, P., 2013. Microfluidic titer plate
for stratified 3D cell culture. Lab. Chip 13, 3548e3554.
Unsworth, B.R., Lelkes, P.I., 1998. Growing tissues in microgravity. Nat. Med. 4, 901e907.
Untucht, C., Rasch, J., Fuchs, E., Rohde, M., Bergmann, S., Steinert, M., 2011. An opti-
mized in vitro bloodebrain barrier model reveals bidirectional transmigration of African
trypanosome strains. Microbiology 157, 2933e2941.
Vanier, G., Segura, M., Gottschalk, M., 2007. Characterization of the invasion of porcine
endothelial cells by Streptococcus suis serotype 2. Can. J. Vet. Res. 71 (2), 81e89.
Wadell, C., Bjork, E., Camber, O., 1999. Nasal drug deliveryeevaluation of an in vitro model
using porcine nasal mucosa. Eur. J. Pharm. Sci. 7, 197e206.
Walpita, D., Hay, E., 2002. Studying actin-dependent processes in tissue culture. Nat. Rev.
Mol. Cell Biol. 3, 137e141.
Wang, Y.-H., Wu, J.-J., Lei, H.-Y., 2009. The autophagic induction in Helicobacter pylori-
infected macrophage. Exp. Biol. Med. 234 (2), 171e180.
Weber, S., Stirnmann, C.U., Wieser, M., Frey, D., Meier, R., Engelhadt, S., Li, X.,
Capitani, G., Kammerer, R.A., Hilbi, H., 2014. A type IV translocated Legionella cysteine
phytase counteracts intracellular growth restriction by phytate. J. Biol. Chem. 289,
34175e34188.
Weksler, B.B., Subileau, E.A., Perriere, N., Charneau, P., Holloway, K., Leveque, M.,
Tricoire-Leignel, H., Nicotra, A., Bourdoulous, S., Turowski, P., Male, D.K.,
Roux, F., Greenwood, J., Romero, I.A., Couraud, P.O., 2005. Bloodebrain
barrier-specific properties of a human adult brain endothelial cell line. FASEB J. 19,
1872e1874.
Wengst, A., Reichl, S., 2010. RPMI 2650 epithelial model and three-dimensional recon-
structed human nasal mucosa as in vitro models for nasal permeation studies. Eur. J.
Pharm. Biopharm. 74, 290e297.
Wewer, C., Seibt, A., Wolburg, H., Greune, L., Schmidt, M.A., Berger, J., Galla, H.J.,
Quitsch, U., Schwerk, C., Schroten, H., Tenenbaum, T., 2011. Transcellular migration
of neutrophil granulocytes through the blood-cerebrospinal fluid barrier after infection
with Streptococcus suis. J. Neuroinflammation 8, 51.
Wiegand, C., Abel, M., Ruth, P., Hipler, U.C., 2009. HaCaT keratinocytes in co-culture
with Staphylococcus aureus can be protected from bacterial damage by polihexanide.
Wound Repair Regen. 17, 730e738.
Wilson, J.W., Ott, C.M., Ramamurthy, R., Porwollik, S., McClelland, M., Pierson, D.L.,
Nickerson, C.A., 2002a. Low-Shear modeled microgravity alters the Salmonella enterica
serovar typhimurium stress response in an RpoS-independent manner. Appl. Environ.
Microbiol. 68, 5408e5416.
Wilson, J.W., Ramamurthy, R., Porwollik, S., McClelland, M., Hammond, T., Allen, P.,
Ott, C.M., Pierson, D.L., Nickerson, C.A., 2002b. Microarray analysis identifies Salmo-
nella genes belonging to the low-shear modeled microgravity regulon. Proc. Natl. Acad.
Sci. U.S.A. 99, 13807e13812.
Wittchen, E.S., 2009. Endothelial signaling in paracellular and transcellular leukocyte
transmigration. Front. Biosci. (Landmark Ed.) 14, 2522e2545.
Woodworth, B.A., Tamashiro, E., Bhargave, G., Cohen, N.A., Palmer, J.N., 2008. An in
vitro model of Pseudomonas aeruginosa biofilms on viable airway epithelial cell
monolayers. Am. J. Rhinol. 22, 235e238.
Wright, J.R., Zlogar, D.F., Taylor, J.C., Zlogar, T.M., Restrepo, C.I., 1999. Effects of endo-
toxin on surfactant protein A and D stimulation of NO production by alveolar
macrophages. Am. J. Physiol. 276, L650eL658.
Wyrick, P.B., Choong, J., Davis, C.H., Knight, S.T., Royal, M.O., Maslow, A.S.,
Bagnell, C.R., 1989. Entry of genital Chlamydia trachomatis into polarized human epithe-
lial cells. Infect. Immun. 57, 2378e2389.
Xu, F., Droemann, D., Rupp, J., Shen, H., Wu, X., Goldmann, T., Hippenstiel, S.,
Zabel, P., Dalhoff, K., 2008. Modulation of the inflammatory response to Streptococcus
pneumoniae in a model of acute lung tissue infection. Am. J. Respir. Cell Mol. Biol.
39, 522e529.
Yamada, K.M., Cukierman, E., 2007. Modeling tissue morphogenesis and cancer in 3D. Cell
130, 601e610.
Yang, Q., Phillips, P.L., Sampson, E.M., Progulske-Fox, A., Jin, S., Antonelli, P.,
Schultz, G.S., 2013. Development of a novel ex vivo porcine skin explant model for
the assessment of mature bacterial biofilms. Wound Repair Regen. 21, 704e714.
Yeaman, C., Grindstaff, K.K., Hansen, M.D., Nelson, W.J., 1999. Cell polarity: Versatile
scaffolds keep things in place. Curr. Biol. 9, R515eR517.
Yeh, T.H., Tsai, C.H., Chen, Y.S., Hsu, W.C., Cheng, C.H., Hsu, C.J., Lee, S.Y., 2007.
Increased communication among nasal epithelial cells in air-liquid interface culture.
Laryngoscope 117, 1439e1444.
Zaman, M.H., Trapani, L.M., Sieminski, A.L., Mackellar, D., Gong, H., Kamm, R.D.,
Wells, A., Lauffenburger, D.A., Matsudaira, P., 2006. Migration of tumor cells in 3D
matrices is governed by matrix stiffness along with cell-matrix adhesion and
proteolysis. Proc. Natl. Acad. Sci. U.S.A. 103, 10889e10894.
Zeana, C., Larson, E., Sahni, J., Bayuga, S.J., Wu, F., Della-Latta, P., 2003. The epidemi-
ology of multidrug-resistant Acinetobacter baumannii: does the community represent a
reservoir? Infect. Control Hosp. Epidemiol. 24, 275e279.
Zeillemaker, A.M., Mul, F.P., Hoynck van Papendrecht, A.A., Leguit, P., Verbrugh, H.A.,
Roos, D., 1996. Neutrophil adherence to and migration across monolayers of human
peritoneal mesothelial cells. The role of mesothelium in the influx of neutrophils during
peritonitis. J. Lab. Clin. Med. 127, 279e286.
Zemans, R.L., Briones, N., Young, S.K., Malcolm, K.C., Refaeli, Y., Downey, G.P.,
Worthen, G.S., 2009. A novel method for long term bone marrow culture and genetic
modification of murine neutrophils via retroviral transduction. J. Immunol. Methods
340, 102e115.
Zhang, E.Y., Phelps, M.A., Banerjee, A., Khantwal, C.M., Chang, C., Helsper, F.,
Swaan, P.W., 2004. Topology scanning and putative three-dimensional structure of
the extracellular binding domains of the apical sodium-dependent bile acid transporter
(SLC10A2). Biochemistry 43, 11380e11392.
Zhang, X., Wang, W., Yu, W., Xie, Y., Zhang, Y., Ma, X., 2005. Development of an in
vitro multicellular tumor spheroid model using microencapsulation and its application
in anticancer drug screening and testing. Biotechnol. Prog. 21, 1289e1296.
Zheng, W., Zhao, Q., 2002. Establishment and characterization of an immortalized Z310
choroidal epithelial cell line from murine choroid plexus. Brain Res. 958, 371e380.
CHAPTER TWO
Science and Art of Cell-Based

Ocular Surface Regeneration
Vivek Singh1, Sachin Shukla1, Charanya Ramachandran1,
Dilip Kumar Mishra2, Kishore R. Katikireddy3, Ikeda Lal4,
Sunil K. Chauhan3 and Virender S. Sangwan5, *
1
Sudhakar and Sreekanth Ravi Stem Cell Biology Laboratory, Prof. Brien Holden Eye Research Centre,
C-TRACER, LV Prasad Eye Institute, Hyderabad, Telangana, India
2
Department of Ocular Pathology, LV Prasad Eye Institute, Hyderabad, Telangana, India
3
4
LV Prasad Eye Institute, Hyderabad, Telangana, India
5
Center for Ocular Regeneration, Dr Paul Dubord Chair in Cornea, LV Prasad Eye Institute, Hyderabad,
Telangana, India
*Corresponding author: E-mail: vsangwan@lvpei.org
Contents
1. Introduction 46
2. Anatomy and Pathology of Ocular Surface 48
2.1 Preocular Tear Film 49
2.2 Conjunctival Epithelium 49
2.3 Limbus 50
3. Cell-Based Therapies for Ocular Surface Regeneration 53
3.1 Embryonic Stem Cells 54
3.2 Induced Pluripotent Stem Cells 54
3.3 Mesenchymal Stem Cells 55
3.3.1 MSCs in corneal reconstruction 56
3.3.2 MSCs in LSCD 56
4. New Material Technologies for Ocular Surface Reconstruction 58
4.1 Biological Materials 60
4.1.1 Fibrin 60
4.1.2 Collagen-based materials 61
4.1.3 Silk fibroin-based materials 62
4.2 Synthetic Materials 63
4.2.1 Contact lenses 63
4.2.2 Polylactic glycolic acid 64
4.2.3 Thermoresponsive substrate 66
4.3 Engineering Limbal Niche 67
5. Animal Models for LSCD 73
5.1 LSCD due to Genetic Defects of Limbal/Corneal Epithelial Cells 73

j
ISSN 1937-6448
46 Vivek Singh et al.
5.2 LSCD due to Chemical and Mechanical Injury of Limbal/Corneal Epithelial Cells 74
5.3 Animal Models for Corneal Wound Healing 77
6. Clinical Outcome of Cell-Based Ocular Surface Reconstructive Procedure 80
6.1 Overview 80
6.2 Surgical Techniques 81
6.2.1 Cultivated limbal epithelial transplantation 81
6.2.2 Simple limbal epithelial transplantation 84
6.2.3 Cultivated oral mucosal epithelial transplantation 89
7. Future Path and Conclusion 91
Acknowledgments 92
References 92
Abstract
The potential cause of blindness worldwide includes diseases of the cornea, ocular sur-
face (limbal stem cell deficiency, allergic conjunctivitis, dry eye diseases), and retinal dis-
eases. The presence of stem cells (limbal stem cells) in the basal region of the limbus
makes it an important tool for the ocular regeneration and also in maintaining the
transparency of eye by replacing the corneal epithelium continuously. Various surgical
modalities have been developed like cultured limbal epithelial transplantation, cultured
oral mucosal epithelial transplantation, simple limbal epithelial transplantation, etc., uti-
lizing the cell-based regenerative properties to treat limbal disorder. Cell-based thera-
pies for ocular repair and regeneration comprise a major hope by therapies involving
the mesenchymal stem cells, embryonic stem cells, and limbal stem cells for the resto-
ration of vision in individuals whose ocular tissue has been irreversibly damaged by dis-
ease or trauma. This review explores critical needs in human disease mainly the ocular
problem where cell-based therapeutics is exceptionally well suited and also the use of
animal models, various artificial scaffolds, as well as advancement in clinical technique
to challenge the current demand to overcome corneal blindness.
1. INTRODUCTION
Advances in basic and clinical research during the last few years on
fetal, amniotic, embryonic, umbilical cord blood, and adult stem cells
have brought countless new dimensions to cell-based/regenerative medicine
by providing with various tools of generating and sustaining various cells.
Cell therapy originated in the nineteenth century when scientists experi-
mented by injecting animal material in an effort to prevent and treat illness
(Starzl, 2000). A well-established and widely used cell-based therapy is the
transplantation of blood stem cells to treat diseases and conditions of the
blood and immune system, or for treatments of specific cancers. The era
of cell-based therapy for ocular surface disorders began with the discovery
Limbal Transplantation: Science and Art 47
of limbal stem cells (LSCs) in the palisades of Vogt. Kenyon and Tseng
(1989) transplanted the healthy limbal tissue from the healthy eye to the
affected eye in cases with severe ocular surface disease. Developed in
1981, the area of stem cell biology has now grown up to adulthood where
the research and clinical trials done so far are mature enough for laying the
foundation of an innovative, promising, and bright path ahead. The stem
cells have the capacity to self-renew as well as the ability to generate differ-
entiated cells (Weissman et al., 2001; Smith, 2001). Based on their origin,
they are broadly classified as embryonic stem cell (ESC) and adult stem
cell. In animal species, in vivo differentiation can be assessed rigorously by
the ability of ESCs to contribute to all somatic lineages and produce germ
line chimerism. Having comparatively lesser constraints on the ethical front,
the adult stem cells have become the choice of priority in the field of regen-
erative medicine. In between, Yamanaka’s group gave new hope to patients
with incurable diseases by developing induced murine pluripotent stem cells
(iPSCs) in 2006 (Takahashi and Yamanaka, 2006) and human iPSCs in 2007
(Takahashi et al., 2007). In the past few years, stem cell-based regenerative
therapies have emerged as a boon for patients suffering from cardiovascular
disorders (Hou et al., 2013), hematological malignancies (Ramdass et al.,
2013), dental problems (Xiao and Nasu, 2014), orthopedics (Riminucci
et al., 2015), and ocular disorders (Kinoshita, 2010; Lal et al., 2013; Sangwan
et al., 2014). Nevertheless, drugs and biologics will always have significant
remedial niches; but there are applications for which cells (cell-based
approach) are better equipped. This review also explores critical needs in hu-
man disease mainly the ocular problem where cell-based therapeutics is
exceptionally well suited.
Ophthalmology, among various branches of medical sciences, was prob-
ably the first to benefit directly from stem cells of regenerative treatment.
Accessibility, ease of follow-up, and the immune-privileged status of eyes
appear to be the key factors behind success stories. Cell-based therapies uti-
lizing cells derived from the ciliary body, iris, and sclera are still waiting for
success in animal trials, but show potential for replacing damaged photore-
ceptors (Dhamodaran et al., 2014). Limbal, corneal, and conjunctival stem
cells have been successfully utilized for ocular surface reconstruction; how-
ever, their potential beyond this is yet to be explored.
This review focuses on the recent advances in LSC culture, various clin-
ical modalities, its application as well as the current understanding at basic
research, and the animal models. We describe how the better anatomical
and cellular knowledge has improved our understanding on critical and
unique functions of these LSCs with an unlimited self-renewal capacity. We

also emphasize on the therapeutic potential of different mesenchymal stem
cells (MSCs) and their in vitro modulation, role of scaffold, and advance-
ment in animal models for treating and curing specific ocular disorders
primarily focusing on LSC disorders. We also summarize some major ad-
vances to translate the experimental models on ex vivo and in vivo expanded
and/or differentiated limbal epithelial stem cells (LESCs) into clinical appli-
cations like cultivated limbal epithelial transplantation (CLET) and simple
limbal epithelial transplantation (SLET) for the advancement of this novel
cellular therapies intended at repairing damaged ocular surface in humans.
2. ANATOMY AND PATHOLOGY OF OCULAR SURFACE

The human eye begins to develop as a pair of outpouching that be-
comes the optic vesicles on each side of the forebrain at the end of the fourth
week of development. Cornea makes anterior 1/6th of the ocular surface
and is the principal refracting medium of eye. The anterior corneal surface
is bathed by the tear film, whereas the posterior surface is in direct contact
with aqueous humor. In the adult, the anterior refractive power of cornea
is 43e43.50 diopters. The vertical diameter of cornea is 1 mm less than
the horizontal diameter which is 11.5e12 mm. It is approximately
0.5 mm thick at the center and 0.7 mm thick at the periphery.
The corneal epithelium is approximately 50 mm thick, which is about
10% of the total thickness of the cornea. The corneal epithelium is stratified
squamous with a smooth surface and it usually consists of five layers of large
squamous cells that have few organelles but often contain glycogen. The su-
perficial cells retain their nuclei, and their external surface forms numerous
fine ridges (microplicae) that help retain moisture on the corneal surface.
Below the superficial cells are wing cells, and lower most layers are basal
cells. The basal cell layer is columnar in shape, and it adheres to the basement
membrane adjacent to Bowman’s layer. Only the basal cells of the corneal
epithelium proliferate. The daughter cells differentiate into wing cells and
subsequently into superficial cells, and gradually migrate to the corneal sur-
face. The differentiation process requires about 7e14 days, after which the
superficial cells are desquamated into the tear film.
The ocular surface in anatomical sense is formed of the entire epithelial
surface of the cornea, limbus, and its major supporting tissue, the conjunc-
tiva. In a wider anatomical, embryological, and also in functional sense, the
eyelid, lacrimal gland, and the lacrimal drainage system also belong to the
ocular surface. The function of ocular surface is protective and refractive.
The major components of ocular surface are discussed below.
2.1 Preocular Tear Film

The function of tears is to keep our eyes moisturized and nourished. It pro-
tects eyes by washing out foreign objects. Tear film is composed of three
different layers: (1) An outermost, oily layer that acts as a sealant to keep tears
from evaporating. The outermost layer of the tear film is secreted by meibo-
mian gland and a small proportion from the glands of Zeis. The oily or lipid
layer is responsible for providing stability to the tear film through interaction
with the aqueousemucin phase, providing a smooth optical surface to the
cornea and acting as a barrier against foreign particles. Meibomian oil secre-
tion is a continuous process, occurring 24 h per day during working and
sleeping hours, and it is aided by blink action. (2) The middle aqueous layer
of the tear film consists of water, electrolytes, proteins, peptide growth
factors, cytokines, immunoglobulins, and antimicrobials, secreted by the
lacrimal glands and accessory tear glands of Krause and Wolfring. The anti-
microbial proteins found in the tear are lysozyme, lactoferrin, and lipocalin.
(3) The innermost mucous layer is largely separated by the conjunctival
goblet cells; however, ocular mucins are also produced by the stratified squa-
mous epithelium of cornea and conjunctiva. Ocular mucus is composed of
mucin, immunoglobulin, salts, glucose, leukocytes, cellular debris, and en-
zymes. Mucins are membrane-associated or secretory. Membrane-associated
mucins form the glycocalyx. Glycocalyx forms the dense barrier at the
epithelial celletear film interface for the pathogen entrance. Secreted mu-
cins move through the tear fluid and collect debris that can be removed
via the nasolacrimal duct during blinking.
2.2 Conjunctival Epithelium

Conjunctiva is the major supporting tissue of the ocular surface. Conjunctiva
is a continuous membrane, and for practical and clinical purposes it is sub-
divided into palpebral, forniceal, and bulbar zones. The palpebral conjunc-
tiva extends from the mucocutaneous junction at the lid margin to the upper
and lower margins of the tarsal plate. Its surface is smooth and contains
several cryptlike infoldings of the epithelium called “Henle’s crypt.” The
forniceal conjunctiva is loosely attached to the orbital septum; it extends
temporally behind the lateral canthus and nasally to the semilunar ducts of
the lacrimal gland that open into the temporal portion of the upper fornix;
into the upper and lower fornices accessory glands of Krause and Wolfring
open. The histomorphology of the conjunctiva differs in all the three zones.
The conjunctiva is covered by two or more layers of stratified squamous and
columnar epithelium at the limbus and the palpebral margins conjunctiva
exhibits a stratified squamous pattern. At the mucocutaneous junction of
the lid margin abrupt transition is found from nonkeratinized stratified squa-
mous mucosal epithelium of palpebral conjunctiva to keratinized epithelium
of skin. The conjunctival basal cells have a thin basement membrane similar
to the basal cells of the corneal epithelium. The midepithelial and superficial
cells appear polygonal and tend to flatten as they approach the surface.
Mucus-secreting goblet cells normally are present in the middle and super-
ficial layers of the epithelium, and they account up to 10% of the basal
epithelial cells of conjunctiva. Pellegrini et al. (1999) reported uniform dis-
tribution of stem cells in bulbar and forniceal conjunctiva.
2.3 Limbus
The limbus forms the border between the transparent cornea and opaque
sclera, contains the pathways of aqueous humor outflow, and is the site of
surgical incisions for cataract and glaucoma (Figure 1). The limbal niche is
a combination of both biochemical (growth factors, cytokines, etc.) and bio-
physical (matrix stiffness, topography, vasculature, etc.) factors that together
modulate cell fate. Anatomically it is already known that the LSCs are
located within the palisades of Vogt, a pigmented region (due to the pres-
ence of melanocytes) by a dense vascular network (Shanmuganathan
et al., 2007; Shortt, 2007 ). The vasculature allows the infiltration of the
Langerhans’ cells and T-lymphocytes while the melanocytes are thought
to protect the resident cells from UV-induced damage.
The epithelial cell border between conjunctiva and cornea possesses
multipotential cells important for differentiation of the respective cell types;
the internal limbal border zone between corneal endothelium and anterior
trabeculum appears to contain specialized cells some of which are activated
to migrate and repopulate the trabecular meshwork after trabecular injury.
The vasculature of the limbus derives in primates primarily from the anterior
ciliary arteries. Their superficial branches form arcades to supply the limbal
conjunctiva and peripheral cornea. Perforating branches contribute to the
vascular supplies of the deep limbal structures and the anterior uvea (Van
Buskirk, 1989) (Figure 1).
As per histological features, the nonkeratinized stratified limbal epithe-
lium can be differentiated from the conjunctival epithelium, as it lacks goblet
(A) (B)
(C) (D)
Figure 1 Morphological and histological architecture of the human limbus. Showing
anatomy and histopathology of normal healthy eye to show the location of limbal
stem cells. (A) Normal eye with healthy cornea, limbal stem cells; (B) Flat mount of
eyeball; (C) Limbal transition zone with undulated limbal epithelium between corneal
and conjuctival layer; (D) Healthy limbus.
cells. Compared to the corneal epithelium, while the superficial epithelial

layers are rather similar, the limbal epithelium contains cell layers, a large
number of mature (activated) and immature epithelial dendritic cells, T lym-
phocytes, highly pigmented melanocytes, and subjacent blood vessels.
Moreover, the basal limbal epithelial cells are unique in that they are the least
differentiated cells of the ocular surface epithelium. These cells are smaller,
less columnar, and have more cytoplasmic organelles (Figure 1). A growing
body of evidence over the past years supports the theory that these cells are
LESCs, giving rise to the more differentiated corneal epithelium (Cotsarelis
et al., 1989). LSCs possess the critical ability to regenerate corneal epithe-
lium, thus offering great therapeutic potential (Davanger and Evensen,
1971; Ramachandran et al., 2014). Multiple diseases that damage LSCs
and their microenvironment can lead to a pathological condition known
as limbal stem cell deficiency (LSCD), reflected in corneal conjunctivaliza-
tion, neovascularization, epithelial defect, and chronic inflammation of
the ocular surface. We can observe the presence of active ocular surface
inflammation such as in cases of acute ocular surface burns and acute

StevenseJohnson syndrome (SJS) as the survival of stem cells is questionable
in presence of ocular inflammation (Figures 2 and 3).
Cells from the corneal surface are desquamated and replaced by prolifer-
ating basal epithelial cells migrated from the periphery. Thoft and Friend
proposed an “XYZ hypothesis” of corneal epithelial maintenance in which
the desquamated cells (Z component) are continuously replaced not only by
the basal cells (X) but also by cells that migrate in from the periphery (Y).
Thus, migration occurs centripetally and circumferentially from the limbus
and vertically from the basal layer forward. LSCs maintain the integrity of
Figure 2 Limbal stem cell deficiency due to thermal and chemical injury. (A) Pterygium:
A degenerative condition, type of solar keratosis in which conjunctival epithelium grows
near/over the cornea. Section (10 magnification) shows folded stratified sqamous
epithelium with few dilated blood vessels, infiltration of inflammatory cells, fibrosis,
and focal area of elastotic degeneration. (B) Acitinic keratosis associated with dysplasia:
A type of solar keratosis in which cells of stratified squamous epithelium shows loss of
differentiation of individual cells (dysplastic cells) in entire thickness of epithelium. Sec-
tion (10 magnification) shows stratified squamous epithelium with full thickness
dysplasia. Stroma shows elastotic degeneration, fibrosis, and few dilated blood vessels.
(C) Pannus: A fibrovascular tissue formed in response to corneal insult. Section (10
magnification) shows folded stratified squamous epithelium with few goblet cells.
Stroma shows dilated blood vessels, fibrosis, and few mononuclear round cells.
Figure 3 Limbal stem cell deficiency due to inflammation. (A) StevenseJohnson syn-
drome: Hypersensitive reaction due to drugs, especially, sulpha drugs and idiopathic
etiology leading to epidermal blistering, necrosis, sloughing and involves less than
10% of body surface area. Section (4 magnification) shows keratinized stratified squa-
mous epithelium with dense inflammation in superficial stroma. (B) Chronic limbitis:
Signifies the chronic inflammation of Limbus associated with atrophic epithelium
and fibrosis of stroma. Section (10 magnification) shows atrophic stratified squamous
epithelium. Underlying stroma shows lymphocytes and plasma cells with few dilated
blood vessels.
the corneal surface, and the limbus might also function as a physical barrier
that prevents conjunctival epithelium from growing onto the cornea. In
LSCD, the corneal epithelium cannot be renewed and eventually replaced
by the conjunctival epithelial cells (Thoft and Friend, 1983).
Limbal Epithelial Cells (LECs) extended from the peripheral aspect of the
undersurface of an interpalisade rete ridge and extended either radially into
the conjunctival stroma parallel to the palisade or circumferentially along the
limbus at right angles to the palisade. The structure was widest at its origin
from the rete ridge and gradually tapered to a narrow extension at its termi-
nation (Dua et al., 2005). The corneal epithelial cells undergo constant
renewal and regeneration. The stem cells responsible for this corneal epithe-
lial renewal are presumed to reside within the basal epithelium at the limbus
(Cotsarelis et al., 1989; Lavker et al., 2004; Schemer et al., 1986).
3. CELL-BASED THERAPIES FOR OCULAR SURFACE

REGENERATION
Stem cells from iris pigment epithelium, ciliary body epithelium, and
choroidal epithelium have shown promise for retinal or neural tissue replace-
ment. Trabecular meshwork, orbital and sclera stem cells have properties
similar to MSCs but their potential is yet to be experimentally validated
(Dhamodaran et al., 2014). The following paragraph summarizes application
of embryonic and adult stem cells in ocular surface reconstruction.
3.1 Embryonic Stem Cells

ESCs are characterized by their capacity to proliferate indefinitely and to
differentiate into any cell type. Homma et al. (2004), for the first time,
attempted the use of ESCs to reconstruct corneal epithelial cells. They suc-
cessfully induced epithelial progenitors, in vitro, from mouse ESCs and were
able to completely reepithelialize the corneal surface within 24 h after trans-
plantation in mouse. ESCs were further shown to be differentiated into
corneal epithelial cells under controlled conditions (Wang et al., 2005).
Ahmad et al. (2007) have shown that ESCs can be differentiated into corneal
epithelial-like cells by in vitro replication of corneal epithelial stem cell
niche. ESCs have also been shown to be differentiated into corneal kerato-
cyte phenotype (Chan et al., 2013) and transplanted on partially wounded
human cornea, in vitro (Hanson et al., 2013). Recently, they have been
shown to enhance the reconstruction of highly proliferative auto-tissue-
engineered lamellar cornea, when cocultured with corneal epithelial cells
(Zhou et al., 2014).
Zhu et al. (2013), reconstructed the ocular surface by acellular porcine
cornea matrix scaffold and LSCs derived from human ESCs. These cells,
when applied to damaged ocular surface in rabbit total LSCD models, suc-
cessfully reconstructed the surface and alleviated the invasion of corneal neo-
vascularization. Corneal epithelial cells were also induced from ESCs by
Kayama et al. (2007), by culturing them on type IV collagen or alternatively
by induction of Pax6 into ESCs. This supports their possible application for
ocular surface reconstruction.
Instead of their wide publicity and successful transdifferentiation into
different types of cells, the ESC-based clinical trials are comparatively fewer
in number than the iPSCs and MSCs. This is due to the safety concerns
related with the teratoma formation and ethical controversies. A recent study
by Schwartz et al. (2015) provides the first evidence of the medium-to-long-
term safety, graft survival, and possible biological activity of ESC progeny in
individuals with any disease. They used the human ESC-derived retinal
pigment epithelium in patients with age-related macular degeneration and
Stargardt’s macular dystrophy with a follow-up of two open-label phase
1/2 studies.
3.2 Induced Pluripotent Stem Cells

After the report of somatic cell reprogramming into a pluripotent ES cell-like
state (termed “induced pluripotent stem cells”) in humans by Takahashi et al.
(2007), revolutionary changes took place in the field of regenerative science.

The application of patient-specific iPSCs for tissue regeneration has been an
exciting area of research in the last few years. Chien et al. (2012), successfully
differentiated corneal keratocytes into iPSCs and effectively used them for
corneal repair with amphiphatic carboxymethyl-hexanoyl chitosan hydrogel.
Later on Hayashi et al. (2012) generated corneal epithelial cells from iPSCs
derived from human dermal fibroblast and corneal limbal epithelium.
Recently, Yu et al. (2013), differentiated mouse iPSCs into corneal epithe-
lial-like cells, while coculturing them with corneal limbal stroma. Some of
the applications of iPSCs in ocular surface stem cell therapies have also
been reviewed by Angunawela et al. (2013). Although the iPSCs are still
not much used for ocular surface regeneration, the above-mentioned find-
ings indicate toward the potential application of iPSCs for corneal repair
and LSCD management in coming years.
3.3 Mesenchymal Stem Cells

MSCs are a kind of multipotent progenitor cells (Du et al., 2005) which
have been suggested as patient-specific drug house for injured tissues.
They are partially defined by their ability to differentiate into tissues
including bone, cartilage, and adipose in vitro, but it is their trophic, para-
crine, and immunomodulatory functions that are supposed to have the
greatest therapeutic impact in vivo. Irrespective of pharmaceutical treat-
ments which deliver a single agent at a specific dose, MSCs are site-regulated
and secrete various bioactive factors and signals at variable concentrations in
response to the local microenvironmental cues. The anti-inflammatory and
immunomodulatory capacities of MSCs may be paramount in the restora-
tion of localized or systemic conditions for normal healing and tissue regen-
eration (Murphy et al., 2013).
MSCs were originally identified in the bone marrow (Friedenstein et al.,
1968), and thereafter reported in many other tissues, including the adipose
(Zannettino et al., 2008), heart (Hoogduijn et al., 2007), Wharton’s jelly
(Hoogduijn et al., 2007), dental pulp (Jo et al., 2007), peripheral blood
(He et al., 2007), cord blood (Oh et al., 2008), menstrual blood (Meng
et al., 2007; Hida et al., 2008; Patel et al., 2008), fallopian tube (Jazedje
et al., 2009), and limbal stroma of the human eye (Polisetty et al., 2008).
In the late 1980s and early 1990s, the heterogeneous population of MSC
from Bone Marrow (BM) was explored and found to be linked to the devel-
opment of various mesenchymal tissues, as well as identifying the first surface
antigens expressed by MSC (cluster of differentiation (CD)73 and CD105)
(Haynesworth et al., 1992). Due to their linkage with the formation of

mesenchymal tissues during embryonic development, these cells were
termed “MSCs” (Caplan, 1991).
3.3.1 MSCs in corneal reconstruction

Cornea appears as an “outer door” of the eye. Continuous corneal regener-
ation is necessary to maintain it in the transparent state that is essential for
vision. Therapy for repair of the damaged anterior cornea is currently
addressed through the transplantation of donor corneas or the delivery of
LESCs to the ocular surface using amniotic membrane (AM) as a supporting
scaffold. However, due to various corneal disorders which commonly
initiate from inflammation, trauma, systemic disease, as well as pathological
changes from adjacent tissues, and eventually result in impaired vision, even
blindness due to vascularization, conjunctivalization, keratinization, corneal
scarring, and opacification, the integrity and transparency of the cornea is
compromised. Because of their anti-inflammatory and modulatory effects
on corneal angiogenesis, MSCs have potential therapeutic value in corneal
reconstruction. They are useful in suppressing corneal transplantation rejec-
tion and facilitating corneal wound healing (Lan et al., 2012; Yao et al.,
2012). Intravenously injected MSCs engrafted to the injured cornea and
promoted wound healing by differentiation, proliferation, and synergy
with hematopoietic stem cells in an in vivo study of the rabbit alkali burn
model (Ye et al., 2006). In order to study the corneal allograft failure, an
in vivo study on a pig model proved that topical application of allogeneic
rat MSCs does not prolong corneal xenograft survival effectively in a pig-
to-rat model (Oh et al., 2009).
However, additional animal model research is needed to address ques-
tions regarding how to transdifferentiate them into corneal epithelial cells,
the most appropriate route and time for applying MSCs for different kinds
of corneal reconstruction, the specific mechanisms (Yao and Bai, 2013).
3.3.2 MSCs in LSCD

LSCs derived from the basal region of limbus are being widely used in treat-
ment of LSCD and associated disorders (Vazirani et al., 2014; Lal et al., 2013;
Bhalekar et al., 2013; Sangwan et al., 2012, 2011) (Figure 4). They have
been described in much detail in the earlier section of this review. Limbal
stroma has been reported to be a good source of MSCs (Hashmani et al.,
2013), and stem cells isolated from the stroma have been observed to contain
immunosuppressive properties (Garfias et al., 2012). MSCs derived from the
(B)
(A)
(F)
(C)
(D) (E)
Figure 4 Potential application of stem cell-based therapies in treating limbal stem cell
deficiency (LSCD). (A) Normal eyes, (B) Unilateral LSCD is treated by CLET/SLET, with re-
ported success. (C) Bilateral LSCD, (D) Transplantation from cadaveric or live related
donor eye remains the only option for bilateral LSCD. However, long-term dependency
on immunosuppressant resulting in moderate-to-severe side effects with chances of
graft rejection limits the success of treatment. (E) Alternatively, stem cells (ES, iPSC,
MSC) of preferably autologous or allogeneic (in case of unavailability of autologous
sources) origin can be transplanted to reduce the chances of graft rejection; however,
the success of such treatment and the underlying mechanism is under question and
needs to be investigated further. (F) Normal eyes after restoration of vision. The green
(gray in print versions), red (dark gray in print versions), and blue (light gray in print ver-
sions) dotted circles represent the healthy limbus, diseased limbus with LSCD, and allo-
geneic limbal graft, respectively.
human limbal niche may participate in angiogenesis and regeneration during

corneal wound healing (Li et al., 2012).
Differentiation of MSCs into human corneal epithelial cells is a critical
step in MSC-based regenerative therapies for corneal repair and ocular sur-
face regeneration. Most recently, Harkin et al. (2015) have evaluated all the
major studies performed with this goal, on the basis of following methodo-
logical criteria: (1) whether or not, the authors have used appropriate
markers to determine transformation into a corneal phenotype; (2) whether
or not, the appropriate methods have been used to evaluate the expression of
corneal markers; (3) whether appropriate controls have been used to validate
results; and (4) whether, the origin of resulting “corneal cells” have been
traced back to the MSCs of noncorneal origin using appropriate markers.
Based on the analysis, Harkin et al. concluded that (1) BM-MSCs have
some ability to produce low levels of cornea-specific keratins, (2) studies
using adipose-derived MSCs indicate toward their likely applicability for
corneal differentiation, and (3) umbilical cord-derived MSCs might provide
Figure 5 Potential questions to be answered in context of successful mesenchymal

stem cell (MSC) transplantation in different corneal epithelial disorders.
a partial surrogate for corneal endothelial cells in case of their failure to trans-
differentiate. Also, Meyer-Blazejewska et al. (2011) used hair follicle-
derived MSC for the treatment of LSCD in a mouse model. However, in
majority of these studies, it is still not very clear as to whether this level of
expression represents true conversion to a functional corneal epithelial cell
phenotype. Therefore, further studies are required with appropriate meth-
odological controls, to validate the transdifferentiation of MSCs into corneal
epithelial phenotype (Figure 5).
4. NEW MATERIAL TECHNOLOGIES FOR OCULAR

SURFACE RECONSTRUCTION
There are two important considerations for tissue regeneration to be
successful. The first is the selection of cells that can achieve the function
of the native tissue (with the associated development of techniques suitable
for isolating and characterizing them). The second consideration is the
choice of material that is used as a carrier or scaffold for cell transplantation.
The latter is as important a consideration as the former, particularly with
stem cells, as the interaction between the cells and the substrate can define
their differentiation characteristics.
For the treatment of LSCD specifically, several substrates have been
developed over the last few years; however, human amniotic membrane
(hAM) remains the gold standard for the culture and transplantation of these
cells for several good reasons as discussed below. The hAM has been used as a
temporary biological material for the regeneration of various tissues such as
exposed pleura, blood vessels, tendon, nerve, bones and extensively in skin
regeneration following burns (Mamede et al., 2012). In ophthalmology, its
first use was reported in the 1940s when it was employed to treat conjunc-
tival defects (Fernandes et al., 2005). Since then, hAM has been the material
of choice in the treatment of various corneal disorders such as persistent
epithelial defects, shield ulcers, microbial keratitis, bullous keratopathy and
in the treatment of LSCD which began in 2000 (Koizumi et al., 2000;
Tsai et al., 2000). The popularity of this material can be attributed to its
anti-inflammatory and antimicrobial properties, in addition to its apparent
low immunogenicity (Sangwan and Basu, 2011).
This fetal membrane normally supports a single layer of columnar
epithelial cells with an underlying layer of stroma containing fibroblasts.
Thus it already possesses the necessary basement membrane components
to promote the migration, proliferation, differentiation and maintain the
viability of epithelial cells that are cultured on it. Studies have also shown
that hAM helps reduce inflammation although the exact mechanism is
not clear. It is known that the cells of the hAM express many of the anti-
inflammatory and antiangiogenic factors such as interleukin (IL)-1, IL-2,
IL-8, interferon g, tumor necrosis factor-b, basic fibroblast growth factor,
platelet-derived growth factor, and thrombospondin-1 (Hao et al., 2000)
thereby pacifying the inflamed ocular surface and preparing it to receive
the transplanted cells. It has also been demonstrated that hAM reduces
fibrosis and scarring associated with wound healing primarily by downregu-
lating the expression of TGFb and its receptors in fibroblast cells.
Finally, hAM is considered to be a nonimmunogenic material. This is
said to be due to the absence of the HLA family of antigens but some studies
have shown that HLA-class 1a and 1b antigens are expressed by the epithe-
lium and fibroblast cells of the AM rendering this tissue immunogenic (Hunt
et al., 1988). The apparent nonimmunogenicity may be due to the
following reasons: (1) the process of cryopreservation in 50% glycerol is
thought to render the cells nonviable and therefore nonimmunogenic and
(2) the expression of immunomodulatory factors such as HLA-G and Fas
ligand that makes this tissue immune-privileged (Kubo et al., 2001).
The hAM has been used extensively for the regeneration of the ocular
surface in patients with LSCD (Sangwan et al., 2011; Nakamura et al.,
2003b; Tsai et al., 2000; Shortt et al., 2008). A recent study analyzing the
outcomes following transplantation of cells cultured on hAM has shown
that the procedure was successful in 70% of 200 individuals. This has thus
far been the largest cohort to be analyzed and the follow-up in this study
ranged from 1 year to 10 years suggesting promising long-term outcomes
(Sangwan et al., 2011). The hAM degraded within 4 weeks leaving behind
a stable ocular surface; it did not elicit any immune response and no other
adverse events were reported. Thus there are several positive attributes
with the use of hAM for ocular surface reconstruction.
However, there are also a few drawbacksdthe most important of these
being the requirement to reduce risk for patients in the use of this human
donor tissue. Thus there is a high cost involved in the testing for pathogens
and the preparation of the tissue and storage under well-managed tissue
bank facilities, all of which require dedicated premises. These factors in addi-
tion to variation in the thickness of the processed tissue, degradation time, and
nontransparency have warranted research in developing materials for replac-
ing the AM. Several such alternatives have been tested and compared to this
“gold standard.” Some of these include thermosensitive substrates, collagen-
based scaffolds, lens capsule, surface-treated contact lenses, synthetic polymer
scaffolds, etc. (Nishida et al., 2004a; Albert et al., 2012; Dravida et al., 2008;
Deshpande et al., 2009, 2013; Brown et al., 2014). Feng et al. have recently
given a more exhaustive description of all the materials that have been devel-
oped for ocular surface reconstruction thus far (Feng et al., 2014).
Thus the challenge is to provide a product that is safe for use in man and
will support the stem cell population sufficiently but allow cells to eventually
adhere to the ocular surface and not produce any toxic breakdown products.
Further it needs to be easily available for widespread use and economical.
Discussed below are some promising alternates to the AM that have been
divided into biological derivatives and synthetic materials.
4.1 Biological Materials

4.1.1 Fibrin
Fibrin is the other most commonly employed substrate outside of hAM for
the culture and transplantation of limbal epithelial cells. Fibrin is made by
combining fibrinogen and thrombin, derived from donor human plasma,
to mimic the last stages of the natural clotting mechanism in the body to
form a mass that adheres to the injury site thus sealing the tissue and allows
for its regeneration. This commercially available material was first employed
by Pelligrini’s group in treating LSCD in 1997 and showed that cells
cultured on fibrin were able to successfully restore the ocular surface in in-
dividuals with chemical burns (Pelligrini et al., 1997). In a larger cohort of
112 patients, this group further showed that transplantation of limbal epithe-
lial cells cultured on fibrin was successful in 76.6% of patients and that the
restored ocular surface was maintained for up to 10 years posttransplantation.
These studies have shown that reepithelialization of the ocular surface
occurred in the first week and a complete transparent surface was achieved
within a month. Studies have shown that when compared to other matrix
materials such as collagen I and Puramatrix, fibrin allows for better cell
growth and adhesion of cells (Forni et al., 2013). Further, fibrin was shown
to be capable of maintaining the holoclone-forming stem cell population in
culture in the presence of mitotically inhibited NIH3T3 feeder cells. These
holoclones were capable of producing at least 90 generations before reaching
senescence (Rama et al., 2001). When transplanted, the fibrin gels degraded
within 3 days in vivo to leave behind a clear transparent epithelial surface
(Talbot et al., 2006) with regression of preoperative vascularization and
opacification.
The near-identical results obtained with the use of fibrin sealant (Rama
et al., 2010) and hAM (Sangwan et al., 2011) suggest that both these mate-
rials are comparable in their ability to support the culture and transplantation
of limbal epithelial cells for the restoration of the ocular surface. However,
the material is just as expensive as the hAM and may not be affordable to all.
4.1.2 Collagen-based materials

Collagen constitutes the most abundant protein in the body and is the main
structural protein of the cornea so it is a natural choice as a substrate for the
limbal epithelial cells. Collagen is biocompatible, exhibits low immunoge-
nicity, and is easily available. Studies have shown that epithelial cells cultured
on it were well adherent and stratified and when applied to the ocular sur-
face induced minimal immune response (Dravida et al., 2008; Fagerholm
et al., 2014). While the use of animal-derived collagen (porcine) has been
reported to be safe for use in vivo (Fagerholm et al., 2014; Hackett et al.,
2011), there is still a concern over its purity and therefore immune rejection.
To overcome this concern, recombinant collagen may be used; however,
this would greatly increase the cost of production since recombinant tech-
nology is not yet amenable to bulk production. Collagen has been shown
to be noncytotoxic and slowly degrading while being replaced by host
tissue.
While collagen is the obvious choice for reconstructing the corneal
stroma or the ocular surface, the main drawback with this material has
been its inherent weakness resulting from its high water content. This has
made handling the material (e.g., suturing) cumbersome. Cross-linking
collagen with chemicals such as 1-ethyl-3-(3-dimethylaminopropyl) carbo-
diimide or N-hydroxysuccinimide has been shown to improve the mechan-
ical strength of the material allowing easy manipulation during surgery
(Hackett et al., 2011). This also serves to help maintain the shape of the
cornea following transplantation. Extensive work in this regard has been
done by the Griffith group where they have used human recombinant
type III collagen (RHCIII) and cross-linked this with the above-mentioned
chemicals for improving its mechanical strength. It was shown that the lim-
bal cells grew well on these constructs, were well stratified, and when trans-
planted into pig’s eyes, there was good integration, reepithelialization, and
reinnervation of the material (Hackett et al., 2011; Dravida et al., 2008).
Further, the transparency of the material was also shown to be comparable
to native cornea. Recently, it was demonstrated by the same group that acel-
lular cross-linked collagen could be used for replacing a part of the corneal
stroma in human subjects with keratoconus or corneal scarring using the
procedure of anterior lamellar keratoplasty (Fagerholm et al., 2014).
Follow-up of these patients over 4 years has shown that these implants
were well integrated into the stroma, transparent, and allowed for epithelial
cell and nerve regeneration. While the procedure replaced only the anterior
layers of stroma, there was no immune rejection reported in any of the sub-
jects and no need for long-term immunosuppression, when compared to us-
ing human donor corneas, which the authors attribute partly to the low
immunogenicity of the material (Fagerholm et al., 2014). This work has
promise in the regeneration of the corneal tissues, provided the material
strength is further improved using better cross-linking procedures or chem-
icals since it was found to be still insufficient to allow peripheral sutures to be
placed anchoring the material to the sclera.
4.1.3 Silk fibroin-based materials

The fibroin protein of silk is derived from the cocoon or glands of the silk
worm and processed to make scaffolds of different thickness and topogra-
phies. This material has been found to be transparent, exhibits good me-
chanical strength, and is biocompatible. Evidence for its biocompatibility
comes from its current use in making medical sutures courtesy of its slow
degradation property and more recently in regeneration of bone, cartilage,
and tooth (Kundu et al., 2013). There are a few studies that have been con-
ducted to determine the ability of this material to support corneal epithelial
cells (Higa et al., 2011; Chirila et al., 2008; Bray et al., 2013). In all these
studies the silk protein has been derived from the commercially available va-
riety of silk worm Bombyx mori. The addition of pores to improve the
permeability was seen to improve cell adhesion and stratification of the
limbal epithelial cells (Higa et al., 2011; Bray et al., 2013). With these
improvements, Bombyx Mori silk has been demonstrated to be as good as

hAM in supporting limbal epithelial cells including the stem cell population
based on marker expression and holoclone formation (Liu et al., 2012; Higa
et al., 2011). Interestingly, the addition of topographies to the silk surface,
especially parallel lines, seems to enhance the adhesion, proliferation, and
spreading of cells on the silk surface (Lawrence et al., 2012; Gil et al.,
2010). Bray et al. (2013) have attempted to construct the corneo-limbal re-
gion by creating a dual-layer of silk with the limbal epithelial cells on the top
layer and stromal cells located in the bottom layer demonstrating that it is
possible to 3D engineer the limbus. Finally the implantation of silk into
the stroma of rabbit’s eyes was found to integrate with the stroma without
eliciting much of an immune response over at least 6 months (Higa et al.,
2011). This could partly be attributed to the avascular nature of the central
cornea, and it remains to be seen if the biocompatibility will hold true if silk
is transplanted into the highly vascular region of limbus or conjunctiva. Thus
silk as a candidate for ocular surface regeneration displays immense potential;
however, further studies to characterize its degradation properties and
biocompatibility will need to be undertaken.
4.2 Synthetic Materials

More recently several synthetic materials are being developed as alternatives
for hAM in LSC transplantation. The main advantages of this approach will
be the nonbiological nature of the material hence less worry about transmit-
ting diseases or eliciting an immune response and their accessibility through
bulk production. Some promising developments are discussed below.
4.2.1 Contact lenses

These have been reported to be simple but effective carrier devices for the
transfer of limbal epithelial cells since the material is biocompatible, FDA
approved, transparent, and nonimmunogenic. Hydrophilic siloxane hydro-
gel contact lenses were the first to be used to regenerate the ocular surface in
patients with LSCD (Di Girolama et al., 2009). This modality was found to
be useful in regenerating a stable ocular surface in these patients with good
improvement in visual acuity and no recurrence of LSCD up to 13 months
of follow-up. However, it is notable there have not been subsequent studies
in man, and so further information on long-term outcomes is unavailable.
The main drawback reported with this technique was the limited transfer
that was achieved during the process since some pockets of cells were
removed along with the removal of contact lenses. Plasma polymer coating
has been done to add acid groups to the contact lens’ surface to enhance cell
adhesion. It has been shown that acrylic acid enhances the hydrophilicity of
the surface (Deshpande et al., 2009) increasing cell adhesion while octadiene
makes the surface hydrophobic resulting in no cell adhesion (Brown et al.,
2014). These surface-coated lenses were able to maintain the stem cell pop-
ulation as evident from the marker expression and also transfer some of these
cells to the ocular surface in rabbits (Deshpande et al., 2009). However,
animal study showed that the transfer was at best partial and that persistent
epithelial defects were noticeable (Brown et al., 2014). Though the contact
lenses score high in terms of their ease of use, biocompatibility, mechanical
properties, etc., transfer of sufficient cells to the ocular surface to reconstruct,
it remains a challenge to overcome.
4.2.2 Polylactic glycolic acid

More recently, synthetic scaffolds made of polylactic and glycolic acid
(PLGA), polymers used in dissolvable sutures (Dunlap et al., 1976), are be-
ing developed as potential transfer materials for limbal epithelial cells.
PLGA has been used for many purposes including drug delivery, diagnos-
tics, and other applications of clinical and basic science research as it is
biodegradable, biocompatible, and FDA approved (Leenslag et al., 1987;
Eppley and Li, 2003; Orr et al., 2003; Sharma et al., 2006; Yasukawa
et al., 2005). The success of PLGA as a product lies in the fact that it de-
grades by hydrolysis in the presence of water to produce the monomers lac-
tic and glycolic acids. Under normal physiological conditions, these are
generated as by-products of various metabolic pathways and are effectively
removed from the body (Tsuji and Ikarashi, 2004). Thus there is minimal
systemic toxicity reported with the use of PLGA in the clinic. This in addi-
tion to the flexibility that it offers to vary the degradation time by simply
altering the ratio of the monomers has made PLGA a preferred choice
for producing various biomedical devices such as sutures, implants, pros-
thetic devices, micro- and nanoparticles.
In a recent study, it was shown that PLGA scaffolds, prepared by the
technique of electrospinning, provide support for the culture of both rabbit
and human limbal cells (Deshpande et al., 2013). The cells retained their
epithelial phenotype as confirmed by marker expression of CK3/12 and
also the stem cell population (p63a). Further, quantification of label-
retaining cells in culture showed that the BrdU positive cells at the end of
the chase period was comparable between culture of cells on the hAM
and the PLGA scaffold suggesting that the PLGA provides a suitable
Figure 6 Quantification of label retaining cells cultured from human limbal explants:
Explant culture of LEC was carried out for 14 days on human amniotic membrane
(hAM) and on fibrin-coated membranes. On the 7th day of culture, BrdU (5 mM) was
added to the cells and incubated for 24 h (pulse). The cells were further cultured in
BrdU-free medium for 7 days (chase). Cells were fixed either at the end of the pulse
or chase period and stained for BrdU positive cells. The number of positive cells was
counted from images as shown in the figure. Percent of label retaining cells at the
two time points was calculated by dividing the number of BrdU (green (white in print
versions)) positive cells by the total number of cells identified using propidium iodide
(PI; red (gray in print versions)) and plotted in the graph. Magnification 10. PLGA, pol-
ylactic and glycolic acid. Generated with permission from Deshpande et al. (2013).
substrate for maintaining the stem cell population (Figure 6). Experiments,
using an ex vivo rabbit corneal model, showed that the material not only
supported the limbal epithelial cell population but it also allowed these
cultured cells to be transferred to the ocular surface. Our own studies in an-
imal models (rabbits) showed that the application of PLGA membranes
attached to the eye with fibrin glue did not induce any adverse immune
response, no evidence of vascularization or inflammation or toxicity during
the 4 weeks in which the animals were studied (Figure 7). Storage and pack-
aging studies showed that these membranes could be stored dry and vacuum
packed at 20 C for at least 2 years and still support corneal regeneration in
Figure 7 Toxicity test of polylactic and glycolic acid (PLGA) scaffolds in rabbits. PLGA
scaffolds were applied to the ocular surface of rabbits and followed for a period of
28 days to check for induced ocular toxicity. The top row of images shows fluorescein
staining before surgery and 7, 15, and 28 days postsurgery. There was no noticeable
local ocular surface toxicity following the application of PLGA. The bottom row of im-
ages shows the corresponding fundus images of these rabbits again indicating that the
retinal structure remains intact following PLGA application.
vitro. Thus PLGA appears as a good alternative to the hAM; however, hu-
man application needs to be evaluated before any conclusions can be drawn.
4.2.3 Thermoresponsive substrate

This is an approach in which the cells are cultured as a sheet on a polymer of
poly(N-isopropylacrylamide) (PIPAAm) on polystyrene dishes. The cells
grow into multilayers on the polymer at 37 C, and then when the temper-
ature is reduced to less than 32 C beneath the lower critical solution tem-
perature of the material, the polymer changes conformation and the cells
are released from the substrate as a sheet without greatly affecting their
integrity. However, the sheet needs to be supported on either nitrocellulose
membrane or gauze for transport from lab to the theater for transplantation
since the cell sheet alone was quite fragile. Nishida et al. (2004b) have
shown that limbal epithelial cells isolated using this technique exhibited
the corneal epithelial morphology, stratified to form multilayered cell com-
plex, expressed the stem cell and differentiated cell markers and retained
their ability to form colonies in culture. Importantly, they showed that
the isolation of cells using this technique when compared to the use of
an enzyme dispase allowed for the extracellular matrix to be retained thus
allowing for easy manipulation and maintenance of the structural integrity
of the cell layer. As expected, the transparency achieved using this tech-
nique was far superior when compared to the use of either fibrin or
hAM as only the cells are being transplanted. In this study, the authors also
showed that it is quite simple to transplant the cultured cells onto the ocular
surface of rabbits without the need for the use of tissue glue or sutures to
stick the cells to the ocular surface with complete restoration of its integrity
(Nishida et al., 2004a). The same group in another study showed that oral
mucosa epithelial cells cultured on the thermoresponsive substrate are
capable of regenerating the ocular surface with little or no complications.
The stratified epithelial sheet was shown to improve the postoperative
best corrected visual acuity, with least evidence for vascularization and
that the status was maintained up to 1.5 years of follow-up time (Nishida
et al., 2004a). Sitalakshmi et al. (2009) used a slightly different composition
of the material commercially available as mebiol gel made of poly(N-iso-
propylacrylamide-co-n-butyl methacrylate) and polyethylene glycol. In
their study, these authors have shown that the gel supports the culture of
limbal epithelial cell sheets from explants that successfully restore the ocular
surface in rabbits with LSCD. Thus the use of thermosensitive surface for
the culture of limbal epithelial cells appears as a viable and a convenient
alternative to hAM with no worry of tissue rejection, toxicity, or infection.
4.3 Engineering Limbal Niche

The reconstruction of the ocular surface using cultured cells from the limbus
can be considered as one of the best success stories in regenerative medicine.
However, there are still many pertinent questions that remain unanswered
that could help in improving the long-term functional outcome and
decrease the failures. For example, what happens to the stem cells that are
transplanted to the ocular surface? Do these home into their anatomical
niche or do they remain on the cornea?
The niche is composed of the mesenchymal cells within the limbal
stroma, and it has been hypothesized that these cells greatly regulate the dif-
ferentiation of the stem cells both in vivo and in vitro. Further, studies have
shown that the limbus has a unique basement composition that is quite
different from the cornea. There is preferential expression of a9 integrin,
N-cadherin, laminin-1, laminin-5, and the a1,a2, and a5 chains of collagen
IV which are not a part of the corneal basement membrane (Stepp et al.,
1995; Hayashi et al., 2007; Matic et al., 1997; Tuori et al., 1996; Ljubimov
et al., 1995; Schl€
otzer-Schrehardt et al., 2007). The differential expression of
these components between the limbus and the cornea may be responsible
for conferring the niche status to the limbus. Thus it becomes apparent
that the material being used for the construction of the limbal niche should
be able to support the different types of cells while allowing for permeabi-
lization of various growth factors and vasculature.
There have been a couple of attempts at engineering the limbal niche
structure using collagen (Levis et al., 2013) and PLGA (Ortega et al.,
2013). In one technique, a combination of microfabrication and electro-
spinning was used to create the 3D architecture of the limbal niche (Ortega
et al., 2013). First crude “niches” of 300 mm were created using microster-
eolithography on polyethylene glycol diacrylate (PEGDA) which func-
tioned as the mold over which PLGA fibers were then electrospun.
While PEGDA by itself is a poor support for cells and required functional-
ization with fibronectin for the cells to adhere, the combined approach of
making a PEDGA template to provide horseshoe-shaped niches within
PLGA, which does not require functionalization, showed that this was sup-
portive of the rabbit LEC.
With collagen, a simple method of adding ridges to collagen during poly-
merization by using ridged hydrophilic porous absorbers was adopted to
make niches of various widths (Levis et al., 2013). Human limbal epithelial
cells including the limbal stromal cells were then seeded onto these ridges
that structurally mimicked the limbal niche. Histological sections through
these niches following stratification showed that the limbal epithelial cells
formed a multilayer of six to seven cells. The top layer of cells were shown
to be the more differentiated squamous epithelium while p63a positive cells
were located at the bottom of the ridge closer to the limbal stromal cells
(Hannah et al., 2013). Both these techniques are promising and could poten-
tially be used to replace or create the limbal niche in vivo. Tables 1 and 2
compare the properties, advantages, and disadvantages of these materials.
In summary, some apparent drawbacks with the use of hAM (safety,
availability, and cost primarily) have driven the field to seek alternatives to
this material. Fibrin has been accepted to be a good alternative to the
hAM and has by far the largest cohort of human subjects with long-term
follow-up second to hAM. Despite this, fibrin has not gained widespread
acceptance in clinical usage possibly due to restricted availability and forbid-
ding cost. For the other alternative materials developed, the potential for
their use in ocular surface reconstruction is promising; however, the avail-
able human data are scarce. More clinical trials using these materials need
to be conducted in order to truly establish their clinical potential. In conclu-
sion, an optimal alternative to the hAM for reconstruction of the ocular sur-
face should be transparent, biocompatible, exhibit sufficient mechanical
strength, remain cost-effective, and support the growth of LSCs.
Limbal Transplantation: Science and Art
Table 1 Material composition, advantages, and disadvantages
Material Source How is it used? Clinical usage? Advantages Disadvantages
Human donor Human Cells are cultured on it Been in clinical use since Known clinical efficacy Small risk of viral disease
amniotic and then cells with 2000 by at least three and compatibility transmission as it is a
membrane membrane are to four groups Reasonable to handle human donor tissue
transplanted to the worldwide and long- Has anti-inflammatory Availability depends on
cornea post removal term (10 years) results and antimicrobial access to materials
of scar tissue reported in largest properties through tissue banks
cohort of 200 patients Supports stem cells well Variability in processing
indicating good and performance
clinical outcomes
(Nakamura et al.,
2004b, 2003a; Shortt
et al., 2008; Sangwan
et al., 2011)
Fibrin Human Cells are cultured on it In use since 2001 by one Known clinical efficacy Made from donor blood
and then cells on the group and long-term and compatibility productsdsmall risk
fibrin are transplanted (10 years) results on Supports stem cells well of disease transmission
to the cornea post 166 patients
removal of scar tissue published indicating
good clinical
outcome (Rama
et al., 2001, 2010)
(Continued)
69
70
Table 1 Material composition, advantages, and disadvantagesdcont'd
Material Source How is it used? Clinical usage? Advantages Disadvantages
Collagen-based Animal Either as compressed Has not been tested in Biocompatible Has poor mechanical
materials (porcine) collagen or cross- human subjects for strength making
and human linked with chemicals limbal cell handling difficult
recombinant to strengthen the transplantation
material
The material along with
cultured cells are
transplanted to the
cornea
Silk fibroin Insect derived The material along with Has not been tested in Known biocompatible High cost of source
cultured cells will be human subjects Easy to handle materials
transplanted to the Small chance of
cornea immune rejection
Contact lenses Synthetic Cells are cultured on the Used for the Known biocompatible Incomplete transfer of
contact lens and transplantation of Easy to handle cultured cells to the
applied to the ocular limbal cells in humans Supports stem cell ocular surface
surface Lens removed (three subjects) (Di population requiring more than
3e4 days Girolama et al., 2009) one application
Vivek Singh et al.

posttransplantation No long-term
outcomes reported
and not widely used
in clinical practice
PLGA Synthetic Cells are cultured on the Has not been tested in Known biocompatible Mechanical strength
material and applied human subjects Supports stem cells well reduces with
to ocular surface Reasonable to handle increased cell culture
time
Thermoresponsive Synthetic Cells are cultured on it Has been used for the No worry of immune Cell sheets are fragile
substrate and then harvested as transplantation of oral rejection without any support
a cell sheet by mucosa cells to ocular Simple technology to and would require
lowering the surface in humans adopt skills for transfer to
temperature. The cell subjects (four ocular surface
sheet is then subjects) (Nishida
transplanted et al., 2004b)
No long-term
outcomes reported
and not widely used
in clinical practice
71
72
Table 2 Material properties
Material Transparency Degradation Mechanical strength Safety Availability
Human Limited 4e6 weeks Reasonable Known biocompatible Restricted to few tissue
amniotic in vivo (humans) and eye banks
membrane
Fibrin Transparent 3e4 days Reasonable Known biocompatible Individual components
in vivo (humans) commercially
available
Collagen Transparency e Poor in native form, No immune response Commercially available
dependent on moderate when reported from animal sources
water content compressed or and as recombinant
cross-linked with protein
other chemicals
Silk fibroin Transparency Slowly degrading Good No immune response Commercially available
dependent on reported
thickness
Contact lenses Transparent NA Good Known biocompatible Commercially available
PLGA Limited 4e6 weeks in vivo Reasonable Reduces Known biocompatible Commercially available
(rabbits) when with time as base polymers
prepared in 50:50
ratio of lactic to
Vivek Singh et al.

glycolic acid polymers
5. ANIMAL MODELS FOR LSCD

Animal models are necessary to understand complex biological inter-
actions and diseases, to investigate new therapies, and to protect the safety of
patients and their environment. Most laboratory animals (95%) are rodents
(rats and mice), because they are small, easily bred, and relatively inexpensive
to house. A variety of genetically altered rodents (knockout and transgenic
models) are available for scientists to study the pathophysiology of various
eye diseases/disorders. In the past, animal research has facilitated many med-
ical breakthroughs. In this section, we aim to review animal models used to
study LSCD, and corneal wound healing. Animal models for LSCD are use-
ful to evaluate the safety and efficacy of cultured cell therapies prior to use of
these treatments in human patients. However, safety and toxicity of culti-
vated LESCs can only be tested in few animal models prior to their use in
human trials.
Pellegrini et al. (1997) first described ex vivo cultured LESCs as a source
for autologous limbal cells in humans. Then many clinical trials using
amniotic membrane transplantation (AMT), keratolimbal grafts, transplan-
tation of ex vivo expansion of limbal epithelial cells, limbal and conjunc-
tival cells (Sangwan et al., 2007; Burman and Sangwan, 2008; Basu
et al., 2012b; Sangwan et al., 2012), oral mucosal cells have been per-
formed to treat LSCD with promising clinical outcome (Gaddipati et al.,
2014a; Kolli et al., 2014). Several animal models with pathophysiological
conditions similar to human LSCD have been created to investigate the
mechanisms involved in the development of LSCD; these animal models
can be divided into two categories: (1) animal models due to genetic de-
fects and (2) animal models due to mechanical injury of limbal/corneal
epithelial cells.
5.1 LSCD due to Genetic Defects of Limbal/Corneal

Epithelial Cells
Mice heterozygous for a PAX6 mutation (PAX6þ/) are characterized by an
overall small body size, small eyes with coloboma, small or lacking lenses with
cataract, abnormal folding of the retina, and reduction of the pigment layer,
causing an eye disease called aniridia. Aniridia-related keratopathy (Lee et al.,
2012) is a corneal deterioration with irregular thickening of the peripheral
corneal epithelium, in-growth of blood vessels from the limbus, stromal
scarring, and accumulation of goblet cells within the corneal epithelium
(Collinson et al., 2004). Heterozygous PAX6þ/ (e.g., PAX6þ/seyNeu)
mice with small eyes have been believed to be an excellent model to study
developmental eye defects and progressive corneal changes (Hill et al.,
1991). However, more recent studies elucidated abnormal differentiation
and higher proliferation of trans-amplifying cells in the basal layer of corneal
epithelium (Davis et al., 2003; Ramaesh et al., 2003) showing only a mild
LESC abnormality in the PAX6þ/ mice. Thus, new genetically changed
models with abnormal corneal epithelium showing a mosaic corneal pattern
have been developed to study LESC: (1) PAX6þ/ XLacZ mice, (2)
PAX6þ/Leca4 XLacZ mice (Mort et al., 2011), (3) PAX6þ/SeyNeu Gli3þ/XtJ
XLacZ double heterozygous mice (Kucerova et al., 2012) (Bargagna-Mohan
et al., 2012), (4) Pax6Sey, (5) Pax6SeyNeu, and (6) Pax6Coop mice (Theiler
et al., 1980; Hogan et al., 1986; Hill et al., 1991; Schmahl et al., 1993;
Lyon et al., 2000). All of these models develop severely reduced or no limbal
cells, and the corneal epithelium may be maintained by itself due to LESC
deficiency or a primary failure of centripetal cell movement (Zhang et al.,
2008). The absence of cell movement leads to increased goblet cell numbers
within the corneal epithelium, which is an indication for LSCD. These
models are helping ophthalmologic researchers to address fundamental ques-
tions about LESCs and their niche environment; however, the above-
discussed models are not best models for LSCD etiology that can mimic
typical LSCD.
5.2 LSCD due to Chemical and Mechanical Injury of Limbal/

Corneal Epithelial Cells
Chemical or mechanical injury of the ocular surface can lead to LSCD. Sur-
gical removal of the limbus and corneal epithelium (Figure 8), and NaOH-
induced chemical burn lead to reduced proliferation of corneal epithelial
cells and finally a dysfunction of the ocular surface, which can cause
LSCD. Although models exist for LSCD caused by chemical burn injury,
no specific animal models exist for SJSda typical LSCD disease. Table 3
summarizes several animal models with LSCD. Rabbits were used in all
studies because of globe size, easy handling, and similar limbal structure
compared to human. Mechanical injury can be achieved by surgical removal
of (1) the limbus (Huang and Tseng, 1991), (2) the entire cornea including
limbus (Koizumi et al., 2000), or (3) corneal epithelium by topical applica-
tion of 1-n-heptanol and by mechanical debridement, followed by 360 sur-
gical removal of a lamellar limbal ring (Ti et al., 2002, 2004).
Chemical injury to the cornea and limbus can be achieved using NaOH
or sulfur mustard (Li et al., 2011). Exposure to the chemical agent sulfur
Figure 8 Limbal stem cell deficiency induced by mechanical abrasion of corneal epithe-
lium. (A) Mechanical removal of half of corneal epithelium and limbus (arrows mark
border of injury). (B) Evaluation of area of injury using fluorescein staining (white dotted
line marks border of injury). (C) Normal transparent mouse cornea. (D) Complete
removal of corneal epithelium and limbus leads to epithelial defect and corneal
opacification.
mustard, leads to inflammation of the anterior segment and corneal erosions

within a few hours (Geeraets et al., 1977; Javadi et al., 2005, 2011). How-
ever, LSCD after SM exposure did not derive from primary damage of the
limbus, unlike other chemical burns following acid or alkali solutions. In all
chemical burns, corneal epithelial stem cells are destroyed leading to the
development of LSCD as well as neovascularization shortly after exposure.
A recent study demonstrated that slow cycling epithelial stem cells in the
limbus are more resistant to SM toxicity than more differentiated epithelial
cells in the central cornea (Kadar et al., 2011), questioning the use of SM to
create LSCD. We have also created our own chemical injury model using
NaoH in both mice and rabbit to study the LSCD and the limbal transplan-
tation (Figure 9(A) and (B)). Although LSCD can be induced in many ways,
currently there is no established animal model that replicates the disease pro-
cesses and the response to treatment in humans. The different distribution of
corneal stem cells in pig, calf, sheep, rabbit, rat, and human even further
hampers clinically relevant research (Lan et al., 2012).
76
Table 3 Animal models for limbal stem cell deficiency (LSCD) by mechanical and chemical injury
Ti et al. (2002), Ti et al.
Group Koizumi et al. (2000) (2004) Du et al. (2003) Luengo et al. (2007) Kadar et al. (2011)
Species Rabbit Rabbit Rabbit Rabbit Rabbit

LSCD Surgical removal of Chemical removal of Chemical removal of Chemical removal of LSCD by sulfur
model entire cornea corneal and limbal corneal and limbal corneal and Limbal mustard (Li et al.,
including limbus as epithelium by epithelium by 1N epithelium by 1N 2011)
well as all n-heptanol followed NaOH followed by NaOH
conjunctival tissue by surgical removal mechanical
within 5 mm of remaining limbal debridement
tissue
Vivek Singh et al.

(A)
(B)
Normal Mice LSCD MICE
Figure 9 (A) Showing the procedure of creating Rabbit LSCD model induced by chem-
ical burn using NAOH. (B) Showing mice LSCD model generated using alkali burn.
5.3 Animal Models for Corneal Wound Healing

According to the World Health Organization, corneal scarring due to infec-
tion remains a major cause of preventable blindness and has been studied
extensively over the last 70 years. Corneal wound healing is a complex
cascade involving cytokine-mediated interactions between epithelial cells,
stromal keratocytes, corneal nerves, lacrimal glands, tear film, and immune
cells. To improve treatments for patients blinded by corneal scarring,
different in vivo and ex vivo animal models are crucial to understand the
pathophysiology of corneal wound healing. Based on the type of animal
model, cell migration, proliferation, differentiation, barrier function, scaring,
reinnervation, and inflammation can be analyzed (Stepp et al., 2014).
In early 1980s Ilene Gipson and Kiorpes (1982) first time used circular
debridement to demonstrate that corneal epithelial wound healing occurs
via sheet movement. They characterized the cytoskeletal components that
are involved in mediating epithelial cell migration and showed differences
between dulled blade debridement and deeper keratectomy wounds. Circu-
lar debridement wounds are ideal to quantify reepithelialization, cell
proliferation rates, reinnervation, and innate immune responses. However,

the diameter of circular wounds can be varied using different size trephines
to demarcate the area within which tissue is removed. Because trephines are
very sharp and are designed to penetrate tissues (Chinnery et al., 2012; Stepp
et al., 2014), a majority of reports use a rotating burr (AlgerbrushII) to injure
the cornea in mice, rats, and rabbitsdincluding removal of LESCs (Mi et al.,
2008; Pal-Ghosh et al., 2011; Chinnery et al., 2012; Lan et al., 2012;
Ferrington et al., 2013). Thermal cautery also induces epithelial injury
creating limbal and corneal epithelial wound models. Dead corneal and lim-
bal epithelia are mechanically scraped using a surgical blade followed by
rinsing with 0.9% NaCl (Lan et al., 2012). These models are used to study
stem cell differentiation in the ocular surface. Clinically oriented scientists
preferably use manual superficial keratectomy (MSK) and photorefractive
keratectomy (PRK) to induce injury in rabbits, rats, and mice (Azar et al.,
1998; Kato et al., 2003; Mohan et al., 2008; Singh et al., 2011, 2013), albeit
the variable depth of the induced injury (Blanco-Mezquita et al., 2013).
Table 4 lists different species and the studies categorized by their major focus
in corneal research.
Basic research of corneal and anterior segment development using xen-
opus (Overton, 1965; Reeve and Wild, 1978) and zebrafish (Norrby, 2006)
has provided insight into how cornea is developed and how the cornea’s
antiangiogenic privilege is maintained (Norrby, 2006). Xenopus and zebra-
fish are widely used animal models to study developmental processes due to
the speed with which genetic studies can be performed, the rapid maturation
of the immature zebrafish to an adult, and relatively low costs. However,
zebrafish and xenopus are not suitable for corneal epithelial wound healing
studies. Chickens were used to study the corneal epithelial basement mem-
brane, collagen molecules in the corneal stroma, and glycosaminoglycan
biosynthesis and matrix assembly (Conrad, 1970; Trelstad, 1973; Conrad
et al., 1977; Svoboda et al., 1988). Recent studies investigated corneal inner-
vation and its regulation during development (Kubilus and Linsenmayer,
2010; Schwend et al., 2012).
For many decades, rabbits have been used to study corneal wound heal-
ing because their cornea is similar in size to the human cornea; this allows
clinicians to use the same instruments and methods of evaluation used in pa-
tients. Rabbit models were critical in studies that lead to the clinical use of
PRK and LASIK procedures (Munnerlyn et al., 1988). The most often used
strain is an albino one called New Zealand White. The rabbit’s cornea was
found to respond similarly to the human cornea in terms of development,
Table 4 Selected corneal studies and the animal model used
Animal Type of corneal study
Xenopus, zebrafish Development (Overton, 1965; Reeve and Wild, 1978)

Chicken Development, angiogenesis, wound healing, cell/organ culture (Bard and Hay, 1975; Norrby, 2006;
Schwend et al., 2012)
Rats Wound healing, cell/organ culture (Gipson and Keezer, 1982; Spurr-Michaud et al., 1988)
Rabbits Scarring, wound healing, dry eye, cell/organ culture, infections (Imanishi et al., 2000; Tandon et al., 2013)
Mice Angiogenesis (Dana and Streilein, 1996; Chung et al., 2009), scarring (Bargagna-Mohan et al., 2012; Blanco-
Mezquita et al., 2013), anterior segment development (Xie et al., 1999; Chikama et al., 2008),
differentiation (Argueso et al., 2009), inflammation (Stepp et al., 2002; Li et al., 2011; Chinnery et al.,
2012; Lee et al., 2012), wound healing (Gipson et al., 1993; Mayo et al., 2008; Mohan et al., 2008;
Pal-Ghosh et al., 2011; Yuan et al., 2013)
Dogs Dry eye (Morgan and Abrams, 1991; Acheampong et al., 1999), wound healing (Murphy et al., 2001;
Gosling et al., 2013)
Cats Scarring (Huxlin et al., 2013), wound healing (Huang et al., 1989; Jester et al., 1994; Petroll et al., 1999),
infections (Kaye and Choudhary, 2006; Gould, 2011)
79
the extent of scarring, and myofibroblast formation (Helena et al., 1998;

Imanishi et al., 2000; Wilson, 2002; Mohan et al., 2003; Santhiago et al.,
2011). Unfortunately, rabbits require higher cost for purchase and mainte-
nance than rats, and the availability of polyclonal antibodies against various
proteins as well as genetically diverse strains and transgenic animals is limited.
Dogs and cats were used in ophthalmology studies to improve treatment
for dogs with corneal defects and to test drugs before human application. For
example, dry eye disease was being treated with cyclosporine in dogs long
before being used in people (Stern et al., 1998). Studies done on cats with
ocular herpes simplex virus (HSV) has improved treatment in humans
(Kaye and Choudhary, 2006). Moreover, Nagy et al. (2007) described
that cats are similar to rabbits and respond similarly to humans in terms of
corneal scarring.
In conclusion, animal models of LSCD and corneal injury have proven
invaluable in investigating human corneal physiology and in developing
therapeutic strategies. However, investigations in these animal models also
encounter limitations, particularly difficulties in studying molecular mecha-
nisms, cellecell interactions, and the stem cells niche due to species variation
(Majo et al., 2008). New transgene, knockout, and gene-editing technolo-
gies will help in developing new animal model, which will continue to
contribute to the future of corneal research.
6. CLINICAL OUTCOME OF CELL-BASED OCULAR

SURFACE RECONSTRUCTIVE PROCEDURE
6.1 Overview
The era of cell-based therapy for ocular surface disorders began with
the discovery of LSCs in the palisades of Vogt. The therapeutic benefit
observed by Kenyon and Tseng (1989) provided evidence that limbal tissue
contained corneal epithelial stem cells. Therefore, definitive treatment of
LSCD involves LSC transplantation. Corneal transplantation alone does
not work in cases of total LSCD as the central corneal tissue does not contain
stem cells and thus the graft develops epithelial healing problems.
Management of LSCD is dependent on various factors that need to be
considered before planning an intervention. Factors to be considered are
the degree of LSCD, laterality of the disease, extent of conjunctival disease,
presence of conjunctival inflammation, nature of the ocular surface, the age
and general health of the patient. Stem cell-based therapy is best avoided in
the presence of active ocular surface inflammation such as in cases of acute

ocular surface burns and acute SJS as the survival of stem cells is questionable
in presence of ocular inflammation (Subramaniam et al., 2013). However,
prompt and adequate management in acute setting can improve the final
outcome and may even prevent the occurrence of stem cell deficiency.
The success of cell-based therapy is dependent on wetness of the ocular sur-
face and is best avoided in cases of severe dry eyes. Similarly, some amount of
ocular surface sensitivity (sensory innervations) is necessary for any recon-
structive procedure to work. In completely anesthetic ocular surface, these
surgeries should not be done. In patients with unilateral LSCD, the limbal
biopsy is taken from the unaffected eye (autologous). In cases with bilateral
stem cell deficiency, the source of limbal tissue is either a close relative or
cadaver (allogeneic). The disadvantage of allogeneic transplantation is pro-
longed need for systemic immunosuppression. Long-term immunosuppres-
sion adds to the cost of therapy and also needs monitoring for development
of associated adverse effects. In this section, we will be describing the current
techniques used for stem cell transplantation and their clinical outcomes.
6.2 Surgical Techniques

6.2.1 Cultivated limbal epithelial transplantation
CLET technique comprises of taking a limbal biopsy from the donor eye,
culturing the stem cells in a laboratory, and the transplantation of these
stem cells to the affected eye. This procedure minimizes the amount of
donor tissue needed for transplantation thereby decreasing the incidence
of iatrogenic LSCD in the donor eyes (Pellegrini et al., 1997; Tsai et al.,
2000; Schwab et al., 2000). The idea came from culture of human keratino-
cytes onto a feeder layer of nonproliferative 3T3 fibroblasts. The cells secrete
some growth factors that are conducive to growth of the keratinocytes. The
idea was extrapolated to LSCs by Pelligrini et al. (1997). In their first report,
two patients with alkali burn were successfully transplanted with 3T3 fibro-
blast feeder-cultured autologous corneal epithelium. Since then, there have
been different modifications of this technique, and various culture methods
have been described in literature (Pellegrini et al., 1997; Tsai et al., 2000;
Meller et al., 2002; Koizumi et al., 2002). The procedure can be performed
under local or general anesthesia depending on the patient’s age and prefer-
ence. After obtaining informed consent, limbal biopsy is taken from the
contralateral healthy eye or healthy area of the affected eye. A 2 2 mm
piece of limbal epithelium with 0.5 mm into clear corneal stromal tissue
from the pigmented limbus was dissected carefully taking care to avoid
buttonholing and deep dissection (Sangwan et al., 2005). If LSCD is bilat-

eral, then the limbal tissue can be harvested either from a live related donor
or a cadaver. A study from our institute has shown that fresh limbal tissue is
preferable to preserved cadaveric limbal tissue (Vemuganti et al., 2004).
Figure 10 shows unilateral LSCD due to chemical burn (acute and chronic
stage). The tissue is transported to laboratory in human corneal epithelium
medium where it is shredded into pieces and cultured. Human Corneal
Epithelial (HCE) medium comprises of 9.7 g/l modified eagle medium,
16.2 g/l Ham’s F12 serum, 10% (vol/vol) autologous serum, 0.01 mg/l
human recombinant epidermal growth factor, 0.25 mg/l insulin, 0.1 mg/l
cholera toxin and hydrocortisone. hAM is used as a carrier. hAM measuring
2.5 5 cm is deepithelialized using 0.25% trypsin and ethylenediamine
tetra-acetic acid (EDTA) solution for 30 min. Our laboratory uses a sub-
merged explant culture system without the use of any feeder cells (Fatima
et al., 2006). The growth is monitored and the culture is terminated
when the monolayer becomes confluent which usually takes 10e14 days
(Sangwan et al., 2011).
6.2.1.1 Surgical steps in CLET

Symblepharon (partial or complete adhesion of the palpebral conjunctiva of
the eyelid to the bulbar conjunctiva of the eyeball) if present is released
before applying the wire speculum. A conjunctival peritomy is performed
360 degrees and pannus is dissected off the cornea using blunt dissection.
The homeostasis is achieved with wet-field cautery. At this stage, the culti-
vated limbal epithelium on hAM is placed over the recipient cornea and
secured in place using either fibrin glue or sutures (Sangwan et al., 2011).
It is surgeon’s choice, glue is preferable to sutures but as glue is expensive,
sutures might be used.
(A) (B)
Figure 10 (A) Total limbal stem cell deficiency status post chemical injury; (B) Corneal
and conjunctival epithelial defect on fluorescein stain in a case of acute chemical
injury.
6.2.1.2 Results of CLET

Pellegrini et al. were the first group to publish results of cultured autolo-
gous corneal epithelial cell transplantation. In three large series, clinical
success was found to be 68% (Rama et al., 2010), 71% (Sangwan et al.,
2011), and 80% (Di Iorio et al., 2010). The eyes with previous AMT or
PK are more prone to failure and it is essential to control the ocular surface
inflammation before going ahead with stem cell transplant (Sangwan et al.,
2011).
In cases with recurrence of LSCD, the results of repeat CLET are encour-
aging in our own study (Basu et al., 2012a). Out of 50 eyes, 33 (66%) main-
tained an epithelialized and clinically stable corneal surface. 38 of the 50
recipient eyes (76%), experienced a 2-line improvement in visual acuity.
None of the donor eyes showed signs of LSCD. The minimum follow-up
in this study was 1 year (Basu et al., 2012a) (Figure 11). For patients with par-
tial LSCD, outcomes are similar and are independent of whether the limbal
biopsy is taken from the healthy or the affected eye (Vazirani et al., 2014). In
case of children younger than 15 years of age, the results of CLET are not as
good as adults. Of the 107 eyes, 50 eyes (46.7%) achieved completely epithe-
lialized, avascular, and stable ocular surface at 3.4 years of mean follow-up. At
the final visit, 58 eyes (54.2%) had improvement in visual acuity of 0.2 or
more log MAR units (Sejpal et al., 2013).
(A) (B)
(C)
Figure 11 Showing outcome of CLET. (A) Total limbal stem cell deficiency after chem-
ical injury; (B) Six months follow-up status post-CLET showing a stable ocular surface; (C)
Healthy donor site.
In our study, patients with unilateral LSCD underwent penetrating ker-

atoplasty either as a single-stage (along with CLET, n ¼ 12) or two-stage (at
least 6 weeks after CLET, n ¼ 35) procedure. The mean follow-up was
4.2 1.9 years. Corneal allograft survival rate at 1 year was significantly
greater in eyes undergoing two-stage (80% 6%; median survival, 4 years)
compared with single-stage (25 13%; median survival, 6 months;
P ¼ 0.0003) procedure. Therefore it can be concluded that the two-stage
approach of CLET followed by PK successfully restores ocular surface stabil-
ity and vision in eyes with chronic ocular burns. The combined single pro-
cedure approach is associated with poorer outcomes (Basu et al., 2011).
Tables 5 and 6 summarize the results of CLET using various techniques
by different authors.
6.2.2 Simple limbal epithelial transplantation

SLET is a single-stage procedure which does not require an expensive lab-
oratory setup unlike CLET and there is no danger of inducing iatrogenic
LSCD in the donor eye. It is convenient for patients and is cost-effective.
Here, ocular surface serves as a natural incubator for cell growth and multi-
plication which leads to in vivo expansion of LSCs (Sangwan et al., 2012).
6.2.2.1 Surgical technique in SLET

A 2 2 mm area of limbal tissue is excised and secured. The preparation of
the recipient bed is same as CLET. The donor tissue is divided into 10e15
small pieces and placed on the AM over the recipient bed, avoiding the
visual axis. These transplants are secured in place with fibrin glue and a
bandage contact lens is applied.
6.2.2.2 Results of SLET

We are the first to describe this novel technique and reported the results of
SLET in six patients with unilateral and total LSCD following ocular surface
burns. A completely stable ocular surface was seen in all recipient eyes at a
mean follow-up of 9.2 months. Visual acuity improved from less than
20/200 in all eyes before surgery to 20/60 or better in four (66.6%) eyes.
None of the donor eyes showed any signs of iatrogenic LSCD (Sangwan
et al., 2012). A representative image of post-SLET improvement is shown
in Figure 12 (Bhalekar et al., 2013) and Figure 13 showing the regeneration
of corneal epithelium as confirmed by corneal epithelium markers like
CK3þ12 post-SLET in LSCD patients. Guillermo et al. have described
a sandwich technique of SLET using cryopreserved AM wherein the
Table 5 Laboratory techniques and clinical outcomes reported in studies on autologous cultivated limbal epithelial transplantation
Follow-up (years)
Culture Air Culture Clinical 2-line visual
Author/Year technique Substrate lifting time (Days) Eyes success (%) gain (%) Mean Range
Feeder-free and xeno-free cell cultures
Marchini et al. (2012) Suspension 3T3 cells No 7e10 16 81.3 NA 1 1e4
Prabhasawat et al. (2012) Explant hAM Yes 14e21 12 66.7 66.7 2.2 0.5e4
Sangwan et al. (2011) Explant hAM No 10e14 200 71 60.5 2.8 1e7.6
Kolli et al. (2010) Explant hAM No 12e14 8 100 63 1.6 1e2.5
Di Girolamo et al. (2009) Explant CL No 10 2 100 50 0.9 0.7e1.1
Shimakazi et al. (2007) Explant hAM No 14.6 16 50 37.5 2.5 0.5e7.1
Nakamura et al. (2006) Explant hAM Yes 15e16 2 100 67 1.2 0.5e1.6
Feeder-free but not xeno-free cell cultures
Baradaran-Rafii et al. (2010) Explant hAM No 14 8 88 63 2.8 0.5e4
Pauklin et al. (2010) Explant hAM No 14 30 77 73 2.4 0.8e6
Shortt et al. (2008) Suspension hAM No 14e21 3 78 22 0.8 0.5e1.1
Sangwan et al. (2006) Explant hAM No 11e15 88 73 37 1.5 0.3e3.3
Sangwan et al. (2003) Explant hAM No 10e14 2 100 50 1 1
Grueterich et al. (2002) Explant hAM No 21 1 100 100 3.1 3.1
Tsai et al. (2000) Explant hAM No 14e21 3 100 50 2 0.3e10
Neither feeder-free nor xeno-free cell cultures
Di Iorio et al. (2010) Suspension Fibrin No NA 166 80 NA NA NA
Rama et al. (2010) Suspension Fibrin No 14e16 107 68 54 2.9 1e10
(Continued)
85
86
Table 5 Laboratory techniques and clinical outcomes reported in studies on autologous cultivated limbal epithelial transplantationdcont'd
Follow-up (years)
Culture Air Culture Clinical 2-line visual
Author/Year technique Substrate lifting time (Days) Eyes success (%) gain (%) Mean Range
Colabelli et al. (2010) Suspension Fibrin No 14e16 6 83 83 2 0.9e2.8
Nakamura et al. (2004b) Explant hAM Yes 23 1 100 100 1.6 1.6
Rama et al. (2001) Suspension Fibrin No 14e16 18 74 33 1.5 1e2.2
Schwab et al. (2000) Suspension hAM Yes 21e28 10 60 36 1.1 0.5e1.6
Schwab (1999) Suspension hAM No 28e35 17 76 16 0.9 0.2e2
Pellegrini et al. (1997) Suspension 3T3 No 16e19 2 100 50 NA NA
hAM, human amniotic membrane; CL, contact lens.
Vivek Singh et al.

Table 6 Laboratory techniques and clinical outcomes reported in studies on allogeneic cultivated limbal epithelial transplantation
Feeder Air Follow-Up
Author/Year Donor Method Cell Serum lifting Xeno-Free Success (months)
Basu et al. (2012a) LR Explant No AS No Yes 71.4% (20/28) 58 (2e114)

Prabhasawat et al. (2012) Cadaveric Explant Yes AS Yes Yes 85.7% (6/7) 28 (4e47)
Pauklin et al. (2010) 4 LR, 10 cadaveric Explant No AS No No 50% (7/14) 28.5 (9e72)
Meller et al. (2012) HLA-donor Explant No AS No Yes 100% (1/1) 31
Shortt et al. (2008) Cadaveric Explant No FCS No No 100% (3/3) 15 (12e18)
Ang et al. (2007) Cadaveric Suspension Yes FCS Yes No 100% (1/1) 48
Shimazaki et al. (2007) 7 LR, 7 cadaveric Explant No AS Yes Yes 50% (10/20) 29 (6e85)
Nakamura et al. (2006) Cadaveric Suspension No AS Yes No 100% (7/7) 14.6 (6e26)
Daya et al. (2005) 1 LR, 9 cadaveric Suspension Yes FCS No No 70% (7/10) 28 (12e50)
Sangwan et al. (2005) LR Explant No AS No Yes 100% (4/4) 15.3 (7e24)
Koizumi et al. (2001b) Cadaveric Explant Yes FCS Yes No 92% (12/13) 6
Koizumi et al. (2001a) Cadaveric Explant Yes FCS Yes No 100% (3/3) 31
Schwab et al. (2000) LR Suspension No AS No No 100% (4/4) 10.5 (2e24)
Schwab (1999) LR Suspension Yes FCS No No 50% (1/2) 13 (16e19)
LR, Live related; AS, Autologous serum; FCS, Fetal calf serum.
87
(A) (B) (C)
(D) (E) (F)
Figure 12 Showing outcome of cultivated limbal epithelial transplantation (CLET). (A)

Clinical photograph of the left eye showing extensive corneal limbal and conjunctival
epithelial defect with corneal stromal haze, 3 days following lime injury. (B) Same eye
6 months after amniotic membrane grafting, showing conjunctivalized ocular surface,
suggestive of Limbal stem cell deficiency (LSCD). (C) Same eye 1 year after autologous
CLET, showing a vascular and conjunctivalized ocular surface suggestive of recurrence
of LSCD following CLET. (D) Postoperative photograph of the left eye 3 weeks after
autologous simple limbal epithelial transplantation (SLET), showing an epithelized
corneal surface with few limbal transplants in place (arrow). (E) 1 year post-SLET the
left eye shows a stable, epithelized and avascular corneal surface with significant
improvement in corneal clarity. (F) Right eye showing healthy donor sites with no
ocular surface deficits. Generated with permission from Bhalekar et al. (2013).
Figure 13 Showing expression of CK3þ12 (corneal epithelium) in cornea after SLET by

using immunofluorescence.
transplants are placed between the two layers of AMG. Their outcomes
were comparable with the outcomes of the original technique in a case series
of four cases (Amescua et al., 2014).
Our recent unpublished data where KaplaneMeier survival analysis from
100 patients undergoing SLET showed an overall success rate of 73% at
1 year and beyond. Successful transplantation was maintained in 72% of
children and 74% of adults and in 91% and 71% of cases of focal and total
LSCD, respectively. Combining SLET with keratoplasty shows significant
high-risk factors of failure. Best Corrected Visual Acuity (BCVA) improved
from 20/200 or worse in all eyes to 20/60 or better in 60% of successful
cases. No adverse effects were noted in the donor fellow eyes 1 h after limbal
extraction (part of result is accepted in ARVO, Basu et al., 2015). These facts
show the obvious opportunity for cell-based therapy for limbal deficiency.
6.2.3 Cultivated oral mucosal epithelial transplantation

In cases with bilateral LSCD, there is no autologous source of LSCs and
either a living or a cadaveric allogeneic donor is required (Fernandes
et al., 2004). However, this necessitates long-term immunosuppression
which predisposes the patients to systemic side effects of these drugs and
thus adds to the cost. An alternative to allogeneic limbal grafting is transplan-
tation of autologous epithelium from nonocular sources. The possibility of
using oral mucosal epithelium was considered because of the phenotypic
similarities between the two epithelial surfaces (Madhira et al., 2008;
Krishnan et al., 2010). The cultivation of oral mucosal epithelial cells
required various animal-derived and xenobiotic materials (Nakamura
et al., 2004a; Nakamura et al., 2011, 2004b; Ang et al., 2005; Satake
et al., 2011). However, recently a technique of xeno-free cultivation similar
to CLET has been described (Gaddipati et al., 2014a).
6.2.3.1 Surgical technique in COMET

All patients undergo an oral examination by a physician to rule out any con-
traindications to a mucosal biopsy. The patients are advised 5% povidone-
iodine mouthwash twice daily for three consecutive days prior to the biopsy.
An oral mucosal biopsy of 3 3 mm is harvested from the inner surface of
the patient’s lower lip under local anesthesia (2% xylocaine submucosal infil-
tration). The biopsied area is left bare and the patient is advised to continue
the mouthwash for 1 week. The biopsy is transferred to laboratory in HCE
medium. The patient’s oral mucosal tissue was divided into small pieces after
separation from the underlying connective and minor salivary glands. The
tissue bits are explanted over the deepithelized hAM epithelial side up. The
culture was transplanted when a monolayer of the cells growing from the
explants became confluent, usually within 15e19 days. The preparation
of the recipient bed is same as described before. The cultivated mucosal
epithelial cells are secured in place with fibrin glue onto the recipient bed
(Gaddipati et al., 2014b).
6.2.3.2 Results of COMET

The indications and results of cultivated oral mucosal epithelial transplanta-
tion (COMET) are highly variable in different studies ranging from 28.5%
to 100% with mean follow-up durations ranging from 12 months to
55 months (Nishida et al., 2004a; Nakamura et al., 2011, 2004a; Ang
et al., 2005; Satake et al., 2011; Burillon et al., 2012). All studies reported
appearance of peripheral superficial corneal vascularization after COMET.
In a recent study by Gaddipati et al., a xeno-free explants culture technique
has been described. Their study included 19 eyes with a mean follow-up of
22.3 months. A stable ocular surface was seen in 37% eyes at the end of 1 year
and vision did not improve in 63% eyes. They concluded that clinical out-
comes of autologous cultivated oral mucosal epithelial transplantation in
eyes with ocular surface burns were poor and the transplanted cells main-
tained the oral phenotype on the corneal surface (Gaddipati et al., 2014b).
Table 7 summarizes the results of COMET by various different groups.
Table 7 Laboratory techniques and clinical outcomes reported in studies on

cultured oral mucosal epithelial transplantation
Stable Mean
Culture ocular Improved follow-up
Author/Year technique Substrate Eyes surface (%) VA (%) (months)
Hirayama et al. Explant Fibrin 16 62.5% 100 12

(2012) hAM 16 43.8% 93.75
Satake et al. (2011) Explant hAM 40 57.5 59 25.4
Takeda et al. Explant hAM 3 66.7 66.7 30
(2011)
Nakamura et al. Explant hAM 19 100 95 55
(2011)
Nishida et al. Explant hAM 4 100 100 14
(2004a)
Inatomi et al. Explant hAM 15 66.7 66.7 20
(2006a,b)
hAM, human amniotic membrane; VA, Visual acuity.
7. FUTURE PATH AND CONCLUSION

The development of new cell-based therapeutic modalities for
various ocular diseases has become necessary while considering the latest
technological developments in the past few years. It has become essential
to combine cutting-edge translational research based upon liberal original
ideas obtained from clinical experience with state-of-the-art basic science
and technology. The current review of literature summarizes various
aspects of cell-based therapies being attempted in different parts of the
world and their likely implications in ocular regeneration. It is clear that
they offer a beam of hope for various eye diseases where existing non-
cell-based therapies either have a poor success rates or face failure due to
recurring graft rejection.
A related issue is the absence of a system to categorize the extent of dam-
age to the limbal niche prior to transplantation. It can be imagined that if
the niche is completely damaged, then the transplanted stem cells might
just remain on the corneal surface and with time undergo differentiation
resulting in the disease recurrence because the cornea is unequipped to sup-
port long-term stem cell survival. While research is presently underway to
answer some of these questions, an alternative approach is also being devel-
oped. This is to reconstruct the niche in vitro so as to give the LSCs a home
to reside in the following transplantation. To do this, a good understanding
of the limbal niche is needed since it is a unique and complex microenvi-
ronment that regulates the proliferation, self-renewal, and differentiation
of the resident stem cell population. In conclusion, an optimal alternative
to the hAM for reconstruction of the ocular surface should be transparent,
biocompatible, exhibit sufficient mechanical strength, remain cost-effec-
tive, and support the growth of LSCs.
The recent discovery of multipotent stem cells in the corneal stroma has
opened up the possibility of developing a cell-based approach to treating
corneal scars as an alternative to keratoplasty. Stromal MSCs may have a
role not only in remodeling but also in preventing inflammation, scarring,
and immune rejection of transplanted corneal tissue (Basu et al., 2012b,
2014). Basu et al. (2014) used therapeutic model in which human limbal
biopsy-derived stromal cells embedded in fibrin gel were applied to the sur-
face of a healing murine debridement wound and report that human stromal
cells derived from limbal biopsies can not only differentiate to functional
keratocytes in vitro but also induce regeneration of damaged stromal tissue
in vivo, resulting in a matrix indistinguishable from that of native cornea. At

our Institute we have already started a pilot study (approved by our institu-
tional review board), using autologous ex vivo cultivated LSC transplanta-
tion for treatment of superficial corneal stromal scars in human patients.
We foresee the ability of a clinician to isolate limbal stromal cells from a
healthy eye, expand the cells, and, after surgically removing the scar tissue
from the wounded eye, apply the patient’s own LSCs to regenerate healthy,
transparent tissue.
Currently, clinicians as well as basic scientists in this field strongly
believe that autologous limbal transplantation is the treatment of choice
in unilateral total or partial LSCD and there is no role of allogeneic proce-
dures in this condition. We additionally advocate SLET over CLET as the
preferred surgical technique not only because it is successful but also
because it offers many other advantages such as being single-staged, more
affordable, and technically feasible in a resource-limited setting. However,
the best therapy for bilateral LSCD is still elusive. In spite of many practical
hurdles, there is emerging need to unravel the secret therapeutic potential
of different stem cells through a combination of clinical and basic science
studies.
ACKNOWLEDGMENTS
The authors thank Dr MacNeil S (Kroto Research Institute, University of Sheffield, Shef-
field, United Kingdom) for her input and editing biomaterial section of the review. We
would also like to thank Chandan Teja and Abhinav Reddy K. (Research fellow at LV Prasad
Eye Institute, Hyderabad) for helping in preparation and designing of some images. Support
in part by Champalimaud foundation and Research Grant by SERB (Vivek Singh: SERB/
LS-599/2013).
REFERENCES
Acheampong, A.A., Shackleton, M., Tang-Liu, D.D., Ding, S., Stern, M.E., Decker, R.,
1999. Distribution of cyclosporin A in ocular tissues after topical administration to albino
rabbits and beagle dogs. Curr. Eye Res. 18, 91e103.
Ahmad, S., Stewart, R., Yung, S., Kolli, S., Armstrong, L., Stojkovic, M., Figueiredo, F.,
Lako, M., 2007. Differentiation of human embryonic stem cells into corneal epithe-
lial-like cells by in vitro replication of the corneal epithelial stem cell niche. Stem Cells
1145e1155.
Albert, R., Vereb, Z., Csomos, K., Moe, M.C., Johnsen, E.O., Olstad, O.K., Nicolaissen, B.,
Rajnav€ olgyi, E., Fés€
us, L., Berta, A., Petrovski, G., 2012. Cultivation and characteriza-
tion of cornea limbal epithelial stem cells on lens capsule in animal material-free medium.
PLoS One 7, 47187.
Amescua, G., Atallah, M., Nikpoor, N., Galor, A., Perez, V.L., 2014. Modified simple limbal
epithelial transplantation using cryopreserved amniotic membrane for unilateral limbal
stem cell deficiency. Am. J. Ophthalmol. 158 (3), 469e475.
Ang, L.P., Sotozono, C., Koizumi, N., Suzuki, T., Inatomi, T., Kinoshita, S., 2007. A com-
parison between cultivated and conventional limbal stem cell transplantation for Stevens-
Johnson syndrome. Am. J. Ophthalmol. 143 (1), 178e180.
Ang, L.P., Tan, D.T., Seah, C.J., Beuerman, R.W., 2005. The use of human serum in sup-
porting the in vitro and in vivo proliferation of human conjunctival epithelial cells. Br. J.
Ophthalmol. 89, 748e752.
Angunawela, R.I., Mehta, J.S., Daniels, J.T., 2013. Ex-vivo ocular surface stem cell therapies:
current techniques, applications, hurdles and future directions. Expert Rev. Mol. Med.
15, e4.
Argueso, P., Guzman-Aranguez, A., Mantelli, F., Cao, Z., Ricciuto, J., Panjwani, N., 2009.
Association of cell surface mucins with galectin-3 contributes to the ocular surface
epithelial barrier. J. Biol. Chem. 284, 23037e23045.
Azar, D.T., Pluznik, D., Jain, S., Khoury, J.M., 1998. Gelatinase B and A expression after
laser in situ keratomileusis and photorefractive keratectomy. Arch. Ophthalmol. 116,
1206e1208.
Baradaran-Rafii, A., Ebrahimi, M., Kanavi, M.R., Taghi-Abadi, E., Aghdami, N.,
Eslani, M., Bakhtiari, P., Einollahi, B., Baharvand, H., Javadi, M.A., 2010. Midterm out-
comes of autologous cultivated limbal stem cell transplantation with or without pene-
trating keratoplasty. Cornea 29, 502e509.
Bard, J.B., Hay, E.D., 1975. The behavior of fibroblasts from the developing avian cornea.
Morphology and movement in situ and in vitro. J. Cell Biol. 67, 400e418.
Bargagna-Mohan, P., Paranthan, R.R., Hamza, A., Zhan, C.G., Lee, D.M., Kim, K.B.,
Lau, D.L., Srinivasan, C., Nakayama, K., Nakayama, K.I., Herrmann, H., Mohan, R.,
2012. Corneal antifibrotic switch identified in genetic and pharmacological deficiency
of vimentin. J. Biol. Chem. 287, 989e1006.
Basu, S., Ali, H., Sangwan, V.S., 2012a. Clinical outcomes of repeat autologous cultivated
limbal epithelial transplantation for ocular surface burns. Am. J. Ophthalmol. 153,
643e650.
Basu, S., Fernandez, M.M., Das, S., Gaddipati, S., Vemuganti, G.K., Sangwan, V.S., 2012b.
Clinical outcomes of xeno-free allogeneic cultivated limbal epithelial transplantation for
bilateral limbal stem cell deficiency. Br. J. Ophthalmol. 96 (12), 1504e1509.
Basu, S., Mohamed, A., Chaurasia, S., Sejpal, K., Vemuganti, G.K., Sangwan, V.S., 2011.
Clinical outcomes of penetrating keratoplasty after autologous cultivated limbal epithelial
transplantation for ocular surface burns. Am. J. Ophthalmol. 152 (6), 917e924.
Basu, S., Hertsenberg, A.J., Funderburgh, M.L., Burrow, M.K., Mann, M.M., Du, Y.,
Lathrop, K.L., Syed-Picard, F.N., Adams, S.M., Birk, D.E., Funderburgh, J.L., 2014.
Human limbal biopsy-derived stromal stem cells prevent corneal scarring. Sci. Transl.
Med. 6 (266), 266ra172.
Basu, S., Surekha, S., Singh, V., Sangwan, V.S., 2015. Simple Limbal Epithelial Transplan-
tation (SLET) for Long-term Corneal Surface Regeneration in Unilateral Ocular Burns.
IOVS 2015 ARVO Abstract 2072.
Bhalekar, S., Basu, S., Lal, I., Sangwan, V.S., 2013. Successful autologous simple limbal
epithelial transplantation (SLET) in previously failed paediatric limbal transplantation
for ocular surface burns. BMJ Case Rep. 2013.
Blanco-Mezquita, J.T., Hutcheon, A.E., Zieske, J.D., 2013. Role of thrombospondin-1 in
repair of penetrating corneal wounds. Invest. Ophthalmol. Vis. Sci. 54, 6262e6268.
Brown, K.D., Low, S., Mariappan, I., Abberton, K.M., Short, R., Zhang, H., Maddileti, S.,
Sangwan, V., Steele, D., Daniell, M., 2014. Plasma polymer-coated contact lenses for the
culture and transfer of corneal epithelial cells in the treatment of limbal stem cell
deficiency. Tissue Eng. Part A 20 (3e4), 646e655.
Burillon, C., Huot, L., Justin, V., Nataf, S., Chapuis, F., Decullier, E., Damour, O., 2012.
Cultured autologous oral mucosal epithelial cell-sheet (CAOMECS) transplantation
for the treatment of corneal limbal epithelial stem cell deficiency. Invest. Ophthalmol.
Vis. Sci. 53, 1325e1331.
Burman, S., Sangwan, V., 2008. Cultivated limbal stem cell transplantation for ocular surface
reconstruction. Clin. Ophthalmol. 2, 489e502.
Bray, L.J., Suzuki, S., Harkin, D.G., Chirila, T.V., 2013. Incorporation of exogenous RGD
peptide and inter-species blending as strategies for enhancing human corneal limbal epithe-
lial cell growth on Bombyx mori silk fibroin membranes. J. Funct. Biomater. 4, 74e88.
Caplan, A.I., 1991. Mesenchymal stem cells. J. Orthop. Res. 9, 641e650.
Chan, A.A., Hertsenberg, A.J., Funderburgh, M.L., Mann, M.M., Du, Y., Davoli, K.A.,
Mich-Basso, J.D., Yang, L., Funderburgh, J.L., 2013. Differentiation of human embry-
onic stem cells into cells with corneal keratocyte phenotype. PLoS One 8 (2), e56831.
Chien, Y., Liao, Y.W., Liu, D.M., Lin, H.L., Chen, S.J., Chen, H.L., Peng, C.H.,
Liang, C.M., Mou, C.Y., Chiou, S.H., 2012. Corneal repair by human corneal kerato-
cyte-reprogrammed iPSCs and amphiphatic carboxymethyl-hexanoyl chitosan
hydrogel. Biomaterials 33 (32), 8003e8016.
Chikama, T., Liu, C.Y., Meij, J.T., Hayashi, Y., Wang, I.J., Yang, L., Nishida, T.,
Kao, W.W., 2008. Excess FGF-7 in corneal epithelium causes corneal intraepithelial
neoplasia in young mice and epithelium hyperplasia in adult mice. Am. J. Pathol. 172,
638e649.
Chinnery, H.R., Mclenachan, S., Binz, N., Sun, Y., Forrester, J.V., Degli-Esposti, M.A.,
Pearlman, E., Mcmenamin, P.G., 2012. TLR9 ligand CpG-ODN applied to the injured
mouse cornea elicits retinal inflammation. Am. J. Pathol. 180, 209e220.
Chirila, T., Barnard, Z., Zainuddin, Harkin, D.G., Schwab, I.R., Hirst, L., 2008. Bombyx
mori silk fibroin membranes as potential substrata for epithelial constructs used in the
management of ocular surface disorders. Tissue Eng. Part A 14 (7), 1203e1211.
Chung, E.S., Saban, D.R., Chauhan, S.K., Dana, R., 2009. Regulation of blood vessel versus
lymphatic vessel growth in the cornea. Invest. Ophthalmol. Vis. Sci. 50, 1613e1618.
Colabelli Gisoldi, R.A., Pocobelli, A., Villani, C.M., Amato, D., Pellegrini, G., 2010. Eval-
uation of molecular markers in corneal regeneration by means of autologous cultures of
limbal cells and keratoplasty. Cornea 29 (7), 715e722.
Collinson, J.M., Chanas, S.A., Hill, R.E., West, J.D., 2004. Corneal development, limbal
stem cell function, and corneal epithelial cell migration in the Pax6(þ/) mouse. Invest.
Ophthalmol. Vis. Sci. 45, 1101e1108.
Conrad, G.W., 1970. Collagen and mucopolysaccharide biosynthesis in mass cultures and
clones of chick corneal fibroblasts in vitro. Dev. Biol. 21, 611e635.
Conrad, G.W., Hamilton, C., Haynes, E., 1977. Differences in glycosaminoglycans synthesized
by fibroblast-like cells from chick cornea, heart, and skin. J. Biol. Chem. 252, 6861e6870.
Cotsarelis, G., Cheng, S.Z., Dong, G., Sun, T.T., Lavker, R.M., 1989. Existence of slow-
cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate:
implications on epithelial stem cells. Cell 57, 201e209.
Dana, M.R., Streilein, J.W., 1996. Loss and restoration of immune privilege in eyes with
corneal neovascularization. Invest. Ophthalmol. Vis. Sci. 37, 2485e2494.
Davanger, M., Evensen, A., 1971. Role of the pericorneal papillary structure in renewal of
corneal epithelium. Nature 229, 560e561.
Davis, J., Duncan, M.K., Robison Jr., W.G., Piatigorsky, J., 2003. Requirement for Pax6 in
corneal morphogenesis: a role in adhesion. J. Cell Sci. 116, 2157e2167.
Daya, S.M., Watson, A., Sharpe, J.R., Giledi, O., Rowe, A., Martin, R., James, S.E., 2005.
Outcomes and DNA analysis of ex vivo expanded stem cell allograft for ocular surface
reconstruction. Ophthalmology 112 (3), 470e477.
Deshpande, P., Notara, M., Bullett, N., Daniels, J.T., Haddow, D.B., MacNeil, S., 2009.
Development of a surface-modified contact lens for the transfer of cultured limbal epithe-
lial cells to the cornea for ocular surface diseases. Tissue Eng. Part A 15, 2889e2902.
Deshpande, P., Ramachandran, C., Sefat, F., Mariappan, I., Johnson, C., McKean, R.,
Hannah, M., Sangwan, V.S., Claeyssens, F., Ryan, A.J., MacNeil, S., 2013. Simplifying
corneal surface regeneration using a biodegradable synthetic membrane and limbal tissue
explants. Biomaterials 34 (21), 5088e5106.
Dhamodaran, K., Subramani, M., Ponnalagu, M., Shetty, R., Das, D., 2014. Ocular stem
cells: a status update! Stem Cell Res. Ther. 5 (2), 56.
Di Girolamo, N., Bosch, M., Zamora, K., Coroneo, M.T., Wakefield, D., Watson, S.L.,
2009. A contact lens-based technique for expansion and transplantation of autologous
epithelial progenitors for ocular surface reconstruction. Transplantation 87, 1571e1578.
Di Iorio, E., Ferrari, S., Fasolo, A., Bohm, E., Ponzin, D., Barbaro, V., 2010. Techniques for
culture and assessment of limbal stem cell grafts. Ocul. Surf. 8, 146e153.
Dravida, S., Gaddipati, S., Griffith, M., Merrett, K., Lakshmi Madhira, S., Sangwan, V.S.,
Vemuganti, G.K., 2008. A biomimetic scaffold for culturing limbal stem cells: a prom-
ising alternative for clinical transplantation. J. Tissue Eng. Regen. Med. 2, 263e271.
Dua, H.S., Shanmuganathan, V.A., Powell-Richards, A.O., Tighe, P.J., Joseph, A., 2005.
Limbal epithelial crypts: a novel anatomical structure and a putative limbal stem cell
niche. Br. J. Ophthalmol. 89 (5), 529e532.
Dunlap, W.A., Purnell, W.D., McPherson Jr., S.D., 1976. Laboratory and clinical evaluation of
a new synthetic absorbable suture for ophthalmic surgery. Adv. Ophthalmol. 33, 49e61.
Du, Y., Chen, J., Funderburgh, J.L., Zhu, X., Li, L., 2003. Functional reconstruction of rab-
bit corneal epithelium by human limbal cells cultured on amniotic membrane. Mol. Vis.
9, 635e643.
Du, Y., Funderburgh, M.L., Mann, M.M., SundarRaj, N., Funderburgh, J.L., 2005. Multi-
potent stem cells in human corneal stroma. Stem Cells 23, 1266e1275.
Eppley, B.L., Li, M., 2003. Long-spanning resorbable plates in cranial vault reconstruction.
J. Craniofac. Surg. 14 (1), 89e91.
Fatima, A., Sangwan, V.S., Iftekhar, G., Reddy, P., Matalia, H., Balasubramanian, D.,
Vemuganti, G.K., 2006. Technique of cultivating limbal derived corneal epithelium
on human amniotic membrane for clinical transplantation. J. Postgrad. Med. 52 (4),
257e261.
Feng, Y., Borrelli, M., Reichl, S., Schrader, S., Geerling, G., 2014. Review of alternative
carrier materials for ocular surface reconstruction. Curr. Eye Res. 39 (6), 541e552.
Fernandes, M., Sangwan, V.S., Rao, S.K., Basti, S., Sridhar, M.S., Bansal, A.K., Dua, H.S.,
2004. Limbal stem cell transplantation. Indian J. Ophthalmol. 52, 5e22.
Fernandes, M., Sridhar, M.S., Sangwan, V.S., et al., 2005. Amniotic membrane transplanta-
tion for ocular surface reconstruction. Cornea 24, 643e653.
Ferrington, D.A., Roehrich, H., Chang, A.A., Huang, C.W., Maldonado, M., Bratten, W.,
Rageh, A.A., Heuss, N.D., Gregerson, D.S., Nelson, E.F., Yuan, C., 2013. Corneal
wound healing is compromised by immunoproteasome deficiency. PLoS One 8, 54347.
Friedenstein, A.J., Petrakova, K.V., Kurolesova, A.I., Frolova, G.P., 1968. Heterotopic of
bone marrow. Analysis of precursor cells for osteogenic and hematopoietic tissues. Trans-
plantation 6, 230e247.
Forni, M.F., Loureiro, R.R., Cristovam, P.C., Bonatti, J.A., Sogayar, M.C., Gomes, J.A.,
2013. Comparison between different biomaterial scaffolds for limbal-derived stem cells
growth and enrichment. Curr. Eye Res. 38 (1), 27e34.
Fagerholm, P., Lagali, N.S., Ong, J.A., Merrett, K., Jackson, W.B., Polarek, J.W.,
Suuronen, E.J., Liu, Y., Brunette, I., Griffith, M., 2014. Stable corneal regeneration
four years after implantation of a cell-free recombinant human collagen scaffold. Bioma-
terials 35 (8), 2420e2427.
Gil, E.S., Mandal, B.B., Park, S.H., Marchant, J.K., Omenetto, F.G., Kaplan, D.L., 2010.
Helicoidal multi-lamellar features of RGD-functionalized silk biomaterials for corneal
tissue engineering. Biomaterials 31 (34), 8953e8963.
Gaddipati, S., Basu, S., Vemuganti, G.K., Ramappa, M., Maddileti, S., Sangwan, V.S.,
2014a. Xeno-free autologous cultivated oral mucosal epithelial transplantation for bilat-
eral ocular surface burns: clinical outcomes and immunohistochemical analysis. Curr.
Indian Eye Res. 1 (1), 17e24.
Gaddipati, S., Muralidhar, R., Sangwan, V.S., Mariappan, I., Vemuganti, G.K.,
Balasubramanian, D., 2014b. Oral epithelial cells transplanted on to corneal surface
tend to adapt to the ocular phenotype. Indian J. Ophthalmol. 62, 644e648.
Garfias, Y., Nieves-Hernandez, J., Garcia-Mejia, M., Estrada-Reyes, C., Jimenez-
Martinez, M.C., 2012. Stem cells isolated from the human stromal limbus possess immu-
nosuppressant properties. Mol. Vis. 18, 2087e2095.
Geeraets, W.J., Abedi, S., Blanke, R.V., 1977. Acute corneal injury by mustard gas. South.
Med. J. 70, 348e350.
Gipson, I.K., Keezer, L., 1982. Effects of cytochalasins and colchicine on the ultrastructure of
migrating corneal epithelium. Invest. Ophthalmol. Vis. Sci. 22, 643e650.
Gipson, I.K., Kiorpes, T.C., 1982. Epithelial sheet movement: protein and glycoprotein
synthesis. Dev. Biol. 92, 259e262.
Gipson, I.K., Spurr-Michaud, S., Tisdale, A., Elwell, J., Stepp, M.A., 1993. Redistribution of
the hemidesmosome components alpha 6 beta 4 integrin and bullous pemphigoid anti-
gens during epithelial wound healing. Exp. Cell Res. 207, 86e98.
Gosling, A.A., Labelle, A.L., Breaux, C.B., 2013. Management of spontaneous chronic
corneal epithelial defects (SCCEDs) in dogs with diamond burr debridement and place-
ment of a bandage contact lens. Vet. Ophthalmol. 16, 83e88.
Gould, D., 2011. Feline herpesvirus-1: ocular manifestations, diagnosis and treatment
options. J. Feline Med. Surg. 13, 333e346.
Grueterich, M., Espana, E.M., Touhami, A., Ti, S.-E., Tseng, S.C.G., 2002. Phenotypic
study of a case with successful transplantation of ex vivo expanded human limbal epithe-
lium for unilateral total limbal stem cell deficiency. Ophthalmology 109, 1547e1552.
Hackett, J.M., Lagali, N., Merrett, K., Edelhauser, H., Sun, Y., Gan, L., Griffith, M.,
Fagerholm, P., 2011. Biosynthetic corneal implants for replacement of pathologic
corneal tissue: performance in a controlled rabbit alkali burn model. Invest. Ophthalmol.
Vis. Sci. 52 (2), 651e657.
Hanson, C., Hardarson, T., Ellerstr€ om, C., Nordberg, M., Caisander, G., Rao, M.,
Hyllner, J., Stenevi, U., 2013. Transplantation of human embryonic stem cells onto a
partially wounded human cornea in vitro. Acta Ophthalmol. 91 (2), 127e130.
Hao, Y., Ma, D.H., Hwang, D.G., Kim, W.S., Zhang, F., 2000. Identification of antiangio-
genic and antiinflammatory proteins in human amniotic membrane. Cornea 19, 348e352.
Harkin, D.G., Foyn, L., Bray, L.J., Sutherland, A.J., Li, F.J., Cronin, B.G., 2015. Can mesen-
chymal stromal cells differentiate into corneal cells? A systematic review of published
data. Stem Cells 33 (3), 785e791.
Hashmani, K., Branch, M.J., Sidney, L.E., Dhillon, P.S., Verma, M., McIntosh, O.D.,
Hopkinson, A., Dua, H.S., 2013. Characterization of corneal stromal stem cells with
the potential for epithelial transdifferentiation. Stem Cell Res. Ther. 4, 75.
Hayashi, R., Ishikawa, Y., Ito, M., Kageyama, T., Takashiba, K., Fujioka, T., Tsujikawa, M.,
Miyoshi, H., Yamato, M., Nakamura, Y., Nishida, K., 2012. Generation of corneal
epithelial cells from induced pluripotent stem cells derived from human dermal fibroblast
and corneal limbal epithelium. PLoS One 7 (9), e45435.
Hayashi, R., Yamato, M., Sugiyama, H., Sumide, T., Yang, J., Okano, T., et al., 2007.
N-Cadherin is expressed by putative stem/progenitor cells and melanocytes in the hu-
man limbal epithelial stem cell niche. Stem Cells 25, 289e296.
Haynesworth, S.E., Baber, M.A., Caplan, A.I., 1992. Cell surface antigens on human
marrow-derived mesenchymal cells are detected by monoclonal antibodies. Bone 13,
69e80.
He, Q., Wan, C., Li, G., 2007. Concise review: multipotent mesenchymal stromal cells in
blood. Stem Cells 25, 69e77.
Helena, M.C., Baerveldt, F., Kim, W.J., Wilson, S.E., 1998. Keratocyte apoptosis after
corneal surgery. Invest. Ophthalmol. Vis. Sci. 39, 276e283.
Hida, N., Nishiyama, N., Miyoshi, S., Kira, S., Segawa, K., Uyama, T., Mori, T., Miyado, K.,
Ikegami, Y., Cui, C., Kiyono, T., Kyo, S., Shimizu, T., Okano, T., Sakamoto, M.,
Ogawa, S., Umezawa, A., 2008. Novel cardiac precursor-like cells from human menstrual
blood-derived mesenchymal cells. Stem Cells 26, 1695e1704.
Higa, K., Takeshima, N., Moro, F., Kawakita, T., Kawashima, M., Demura, M., Shimazaki, J.,
Asakura, T., Tsubota, K., Shimmura, S., 2011. Porous silk fibroin film as a transparent car-
rier for cultivated corneal epithelial sheets. J. Biomater. Sci. Polym. 22, 2261e2276.
Hill, R.E., Favor, J., Hogan, B.L., Ton, C.C., Saunders, G.F., Hanson, I.M., Prosser, J.,
Jordan, T., Hastie, N.D., Van Heyningen, V., 1991. Mouse small eye results from mu-
tations in a paired-like homeobox-containing gene. Nature 354, 522e525.
Hirayama, M., Satake, Y., Higa, K., Yamaguchi, T., Shimazaki, J., 2012. Transplantation of
cultivated oral mucosal epithelium prepared in fibrin-coated culture dishes. Invest. Oph-
thalmol. Vis. Sci. 53 (3), 1602e1609.
Hogan, B.L., Horsburgh, G., Cohen, J., Hetherington, C.M., Fisher, G., Lyon, M.F., 1986.
Small eyes (Sey): a homozygous lethal mutation on chromosome 2 which affects the dif-
ferentiation of both lens and nasal placodes in the mouse. J. Embryol. Exp. Morphol. 97,
95e110.
Homma, R., Yoshikawa, H., Takeno, M., Kurokawa, M.S., Masuda, C., Takada, E.,
Tsubota, K., Ueno, S., Suzuki, N., 2004. Induction of epithelial progenitors in vitro
from mouse embryonic stem cells and application for reconstruction of damaged cornea
in mice. Invest. Ophthalmol. Vis. Sci. 45 (12), 4320e4326.
Hoogduijn, M.J., Crop, M.J., Peeters, A.M., Van Osch, G.J., Balk, A.H., Ijzermans, J.N.,
Weimar, W., Baan, C.C., 2007. Human heart, spleen, and perirenal fat-derived mesen-
chymal stem cells have immunomodulatory capacities. Stem Cells Dev. 16, 597e604.
Hou, J., Wang, L., Jiang, J., Zhou, C., Guo, T., Zheng, S., Wang, T., 2013. Cardiac stem
cells and their roles in myocardial infarction. Stem Cell Rev. 9 (3), 326e338.
Huang, A.J., Tseng, S.C., 1991. Corneal epithelial wound healing in the absence of limbal
epithelium. Invest. Ophthalmol. Vis. Sci. 32, 96e105.
Huang, P.T., Nelson, L.R., Bourne, W.M., 1989. The morphology and function of healing
cat corneal endothelium. Invest. Ophthalmol. Vis. Sci. 30, 1794e1801.
Hunt, J.S., Andrews, G.K., Fishback, J.L., Feess, M., Wood, G.W., 1988. Amnion mem-
brane epithelial cells express class I HLA and contain class I HLA mRNA. J. Immunol.
140, 2790e2795.
Huxlin, K.R., Hindman, H.B., Jeon, K.I., Buhren, J., Macrae, S., Demagistris, M., Ciufo, D.,
Sime, P.J., Phipps, R.P., 2013. Topical rosiglitazone is an effective anti-scarring agent in
the cornea. PLoS One 8, e70785.
Imanishi, J., Kamiyama, K., Iguchi, I., Kita, M., Sotozono, C., Kinoshita, S., 2000. Growth
factors: importance in wound healing and maintenance of transparency of the cornea.
Prog. Retin. Eye Res. 19, 113e129.
Inatomi, T., Nakamura, T., Koizumi, N., Sotozono, C., Yokoi, N., Kinoshita, S., 2006a.
Midterm results on ocular surface reconstruction using cultivated autologous oral
mucosal epithelial transplantation. Am. J. Ophthalmol. 141, 267e275.
Inatomi, T., Nakamura, T., Kojyo, M., Koizumi, N., Sotozono, C., Kinoshita, S., 2006b. Ocular
surface reconstruction with combination of cultivated autologous oral mucosal epithelial
transplantation and penetrating keratoplasty. Am. J. Ophthalmol. 142 (5), 757e764.
Javadi, M.A., Jafarinasab, M.R., Feizi, S., Karimian, F., Negahban, K., 2011. Management of
mustard gas-induced limbal stem cell deficiency and keratitis. Ophthalmology 118,
1272e1281.
Javadi, M.A., Yazdani, S., Sajjadi, H., Jadidi, K., Karimian, F., Einollahi, B., Ja’farinasab, M.R.,
Zare, M., 2005. Chronic and delayed-onset mustard gas keratitis: report of 48 patients and
review of literature. Ophthalmology 112, 617e625.
Jazedje, T., Perin, P.M., Czeresnia, C.E., Maluf, M., Halpern, S., Secco, M., Bueno, D.F.,
Vieira, N.M., Zucconi, E., Zatz, M., 2009. Human fallopian tube: a new source of
multipotent adult mesenchymal stem cells discarded in surgical procedures. J. Transl.
Med. 7, 46.
Jester, J.V., Barry, P.A., Lind, G.J., Petroll, W.M., Garana, R., Cavanagh, H.D., 1994.
Corneal keratocytes: in situ and in vitro organization of cytoskeletal contractile
proteins. Invest. Ophthalmol. Vis. Sci. 35, 730e743.
Jo, Y.Y., Lee, H.J., Kook, S.Y., Choung, H.W., Park, J.Y., Chung, J.H., Choung, Y.H.,
Kim, E.S., Yang, H.C., Choung, P.H., 2007. Isolation and characterization of postnatal
stem cells from human dental tissues. Tissue Eng. 13, 767e773.
Kadar, T., Horwitz, V., Sahar, R., Cohen, M., Cohen, L., Gez, R., Tveria, L., Gutman, H.,
Buch, H., Fishbine, E., Brandeis, R., Dachir, S., Amir, A., 2011. Delayed loss of corneal
epithelial stem cells in a chemical injury model associated with limbal stem cell deficiency
in rabbits. Curr. Eye Res. 36, 1098e1107.
Kato, T., Chang, J.H., Azar, D.T., 2003. Expression of type XVIII collagen during healing of
corneal incisions and keratectomy wounds. Invest. Ophthalmol. Vis. Sci. 44, 78e85.
Kayama, M., Kurokawa, M.S., Hiroki, U., Suzuki, N., 2007. Recent advances in corneal
regeneration and possible application of embryonic stem cell-derived corneal epithelial
cells. Clin. Ophthalmol. 1 (4), 373e382.
Kaye, S., Choudhary, A., 2006. Herpes simplex keratitis. Prog. Retin. Eye Res. 25, 355e380.
Kenyon, K.R., Tseng, S.C., 1989. Limbal autograft transplantation for ocular surface
disorders. Ophthalmology 96 (5), 709e722.
Kinoshita, S., 2010. Research and development for treating devastating corneal diseases.
Nihon Ganka Gakkai Zasshi 114, 161e199.
Koizumi, N., Cooper, L.J., Fullwood, N.J., Nakamura, T., Inoki, K., Tsuzuki, M.,
Kinoshita, S., 2002. An evaluation of cultivated corneal limbal epithelial cells, using
cell-suspension culture. Invest. Ophthalmol. Vis. Sci. 43 (7), 2114e2121.
Koizumi, N., Inatomi, T., Quantock, A.J., Fullwood, N.J., et al., 2000. Amniotic membrane
as a substrate for cultivating limbal corneal epithelial cells for autologous transplantation
in rabbits. Cornea 19 (1), 65e71.
Koizumi, N., Inatomi, T., Suzuki, T., et al., 2001a. Cultivated corneal epithelial transplan-
tation for ocular surface reconstruction in acute phase of Stevens-Johnson syndrome.
Arch. Ophthalmol. 119, 298e300.
Koizumi, N., Inatomi, T., Suzuki, T., Sotozono, C., Kinoshita, S., 2001b. Cultivated corneal
epithelial stem cell transplantation in ocular surface disorders. Ophthalmology 108 (9),
1569e1574.
Kolli, S., Ahmad, S., Lako, M., Figueiredo, F., 2010. Successful clinical implementation of
corneal epithelial stem cell therapy for treatment of unilateral limbal stem cell deficiency.
Stem Cells 28, 597e610.
Kolli, S., Ahmad, S., Mudhar, H.S., Meeny, A., Lako, M., Figueiredo, F.C., 2014. Successful
application of ex vivo expanded human autologous oral mucosal epithelium for the treat-
ment of total bilateral limbal stem cell deficiency. Stem Cells 32, 2135e2146.
Krishnan, S., Iyer, G.K., Krishnakumar, S., 2010. Culture & characterisation of limbal epithe-
lial cells & oral mucosal cells. Indian J. Med. Res. 131, 422e428.
Kubilus, J.K., Linsenmayer, T.F., 2010. Developmental corneal innervation: interactions
between nerves and specialized apical corneal epithelial cells. Invest. Ophthalmol. Vis.
Sci. 51, 782e789.
Kucerova, R., Dora, N., Mort, R.L., Wallace, K., Leiper, L.J., Lowes, C., Neves, C.,
Walczysko, P., Bruce, F., Fowler, P.A., Rajnicek, A.M., Mccaig, C.D., Zhao, M.,
West, J.D., Collinson, J.M., 2012. Interaction between hedgehog signalling and PAX6
dosage mediates maintenance and regeneration of the corneal epithelium. Mol. Vis. 18,
139e150.
Kundu, B., Rajkhowa, R., Kundu, S.C., Wang, X., 2013. Silk fibroin biomaterials for tissue
regenerations. Adv. Drug Deliv. Rev. 65 (4), 457e470.
Kubo, M., Sonoda, Y., Muramatsu, R., Usui, M., 2001. Immunogenicity of human amniotic
membrane in experimental xenotransplantation. Invest. Ophthalmol. Vis. Sci. 42 (7),
1539e1546.
Lawrence, B.D., Pan, Z., Liu, A., Kaplan, D.L., Rosenblatt, M.I., 2012. Human corneal
limbal-epithelial cell response to varying silk film geometric topography in vitro. Acta
Biomater. 8 (10), 3732e3743.
Levis, H.J., Massie, I., Dziasko, M.A., Kaasi, A., Daniels, J.T., 2013. Rapid tissue engineering
of biomimetic human corneal limbal crypts with 3D niche architecture. Biomaterials 34,
8860e8868.
Lal, I., Panchal, B.U., Basu, S., Sangwan, V.S., 2013. In-vivo expansion of autologous limbal
stem cell using simple limbal epithelial transplantation for treatment of limbal stem cell
deficiency. BMJ Case. Rep. 2013.
Lan, Y., Kodati, S., Lee, H.S., Omoto, M., Jin, Y., Chauhan, S.K., 2012. Kinetics and func-
tion of mesenchymal stem cells in corneal injury. Invest. Ophthalmol. Vis. Sci. 53 (7),
3638e3644.
Lavker, R.M., Tseng, S.C., Sun, T.T., 2004. Corneal epithelial stem cells at the limbus: look-
ing at some old problems from a new angle. Exp. Eye Res. 78, 433e446.
Lee, H.S., Hattori, T., Park, E.Y., Stevenson, W., Chauhan, S.K., Dana, R., 2012. Expres-
sion of toll-like receptor 4 contributes to corneal inflammation in experimental dry eye
disease. Invest. Ophthalmol. Vis. Sci. 53, 5632e5640.
Li, Z., Burns, A.R., Han, L., Rumbaut, R.E., Smith, C.W., 2011. IL-17 and VEGF are
necessary for efficient corneal nerve regeneration. Am. J. Pathol. 178, 1106e1116.
Li, G.G., Zhu, Y.T., Xie, H.T., Chen, S.Y., Tseng, S.C., 2012. Mesenchymal stem cells
derived from human limbal niche cells. Invest. Ophthalmol. Vis. Sci. 53, 5686e5697.
Liu, J., Lawrence, B.D., Liu, A., Schwab, I.R., Oliveira, L.A., Rosenblatt, M.I., 2012. Silk
Fibroin as a Biomaterial Substrate for Corneal Epithelial Cell Sheet Generation. Invest.
Ophthal. Vision Science 53 (7), 4130e4138.
Ljubimov, A.V., Burgeson, R.E., Butkowski, R.J., Michael, A.F., Sun, T.T., Kenney, M.C.,
1995. Human corneal basement membrane hetero-geneity: topographical differences in
the expression of type IV collagen and laminin isoforms. Lab. Invest. 72, 461e473.
Luengo Gimeno, F., Lavigne, V., Gatto, S., Croxatto, J.O., Correa, L., Gallo, J.E., 2007.
Advances in corneal stem-cell transplantation in rabbits with severe ocular alkali burns.
J. Cataract Refract. Surg. 33, 1958e1965.
Lyon, M.F., Bogani, D., Boyd, Y., Guillot, P., Favor, J., 2000. Further genetic analysis of
two autosomal dominant mouse eye defects, Ccw and Pax6 (coop). Mol. Vis. 6,
199e203.
Leenslag, J.W., Pennings, A.J., Bos, R.R., Rozema, F.R., Boering, G., 1987. Resorbable
materials of poly(L-lactide). VI. Plates and screws for internal fracture fixation. Biomate-
rials 8 (1), 70e73.
Madhira, S.L., Vemuganti, G., Bhaduri, A., Gaddipati, S., Sangwan, V.S., Ghanekar, Y.,
2008. Culture and characterization of oral mucosal epithelial cells on human amniotic
membrane for ocular surface reconstruction. Mol. Vis. 14, 189e196.
Majo, F., Rochat, A., Nicolas, M., Jaoude, G.A., Barrandon, Y., 2008. Oligopotent stem
cells are distributed throughout the mammalian ocular surface. Nature 456, 250e254.
Mamede, A.C., Carvalho, M.J., Abrantes, A.M., Laranjo, M., Maia, C.J., Botelho, M.F.,
2012. Amniotic membrane: from structure and functions to clinical applications. Cell
Tissue Res. 349 (2), 447e458.
Marchini, G., Pedrotti, E., Pedrotti, M., Barbaro, V., Di Iorio, E., Ferrari, S., Bertolin, M.,
Ferrari, B., Passilongo, M., Fasolo, A., Ponzin, D., 2012. Long-term effectiveness of
autologous cultured limbal stem cell grafts in patients with limbal stem cell deficiency
due to chemical burns. Clin. Exp. Ophthalmol. 40 (3), 255e267.
Matic, M., Petrov, I.N., Chen, S., Wang, C., Dimitrijevich, S.D., Wolosin, J.M., 1997. Stem
cells of the corneal epithelium lack connexins and metabolite transfer capacity. Differen-
tiation 61, 251e260.
Mayo, C., Ren, R., Rich, C., Stepp, M.A., Trinkaus-Randall, V., 2008. Regulation by
P2X7: epithelial migration and stromal organization in the cornea. Invest. Ophthalmol.
Vis. Sci. 49, 4384e4391.
Meller, D., Pires, R.T., Tseng, S.C., 2002. Ex vivo preservation and expansion of human
limbal epithelial stem cells on amniotic membrane cultures. Br. J. Ophthalmol. 86 (4),
463e471.
Meller, D., Thomasen, H., Steuhl, K.P., 2012. Ocular surface reconstruction in limbal stem
cell deficiency: transplantation of cultivated limbal epithelium. Ophthalmologe 109 (9),
863e868.
Meng, X., Ichim, T.E., Zhong, J., Rogers, A., Yin, Z., Jackson, J., Wang, H., Ge, W.,
Bogin, V., Chan, K.W., Thebaud, B., Riordan, N.H., 2007. Endometrial regenerative
cells: a novel stem cell population. J. Transl. Med. 5, 57.
Meyer-Blazejewska, E.A., Call, M.K., Yamanaka, O., Liu, H., Schl€ otzer-Schrehardt, U.,
Kruse, F.E., Kao, W.W., 2011. From hair to cornea: toward the therapeutic use of
hair follicle-derived stem cells in the treatment of limbal stem cell deficiency. Stem Cells
29 (1), 57e66.
Mi, S., Yang, X., Zhao, Q., Qu, L., Chen, S., Meek, K.M., Dou, Z., 2008. Reconstruction
of corneal epithelium with cryopreserved corneal limbal stem cells in a goat model. Mol.
Reprod. Dev. 75, 1607e1616.
Mohan, R.R., Hutcheon, A.E., Choi, R., Hong, J., Lee, J., Ambrosio Jr., R., Zieske, J.D.,
Wilson, S.E., 2003. Apoptosis, necrosis, proliferation, and myofibroblast generation in
the stroma following LASIK and PRK. Exp. Eye Res. 76, 71e87.
Mohan, R.R., Stapleton, W.M., Sinha, S., Netto, M.V., Wilson, S.E., 2008. A novel
method for generating corneal haze in anterior stroma of the mouse eye with the excimer
laser. Exp. Eye Res. 86, 235e240.
Morgan, R.V., Abrams, K.L., 1991. Topical administration of cyclosporine for treatment of
keratoconjunctivitis sicca in dogs. J. Am. Vet. Med. Assoc. 199, 1043e1046.
Mort, R.L., Bentley, A.J., Martin, F.L., Collinson, J.M., Douvaras, P., Hill, R.E., Morley, S.D.,
Fullwood, N.J., West, J.D., 2011. Effects of aberrant Pax6 gene dosage on mouse corneal
pathophysiology and corneal epithelial homeostasis. PLoS One 6, e28895.
Munnerlyn, C.R., Koons, S.J., Marshall, J., 1988. Photorefractive keratectomy: a technique
for laser refractive surgery. J. Cataract Refract. Surg. 14, 46e52.
Murphy, C.J., Marfurt, C.F., Mcdermott, A., Bentley, E., Abrams, G.A., Reid, T.W.,
Campbell, S., 2001. Spontaneous chronic corneal epithelial defects (SCCED) in dogs:
clinical features, innervation, and effect of topical SP, with or without IGF-1. Invest.
Murphy, M.B., Moncivais, K., Caplan, A.I., 2013. Mesenchymal stem cells: environmentally
responsive therapeutics for regenerative medicine. Exp. Mol. Med. 45, e54.
Nagy, L.J., Macrae, S., Yoon, G., Wyble, M., Wang, J., Cox, I., Huxlin, K.R., 2007. Pho-
torefractive keratectomy in the cat eye: biological and optical outcomes. J. Cataract
Refract. Surg. 33, 1051e1064.
Nakamura, T., Endo, K., Cooper, L.J., et al., 2003a. The successful culture and autologous
transplantation of rabbit oral mucosal epithelial cells on amniotic membrane. Invest.
Nakamura, T., Koizumi, N., Tsuzuki, M., Inoki, K., Sano, Y., Sotozono, C., et al., 2003b.
Successful regrafting of cultivated corneal epithelium using amniotic membrane as a car-
rier in severe ocular surface disease. Cornea 22, 70e71.
Nakamura, T., Inatomi, T., Sotozono, C., Amemiya, T., Kanamura, N., Kinoshita, S.,
2004a. Transplantation of cultivated oral mucosal epithelial cells in patients with severe
ocular surface disorders. Br. J. Ophthalmol. 88, 1280e1284.
Nakamura, T., Inatomi, T., Sotozono, C., Koizumi, N., Kinoshita, S., 2004b. Successful pri-
mary culture and autologous transplantation of corneal limbal epithelial cells from min-
imal biopsy for unilateral severe ocular surface disease. Acta Ophthalmol. Scand. 82,
468e471.
Nakamura, T., Inatomi, T., Sotozono, C., Ang, L.P.K., Koizumi, N., Yokoi, N.,
Kinoshita, S., 2006. Transplantation of autologous serum-derived cultivated corneal
epithelial equivalents for the treatment of severe ocular surface disease. Ophthalmology
113, 1765e1772.
Nakamura, T., Takeda, K., Inatomi, T., Sotozono, C., Kinoshita, S., 2011. Long-term results
of autologous cultivated oral mucosal epithelial transplantation in the scar phase of severe
ocular surface disorders. Br. J. Ophthalmol. 95, 942e946.
Nishida, K., Yamato, M., Hayashida, et al., 2004a. Corneal reconstruction with tissue engi-
neered cell sheets composed of autologous oral mucosal epithelium. N. Engl. J. Med.
351, 1187e1196.
Nishida, K., Yamato, M., Hayashida, Y., et al., 2004b. Functional bioengineered corneal
epithelial sheet grafts from corneal stem cells expanded ex vivo on a temperature-respon-
sive cell culture surface. Transplantation 77, 379e385.
Norrby, K., 2006. In vivo models of angiogenesis. J. Cell. Mol. Med. 10, 588e612.
Oh, J.Y., Kim, M.K., Ko, J.H., Lee, H.J., Lee, J.H., Wee, W.R., 2009. Rat allogeneic
mesenchymal stem cells did not prolong the survival of corneal xenograft in a pig-to-
rat model. Vet. Ophthalmol. 12 (Suppl 1), 35e40.
Oh, W., Kim, D.S., Yang, Y.S., Lee, J.K., 2008. Immunological properties of umbilical cord
blood-derived mesenchymal stromal cells. Cell Immunol. 251, 116e123.
Orr, J.W., Montz, F.J., Barter, J., et al., 2003. Continuous abdominal fascial closure: a ran-
domized controlled trial of poly(L-lactide/glycolide). Gynecol. Oncol. 90, 342e347.
Ortega, I., Ryan, A.J., Deshpande, P., MacNeil, S., Claeyssens, F., 2013. Combined micro-
fabrication and electrospinning to produce 3-D architectures for corneal repair. Acta Bio-
mater. 9 (3), 5511e5520.
Overton, J., 1965. Changes in cell fine structure during lens regeneration in Xenopus Laevis.
J. Cell Biol. 24, 211e222.
Pal-Ghosh, S., Blanco, T., Tadvalkar, G., Pajoohesh-Ganji, A., Parthasarathy, A.,
Zieske, J.D., Stepp, M.A., 2011. MMP9 cleavage of the beta4 integrin ectodomain leads
to recurrent epithelial erosions in mice. J. Cell Sci. 124, 2666e2675.
Patel, A.N., Park, E., Kuzman, M., Benetti, F., Silva, F.J., Allickson, J.G., 2008. Multipotent
menstrual blood stromal stem cells: isolation, characterization, and differentiation. Cell
Transplant 17, 303e311.
Pauklin, M., Fuchsluger, T.A., Westekemper, H., Steuhl, K.-P., Meller, 2010. Midterm re-
sults of cultivated autologous and allogeneic limbal epithelial transplantation in limbal
stem cell deficiency. Dev. Ophthalmol. 45, 57e70.
Pellegrini, G., Traverso, C.E., Franzi, A.T., Zingirian, M., Cancedda, R., De Luca, M., 1997.
Long-term restoration of damaged corneal surfaces with autologous cultivated corneal
epithelium. Lancet 349, 990e993.
Pellegrini, G., Golisano, O., Paterna, P., Lambiase, A., Bonini, S., Rama, P., De Luca, M.,
1999. Location and clonal analysis of stem cells and their differentiated progeny in the
human ocular surface. J. Cell Biol. 145 (4), 769e782.
Petroll, W.M., Jester, J.V., Bean, J., Cavanagh, H.D., 1999. Labeling of cycling corneal endo-
thelial cells during healing with a monoclonal antibody to the Ki67 antigen (MIB-1).
Cornea 18, 98e108.
Polisetty, N., Fatima, A., Madhira, S.L., Sangwan, V.S., Vemuganti, G.K., 2008. Mesen-
chymal cells from limbal stroma of human eye. Mol. Vis. 14, 431e442.
Prabhasawat, P., Ekpo, P., Uiprasertkul, M., Chotikavanich, S., Tesavibul, N., 2012. Efficacy
of cultivated corneal epithelial stem cells for ocular surface reconstruction. Clin. Oph-
thalmol. 6, 1483e1492.
Rama, P., Bonini, S., Lambiase, A., Golisano, O., Paterna, P., De Luca, M.,
Pellegrini, G., 2001. Autologous fibrin-cultured limbal stem cells permanently
restore the corneal surface of patients with total limbal stem cell deficiency. Trans-
plantation 72, 1478e1485.
Rama, P., Matuska, S., Paganoni, G., Spinelli, A., De Luca, M., Pellegrini, G., 2010.
Limbal stem-cell therapy and long-term corneal regeneration. N. Engl. J. Med. 363,
147e155.
Ramachandran, C., Basu, S., Sangwan, V.S., Balasubramanian, D., 2014. Concise review: the
coming of age of stem cell treatment for corneal surface damage. Stem Cells Transl. Med.
3, 1160e1168.
Ramaesh, T., Collinson, J.M., Ramaesh, K., Kaufman, M.H., West, J.D., Dhillon, B., 2003.
Corneal abnormalities in Pax6þ/ small eye mice mimic human aniridia-related
keratopathy. Invest. Ophthalmol. Vis. Sci. 44, 1871e1878.
Ramdass, B., Chowdhary, A., Koka, P.S., 2013. Hematological malignancies: disease path-
ophysiology of leukemic stem cells. J. Stem Cells 8 (3e4), 151e187.
Reeve, J.G., Wild, A.E., 1978. Lens regeneration from cornea of larval Xenopus laevis in the
presence of the lens. J. Embryol. Exp. Morphol. 48, 205e214.
Riminucci, M., Remoli, C., Robey, P.G., Bianco, P., 2015. Stem cells and bone diseases:
new tools, new perspective. Bone 70C, 55e61.
Sangwan, V.S., Basu, S., MacNeil, S., Balasubramanian, D., 2012. Simple limbal epithelial
transplantation (SLET): a novel surgical technique for the treatment of unilateral limbal
stem cell deficiency. Br. J. Ophthalmol. 96 (7), 931e934.
Sangwan, V.S., Basu, S., Vemuganti, G.K., Sejpal, K., Subramaniam, S.V., Bandyopadhyay, S.,
et al., 2011. Clinical outcomes of xeno-free autologous cultivated limbal epithelial trans-
plantation: a 10-year study. Br. J. Ophthalmol. 95 (11), 1525e1529.
Sangwan, V.S., Basu, S., 2011. Antimicrobial properties of amniotic membrane. Br. J. Oph-
thalmol. 95, 1e2.
Sangwan, V.S., Matalia, H.P., Vemuganti, G.K., et al., 2006. Clinical outcome of autologous
cultivated limbal epithelium transplantation. Indian J. Ophthalmol. 54, 29e34.
Sangwan, V.S., Matalia, H.P., Vemuganti, G.K., et al., 2005. Early results of penetrating ker-
atoplasty after cultivated limbal epithelium transplantation. Arch. Ophthalmol. 123 (3),
334e340.
Sangwan, V.S., Vemuganti, G.K., Singh, S., Balasubramanian, D., 2003. Successful recon-
struction of damaged ocular outer surface in humans using limbal and conjunctival
stem cell culture methods. Biosci. Rep. 23, 169e174.
Sangwan, V.S., Burman, S., Tejwani, S., Mahesh, S.P., Murthy, R., 2007. Amniotic mem-
brane transplantation: a review of current indications in the management of ophthalmic
disorders. Indian J. Ophthalmol. 55, 251e260.
Sangwan, V.S., Jain, R., Basu, S., Bagadi, A.B., Sureka, S., Mariappan, I., Macneil, S., 2014.
Transforming ocular surface stem cell research into successful clinical practice. Indian J.
Ophthalmol. 62 (1), 29e40.
Santhiago, M.R., Singh, V., Barbosa, F.L., Agrawal, V., Wilson, S.E., 2011. Monocyte
development inhibitor PRM-151 decreases corneal myofibroblast generation in
rabbits. Exp. Eye Res. 93 (5), 786e789.
Satake, Y., Higa, K., Tsubota, K., Shimazaki, J., 2011. Long-term outcome of cultivated oral
mucosal epithelial sheet transplantation in treatment of total limbal stem cell deficiency.
Ophthalmology 118, 1524e1530.
Schermer, A., Galvin, S., Sun, T.T., 1986. Differentiation-related expression of a major 64K
corneal keratin in vivo and in culture suggests limbal location of corneal epithelial stem
cells. J. Cell Biol. 103, 49e62.
Schl€otzer-Schrehardt, U., Dietrich, T., Saito, K., Sorokin, L., Sasaki, T., Paulsson, M., et al.,
2007. Characterization of extracellular matrix components in the limbal epithelial stem
cell compartment. Exp. Eye Res. 85, 845e860.
Schmahl, W., Knoedlseder, M., Favor, J., Davidson, D., 1993. Defects of neuronal migration
and the pathogenesis of cortical malformations are associated with small eye (Sey) in the
mouse, a point mutation at the Pax-6-locus. Acta Neuropathol. 86, 126e135.
Schwab, I.R., Reyes, M., Isseroff, R.R., 2000. Successful transplantation of bioengineered
tissue replacements in patients with ocular surface disease. Cornea 19 (4), 421e426.
Schwab, I.R., 1999. Cultured corneal epithelia for ocular surface disease. Trans. Am. Oph-
thalmol. Soc. 97, 891e986.
Schwartz, S.D., Regillo, C.D., Lam, B.L., Eliott, D., Rosenfeld, P.J., Gregori, N.Z.,
Hubschman, J.P., Davis, J.L., Heilwell, G., Spirn, M., Maguire, J., Gay, R.,
Bateman, J., Ostrick, R.M., Morris, D., Vincent, M., Anglade, E., Del Priore, L.V.,
Lanza, R., 2015. Human embryonic stem cell-derived retinal pigment epithelium in pa-
tients with age-related macular degeneration and Stargardt’s macular dystrophy: follow-up
of two open-label phase 1/2 studies. Lancet 385 (9967), 509e516.
Schwend, T., Lwigale, P.Y., Conrad, G.W., 2012. Nerve repulsion by the lens and cornea
during cornea innervation is dependent on Robo-Slit signaling and diminishes with
neuron age. Dev. Biol. 363, 115e127.
Sejpal, K., Ali, M.H., Maddileti, S., Basu, S., Ramappa, M., Kekunnaya, R., Vemuganti, G.K.,
Sangwan, V.S., 2013. Cultivated limbal epithelial transplantation in children with ocular
surface burns. JAMA Ophthalmol. 131 (6), 731e736.
Shanmuganathan, V.A., Foster, T., Kulkarni, B.B., Hopkinson, A., Gray, T., Powe, D.G.,
Lowe, J., Dua, H.S., 2007. Morphological characteristics of the limbal epithelial crypt.
Br. J. Ophthalmol. 91 (4), 514e519.
Sharma, V., Nemet, A., Ghabrial, R., Martin, P.A., Kourt, G., Danks, J.J., Marcells, G.,
2006. A technique for medial canthal fixation using resorbable poly-L-lactic acid-poly-
glycolic acid fixation kit. Arch. Ophthalmol. 124 (8), 1171e1174.
Shimazaki, J., Higa, K., Morito, F., Dogru, M., Kawakita, T., Satake, Y., Shimmura, S.,
Tsubota, K., 2007. Factors influencing outcomes in cultivated limbal epithelial trans-
plantation for chronic cicatricial ocular surface disorders. Am. J. Ophthalmol. 143,
945e953.
Shortt, A.J., Secker, G.A., Madhavan, S., Meligonis, G., Dart, J.K., et al., 2008. Ex vivo expan-
sion and transplantation of limbal epithelial stem cells. Ophthalmology 115, 1989e1997.
Shortt, A.J., Secker, G.A., Munro, P.M., Khaw, P.T., Tuft, S.J., Daniels, J.T., 2007. Char-
acterization of the limbal epithelial stem cell niche: novel imaging techniques permit in
vivo observation and targeted biopsy of limbal epithelial stem cells. Stem Cells 25 (6),
1402e1409.
Singh, V., Santhiago, M.R., Barbosa, F.L., Agrawal, V., Singh, N., Ambati, B.K.,
Wilson, S.E., 2011. Effect of TGFbeta and PDGF-B blockade on corneal myofibroblast
development in mice. Exp. Eye Res. 93, 810e817.
Singh, V., Jaini, R., Torricelli, A.A.M., Tuohy, V.K., Wilson, S.E., 2013. A method to
generate enhanced GFPþ chimeric mice to study the role of bone marrow-derived cells
in the eye. Exp. Eye Res. 116, 366e370.
Sitalakshmi, G., Sudha, B., Madhavan, H.N., Vinay, S., Krishnakumar, S., Mori, Y.,
Yoshioka, H., Abraham, S., 2009. Ex vivo cultivation of corneal limbal epithelial cells
in a thermoreversible polymer (Mebiol gel) and their transplantation in rabbits: an animal

model. Tissue Eng. Part A 15 (2), 407e415.
Smith, A.G., 2001. Embryo-derived stem cells: of mice and men. Annu. Rev. Cell Dev. Biol.
17, 435e462.
Spurr-Michaud, S.J., Barza, M., Gipson, I.K., 1988. An organ culture system for study of
adherence of Pseudomonas aeruginosa to normal and wounded corneas. Invest. Ophthal-
mol. Vis. Sci. 29, 379e386.
Starzl, T.E., 2000. History of clinical transplantation. World J. Surg. 24 (7), 759e782.
Stepp, M.A., Zhu, L., Sheppard, D., Cranfill, R.L., 1995. Localized distribution of alpha 9
integrin in the cornea and changes in expression during corneal epithelial cell
differentiation. J. Histochem. Cytochem. 43, 353e362.
Stepp, M.A., Gibson, H.E., Gala, P.H., Iglesia, D.D., Pajoohesh-Ganji, A., Pal-Ghosh, S.,
Brown, M., Aquino, C., Schwartz, A.M., Goldberger, O., Hinkes, M.T., Bernfield, M.,
2002. Defects in keratinocyte activation during wound healing in the syndecan-1-deficient
mouse. J. Cell Sci. 115, 4517e4531.
Stepp, M.A., Zieske, J.D., Trinkaus-Randall, V., Kyne, B.M., Pal-Ghosh, S., Tadvalkar, G.,
Pajoohesh-Ganji, A., 2014. Wounding the cornea to learn how it heals. Exp. Eye Res.
121, 178e193.
Stern, M.E., Beuerman, R.W., Fox, R.I., Gao, J., Mircheff, A.K., Pflugfelder, S.C., 1998.
The pathology of dry eye: the interaction between the ocular surface and lacrimal
glands. Cornea 17, 584e589.
Subramaniam, S.V., Sejpal, K., Fatima, A., Gaddipati, S., Vemuganti, G.K.,
Sangwan, V.S., 2013. Co-culture of autologous limbal and conjunctival epithelial
cells to treat severe ocular surface disorders: long-term survival analysis. Indian J.
Ophthalmol. 61, 202e207.
Svoboda, K.K., Nishimura, I., Sugrue, S.P., Ninomiya, Y., Olsen, B.R., 1988. Embryonic
chicken cornea and cartilage synthesize type IX collagen molecules with different
amino-terminal domains. Proc. Natl. Acad. Sci. U.S.A. 85, 7496e7500.
Talbot, M., Carrier, P., Giasson, C.J., Deschambeault, A., Guérin, S.L., Auger, F.A.,
Bazin, R., Germain, L., 2006. Autologous transplantation of rabbit limbal epithelia
cultured on fibrin gels for ocular surface reconstruction. Mol. Vis. 12, 65e75.
Takahashi, K., Tanabe, K., Ohnuki, M., Narita, M., Ichisaka, T., Tomoda, K., Yamanaka, S.,
2007. Induction of pluripotent stem cells from adult human fibroblasts by defined factors.
Cell 131 (5), 861e872.
Takahashi, K., Yamanaka, S., 2006. Induction of pluripotent stem cells from mouse embry-
onic and adult fibroblast cultures by defined factors. Cell 126 (4), 663e676.
Takeda, K., Nakamura, T., Inatomi, T., Sotozono, C., Watanabe, A., Kinoshita, S., 2011.
Ocular surface reconstruction using the combination of autologous cultivated oral
mucosal epithelial transplantation and eyelid surgery for severe ocular surface disease.
Am. J. Ophthalmol. 152 (2), 195e201.
Tandon, A., Sharma, A., Rodier, J.T., Klibanov, A.M., Rieger, F.G., Mohan, R.R., 2013.
BMP7 gene transfer via gold nanoparticles into stroma inhibits corneal fibrosis in vivo.
PLoS One 8, 66434.
Theiler, K., Varnum, D.S., Stevens, L.C., 1980. Development of Dickie’s small eye: an early
lethal mutation in the house mouse. Anat. Embryol. 161, 115e120.
Thoft, R.A., Friend, J., 1983. The X, Y, Z hypothesis of corneal epithelial maintenance.
Invest. Ophthalmol. Vis. Sci. 24, 1442e1443.
Ti, S.E., Anderson, D., Touhami, A., Kim, C., Tseng, S.C., 2002. Factors affecting outcome
following transplantation of ex vivo expanded limbal epithelium on amniotic membrane
for total limbal deficiency in rabbits. Invest. Ophthalmol. Vis. Sci. 43, 2584e2592.
Ti, S.E., Grueterich, M., Espana, E.M., Touhami, A., Anderson, D.F., Tseng, S.C., 2004.
Correlation of long term phenotypic and clinical outcomes following limbal epithelial
transplantation cultivated on amniotic membrane in rabbits. Br. J. Ophthalmol. 88, 422e

427.
Trelstad, R.L., 1973. The developmental biology of vertebrate collagens. J. Histochem.
Cytochem. 21, 521e528.
Tsai, R.J., Li, L.M., Chen, J.K., 2000. Reconstruction of damaged corneas by transplantation
of autologous limbal epithelial cells. N. Engl. J. Med. 343 (2), 86e93.
Tsuji, H., Ikarashi, K., 2004. In vitro hydrolysis of poly(L-lactide) crystalline residues as
extended-chain crystallites. Part I: long-term hydrolysis in phosphate-buffered solution
at 37 degrees C. Biomaterials 25 (24), 5449e5455.
Tuori, A., Uusitalo, H., Burgeson, R.E., Terttunen, J., Virtanen, I., 1996. The immunohisto-
chemical composition of the human corneal basement membrane. Cornea 15, 286e294.
Van-Buskirk, E.M., 1989. The anatomy of the limbus. Eye (Lond) 3 (2), 101e108.
Vazirani, J., Basu, S., Kenia, H., Ali, M.H., Kacham, S., Mariappan, I., Sangwan, V., 2014.
Unilateral partial limbal stem cell deficiency: contralateral versus ipsilateral autologous
cultivated limbal epithelial transplantation. Am. J. Ophthalmol. 157, 584e590.
Vemuganti, G.K., Kashyap, S., Sangwan, V.S., Singh, S., 2004. Ex-vivo potential of cadaveric
and fresh limbal tissues to regenerate cultured epithelium. Indian J. Ophthalmol. 52 (2),
113e120.
Wang, Z., Ge, J., Huang, B., Gao, Q., Liu, B., Wang, L., Yu, L., Fan, Z., Lu, X., Liu, J.,
2005. Differentiation of embryonic stem cells into corneal epithelium. Sci. China C
Life Sci. 48 (5), 471e478.
Weissman, I.L., Anderson, D.J., Gage, F., 2001. Stem and progenitor cells: origins, pheno-
types, lineage commitments, and transdifferentiations. Annu. Rev. Cell Dev. Biol. 17,
387e403.
Wilson, S.E., 2002. Analysis of the keratocyte apoptosis, keratocyte proliferation, and myo-
fibroblast transformation responses after photorefractive keratectomy and laser in situ
keratomileusis. Trans. Am. Ophthalmol. Soc. 100, 411e433.
Xiao, L., Nasu, M., 2014. From regenerative dentistry to regenerative medicine: prog-
ress, challenges, and potential applications of oral stem cells. Stem Cells Cloning 7,
89e99.
Xie, W., Chow, L.T., Paterson, A.J., Chin, E., Kudlow, J.E., 1999. Conditional expression of
the ErbB2 oncogene elicits reversible hyperplasia in stratified epithelia and up-regulation
of TGFalpha expression in transgenic mice. Oncogene 18, 3593e3607.
Yao, L., Li, Z.R., Su, W.R., Li, Y.P., Lin, M.L., Zhang, W.X., Liu, Y., Wan, Q., Liang, D.,
2012. Role of mesenchymal stem cells on cornea wound healing induced by acute alkali
burn. PLoS One 7 (2), e30842.
Yao, L., Bai, H., 2013. Review: mesenchymal stem cells and corneal reconstruction. Mol.
Vis. 19, 2237e2243.
Yasukawa, T., Ogura, Y., Sakurai, E., Tabata, Y., Kimura, H., 2005. Intraocular sustained drug
delivery using implantable polymeric devices. Adv. Drug Deliv. Rev. 57 (14), 2033e2046.
Ye, J., Yao, K., Kim, J.C., 2006. Mesenchymal stem cell transplantation in a rabbit corneal
alkali burn model: engraftment and involvement in wound healing. Eye (Lond) 20,
482e490.
Yu, D., Chen, M., Sun, X., Ge, J., 2013. Differentiation of mouse induced pluripotent stem
cells into corneal epithelial-like cells. Cell Biol. Int. 37 (1), 87e94.
Yuan, Y., Yeh, L.K., Liu, H., Yamanaka, O., Hardie, W.D., Kao, W.W., Liu, C.Y., 2013.
Targeted overexpression of TGF-alpha in the corneal epithelium of adult transgenic
mice induces changes in anterior segment morphology and activates noncanonical
Wnt signaling. Invest. Ophthalmol. Vis. Sci. 54, 1829e1837.
Zannettino, A.C., Paton, S., Arthur, A., Khor, F., Itescu, S., Gimble, J.M., Gronthos, S.,
2008. Multipotential human adipose-derived stromal stem cells exhibit a perivascular
phenotype in vitro and in vivo. J. Cell Physiol. 214, 413e421.
Zhang, W., Zhao, J., Chen, L., Urbanowicz, M.M., Nagasaki, T., 2008. Abnormal epithelial
homeostasis in the cornea of mice with a destrin deletion. Mol. Vis. 14, 1929e1939.
Zhou, Q., Liu, Z., Wu, Z., Wang, X., Wang, B., Li, C., Liu, Y., Li, L., Wan, P., Huang, Z.,
Wang, Z., 2014. Reconstruction of highly proliferative auto-tissue-engineered lamellar
cornea enhanced by embryonic stem cell. Tissue Eng. Part C Methods 21 (7), 639e648.
Zhu, J., Zhang, K., Sun, Y., Gao, X., Li, Y., Chen, Z., Wu, X., 2013. Reconstruction
of functional ocular surface by acellular porcine cornea matrix scaffold and limbal stem
cells derived from human embryonic stem cells. Tissue Eng. Part A 19 (21e22),
2412e2425.
CHAPTER THREE
Eukaryotic Ribosome Assembly

and Nuclear Export
Purnima Nerurkar1, 2, a, Martin Altvater1, 2, a, Stefan Gerhardy1, 3, a,
€ tz1, 2, a, Ute Fischer1, a, Christine Weirich1, a and
Sabina Schu
Vikram Govind Panse1, *
1
Institute of Biochemistry (IBC), Department of Biology (D-BIOL), ETH Zurich, Zurich, Switzerland
2
3
Biomolecular Structure and Mechanism (BSM) Graduate School, Zurich, Switzerland
*Corresponding author: E-mail: vikram.panse@bc.biol.ethz.ch
Contents
1. Introduction 108
1.1 Experimental Approaches to Understanding Ribosome Biogenesis 109
2. Assembly of 90S Preribosome, Earliest Ribosomal Precursor 110
3. Nuclear Maturation of Preribosomal Particles 111
3.1 Nuclear Maturation of Pre-60S Subunits 111
3.2 Nuclear Maturation of Pre-40S Subunits 113
4. Export of Preribosomal Subunits 114
4.1 Shared Export Factors 115
4.2 Nuclear Export of Pre-60S Subunits 117
4.3 Nuclear Export of Pre-40S Subunits 121
5. Cytoplasmic Maturation of Preribosomal Particles 124
5.1 Cytoplasmic Maturation of Pre-60S Subunits 124
5.2 Cytoplasmic Maturation of Pre-40S Subunits 127
6. Concluding Remarks 131
References 132
Abstract
Accurate translation of the genetic code into functional polypeptides is key to cellular
growth and proliferation. This essential process is carried out by the ribosome, a ribonu-
cleoprotein complex of remarkable size and intricacy. Although the structure of the
mature ribosome has provided insight into the mechanism of translation, our knowledge
regarding the assembly, quality control, and intracellular targeting of this molecular ma-
chine is still emerging. Assembly of the eukaryotic ribosome begins in the nucleolus and
requires more than 350 conserved assembly factors, which transiently associate with the
preribosome at specific maturation stages. After accomplishing their tasks, early-acting
a
These authors contributed equally.
j
ISSN 1937-6448
108 Purnima Nerurkar et al.
assembly factors are released, preparing preribosomes for nuclear export. Export compe-
tent preribosomal subunits are transported through nuclear pore complexes into the
cytoplasm, where they undergo final maturation steps, which are closely connected
to quality control, before engaging in translation. In this chapter, we focus on the final
events that commit correctly assembled ribosomal subunits for translation.
1. INTRODUCTION
The ribosome is responsible for the final step of translating genetic
codes into functional proteins. This molecular machine consists of two uni-
versally conserved subunits, both of which are composed of ribosomal
RNAs (rRNAs) and ribosomal proteins (r-proteins). Eukaryotic ribosomal
subunits are larger and more complex, and, with 65% rRNA, contain pro-
portionately more rRNA than their prokaryotic counterparts, which
contain 35% rRNA (Melnikov et al., 2012). Nevertheless, both prokaryotic
and eukaryotic ribosomal subunits share a common morphology, core struc-
ture, and enzymatic mechanism. Additional eukaryotic rRNA and r-protein
elements are largely restricted to insertions that emerge from the ribosome
core, suggesting that these regions are involved in regulatory events rather
than directly affecting peptide bond synthesis. This chapter will focus exclu-
sively on the assembly and transport of the yeast ribosome that is the model
of choice for studying this process.
Synthesis of ribosomes is a central cellular event that consumes significant
amounts of metabolites and energy. Two polymerases generate rRNAs in
parallel: RNA Polymerase I and RNA Polymerase III (see below). In addi-
tion, RNA Polymerase II is required to transcribe the 139 ribosomal protein
genes (RPGs) in a tightly regulated manner. Notably, 102 of the 139 RPGs in
budding yeast contain introns. Although such intron-containing genes repre-
sent less than 5% of genes in yeast, they account for nearly one-third of total
cellular transcription, making splicing a crucial cotranscriptional process. In
total, the concerted activity of all three-transcriptional machineries (RNA
polymerases I, II, and III), the splicing apparatus and the cellular transport sys-
tem are required to ensure highly efficient and accurate ribosome biogenesis.
In addition to rRNA and r-protein components, eukaryotic ribosomal
subunit assembly requires >350 nonribosomal factors (Gerhardy et al.,
2014). In a growing yeast cell, this spatially and temporally coordinated pro-
cess produces up to 40 nascent ribosomes per second, each of which must
travel from the nucleolus, the primary site of assembly, to the cytoplasm,
the site of translation (Warner, 1999). Although the structure and function
Ribosome Assembly and Transport 109
of the mature ribosome has been characterized at the molecular level, our
knowledge regarding the assembly pathway of this universal translating
machine is still very rudimentary.
1.1 Experimental Approaches to Understanding Ribosome

Biogenesis
Early biochemical work indicated that the process of ribosome synthesis in
budding yeast is similar to that in metazoans (Udem and Warner, 1972;
Trapman et al., 1975; Warner, 1971). This similarity validated the use of
yeast as a general model system, to which powerful genetic tools could be
applied. In the 1990s, the genetic screens in yeast led to the identification
of approximately 30 trans-acting factors involved in ribosome biogenesis
(Kressler et al., 1999). The rRNA processing analyses carried out in these
mutant yeast strains have laid the groundwork for understanding the highly
ordered steps and positioning of the cleavage sites within intermediates along
the pre-RNA processing pathway (Venema and Tollervey, 1995).
In addition to these biochemical and genetic studies, cell-biological
approaches were developed to identify factors involved in the intracellular
transport of ribosomal subunits. These in vivo transport assays employ both
large-subunit (uL23eGFP and uL5eGFP, new nomenclature according
to Ban et al. (2014)) and small-subunit (uS5eGFP) reporters and have
revealed that nucleoporins, the Ran cycle, and the general export factor
Xpo1/Crm1 are required for nuclear export of both 60S and 40S subunits
(Stage-Zimmermann et al., 2000; Milkereit et al., 2003; Hurt et al.,
1999). Visual screening of temperature sensitive mutant collections, based
on the accumulation of ribosomal reporters in the nucleolus or nucleoplasm,
has also revealed several mutants specifically defective in ribosome biogenesis
(Gerhardy et al., 2014).
Despite these fundamental advances, the composition of preribosomal
particles remained largely unknown until the past decade, when the advent
of tandem affinity purification protocols in budding yeast combined with
sensitive mass spectrometry allowed the isolation and compositional analysis
of maturing pre-60S and pre-40S particles (Rigaut et al., 1999). These
studies have provided a picture of how preribosomal particles progress
sequentially through their biogenesis pathways and have generated initial
“biochemical snapshots” of this highly complex and dynamic assembly pro-
cess (Nissan et al., 2002; Grandi et al., 2002; Schafer et al., 2003). In addi-
tion, these approaches have expanded the inventory of factors known to
be involved in ribosome biogenesis and export. Nevertheless, the precise
roles of these factors and their sites of action on maturing preribosomal par-
ticles are only beginning to be elucidated.
2. ASSEMBLY OF 90S PRERIBOSOME, EARLIEST

RIBOSOMAL PRECURSOR
Pioneering work performed in the early 1970s by the Planta and
Warner laboratories identified the earliest eukaryotic preribosomal particle,
termed 90S based on its apparent size in sedimentation experiments. The
90S was later found to be processed to the smaller 66S and 43S particles,
which are the precursors to the mature 60S and 40S subunits, respectively
(Gerhardy et al., 2014). These particles are now known to contain pre-
rRNAs, r-proteins, and multiple unknown assembly factors that are
removed as preribosomes are transported from the nucleolus to the
cytoplasm.
The 90S particle is built upon the 35S pre-rRNA, which is transcribed
from rDNA repeats in the nucleolus by RNA Polymerase I (see Figure 1).
This rRNA substrate is cotranscriptionally modified by both methylation
and pseudouridylation (Woolford and Baserga, 2013). It is also cotranscrip-
tionally loaded with a subset of small subunit (SSU) r-proteins and trans-
acting factors. Strikingly, the 90S particle does not contain r-proteins of
the large subunit or trans-acting factors involved in the 60S biogenesis
Figure 1 Overview of eukaryotic ribosome maturation. Cotranscriptional binding of

early trans-acting factors to the 35S rRNA forms the 90S preribosome (purple (dark
gray in print versions)). Cleavage at the A2 site separates the pre-60S subunit (blue
(gray in print versions)) and the pre-40S subunit (green (light gray in print versions)),
which undergo independent maturation pathways in the nucleoplasm and, eventually,
in the cytoplasm. Final maturation in the cytoplasm allows joining of both subunits to
form the functional 80S ribosome. Adapted from Gerhardy et al. (2014).
pathway. Cleavage at the A2 site releases the 20S pre-rRNA and associated
proteins as the pre-40S particle. The remaining 27SA2 pre-rRNA then
recruits r-proteins of the large subunit and pre-60 biogenesis/maturation
factors to form the pre-60S particle (Gerhardy et al., 2014).
Nucleolar ribosome assembly requires a growing yeast cell to import
w14 million r-proteins into the nucleus per generation time. The yeast
importin Kap123 is the main mediator of r-protein nuclear import but
can be functionally substituted by the importins Pse1, Sxm1, and Nmd5
(Rout et al., 1997; Sydorskyy et al., 2003; Schlenstedt et al., 1997).
R-proteins contain large basic and unstructured regions that are prone to
nonspecific interactions with nucleic acids, aggregation, and proteolytic
degradation in their nonassembled state (Jakel and Gorlich, 1998; Jakel
et al., 2002; Hasgall et al., 2011; Klinge et al., 2011; Rabl et al., 2011).
For many years, it was unclear how these intrinsically unstable and aggrega-
tion-prone proteins are targeted to the ribosome assembly site in the nucle-
olus. Work from the Panse laboratory recently identified a first carrier,
termed an escortin, that fulfills this targeting role, linking the nuclear import
machinery with the ribosome assembly pathway (Sch€ utz et al., 2014). Spe-
cifically, the escortin Tsr2 coordinates the transfer of the r-protein eS26 after
nuclear import to the assembling 90S preribosome. Tsr2 extracts eS26 from
its importins in an atypical RanGTP-independent mechanism to terminate
its import process. Subsequently, Tsr2 binds and protects the released eS26
from aggregation and proteolysis enabling its safe transfer to the 90S preri-
bosome. It has been proposed that additional escortins exist to connect
the nuclear import of additional r-proteins with ribosome assembly.
3. NUCLEAR MATURATION OF PRERIBOSOMAL

PARTICLES
3.1 Nuclear Maturation of Pre-60S Subunits
After the 90S is separated into pre-40S and pre-60S particles, these
two precursors follow independent biogenesis pathways. Within the pre-
60S, the 27SA2 pre-rRNA undergoes a complex set of endonucleolytic
and exonucleolytic cleavage events to generate mature 25S and 5.8S
rRNAs. In addition to these highly processed rRNAs, the 60S also contains
5S rRNA, which is transcribed by RNA Polymerase III and processed
separately by a set of dedicated maturation factors (van Hoof et al., 2000;
Li et al., 2009).
Incorporation of the 5S rRNA, together with the r-proteins uL18

(Rpl5) and uL5 (Rpl11), is one of the earliest steps in the 60S biogenesis
pathway. Assembly of this subcomplex requires the co-import of the two
r-proteins by Syo1, which is an adaptor for the importin Kap104, into the
nucleus. Once imported, the Syo1duL18 (Rpl5)duL5 (Rpl11) complex
is released from Kap104 in a RanGTP-dependent manner. Subsequently,
uL18 and uL5 are loaded onto the 5S rRNA and incorporated into the
maturing pre-60S particles, facilitated by the assembly factors Rpf1 and
Rpf2 (Kressler et al., 2012a; Ciganda and Williams, 2011; Zhang et al.,
2007). Despite the identification of the factors involved in 5S RNP biogen-
esis, the exact mechanism and timing of 5S rRNA incorporation remains
unknown.
On its way through the nucleoplasm, the 60S preribosome transiently
associates with >150 maturation/assembly factors. These factors are thought
to be required to promote and/or mark distinct maturation stages, after
which they are released from the preribosome to allow their participation
in new rounds of biogenesis. Energy-consuming enzymes, such as AAA-
ATPases, ABC-ATPases, GTPases, and ATP-dependent helicases, are
thought to be required to trigger sequential factor release on maturing par-
ticles, thereby conferring directionality (and irreversibility) to the assembly
process. Although the details of many of these interactions remain unclear,
recent studies have begun to address the roles and binding sites of these
enzymes (Kressler et al., 2012b; Strunk et al., 2012).
The roles of the essential AAA-ATPases Rix7 and Rea1, which are
responsible for the release and recycling of specific pre-60S biogenesis fac-
tors, are the best-studied examples of enzyme-driven assembly steps. Rix7
and Drg1 are closely related, and both display strong sequence similarity
to the well-characterized Cdc48/p97 ATPase. Rix7 is the earliest identified
AAA-ATPase acting in the pre-60S maturation pathway and is responsible
for early nucleolar 60S maturation steps (Kappel et al., 2012). Although clear
evidence remains elusive, Rix7 has been specifically implicated in the release
of Nsa1, an assembly factor, from pre-60S particles. This release step is crit-
ical to facilitate the nucleolar to nucleoplasmic movement of 60S preriboso-
mal particles (Gadal et al., 2001a; Panse et al., 2006; Kressler et al., 2008).
Rea1, which is related to the dynein heavy chain, is involved in multiple
nuclear steps in the 60S biogenesis pathway (Bassler et al., 2010; Kressler
et al., 2012b; Ulbrich et al., 2009). Rea1 is the largest protein in yeast and
forms a ringlike structure containing six ATPase modules and a taillike struc-
ture carrying the MIDAS (metal ion-dependent adhesion site) domain.
Through its ring domain, Rea1 binds the nuclear preribosome in close prox-
imity to the 60S biogenesis Rix1-Ipi3-Ipi1 subcomplex. In addition to ring
domain interactions, the flexible MIDAS domain of Rea1 contacts the
MIDAS interacting domains (MIDOs) of Ytm1 and Rsa4, thereby trig-
gering the release of these two factors in an ATP-dependent manner (Bassler
et al., 2010; Ulbrich et al., 2009; Nissan et al., 2002). The active release of
Ytm1, along with its cofactors Erb1 and Nop7, in the nucleolus has been
proposed to further facilitate the displacement of neighboring assembly fac-
tors, conferring directionality to the ribosome maturation process (Nissan
et al., 2002; Ulbrich et al., 2009). In addition to its early nucleolar role,
recent studies suggest that Rea1 is involved in a late nuclear checkpoint
step that regulates the nuclear export of 60S preribosomes through the
release of the GTPase Nug2 (Nog2). In addition to its role in the nuclear
maturation of the preribosome, Nug2 acts as placeholder for the 60S export
factor Nmd3 (Matsuo et al., 2014; Sengupta et al., 2010; Bassler et al., 2001;
Saveanu et al., 2003). The release of Nug2 from nuclear pre-60S ribosomes
depends both on its GTPase activity and the ATPase activity of Rea1. Once
Nug2 has been released, Nmd3 is able to bind and promote 60S nuclear
export (Matsuo et al., 2014; Sengupta et al., 2010). Taken together, Rea1
plays a major role in coordinating and controlling multiple nuclear matura-
tion steps and preparing 60S preribosomes for nuclear export.
3.2 Nuclear Maturation of Pre-40S Subunits

After cleavage of the 35S rRNA at sites A0, A1, and A2, the 20S pre-rRNA
and associated proteins are released from the 90S as a pre-40S particle. At this
stage, most components of the SSU processome are released, and nearly all r-
proteins of the SSU are already associated with the pre-40S (Ferreira-Cerca
et al., 2012).
Perhaps unsurprisingly, r-proteins themselves play critical roles in ribo-
some assembly. Specific examples include the role of uS13 in nucleolar
maturation events (Leger-Silvestre et al., 2004) and the dependence of
nuclear export on r-proteins uS2, uS5, uS3, eS10, and uS19 (Ferreira-Cerca
et al., 2012). More broadly, r-proteins assist in rRNA folding, trigger
conformational changes in pre-rRNA and form binding platforms for
assembly and translation factors. r-proteins involved in early pre-rRNA pro-
cessing steps are located primarily toward the 50 end of the 18S rRNA,
whereas those important for late processing steps are generally located in
close proximity to the head of the pre-40S subunit, which contains the 30
major domain of the 18S rRNA. Some r-proteins perform multiple roles
during 40S biogenesis. For example, uS5 and uS11 (Rps14) are implicated in
early pre-rRNA folding and processing (Ferreira-Cerca et al., 2012;
Jakovljevic et al., 2004; Moritz et al., 1990) as well as in assembling the
head and platform domains of the pre-40S subunit. This latter role involves
sequences close to the 30 end of the 20S pre-rRNA and is essential for cyto-
plasmic pre-rRNA processing steps (Neueder et al., 2010).
In contrast to pre-60S particles, relatively few assembly factors are known
to associate with the nucleoplasmic pre-40S subunit. Some factors are
already incorporated into 90S preribosomes prior to the first pre-rRNA
cleavage steps (Faza et al., 2012; Grandi et al., 2002). Others, including
Nob1, Rio2, Ltv1, and Tsr1, appear to bind after A0-A2 cleavage (Schafer
et al., 2003). These late associating factors remain bound to the pre-40S sub-
unit and contribute to its final maturation in the cytoplasm. Energy
consuming-enzymes, including the kinase Hrr25 and the kinases/ATPases
Rio1 and Rio2, are thought to prepare pre-40S subunits for nuclear export
and/or cytoplasmic maturation (Geerlings et al., 2003; Vanrobays et al.,
2003). However, the nuclear functions of these factors remain poorly
understood.
During their travel through the nucleoplasm, pre-40S subunits un-
dergo few compositional changes. These particles already display most of
the structural hallmarks of their mature form, including the head, platform,
and body, but they lack the characteristic beak structure (Schafer et al.,
2006). In these pre-40S particles, a trimeric subcomplex composed of the
assembly factors Enp1, Ltv1, and the r-protein uS3 is bound closely to
the prospective beak structure. Phosphorylation of uS3 and Enp1 by the ki-
nase Hrr25 weakens the pre-40S association of this subcomplex and in-
creases the conformational flexibility of the head region, which has been
speculated to be necessary for efficient nuclear export, since a prematurely
formed beak might hinder transport through the NPCs (Schafer et al.,
2006).
4. EXPORT OF PRERIBOSOMAL SUBUNITS

All transport events between the nucleus and cytoplasm occur through
nuclear pore complexes (NPCs), which are composed of approximately 30
proteins repeated in an eightfold radial symmetry (Frey and Gorlich, 2007).
Whereas small proteins (<40 kDa) and ions diffuse passively through this
transport channel, larger proteins (Bonner, 1975; Paine et al., 1975) and
macromolecular complexes require active transport due to the hydrophobic

selectivity barrier formed by phenylalanineeglycine (FG) repeat proteins
that line the central pore of the NPC.
In order to facilitate this active transport large cargos interact with spe-
cific transport receptors, primarily members of the importin-b-like transport
receptor family such as Crm1 (Cook et al., 2007). Crm1 is the major protein
export receptor and forms a complex with RanGTP and a “cargo” protein
containing a nuclear export signal (NES) (Stade et al., 1997; Fornerod et al.,
1997). This trimeric complex is dissociated upon entry into the cytoplasm
due to its contact with Ran GAP, which stimulates the GTPase activity
of Ran and provides directionality to the export process. Variants of
Crm1 that are sensitive to the inhibitor Leptomycin B (LMB) have been
widely used in screens to identify NES cargos of this receptor, including
Nmd3 (see below). However, NESs have not been identified on r-proteins,
indicating the requirement for additional transiently interacting transport
adaptors.
In general, 1:1 stoichiometry between transport receptor and cargo is
sufficient to trigger active transport through the NPC. However, the trans-
port rate per cargo:receptor complex correlates negatively with the cargo
size (Ribbeck et al., 1998). Based on this observation, a single export factor
has been proposed to be insufficient for the export of the huge (>2 MDa)
ribosomal subunit at rates required for cellular growth. Therefore, eukary-
otes are thought to utilize multiple export factors to ensure the rapid export
of preribosomal particles (Gerhardy et al., 2014).
4.1 Shared Export Factors

The export of pre-40S and pre-60S particles requires common factors as
well as export factors that are unique to each export pathway. As described
above, the first factors identified to be involved in preribosome export were
components of the Ran gradient and protein export pathways, including
Crm1. In addition, Mex67/Mtr2 and Rrp12 have been identified as com-
mon factors that promote the export of both pre-40S and pre-60S particles
(Oeffinger et al., 2004; Faza et al., 2012; Yao et al., 2008).
The Mex67-Mtr2 heterodimer (p15-NTF2 in humans) was first identi-
fied as an essential mRNA export factor and later found to play a role in
ribosome subunit export (Conti and Izaurralde, 2001; Weis, 2002; Cole
and Scarcelli, 2006). Mex67 consists of three different domains with
different functions: an N-terminal leucine-rich repeat (LRR) domain
required for mRNA binding (Strasser et al., 2000; Stutz et al., 2000), a
central NTF2-like domain, and a C-terminal ubiquitin-associated (UBA)

domain. Mtr2 also contains an NTF2-like domain (Bayliss et al., 2002),
and these matching domains mediate dimerization between Mex67 and
Mtr2. In addition, both NTF2 domains, together with the UBA-like
domain of Mex67, interact directly with FG-repeat nucleoporins to facilitate
the movement of cargo through the NPC (Strasser et al., 2000; Strawn et al.,
2001; Herold et al., 2000; Segref et al., 1997; Santos-Rosa et al., 1998).
Mex67 and Mtr2 contain loop insertions in the NTF2-like domains that
are not present in their metazoan homologues. These loops contribute to
pre-60S and pre-40S subunit binding (Yao et al., 2007; Fribourg and Conti,
2003; Senay et al., 2003; Faza et al., 2012) suggesting a versatile common
interaction platform on Mex67-Mtr2. Specifically, cross-linking experi-
ments demonstrated that Mex67 interacts with the 20S pre-rRNA of the
pre-40S and the 35S and 5.8S rRNAs of the pre-60S (Tuck and Tollervey,
2013; Yao et al., 2007). In addition, the NTF2-domain loops are predicted
to be distinct from the surfaces that are required to interact with FG-repeat
nucleoporins, suggesting that Mex67-Mtr2 could bind FG-repeat nucleo-
porins and preribosomal particles simultaneously (Fribourg and Conti,
2003; Senay et al., 2003). Deleting or mutating these loops specifically
impairs preribosome export (Faza et al., 2012; Yao et al., 2008) but not
mRNA export, indicating that Mex67-Mtr2 interacts with its distinct classes
of export cargos using different interaction surfaces and that these cargo
interactions may be differentially regulated (Faza et al., 2012; Yao et al.,
2008). However, the loop insertions interact with the nucleoporin Nup85
during both preribosome and mRNA export suggesting that these export
processes share a common route through the NPC (Yao et al., 2008).
In contrast to Crm1 and other importin-b-related transport factors,
Mex67/Mtr2 does not rely directly on the Ran-cycle for export direction-
ality. This apparent redundancy of RanGTP-dependent and -independent
export mechanisms could be required for optimal ribosome subunit export.
In addition, Mex67-Mtr2 is unique in that it contributes to the export of
three major cargos: mRNA, pre-60S, and pre-40S subunits. The reliance
of these three export pathways on a single transport factor could provide a
cellular mechanism to coregulate the availability of ribosome subunits and
translation targets.
Rrp12 is a second factor required for the export of both pre-60S and pre-
40S subunits (Oeffinger et al., 2004). Like importin-b-like transport recep-
tors, Rrp12 contains helical HEAT (Huntingtin, elongation factor 3, protein
phosphatase 2A, and TOR1) repeats, which have been shown to interact
with FG-repeat containing nucleoporins. Rrp12 has also been shown to

interact with both FG-repeat containing nucleoporins and RanGTP, and
its depletion inhibits 5.8S rRNA processing (Oeffinger et al., 2004). Based
on these observations, Rrp12 has been proposed to interact with preribo-
somes, supporting the Crm1-mediated export of these macromolecular as-
semblies through its interactions with RanGTP and FG-repeats (Oeffinger
et al., 2004). However, the detailed mechanism of how Rrp12 facilitates
ribosomal export remains to be elucidated.
4.2 Nuclear Export of Pre-60S Subunits

In addition to the shared ribosome export factors Crm1, Mex67/Mtr2, and
Rrp12 (Faza et al., 2012; Oeffinger et al., 2009; Yao et al., 2008), pre-60S
ribosome export requires additional factors that are specific to its export
pathway. These factors include the essential NES-containing adaptor
Nmd3, as well as the nonessential receptors Arx1, Bud20, and Ecm1 (Bassler
et al., 2012; Bradatsch et al., 2007; Yao et al., 2010). These auxiliary factors
are thought to share a common mechanism that relies on interactions with
FG-rich nucleoporins to directly facilitate the translocation of preribosomal
particles through the NPC channel (Figure 2).
The RNA binding protein Npl3, which was originally identified as an
adaptor for the interaction of Mex67/Mtr2 with mRNA substrates (Gilbert
and Guthrie, 2004), is also involved in pre-60S export. In npl3 mutants,
uL5-GFP accumulates in the nucleus (Stage-Zimmermann et al., 2000).
In addition, Npl3 copurifies with pre-60S ribosomal particles, and Npl3-
GFP accumulates in the nuclei of cells mutant for pre-60S export factors
(Hackmann et al., 2011; Stage-Zimmermann et al., 2000; Windgassen
et al., 2004). In contrast to its mRNA export role, Npl3 is not an adaptor
for Mex67/Mtr2 during the pre-60S export process, since Mex67 copurifies
with uL23 in npl3D cells (Hackmann et al., 2011). Instead, Nmd3 has been
proposed to be an adaptor for the recruitment of Crm1 to preribosomes,
based on its mislocalization in yeast cells treated with LMB. In support of
a role in the Crm1 pathway, genetic interactions between NPL3, NMD3,
and MTR2 also suggest that NPL3 functions independently of MEX67/
MTR2 (Hackmann et al., 2011).
During the passage of pre-60S particles through the NPC, both Npl3
and Crm1 are thought to interact with FG-repeat containing nucleoporins,
since pull downs with nucleoporins involved in pre-60S export, including
Nup82, Nsp1, Nic96, and Nup85, copurify both export factors (Fornerod
et al., 1997; Askjaer et al., 1999). In support of this model, Npl3 has also
pre-40S pre-60S
Nucleoplasm
Mex67
Npl3 Mtr2
Nup116
Nmd3 Gle2
Mex67 Ecm1
Mtr2
Bud20
Cytoplasm Arx1
Alb1
5S RNP
Rio2 Nob1 Rsa4
Mrt4
Dim1 Nog1
Tsr1 eIF6
Rlp24
Arx1
ES27L
Figure 2 Export of preribosomes through the NPC channel. Export of preribosomes is

facilitated by indicated export factors (yellow (light gray in print versions)) that interact
with the FG-repeats of the NPC channel. Cryo-EM maps of late preribosomes isolated by
Rio2-TAP and Alb1-TAP. Adapted from Gerhardy et al. (2014).
been shown to bind FG-repeats in vitro (Hackmann et al., 2011). Therefore,

Npl3 and Crm1 may act together to shield pre-60S particles from the hydro-
phobic environment of the NPC during export.
Nmd3 is composed of an N-terminal Zn2þ-binding domain and a C-
terminal domain containing two leucine-rich NESs (Ho et al., 2000; Belk
et al., 1999). The N-terminus of Nmd3 binds to the interface of the pre-
60S subunit adjacent to the uL16 binding site (Sengupta et al., 2010; Hedges
et al., 2005). Although Nmd3 is the only known NES-containing adaptor,
additional export adaptors are also required to shield the large and highly
charged surface area of the pre-60S particles during translocation through
the NPC. These additional factors include Ecm1, Bud20, and Arx1, which
share affinity for the FG-repeats of nucleoporins as well as common genetic
interactions with other export factors (Bradatsch et al., 2007; Bassler et al.,
2012; Altvater et al., 2012). However, a mutant nmd3 lacking the NES is
unable to complement the lethality caused by nmd3D (Ho et al., 2000; Gadal
et al., 2001b). In addition, overexpression of NMD3DNES causes nuclear
accumulation of the mutant protein, as well as dominant inhibition of 60S
biogenesis and export (Ho et al., 2000). These results suggest that, although
additional transport factors may promote the rapid and efficient export of the
pre-60S particle, NMD3 is an essential factor in this process.
The trans-acting factor Arx1 plays an auxiliary role in pre-60S subunit
nuclear export (Bradatsch et al., 2007; Hung et al., 2008). ARX1 encodes
a pre-60S export factor that interacts directly with FG-repeats. In addition,
ARX1 displays genetic interactions with known export factors and nucleo-
porins (Bradatsch et al., 2007), and these double mutant strains display pre-
60S export defects. Strikingly, arx1D is synthetically lethal with both ecm1D
and bud20D. Arx1 contains a methionine aminopeptidase (MetAP)-like
fold, which is present in a family of proteins that remove the N-terminal
methionine from nascent polypeptides as they emerge from the ribosome.
However, Arx1 lacks methionine aminopeptidase activity, and mutations
in the methionine-binding pocket cause defects in pre-60S subunit export
in vivo and reduced FG-repeat interactions in vitro suggesting that this fold
might interact with FG-repeat containing nucleoporins (Bradatsch et al.,
2007). However, two recent structural studies revealed that this pocket
points toward the exit tunnel of the 60S subunit, suggesting a role in inter-
acting with the pre-60S (Greber et al., 2012; Bradatsch et al., 2012). In addi-
tion to covering the exit tunnel of the 60S subunit, Arx1 also contacts the
conserved rRNA expansion segment 27 (ES27), immobilizing this dynamic
RNA region into the so-called tunnel conformation. These data suggest that
Arx1 constrains the conformation of ES27, perhaps to facilitate translocation
through the NPC or to prevent the inappropriate recruitment of translation
factors (Bradatsch et al., 2012; Greber et al., 2012).
Ecm1 is a second auxiliary factor involved in pre-60S export. Deletion of
ECM1 together with deletion of specific nucleoporins or other export fac-
tors, including ARX1, causes a strong pre-60S export defect (Yao et al.,
2010). In addition, polysome profile analysis of ecm1Darx1D double mutants
revealed a severe decrease in free 60S subunits and a polysome “halfmer”
phenotype, further indicating a pre-60S export defect. Intriguingly, deletion
of ECM1 or ARX1 alone does not impair export of pre-60 particles, sug-
gesting that these export pathways are highly redundant (Altvater et al.,
2012; Bassler et al., 2012; Bradatsch et al., 2007).
The nonessential nucleocytoplasmic shuttling protein Bud20 also dis-
plays characteristics of a pre-60S export factor, including the ability to
interact with preribosomal particles and FG-repeat containing nucleoporins.
In addition, bud20D cells are cold sensitive and display defects in pre-60S
export without the presence of additional mutations (Altvater et al., 2012;
Bassler et al., 2012). Further, strong genetic interactions between BUD20

and known export factors, including MEX67, ARX1, ECM1, and
NMD3, suggest a function in export. The N-terminus of Bud20 contains
a potential leucine-rich NES (Bassler et al., 2012), and addition of this N-
terminal sequence to an allele of NMD3 lacking its NESs complements
the lethality of nmd3D cells, suggesting that Bud20 functions as an export
adaptor for Crm1. However, in vitro experiments showed that Bud20
does not form a trimeric complex with Crm1 and RanGTP, but instead in-
teracts with different FG-nucleoporins through its Zn2þ-domain (Altvater
et al., 2012). These results suggest that Bud20 also acts as an export factor
that binds directly to FG-repeats.
In addition to the mechanism of FG-repeat-mediated transport through
the NPC shared by these export factors, an additional pathway is also used to
facilitate pre-60S export. The export factor Gle2 does not interact with FG-
repeat nucleoporins or copurify with nucleoplasmic or late cytoplasmic pre-
60S particles. Instead, Gle2 interacts specifically with shuttling ribosomes
containing Arx1. Curiously, the recruitment of Gle2 to pre-60S subunits
requires its prior tethering to the Gle2-binding sequence (GLEBS) of the
nucleoporin Nup116. However, like ARX1, ECM1, and BUD20, GLE2
genetically interacts with pre-60S export factors and double mutants display
pre-60S export defects (Occhipinti et al., 2013).
Gle2 is also involved in the mRNA export pathway, especially under
stress conditions (Occhipinti et al., 2013). Similar to mutations in MEX67
and MTR2, alleles of GLE2 have been identified that specifically impair
pre-60S export but leave mRNA export intact, indicating that the two
export roles of Gle2 are separable (Oeffinger et al., 2009; Yao et al.,
2008). Thus, Gle2 could utilize distinct interaction surfaces to prevent
kinetic delays experienced by mRNPs and pre-60S subunits during translo-
cation through the NPC channel, especially in the case when cargos have
failed to recruit its optimal complement of export factors.
One outstanding issue is the precise roles of specialized transport recep-
tors in pre-60S export. Although ECM1, ARX1, BUD20, and GLE2 are
nonessential genes, their deletions are synthetically lethal with deletions in
additional export factors or NPC components. These interactions suggest
that these proteins perform redundant functions during pre-60S export,
and also imply that ribosomal subunits require a quorum of factors, rather
than the full complement of export factors for their transport. However,
if multiple solutions to the problems of export are possible, it is unclear
how cells distinguish between export competent and incompetent particles.
In addition, since export factors generally bind pre-60S particles during late
nucleoplasmic maturation steps, these interactions could also represent po-
tential quality control mechanisms to ensure that these particles have
completed upstream biogenesis events prior to export.
Another major question in preribosomal subunit transport is how trans-
port factors successfully mask their large, highly charged surfaces to allow
transport through the hydrophobic NPC. Structural analyses and interaction
studies have shown that export factors are distributed over the surface of the
pre-60S particle. For example, Arx1 binds near the ribosome exit tunnel
(Bradatsch et al., 2012; Greber et al., 2012), whereas Nmd3 binds at the sub-
unit interface adjacent to uL16 (Hedges et al., 2005; Sengupta et al., 2010),
and Mex67/Mtr2 interacts with the 5S rRNA near uL18 and uL5 (Yao
et al., 2007). The diverse binding sites on the preribosome utilized by these
factors suggests that they work together, shielding the hydrophilic surface of
the preribosomal particle to allow interactions with FG-repeat nucleoporins
and passage through the NPC. This cooperative shielding could also explain
the presence of multiple, nonessential export factors. Additional structural
studies will allow additional transport factor mapping and may provide
insight into how preribosomal particles are oriented during translocation
through the NPC.
In addition to redundancy among export factors, there appear to be mul-
tiple, independent mechanisms to export pre-60S particles. Overexpression
of MEX67 or NMD3 rescues the slow growth and impaired pre-60S export
defects of arx1D, bud20D, and ecm1D cells, suggesting that export of the 60S
relies on the presence of multiple copies of export factors, rather than
requiring a set consisting of each factor (Yao et al., 2010; Altvater et al.,
2012). In addition, general and essential export receptors, including
Crm1, independently mediate the export of preribosomal particles. The
use of multiple export pathways may allow yeast cells to continue growth
under conditions where one pathway is overwhelmed by transport cargos
or otherwise unable to support pre-60S export.
4.3 Nuclear Export of Pre-40S Subunits

Export of the pre-40S subunit shares several characteristics with pre-60S
export, including the presence of multiple export factors that bind to
different sites on the particle, working together to shield the surface of the
pre-40S and ensuring efficient export. However, in contrast to pre-60S
export, no essential NES-containing export factor, analogous to Nmd3,
has been identified for pre-40S export. The assembly factors Ltv1 and
hRio2 have been identified through studies using nuclear accumulation of

uS5-GFP and mislocalization of ITS1 to the nucleoplasm as reporters for
pre-40S export defects. These pre-40S factors are thought to share a com-
mon mechanism that relies on Crm1 for their export functions (Zemp
et al., 2009).
Ltv1 was identified as a pre-40S export factor based on its association
with 43S/40S subunits, nuclear mislocalization in CRM1 mutants and its
interaction with Crm1 by yeast two-hybrid system (Ito et al., 2001; Neuber
et al., 2008; Seiser et al., 2006). In support of a Crm1-mediated export
mechanism, Ltv1 contains a leucine-rich NES (Merwin et al., 2014). Similar
to overexpression of nmd3DNES (Ho et al., 2000), overexpression of
ltv1DNES causes accumulation of the SSU marker uS3-GFP in the nucleus,
even in the presence of wildtype LTV1. However, in contrast to
nmd3DNES, this mutant does not cause a growth defect, suggesting that
Ltv1 is not the only NES containing adaptor of the pre-40S. In addition
to its role in export, Ltv1 may also play a role in cytoplasmic maturation
events of this subunit, since it binds the beak of the pre-40S, along with
Enp1 and uS3 (see below).
Rio2 represents a second NES-containing pre-40S adaptor. Human
Rio2 (hRio2) interacts with Crm1 in a RanGTP-dependent manner and
carries a conserved leucine-rich NES at its C-terminus. Mutation or deletion
of the NES reduces its interaction with Crm1 in vitro and impairs pre-40S
ribosome export in vivo (Zemp and Kutay, 2007). In yeast, the export of
Rio2 (yRio2) has been shown to be Crm1-depentent, since yRio2 misloc-
alizes to the nucleus upon LMB treatment (Schafer et al., 2003; Vanrobays
et al., 2003). Although RIO2 is essential, its NES is not, since yeast cells
expressing a mutant lacking the NES, RIO2DC90, are viable (Ferreira-
Cerca et al., 2012). Therefore, it does not appear to encode an essential
adaptor for Crm1-mediated pre-40S export.
Recently, Slx9 was identified as a RanGTP-binding protein that
facilitates Crm1 recruitment to Rio2 (Fischer et al., 2015). In vitro, Slx9
binds Rio2 and RanGTP to form a ternary complex that recruits Crm1
in a stepwise fashion. In addition, a mutation in Slx9 that causes defects in
Crm1-export complex formation in vitro impairs 40S preribosome export
in vivo. These data support a model in which Slx9 functions to optimally pre-
sent RanGTP and the NES to Crm1, increasing the rate of complex
formation and triggering 40S preribosome export. A family of similar
RanGTP-binding proteins could provide a mechanism to efficiently recruit
Crm1 to low-affinity NESs.
Another critical factor for pre-40S export is Yrb2, a small, shuttling

RanGTP-binding protein that localizes predominantly to the nucleus under
steady state conditions (Moy and Silver, 2002; Taura et al., 1998). Geneti-
cally, YRB2 interacts with pre-40S export factors MEX67 and MTR2 (Faza
et al., 2012). The heterodimer is present on the pre-40S during early nuclear
maturation steps and is released in the cytoplasm (Faza et al., 2012). In addi-
tion, yrb2D cells accumulate pre-40S subunits in the nucleus and have
reduced levels of pre-40S particles (Faza et al., 2012; Moy and Silver,
2002). Yrb2 interacts with Crm1 and RanGTP in vitro (Maurer et al.,
2001), and Yrb2 and Crm1 interact in the yeast two-hybrid system (Taura
et al., 1998), suggesting a role for Yrb2 in directing Crm1 and RanGTP
to NES-containing proteins. Further studies support this role, since the hu-
man homologue of Yrb2 (RanBP3) triggers the loading of Xpo1 (Crm1)
and RanGTP on specific cargos in vitro (Englmeier et al., 2001). However,
the direct function of Yrb2 in pre-40S particle export remains unclear.
Interestingly Yrb2 lacks a leucine-rich NES, and human Crm1 is thought
to interact with the FG-repeat domain of RanPB3 (Lindsay et al., 2001).
In agreement with this model, expression of a Yrb2 variant lacking FG-
repeat sequences mislocalizes to the cytoplasm (Taura et al., 1998). One pos-
sibility that remains to be tested is whether Yrb2 increases the affinity of
Crm1 for the NES-containing export adaptors Rio2 and Ltv1, thereby pro-
moting the formation of stable export complexes on the surface of pre-40S
particles.
Although considerable effort has been directed toward identifying pre-
40S export factors, relatively few have been identified. One explanation
for the lower number of identified pre-40S export factors is that this
export pathway might be less complex compared to the pre-60S, mirror-
ing the simpler biogenesis process of this subunit. In addition, it appears
that pre-40S ribosome export may not rely on the action of a single, essen-
tial NES-containing adaptor. Instead, export may utilize highly redundant,
nonessential adaptors, making it unique among the transport pathways.
This redundancy model is supported by genetic data, including the
observed synthetic lethality between ltv1D and rio2D328-425 (Ferreira-
Cerca et al., 2012). Further genetic and functional studies will be required
to determine how Rio2 and Ltv1 coordinate Crm1-dependent pre-40S
export. However, similar to pre-60S export, export of the pre-40S is likely
to require the coordination of different export factors to drive the efficient
export of preribosomal particles to the cytoplasm. More broadly, the
involvement of shared factors such as Mex67/Mtr2 suggests that the cellular
components required for protein translation are exported in a coregulated

fashion.
5. CYTOPLASMIC MATURATION OF PRERIBOSOMAL

PARTICLES
Although the assembly of ribosomal subunits begins in the nucleolus,
the final maturation steps take place in the cytoplasm. Cytoplasmic matura-
tion steps are thought to be critical both for the final structure of the ribo-
some and as a quality control mechanism to ensure that ribosomal subunits
are competent for translation. These maturation steps for both the pre-60S
and pre-40S particle are described below.
5.1 Cytoplasmic Maturation of Pre-60S Subunits

In addition to export factors, additional proteins accompany pre-60S parti-
cles into the cytoplasm, including Rlp24, Tif6, Nog1, and Arx1 (Figure 3).
Removal of these factors is critical for final maturation of the pre-60S and is
mediated by energy-consuming enzymes. These GTPases and ATPases only
Figure 3 Cytoplasmic maturation of pre-60S ribosomes. Exported pre-60S subunits are

bound by export factors (yellow (light gray in print versions)) and shuttling factors
(green (gray in print versions)), which are released in the cytoplasm. The ATPase
Drg1 releases Rlp24 from the preribosomal particles, which triggers subsequent matu-
ration steps. Arx1 and Alb1 require Rei1, Jjj1, and Ssa1/Ssa2 for their release, whereas
stalk assembly can only occur after the release of Mrt4 by Yvh1. Recruitment of uL10
(Rpp0) releases Yvh1, which allows further assembly of the P1 (Rpp1) and P2 (Rpp2) het-
erodimer onto the stalk. Loading of uL16 (Rpl10) triggers the final maturation steps. The
GTPase Efl1 and its cofactor Sdo1 release Tif6, and the GTPase Kre35 (Lsg1) removes
Nmd3. The release mechanisms/factors of shuttling assembly factors, including
Nog1, Nug1, and Nsa2 and the transport factors Mex67-Mtr2, Bud20, Ecm1, and Gle2
(depicted in gray) remain to be determined. Adapted from Gerhardy et al. (2014).
transiently bind to cytoplasmic ribosomes. Release steps are thought to

occur in a specific order, but how they are linked to each other remains
elusive (Lo et al., 2010).
One early step in cytoplasmic maturation of the pre-60S particle is the
removal of the shuttling assembly factor Rlp24. Following Rlp24 release,
the r-protein eL24 (Rpl24) is incorporated into the nascent pre-60S particle
(Saveanu et al., 2003). Rlp24 and eL24 show significant sequence similarity
and both proteins are thought to recognize the same binding site on ribo-
somal particles, with Rlp24 acting as a placeholder for eL24 (Saveanu
et al., 2003).
Rlp24 release is catalyzed by the essential AAA ATPase Drg1, which
consists of an N-terminal domain and two ATPase domains, termed D1
and D2. ATP-bound Drg1 forms hexamers, which are recruited to cyto-
plasmic pre-60S particles (Thorsness et al., 1993; Wendler et al., 1997;
Zakalskiy et al., 2002). Rlp24 recruits the hexameric form of Drg1 and,
through its C-terminal domain, triggers ATP hydrolysis by both Drg1
AAA domains (Kappel et al., 2012). These ATP hydrolysis events have
distinct functions: D2 hydrolysis causes release of Rlp24 from ribosomal sub-
units, whereas D1 hydrolysis is required for the dissociation of Drg1 from
Rlp24. In addition, Rlp24 dissociation from the pre-60S is promoted by in-
teractions between Drg1 and the nucleoporin Nup116. Interestingly, the C-
terminal ATPase activating domain of Rlp24 is not shared with eL24,
providing a mechanism for this placeholder protein to direct its own release.
Accordingly, when Drg1 function is impaired, decreased levels of ribosome
bound eL24 are observed (Kappel et al., 2012).
Inhibition of Rlp24 release can also be achieved by expressing a domi-
nant negative allele of DRG1 encoding a catalytically inactive D2
(DRG1DN (E617Q)) (Pertschy et al., 2007) or by applying the drug diaza-
borine, which blocks ATP hydrolysis by D2 (Kappel et al., 2012). These
and other experiments demonstrated that, in addition to Rlp24 release,
ATP hydrolysis by Drg1 is a prerequisite for the release of Tif6, Mex67-
Mtr2, and Mrt4. In addition, Drg1 is required to load the late cytoplasmic
factors Rei1, Sqt1, and Yvh1 onto pre-60S particles (Kappel et al., 2012;
Lo et al., 2010; Pertschy et al., 2007).
The second major cytoplasmic event in pre-60S maturation is the release
of Arx1. Based on the homology between Arx1 and MetAPs, Arx1 has been
proposed to act as MetAP analogue that prevents the premature binding of
MetAP (Bradatsch et al., 2007). In addition, Arx1 may prevent the prema-
ture engagement of preribosomes by other translation factors. For example,
Arx1 binds the pre-60S ribosome near uL23 (Rpl25), which sits at the poly-
peptide exit tunnel, a region critical for ribosome interaction with the signal
recognition particle (SRP) and subsequent targeting to the endoplasmic re-
ticulum (Bradatsch et al., 2012; Dalley et al., 2008; Greber et al., 2012). The
release of Arx1 and its binding partner, Alb1, is catalyzed by the zinc-finger
protein Rei1, which acts together with the ATPase Ssa1/Ssa2 (Hsp70) and
DnaJ domain-containing protein Jjj1 (Lebreton et al., 2006; Hung and
Johnson, 2006; Demoinet et al., 2007; Lo et al., 2010; Meyer et al., 2007,
2010).
Pre-60S maturation also requires the assembly of the ribosomal stalk.
This structure plays an essential role in translation by recruiting and acti-
vating translation and elongation factors (Ballesta and Remacha, 1996;
Berk and Cate, 2007; Gonzalo and Reboud, 2003). The stalk is built of a
single copy of uL10 (Rpp0) and two heterodimers of P1 (Rpp1) and P2
(Rpp2). In mature ribosomes, the stalk is anchored through the interaction
of uL10 with rRNA and uL11 (Rpl12). However, during early maturation,
Mrt4 acts as a placeholder for uL10. Only after Mrt4 is released and uL10 is
loaded, the maturation of the stalk progresses.
Removal of Mtr4 is catalyzed by the phosphatase Yvh1 (Kemmler et al.,
2009; Lo et al., 2009, 2010). The recruitment of Yvh1 requires the exchange
of Rlp24 for eL24, indicating a mechanism for generating directionality in
the cytoplasmic maturation process. Interestingly, the zinc-binding domain,
but not the phosphatase activity, of Yvh1 is required to release Mrt4 from
cytoplasmic pre-60S particles (Kemmler et al., 2009; Lo et al., 2009). The
exact mechanism of this release and the assembly of the stalk are still not fully
understood.
Finally, the removal of Tif6 is a critical late step in cytoplasmic pre-60S
maturation. Tif6 is thought to prevent the premature joining of 60S and 40S
preribosomal subunits in the cytoplasm (Russell and Spremulli, 1979; Valen-
zuela et al., 1982). The release of both Arx1 and Mrt4 is important and in-
dependent prerequisites for the release of Tif6 from maturing pre-60S
subunits, and failure to release Arx1 causes accumulation of Tif6 on these
particles (Hung and Johnson, 2006; Lebreton et al., 2006; Lo et al., 2010).
In addition to these indirect requirements, two proteins actively remove
Tif6: the GTPase Efl1 and the yeast ortholog of the protein mutated in
the ShwachmaneDiamond syndrome, Sdo1 (Becam et al., 2001; Menne
et al., 2007; Senger et al., 2001). Intriguingly, the Tif6 releasing factor
Efl1 is closely related to the elongation factor eEF2 (Berk and Cate, 2007;
Senger et al., 2001). This “translation-like” binding by Efl1 suggests that
its recruitment by the ribosome stalk functions both to release Tif6 and test
whether the ribosome is correctly assembled to initiate translation.
After release of Tif6, Nmd3 must be released from pre-60S particles.
Nmd3 release also requires Efl1 and Sdo1, since genetic analysis of EFL1
and SDO1 indicated that they act upstream of Nmd3 release (Lo et al.,
2010). Its also depends on the r-protein uL16 and the GTPase Kre35
(Lsg1), since uL16 and KRE35 mutants inhibit the release of Nmd3 from
pre-60S subunits. One model is that Kre35 recruits uL16, which triggers
the release of Nmd3 (Hedges et al., 2005; Karl et al., 1999; West et al.,
2005). However, how these factors work together to release Nmd3 remains
poorly understood.
In addition to these factors, which contribute to specific steps in
cytoplasmic pre-60S maturation, additional proteins have recently been
identified as bound to the pre-60S in the cytoplasm. These factors were
identified through a combination of genetic trapping, affinity purification,
and a targeted proteomic approach based on selected reaction monitoring
mass spectrometry (Altvater et al., 2012). Several unanticipated shuttling as-
sembly factors, including Nug1, Nsa2, and Rli1 were found to be released
only after Drg1-mediated release of Rlp24 in the cytoplasm. The signifi-
cance of shuttling behavior of these assembly factors is unknown. One pos-
sibility is that they participate directly in the transport and/or final functional
proofreading of pre-60S subunits. Both the functions of these factors as well
as the proteins that trigger their release in the cytoplasm remain to be
discovered.
5.2 Cytoplasmic Maturation of Pre-40S Subunits

Currently, eight trans-acting factors (Dim1, Dim2, Enp1, Hrr25, Ltv1,
Nob1, Rio2, and Tsr1) are known to accompany pre-40S subunits to the
cytoplasm. Upon arrival in the cytoplasm, the pre-40S undergoes two major
maturation events: first, a structural rearrangement occurs to generate the
beak structure of the mature 40S subunit, and second, the final 20S pre-
rRNA is cleaved to yield mature 18S rRNA (Figure 4).
As described above, the beak regions of the pre-40S subunit is bound by
assembly factors Enp1 and Ltv1, which form a salt-resistant complex with
uS3 (Schafer et al., 2006). This complex is phosphorylated by the kinase
Hrr25, which renders the association between the complex and pre-40S
salt-sensitive. Subsequent dephosphorylation of uS3 allows its stable incor-
poration into the cytoplasmic pre-40S subunit, generating the mature
beak structure (Schafer et al., 2006). The human homologue of Hrr25,
Figure 4 Cytoplasmic maturation of pre-40S ribosomes. Exported pre-40S ribosomes

are bound by export factors (shown in yellow (light gray in print versions)), and shut-
tling factors (shown in green (gray in print versions)). In one pathway (upper part), bind-
ing of the GTPase eIF5B to the pre-40S particles promotes joining of the 60S particle.
The formation of this 80S-like particle allows the release of Rio2 and Tsr1. Subsequently,
the Rli1-dependant dissociation of the 80S-like particle and processing of the 20S pre-
rRNA allow the release of Dim1, Pno1, Nob1, and Enp1. In an alternative pathway (lower
part), Rio1 associates with the pre-40S and induces the release of Enp1, Ltv1, Dim1,
Rio2, and Tsr1. Association of the 60S then triggers the release of Pno1, which triggers
20S pre-rRNA processing and the release of Rio1 and Nob1. How the 60S is dissociated
in this pathway remains to be determined.
CK1 has also been shown to phosphorylate Enp1 and Ltv1, and depletion of
CK1 leads both to 20S pre-rRNA processing defects and failure to release
Enp1, Ltv1, Rrp12, Dim2, Rio2, and Nob1 from the pre-40S particles
(Zemp et al., 2014), suggesting that beak formation and 20S pre-rRNA
are closely related. Recently, Ghalei et al. (2015) showed that Hrr25-depen-
dent release of Ltv1 is required for 60S subunit joining to allow the transla-
tion-like cycle to occur (see below).
The second essential cytoplasmic maturation step that renders pre-40S
subunits translation competent is the endonucleolytic cleavage of 20S pre-
rRNA to 18S rRNA. This processing involves two conserved events.
First, the dimethylase Dim1 modifies two consecutive conserved adenines
at the 30 end of the 20S rRNA. Second, the endonuclease Nob1 cleaves
the 20S pre-rRNA at endonucleolytic D-site to generate the mature 18S
rRNA.
Although Dim1 methylation of the 20S pre-rRNA occurs in the cyto-
plasm, Dim1 is associated with the 90S preribosome and is required for early
nucleolar processing events (Lafontaine et al., 1995, 1998; Schafer et al.,
2003). Dim1 is thought to be recruited to the nucleolus by Dim2/Pno1,
another factor involved in early nucleolar cleavage events (Vanrobays et al.,

2004; Peng et al., 2003; Senapin et al., 2003). In contrast to its essential role
in early pre-rRNA processing, dimethylation by Dim1 is not essential for
20S pre-rRNA processing. Instead, this modification has been proposed
to fine-tune translation (Lafontaine et al., 1998).
Nob1 also binds the pre-40S particle in the nucleus (Pertschy et al., 2009;
Lamanna and Karbstein, 2009). Recent studies demonstrated that Pno1 in-
teracts directly with Nob1 (Campbell and Karbstein, 2011). This interaction
led to the model that the function of Pno1 is rather to recruit Nob1 on the
pre-40S ribosomes (Campbell and Karbstein, 2011). The regulation of
Dim1 and Nob1, which are both present on nuclear pre-40S particles but
act only in the cytoplasm, remains unclear.
One possibility for the regulation of Dim1 and Nob1 activity is the pres-
ence of additional factors that regulate their access to the 20S pre-rRNA.
One such factor is the export factor Ltv1, which is also involved in cyto-
plasmic pre-40S maturation (Ball, 2011; Fassio et al., 2010; Pertschy et al.,
2009). Cryo-EM structures of pre-40S particles have shown that Ltv1,
Enp1, and uS3 obscure the hinge region formed upon opening of the
mRNA channel, suggesting that these proteins directly regulate the access
of Nob1 to its cleavage site on the 20S pre-rRNA (Strunk et al., 2011).
LTV1 displays strong genetic interactions with NOB1, PRP43, which en-
codes a DEAH-box helicase, and PFA1, which encodes a cofactor for
Prp43 (Pertschy et al., 2009). Cross-linking experiments showed that
Prp43 binds proximal to the cleavage site at the base of helix 44, which is
near the D-site of 20S pre-rRNA (Bohnsack et al., 2009). In addition,
ltv1Dpfa1D mutants display strong cytoplasmic accumulation of 20S pre-
rRNA, which can be partially rescued by introducing low copy expression
of NOB1 (Pertschy et al., 2009). These data strongly suggest that Ltv1,
Prp43, and Pfa1 promote access to the endonuclease Nob1 for efficient
cleavage.
In addition to specific interactions on the 18S rRNA, cryo-EM image
reconstructions previously revealed that several shuttling maturation factors
occupy positions on late cytoplasmic 40S preribosomes that prevent interac-
tions with translation initiation factors, mRNAs and tRNAs (Strunk et al.,
2011), providing a mechanism to proofread ribosomes before their release
into the translating pool. As described above, Enp1 and Ltv1 bind to uS3,
and these proteins also block the mRNA channel opening. In addition,
Tsr1, Dim1, and Rio2 bind to the subunit interface of the pre-40S subunit,
preventing 60S subunit joining and binding of initiation factor eIF1A.
Finally, Pno1 and Nob1 preclude the binding of the translation initiation
factor eIF3 (Strunk et al., 2011).
Intriguingly, these assembly factors partially mimic the translation initia-
tion state of mature 40S subunits on late cytoplasmic pre-40S particles.
Therefore, these interactions could probe the ability of pre-40S subunits
to interact with mature 60S subunits, providing a test of translation ability
(Strunk et al., 2012; Lebaron et al., 2012). Evidence for this includes the
observation that mutations in the r-protein uL3 that reduce its affinity for
translation elongation factors specifically impair 20S pre-rRNA processing
(Garcia-Gomez et al., 2014). In addition to ribosome assembly factors, the
GTPase Fun12 (eIF5B) promotes formation of this translation-like interac-
tion, triggering 20S pre-rRNA to 18S rRNA processing by Nob1 and
mimicking its role in subunit joining during translation (Strunk et al.,
2012; Lebaron et al., 2012). After the processing of 20S pre-rRNA within
these 80S-like particles, they are dissociated by the ABC-type ATPase
Rli1 and the tRNA mimic Dom34 (Becker et al., 2012), releasing Nob1
(Strunk et al., 2012).
Although 20S pre-rRNA processing is essential, FUN12 is not an essen-
tial gene and its depletion results in only a slight cytoplasmic accumulation of
20S pre-rRNA, suggesting that other pathways also contribute to this pro-
cessing event (Lebaron et al., 2012). Recently, an alternative pathway was
found that promotes rRNA cleavage in 80S-like particles (Turowski
et al., 2014). This second pathway requires ATP binding by Rio1, which
joins pre-40S subunits during late biogenesis, via its conserved C-terminal
domain (Ferreira-Cerca et al., 2012; Turowski et al., 2014). In vitro, RNA
cleavage is strongly stimulated by Rio1-associated pre-40S particles and is
dependent on ATP binding to Rio1. CRAC analyses on actively growing
cells have allowed the identification of binding sites for Pno1, Rio1, and
Nob1 on the pre-40S. As expected, Nob1 and Pno1 bind to overlapping
sites on ITS1 (Turowski et al., 2014). Intriguingly, Rio1 binds to the core
of the 18S rRNA structure, distinct from the sites bound by Nob1 and
Pno1. Based on these data, it has been proposed that the binding of Rio1
on pre-40S particles triggers the release of most assembly factors, with the
exception of Nob1 and Pno1/Dim2. Subsequently, joining of the mature
60S allows the release of Pno1/Dim2, followed by cleavage at site D by
Nob1 (Turowski et al., 2014).
In addition to interactions on the pre-40S, interactions with correctly
assembled mature 60S subunits are thought to be required for cleavage of
the 20S pre-rRNA (Strunk et al., 2012; Lebaron et al., 2012). One strength
of this model is that it provides a mechanism to test the translational ability of

60S subunits and pre-40S/40S subunits simultaneously. The presence of
both mature and immature ribosomal subunits in the cytoplasm suggests
that strategies have evolved to assess their correct assembly, maturation state,
and functionality. Therefore, mature 60S and 40S subunits might be
constantly interacting with each other and sensing their “decoding” ability,
potentially by segregating and targeting nonfunctional ribosomes for disas-
sembly and degradation.
6. CONCLUDING REMARKS
Despite the initial description of eukaryotic ribosome assembly, which
now dates back nearly 40 years, our understanding of this fundamental pro-
cess remains rudimentary. The development of visual reporters and proteo-
mic approaches in budding yeast has provided crucial tools to further dissect
eukaryotic ribosome biogenesis. These methodologies have expanded the
inventory of assembly and transport factors that aid ribosome production.
However, uncovering the functional contributions of these assembly factors
during ribosome formation remains a formidable task. A combination of
classical genetic approaches in budding yeast with modern structural ap-
proaches should facilitate functional analyses of this highly complex pathway
(Armache et al., 2010; Ben-Shem et al., 2011; Klinge et al., 2011; Rabl et al.,
2011; Bradatsch et al., 2012; Greber et al., 2012).
The importance of producing translation competent ribosomes is re-
flected by the growing list of human diseases that are linked to defects in
ribosome assembly. One example is the rare genetic disease Diamonde
Blackfan anemia (DBA), which causes bone marrow failure and severe ane-
mia (Ellis and Gleizes, 2011; Ellis and Lipton, 2008). DBA is characterized by
mutations in various proteins of the small and the large subunit, resulting in
reduced translation capacity of the cell. Recently, mutations in the r-protein
uL16 (Rpl10) and uL18 (Rpl5) were also found to be associated with T-cell
acute lymphoblastic leukemia (De Keersmaecker et al., 2013). In addition,
mutations in the SSU component hUTP4 are proposed to be responsible
for North American Indian childhood cirrhosis (Freed et al., 2012). Finally,
ShwachmaneBodianeDiamond syndrome arises from the inability to
release Tif6 from pre-60S subunits (Menne et al., 2007; Finch et al.,
2011; Wong et al., 2011). In addition to these genetic diseases, ribosome as-
sembly is an important target for cancer treatments, since it is a critical
process for growing and proliferating cells. Therefore, unraveling the path-
ways and mechanisms by which eukaryotes build ribosomes will generate
fundamental knowledge and therefore facilitate rational design of therapeu-
tics in the treatment of malignant cancers.
REFERENCES
Altvater, M., Chang, Y., Melnik, A., Occhipinti, L., Sch€ utz, S., Rothenbusch, U., Picotti, P.,
Panse, V.G., 2012. Targeted proteomics reveals compositional dynamics of 60S pre-
ribosomes after nuclear export. Mol. Syst. Biol. 8, 628.
Armache, J.P., Jarasch, A., Anger, A.M., Villa, E., Becker, T., Bhushan, S., Jossinet, F.,
Habeck, M., Dindar, G., Franckenberg, S., Marquez, V., Mielke, T., Thomm, M.,
Berninghausen, O., Beatrix, B., Soding, J., Westhof, E., Wilson, D.N.,
Beckmann, R., 2010. Cryo-EM structure and rRNA model of a translating eukaryotic
80S ribosome at 5.5-A resolution. Proc. Natl. Acad. Sci. U.S.A. 107, 19748e19753.
Askjaer, P., Bachi, A., Wilm, M., Bischoff, F.R., Weeks, D.L., Ogniewski, V., Ohno, M.,
Niehrs, C., Kjems, J., Mattaj, I.W., Fornerod, M., 1999. RanGTP-regulated interactions
of CRM1 with nucleoporins and a shuttling DEAD-box helicase. Mol. Cell Biol. 19,
6276e6285.
Ball, S., 2011. Diamond Blackfan anemia. Hematol. Am. Soc. Hematol. Educ. Program
2011, 487e491.
Ballesta, J.P., Remacha, M., 1996. The large ribosomal subunit stalk as a regulatory element of
the eukaryotic translational machinery. Prog. Nucleic Acid. Res. Mol. Biol. 55, 157e193.
Ban, N., Beckmann, R., Cate, J.H., Dinman, J.D., Dragon, F., Ellis, S.R., Lafontaine, D.L.,
Lindahl, L., Liljas, A., Lipton, J.M., Mcalear, M.A., Moore, P.B., Noller, H.F.,
Ortega, J., Panse, V.G., Ramakrishnan, V., Spahn, C.M., Steitz, T.A.,
Tchorzewski, M., Tollervey, D., Warren, A.J., Williamson, J.R., Wilson, D.,
Yonath, A., Yusupov, M., 2014 Feb. A new system for naming ribosomal proteins.
Curr. Opin. Struct. Biol. 24, 165e169.
Bassler, J., Grandi, P., Gadal, O., Lessmann, T., Petfalski, E., Tollervey, D., Lechner, J.,
Hurt, E., 2001. Identification of a 60S preribosomal particle that is closely linked to nu-
clear export. Mol. Cell 8, 517e529.
Bassler, J., Kallas, M., Pertschy, B., Ulbrich, C., Thoms, M., Hurt, E., 2010. The AAA-
ATPase Rea1 drives removal of biogenesis factors during multiple stages of 60S ribosome
assembly. Mol. Cell 38, 712e721.
Bassler, J., Klein, I., Schmidt, C., Kallas, M., Thomson, E., Wagner, M.A., Bradatsch, B.,
Rechberger, G., Strohmaier, H., Hurt, E., Bergler, H., 2012. The conserved Bud20
zinc finger protein is a new component of the ribosomal 60S subunit export
machinery. Mol. Cell Biol. 32, 4898e4912.
Bayliss, R., Leung, S.W., Baker, R.P., Quimby, B.B., Corbett, A.H., Stewart, M., 2002.
Structural basis for the interaction between NTF2 and nucleoporin FxFG repeats.
EMBO J. 21, 2843e2853.
Becam, A.M., Nasr, F., Racki, W.J., Zagulski, M., Herbert, C.J., 2001. Ria1p (Ynl163c), a
protein similar to elongation factors 2, is involved in the biogenesis of the 60S subunit of
the ribosome in Saccharomyces cerevisiae. Mol. Genet. Genomics 266, 454e462.
Becker, T., Franckenberg, S., Wickles, S., Shoemaker, C.J., Anger, A.M., Armache, J.P.,
Sieber, H., Ungewickell, C., Berninghausen, O., Daberkow, I., Karcher, A.,
Thomm, M., Hopfner, K.P., Green, R., Beckmann, R., 2012. Structural basis of highly
conserved ribosome recycling in eukaryotes and archaea. Nature 482, 501e506.
Belk, J.P., He, F., Jacobson, A., 1999. Overexpression of truncated Nmd3p inhibits protein
synthesis in yeast. RNA 5, 1055e1070.
Ben-Shem, A., Garreau De Loubresse, N., Melnikov, S., Jenner, L., Yusupova, G.,
Yusupov, M., 2011. The structure of the eukaryotic ribosome at 3.0 A resolution.
Science 334, 1524e1529.
Berk, V., Cate, J.H., 2007. Insights into protein biosynthesis from structures of bacterial
ribosomes. Curr. Opin. Struct. Biol. 17, 302e309.
Bohnsack, M.T., Martin, R., Granneman, S., Ruprecht, M., Schleiff, E., Tollervey, D.,
2009. Prp43 bound at different sites on the pre-rRNA performs distinct functions in
ribosome synthesis. Mol. Cell 36, 583e592.
Bonner, W.M., 1975. Protein migration into nuclei. I. Frog oocyte nuclei in vivo accumulate
microinjected histones, allow entry to small proteins, and exclude large proteins. J. Cell
Biol. 64, 421e430.
Bradatsch, B., Katahira, J., Kowalinski, E., Bange, G., Yao, W., Sekimoto, T.,
Baumgartel, V., Boese, G., Bassler, J., Wild, K., Peters, R., Yoneda, Y., Sinning, I.,
Hurt, E., 2007. Arx1 functions as an unorthodox nuclear export receptor for the 60S
preribosomal subunit. Mol. Cell 27, 767e779.
Bradatsch, B., Leidig, C., Granneman, S., Gn€adig, M., Tollervey, D., B€ ottcher, B.,
Beckmann, R., Hurt, E., 2012. Structure of the pre-60S ribosomal subunit with
nuclear export factor Arx1 bound at the exit tunnel. Nat. Struct. Mol. Biol. 19,
1234e1241.
Campbell, M.G., Karbstein, K., 2011. Protein-protein interactions within late pre-40S
ribosomes. PLoS One 6, e16194.
Ciganda, M., Williams, N., 2011. Eukaryotic 5S rRNA biogenesis. Wiley Interdiscip. Rev.
RNA 2, 523e533.
Cole, C.N., Scarcelli, J.J., 2006. Transport of messenger RNA from the nucleus to the
cytoplasm. Curr. Opin. Cell Biol. 18, 299e306.
Conti, E., Izaurralde, E., 2001. Nucleocytoplasmic transport enters the atomic age. Curr.
Opin. Cell Biol. 13, 310e319.
Cook, A., Bono, F., Jinek, M., Conti, E., 2007. Structural biology of nucleocytoplasmic
transport. Annu. Rev. Biochem. 76, 647e671.
Dalley, J.A., Selkirk, A., Pool, M.R., 2008. Access to ribosomal protein Rpl25p by the signal
recognition particle is required for efficient cotranslational translocation. Mol. Biol. Cell
19, 2876e2884.
De Keersmaecker, K., Atak, Z.K., Li, N., Vicente, C., Patchett, S., Girardi, T., Gianfelici, V.,
Geerdens, E., Clappier, E., Porcu, M., Lahortiga, I., Luca, R., Yan, J., Hulselmans, G.,
Vranckx, H., Vandepoel, R., Sweron, B., Jacobs, K., Mentens, N., Wlodarska, I.,
Cauwelier, B., Cloos, J., Soulier, J., Uyttebroeck, A., Bagni, C., Hassan, B.A.,
Vandenberghe, P., Johnson, A.W., Aerts, S., Cools, J., 2013. Exome sequencing iden-
tifies mutation in CNOT3 and ribosomal genes RPL5 and RPL10 in T-cell acute
lymphoblastic leukemia. Nat. Genet. 45, 186e190.
Demoinet, E., Jacquier, A., Lutfalla, G., Fromont-Racine, M., 2007. The Hsp40 chaperone
Jjj1 is required for the nucleo-cytoplasmic recycling of preribosomal factors in Saccharo-
myces cerevisiae. RNA 13, 1570e1581.
Ellis, S.R., Gleizes, P.E., 2011. Diamond Blackfan anemia: ribosomal proteins going rogue.
Semin. Hematol. 48, 89e96.
Ellis, S.R., Lipton, J.M., 2008. Diamond Blackfan anemia: a disorder of red blood cell
development. Curr. Top. Dev. Biol. 82, 217e241.
Englmeier, L., Fornerod, M., Bischoff, F.R., Petosa, C., Mattaj, I.W., Kutay, U., 2001.
RanBP3 influences interactions between CRM1 and its nuclear protein export
substrates. EMBO Rep. 2, 926e932.
Fassio, C.A., Schofield, B.J., Seiser, R.M., Johnson, A.W., Lycan, D.E., 2010. Dominant
mutations in the late 40S biogenesis factor Ltv1 affect cytoplasmic maturation of the small
ribosomal subunit in Saccharomyces cerevisiae. Genetics 185, 199e209.
Faza, M.B., Chang, Y., Occhipinti, L., Kemmler, S., Panse, V.G., 2012. Role of Mex67-
Mtr2 in the nuclear export of 40S pre-ribosomes. PLoS Genet. 8, e1002915.
Ferreira-Cerca, S., Sagar, V., Schafer, T., Diop, M., Wesseling, A.M., Lu, H., Chai, E.,
Hurt, E., Laronde-Leblanc, N., 2012. ATPase-dependent role of the atypical kinase
Rio2 on the evolving pre-40S ribosomal subunit. Nat. Struct. Mol. Biol. 19,
1316e1323.
Finch, A.J., Hilcenko, C., Basse, N., Drynan, L.F., Goyenechea, B., Menne, T.F., Gonzalez
Fernandez, A., Simpson, P., D’Santos, C.S., Arends, M.J., Donadieu, J., Bellanne-
Chantelot, C., Costanzo, M., Boone, C., Mckenzie, A.N., Freund, S.M.,
Warren, A.J., 2011. Uncoupling of GTP hydrolysis from eIF6 release on the ribosome
causes Shwachman-Diamond syndrome. Genes Dev. 25, 917e929.
Fischer, U., Schauble, N., Sch€ utz, S., Altvater, M., Chang, Y., Boulos Faza, M., Panse, V.G.,
2015. A non-canonical mechanism for Crm1-export cargo complex assembly. eLife 4.
Fornerod, M., Ohno, M., Yoshida, M., Mattaj, I.W., 1997. CRM1 is an export receptor for
leucine-rich nuclear export signals. Cell 90, 1051e1060.
Freed, E.F., Prieto, J.L., Mccann, K.L., Mcstay, B., Baserga, S.J., 2012. NOL11, implicated in
the pathogenesis of North American Indian childhood cirrhosis, is required for pre-
rRNA transcription and processing. PLoS Genet. 8, e1002892.
Frey, S., Gorlich, D., 2007. A saturated FG-repeat hydrogel can reproduce the permeability
properties of nuclear pore complexes. Cell 130, 512e523.
Fribourg, S., Conti, E., 2003. Structural similarity in the absence of sequence homology of
the messenger RNA export factors Mtr2 and p15. EMBO Rep. 4, 699e703.
Gadal, O., Strauss, D., Braspenning, J., Hoepfner, D., Petfalski, E., Philippsen, P.,
Tollervey, D., Hurt, E., 2001a. A nuclear AAA-type ATPase (Rix7p) is required for
biogenesis and nuclear export of 60S ribosomal subunits. EMBO J. 20, 3695e3704.
Gadal, O., Strauss, D., Kessl, J., Trumpower, B., Tollervey, D., Hurt, E., 2001b. Nuclear
export of 60s ribosomal subunits depends on Xpo1p and requires a nuclear export
sequence-containing factor, Nmd3p, that associates with the large subunit protein
Rpl10p. Mol. Cell Biol. 21, 3405e3415.
Garcia-Gomez, J.J., Fernandez-Pevida, A., Lebaron, S., Rosado, I.V., Tollervey, D.,
Kressler, D., De La Cruz, J., 2014. Final Pre-40S maturation depends on the functional
integrity of the 60S subunit ribosomal protein L3. PLoS Genet. 10, e1004205.
Geerlings, T.H., Faber, A.W., Bister, M.D., Vos, J.C., Raue, H.A., 2003. Rio2p, an
evolutionarily conserved, low abundant protein kinase essential for processing of 20 S
Pre-rRNA in Saccharomyces cerevisiae. J. Biol. Chem. 278, 22537e22545.
Gerhardy, S., Menet, A.M., Pena, C., Petkowski, J.J., Panse, V.G., 2014. Assembly and nu-
clear export of pre-ribosomal particles in budding yeast. Chromosoma 123, 327e344.
Ghalei, H., Schaub, F.X., Doherty, J.R., Noguchi, Y., Roush, W.R., Cleveland, J.L.,
Stroupe, M.E., Karbstein, K., 2015. Hrr25/CK1delta-directed release of Ltv1 from
pre-40S ribosomes is necessary for ribosome assembly and cell growth. J. Cell Biol.
208, 745e759.
Gilbert, W., Guthrie, C., 2004. The Glc7p nuclear phosphatase promotes mRNA export by
facilitating association of Mex67p with mRNA. Mol. Cell 13, 201e212.
Gonzalo, P., Reboud, J.P., 2003. The puzzling lateral flexible stalk of the ribosome. Biol.
Cell 95, 179e193.
Grandi, P., Rybin, V., Bassler, J., Petfalski, E., Strauss, D., Marzioch, M., Schafer, T.,
Kuster, B., Tschochner, H., Tollervey, D., Gavin, A.C., Hurt, E., 2002. 90S pre-ribo-
somes include the 35S pre-rRNA, the U3 snoRNP, and 40S subunit processing factors
but predominantly lack 60S synthesis factors. Mol. Cell 10, 105e115.
Greber, B.J., Boehringer, D., Montellese, C., Ban, N., 2012. Cryo-EM structures of Arx1
and maturation factors Rei1 and Jjj1 bound to the 60S ribosomal subunit. Nat. Struct.
Mol. Biol. 19, 1228e1233.
Hackmann, A., Gross, T., Baierlein, C., Krebber, H., 2011. The mRNA export factor
Npl3 mediates the nuclear export of large ribosomal subunits. EMBO Rep. 12,
1024e1031.
Hasgall, P.A., Hoogewijs, D., Faza, M.B., Panse, V.G., Wenger, R.H., Camenisch, G., 2011.
The putative RNA helicase HELZ promotes cell proliferation, translation initiation and
ribosomal protein S6 phosphorylation. PLoS One 6, e22107.
Hedges, J., West, M., Johnson, A.W., 2005. Release of the export adapter, Nmd3p, from the
60S ribosomal subunit requires Rpl10p and the cytoplasmic GTPase Lsg1p. EMBO J. 24,
567e579.
Herold, A., Suyama, M., Rodrigues, J.P., Braun, I.C., Kutay, U., Carmo-Fonseca, M.,
Bork, P., Izaurralde, E., 2000. TAP (NXF1) belongs to a multigene family of putative
RNA export factors with a conserved modular architecture. Mol. Cell Biol. 20,
8996e9008.
Ho, J.H., Kallstrom, G., Johnson, A.W., 2000. Nascent 60S ribosomal subunits enter the free
pool bound by Nmd3p. RNA 6, 1625e1634.
Hung, N.J., Johnson, A.W., 2006. Nuclear recycling of the pre-60S ribosomal subunit-
associated factor Arx1 depends on Rei1 in Saccharomyces cerevisiae. Mol. Cell Biol. 26,
3718e3727.
Hung, N.J., Lo, K.Y., Patel, S.S., Helmke, K., Johnson, A.W., 2008. Arx1 is a nuclear export
receptor for the 60S ribosomal subunit in yeast. Mol. Biol. Cell 19, 735e744.
Hurt, E., Hannus, S., Schmelzl, B., Lau, D., Tollervey, D., Simos, G., 1999. A novel in vivo
assay reveals inhibition of ribosomal nuclear export in ran-cycle and nucleoporin
mutants. J. Cell Biol. 144, 389e401.
van Hoof, A., Lennertz, P., Parker, R., 2000. Three conserved members of the RNase D
family have unique and overlapping functions in the processing of 5S, 5.8S, U4, U5,
RNase MRP and RNase P RNAs in yeast. EMBO J. 19, 1357e1365.
Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M., Sakaki, Y., 2001. A comprehensive
two-hybrid analysis to explore the yeast protein interactome. Proc. Natl. Acad. Sci.
U.S.A. 98, 4569e4574.
Jakel, S., Gorlich, D., 1998. Importin beta, transportin, RanBP5 and RanBP7 mediate
nuclear import of ribosomal proteins in mammalian cells. EMBO J. 17, 4491e4502.
Jakel, S., Mingot, J.M., Schwarzmaier, P., Hartmann, E., Gorlich, D., 2002. Importins fulfil a
dual function as nuclear import receptors and cytoplasmic chaperones for exposed basic
domains. EMBO J. 21, 377e386.
Jakovljevic, J., De Mayolo, P.A., Miles, T.D., Nguyen, T.M., Leger-Silvestre, I., Gas, N.,
Woolford Jr., J.L., 2004. The carboxy-terminal extension of yeast ribosomal protein
S14 is necessary for maturation of 43S preribosomes. Mol. Cell 14, 331e342.
Kappel, L., Loibl, M., Zisser, G., Klein, I., Fruhmann, G., Gruber, C., Unterweger, S.,
Rechberger, G., Pertschy, B., Bergler, H., 2012. Rlp24 activates the AAA-ATPase
Drg1 to initiate cytoplasmic pre-60S maturation. J. Cell Biol. 199, 771e782.
Karl, T., Onder, K., Kodzius, R., Pichova, A., Wimmer, H., Th, R.,A., Hundsberger, H.,
Loffler, M., Klade, T., Beyer, A., Breitenbach, M., Koller, L., 1999. GRC5 and
NMD3 function in translational control of gene expression and interact genetically.
Curr. Genet. 34, 419e429.
Kemmler, S., Occhipinti, L., Veisu, M., Panse, V.G., 2009. Yvh1 is required for a late matu-
ration step in the 60S biogenesis pathway. J. Cell Biol. 186, 863e880.
Klinge, S., Voigts-Hoffmann, F., Leibundgut, M., Arpagaus, S., Ban, N., 2011. Crystal struc-
ture of the eukaryotic 60S ribosomal subunit in complex with initiation factor 6. Science
334, 941e948.
Kressler, D., Bange, G., Ogawa, Y., Stjepanovic, G., Bradatsch, B., Pratte, D., Amlacher, S.,
Strauss, D., Yoneda, Y., Katahira, J., Sinning, I., Hurt, E., 2012a. Synchronizing nuclear
import of ribosomal proteins with ribosome assembly. Science 338, 666e671.
Kressler, D., Hurt, E., Bergler, H., Bassler, J., 2012b. The power of AAA-ATPases on the
road of pre-60S ribosome maturationemolecular machines that strip pre-ribosomal
particles. Biochim. Biophys. Acta 1823, 92e100.
Kressler, D., Linder, P., De La Cruz, J., 1999. Protein trans-acting factors involved in ribo-
some biogenesis in Saccharomyces cerevisiae. Mol. Cell Biol. 19, 7897e7912.
Kressler, D., Roser, D., Pertschy, B., Hurt, E., 2008. The AAA ATPase Rix7 powers pro-
gression of ribosome biogenesis by stripping Nsa1 from pre-60S particles. J. Cell Biol.
181, 935e944.
Lafontaine, D., Vandenhaute, J., Tollervey, D., 1995. The 18S rRNA dimethylase Dim1p is
required for pre-ribosomal RNA processing in yeast. Genes Dev. 9, 2470e2481.
Lafontaine, D.L., Preiss, T., Tollervey, D., 1998. Yeast 18S rRNA dimethylase Dim1p: a
quality control mechanism in ribosome synthesis? Mol. Cell Biol. 18, 2360e2370.
Lamanna, A.C., Karbstein, K., 2009. Nob1 binds the single-stranded cleavage site D at the
30 -end of 18S rRNA with its PIN domain. Proc. Natl. Acad. Sci. U.S.A. 106,
14259e14264.
Lebaron, S., Schneider, C., Van Nues, R.W., Swiatkowska, A., Walsh, D., Bottcher, B.,
Granneman, S., Watkins, N.J., Tollervey, D., 2012. Proofreading of pre-40S ribosome
maturation by a translation initiation factor and 60S subunits. Nat. Struct. Mol. Biol. 19,
744e753.
Lebreton, A., Saveanu, C., Decourty, L., Rain, J.C., Jacquier, A., Fromont-Racine, M.,
2006. A functional network involved in the recycling of nucleocytoplasmic pre-60S
factors. J. Cell Biol. 173, 349e360.
Leger-Silvestre, I., Milkereit, P., Ferreira-Cerca, S., Saveanu, C., Rousselle, J.C.,
Choesmel, V., Guinefoleau, C., Gas, N., Gleizes, P.E., 2004. The ribosomal protein
Rps15p is required for nuclear exit of the 40S subunit precursors in yeast. EMBO J.
23, 2336e2347.
Li, Z., Lee, I., Moradi, E., Hung, N.J., Johnson, A.W., Marcotte, E.M., 2009. Rational
extension of the ribosome biogenesis pathway using network-guided genetics. PLoS
Biol. 7, e1000213.
Lindsay, M.E., Holaska, J.M., Welch, K., Paschal, B.M., Macara, I.G., 2001. Ran-binding
protein 3 is a cofactor for Crm1-mediated nuclear protein export. J. Cell Biol. 153,
1391e1402.
Lo, K.Y., Li, Z., Bussiere, C., Bresson, S., Marcotte, E.M., Johnson, A.W., 2010.
Defining the pathway of cytoplasmic maturation of the 60S ribosomal subunit. Mol.
Cell 39, 196e208.
Lo, K.Y., Li, Z., Wang, F., Marcotte, E.M., Johnson, A.W., 2009. Ribosome stalk assembly
requires the dual-specificity phosphatase Yvh1 for the exchange of Mrt4 with P0. J. Cell
Biol. 186, 849e862.
Matsuo, Y., Granneman, S., Thoms, M., Manikas, R.G., Tollervey, D., Hurt, E., 2014.
Coupled GTPase and remodelling ATPase activities form a checkpoint for ribosome
export. Nature 505, 112e116.
Maurer, P., Redd, M., Solsbacher, J., Bischoff, F.R., Greiner, M., Podtelejnikov, A.V.,
Mann, M., Stade, K., Weis, K., Schlenstedt, G., 2001. The nuclear export receptor
Xpo1p forms distinct complexes with NES transport substrates and the yeast Ran bind-
ing protein 1 (Yrb1p). Mol. Biol. Cell 12, 539e549.
Melnikov, S., Ben-Shem, A., Garreau De Loubresse, N., Jenner, L., Yusupova, G.,
Yusupov, M., 2012. One core, two shells: bacterial and eukaryotic ribosomes. Nat.
Struct. Mol. Biol. 19, 560e567.
Menne, T.F., Goyenechea, B., Sanchez-Puig, N., Wong, C.C., Tonkin, L.M., Ancliff, P.J.,
Brost, R.L., Costanzo, M., Boone, C., Warren, A.J., 2007. The Shwachman-Bodian-
Diamond syndrome protein mediates translational activation of ribosomes in yeast.
Nat. Genet. 39, 486e495.
Merwin, J.R., Bogar, L.B., Poggi, S.B., Fitch, R.M., Johnson, A.W., Lycan, D.E., 2014. Ge-
netic analysis of the ribosome biogenesis factor Ltv1 of Saccharomyces cerevisiae. Genetics
198, 1071e1085.
Meyer, A.E., Hoover, L.A., Craig, E.A., 2010. The cytosolic J-protein, Jjj1, and Rei1 func-
tion in the removal of the pre-60 S subunit factor Arx1. J. Biol. Chem. 285, 961e968.
Meyer, A.E., Hung, N.J., Yang, P., Johnson, A.W., Craig, E.A., 2007. The specialized cyto-
solic J-protein, Jjj1, functions in 60S ribosomal subunit biogenesis. Proc. Natl. Acad. Sci.
U.S.A. 104, 1558e1563.
Milkereit, P., Strauss, D., Bassler, J., Gadal, O., Kuhn, H., Schutz, S., Gas, N., Lechner, J.,
Hurt, E., Tschochner, H., 2003. A Noc complex specifically involved in the formation
and nuclear export of ribosomal 40 S subunits. J. Biol. Chem. 278, 4072e4081.
Moritz, M., Paulovich, A.G., Tsay, Y.F., Woolford Jr., J.L., 1990. Depletion of yeast ribo-
somal proteins L16 or rp59 disrupts ribosome assembly. J. Cell Biol. 111, 2261e2274.
Moy, T.I., Silver, P.A., 2002. Requirements for the nuclear export of the small ribosomal
subunit. J. Cell Sci. 115, 2985e2995.
Neuber, A., Franke, J., Wittstruck, A., Schlenstedt, G., Sommer, T., Stade, K., 2008. Nuclear
export receptor Xpo1/Crm1 is physically and functionally linked to the spindle pole
body in budding yeast. Mol. Cell Biol. 28, 5348e5358.
Neueder, A., Jakob, S., Poll, G., Linnemann, J., Deutzmann, R., Tschochner, H.,
Milkereit, P., 2010. A local role for the small ribosomal subunit primary binder rpS5
in final 18S rRNA processing in yeast. PLoS One 5, e10194.
Nissan, T.A., Bassler, J., Petfalski, E., Tollervey, D., Hurt, E., 2002. 60S pre-ribosome for-
mation viewed from assembly in the nucleolus until export to the cytoplasm. EMBO J.
21, 5539e5547.
Occhipinti, L., Chang, Y., Altvater, M., Menet, A.M., Kemmler, S., Panse, V.G., 2013.
Non-FG mediated transport of the large pre-ribosomal subunit through the nuclear
pore complex by the mRNA export factor Gle2. Nucleic Acids Res. 41, 8266e8279.
Oeffinger, M., Dlakic, M., Tollervey, D., 2004. A pre-ribosome-associated HEAT-repeat
protein is required for export of both ribosomal subunits. Genes Dev. 18, 196e209.
Oeffinger, M., Zenklusen, D., Ferguson, A., Wei, K.E., El Hage, A., Tollervey, D.,
Chait, B.T., Singer, R.H., Rout, M.P., 2009. Rrp17p is a eukaryotic exonuclease
required for 5’ end processing of Pre-60S ribosomal RNA. Mol. Cell 36, 768e781.
Paine, P.L., Moore, L.C., Horowitz, S.B., 1975. Nuclear envelope permeability. Nature 254,
109e114.
Panse, V.G., Kressler, D., Pauli, A., Petfalski, E., Gnadig, M., Tollervey, D., Hurt, E., 2006.
Formation and nuclear export of preribosomes are functionally linked to the small-ubiq-
uitin-related modifier pathway. Traffic 7, 1311e1321.
Peng, W.T., Robinson, M.D., Mnaimneh, S., Krogan, N.J., Cagney, G., Morris, Q.,
Davierwala, A.P., Grigull, J., Yang, X., Zhang, W., Mitsakakis, N., Ryan, O.W.,
Datta, N., Jojic, V., Pal, C., Canadien, V., Richards, D., Beattie, B., Wu, L.F.,
Altschuler, S.J., Roweis, S., Frey, B.J., Emili, A., Greenblatt, J.F., Hughes, T.R.,
2003. A panoramic view of yeast noncoding RNA processing. Cell 113, 919e933.
Pertschy, B., Saveanu, C., Zisser, G., Lebreton, A., Tengg, M., Jacquier, A., Liebminger, E.,
Nobis, B., Kappel, L., Van Der Klei, I., Hogenauer, G., Fromont-Racine, M.,
Bergler, H., 2007. Cytoplasmic recycling of 60S preribosomal factors depends on the
AAA protein Drg1. Mol. Cell Biol. 27, 6581e6592.
Pertschy, B., Schneider, C., Gnadig, M., Schafer, T., Tollervey, D., Hurt, E., 2009. RNA
helicase Prp43 and its co-factor Pfa1 promote 20 to 18 S rRNA processing catalyzed
by the endonuclease Nob1. J. Biol. Chem. 284, 35079e35091.
Rabl, J., Leibundgut, M., Ataide, S.F., Haag, A., Ban, N., 2011. Crystal structure of the
eukaryotic 40S ribosomal subunit in complex with initiation factor 1. Science 331,
730e736.
Ribbeck, K., Lipowsky, G., Kent, H.M., Stewart, M., Gorlich, D., 1998. NTF2 mediates
nuclear import of Ran. EMBO J. 17, 6587e6598.
Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., Seraphin, B., 1999. A generic
protein purification method for protein complex characterization and proteome
exploration. Nat. Biotechnol. 17, 1030e1032.
Rout, M.P., Blobel, G., Aitchison, J.D., 1997. A distinct nuclear import pathway used by
ribosomal proteins. Cell 89, 715e725.
Russell, D.W., Spremulli, L.L., 1979. Purification and characterization of a ribosome disso-
ciation factor (eukaryotic initiation factor 6) from wheat germ. J. Biol. Chem. 254,
8796e8800.
Santos-Rosa, H., Moreno, H., Simos, G., Segref, A., Fahrenkrog, B., Pante, N., Hurt, E.,
1998. Nuclear mRNA export requires complex formation between Mex67p and
Mtr2p at the nuclear pores. Mol. Cell Biol. 18, 6826e6838.
Saveanu, C., Namane, A., Gleizes, P.E., Lebreton, A., Rousselle, J.C., Noaillac-Depeyre, J.,
Gas, N., Jacquier, A., Fromont-Racine, M., 2003. Sequential protein association with
nascent 60S ribosomal particles. Mol. Cell Biol. 23, 4449e4460.
Schafer, T., Maco, B., Petfalski, E., Tollervey, D., Bottcher, B., Aebi, U., Hurt, E., 2006.
Hrr25-dependent phosphorylation state regulates organization of the pre-40S subunit.
Nature 441, 651e655.
Schafer, T., Strauss, D., Petfalski, E., Tollervey, D., Hurt, E., 2003. The path from nucleolar
90S to cytoplasmic 40S pre-ribosomes. EMBO J. 22, 1370e1380.
Schlenstedt, G., Smirnova, E., Deane, R., Solsbacher, J., Kutay, U., Gorlich, D.,
Ponstingl, H., Bischoff, F.R., 1997. Yrb4p, a yeast ran-GTP-binding protein
involved in import of ribosomal protein L25 into the nucleus. EMBO J. 16,
6237e6249.
Sch€utz, S., Fischer, U., Altvater, M., Nerurkar, P., Pena, C., Gerber, M., Chang, Y.,
Caesar, S., Schubert, O.T., Schlenstedt, G., Panse, V.G., 2014. A RanGTP-independent
mechanism allows ribosomal protein nuclear import for ribosome assembly. eLife 3,
e03473.
Segref, A., Sharma, K., Doye, V., Hellwig, A., Huber, J., Luhrmann, R., Hurt, E., 1997.
Mex67p, a novel factor for nuclear mRNA export, binds to both poly(A)þ RNA and
nuclear pores. EMBO J. 16, 3256e3271.
Seiser, R.M., Sundberg, A.E., Wollam, B.J., Zobel-Thropp, P., Baldwin, K., Spector, M.D.,
Lycan, D.E., 2006. Ltv1 is required for efficient nuclear export of the ribosomal small
subunit in Saccharomyces cerevisiae. Genetics 174, 679e691.
Senapin, S., Clark-Walker, G.D., Chen, X.J., Seraphin, B., Daugeron, M.C., 2003. RRP20,
a component of the 90S preribosome, is required for pre-18S rRNA processing in Saccha-
romyces cerevisiae. Nucleic Acids Res. 31, 2524e2533.
Senay, C., Ferrari, P., Rocher, C., Rieger, K.J., Winter, J., Platel, D., Bourne, Y., 2003. The
Mtr2-Mex67 NTF2-like domain complex. Structural insights into a dual role of Mtr2
for yeast nuclear export. J. Biol. Chem. 278, 48395e48403.
Senger, B., Lafontaine, D.L., Graindorge, J.S., Gadal, O., Camasses, A., Sanni, A.,
Garnier, J.M., Breitenbach, M., Hurt, E., Fasiolo, F., 2001. The nucle(ol)ar Tif6p and
Efl1p are required for a late cytoplasmic step of ribosome synthesis. Mol. Cell 8,
1363e1373.
Sengupta, J., Bussiere, C., Pallesen, J., West, M., Johnson, A.W., Frank, J., 2010. Character-
ization of the nuclear export adaptor protein Nmd3 in association with the 60S ribosomal
subunit. J. Cell Biol. 189, 1079e1086.
Stade, K., Ford, C.S., Guthrie, C., Weis, K., 1997. Exportin 1 (Crm1p) is an essential nuclear
export factor. Cell 90, 1041e1050.
Stage-Zimmermann, T., Schmidt, U., Silver, P.A., 2000. Factors affecting nuclear export of
the 60S ribosomal subunit in vivo. Mol. Biol. Cell 11, 3777e3789.
Strasser, K., Bassler, J., Hurt, E., 2000. Binding of the Mex67p/Mtr2p heterodimer to FXFG,
GLFG, and FG repeat nucleoporins is essential for nuclear mRNA export. J. Cell Biol.
150, 695e706.
Strawn, L.A., Shen, T., Wente, S.R., 2001. The GLFG regions of Nup116p and Nup100p
serve as binding sites for both Kap95p and Mex67p at the nuclear pore complex. J. Biol.
Chem. 276, 6445e6452.
Strunk, B.S., Loucks, C.R., Su, M., Vashisth, H., Cheng, S., Schilling, J., Brooks 3rd, C.L.,
Karbstein, K., Skiniotis, G., 2011. Ribosome assembly factors prevent premature trans-
lation initiation by 40S assembly intermediates. Science 333, 1449e1453.
Strunk, B.S., Novak, M.N., Young, C.L., Karbstein, K., 2012. A translation-like cycle is a
quality control checkpoint for maturing 40S ribosome subunits. Cell 150, 111e121.
Stutz, F., Bachi, A., Doerks, T., Braun, I.C., Seraphin, B., Wilm, M., Bork, P., Izaurralde, E.,
2000. REF, an evolutionary conserved family of hnRNP-like proteins, interacts with
TAP/Mex67p and participates in mRNA nuclear export. RNA 6, 638e650.
Sydorskyy, Y., Dilworth, D.J., Yi, E.C., Goodlett, D.R., Wozniak, R.W., Aitchison, J.D.,
2003. Intersection of the Kap123p-mediated nuclear import and ribosome export
pathways. Mol. Cell Biol. 23, 2042e2054.
Taura, T., Krebber, H., Silver, P.A., 1998. A member of the Ran-binding protein family,
Yrb2p, is involved in nuclear protein export. Proc. Natl. Acad. Sci. U.S.A. 95,
7427e7432.
Thorsness, P.E., White, K.H., Ong, W.C., 1993. AFG2, an essential gene in yeast, encodes a
new member of the Sec18p, Pas1p, Cdc48p, TBP-1 family of putative ATPases. Yeast 9,
1267e1271.
Trapman, J., Retel, J., Planta, R.J., 1975. Ribosomal precursor particles from yeast. Exp. Cell
Res. 90, 95e104.
Tuck, A.C., Tollervey, D., 2013. A transcriptome-wide atlas of RNP composition reveals
diverse classes of mRNAs and lncRNAs. Cell 154, 996e1009.
Turowski, T.W., Lebaron, S., Zhang, E., Peil, L., Dudnakova, T., Petfalski, E.,
Granneman, S., Rappsilber, J., Tollervey, D., 2014. Rio1 mediates ATP-dependent final
maturation of 40S ribosomal subunits. Nucleic Acids Res. 42, 12189e12199.
Udem, S.A., Warner, J.R., 1972. Ribosomal RNA synthesis in Saccharomyces cerevisiae. J. Mol.
Biol. 65, 227e242.
Ulbrich, C., Diepholz, M., Bassler, J., Kressler, D., Pertschy, B., Galani, K., Bottcher, B.,
Hurt, E., 2009. Mechanochemical removal of ribosome biogenesis factors from nascent
60S ribosomal subunits. Cell 138, 911e922.
Valenzuela, D.M., Chaudhuri, A., Maitra, U., 1982. Eukaryotic ribosomal subunit anti-
association activity of calf liver is contained in a single polypeptide chain protein of
Mr ¼ 25,500 (eukaryotic initiation factor 6). J. Biol. Chem. 257, 7712e7719.
Vanrobays, E., Gelugne, J.P., Caizergues-Ferrer, M., Lafontaine, D.L., 2004. Dim2p, a KH-
domain protein required for small ribosomal subunit synthesis. RNA 10, 645e656.
Vanrobays, E., Gelugne, J.P., Gleizes, P.E., Caizergues-Ferrer, M., 2003. Late cytoplasmic
maturation of the small ribosomal subunit requires RIO proteins in Saccharomyces
cerevisiae. Mol. Cell Biol. 23, 2083e2095.
Venema, J., Tollervey, D., 1995. Processing of pre-ribosomal RNA in Saccharomyces cerevisiae.
Yeast 11, 1629e1650.
Warner, J.R., 1971. The assembly of ribosomes in yeast. J. Biol. Chem. 246, 447e454.
Warner, J.R., 1999. The economics of ribosome biosynthesis in yeast. Trends Biochem. Sci.
24, 437e440.
Weis, K., 2002. Nucleocytoplasmic transport: cargo trafficking across the border. Curr. Opin.
Cell Biol. 14, 328e335.
Wendler, F., Bergler, H., Prutej, K., Jungwirth, H., Zisser, G., Kuchler, K., Hogenauer, G.,
1997. Diazaborine resistance in the yeast Saccharomyces cerevisiae reveals a link between
YAP1 and the pleiotropic drug resistance genes PDR1 and PDR3. J. Biol. Chem. 272,
27091e27098.
West, M., Hedges, J.B., Chen, A., Johnson, A.W., 2005. Defining the order in which
Nmd3p and Rpl10p load onto nascent 60S ribosomal subunits. Mol. Cell Biol. 25,
3802e3813.
Windgassen, M., Sturm, D., Cajigas, I.J., Gonzalez, C.I., Seedorf, M., Bastians, H.,
Krebber, H., 2004. Yeast shuttling SR proteins Npl3p, Gbp2p, and Hrb1p are part of
the translating mRNPs, and Npl3p can function as a translational repressor. Mol. Cell
Biol. 24, 10479e10491.
Wong, C.C., Traynor, D., Basse, N., Kay, R.R., Warren, A.J., 2011. Defective ribosome
assembly in Shwachman-Diamond syndrome. Blood 118, 4305e4312.
Woolford Jr., J.L., Baserga, S.J., 2013. Ribosome biogenesis in the yeast Saccharomyces
cerevisiae. Genetics 195, 643e681.
Yao, W., Lutzmann, M., Hurt, E., 2008. A versatile interaction platform on the Mex67-Mtr2
receptor creates an overlap between mRNA and ribosome export. EMBO J. 27, 6e16.
Yao, W., Roser, D., Kohler, A., Bradatsch, B., Bassler, J., Hurt, E., 2007. Nuclear export of
ribosomal 60S subunits by the general mRNA export receptor Mex67-Mtr2. Mol. Cell
26, 51e62.
Yao, Y., Demoinet, E., Saveanu, C., Lenormand, P., Jacquier, A., Fromont-Racine, M.,
2010. Ecm1 is a new pre-ribosomal factor involved in pre-60S particle export. RNA
16, 1007e1017.
Zakalskiy, A., Hogenauer, G., Ishikawa, T., Wehrschutz-Sigl, E., Wendler, F., Teis, D.,
Zisser, G., Steven, A.C., Bergler, H., 2002. Structural and enzymatic properties of the
AAA protein Drg1p from Saccharomyces cerevisiae. Decoupling of intracellular function
from ATPase activity and hexamerization. J. Biol. Chem. 277, 26788e26795.
Zemp, I., Kutay, U., 2007. Nuclear export and cytoplasmic maturation of ribosomal
subunits. FEBS Lett. 581, 2783e2793.
Zemp, I., Wandrey, F., Rao, S., Ashiono, C., Wyler, E., Montellese, C., Kutay, U., 2014.
CK1delta and CK1epsilon are components of human 40S subunit precursors required for
cytoplasmic 40S maturation. J. Cell Sci. 127, 1242e1253.
Zemp, I., Wild, T., O’Donohue, M.F., Wandrey, F., Widmann, B., Gleizes, P.E., Kutay, U.,
2009. Distinct cytoplasmic maturation steps of 40S ribosomal subunit precursors require
hRio2. J. Cell Biol. 185, 1167e1180.
Zhang, J., Harnpicharnchai, P., Jakovljevic, J., Tang, L., Guo, Y., Oeffinger, M.,
Rout, M.P., Hiley, S.L., Hughes, T., Woolford Jr., J.L., 2007. Assembly factors Rpf2
and Rrs1 recruit 5S rRNA and ribosomal proteins rpL5 and rpL11 into nascent
ribosomes. Genes Dev. 21, 2580e2592.
CHAPTER FOUR
Transmembrane 4 L Six Family

Member 5 (TM4SF5)-Mediated
EpithelialeMesenchymal
Transition in Liver Diseases
Jung Weon Lee
Department of Pharmacy, Research Institute of Pharmaceutical Sciences, Tumor Microenvironment Global
Core Research Center, Medicinal Bioconvergence Research Center, College of Pharmacy, Seoul National
University, Seoul, Korea
E-mail: jwl@snu.ac.kr
Contents
1. Introduction 142
2. TM4SF5 as Component of Tetraspanin-Enriched Microdomains 142
3. TM4SF5-Mediated EpithelialeMesenchymal Transition 143
3.1 TM4SF5-Mediated Development of Muscle Cells in Zebrafish 145
3.2 TM4SF5-Mediated Liver Fibrosis 146
3.2.1 Regulation of TM4SF5 expression by TGFb1 signaling 147
3.2.2 Cross-talk between tetraspanins 148
4. TM4SF5-Mediated Metastatic Potential 150
4.1 TM4SF5-Mediated FAK Activation for Direct Migration 150
4.2 TM4SF5-Mediated c-Src Regulation of Invasion 152
5. TM4SF5-Dependent Drug Resistance 153
6. TM4SF5-Dependent Self-Renewal Capacity 156
7. Conclusion 157
Acknowledgments 158
References 158
Abstract
The membrane protein TM4SF5, a member of the transmembrane 4L six family, forms a
tetraspanin-enriched microdomain (TEM) on the cell surface, where many different mem-
brane proteins and receptors form a massive proteineprotein complex to regulate
cellular functions including transdifferentiation, migration, and invasion. We recently re-
ported that TM4SF5 causes epithelialemesenchymal transition (EMT), eventually contrib-
uting to aberrant multilayer cellular growth, drug resistance, enhanced migration,
invasion, its circulation in the blood, tumor initiation for successful metastasis, and muscle
development in zebrafish. In this review, I summarize the information on the role of
TM4SF5 in EMT-related functions at TM4SF5-enriched microdomain (T5EM) on cell sur-
face, where proteins such as TM4SF5, CD151, CD44, integrins, and epidermal growth
j
ISSN 1937-6448
142 Jung Weon Lee
factor receptor (EGFR) can form numerous protein complexes. TM4SF5-mediated EMT
contributes to diverse cellular functions, leading to fibrotic phenotypes and initiating
and maintaining tumors in primary and/or metastatic regions, in addition to its role in
muscle development in zebrafish. Anti-TM4SF5 strategies for addressing the protein
networks can lead to regulation of the fibrotic, tumorigenic, and tumor-maintaining func-
tions of TM4SF5-positive hepatic cells. This review is for us to (re)consider the antifibrotic
or antitumorigenic (i.e., anti-EMT-related diseases) strategies of dealing with protein net-
works that would be involved in cross-talks to regulate various cellular functions during
TM4SF5-dependent progression from fibrotic to cancerous hepatic cells.
1. INTRODUCTION
The plasma membrane is a structure that is fundamental to the trans-
duction of signals between the intracellular and extracellular environment.
Diverse membrane proteins laterally diffuse and functionally interact with
each other for signal transduction (Yanez-Mo et al., 2009). The organization
of membrane proteins is dynamic within certain compartmentalized local
areas called membrane (micro)domains. The tetraspanin-enriched microdo-
main (TEM or tetraspanin web) has an independent organization and plays
roles in regulating cellular functions, through numerous proteineprotein in-
teractions between tetraspanins and other membrane receptors including
integrins and growth factor receptors, contributing to adhesion, prolifera-
tion, and migration (Detchokul et al., 2013; Lee et al., 2011).
A member of the tetraspanin family, transmembrane 4 L six family mem-
ber 5 (TM4SF5), is expected to form TM4SF5-enriched microdomains
(T5EMs) in regulating cellular functions. In this review, TM4SF5, which in-
teracts with other membrane proteins including integrins (Lee et al., 2011), is
discussed in regard to its roles in biological functions based on its induction of
epithelialemesenchymal transition (EMT) (Figure 1).
2. TM4SF5 AS COMPONENT OF TETRASPANIN-

ENRICHED MICRODOMAINS
TM4SF5 together with TM4SF1 (L6, L6-Ag), TM4SF4 (IL-TIMP),
TM4SF18 (L6D), TM4SF19, and TM4SF20 forms transmembrane 4 L six
family (Lee et al., 2011; Wright et al., 2000) (Figure 2). The transmembrane
4 L six family is very similar to genuine tetraspanins (TM4SFs) in terms of
membrane topology with four transmembrane domains, two short cytosolic
N- and C-terminal tails, and two extracellular loops (short extracellular loop 1
Biological Roles of Tetraspanin TM4SF5 143
E-Cad
TM4SF5
β -actin
Neg Pos
SNU449
Figure 1 TM4SF5 expression reduced mRNA for E-cadherin (E-cad).
(SEL) and long extracellular loop 2 (LEL)), but the members show slight dif-
ferences in their LELs. The transmembrane 4 L six family has relatively var-
iable LELs, whereas genuine TM4SFs have a conserved region and a variable
region, including a CCG moiety and four conserved cysteine residues in the
LEL (Detchokul et al., 2013; Veenbergen and van Spriel, 2011). There have
been 33 reported mammalian tetraspanins (TM4SFs) and they are small pro-
teins of 20e30 kDa (Yanez-Mo et al., 2009).
At TEMs, tetraspanins form large protein complexes homophilically or
heterophilically with tetraspanins, integrins, and growth factor receptors
(Berditchevski, 2001). Tetraspanins in TEM have interactions at three different
levels: the primary interactions between tetraspanins and nontetraspanin
(binding) partners, secondary interactions between tetraspanins (Rubinstein,
2011), and tertiary interactions between tetraspanins and cytosolic signaling
molecules (Hemler, 2005; Yanez-Mo et al., 2009). These associations are
important biological phenomena in the plasma membrane during the cellular
activities of living cells, such as cell adhesion, migration, and invasion (Yanez-
Mo et al., 2009). As a member of the transmembrane 4 L six family similar to
genuine tetraspanins, TM4SF5 appears to form a T5EM. TM4SF5 was shown
to interact with integrins a2b1 (Lee et al., 2006) and a5 (Choi et al., 2009),
epidermal growth factor (EGF) receptor (EGFR) (Lee et al., 2012), transform-
ing growth factor b receptor (TGFbR) (Kang et al., 2012a), and CD151 (Kang
et al., 2014). Therefore, TM4SF5 may play important roles in diverse cellular
functions via its involvement in a T5EM.
3. TM4SF5-MEDIATED EPITHELIALeMESENCHYMAL
TRANSITION
Tetraspanins are correlated not only with the progression of a variety
of cancers (Sala-Valdes et al., 2012) but also with development (Kashef et al.,
(A) TM4SF1 (B)
144
Phylogenetic tree (distance)
--------MCYGKCARCIGHSLVGLALLCIAANIL-LYFPNGETKYASENHLSRFVWFFS 51
TM4SF18 --------MGSRKCGGCLSCLLIPLALWSIIVNIL-LYFPNGQTSYASSNKLTNYVWYFE 51
TM4SF4 --------MCTGGCARCLGGTLIPLAFFGFLANIL-LFFPGGKVIDDN-DHLSQEIWFFG 50
TM4SF5 --------MCTGKCARCVGLSLITLCLVCIVANAL-LLVPNGETSWTNTNHLSLQVWLMG 51
TM4SF19 MVSSPCTPASSRTCSRILGLSLGTAALFAAGANVA-LLLPNWDVTYLLRGLLGRHAMLGT 59
TM4SF20 -------MTCCEGWTSCNGFSLLVLLLLGVVLNVIPLIVSLVEEDQFSQNPISCFEWWFP 53
. * : * * .. . . :
TM4SF1 GIVGGGLLMLLPAFVFIGLEQDDCCGCCGHENCGKRCAMLSSVLAALIGIAGSGYCVIVA 111
TM4SF18 GICFSGIMMLIVTTVLLVLENNNNYKCCQSENCSKKYVTLLSIIFSSLGIAFSGYCLVIS 111
TM4SF4 GILGSGVLMIFPALVFLGLKNNDCCGCCGNEGCGKRFAMFTSTIFAVVGFLGAGYSFIIS 110
TM4SF5 GFIGGGLMVLCPGIAAVRAGGK---GCCGAGCCGNRCRMLRSVFSSAFGVLGAIYCLSVS 108
TM4SF19 GLWGGGLMVLTAAILISLMGWR---YGCFS-KSGLCRSVLTALLSGGLALLGALICFVTS 115
TM4SF20 GIIGAGLMAIPATTMSLTARKR--------ACCNNRTGMFLSSFFSVITVIGALYCMLIS 105
*: .*:: : .. : : : . . . : .. :
TM4SF1 ALGLAEGPLCLDSLG--------QWNYTFASTEG---QYLLDTSTWS-ECTEPK------ 153
TM4SF5:0.28078
TM4SF1:0.23685
TM4SF18 ALGLVQGPYCR-TLD--------GWEYAFEGTAG---RFLTDSSIWI-QCLEPA------ 152
TM4SF4:0.27044
TM4SF4 AISINKGPKCLMANS--------TWGYPFH--DG---DYLNDEALWN-KCREPL------ 150
TM4SF5 GAGLRNGPRCLMNG---------EWGYHFEDTAG---AYLLNRTLWD-RCEAPP------ 149
TM4SF19 GVALKDGPFCMFDVSSFNQTQAWKYGYPFKDLHSR--NYLYDRSLWNSVCLEPS------ 167
TM4SF19:0.36632
TM4SF18:0.31315
TM4SF20 IQALLKGPLMCNSPSNSN----ANCEFSLKNISDIHPESFNLQWFFNDSCAPPTGFNKPT 161
.: .** : : . : : * *
TM4SF1 ---------------------HIVEWNVSLFSILLALGGIEFILCLIQVINGVLGGICG- 191
TM4SF20:0.43160
TM4SF18 ---------------------HVVEWNIILFSILITLSGLQVIICLIRVVMQLSKILCGS 191
TM4SF4 ---------------------NVVPWNLTLFSILLVVGGIQMVLCAIQVVNGLLGTLCGD 189
TM4SF5 ---------------------RVVPWNVTLFSLLVAASCLEIVLCGIQLVNATIGVFCGD 188
TM4SF19 ---------------------AAVVWHVSLFSALLCISLLQLLLVVVHVINSLLGLFCSL 206
TM4SF2O SNDTMASGWRASSFHFDSEENKHRLIHFSVFLGLLLVGILEVLFGLSQIVIGFLGCLCGV 221
:. :* *: . ::.:: ::: :*.
TM4SF1 FCCSHQQQMTAKRTNPGQSHNLPLFHCNLYISLVFICKTLY 232
TM4SF18 YSVIFQPGII------------------------------- 201
TM4SF4 CQCCGCCGGDGPV---------------------------- 202
Jung Weon Lee

TM4SF5 CRKKQDTPH-------------------------------- 197
TM4SF19 CEK-------------------------------------- 209
TM4SF20 SKRRSQIV--------------------------------- 229
Figure 2 Blast alignment of the TM4SF members (A) and their phylogenetic tree (B).
2013). EMT is involved in diverse biological functions, including develop-

ment, organ fibrosis, and cancer metastasis, that are related to the morpho-
logical adaptability and migratory capacities of cells (Cannito et al., 2010).
The expression of TM4SF5 in hepatocytes causes EMT (Lee et al., 2008).
Thus, it is reasonable to assume that TM4SF5 may play important roles in
different biological functions that depend on EMT.
TM4SF5 expression decreases the expression of E-cadherin (Figure 1),
and zonula occludens 1 (ZO1) and increases the expression of a-smooth
muscle actin (a-SMA) leading to the loss of cellecell contacts. Depending
on cytosolic p27Kip1-mediated Ras homolog gene family, member A
(RhoA) inactivation, and morphological changes, the TM4SF5-mediated
EMT is blocked by the suppression of TM4SF5 or p27Kip1 (Lee et al.,
2008). Additionally, hepatocyte growth factor (HGF)-mediated EMT of
endogenous TM4SF5-expressing hepatocytes was also blocked by
TM4SF5 suppression (Lee et al., 2008). TGFb1 could induce TM4SF5
expression and the subsequent EMT of normal hepatocytes, presumably
leading to transdifferentiation to activated myofibroblasts. Treatment with
conditioned media from activated hepatic stellate cell (HSC) cultures
(LX2 cells) also caused TM4SF5 expression in normal hepatocytes, leading
to the acquisition of mesenchymal features for EMT, such as cell scattering
and a-SMA expression (Kang et al., 2012a). TM4SF5-mediated EMT leads
to persistent proliferation even under confluent conditions (Lee et al., 2008)
and accelerates S-phase entry (Kim et al., 2010). In addition to an aberrant
proliferation pattern, TM4SF5-mediated EMT was shown to play impor-
tant roles in diverse EMT-related cellular functions, as explained below.
3.1 TM4SF5-Mediated Development of Muscle Cells in

Zebrafish
EMT plays roles in homeostatic physiological processes such as develop-
mental embryogenesis (Cannito et al., 2010). It is also plausible that
TM4SF5 plays a role in embryogenesis by regulating cellular migration
via the induction of EMT, because the correct migration patterns are critical
for the control of development in multicellular organisms (Kenneth and
Duckett, 2012). We recently observed that tm4sf5 was highly expressed in
somites and the myotome during somitogenesis of zebrafish. The knock-
down of tm4sf5 caused abnormal morphology of muscle fibers and impaired
muscle mass, leading to impaired development of the trunk and tail (Choi
et al., 2014).
146 Jung Weon Lee
TM4SF5 is conserved in zebrafish and other vertebrates and binds to

integrin a5 to regulate focal adhesion kinase (FAK)/c-Src activities for
gene regulations in hepatocytes (Choi et al., 2009). The cross-talk depend-
ing on a physical interaction between TM4SF5 and integrin a5 in zebrafish
may also be functional. The somite boundaries in zebrafish mature into the
V-shaped myotome boundaries through the accumulation of fibronectin
(FN) and focal adhesion (FA) components such as FAK, integrin, and pax-
illin (Crawford et al., 2003). In particular, the FN matrix is assembled
depending on integrin a5, which is a crucial component for boundary main-
tenance, through its functions in the epithelialization of somitic border cells
(Koshida et al., 2005).
fak messenger ribonucleic acid (mRNA) is expressed in developing so-
mites, and FAK localizes at intersegmental boundaries in zebrafish as the
myotome matures (Henry et al., 2001). Thus, TM4SF5 may function as a
regulator of myogenesis via its cooperation with integrin a5. Knockdown
of tm4sf5 further disturbed the localization of integrin a5 and disrupted
the localization of extracellular FN and intracellular FAK, vinculin, and actin
at the somite boundaries. The effect was abolished by the concomitant
introduction of zebrafish tm4sf5 complementary deoxyribonucleic acid
(cDNA) (Koshida et al., 2005). tm4sf5 mRNA expression was elevated dur-
ing somitogenesis and was also detected in developing somites, such as fak
mRNA. A role for tm4sf5 in developmental regulation during somite and
myotome formation could be via its interaction with FAK (Jung et al.,
2012) or integrin a5 (Choi et al., 2009).
Because the proper regulation of cell migration and EMT is crucial for
normal development during embryonic morphogenesis (Moustakas and
Heldin, 2007), it was inferred that zebrafish tm4sf5 might also play an impor-
tant role as a regulator in developmental processes via its involvement in the
proper translocation (migration) of muscle progenitor cells toward and celle
cell contacts around somite boundaries, presumably via the interaction with
signaling molecules such as FAK, integrin receptors, and growth factor re-
ceptors (Jung et al., 2012; Kang et al., 2012a; Lee et al., 2006, 2008, 2010).
3.2 TM4SF5-Mediated Liver Fibrosis

TM4SF5, as an inducer of the EMT process, can also be involved in organ
fibrosis. Additionally, because hepatocellular carcinoma can be driven from
liver fibrosis/cirrhosis (Severi et al., 2010), it is reasonable to hypothesize
that TM4SF5 overexpression in liver cancer tissues (Lee et al., 2008) could
be involved in liver fibrosis. TM4SF5 expression is induced during
CCl4-mediated mouse liver injury with fibrotic phenotypes, which includes

a-SMA expression and collagen I deposition (Kang et al., 2012b) supported
by signaling activities of the TGFb1- and EGFR-mediated signaling pathways
(Kang et al., 2012a). TM4SF5 expression correlates with a-SMA expression
in CCl4-administered mice livers, possibly indicating an EMT process and/or
HSC activation (Zhong et al., 2009). During liver fibrosis following CCl4
administration to mice, TGFb1-mediated EMT results in the activation of
HSCs and myofibroblasts, which are marked by enhanced a-SMA expres-
sion. CCl4-treated livers show TM4SF5 expression along the fibrotic septa,
where the localization of collagen I and a-SMA expressions is apparent
(Gressner and Weiskirchen, 2006; Meindl-Beinker and Dooley, 2008).
Thus TM4SF5 may be involved in the process of activating HSCs or myofi-
broblasts, which eventually produces and deposits collagen I for fibrosis (Kang
et al., 2012b). These liver injury and fibrotic phenotypes, including collagen I
deposition, a-SMA expression, and fibrotic septa formation in CCl4-admin-
istered mouse livers, are attenuated by further injection of the anti-TM4SF5
reagent, 4’-(p-toluenesulfonylamido)-4-hydroxychalcone (TSAHC) (Kang
et al., 2012b; Lee et al., 2009). Therefore, TM4SF5 is an important player
during TGFb1- and soluble factor-mediated activation of myofibroblasts dur-
ing the development of liver fibrosis.
3.2.1 Regulation of TM4SF5 expression by TGFb1 signaling

TM4SF5 is involved in liver fibrosis and tumorigenesis, presumably through the
regulation of cellular morphology and migration via the EMT process. During
such biological processes, cytokines such as TGFb1 and growth factors may play
important roles. TM4SF5 is induced during the development of liver fibrotic
phenotypes in mice administered with CCl4 (Kang et al., 2012b); additionally,
clinical hepatocellular carcinoma tissues showed overexpression of TM4SF5
compared with their counterpart normal liver tissues (Lee et al., 2008).
Thus, soluble factors may regulate TM4SF5 expression. As expected from
the multifunctional cytokine TGFb1 being involved in fibrosis and tumorigen-
esis via the induction of EMT (Reichl et al., 2012), TM4SF5 is regulated by
the TGFb1-related signaling pathway (Kang et al., 2012a). TGFb1 induces
TM4SF5 expression and consequently acquires mesenchymal features in
TM4SF5-expressing hepatocytes via an activated Mothers against DPP homo-
logs (SMAD) mediated cross-talk with the EGFR/extracellular regulated
MAP kinase (Erk) signaling pathway. These effects are abolished by the inhi-
bition of TGFb1-mediated receptor-regulated SMAD (R-SMAD) activity,
EGFR activity, or TM4SF5 function (Kang et al., 2012a).
148 Jung Weon Lee
A signaling link between activated SMADs and EGFR resulting

in TM4SF5 induction is reminiscent of a signaling link between Ser/Thr
kinase and Tyr kinase signaling pathways. SMAD4 overexpression in Chang
cells does not increase EGFR expression levels, although its overexpression
enhances EGFR/Erk2 phosphorylation and TM4SF5 expression even
without TGFb1 treatment. Consistently, the livers of TM4SF5-
overexpressing transgenic mice showed enhanced SMAD2/3 and EGFR
phosphorylation, indicating a close connection between TGFb1 and
EGFR signaling activation that is relevant to TM4SF5 expression (Kang
et al., 2012a). Activated SMADs-mediated EGFR activation for TM4SF5
induction does not involve de novo transcription of EGFR or recycling
of internalized EGFR. However, activated SMADs-mediated retention of
EGFR on the membrane surface may be involved in TGFb1-mediated
TM4SF5 expression. TGFb1 also enhances EGFR surface expression on
normal rat kidney fibroblasts (Assoian et al., 1984). Alternatively, the metal-
loprotease ADAM metallopeptidase domain 17 (ADAM 17) may be stimu-
lated by TGFb1 treatment to rapidly induce the shedding of EGF family
member(s) (Murillo et al., 2005), and so TGFb1 treatment may be linked
to EGFR activation. Also, it cannot be ruled out that additional molecule(s)
may directly transduce the signaling from the activated SMADs to EGFR via
proteineprotein interactions, leading to TM4SF5 induction.
3.2.2 Cross-talk between tetraspanins

Tetraspanins located in TEMs play roles in adhesion, migration, and invasion
via homophilic and/or heterophilic interactions among tetraspanins, integrins,
and growth factor receptors (Hemler, 2005). Among the 33 genuine tetraspa-
nins, CD9, CD63, and CD151 have been studied to evaluate their relation-
ship with TM4SF5, although there are possible correlations to other TM4SF
members. Although TM4SF5 does not belong to the genuine tetraspanin
family (Wright et al., 2000), TM4SF5 can form TEMs with other tetraspanins
to play critical roles in the regulation of metastatic potentials. TM4SF5 was
shown to interact with integrins a2b1 (Lee et al., 2006) and a5 (Choi
et al., 2009), EGFR (Lee et al., 2012), and TGFbR (Kang et al., 2012a). In
addition to these binding partners, TM4SF5 was shown to bind tumorigenic
CD151, but not tumor-suppressive CD63 (Kang et al., 2014), supporting
roles for TM4SF5 in tumor progression as a tetraspanin molecule presumably
at T5EMs (Figure 3).
CD151 (Tspan24) was first identified as a promoter of metastasis (Testa
et al., 1999) and was positively correlated with metastatic potential (Ke et al.,
Figure 3 A working model for cellular functions depending on EMT supported by

TM4SF5 at TM4SF5-enriched microdomains (i.e, T5EM). TM4SF5 forms T5EM, while regu-
lating cellular functions via formation of protein complexes with tetraspanin CD151,
integrins, CD44, and other membrane receptors. hKIS, kinase interacting with stathmin;
Akt, v-akt murine thymoma viral oncogene homolog 1; ECM, extracellular matrix; GRAF,
GTPase regulator associated with FAK.
2009, 2011). Tetraspanin CD151 is widely expressed in different cell types

including platelets and epithelial cells, endothelial cells, dendritic cells, and
muscle cells, and the expression of CD151 is increased in liver cancer,
compared with normal cells (Sincock et al., 1997). CD151 interacts with
integrins a3b1 (Yauch et al., 1998), a6b1 (Serru et al., 1999), a6b4 (Sterk
et al., 2000), and a7b1 (Sterk et al., 2002). Through forming and regulating
these protein complexes, CD151 plays roles in cellular migration and inva-
sion and also contributes to angiogenesis and drug resistance (Sala-Valdes
et al., 2012; Takeda et al., 2007; Yang et al., 2010). Thus, the interaction
between TM4SF5 and CD151 can synergistically promote the development
of tumorigenic phenotypes.
Unlike protumorigenic CD151, CD63 (Tspan30) is a tumor suppressor
that is expressed in the endosomes and lysosomes and at the cell surface (Pols
and Klumperman, 2009). CD63 interacts with integrins a4b1, a3b1, a6b1,
and b2, as well as CD81, CD82, CD9, and CD151 (Pols and Klumperman,
150 Jung Weon Lee
2009). CD63 is an abundantly expressed surface antigen in the early mela-

noma stage, but its level decreases during malignant progression, so a nega-
tive correlation exists between the CD63 surface levels and invasiveness
(Radford et al., 1997). CD63 at the membrane surface was shown to asso-
ciate with TM4SF1 (L6-Ag, another transmembrane 4 L six family member)
for cell migration, and CD63 localization at the membrane surface increases
following TM4SF1 suppression (Lekishvili et al., 2008). The regulation of
one TEM component thus controls other components of the TEM, leading
to the regulation of cell motility. Similar to TM4SF1, TM4SF5 expression
causes internalization of CD63, decreasing its level at the membrane surface
and disrupting its tumor-suppressing action (Kang et al., 2014). Therefore,
TM4SF5 seems to be similar to TM4SF1 in terms of regulating the CD63
level on the membrane surface.
Although TM4SF5, CD151, and CD63 can localize at the membrane sur-
face as well as at the endosomal membranes, their localization at the mem-
brane surface is critically regulated as part of their roles in tumorigenic
progression and fibrotic phenotype development in livers. TM4SF5 interacts
with CD151 but not with CD63. The interaction might involve other mem-
brane proteins including tetraspanins, integrins, or growth factor receptors. At
T5EM enriched with TM4SF5 (Figure 3), numerous interactions among
membrane proteins mediate intracellular signaling pathways for regulating
cellular functions, such as migration and EMT (Radisky et al., 2007).
4. TM4SF5-MEDIATED METASTATIC POTENTIAL

4.1 TM4SF5-Mediated FAK Activation for Direct
Migration
TM4SF5 mediates EMT (Lee et al., 2008) and leads to enhanced
migration and invasion (Jung et al., 2012; Lee et al., 2010). It is highly
expressed in hepatocarcinoma, and plays roles in aberrant cell proliferation
and enhanced metastatic potential (Lee et al., 2011). Therefore, TM4SF5
may transduce intracellular signal activations to regulate cellular behaviors.
TM4SF5 has short N- and C-termini in the cytosol, leading to interactions
with certain intracellular molecules to transduce signaling pathways and pro-
mote tumorigenic functions. In addition to binding membrane proteins, as
explained above, TM4SF5 also binds to FAK (Jung et al., 2012) and c-Src
family kinases (SFKs, (Jung et al., 2013)) to form a complex with different
actin-organizing molecules such as actin-related protein 2 (Arp2), neuronal
WiskotteAldrich Syndrome protein (N-WASP), and cortactin (Jung et al.,

2012). Therefore, TM4SF5 may activate FAK signaling activity, leading to
actin organization in the leading edges of a migratory cell.
Integrin engagement of the extracellular matrix causes activation and Tyr
phosphorylation of diverse signaling or adaptor molecules in FAs, including
FAK (Cox et al., 2006). FAK has the N-terminal four-point-one, ezrin,
radixin, moesin (FERM) domain with three lobes, a kinase domain, and a
C-terminal focal adhesion targeting domain (Lietha et al., 2007). An intramo-
lecular interaction between the FERM domain (F2 lobe) and the FAK kinase
domain (C-lobe centered on Phe596) buries the Tyr397 residue in the inactive
form; however, upon cell adhesion, this inhibitory interaction is interrupted,
allowing Tyr397 to be released and trans-autophosphorylated (Lietha et al.,
2007). Although membrane receptors that bind the FAK-FERM domain
have been reported, the resulting FAK activation was not shown to depend
on cell adhesion. The FAK-FERM domain binds phosphatidylinositol-4,5-
bisphosphate (Cai et al., 2008) and interacts with the c-Met receptor to mediate
FAK phosphorylation during HGF-enhanced invasion (Chen and Chen,
2006). FAK Tyr194 phosphorylation by c-Met interferes with the inhibitory
intramolecular interaction leading to its activation (Chen et al., 2011). The
FAK-FERM domain also binds EGFR to promote migration via the interme-
diary protein SRC3D4 [steroid receptor coactivator 3 (SRC3) splice form with
a deletion of exon 4] after EGF treatment (Long et al., 2010).
As a membrane protein, TM4SF5 appears to directly bind and activate
FAK during cell adhesion. The intracellular loop (ICL) of TM4SF5 directly
binds the F1 lobe of the FAK-FERM domain, which releases the inhibitory
intramolecular interaction in FAK, leading to autophosphorylation
(pY397FAK) and activation (pY577FAK) upon cell adhesion. The binding
between the F1 lobe and ICL induces secondary structural elements in
the ICL peptide inside the F1 lobe groove consisting of L37, L98, and
F111 residues; therefore L37A or L98A mutation in FAK results in dramat-
ically enhanced Tyr397, Tyr577 FAK, and Tyr118 paxillin phosphoryla-
tions (Jung et al., 2012). Similarly, another group reported that the K38A
mutation in the FAK-FERM F1 lobe activates FAK and promotes cell
motility (Cohen and Guan, 2005). The interaction between TM4SF5 and
FAK causes further FAK activation and actin polymerization at the leading
edges of migratory cells via complex formation with Arp2, N-WASP, and
cortactin (Jung et al., 2012), enhancing the metastatic potential.
TM4SF5/FAK interaction-mediated FAK activation depends on cell
adhesion but does not completely depend on integrins. It is currently
152 Jung Weon Lee
unknown how TM4SF5 achieves cell adhesion or which TM4SF5 ligands

are involved. TM4SF5-expressing cells plated on poly-L-lysine-coated
dishes or pretreated with integrin b1-neutralizing antibody before being
plated on FN-coated dishes showed lower FAK phosphorylation than
FN-adherent cells but higher phosphorylation than suspended TM4SF5-
expressing cells or FN-adherent TM4SF5-null cells (Jung et al., 2012).
Thus, it is likely that TM4SF5 alone can activate FAK at earlier adhesion
times, and as more time passes after adhesion, TM4SF5 may collaborate
with integrins for FAK activation, presumably via complex formation be-
tween TM4SF5, FAK, and integrins. It is also possible that TM4SF5 binds
and activates FAK at unique locations (e.g., T5EM) earlier in the adhesion
process, and then additional FAK is recruited and further activated at integ-
rin-clustered FAs in a TM4SF5- and integrin-dependent manner at a later
time during cell adhesion. The interaction between TM4SF5 and FAK leads
to FAK activation in a cell adhesion-dependent manner, thus recapitulating
a cell adhesion-dependent mechanism in which a membrane receptor acti-
vates FAK at the leading edges of migratory cells.
In addition to directional migration, TM4SF5-mediated FAK activity
can lead to an escape from immunological action by IL-6-mediated activa-
tion of signal transducer and activator of transcription (STAT) 3 in
TM4SF5-positive cancer cells (Ryu et al., 2014). To achieve this escape,
TM4SF5-positive cancer cells express negligible level of IL-6, whereas
normal hepatocytes with ectopic TM4SF5 expression express more IL-6,
which can downregulate the TM4SF5-FAK signaling activity (Ryu et al.,
2014). Furthermore, in TM4SF5-positive cancer cells, STAT3 activity is in-
dependent of IL-6, which clearly leads to the hypothesis that IL-6 could be
irrelevant to TM4SF5 signaling. IL-6-independent STAT3 activity in
TM4SF5-positive cancer cells can be highly enhanced for migration and in-
vasion (Ryu et al., 2014).
4.2 TM4SF5-Mediated c-Src Regulation of Invasion

The C-terminal tail of TM4SF5 directly binds and regulates c-Src (Jung
et al., 2013). Although the C-terminal tail of TM4SF5 has no specific pro-
tein subdomains, it interacts with c-Src regions rather than the SH2SH3
domain. Interestingly, the C-terminus of TM4SF5 binds inactive (i.e., non-
phosphorylated at Y416) or kinase-dead c-Src more efficiently than active or
Y416-phosphorylated c-Src, although active c-Src can still bind TM4SF5
(Jung et al., 2013). Additionally, the interaction between TM4SF5 and
SFKs were more efficient in suspended cells (where c-Src would generally
be inactive) than in adherent cells (Jung et al., 2013). Therefore, it is likely

that TM4SF5 recruits inactive c-Src to its C-terminus to activate it.
The regulation of c-Src activity via a direct interaction between c-Src
and TM4SF5 requires EGFR phosphorylation at Y845 but not Y992 or
Y1173, eventually causing effective formation of invasive protrusions
(Jung et al., 2013). Y845 phosphorylation of EGFR is a c-Src phosphoryla-
tion site even without ligand binding (Tice et al., 1999), whereas Y992 or
Y1173 phosphorylation is autophosphorylated by EGF binding (Foley
et al., 2010). Y845F EGFR transfection does not alter c-Src activity but in-
hibits the formation of invasive protrusions of TM4SF5 wild-type (WT)
cells, suggesting a signaling link from c-Src to EGFR for invasive protrusion
formation (Jung et al., 2013). With the capacity to interact physically with
other membrane receptors, TM4SF5 can dynamically regulate membrane
activities to form local protrusions or retractions, presumably depending
on differential microenvironments near the membrane boundaries.
TM4SF5 WT cells show pY845EGFR and invasive protrusions depending
on serum or EGF treatment, so TM4SF5 WT-expressing cells may activate
c-Src via cell adhesion and growth factor stimulation. It is currently being
explored how the short C-terminus of TM4SF5 mechanistically causes
the activation of c-Src.
5. TM4SF5-DEPENDENT DRUG RESISTANCE

Similar to the TGFbR and c-MET signaling pathways, the tetraspanin
TM4SF5 can induce EMT (Lee et al., 2008), which enhances not only cell
proliferation, migration, and invasion (Muschel and Gal, 2008) but also resis-
tance to EGFR kinase inhibitors or tyrosine kinase inhibitors (TKIs) like
gefitinib (Thomson et al., 2005). TKI-resistant cells may adapt an alternative
signaling pathway to bypass EGFR-dependent proliferation and survival
signaling pathways via integrative roles through other membrane receptors
such as tetraspanins or c-MET.
Thus far, there is evidence to suggest that gefitinib resistance of cancer
cells can be attributed to acquired mutations in EGFR (e.g., T790M) dur-
ing TKI therapy, which sterically hinders gefitinib and enhances the affin-
ity of mutated EGFR for adenosine triphosphate (Suda et al., 2009),
nullifying the hypersensitivity of activating EGFR mutations (Nguyen
et al., 2009). T790M mutation-mediated gefitinib resistance accounts for
50% of TKI resistance in patients carrying an EGFR mutation (Nguyen
154 Jung Weon Lee
et al., 2009). The resistance has also been attributed to the contribution of
other membrane receptors or their downstream effector(s). Gefitinib
resistance of A549 cells is caused by phosphatidylinositol 3-kinase (PI3K)
activation and the insulinlike growth factor 1 receptor (IGF1R) pathway
(Guix et al., 2008). c-Met amplification, which accounts for 20% of TKI-
resistant tumors (Nguyen et al., 2009), also results in gefitinib resistance
of non-small cell lung cancer (NSCLC) (Engelman et al., 2007). Further-
more, loss of molecules at cellecell contacts has recently been reported to
be a possible determinant of the sensitivity of NSCLC cells and xenografts
to erlotinib, another EGFR kinase inhibitor (Thomson et al., 2005). c-Met
as a receptor for HGF, is well known to cause cell scattering and EMT
(Naldini et al., 1991). Ectopic expression of TM4SF5 in NSCLC leads
to EMT and gefitinib resistance but does not accompany the T790M
EGFR mutation, so TM4SF5-mediated gefitinib resistance may be irrele-
vant to EGFR. Interestingly, TM4SF5, as a membrane protein that induces
EMT (Lee et al., 2008; Muller-Pillasch et al., 1998), causes gefitinib resis-
tance, and gefitinib-resistant cells with the T790M EGFR mutation (i.e.,
NCI-H1975 cells) enhance the expression of a-SMA and TM4SF5 (Lee
et al., 2012). Stable transfection of T790M EGFR into HCC827 cells leads
to gefitinib resistance and enhanced TM4SF5 expression and is correlated
with EMT phenotypes.
TM4SF5 expression in gefitinib-sensitive NSCLC cells results in EMT
phenotypes and gefitinib resistance, depending on cytosolic p27Kip1 (Lee
et al., 2012). The stabilization of cytosolic p27Kip1 is shown in diverse tumor
tissues (Chu et al., 2008), accounts for the inactivation of RhoA guanosine
triphosphatase via a direct interaction, and plays regulatory roles in actin
dynamics and cell migration (Besson et al., 2004, 2008). The suppression
of either TM4SF5 or p27Kip1 recovers E-cadherin expression at cellecell
contacts (Lee et al., 2008), and suppression of p27Kip1 in gefitinib-resistant
cells renders these cells sensitive to gefitinib (Lee et al., 2012). Therefore,
cytosolic p27Kip1 appears to be tumorigenic by being involved in
TM4SF5-mediated EMT and drug resistance, although nuclear p27Kip1
can be antitumorigenic as an inhibitor of cyclin-dependent kinases
(Coqueret, 2003).
TM4SF5 as a tetraspanin can form the T5EM in the cell plasma mem-
brane (Figure 3), which plays roles in organizing the integrity or surface
retention of membrane receptors, including integrins and growth factor
receptors, through numerous proteineprotein interactions similar to other
tetraspanins (Berditchevski, 2001; Yanez-Mo et al., 2009; Yang et al.,
2004). TM4SF5 and five other structurally similar proteins (TM4SF1, 4,

18, 19, and 20), which have overall sequence homology with the genuine
tetraspanins, could confer drug resistance to cancer (Sala-Valdes et al.,
2012). TM4SF1 is involved in the p23 (a heat shock protein 90)-mediated
drug resistance mechanism for etoposide and doxorubicin in human breast
adenocarcinoma MCF-7 cells overexpressing p23 (Carloni et al., 2013).
TM4SF5 further interacts with EGFR (Lee et al., 2012). Therefore,
TM4SF5-mediated EMT may confer TM4SF5-expressing cells with char-
acteristics that resemble the T790M mutation in terms of the binding affin-
ity of EGFR to TKIs. The TM4SF5-mediated regulation of membrane
receptor networks including IGF1R on the cell surface may also lead to
IGF1R-mediated gefitinib resistance (Guix et al., 2008). Integrin a5 binds
to TM4SF5 and is more efficiently retained on the surface of hepatocytes
that express TM4SF5 than on the surface of cells lacking TM4SF5 (Choi
et al., 2009). EGFR is also retained more on the membrane surface of
TM4SF5-mediated gefitinib-resistant cells (Lee et al., 2012). Similarly,
CD63 and CD82 are more likely to be found on the surface of cells that
overexpress L6-Ag (Lekishvili et al., 2008), another member of the trans-
membrane 4 L six family (Wright et al., 2000). Therefore, TM4SF5-medi-
ated effects, including EMT and membrane retention of EGFR via protein
interactions, may be involved in gefitinib resistance of cancer cells.
Tetraspanins such as CD9 (Tspan29), CD81 (Tspan28), CD82
(Tspan27), and CD151 (Tspan24) are expressed in various cancer cells and
tissues (Sala-Valdes et al., 2012). The expression levels of tetraspanin in
many different tumors can confer drug resistance, and the presence of tetra-
spanin correlates with resistance to anticancer drugs in several different
tumor types (Detchokul et al., 2013; Sala-Valdes et al., 2012). CD151 inter-
acts with laminin-binding integrins (a6b4, a3b1) and laminin-5 via FAK,
and these interactions are involved in the drug resistance of breast cancer
cells to Erb2 antagonists (Hemler, 2008; Yang et al., 2010). The expression
of CD9 in small cell lung cancer cells is resistant to cisplatin or etoposide
(Kohmo et al., 2010). Tetraspanin CD81/CD9 is involved in the regulation
of colon cancer cell fusion, whose occurrence in a metastatic model of colon
carcinoma causes the appearance of cells resistant to both 5-fluorouracil and
oxaliplatin (Simpson et al., 2010).
TM4SF5-mediated EMT and the regulation of the activity or integrity
of membrane receptors including tetraspanins, EGFR, IGF1R, and/or
c-Met on the cell surface, may likely be important for the drug resistance
of cancer cells.
156 Jung Weon Lee
6. TM4SF5-DEPENDENT SELF-RENEWAL CAPACITY

TM4SF5 in hepatocytes induces EMT (Lee et al., 2008) and causes
gefitinib resistance of NSCLC (Lee et al., 2012). Interestingly, stemness
(i.e., self-renewal capacity) of cancer cells is linked to EMT (Scheel and
Weinberg, 2012). Overexpression of the E-cadherin suppressor Snail or
Twist results in EMT, which renders stem cell properties to form mammo-
spheres with CD44high/CD24low expression (Mani et al., 2008). Cancer
stem cells (CSCs) are usually examined in terms of the capacity to form
spheroids with a lower adhesive environment, as well as aldehyde dehydro-
genase (ALDH) activity, markers, and tumor formation in xenografts with
fewer cell numbers. Interestingly, the expression of TM4SF5 in hepatocytes
including SNU449, SNU761, and HepG2 cells results in spheroid forma-
tion under a nonadhesive condition, and its suppression or mutation of
TM4SF5 at N-glycosylation residues blocks spheroid formation. ALDH
activities depend on TM4SF5 expression or suppression. Tumor formations
occur after xenograft injection of 500e5000 cells, and serial xenografts show
further aggressive tumor formation. Furthermore, TM4SF5-dependent
spheroids reveal no expression of CD24 and expression of CD44, resulting
in spheroids with the CD24/CD44þ/ALDHþ/TM4SF5þ phenotype.
CD44, a hyaluronan receptor, is also involved in reducing apoptosis and
promoting proliferation, adhesion, and migration (Naor et al., 1997), and
the expression of CD44 in the stroma of tumors can facilitate the recruit-
ment of monocytes and tumor-associated macrophages, which releases
soluble factors for angiogenesis and lymphangiogenesis (Schoppmann
et al., 2002), leading to the formation of a microenvironment favored by
tumor cells (Negi et al., 2012). CD44 is bound to TM4SF5 in TM4SF5-
positive spheroids, but disruption of the interaction using anti-TM4SF5
reagent (TSAHC, (Lee et al., 2009)) or via the N-glycosylation TM4SF5
mutant abolishes spheroid formation, indicating that the interaction
between TM4SF5 and CD44 depends on the N-glycosylation of
TM4SF5 and is important for spheroid formation and self-renewal capacity.
Furthermore, the TM4SF5-mediated self-renewal capacity appears to be
involved in the circulation of TM4SF5-positive tumor cells. Cells expressing
TM4SF5 and CD44 are observed in the bloodstream 6 weeks after ortho-
topic injection into mouse livers, whereas suppression of either TM4SF5
or CD44 prevents these cells from circulating through the bloodstream.
Consistent with this view, CD44 lacking the cytosolic region still binds to
TM4SF5. CD44 in tumors is known to recruit monocytes and tumor-

associated macrophages that release soluble factors for the formation of a
microenvironment favored by tumor cells (Negi et al., 2012). TM4SF5
expression correlates with enhanced STAT3 activity in hepatic cancerous
spheroids, and pharmacological inhibition of STAT3 inhibits spheroid
formation. Additionally, STAT3 activity is correlated with increases in the
expression of the EMT-promoting Twist and stemness Bmi1. Additionally,
STAT3/Snail/Twist/stemness has been shown to be the CSC property of
head and neck squamous cell carcinoma cells (Chen et al., 2010). Therefore,
TM4SF5 expression can induce hepatic cancer cells to acquire stemness and
circulate through the bloodstream.
7. CONCLUSION
We have previously summarized the findings of TM4SF5; primarily,
the TM4SF5-mediated communication with the tumor microenvironment
via cross-talk between TM4SF5 and integrins leading to actin organization,
EMT, proliferation, and angiogenesis (Lee et al., 2011). Here we further
evaluate our data to reveal the biological significance of TM4SF5-mediated
EMT. The TM4SF5-dependent EMT process is involved in the develop-
ment of muscle cells in zebrafish, liver fibrosis, enhanced migration and
invasion through direct binding to FAK/c-Src, TKI resistance, and self-
renewal stemness. Therefore, via EMT phenotype induction, TM4SF5 plays
roles in the development of liver fibrosis as well as tumors and their
maintenance.
However, the following questions regarding diverse aspects of TM4SF5
persist. Although the components that bind to TM4SF5 have been identified,
how does TM4SF5 perform its functions in the TM4SF5-enriched microdo-
main? How are the roles of TM4SF5 different from those of other transmem-
brane 4 L six family members? How does the C-terminus of TM4SF5 play
negative regulatory roles, if any, in TM4SF5-mediated signaling and cell
behavior? Furthermore, we identified certain point mutations in TM4SF5
from open databases (Figure 4), and we are currently exploring the tumorigenic
significance of these point mutations. Additionally, we are gathering the tools
and reagents, such as small compounds, peptides, antibodies, and natural prod-
ucts, to regulate TM4SF5-mediated effects. The successful drug development
of anti-TM4SF5 compounds through further investigations of TM4SF5 would
facilitate therapeutic strategies against liver, colon, and prostate cancers.
158 Jung Weon Lee
Transmembrane 4 L six family 5

(TM4SF5)
Figure 4 Clinically identified mutations in TM4SF5 summarized from the following
open databases: (1) http://www.cbioportal.org/public-portal/cross_cancer.do?cancer_
study_id¼all&data_priority¼0&case_ids¼&gene_set_choice¼user-defined-list&gene_
list¼TM4SF5&clinical_param_selection¼null&tab_index¼tab_download&Action¼Submit
#crosscancer/overview/0/TM4SF5; (2) http://cancer.sanger.ac.uk/cosmic/gene/analysis?
ln¼TM4SF5#histo; (3) https://dcc.icgc.org/#/genes/ENSG00000142484. Cl, colon; Lu,
lung; Lv, liver; Es, esophagus; Sk, skin; Ut, uterine; HN, head and neck; St, stomach;
Sp, splicing.
ACKNOWLEDGMENTS
The author’s own work was supported by the National Research Foundation of Korea
(NRF) grant for the Tumor Microenvironment Global Core Research Center (GCRC)
funded by the Korea government (Ministry of Science, ICT & Future Planning) (2011e
0030001), for senior researchers program (Leap research, 2010e0015029/2013e035235),
and for Medicinal Bioconvergence Research Center (NRF-2013M3A6A4044019) to JWL.
REFERENCES
Assoian, R.K., Frolik, C.A., Roberts, A.B., Miller, D.M., Sporn, M.B., 1984. Transforming
growth factor-b controls receptor levels for epidermal growth factor in NRK fibroblasts.
Cell 36, 35e41.
Berditchevski, F., 2001. Complexes of tetraspanins with integrins: more than meets the eye.
J. Cell Sci. 114, 4143e4151.
Besson, A., Dowdy, S.F., Roberts, J.M., 2008. CDK inhibitors: cell cycle regulators and
beyond. Dev. Cell 14, 159e169.
Besson, A., Gurian-West, M., Schmidt, A., Hall, A., Roberts, J.M., 2004. p27Kip1
modulates cell migration through the regulation of RhoA activation. Genes Dev. 18,
862e876.
Cai, X., Lietha, D., Ceccarelli, D.F., Karginov, A.V., Rajfur, Z., Jacobson, K., Hahn, K.M.,
Eck, M.J., Schaller, M.D., 2008. Spatial and temporal regulation of focal adhesion kinase
activity in living cells. Mol. Cell. Biol. 28, 201e214.
Cannito, S., Novo, E., di Bonzo, L.V., Busletta, C., Colombatto, S., Parola, M., 2010.
Epithelial-mesenchymal transition: from molecular mechanisms, redox regulation to
implications in human health and disease. Antioxid. Redox Signal. 12, 1383e1430.
Carloni, V., Mazzocca, A., Mello, T., Galli, A., Capaccioli, S., 2013. Cell fusion promotes
chemoresistance in metastatic colon carcinoma. Oncogene 32, 2649e2660.
Chen, S.Y., Chen, H.C., 2006. Direct interaction of focal adhesion kinase (FAK) with Met is
required for FAK to promote hepatocyte growth factor-induced cell invasion. Mol. Cell.
Biol. 26, 5155e5167.
Chen, T.H., Chan, P.C., Chen, C.L., Chen, H.C., 2011. Phosphorylation of focal adhesion
kinase on tyrosine 194 by Met leads to its activation through relief of autoinhibition.
Oncogene 30, 153e166.
Chen, Y.W., Chen, K.H., Huang, P.I., Chen, Y.C., Chiou, G.Y., Lo, W.L., Tseng, L.M.,
Hsu, H.S., Chang, K.W., Chiou, S.H., 2010. Cucurbitacin I suppressed stem-like
property and enhanced radiation-induced apoptosis in head and neck squamous carci-
nomaederived CD44(þ)ALDH1(þ) cells. Mol. Cancer Ther. 9, 2879e2892.
Choi, S., Lee, S.A., Kwak, T.K., Kim, H.J., Lee, M.J., Ye, S.K., Kim, S.H., Kim, S.,
Lee, J.W., 2009. Cooperation between integrin a5 and tetraspan TM4SF5 regulates
VEGF-mediated angiogenic activity. Blood 113, 1845e1855.
Choi, Y.J., Kim, H.H., Kim, J.G., Kim, H.J., Kang, M., Lee, M.S., Ryu, J., Song, H.E.,
Nam, S.H., Lee, D., Kim, K.W., Lee, J.W., 2014. TM4SF5 suppression disturbs
integrin a5-related signaling and muscle development in zebrafish. Biochem. J.
462, 89e101.
Chu, I.M., Hengst, L., Slingerland, J.M., 2008. The Cdk inhibitor p27 in human cancer:
prognostic potential and relevance to anticancer therapy. Nat. Rev. Cancer 8, 253e267.
Cohen, L.A., Guan, J.L., 2005. Residues within the first subdomain of the FERM-like
domain in focal adhesion kinase are important in its regulation. J. Biol. Chem. 280,
8197e8207.
Coqueret, O., 2003. New roles for p21 and p27 cell-cycle inhibitors: a function for each cell
compartment? Trends Cell Biol. 13, 65e70.
Cox, B.D., Natarajan, M., Stettner, M.R., Gladson, C.L., 2006. New concepts regarding
focal adhesion kinase promotion of cell migration and proliferation. J. Cell. Biochem.
99, 35e52.
Crawford, B.D., Henry, C.A., Clason, T.A., Becker, A.L., Hille, M.B., 2003. Activity and
distribution of paxillin, focal adhesion kinase, and cadherin indicate cooperative roles
during zebrafish morphogenesis. Mol. Biol. Cell. 14, 3065e3081.
Detchokul, S., Williams, E.D., Parker, M.W., Frauman, A.G., 2013. Tetraspanins as regula-
tors of the tumour microenvironment: implications for metastasis and therapeutic
strategies. Br. J. Pharmacol. 171 (24), 5462e5490.
Engelman, J.A., Zejnullahu, K., Mitsudomi, T., Song, Y.C., Hyland, C., Park, J.O.,
Lindeman, N., Gale, C.M., Zhao, X.J., Christensen, J., Kosaka, T., Holmes, A.J.,
Rogers, A.M., Cappuzzo, F., Mok, T., Lee, C., Johnson, B.E., Cantley, L.C.,
Janne, P.A., 2007. MET amplification leads to gefitinib resistance in lung cancer by acti-
vating ERBB3 signaling. Science 316, 1039e1043.
Foley, J., Nickerson, N.K., Nam, S., Allen, K.T., Gilmore, J.L., Nephew, K.P.,
Riese 2nd, D.J., 2010. EGFR signaling in breast cancer: bad to the bone. Semin. Cell
Dev. Biol. 21, 951e960.
160 Jung Weon Lee
Gressner, A.M., Weiskirchen, R., 2006. Modern pathogenetic concepts of liver fibrosis sug-
gest stellate cells and TGF-beta as major players and therapeutic targets. J. Cell. Mol.
Med. 10, 76e99.
Guix, M., Faber, A.C., Wang, S.E., Olivares, M.G., Song, Y., Qu, S., Rinehart, C.,
Seidel, B., Yee, D., Arteaga, C.L., Engelman, J.A., 2008. Acquired resistance to
EGFR tyrosine kinase inhibitors in cancer cells is mediated by loss of IGF-binding
proteins. J. Clin. Invest. 118, 2609e2619.
Hemler, M.E., 2005. Tetraspanin functions and associated microdomains. Nat. Rev. Mol.
Cell Biol. 6, 801e811.
Hemler, M.E., 2008. Targeting of tetraspanin proteinsepotential benefits and strategies. Nat.
Rev. Drug Discov. 7, 747e758.
Henry, M.D., Satz, J.S., Brakebusch, C., Costell, M., Gustafsson, E., Fassler, R.,
Campbell, K.P., 2001. Distinct roles for dystroglycan, b1 integrin and perlecan in cell
surface laminin organization. J. Cell Sci. 114, 1137e1144.
Jung, O., Choi, S., Jang, S.B., Lee, S.A., Lim, S.T., Choi, Y.J., Kim, H.J., Kim, D.H.,
Kwak, T.K., Kim, H., Kang, M., Lee, M.S., Park, S.Y., Ryu, J., Jeong, D.,
Cheong, H.K., Park, K.H., Lee, B.J., Schlaepfer, D.D., Lee, J.W., 2012. Tetraspan
TM4SF5-dependent direct activation of FAK and metastatic potential of hepatocarci-
noma cells. J. Cell Sci. 125, 5960e5973.
Jung, O., Choi, Y.J., Kwak, T.K., Kang, M., Lee, M.S., Ryu, J., Kim, H.J., Lee, J.W.,
2013. The COOH-terminus of TM4SF5 in hepatoma cell lines regulates c-Src to
form invasive protrusions via EGFR Tyr845 phosphorylation. Biochim. Biophys.
Acta 1833, 629e642.
Kang, M., Choi, S., Jeong, S.J., Lee, S.A., Kwak, T.K., Kim, H., Jung, O., Lee, M.S., Ko, Y.,
Ryu, J., Choi, Y.J., Jeong, D., Lee, H.J., Ye, S.K., Kim, S.H., Lee, J.W., 2012a. Cross-
talk between TGFb1 and EGFR signalling pathways induces TM4SF5 expression and
epithelial-mesenchymal transition. Biochem. J. 443, 691e700.
Kang, M., Jeong, S.J., Park, S.Y., Lee, H.J., Kim, H.J., Park, K.H., Ye, S.K., Kim, S.H.,
Lee, J.W., 2012b. Antagonistic regulation of transmembrane 4 L6 family member 5
attenuates fibrotic phenotypes in CCl(4) -treated mice. FEBS J. 279, 625e635.
Kang, M., Ryu, J., Lee, D., Lee, M.S., Kim, H.J., Nam, S.H., Song, H.E., Choi, J.,
Lee, G.H., Kim, T.Y., Lee, H., Kim, S.J., Ye, S.K., Kim, S., Lee, J.W., 2014. Correla-
tions between Transmembrane 4 L6 family member 5 (TM4SF5), CD151, and CD63 in
liver fibrotic phenotypes and hepatic migration and invasive capacities. PLoS One 9,
e102817.
Kashef, J., Diana, T., Oelgeschlager, M., Nazarenko, I., 2013. Expression of the tetraspanin
family members Tspan3, Tspan4, Tspan5 and Tspan7 during Xenopus laevis embryonic
development. Gene Expr. Patterns 13, 1e11.
Ke, A.W., Shi, G.M., Zhou, J., Huang, X.Y., Shi, Y.H., Ding, Z.B., Wang, X.Y.,
Devbhandari, R.P., Fan, J., 2011. CD151 amplifies signaling by integrin a6b1 to
PI3K and induces the epithelial-mesenchymal transition in HCC cells. Gastroenterology
140, 1629e1641 e1615.
Ke, A.W., Shi, G.M., Zhou, J., Wu, F.Z., Ding, Z.B., Hu, M.Y., Xu, Y., Song, Z.J.,
Wang, Z.J., Wu, J.C., Bai, D.S., Li, J.C., Liu, K.D., Fan, J., 2009. Role of overexpression
of CD151 and/or c-Met in predicting prognosis of hepatocellular carcinoma. Hepatol-
ogy 49, 491e503.
Kenneth, N.S., Duckett, C.S., 2012. IAP proteins: regulators of cell migration and
development. Curr. Opin. Cell Biol. 24, 871e875.
Kim, H., Kang, M., Lee, S.A., Kwak, T.K., Jung, O., Lee, H.J., Kim, S.H., Lee, J.W., 2010.
TM4SF5 accelerates G1/S phase progression via cytosolic p27(Kip1) expression and
RhoA activity. Biochim. Biophys. Acta 1803, 975e982.
Kohmo, S., Kijima, T., Otani, Y., Mori, M., Minami, T., Takahashi, R., Nagatomo, I.,
Takeda, Y., Kida, H., Goya, S., Yoshida, M., Kumagai, T., Tachibana, I., Yokota, S.,
Kawase, I., 2010. Cell surface tetraspanin CD9 mediates chemoresistance in small cell
lung cancer. Cancer Res. 70, 8025e8035.
Koshida, S., Kishimoto, Y., Ustumi, H., Shimizu, T., Furutani-Seiki, M., Kondoh, H.,
Takada, S., 2005. Integrinalpha5-dependent fibronectin accumulation for maintenance
of somite boundaries in zebrafish embryos. Dev. Cell 8, 587e598.
Lee, M.S., Kim, H.P., Kim, T.Y., Lee, J.W., 2012. Gefitinib resistance of cancer cells corre-
lated with TM4SF5-mediated epithelial-mesenchymal transition. Biochim. Biophys.
Acta 1823, 514e523.
Lee, S.A., Kim, T.Y., Kwak, T.K., Kim, H., Kim, S., Lee, H.J., Kim, S.H., Park, K.H.,
Kim, H.J., Cho, M., Lee, J.W., 2010. Transmembrane 4 L six family member 5
(TM4SF5) enhances migration and invasion of hepatocytes for effective metastasis.
J. Cell. Biochem. 111, 59e66.
Lee, S.A., Lee, S.Y., Cho, I.H., Oh, M.A., Kang, E.S., Kim, Y.B., Seo, W.D., Choi, S.,
Nam, J.O., Tamamori-Adachi, M., Kitajima, S., Ye, S.K., Kim, S., Hwang, Y.J.,
Kim, I.S., Park, K.H., Lee, J.W., 2008. Tetraspanin TM4SF5 mediates loss of contact
inhibition through epithelial-mesenchymal transition in human hepatocarcinoma.
J. Clin. Invest. 118, 1354e1366.
Lee, S.A., Park, K.H., Lee, J.W., 2011. Modulation of signaling between TM4SF5 and integ-
rins in tumor microenvironment. Front. Biosci. 16, 1752e1758.
Lee, S.A., Ryu, H.W., Kim, Y.M., Choi, S., Lee, M.J., Kwak, T.K., Kim, H.J., Cho, M.,
Park, K.H., Lee, J.W., 2009. Blockade of four-transmembrane L6 family member 5
(TM4SF5)-mediated tumorigenicity in hepatocytes by a synthetic chalcone derivative.
Hepatology 49, 1316e1325.
Lee, S.Y., Kim, Y.T., Lee, M.S., Kim, Y.B., Chung, E., Kim, S., Lee, J.W., 2006. Focal
adhesion and actin organization by a cross-talk of TM4SF5 with integrin a2 are regulated
by serum treatment. Exp. Cell Res. 312, 2983e2999.
Lekishvili, T., Fromm, E., Mujoomdar, M., Berditchevski, F., 2008. The tumour-associated
antigen L6 (L6-Ag) is recruited to the tetraspanin-enriched microdomains: implication
for tumour cell motility. J. Cell Sci. 121, 685e694.
Lietha, D., Cai, X., Ceccarelli, D.F., Li, Y., Schaller, M.D., Eck, M.J., 2007. Structural basis
for the autoinhibition of focal adhesion kinase. Cell 129, 1177e1187.
Long, W., Yi, P., Amazit, L., LaMarca, H.L., Ashcroft, F., Kumar, R., Mancini, M.A.,
Tsai, S.Y., Tsai, M.J., O’Malley, B.W., 2010. SRC-3D4 mediates the interaction of
EGFR with FAK to promote cell migration. Mol. Cell 37, 321e332.
Mani, S.A., Guo, W., Liao, M.J., Eaton, E.N., Ayyanan, A., Zhou, A.Y., Brooks, M.,
Reinhard, F., Zhang, C.C., Shipitsin, M., Campbell, L.L., Polyak, K., Brisken, C.,
Yang, J., Weinberg, R.A., 2008. The epithelial-mesenchymal transition generates cells
with properties of stem cells. Cell 133, 704e715.
Meindl-Beinker, N.M., Dooley, S., 2008. Transforming growth factor-beta and hepato-
cyte transdifferentiation in liver fibrogenesis. J. Gastroenterol. Hepatol. 23 (Suppl 1),
S122eS127.
Moustakas, A., Heldin, C.H., 2007. Signaling networks guiding epithelial-mesenchymal
transitions during embryogenesis and cancer progression. Cancer Sci. 98, 1512e1520.
Muller-Pillasch, F., Wallrapp, C., Lacher, U., Friess, H., Buchler, M., Adler, G., Gress, T.M.,
1998. Identification of a new tumour-associated antigen TM4SF5 and its expression in
human cancer. Gene 208, 25e30.
Murillo, M.M., del Castillo, G., Sanchez, A., Fernandez, M., Fabregat, I., 2005. Involvement
of EGF receptor and c-Src in the survival signals induced by TGF-b1 in hepatocytes.
Oncogene 24, 4580e4587.
162 Jung Weon Lee
Muschel, R.J., Gal, A., 2008. Tetraspanin in oncogenic epithelial-mesenchymal transition.

J. Clin. Invest. 118, 1347e1350.
Naldini, L., Weidner, K.M., Vigna, E., Gaudino, G., Bardelli, A., Ponzetto, C.,
Narsimhan, R.P., Hartmann, G., Zarnegar, R., Michalopoulos, G.K., et al., 1991. Scat-
ter factor and hepatocyte growth factor are indistinguishable ligands for the MET
receptor. EMBO J. 10, 2867e2878.
Naor, D., Sionov, R.V., Ish-Shalom, D., 1997. CD44: structure, function, and association
with the malignant process. Adv. Cancer Res. 71, 241e319.
Negi, L.M., Talegaonkar, S., Jaggi, M., Ahmad, F.J., Iqbal, Z., Khar, R.K., 2012. Role of
CD44 in tumour progression and strategies for targeting. J. Drug Target. 20, 561e573.
Nguyen, K.S., Kobayashi, S., Costa, D.B., 2009. Acquired resistance to epidermal growth
factor receptor tyrosine kinase inhibitors in non-small-cell lung cancers dependent on
the epidermal growth factor receptor pathway. Clin. Lung Cancer 10, 281e289.
Pols, M.S., Klumperman, J., 2009. Trafficking and function of the tetraspanin CD63. Exp.
Cell Res. 315, 1584e1592.
Radford, K.J., Thorne, R.F., Hersey, P., 1997. Regulation of tumor cell motility and migra-
tion by CD63 in a human melanoma cell line. J. Immunol. 158, 3353e3358.
Radisky, D.C., Kenny, P.A., Bissell, M.J., 2007. Fibrosis and cancer: do myofibroblasts come
also from epithelial cells via EMT? J. Cell. Biochem. 101, 830e839.
Reichl, P., Haider, C., Grubinger, M., Mikulits, W., 2012. TGF-beta in epithelial to mesen-
chymal transition and metastasis of liver carcinoma. Curr. Pharm. Des. 18, 4135e4147.
Rubinstein, E., 2011. The complexity of tetraspanins. Biochem. Soc. Trans. 39, 501e505.
Ryu, J., Kang, M., Lee, M.S., Kim, H.J., Nam, S.H., Song, H.E., Lee, D., Lee, J.W., 2014.
Cross-talk between the TM4SF5/FAK and the IL6/STAT3 pathways renders immune
escape of human liver cancer cells. Mol. Cell. Biol. 34, 2946e2960.
Sala-Valdes, M., Ailane, N., Greco, C., Rubinstein, E., Boucheix, C., 2012. Targeting tet-
raspanins in cancer. Expert Opin. Ther. Targets 16, 985e997.
Scheel, C., Weinberg, R.A., 2012. Cancer stem cells and epithelial-mesenchymal transition:
concepts and molecular links. Semin. Cancer Biol. 22, 396e403.
Schoppmann, S.F., Birner, P., Stockl, J., Kalt, R., Ullrich, R., Caucig, C., Kriehuber, E.,
Nagy, K., Alitalo, K., Kerjaschki, D., 2002. Tumor-associated macrophages express
lymphatic endothelial growth factors and are related to peritumoral
lymphangiogenesis. Am. J. Pathol. 161, 947e956.
Serru, V., Le Naour, F., Billard, M., Azorsa, D.O., Lanza, F., Boucheix, C., Rubinstein, E.,
1999. Selective tetraspan-integrin complexes (CD81/a4b1, CD151/a3b1, CD151/
a6b1) under conditions disrupting tetraspan interactions. Biochem. J. 340, 103e111.
Severi, T., van Malenstein, H., Verslype, C., van Pelt, J.F., 2010. Tumor initiation and pro-
gression in hepatocellular carcinoma: risk factors, classification, and therapeutic targets.
Acta Pharmacol. Sin. 31, 1409e1420.
Simpson, N.E., Lambert, W.M., Watkins, R., Giashuddin, S., Huang, S.J., Oxelmark, E.,
Arju, R., Hochman, T., Goldberg, J.D., Schneider, R.J., Reiz, L.F., Soares, F.A.,
Logan, S.K., Garabedian, M.J., 2010. High levels of Hsp90 cochaperone p23 promote
tumor progression and poor prognosis in breast cancer by increasing lymph node metas-
tases and drug resistance. Cancer Res. 70, 8446e8456.
Sincock, P.M., Mayrhofer, G., Ashman, L.K., 1997. Localization of the transmembrane 4 su-
perfamily (TM4SF) member PETA-3 (CD151) in normal human tissues: comparison
with CD9, CD63, and a5b1 integrin. J. Histochem. Cytochem. 45, 515e525.
Sterk, L.M., Geuijen, C.A., Oomen, L.C., Calafat, J., Janssen, H., Sonnenberg, A., 2000.
The tetraspan molecule CD151, a novel constituent of hemidesmosomes, associates
with the integrin a6b4 and may regulate the spatial organization of hemidesmosomes.
J. Cell Biol. 149, 969e982.
Sterk, L.M., Geuijen, C.A., van den Berg, J.G., Claessen, N., Weening, J.J., Sonnenberg, A.,
2002. Association of the tetraspanin CD151 with the laminin-binding integrins a3b1,
a6b1, a6b4 and a7b1 in cells in culture and in vivo. J. Cell Sci. 115, 1161e1173.
Suda, K., Onozato, R., Yatabe, Y., Mitsudomi, T., 2009. EGFR T790M mutation: a double
role in lung cancer cell survival? J. Thorac. Oncol. 4, 1e4.
Takeda, Y., Kazarov, A.R., Butterfield, C.E., Hopkins, B.D., Benjamin, L.E., Kaipainen, A.,
Hemler, M.E., 2007. Deletion of tetraspanin CD151 results in decreased pathologic
angiogenesis in vivo and in vitro. Blood 109, 1524e1532.
Testa, J.E., Brooks, P.C., Lin, J.M., Quigley, J.P., 1999. Eukaryotic expression cloning with
an antimetastatic monoclonal antibody identifies a tetraspanin (PETA-3/CD151) as an
effector of human tumor cell migration and metastasis. Cancer Res. 59, 3812e3820.
Thomson, S., Buck, E., Petti, F., Griffin, G., Brown, E., Ramnarine, N., Iwata, K.K.,
Gibson, N., Haley, J.D., 2005. Epithelial to mesenchymal transition is a determinant
of sensitivity of non-small-cell lung carcinoma cell lines and xenografts to epidermal
growth factor receptor inhibition. Cancer Res. 65, 9455e9462.
Tice, D.A., Biscardi, J.S., Nickles, A.L., Parsons, S.J., 1999. Mechanism of biological synergy
between cellular Src and epidermal growth factor receptor. Proc. Natl. Acad. Sci. U.S.A.
96, 1415e1420.
Veenbergen, S., van Spriel, A.B., 2011. Tetraspanins in the immune response against cancer.
Immunol. Lett. 138, 129e136.
Wright, M.D., Ni, J., Rudy, G.B., 2000. The L6 membrane proteinsea new four-transmem-
brane superfamily. Protein Sci. 9, 1594e1600.
Yanez-Mo, M., Barreiro, O., Gordon-Alonso, M., Sala-Valdes, M., Sanchez-Madrid, F.,
2009. Tetraspanin-enriched microdomains: a functional unit in cell plasma
membranes. Trends Cell Biol. 19, 434e446.
Yang, X., Kovalenko, O.V., Tang, W., Claas, C., Stipp, C.S., Hemler, M.E., 2004. Palmi-
toylation supports assembly and function of integrin-tetraspanin complexes. J. Cell Biol.
167, 1231e1240.
Yang, X.H., Flores, L.M., Li, Q., Zhou, P., Xu, F., Krop, I.E., Hemler, M.E., 2010. Disrup-
tion of laminin-integrin-CD151-focal adhesion kinase axis sensitizes breast cancer cells to
ErbB2 antagonists. Cancer Res. 70, 2256e2263.
Yauch, R.L., Berditchevski, F., Harler, M.B., Reichner, J., Hemler, M.E., 1998. Highly stoi-
chiometric, stable, and specific association of integrin a3b1 with CD151 provides a ma-
jor link to phosphatidylinositol 4-kinase, and may regulate cell migration. Mol. Biol. Cell
9, 2751e2765.
Zhong, W., Shen, W.F., Ning, B.F., Hu, P.F., Lin, Y., Yue, H.Y., Yin, C., Hou, J.L.,
Chen, Y.X., Zhang, J.P., Zhang, X., Xie, W.F., 2009. Inhibition of extracellular
signal-regulated kinase 1 by adenovirus mediated small interfering RNA attenuates
hepatic fibrosis in rats. Hepatology 50, 1524e1536.
CHAPTER FIVE
Emerging Roles of JmjC

Domain-Containing Proteins
Sandra L. Accari1 and Paul R. Fisher2, *
1
Professional and Continuing Education, Turitea Campus, Massey University, Palmerston North, New Zealand
2
Discipline of Microbiology, La Trobe University, Melbourne, VIC, Australia
*Corresponding author: E-mail: P.Fisher@latrobe.edu.au
Contents
1. Introduction 166
2. Histone Modification and Methylation 167
3. Histone Demethylation and Demethylases 171
3.1 Peptidylarginine Deiminase 4 (PADI4/PAD4) and Amine Oxidase (LSD1/KDM1) 171
3.2 Jumonji C Domain-Bearing Histone Demethylases 174
3.2.1 Jumonji histone demethylase subgroups 176
4. Plant JmjC Histone Demethylation 210
4.1 Roles of Nondemethylating JmjC Domain-Containing Proteins 210
4.2 Plant JmjC Histone Demethylases 211
5. Conclusions 212
References 213
Abstract
Jumonji C (JmjC) domain-containing proteins are a diverse superfamily of proteins con-
taining a characteristic, evolutionarily conserved b-barrel structure that normally con-
tains binding sites for Fe(II) and a-ketoglutarate. In the best studied JmjC-domain
proteins, the JmjC barrel has a histone demethylase catalytic activity. Histones are
evolutionarily conserved proteins intimately involved in the packaging of DNA within
the nucleus of eukaryotic organisms. The N-termini (“tails”) of the histone proteins are
subject to a diverse array of posttranslational modifications including methylation.
Unlike many of the other histone modifications which are transient, methylation was
thought to be permanent, until the relatively recent identification of the first demethy-
lases. Jumonji C domain-containing proteins were first identified with a role in the
modulation of histone methylation marks. This family of proteins is broken up into
seven distinct subgroups based on domain architecture and their ability to antagonize
specific histone methylation marks. Their biological functions derive from their ability to
regulate gene expression and include roles in cell differentiation, growth, proliferation,
and stress responses. However, one subgroup remains, the largest, in which the JmjC
domain has no known biochemical function. These proteins belong to the JmjC-
domain-only subgroup and as their name suggests, the only bioinformatically recogniz-
able domain they contain is the highly conserved JmjC domain.
j
ISSN 1937-6448
166 Sandra L. Accari and Paul R. Fisher
1. INTRODUCTION
The Jumonji C (JmjC) domain-containing proteins are a family of
redox enzymes that are able to catalyze a wide variety of oxidative reactions.
These proteins have been identified in all living organisms from bacteria to
higher eukaryotes and are characterized by the highly conserved JmjC
domain. This novel domain was first described by Takeuchi et al. (1995)
in the protein Jumonji (meaning cruciform in Japanese), mutations in which
were shown to cause a cross-like malformation in the neural plate during
development in mutant embryos. Since the initial identification of this novel
domain, the JmjC domain has been annotated in over 10,000 proteins in
public databases including Pfam, SMART, UniProt, and InterPro. Such a
large number of proteins identified with this domain indicate an expansion
of this superfamily (Hahn et al., 2008).
The JmjC domain has a characteristic structural topology consisting of a
double-stranded b helix (DSBH) fold. The fold consists of eight antiparallel
b sheets to form a barrellike structure (Figure 1). This topology has been
shown to be characteristic of cupin metalloenzymes to which the JmjC
domain-containing proteins are thought to belong (Dunwell et al., 2001).
The annotation cupin was given to the group on the basis of the putative
b barrel shape conserved among the members of the family (“cupa” being
the Latin for small barrel) (Dunwell et al., 2001). Cupin metalloenzymes
are characterized by containing at least one DSBH surrounded by other sec-
ondary structure elements (Clifton et al., 2006). The JmjC domain places
proteins bearing it into the 2-oxoglutarate oxygenase class of cupin proteins
using the domain’s b barrel topology to coordinate iron II (Fe(II)) and
2-oxoglutarate (also known as a-ketoglutarate) to carry out their functions
(Clifton et al., 2006).
The JmjC domain superfamily is large and contains a diverse range of both
enzymatically active (including many histone demethylases) and nonactive
members (Table 1). The family consists of two distinct functional subfamilies:
one shown to be able to remove the methylation marks (“marks” refers to the
posttranslation addition of methyl groups to specific lysine or arginine residues
on the N-termini of histone proteins) on histone tails (KDM2 to KDM7 in
Table 1) and the other with unknown function (Group JmjC only in Table 1).
Both of these subfamilies will be discussed in the following sections. All of the
well-studied JmjC domain proteins are histone demethylases, other histone
demethylases (PADI and KDM1 in Table 1) are also briefly considered.
Roles of JmjC Domain-Containing Proteins 167
Figure 1 Tertiary structure of factor-inhibiting HIF-1 alpha. The first crystal structure
solved for a JmjC domain-containing protein was factor-inhibiting hypoxia inducible
factor alpha. This structure shows the classical double-stranded b helix consisting of
eight antiparallel b sheets. The highlighted residues indicate the beginning of the
JmjC domain (pink (light gray in print versions), labeled J), the conserved residues
for Fe(II) binding (red (dark gray in print versions), labeled F), and a-ketoglutarate bind-
ing (green (gray in print versions), labeled O). (Figure generated using the SWISS MODEL
online software in the automated mode at http://swissmodel.expasy.org/.). Template
2w0xA (factor-inhibiting HIF-1 alpha with pyridine 2,4 dicarboxylic acid, Chain A) was
used.
2. HISTONE MODIFICATION AND METHYLATION

Histones are evolutionarily conserved proteins shown to be acutely
involved in the packaging of DNA within the nucleus of eukaryotic organ-
isms. The classical histone core, around which 146 bp of DNA is wrapped,
consists of two, each of, histone proteins H3, H4, H2A, and H2B (Anand
and Marmorstein, 2007). The N-termini of the histone proteins project
out and are termed “tails”; these are subject to a diverse array of posttransla-
tional modifications which are signals for numerous proteins. Modifications
include phosphorylation, SUMOylation, acetylation, ubiquitination, and
168
Table 1 Histone demethylases and JmjC domain-containing proteins
Enzyme family Histone demethylase Group Name/synonym Substrate
PADI Yes PADI PADI4/PAD4 H3R2; H3R8; H3R17;

H3R26; H4R3
LSD Yes KDM1 LSD1 H3K4me2/me1; H3K9me2/1
LSD2 H3K4me2/1
JMJC Yes KDM2 JHDM1A/FBXL11/Ndy2/ H3K36me2/me1; H3K4me3
KDM2A
JHDM1B/FBXL10/Ndy1/ H3K36me2/1
KDM2B
Yes KDM3 JHDM2A/TSGA/JMJD1A/ H3K9me2/me1
KDM3A
JHDM2B/5qCNA/JMJD1B/ H3K9me2/1
KDM3B
JHDM2C/TRIP8/JMJD1C/ H3K9me2/1?
KDM3C
HR ND
Sandra L. Accari and Paul R. Fisher

Yes KDM4 JHDM3A/JMJD2A/KDM4A H3K9me3/me2;
H3K36me3/me2
JMJD2B/KDM4B H3K9me3/me2;
H3K36me3/me2
JMJD2C/GASC1/KDM4C H3K9me3/me2;
H3K36me3/me2
JMJD2D/KDM4D H3K9me3/me2/me1;
H3K36me3/me2
Roles of JmjC Domain-Containing Proteins
Yes KDM5 JARID1A/RBP2/KDM5A H3K4me3/me2
JARID1B/PLU-1/KDM5B H3K4me3/me2
JARID1C/SMCX/KDM5C H3K4me3/me2
JARID1D/SMCY/KDM5D H3K4me3/me2
JARID2/Jumonji ND
Yes KDM6 UTX/KDM6A H3K27me3/me2
UTY ND
JMJD3/KDM6B H3K27me3/me2
Yes KDM7 KIAA1718/KDM7A H3K27me2/me1; H3K9me2/1
PHF8/KDM7B H3K9me2; H4K20me1
PHF2/KDM7C H3K9me1
Yes and Nob JmjC only JMJD6/PTDSR H3R2me2, H4R3me2a
NO66 H3K3me3/me2, H3K36me3/
me2
KDM, Lysine demethylase; H, histone; K, lysine; R, arginine; me1/2/3, mono-, di-, or trimethylation; ND, not determined.
a
There is still some debate as to whether JMJD6 is a true arginine demethylase as the results have not been able to be repeated.
b
Many members of this subgroup remain uncategorized. Therefore, it is unknown whether all members possess histone demethylase activity.
Information compiled from Cloos et al. (2008), Fortschegger and Shiekhattar (2011), Klose et al. (2006), Pedersen and Helin (2010)
169
methylation (Anand and Marmorstein, 2007; Lin et al., 2008). Multiple

modifications can decorate these tails, with some key amino acids within
the tail being able to be modified several different ways (Schneider and
Shilatifard, 2006). Many of these modifications are transient, although
methylation was previously thought to be permanent until the identification
of the first demethylases (Anand and Marmorstein, 2007; Schneider and
Shilatifard, 2006).
Histone methylation has been shown to be involved in many key areas of
epigenetics, including gene activation and silencing, DNA damage re-
sponses, and chromatin formation (Shi, 2007). Once the process of histone
methylation was thought to be static, with methylation only reversed by
proteolytic removal of the histone tail or complete removal of the histone
and replacement with an unmodified one (Agger et al., 2008; Lin et al.,
2008; Shi et al., 2004). However, this is no longer the case. Recent discov-
eries have shown that methylation is, in fact, a dynamic process with histone
methylation marks being antagonized by their own demethylases (Lin et al.,
2008). Methylation of histone tails has been shown to occur mainly on H3
and H4 and individual residues linked with either transcriptionally active or
silenced genes (Figure 2). Several residues to date have been shown to be
involved, including (but not necessarily limited to) H3K4, H3K9, H3K27,
and H3K36, which are located in the tail as well as H3K79, which is located
in the core of H3. H4K20, to date, is the only known methylation mark on
H4 to be antagonized (term used for the demethylation of a particular
methylation state) by a demethylase (Agger et al., 2008). Methylation of
lysine residues can occur in three distinct states, these being mono-, di-,
or trimethylation (Figure 2) (Agger et al., 2008).
Methylation is not restricted to lysine residues in histone tails, but also
occurs on arginine residues which can be either mono- or dimethylated
(which can occur either symmetrically or asymmetrically) (Figure 3). Again,
the methylation marks are usually found on H3 and H4 residuesdthese be-
ing H3R2, H3R8, H3R17, and H3R26 as well as H4R3 (Agger et al.,
2008; Klose and Zhang, 2007). Methylation marks on arginine residues
have been shown to occur in close proximity to other modifications, and
there have been suggestions that there is some cross-talk between them
(Agger et al., 2008).
Methylation of a particular histone variant, as well as the position and de-
gree of methylation, may signal different outcomes. These differences allow
for a vast array of potential downstream effects from transcription activity or
inactivity to the spread of heterochromatin or the potential for disease states.
Figure 2 The major lysine residues methylated on H3 and H4. The numbers represent
the methylated amino acid residue on each histone, H3 (dark gray) and H4 (light gray).
The general function of the methylation state (mono-, di-, and tri-) is depicted by
the different symbols (refer to key in the figure). Figure modified from Figure 1 of
Mosammaparast and Shi (2010).
Therefore, the responsible enzymes and mutations in these enzymes have

the potential to cause many harmful or beneficial outcomes.
3. HISTONE DEMETHYLATION AND DEMETHYLASES

Histone demethylation is achieved by three distinct reaction mecha-
nisms, respectively, carried out by amine oxidases, peptidylarginine deimi-
nases, and by the JmjC domain-bearing lysine demethylases.
3.1 Peptidylarginine Deiminase 4 (PADI4/PAD4) and Amine

Oxidase (LSD1/KDM1)
The first histone demethylase to be discovered was PADI4/PAD4.
Mammals have been shown to encode four peptidylarginine deiminase
enzymes; however, only PADI4 is localized in the nucleus, leading to it be-
ing suggested as the first histone arginine demethylase (Klose and Zhang,
2007). PADI4 is an arginine demethylase that is able to remove arginine
methylation by the process of deimination/demethylimiation that converts
(A) CH3 CH3 CH3 CH3

H 2N NH2+ HN NH2+ HN NH+ N NH2+
H3C
HN HN HN HN
HMT HMT or
O O O O
N N N N
H O H O H O H O
Arginine Mono-methyl Symmetrical Asymmetrical
arginine di-methyl di-methyl
arginine arginine
(B) CH3 CH3

H 3C H3C H 3C
NH3+ NH2+ NH+ N+ CH3
HMT HMT HMT
O O O O
N N N N
H H H H
O O O O
Lysine Mono-methyl Di-methyl Tri-methyl
lysine lysine lysine
Figure 3 Methylation states of lysine and arginine residues in histones. (A) Methylation
of arginine residues can occur to form mono-methyl, symmetrical di-methyl, and asym-
metrical di-methylarginine. (B) Methylation of lysine residues can occur to form mono-,
di-, or trimethyllysine. The ellipses highlight the methyl groups attached at each of the
different methylation states. (From Figure 1 of Klose and Zhang (2007).). HMT ¼ Histone
Methyltransferase.
arginine or methylarginine to citrulline (Figure 4) (Anand and Marmorstein,

2007; Shi et al., 2004). However, it is moot whether this enzyme should be
regarded as a true demethylase because this process does not produce an un-
modified residue but a modified residue. Furthermore PADI4/PAD4 will
catalyze the deimination reaction irrespective of the methylation state of
the substrate arginine. This also suggests it should not be considered a true
demethylase (Shi et al., 2004).
PADI4 demethylimination of the methylarginine produces equal
amounts of both citrulline and methylamine (Cuthbert et al., 2004; Klose
and Zhang, 2007; Wang et al., 2004). The active form of PADI4 is a dimer
in which the subunits bind head to tail to form an active site cleft that is
responsible for the deimination (Klose and Zhang, 2007). The binding of
two Ca2þ ions close to the active site allows for a structural conformational
change which stabilizes the dimer and results in the functional enzyme
CH3
H 2N NH2+ H2 N O
H2 N NH+ H2 N O
HN PADI4 HN
HN PADI4 HN
Ca2+
Ca2+
H2O NH3
O O H2O CH3
N N O O
H H N N
O O NH3+ H
H
O O
Arginine Citrulline Mono-methyl Citrulline
arginine
Figure 4 Deimination and demethylimination of arginine and mono-methylated

arginine. Mechanisms of action of PADI4 in antagonizing methyl marks on arginine res-
idues. This method of demethylation is not a true demethylation as it does not produce
an unmodified arginine residue but citrulline which therefore cannot be methylated
again. Figure from Figure 2(A) of Klose and Zhang (2007).
(Figure 4) (Klose and Zhang, 2007). PADI4 has been shown to have a very
broad substrate range including the H3- and H4-methylated residues, as well
as nonhistone arginine residues. To carry out the removal of the methyl
mark, PADI4 binds to the corresponding residue which allows for direction
of the residue into the active site cleft, this produces a conformational change
of PADI4 usually in conjunction with five normally unstructured amino
acids of the histone peptide (Cuthbert et al., 2004). The five unstructured
amino acids form an ordered b-turn-like conformation. PADI4 does not
recognize a specific binding sequence but instead the presence of the un-
structured amino acids surrounding the target residues seems to signal the
binding (Cuthbert et al., 2004). This explains why PADI4 is able to bind
to such a wide number of substrates (Klose and Zhang, 2007).
The second demethylase to be discovered was lysine-specific histone
demethylase 1 (LSD1/KDM1) (Agger et al., 2008; Lan et al., 2008; Shi,
2007). LSD1 is a flavin-dependant amine oxidase (Culhane and Cole,
2007). It belongs to the family of monoamine oxidases based on sequence
analysis and crystal structures (Culhane and Cole, 2007).
LSD1 was the first protein shown to demethylate histone lysine residue
H3K4 from either the mono- or dimethylated state to unmethylated lysine
(Culhane and Cole, 2007; Shi, 2007). LSD1, however, is unable to remove
trimethylation from its target residues. Flavin-dependant demethylases such
as LSD1 produce hydrogen peroxide (H2O2) as a by-product of the deme-
thylation reaction. Each demethylation cycle requires electrons to be shut-
tled to molecular O2 via an FAD/FADH moiety (Figure 5) (Forneris et al.,
LSD1 H2 O O
Histone-Lys Histone-Lys Histone-Lys +
H H
FAD FADH2
H3C NH H3C N H2 N +
+ CH + CH
3
2 CH3
H2O2 O2
Figure 5 Action of LSD1 against mono-methylated H3K4. LSD1 is able to antagonize

methyllysine using FAD and an electron acceptor. Figure from Figure 4(A) of
Marmorstein and Trievel (2009).
2008). LSD1 has been shown to have a conserved role, indicating the impor-
tance of such demethylating proteins (Lan et al., 2008).
LSD1 has also been linked to the removal of methylation marks from
H3K9. This suggests that dynamic demethylation of histones is not limited
to H3K4 and that all methylation marks are likely to have a demethylase that
antagonizes their presence. Many methylation marks have been linked to
different diseases, in particular, those related to overactivation or repression
of a particular gene. This is revealed in the potential link between decreased
H3K4 methylation and enrichment of H3K9 methylation with certain tu-
mors (Forneris et al., 2008). It would suggest that LSD1 and other histone
demethylases linked to disease would provide potential drug targets.
The identification of LSD1 as a histone demethylase laid to rest the
previous dogma stating that histone methylation was a static modification
(Culhane and Cole, 2007). Finding one demethylase stimulated the search
for more proteins that may antagonize methylation marks at different posi-
tions on H3 and other histones. This led to the discovery of other histone
demethylases belonging to the JmjC family.
3.2 Jumonji C Domain-Bearing Histone Demethylases

JmjC domain-containing proteins were first identified in 1995 by Takeuchi
et al. with the identification of Jumonji (Takeuchi et al., 1995), but the role
they had in histone demethylation took longer to discover. JmjC domain-
containing proteins were initially hypothesized to have the ability to
demethylate substrates (Figure 6) based on sequence homology of the
JmjC domain to the catalytic region of the DNA repair enzyme AlkB
(Marmorstein and Trievel, 2009). The AlkB protein demethylates alkylated
DNA bases, thereby removing toxic methylation damage to DNA (Trewick
et al., 2002). AlkB and JmjC domain-containing histone demethylases
Figure 6 Multiple sequence alignment of JmjC domains from several subgroups with
AlkB. Using ClustalW in BioEdit 7.0.5.3, an alignment was carried out to show that
the residues for Fe(II) are conserved from bacteria to higher eukaryotes. This conserva-
tion led to the hypothesis that JmjC domain-containing proteins may be able to antag-
onize histone methylation in a similar manner to that of the DNA repair enzyme AlkB.
This information led to the catalytic core of HxD/EXnH being identified for JmjC
domain-containing proteins (shaded squares represent the conserved residues). Ec,
Escherichia coli; Ce, Caenorhabditis elegans; Dm, Drosophila melanogaster; Hs, Homo
sapiens; Mm, Mus musculus; Rn, Rattus norwegicus; Sc, Saccharomyces cerevisiae.
(JmjC HDM) both belong to the same class of cupin mononuclear

Fe(II)-dependent oxygenases (Marmorstein and Trievel, 2009).
The key feature of all the JmjC HDM proteins in the process of deme-
thylation is the presence of the JmjC domain. It has been shown that the
JmjC domain is required for oxidative catalytic demethylation of lysine/
arginine residues in conjunction with its two cofactors Fe(II) and
a-ketoglutarate (a-KG) (a-ketoglutarate is also termed 2-oxoglutarate
(2-OG)) (Agger et al., 2008). The domain contains a conserved motif
(HXD/EXnH) required for catalytic activity and Fe(II) binding. It includes
two essential histidine residues which bind a-KG (Cloos et al., 2008). The
motif must also be contained within the characteristic b barrel structure of
the JmjC domain to carry out the binding of its cofactors and therefore
the demethylation process. The key differences between JmjC HDMs and
other histone demethylases are their ability to remove all three states of
methylation on lysine residues through hydroxylation. This mechanism
means that, unlike LSD1, these proteins do not require a protonated lysine
ε-amine group on the substrate (Figure 7) (Marmorstein and Trievel, 2009).
The reaction catalyzed by JmjC HDMs occurs instead, via the hydroxylation
of z-methyl group of the substrate methyllysine, utilizing a-KG and molec-
ular oxygen to produce a hydroxyl-methyllysine intermediate, succinate,
Figure 7 Mechanism of histone demethylation carried out by JmjC domain-containing

proteins. Schematic of demethylation carried out by JmjC domain-containing histone
demethylases. (A) Hydroxylation of a lysine methyl group using a radical-based mech-
anism with O2 and a-ketoglutarate (2-OG) to produce carbon dioxide and succinate. (B)
The hydroxyl-methyl ε-ammonium hemiaminal intermediate is decomposed to yield
formaldehyde and demethylated lysine. Modified from Figure 4(B) of Marmorstein and
Trievel (2009).
and CO2 (Figure 7(A)) (Marmorstein and Trievel, 2009). The hydroxyl-
methyl ε-amine intermediate is unstable and decomposes spontaneously to
produce demethylated lysine and formaldehyde (Figure 7(B)) (Marmorstein
and Trievel, 2009).
Since JmjC HDMs are able to demethylate all three states of histone
lysine methylation incrementally, they provide cells with a mechanism to
dynamically control all degrees of histone lysine methylation and thereby
regulate various functions of the target residue (Marmorstein and Trievel,
2009). This provides the ability to regulate the state that chromatin is in,
enabling the activation or inhibition of various transcription states and affects
the ability of DNA to be repaired (Mosammaparast and Shi, 2010). Figure 8
identifies some of the known roles of the different JmjC HDM proteins to
be discussed further in subsequent sections.
3.2.1 Jumonji histone demethylase subgroups

The JmjC domain proteins, including the demethylases, have been classified
into subgroups on the basis of their domain architecture and the similarity of
the JmjC domain sequence (Cloos et al., 2008; Klose et al., 2006).
3.2.1.1 KDM2: Jumonji histone demethylase 1(JHDM1) subgroup

The first subgroup of JmjC domain-containing proteins to be identified
with histone demethylase activity was the JDHM1 subgroup, later renamed
KDM2 (for lysine demethylase 2). Tsukada et al. (2006) used a series of
biochemical assays to isolate a protein able to antagonize the methylation
mark on H3K36. Using fractionation and chromatography, they identified
Figure 8 Substrate specificities of the various JmjC HDMs. The dashed lines point to
the methylated residue(s) that are demethylated by the enzymes indicated above
them. The numbers on each of the histone tales represent the lysine residues that
are methylated. The symbols indicate the biological function being regulated (see
legend in figure) and the number of dots represents the level of methylation (i.e.,
mono-, di-, or trimethylation). Figure modified from Figure 3 of Mosammaparast and
Shi (2010).
the fraction containing the protein able to antagonize the mark of interest
and, using mass spectrometry, they identified the protein as F-box
leucine-rich repeat protein 11 (Tsukada et al., 2006). From this starting
point, they were then able to identify a second highly related protein
FBXL10, which also exhibited H3K36 demethylase activity (Tsukada
et al., 2006).
From this first identification, several additional proteins were found to
belong to this subgroup. They are distributed across widely divergent phylo-
genetic groups, from yeast to mammals (Agger et al., 2008; Klose et al.,
2006). Members identified to date include Epe1 (Schizosaccharomyces pombe),
Jhd1 (Saccharomyces cerevisiae), CG11033 (Drosophila melanogaster), 3H549
(Caenorhabditis elegans), and JHDM1A and B (Homo sapiens, Mus musculus)
(Frescas et al., 2007; Klose et al., 2006; Pedersen and Helin, 2010; Trewick
et al., 2005). As the JmjC domain-containing protein family is relatively
new, there will doubtless be more members identified as more analysis is
conducted.
In the initial identification of proteins belonging to this subgroup,

FBXL11 and FBXL10 both had very similar domain architecture. These
proteins contained the essential JmjC domain and F-box, as well as the
leucine rich repeats, which initially characterized these proteins. They also
contained a CxxC zinc finger and a plant homeodomain (PHD)
(Figure 9(A)) (Klose et al., 2006; Tsukada et al., 2006). The subgroup mem-
bers in lower organisms either contain only the JmjC domain alone or, in the
case of S. cerevisiae, the additional PHD (Figure 9(A)) (Klose et al., 2006).
This implies that in different lineages, domains have been added to or
removed from the ancestral protein and reinforces the importance of classi-
fying the JmjC domain proteins not only on the basis of their domain archi-
tecture but also on the similarity of the JmjC domain sequence (Figure 9(B)).
3.2.1.1.1 KDM2A/KDM2B Of the members of this subgroup, KDM2A

and KDM2B, in particular KDM2B, are the best studied. These proteins
were initially identified on the basis of being able to carry out H3K36 deme-
thylation by Tsukada et al. (2006). Tsukada et al. (2006) carried out a series
of domain deletions on KDM2A to confirm that the JmjC domain was
responsible for catalyzing the demethylation process (Tsukada et al.,
2006). As this was the first attributed function of the protein, and the
JmjC domain carried out the enzymatic activity, they renamed the protein
JHDM1A (JmjC domain-containing histone demethylase 1A) and subse-
quently renamed FBXL10, JHDM1B (Tsukada et al., 2006). However,
this may not be the only role that these proteins carry out. As seen in Table
1, these proteins have a number of names depending on the organism in
which they were identified. This has led to much confusion over the roles
of these proteins which are, in most cases, dependent on the JmjC domain.
KDM2A (JHDM1A/FBXL11) is a H3K36me2/me1 demethylase local-
ized to the nucleus of cells (Frescas et al., 2008). In the nucleus, KDM2A has
been shown to interact with the heterochromatin protein, HP1, and
through this interaction, KDM2A binds to regions of heterochromatin
and demethylates H3K36me2 at pericentric regions (Frescas et al., 2008).
This has an overall effect of the stability of the heterochromatin and also
acts as a repressor for satellite RNA genes embedded in the centeromeric re-
peats (Tanaka et al., 2010).
The demethylase activity of KDM2A is not limited to histone targets.
KDM2A has been shown to be a negative regulator of the immune response
factor, NFkB (Lu et al., 2009). When expression is decreased the opposite
effect is noted, whereby there is increased accessibility to NFkB and this is
Figure 9 Members of the KDM2 subgroup. (A) A schematic diagram of the domain
architecture of several KDM2 members from different organisms. It can be seen that
in higher eukaryotes that they have a similar domain architecture, whereas in lower
organisms, there is a loss of key domains. (B) A multiple sequence alignment of the
JmjC domain from the selected organisms indicating the highly conserved residues
for Fe(II) and a-KG binding suggesting that these, with the exception of Epe1, are all
enzymatically active. Hs, Homo sapiens; mm, Mus musculus; dm, Drosophila melanogaster;
sp. Schizosaccharomyces pombe; sc, Saccharomyces cerevisiae. Figure modified from
Figure 4 of Klose et al. (2006).
directly related to the activity of the JmjC domain (Lu et al., 2009). When a
single mutation in the catalytic triad is created (H212A), this function is
abolished (Lu et al., 2009). The ability to act on this pathway is based on
the recognition by the JmjC domain of the K218 and K221 methylated
residues on the p65 subunit of NFkB (Lu et al., 2010b). KDM2A is also able
to interact with histones at ribosomal DNA (rDNA) promoters demethylat-
ing H3K36me1 and H3K36me2 (Frescas et al., 2007). This function has
been shown to occur in mammalian cells under starvation conditions
(Frescas et al., 2007; Tanaka et al., 2010). This study also showed that the
succinate produced as a by-product of demethylation is able to inhibit the
function of KDM2A, suggesting that there may be a negative feedback
loop with respect to succinate (Tanaka et al., 2010).
In a similar study conducted for KDM2A, Frescas et al. (2007) found that
KDM2B is a nucleolar protein. Nucleolar localization, as well as the pres-
ence of the CXXC zinc finger DNA binding domain, led Frescas et al.
(2007) to suggest that KDM2B is able to interact with rDNA. The JmjC
domain, CXXC zinc finger domain and the nucleolus localization signal
were all shown to be required for binding and repression of rDNA transcrip-
tion through the demethylation of H3K4me3 (Frescas et al., 2007). Inhibi-
tion of rDNA transcription led to decreased cell growth and proliferation as
these processes are linked directly to rRNA synthesis. These results thus
revealed a biological function of KDM2B (Frescas et al., 2007).
KDM2B has also been shown to interact with the transcription factor
c-JUN in a JmjC domain-independent manner. The interaction is based
on other domains present (F-Box and LRR) (Koyama-Nasu et al., 2007).
KDM2B is also important in cell death and cell cycle progression, again in
manner independent of its JmjC domain and its demethylase activity
(Koyama-Nasu et al., 2007). Together, this suggests that JmjC domain-
containing proteins with multiple domains will have activities based on
which particular domains are present.
In mice, Ndy2 (KDM2A/JHDM1A) and Ndy1 (KDM2B/JHDM1B)
have been shown to contribute to the induction and or progression of
MMLV-induced T cell lymphomas (Pfau et al., 2008). Pfau et al. (2008)
also showed that Ndy1 has a role in cell senescence. Both proteins have
also been shown, when overexpressed, to be involved in immortalization
of mouse embryonic fibroblasts in culture (Pfau et al., 2008). These
phenotypes have been linked to various cancers including the already
identified lymphoma and acute myeloid leukemia (He et al., 2011; Pfau
et al., 2008).
Mouse Ndy1 has also been implicated in protecting cells from oxi-
dative stress (Polytarchou et al., 2008). Through its demethylase activity,
Ndy1 is able to regulate the expression of redox regulatory genes, directly
targeting the promoters for the antioxidant enzymes NAD(P)H quinone
oxidoreductase-1 (Nqo1) and peroxiredoxin-4 (Prdx4). Nqo1 is a ubiquitous

homodimeric flavoprotein located in the cytosol, which catalyzes the two-
electron reduction of quinones to hydroquinones (Polytarchou et al., 2008).
Prdx4 is one of six peroxiredoxins found in cells, which converts H2O2 and
alkyl-hydroperoxides to H2O or alcohols, respectively. Prdx4 is localized to
the endoplasmic reticulum and the extracellular space (Polytarchou et al.,
2008). There is some evidence that links changes in the metabolism of reac-
tive oxygen species are important in the establishment of cell senescence.
This links the demethylation of H3K36me2 and H3K4me3 with alterations
in both these processes (Polytarchou et al., 2008).
KDM2A and KDM2B have both been linked through aberrations in his-
tone demethylation to various cancers (Klose et al., 2006). The alterations in
levels of demethylation whether through increased or decreased levels have
been shown to effect cell proliferation and cell death. The ability of these
proteins to effect the regulation of specific pathways is what is suggested
to lead to this outcome. However, with the JmjC domain not being the
only domain present in these proteins, the presence of other domains allows
for a spectrum of possible interactions that are yet to be explored.
3.2.1.1.2 Jhd1 Unlike other organisms that have many methylatable

lysine residues on histone H3, the budding yeast S. cerevisiae has only three,
K4, K36, and K79 (Fang et al., 2007). Saccharomyces cerevisiae differs from
other organisms in that it lacks an LSD 1 homologue, though it still exhibits
significant levels of lysine methylation on its histones (Tu et al., 2007). For
histone lysine demethylation to occur, S. cerevisiae would therefore be
expected to possess members of the JmjC domain histone demethylating
subgroups.
In a screen for homologues to JHDM1A, Jdh1 was identified (Tu et al.,
2007). Like other members of this subgroup, Jhd1 has been shown to be a
histone demethylase (Fang et al., 2007). Unlike other members of this sub-
group, Jhd1 lacks the F-box, zinc finger, and LRR domains (Fang et al.,
2007; Tu et al., 2007). However, the JmjC domain is highly similar to
that of other members of this subgroup and like them is a H3K36 demethy-
lase. The demethylase activity depends on both the JmjC domain and adja-
cent sequence (Fang et al., 2007). It has been shown that the Jhd1 does not
form a stable complex with other proteins but is able to form a homoligomer
(Fang et al., 2007). The role of Jhd1 is yet to be elucidated, but it has been
suggested that there is a link between different methylation states and the
level of transcription (Kim and Buratowski, 2007).
3.2.1.1.3 Epe1 Unlike other members of this subgroup, Epe1 (the

JmjC domain-containing protein identified in the fission yeast S. pombe)
has been shown to be involved in the stability of heterochromatin rather
than the demethylation of histone lysine residues (Ayoub et al., 2003; Isaac
et al., 2007; Trewick et al., 2005; Zofall and Grewal, 2006). Epe1 contains a
naturally occurring mutation in the third Fe(II) binding residue (HiseTyr)
which abolishes its ability to be a histone demethylase (Klose et al., 2006).
The role in heterochromatin stability was first identified in a screen carried
out by Ayoub et al. (2003) who were looking for the conditions required to
overcome the heterochromatin barrier at the distal ends of centromeres in
the mating locus. Heterochromatin at the centromere distal and proximal
ends has distinct barriers that prevent the spread of heterochromatin into
neighboring euchromatin (Ayoub et al., 2003). The maintenance of the
heterochromatin region between these barriers relies on the heterochromat-
in association protein Swi6 and methylation patterns on H3, in particular
Lys-9 (distal) and Lys-4 (proximal) (Ayoub et al., 2003; Zofall and Grewal,
2006).
Ayoub et al. (2003) showed that inactivation of Epe1 enhances hetero-
chromatization; while overexpression causes disruption of heterochromatin
and centeromere function. Zofall and Grewal (2006) subsequently showed
that Epe1 localizes with and antagonizes the heterochromatin protein
1 homologue, Swi6, and that direct interaction with Swi6 is required for
localization (Zofall and Grewal, 2006). Isaac et al. (2007) showed that the
Epe1 interacts with the heterochromatin assembly machinery to help regu-
late gene expression in S. pombe (Isaac et al., 2007).
The role of Epe1 in a function other than demethylation of histone
lysine residues indicates that demethylation is not the only role for the
JmjC domain. The ability for the JmjC domain to have alternative roles
has been highlighted by this subgroup as many of its members have been
shown to have roles outside of the classical histone demethylase function.
3.2.1.2 KDM3: JHDM2 (JMJD1) subgroup

Of the histone demethylating JmjC domain-containing proteins, there is
little known about the KDM3 subgroup. In mammals this group contains
four proteins: KDM3A (TSGA/JHDM2A/JMJD1A), KDM3B (5qCNA/
JHDM2B/JMJD1B), KDM3C (TRIP8/JHDM2C/JMJD1C), and hairless
(a protein known to be involved in congenital alopecia) (Cloos et al.,
2008; Klose et al., 2006). The domain architecture of this subgroup consists
of the C-terminal JmjC domain and a modified zinc finger domain in the
central region of the sequence (Figure 10(A)) (Klose et al., 2006). Later pa-
pers no longer show the modified zinc finger (Figure 10(B)), but no expla-
nation, as to why the domain is no longer shown, has yet been published
(Cloos et al., 2008; Pedersen and Helin, 2010). This omission could be
due to poor similarities or incorrect identification of the domain in initial
analysis. In our own analysis of members of this subgroup (KDM3A,
KDM3B, and KDM3C human and mouse) using the Simple Modular Anal-
ysis Research Tool database (SMART) we were unable to identify the
modified zinc finger domain (Letunic et al., 2006; Schultz et al., 1998).
The best studied member of this subgroup is KDM3A; the other members
remain poorly understood.
3.2.1.2.1 KDM3A The KDM3 family of proteins remains poorly stud-

ied even though they were one of the first families identified with the ability
to demethylate histones. KDM3A was originally identified as a male germ
line-specific transcript and therefore given the designation of testis-specific
gene A (TSGA) (Hoog et al., 1991). The demethylase activity was discov-
ered in a series of experiments using G9a-methylated histone substrates
(G9a is a methyltransferase) during a parallel study with KDM2A
(JHDM1A) and was then renamed JmjC domain-containing histone deme-
thylase 2A (JHDM2A) (Yamane et al., 2006). Yamane et al. (2006) showed
that KDM3A is a H3K9 (me2/1)-specific demethylase in vitro and in vivo,
but is unable to demethylate trimethylated H3K9 (Tian and Fang, 2007;
Yamane et al., 2006). It was shown that both the JmjC domain and the re-
gion of the putative zinc finger are essential for the demethylase activity
(Cloos et al., 2008; Yamane et al., 2006).
The H3K9 methylation mark has been shown to be a key negative
regulator in transcription and heterochromatin formation so that a demethy-
lase for this mark will antagonize any associated gene silencing (Martin and
Zhang, 2005; Yamane et al., 2006). Thus Yamane et al. (2006) showed that
a knockdown of KDM3A caused a downregulation of genes responsible for
differentiation and pluripotency in mice (Yamane et al., 2006). This has also
been shown by KDM3A involvement in spermatogenesis and embryonic
development in mice through interaction with the transcription factor
ER71 (Knebel et al., 2006). Okada et al. (2007) reported that KDM3A-
deficient male mice had abnormal chromatin condensation leading to low
mature sperm cell counts (Okada et al., 2007). Loh et al. (2007) showed
that maintenance of the H3K9 methylation state is important in maintaining
the plasticity and pluripotency of mouse embryonic stem cells by allowing
Figure 10 KDM3 subgroup of histone demethylases. (A) Schematic diagram of the

domain architecture of the KDM3 subgroup of JmjC domain-containing proteins.
This subgroup is characterized by the presence of a zinc finger-like binding domain
setting it apart from the JmjC-domain-only subgroup. (B) A multiple sequence align-
ment of the prominent members of the KDM3 subgroup. Conserved residues for
a-KG and Fe binding are indicted. A striking feature of this subgroup is the presence
of a large insertion (boxed) not found in other subgroups. (Modified from Figure 9 of
Klose et al. (2006).). hs, Homo sapiens; mm, Mus musculus; dm, Drosophila melanogaster.
the permissive transcription state of low H3K9 methylation at promoter sites

through the binding of the Oct4 transcription factor (Loh et al., 2007). The
regulation of H3K9 methylation has also been linked to smooth muscle cell
differentiation (Lockman et al., 2007).
As mentioned above, there are four closely related proteins in this family,
one being KDM3C (TRIP8). KDM3C was originally identified as a thyroid
hormone receptor-interacting protein, leading Yamane et al. (2006) to
consider KDM3A demethylase activity as a possible means of transcriptional
regulation by nuclear receptors. KDM3A has an LXXLL motif that has been
shown to be involved in nuclear hormone receptor interactions (Heery
et al., 1997). Subsequently KDM3A was shown to associate with the
androgen receptor (AR) in a ligand-dependent manner (Cloos et al.,
2008; Tian and Fang, 2007; Yamane et al., 2006), thereby linking this partic-
ular demethylase activity with hormone-dependent transcriptional activity
(Tian and Fang, 2007).
Recent studies have also indicated that KDM3A has a role in regulating
hypoxia-inducible genes through interaction with transcription factors that
are targeted to KDM3A under hypoxic conditions (Krieg et al., 2010; Lim
et al., 2010; Pollard et al., 2008; Sar et al., 2009; Wellmann et al., 2008).
Hypoxic conditions have been linked to enhanced tumor growth (Lim
et al., 2010). Hypoxia is commonly found in solid tumors which also restrict
the access of anticancer drugs and allow a selective environment for aggres-
sive cancer cells (Sar et al., 2009). KDM3A has been shown to maintain
some demethylase activity even under severe hypoxic conditions (Beyer
et al., 2008). The regulation of transcription has not only been shown in
tumor cells in hypoxic conditions but also macrophages (Tausendschon
et al., 2011). During wound healing, immune cells such as macrophages
need to adapt to hypoxic environments through alteration in gene expres-
sion patterns. Under these conditions macrophage KDM3A is down regu-
lated and therefore H3K9 me2/me3 levels are increased switching off
pathways not required under those conditions (Tausendschon et al., 2011).
3.2.1.2.2 KDM3B The second member of this family to be studied is

KDM3B. Very little is known about this member of the subgroup. It was
first identified during a screen of proteins on chromosome 5 which are
frequently deleted in malignant myeloid disorders (including myelodysplasia
and acute myeloid leukemia) (Clausen et al., 2009; Hu et al., 2001). From
this first identification, KDM3B was shown to have the conserved JmjC
domain and as a potential histone demethylase would have the ability to
regulate gene expression (Hu et al., 2001). The loss of KDM3B (5qNCA)
from tumor cells could thus lead to the indiscriminate growth seen in these
conditions. When expressed in trans in cells which have the region encoding
KDM3B deleted, the protein was shown to suppress tumor growth (Hu
et al., 2001). This suggests that it may act as a tumor suppressor. The role
in tumor suppression has been further elucidated with studies showing
that KDM3B may act as a corepressor for the retinoic acid receptor and
act in complex with c-Myb to modulate neutrophil differentiation
(Kravarusic et al., 2004; Westbrook et al., 2000). To confirm this role,
further studies will need to be conducted. Mikhaleva et al. (2011) have sug-
gested recently that KDM3B may have a role in delta sleep-induced peptide
production (Mikhaleva et al., 2011). This would provide a novel function
for this and other members of the JmjC HDM family.
3.2.1.2.3 KDM3C Though the first identified member of this family

was KDM3C (TRIP8), it is possibly the least studied of the family. It was
identified by Katoh and Katoh (2003) using a bioinformatic approach.
KDM3C (TRIP8) belongs to the thyroid hormone receptor b-binding pro-
tein family. 15 proteins were identified in this particular family, but TRIP8
itself remained uncharacterized. Using the bioinformatic approach, Katoh
and Katoh (2003) showed that TRIP8 contained TRI8H1 and TRI8H2
domains as well as the previously uncharacterized JmjC domain (Katoh
and Katoh, 2003). A BLASTP search showed that TRIP8 was homologous
to 5qCNA and TSGA based on the JmjC domain (Katoh and Katoh, 2003).
These two proteins were later designated as KDM3A (JHDM2A) and
KDM3B (JHDM2B).
Since its first identification, little work has been carried out on KDM3C.
Like the other members of this subgroup, KDM3C has the conserved cata-
lytic amino acids required for histone demethylase activity, however, as yet,
this predicted catalytic activity has not been verified experimentally (Cloos
et al., 2008). Wolf et al. (2007b) showed that a small splice variant of
KDM3C is a selective coactivator of the AR. The variant protein is shorter
than the reported full protein, and differs from that protein in 55aa (Wolf
et al., 2007b). This variant was also identified in a bioinformatic study carried
out by Katoh and Katoh (2007). They showed that both variants were
expressed in undifferentiated ES cells (Katoh and Katoh, 2007).
Wolf et al. (2007b) showed that the tissue distribution of the smaller
variant of KDM3C coincides largely with that of AR and responds to the
alteration of systemic levels of androgen within the tissues in which it is
located. As well as this, they showed that this variant is decreased in breast
cancer tumors by approximately 20e60%. This suggests that it may have
a role in the regulation of genes involved in the inhibition of tumor forma-
tion (Wolf et al., 2007b).
In a recent study, Kim et al. (2010) have shown that KDM3C interacts
with the histone methyltransferase WHISTLE in the mouse testis during
steroidogenesis, the process whereby cholesterol is converted in the mito-
chondrial inner membrane into testosterone (Kim et al., 2010). It was shown
in differential timescale occupancy of the promoter of the steroidogenic
marker p450c17, that KDM3C (JMJD1C) and WHISTLE occupy the
promoter to regulate transcription via interaction with SF-1 (steroidogenic
factor 1) during the development of the testis (Kim et al., 2010). They sug-
gest that a possible transcriptional regulatory mechanism during develop-
ment of the mouse testis occurs via a coordinated regulation of both
histone methylation and demethylation (Kim et al., 2010). This is the first
reference to KDM3C being a H3K9me2/me1 demethylase, and again it
was shown that both the JmjC and zinc finger-like domains were essential
for this function (Kim et al., 2010).
3.2.1.3 KDM4: JHDM3/JMJD2 subgroup

The third group of the JmjC domain-containing proteins to be described as
having histone demethylase properties are the KDM4 (JHDM3/JMJD2)
subgroup. Members of this subgroup have been shown to be conserved
from yeast to mammals (Figure 11) (Klose et al., 2006; Marmorstein and
Trievel, 2009). The KDM4 subgroup consists of six members (JMJD2A,
JMJD2B, JMJD2C, JMJD2D, JMJD2E, and JMJD2F) first identified in silico
using the TRIP8 JmjC-domain sequence as a query in an RPS-BLAST
search and the novel sequences so obtained in subsequent TBLASTN
searches (Katoh and Katoh, 2004).
KDM4A-C (JMJD2A-C) were the first identified. These proteins all
contained a similar domain structure with both JmjC and JmjN domains
as well as conserved TUDOR and PH domains (Figure 11(A)) (Katoh
and Katoh, 2004). This group also revealed a new domain (JD2H) in these
proteins, however to date, nothing is known about the properties of this
particular domain and it has not been described elsewhere (Katoh and
Katoh, 2004).
The next set of KDM4/JMJD2 genes to be identified was JMJD2D-F.
The encoded proteins differed from KDM4A-C in that they not only
lack the TUDOR and PH domains but also are not encoded within the
Figure 11 Members of the KDM4 subgroup. (A) A schematic diagram of the domain ar-
chitecture of several KDM4 members from different organisms. It can be seen in higher
eukaryotes that they have a similar domain architecture, whereas in more phylogenet-
ically distant lineages, there is a greater divergence in key domains other than the
conserved JmjN and JmjC domains. (B) A multiple sequence alignment of the JmjC
domain from the selected organisms showing the highly conserved residues for bind-
ing Fe(II) and a-KG. hs, Homo sapiens; mm, Mus musculus; dm, Drosophila melanogaster;
ce, Caenorhabditis elegans; sc, Saccharomyces cerevisiae. Modified from Figure 7 of Klose
et al. (2006).
KDM4A-C gene cluster found in humans (Katoh and Katoh, 2004).

JMJD2D, JMJD2E and JMJD2F have only the JmjN and JmjC domains
in common with the rest of their subgroup, but they nonetheless share
the high sequence homology found among all the proteins in the group
(Katoh and Katoh, 2004). It was later reported that JMJD2E and JMJD2F
appear to be pseudogenes and that KDM4A-D exhibit the histone demethy-
lase activity (Katoh and Katoh, 2004; Shin and Janknecht, 2007).
These four main proteins are only found in higher organisms. Nonethe-
less, KDM4 subgroup members lacking the TUDOR and PH domains have
also been identified in yeast, flies, and worms (Klose et al., 2006). The other
members in the KDM4 subgroup are poorly studied in higher organisms,
with little information beyond putative roles in demethylating the differing
states of H3K9 and H3K36. KDM4B and KDM4C have similar domain
structures to that of KDM4A. They may have similar functions but are
expressed at different times, and their function may also be dependent on
other activator proteins being present. Unlike the other members of the
key group in this family, KDM4D lacks the PH and Tudor domains (Shin
and Janknecht, 2007), but it is unclear whether this affects the protein’s abil-
ity to demethylate its target histone residues in vivo.
3.2.1.3.1 KDM4A (JMJD2A) KDM4A is one of the most studied pro-

teins of all the KDM4 family. It was the first identified member to have his-
tone demethylase activity targeted against H3K9 and H3K36 (Whetstine
et al., 2006). The dual specificity was previously unidentified in any other
members of the JHDM proteins (Mosammaparast and Shi, 2010). It is
the presence of the dual domains that has been suggested to be essential in
the specificity of KDM4A. The protein is found both in the nucleus
and the cytosol where it is able to associate with different complexes to
remove methyl marks in its targets, including histone deacetylases and the
N-CoR (Nuclear Receptor Corepressor) (Gray et al., 2005; Zhang et al.,
2005).
To date KDM4A is the only member of this subgroup to have a high
resolution structure published. The protein has a catalytic core that consists
of all its major domains: JmjN, JmjC, a zinc finger domain in its C-terminus,
and a b-hairpin (this along with a mixed domain is used to connect the JmjN
and JmjC domains) (Figure 12) (Chen et al., 2007; Couture et al., 2007;
Mosammaparast and Shi, 2010; Ng et al., 2007). This structure provides
the necessary stability to the JmjC domain catalytic core and is therefore
required for the activity of this protein. The structure of KDM4A is also
Figure 12 Crystal structure of KDM4A (JMJD2A) bound to H3K36me3 and N-oxalygly-

cine. This is the first high resolution crystal structure of a JmjC domain-containing pro-
tein. The JmjC domain with its characteristic eight antiparallel b barrel is highlighted by
the shaded ellipse and the other features are as indicated. (Modified from Figure 1(A) of
Chen et al. (2007).). JmjC, Jumonji C-terminal domain; JmjN, Jumonji N-terminal domain;
NOG, N-oxalylglycine.
important in the ability of this protein to have dual specificity and is selective
toward the trimethyl state of H3K9 and H3K36 (Couture et al., 2007).
The dual TUDOR domains, which have been shown to bind to H3K4
and H4K20, are suggested to be involved in targeting the protein to all sub-
strates for demethylation to occur (Huang et al., 2006; Mosammaparast and
Shi, 2010). These domains when bound to H3K4me3 form a bilobal,
saddlelike structure (Figure 13) which is suggested to allow the protein to
bind to chromatin enriched with either H3K4me3 or H4K20me3, but
this is yet to be established (Huang et al., 2006).
KDM4A has been shown to be involved in the cell cycle through its ac-
tivity in demethylating H3K36me3. The chromatin state is highly regulated
throughout replication and therefore it is important to maintain the state to
allow for proper cell cycle progression. It has been shown that chromatin
accessibility is a critical determinant in the timing of DNA replication.
KDM4A has been shown to be able to antagonize heterochromatin protein
1 (HP1) family member HP1g and thereby help cells to maintain their cell
cycle progression (Black et al., 2010). The ability of KDM4A to interact
with N-CoR also has links to the ability of this protein to affect cell prolif-
eration and differentiation (Zhang et al., 2005). The role of specific
Figure 13 Crystal structure of the TUDOR domains in KDM4A (bound to H3K4me3). A

ribbon representation of the two TUDOR domains of KDM4A in complex with H3K4
peptide (ball and stick model). The two TUDOR motifs are indicated as light (first)
and dark (second), flattened arrows (b-strands). It is suggested that this structure allows
for the dual specificity of KDM4A with respect to H3K4 and H4K20. Modified from
Figure 1(C) of Huang et al. (2006).
demethylases in key areas of cell cycle and DNA replication suggests that
alterations in function may lead to disease states.
3.2.1.3.2 KDM4B (JMJD2B) Like KDM3A (JMJD1A), KDM4B has

been shown to have a role in the regulation of transcription under hypoxic
conditions (Beyer et al., 2008; Pollard et al., 2008). KDM4B is able to antag-
onize the methylation state of H3K9 as are other members of this subgroup.
Fodor et al. (2006) showed that KDM4B acts at the pericentric region in
cells, specifically the H3K9me3 state (Fodor et al., 2006). They also showed
that it was necessary to have a functional JmjN, as well as a functional JmjC,
domain to carry out this activity.
The ability of KDM4B to act under hypoxic conditions has led to a sug-
gested role of this demethylase in tumor invasion and metastasis through
interaction with HIF (hypoxia-inducible factor) (Beyer et al., 2008; Pollard
et al., 2008). HIF subunit 1a is able to interact with specific marks to upre-
gulate the activity of KDM4B under hypoxic conditions. This is contrary to
what is known about the activity of other JmjC domain-containing proteins,
which have been shown to lose their activity under these conditions (Beyer
et al., 2008). It may be that loss of H3K9 or H3K36 methylation is important
in the stability of genomic transcripts under these conditions.
3.2.1.3.3 KDM4C (JMJD2C/GASC1) KDM4C was originally identi-

fied in a screen of genes in a region 9p23-24 that is often deleted in many
malignancies including espohageal cancer (Yang et al., 2000). Cloos et al.
(2006) showed that like other members of this subgroup, it is able to de-
methylate the histone mark H3K9me3/me2. By incubating KDM4C
with various histones and evaluating the methylation status they were able
to show that H3K9 was the most effected by KDM4C activity and that
this applied to both me3 and me2 states (Cloos et al., 2006).
KDM4C is also able to interact with LSD1 to form a complex with it and
the AR (Wissmann et al., 2007). In a series of experiments carried out by
Wissmann et al. (2007), it was shown that KDM4C directly interacts with
the amino-terminal, DNA-binding, and ligand-binding domains of the
AR. This, in conjunction with the binding of LSD1, allows for the assembly
of a multiple-specificity demethylase complex (Wissmann et al., 2007). It
was shown that inhibition of either LSD1 or KDM4C effects the prolifera-
tion of prostate tumors, therefore providing a possible therapeutic strategy
for the control of AR activity in diseased cells (Wissmann et al., 2007).
A number of studies have shown that KDM4 family members are over-
expressed in different cancers (Ponnaluri et al., 2009). This would be consis-
tent with their roles in making genes accessible through the alteration in
methylation state. It has been shown that JmjC domain-containing proteins
of this family do not act only on histone substrates (Ponnaluri et al., 2009).
The ability to act on substrates aside from histones, may play an important
role in the way that these proteins effect cell cycle progression and differen-
tiation leading to disease states by altering levels of transcription (Ponnaluri
et al., 2009).
3.2.1.3.4 Yeast KDM4 members Two members of this family have

been identified in the budding yeast, S. cerevisiae (Gis1 and Rph1) and a sin-
gle member in the fission yeast S. pombe ( jmjd2) (Huarte et al., 2007; Tu
et al., 2007). The S. cerevisiae proteins were first identified in a screen for pro-
teins involved in transcriptional repression of DNA damage in response to
UV radiation (Jang et al., 1999). They, like other members of this subgroup,
contain both the JmjC and JmjN domains as well as two conserved C5HC2
zinc finger domains (Klose et al., 2006). Gis1 is involved in DNA repair,
transcription and sumoylation (Tronnersjo et al., 2007). These processes
rely on the presence of both the JmjN and JmjC domains (Tronnersjo
et al., 2007). Rph1 has been shown to be able to demethylate both
H3K9 and H3K36 methyl marks like other members of this family (Klose
et al., 2007a). However, the H3K9 demethylase properties of Rph1 seem
to be due to the promiscuity of this protein rather than direct targeting to
this methylation mark (Klose et al., 2007a). Unlike other members of this
subgroup, Rph1 does not form complexes with other proteins but a homo-
tetramer (Klose et al., 2007a).
3.2.1.4 KDM5: JARID subgroup

Probably one of the most widely studied of all the JmjC domain-containing
groups, this cluster of proteins is broken into two subgroups in mammalsd
JARID1 consisting of four proteins: KDM5A ( JARID1A/RBP2), KDM5B
( JARID1B/PLU-1), KDM5C ( JARID1C/SMCX), and KDM5D (JARID1D/
SMCY) and JARID2 consisting currently of a single protein Jumonji
( JARID2) (Cloos et al., 2008; Klose et al., 2006). Proteins belonging to
the JARID group have been shown to be conserved from amoeba to
humans with a single member identified in Dictyostelium discoideum (Accari,
2012). Many of these proteins have been linked to defects in cellular differ-
entiation and cancer (Agger et al., 2008; Blair et al., 2011).
The KDM5 subgroup has been linked to the demethylation of tri- and
dimethylation of H3K4 (Christensen et al., 2007; Iwase et al., 2007; Klose
et al., 2007b; Lee et al., 2007a; Tahiliani et al., 2007; Yamane et al.,
2007). H3K4 methylation has been linked to transcriptionally active regions
of chromatin; therefore, it is suggested that the removal of methyl marks acts
as transcriptional repression (Cloos et al., 2008; Klose et al., 2006). The
removal of H3K4me3/me2 has been shown to recruit other repressive chro-
matin modifiers thereby turning off transcription of the target regions (Cloos
et al., 2008; Klose et al., 2006).
The domain architecture of JARID1 members encompasses JmjC, JmjN,
ARID/Bright, and a zinc finger as well as one or more PH domains
(Figure 14) (Cloos et al., 2008; Klose et al., 2006). The domain architecture
of JARID2 is slightly different with the loss of the PH domain as well as, in
many cases, the zinc finger domain (Figure 14).
3.2.1.4.1 KDM5A (JARID1A/RBP2) and Lid JARID1A (RBP2) has

been shown to be involved in the demethylation of H3K3me3/me2
(Agger et al., 2008; Cloos et al., 2008). However, there is little known
about the physiological roles of this protein. Experiments carried out by
Christensen et al. (2007) showed the roles of not only KDM5A (RBP2),
but also KDM5C (SMCX) and KDM5B (PLU-1) in the ability to
demethylate trimethylated H3K4. Prior to this, no protein had been
identified as removing such trimethyllysine marks and this revealed yet
another role of the JmjC domain in histone demethylation (Christensen
et al., 2007).
Figure 14 KDM5 subgroup of histone demethylases. (A) Domain architecture of some

of the members of the JARID1/JARID2 subgroup. (B) Multiple sequence alignment of
the JmjC domain from some of the members of the JARID1 subgroup. Conserved cat-
alytic residues for Fe(II) binding and a-KG binding are indicated. As can be seen not all
members retain the necessary amino acids for catalytic activity suggesting that they
may have roles aside from histone demethylation. hs, Homo sapiens; dm, Drosophila
melanogaster; ce, Caenorhabditis elegans; sp, Schizosaccharomyces pombe; sc, Saccharo-
myces cerevisiae. Modified from Figure 6 of Klose et al. (2006).
KDM5A was originally identified as a potential binding partner of the

pRB binding protein, and the resulting complex formed is associated
with signals that affect cell cycle exit and induction of differentiation
(Benevolenskaya, 2007). The orthologue of KDM5A in D. melanogaster
is Lid (Little Imaginal Disc), a member of the trithorax-group chromatin
modifiers (Eissenberg et al., 2007). Like KDM5A, Lid shares all the same
domains and therefore provides an excellent model to study the effects
of this protein in vivo (Blair et al., 2011).
Both Lid and KDM5A have been confirmed to affect H3K4me3/me2
methylation states (Benevolenskaya, 2007; Eissenberg et al., 2007).
Beneveloskaya (2007) showed that KDM5A was important in the regulation
of genes that provide control over the decision to withdraw from cell cycle
and differentiation through interaction with pRB, and repression of pro-
moters (Benevolenskaya, 2007; Lopez-Bigas et al., 2008). Recent work
has shown that KDM5A has a role in NOTCH signaling through interac-
tion with RBP-J (Liefke et al., 2010). The multiple pathways in which
KDM5A is involved highlight the role of this protein in the removal of
methylation at genes essential for cellular differentiation and fate. Lid also
is able to interact with NOTCH signaling in vivo to limit growth and
tumorigenesis (Liefke et al., 2010; Secombe et al., 2007).
KDM5A plays conflicting roles in cancer. As a potential blocker of cell
cycle progression in complex with pRB, it acts as a potential tumor suppres-
sor. However, recent studies have shown that it may also have a role in drug
resistance (Lim et al., 2010). It has been shown that KDM5A is overex-
pressed in some gastric and cervical cancers. KDM5A also interacts with
other proteins involved in oncogenesis such as p107, TBP, Myc, and
some nuclear receptors (Blair et al., 2011). Loss of KDM5A dramatically
inhibits tumorigenesis in a cancer mouse model. With these apparently con-
flicting results, further work needs to be conducted to clarify the role of
KDM5A and loss of H3K4 methylation in diseased cell states.
3.2.1.4.2 KDM5B (JARID1B/PLU-1) KDM5B was identified in a study

looking at genes that are upregulated in breast cancer (Lu et al., 1999). The
protein is localized to the nucleus and its expression is low in normal adult
tissue, but in breast cancer and breast cancer cell lines there is higher expres-
sion (Barrett et al., 2002, 2007; Lu et al., 1999). KDM5B was also shown to
be downregulated when the tyrosine kinase, HER2, was inhibited, but
expression was nonetheless shown to be independent of HER2 expression
in breast cancer (Barrett et al., 2007).
Like other members of this subgroup, KDM5B has been shown to have
H3K4 demethylase activity. This activity, like KDM5A, has been linked to
transcriptional repression as H3K4me3/me2 is a transcriptional active mark
(Yamane et al., 2007). The demethylase activity of KDM5B plays an impor-
tant role in proliferation of breast cancer through direct repression of the
tumor suppressor gene BRAC1 (Barrett et al., 2007; Scibetta et al., 2007;
Yamane et al., 2007). It has also been shown to affect the expression of other
genes implicated in breast cancers including CAV1 and HOXA5 (Cloos
et al., 2008; Yamane et al., 2007).
KDM5B is also important in directly regulating cell fate decisions
through its ability to control cell cycle, cell differentiation, and cell lineage
(Dey et al., 2008). This indicates the key role for KDM5B in cells. Through
its ability to regulate the expression of key genes, KDM5B is able to deter-
mine cell fate by altering histone methylation at genes specific for pathways
to remain pluripotent, become progenitor cells, or continue through the cell
cycle (Dey et al., 2008). Figure 15 shows the possible mechanisms outlined
by Dey et al. (2008) as to how KDM5B and other members of this subgroup
are able to effect cellular decisions.
KDM5B has been shown to be a potent transcriptional repressor. This is
through not only its ability to demethylate H3K4 but also through its inter-
actions with histone deacetylases class I and II. It has also been shown to
interact with the transcriptional corepressor N-CoR to mediate repression
of tumor suppressor genes (Barrett et al., 2007). The interaction with
N-CoR, however, is indirect rather than the direct interaction seen with
the deacetylases (Barrett et al., 2007).
KDM5B has also been linked to prostate cancer by Xiang et al. (2007a)
who showed that KDM5B is able to demethylate all three methyl states of
H3K4 and through this activity plays a role in the regulation of AR
transcriptional activity. It remains unclear whether this activity promotes
prostate cancer development and progression (Xiang et al., 2007a). Apart
from these effects on various cancers, the normal biological roles of
KDM5B are unknown.
3.2.1.4.3 KDM5C (JARID1C/SMCX) and KDM5D (JARID1D/SMCY)

The genes encoding KDM5C (JARID1C/SMCX) and KDM5D
(JARID1D/SMCY) are found on the X and Y chromosomes of mammals,
respectively. Little is known about the biological roles of these proteins in
cells. Like other members of the KDM5 subgroup, they have been shown
to have H3K4me3/me2 demethylase activity (Cloos et al., 2008). Unlike
Figure 15 A model for the function of KDM5B in early development. (A) In the process
of differentiation KDM5B is downregulated, it is suggested that other members of this
subgroup are upregulated to allow for various subsets of target genes to be expressed
in terminally differentiated cells. (B) In stem cells, KDM5B may be directed to target
genes based on extracellular signals; this allows the cell to remain pluripotent. From
Figure 8 of Dey et al. (2008).
many genes located on the X chromosome, KDM5C is able to escape

X-inactivation, therefore results in dimorphic expression in females (Xu
and Andreassi, 2010). It remains unclear as to why this occurs.
KDM5C has been shown to be highly expressed in brain tissue (Xu et al.,
2008a). Mutations in KDM5C have been linked to X-linked mental retar-
dation through aberrant histone methylation and neuronal survival and den-
dritic development (Iwase et al., 2007). However, this is not the only
phenotype, KDM5C mutations have also been linked to autism spectrum
disorder (ASD) as well as X-linked short stature and hyperreflexia related
to mental retardation (Abidi et al., 2008; Adegbola et al., 2008). To date,
more than 20 mutations have been linked to KDM5C, loss of demethylase
activity, and the resulting phenotypes of mental retardation (Xu and
Andreassi, 2010). This shows the direct link between loss of demethylase
activity and X-linked mental retardation (Metzger and Schule, 2007).
Further analysis is needed to determine the possible significance of the sexu-
ally dimorphic KDM5C expression in the brain.
The role of KDM5C has been further elucidated by a link between
RE-1 silencing transcription factor (REST) and KDM5C. Mutations in
KDM5C prevent the recruitment of REST and therefore impair REST-
mediated neuronal gene regulation (Tahiliani et al., 2007). Tahiliani et al.
(2007) were able to isolate a complex from HeLa cells showing that
KDM5C was specifically associated with the transcriptional repressor
E2F6 and its partners DPI, Max, and Mga; NCoR1; REST as well as His-
tone Deacetylase 1 (HDAC1) and HDAC2. They also showed interaction
with G9a, heterochromatin protein 1g (HP1g). Together this points to
the role of KDM5C as a transcriptional repressor (Tahiliani et al., 2007).
This indicates that when mutations occur in KDM5C and it no longer is
able to associate with its binding partners, repression is no longer present,
leading to active transcription at incorrect times with the outcome in the
brain of mental retardation.
KDM5C has also been linked to the repression of Smad3 expression.
Smad3 is known to be a mediator of transforming growth factor-b
(TGF-b), which is important in cell differentiation, growth, and prolifera-
tion (Kim et al., 2008). The link with Smad3 has been suggested to point
to KDM5C having potential to be an oncoprotein, in that the repression
of Smad3 expression has been shown to increase the potential of developing
colorectal cancer through obstructions in the TGF-b pathway (Kim et al.,
2008). The potential role of KDM5C in this pathway needs to be further
investigated.
The KDM5D gene is found on the Y chromosome of mammals and the

protein has the conserved domain architecture of other members of this
subgroup. However, little is known about the biological function of this
protein. It contains the conserved residues required for histone demethyla-
tion. There has been some work showing a possible interaction with Poly-
comb-like proteins such as Ring6a, but further work, to understand the role
this protein has in cells, needs to be conducted (Cloos et al., 2008). There is
also a potential link between KDM5D and prostate cancer, in that 52% of
human prostate cancers carry a deletion of the KDM5D gene (Blair et al.,
2011). Further work needs to be conducted to clarify the role of KDM5D.
3.2.1.4.4 JARID2-Jumonji Although it was the first of the JmjC

domain proteins to be discovered, Jumonji (JARID2) is not as well studied
as other members. Jumonji was first identified in a gene trap experiment
carried out by Takeuchi et al. (1995) in mice. They found a protein involved
in neural tube formation, naming it Jumonji due to the morphology of the
abnormal neural groove formed that appeared to resemble a cross (Jumonji
translated to cruciform in Japanese) (Takeuchi et al., 1995). Toyoda et al.
(2000) showed that Jumonji localizes to the nucleus of cells and has a nega-
tive role on cell proliferation of cardiac myocytes and megakaryocytes, as
well as neural tube cells (Toyoda et al., 2000).
Unlike other members of the JARID subgroup, Jumonji appears to lack
histone demethylase activity due to the lack of the necessary conserved
residues in the JmjC domain for catalytic activity (Figure 14) (Cloos et al.,
2008; Klose et al., 2006; Shirato et al., 2008). A study conducted by Shirato
et al. (2008) has shown that Jumonji is able to interact with the histone
methyltransferase that mediates H3K9 methylation. This was shown to
repress cyclin D1 transcription by acting on the promoter region and had
the effect of repressing cell proliferation of cardiac myocytes during devel-
opment (Shirato et al., 2008). This is consistent with the previous finding
that Jumonji negatively regulates cell proliferation. As with Epe1 in yeast,
this shows that lack of catalytic activity does not exclude key roles of these
proteins in transcriptional activation or repression.
3.2.1.5 KDM6: UTX/UTY/JMJD3 subgroup

The KDM6 group of JmjC domain-containing proteins are also linked to
histone demethylation, in particular H3K27 demethylation. Trimethylation
of H3K27 is considered to be a very important epigenetic mark. It has been
considered important for maintaining the pluripotency and plasticity of
embryonic stem cells during embryonic development as well as Polycomb-

mediated gene silencing and X chromosome inactivation (Hong et al.,
2007). Therefore the demethylases able to antagonize this particular mark
may be expected to have a significant role in the development of embryos.
The KDM6 group consists of three proteinsdUTX (ubiquitously tran-
scribed tetratricopeptide repeat on the X chromosome), UTY (ubiquitously
transcribed tetratricopeptide repeat on the Y chromosome), and JMJD3
(Hong et al., 2007). Both UTX and UTY have a domain architecture con-
sisting of the JmjC domain as well as six tetratricopeptide repeats (TPR)
(Figure 16(A)) (Hong et al., 2007; Klose et al., 2006). These two proteins
exhibit 88% homology across the entire protein (Figure 16(B)) (Hong
et al., 2007; Klose et al., 2006).
The third member of this family, JMJD3 lacks the TPR domain found in
the other two members of this family (Figure 16(A)), but has been shown to
have significant homology not only in the region of the JmjC domain but
also areas in the rest of the protein (Figure 16(B)). This broad sequence
similarity places JMJD3 in this group rather than the JmjC only, where it
would otherwise be classified based on domain architecture (Hong et al.,
2007). There have also been four proteins identified in C. elegans that
have significant homology to this family, but like JMJD3 they also lack
the TPR domain (Figure 16) (Agger et al., 2007). The high homology of
these proteins to the identified proteins in mammals suggests that the roles
of these proteins may be similar to those of UTX/UTY/JMJD3 (Klose
et al., 2006). These proteins all contain the JmjC domain and have been
shown to have the conserved residues required of the binding of the cofac-
tors Fe(II) and a-ketoglutarate known to be necessary for these proteins to
function as histone demethylases (Klose et al., 2006). UTY has been shown
to be widely but not ubiquitously expressed in many male tissues, however
to date, no activity has been assigned to it (Cloos et al., 2008; Hong et al.,
2007; Lan et al., 2007). Therefore the focus of this subsection will be on
KDM6A (UTX) and KDM6B (JMJD3).
3.2.1.5.1 KDM6A (UTX) KDM6A (UTX), as the original name states,

is located on the X chromosome in mammals. The X-linked UTX in fe-
males is not subject to X-inactivation and has been shown to be ubiquitously
expressed in both mice and humans (Hong et al., 2007; Klose et al., 2006). It
is highly expressed in brain tissue but the significance of this remains to be
elucidated (Xu et al., 2008b). It was one of the first proteins to be identified
with H3K27 demethylase activity (Hong et al., 2007; Lan et al., 2007;
Figure 16 KDM6 subgroup of histone demethylases. (A) Schematic domain architec-

ture of members of this subgroup. (B) A multiple sequence alignment of the JmjC
domain of the conserved members of this subgroup. Key conserved amino acids for
Fe(II) binding and those required for a-KG binding are indicated. hs, Homo sapiens;
mm, Mus musculus; dm, Drosophila melanogaster; ce, Caenorhabditis elegans. Modified
from Figure 8 of Klose et al. (2006).
Swigut and Wysocka, 2007). H3K27me3/me2 is usually a mark for tran-

scriptionally inactive chromatin (Swigut and Wysocka, 2007). KDM6A
demethylase activity is specific for H3K27 and does not affect the levels of
other key methylation marks (Lee et al., 2007b).
KDM6A has been shown to interact with Hox genes to regulate expres-
sion during development (Agger et al., 2007; Lee et al., 2007b). Agger et al.
(2007) showed that KDM6A had limited demethylase activity toward me3 in
vitro, but did show that in vivo there was a marked increase in H3K27me3
when KDM6A levels were decreased. This suggests that for correct demethy-
lase activity, KDM6A requires other cofactors aside from Fe(II) and a-KG to
carry out its activity (Agger et al., 2007). From their experiments, Agger et al.
(2007) suggest that KDM6A is important in maintaining steady state
H3K27me3/me2 levels in proliferating cells. KDM6A is part of the MLL2
methyltranferase complex suggesting an elegant method of removing a
repressive mark while simultaneously depositing an active mark to lead to
activation of target genes (Agger et al., 2007; Cho et al., 2007).
The mutually antagonistic methyltransferase and KDM6A demethylase
activities are reminiscent of the mutually antagonistic functions of the Poly-
comb-group (PcG) and the Trithorax (TrxG) transcriptional regulators.
PcG-mediated repression of chromatin and TrxG-mediated induction
work antagonistically to facilitate tissue-specific patterning of gene expres-
sion during development. This requires the presence of H3K27me3/me2
marks and alterations which will change the expression of genes involved
in proliferation (Seenundun et al., 2010). The action of KDM6A is a key
player in these pathways to allow correct expression of genes during cell di-
vision, differentiation, and proliferation. It has been shown to be important
in regulating myogenesis (Seenundun et al., 2010), embryonic development
(Gao et al., 2010), and b-globin expression (Hosey et al., 2010) to name a
few examples.
Inactivating mutations of KDM6A have been identified in multiple
cancer types, including multiple myeloma, breast and colorectal cancers,
oesophageal squamous cell carcinoma, renal cell carcinoma, myeloid leuke-
mia, and glioblastoma (Blair et al., 2011). KDM6A affects the expression of
Notch pathway proteins which, in turn, controls the expression of Rb
(a known tumor suppressor) thereby affecting tumor formation (Herz
et al., 2010; Tsai et al., 2010; Wang et al., 2010).
3.2.1.5.2 KDM6B (JMJD3) KDM6B (JMJD3) is highly homologous to

the other members of this subgroup, however, as previously stated, it lacks
the TPR domains found in UTX and UTY (Agger et al., 2007). The
expression pattern of KDM6B remains undetermined. The JmjC domain
of KDM6B contains a lysine to arginine substitution in the second
a-ketoglutarate binding site. Since this substitution involves an amino acid
with very similar charge properties, it is considered unlikely to cause an
abrogation of cofactor binding and KDM6B has in fact been shown to still
be a functional enzyme (Klose et al., 2006).
Like KDM6A, KDM6B has been found to be associated with Hox gene
expression during development (Agger et al., 2007). Agger et al. (2007)
showed that KDM6B was able to demethylate H3K27me3 and me2 but
not me1. This activity has also been shown by others (Hong et al., 2007;
Xiang et al., 2007b) and, like that of KDM6A, is suggested to play roles
in cellular development, differentiation, and proliferation (Agger et al.,
2007). Consistent with such roles, KDM6B is involved in the INK4A-ARF
locus which is known to be important in the regulation of cell senescence
(Agger et al., 2009; Barradas et al., 2009). KDM6B expression in relation
to this locus is induced by the RAS-RAF pathway, suggesting functions
in oncogenic stress responses (Agger et al., 2009; Barradas et al., 2009).
KDM6B is also associated with PcG silencing (De Santa et al., 2007)
linking it, like KDM6A, to tissue-specific regulation of gene expression
and control of self-renewing tissues in mammals (Sen et al., 2008). The
removal of the H3K27me3 marks on promoters is related to cellular differ-
entiation and loss of PcG binding (Sen et al., 2008). KDM6B has also been
associated with the inflammatory response again in relation to the loss of
H3K27me3 and silencing of members of the PcG of proteins (De Santa
et al., 2007). KDM6B also contributes to the control of gene expression
in macrophages that have been activated in response to bacterial infection
(De Santa et al., 2009). This shows the potency of H3K27 methylation
and demethylation in various pathways.
3.2.1.6 KDM7: PHF2/PHF8/KIAA1718 subgroup

The KDM7 or PHD subgroup of the JmjC domain-containing proteins is
found from worms to higher eukaryotes. These proteins contain an
N-terminal PHD zinc finger in addition to the JmjC domain but no other
identifiable domains (Fortschegger and Shiekhattar, 2011; Suganuma and
Workman, 2010). The crystal structure of the PHD and JmjC domains in
these proteins has now been solved which opened insights into the function
of these enzymes (Fortschegger and Shiekhattar, 2011). In addition to the
conserved zinc chelating residues of the PHD, a patch of phenylalanine
and tyrosine residues were identified in an aromatic cagelike structure

(Fortschegger and Shiekhattar, 2011). These residues are able to interact
with methylated lysine. Similar aromatic regions are not limited to this pro-
tein family and have been shown in some other domains (Fortschegger and
Shiekhattar, 2011).
PHDs belong to a class of zinc fingers and mediate specific proteine
protein interactions. Several chromatin-associating factors containing this
domain have been shown to specifically interact with trimethylated H3K4
(Feng et al., 2010). The PHD is distinct from other known zinc finger motifs
and has been thought to belong to transcriptional regulators, consistent with
the fact that many of the JmjC domain-containing proteins have also been
identified to contain this domain (Hasenpusch-Theil et al., 1999).
The KDM7 subgroup consists of three membersdKDM7C (PHF2),
KDM7B (PHF8), and KDM7A (KIAA1718). Phylogenetic analysis and
alignment studies have shown that this subgroup is most closely related to
KDM2 subgroup (Cloos et al., 2008; Klose et al., 2006). KDM7C and
KDM7B were initially identified because their absence causes severe
mental retardation and the facial deformity of either cleft palate or lip or
both. Both proteins, but in particular KDM7B, have been linked to
X-linked mental retardation including Siderius-Hamel CL/P syndrome
(Abidi et al., 2007). KDM7A was identified due to its high homology to
KDM7B and KDM7C (Klose et al., 2006).
Little is known about the biological role of KDM7C. It lacks the
conserved catalytic amino acids required for histone demethylation (Cloos
et al., 2008) and is ubiquitously expressed. In mouse models, it is concentrated
in neuronal tubes and root ganglia (Klose et al., 2006). KDM7C plays a role in
mental retardation, but this has not been linked to any catalytic activity of the
JmjC domain (Hasenpusch-Theil et al., 1999). As is the case with Epe1 in S.
pombe, which lacks the catalytic triad, the JmjC domain of KDM7C may have
an as yet undefined activity aside from histone lysine demethylation. One
recent study identified a potential H3K9me1 demethylase activity for
KDM7C, based on the binding of the PHD to H3K4me2/3 (Wen et al.,
2010). However, the ability to demethylate this residue was not confirmed
to be carried out by the JmjC domain in this study (Wen et al., 2010).
3.2.1.6.1 KDM7B (PHF8) KDM7B is the best studied protein of

KDM7 subgroup. It was initially identified due to its role in X-linked
mental retardation along with its fellow member KDM7C (Laumonnier
et al., 2005). Since then KDM7B has been shown to have several potential
roles within cells, in the activation of genes involved in the transcription of

rRNA through their ability to bind to H3K4 trimethylation sites and the
demethylation of H3K9 mono-and dimethylation states. The ability to
demethylate H4K20/H3K9 in zebra fish allows KDM7B to alter cellular
differentiation and also plays a role in cell cycle progression (Feng et al.,
2010; Liu et al., 2010; Qi et al., 2010; Qiu et al., 2010). These functions
have been linked to the presence of both the PHD and JmjC domain.
Like all members of this subgroup, KDM7B has the characteristic
domain architecture with the PHD in the N-terminal region of the protein
followed by the JmjC domain in the C-terminal region (Figure 17) (Klose
et al., 2006). The first studies carried out identified PHF8 as having a signif-
icant role in mental retardation, particularly in relation to the cleft lip/palate
Figure 17 KDM7 subgroup of histone demethylases. (A) A schematic diagram of the

domain architecture of some of the members of KDM7. (B) A multiple sequence align-
ment of the JmjC domain from KDM7. Fe(II) binding residues and a-KG binding residues
are indicated. Hs, Homo sapiens; mm, Mus musculus; ce, Caenorhabditis elegans. Modified
from Figure 5 of Klose et al. (2006).
X-linked mental retardation (XLMR) (Abidi et al., 2007; Koivisto et al.,

2007; Laumonnier et al., 2005). This was due to a mutation or loss of func-
tion of the JmjC domain. Following these studies little else was discovered
about the role of PHF8 in cells until recently.
Although the JmjC domain of KDM7B had an established role in the
XLMR and the oral cleft phenotypes, little was understood about the
molecular mechanisms of how this was carried out. Only recently did Yu
et al. (2010) establish that KDM7B is a histone demethylase. They identified
the target of KDM7B as H3K9me2/1 and showed that the presence of the
PHD is likely to be required for accurate substrate binding (Yu et al., 2010).
Yue et al. (2010) also published the crystal structure of KDM7B showing the
presence of deep binding pockets to allow for substrate binding (Yu et al.,
2010; Yue et al., 2010). The mutation linked to XLMR and the oral cleft
phenotypes was shown to alter the substrate binding within the pocket
(Yue et al., 2010).
The KDM7B domain structure is the key to understanding the activity
of the enzyme. It reveals that the PHD finger binds to trimethylated H3K4
thereby allowing efficient demethylation of H3K9. It is suggested that this
binding is an anchor to allow for positioning of the active site of the JmjC
domain in the correct orientation for insertion of the methylated histone res-
idue (Feng et al., 2010). It is the combination of both the JmjC domain and
the PHD that allows KDM7B to carry out its function. Without either of
these domains, it would lose its functions as shown by the experiments car-
ried out with mutants containing point mutations in either or both domains
(Feng et al., 2010).
KDM7B has been suggested to be involved in the regulation of the tran-
scription of rDNA genes (Feng et al., 2010). KDM2B (JHDM1B/FBXL10)
was the first of the JmjC domain-containing proteins to have been linked
with this process (Frescas et al., 2007) (See Section 3.2.1.1.1). This may
be through its ability to demethylate H3K4. KDM7B was determined
experimentally to be enriched in the nucleoli, strengthening the case that
it is involved in demethylating histone residues and in the transcription of
rDNA genes (Feng et al., 2010).
Feng et al. (2010) showed that KDM7B associated with hypomethylated
rDNA and PolI to mediate the transcription of these genes. To understand
the mechanism behind KDM7B-dependant rDNA transcription, mutants of
KDM7B with point mutations in both the PHD and JmjC domain were
analyzed. These mutants showed that both domains need to be functional
to allow for this protein to play its role in both demethylation and binding.
Furthermore, the protein loses function with the introduction of a single

base substitution in the JmjC domain (F279S) that mimics the mutant allele
in families with X-linked mental retardation and cleft lip/palate (Feng et al.,
2010). This mutant protein was no longer able to activate PolI transcription,
suggesting that impaired PolI activity contributes to the phenotype.
KDM7B has also been shown to have histone demethylase activity
similar to that of many of the JmjC domain-containing proteins. Specifically
KDM7B has been linked to the demethylation of H3K9me2 (Zhu et al.,
2010). This demethylase activity has been linked to its role in transcription
of rDNA by PolII (Zhu et al., 2010). H3K9 demethylation is an important
epigenetic modification shown to be linked with areas of silenced euchro-
matin (unlike others linked with the formation of heterochromatin) and
associated with transcriptional silencing (Zhu et al., 2010).
3.2.1.6.2 KDM7A (KIAA1718) The second protein in the KDM7 sub-

group to be studied is KDM7A (KIAA1718). This protein is conserved from
worms to humans and is probably best studied in these two organisms. The
crystal structure has been solved for KDM7A in C. elegans and humans, and
like KDM4, it has dual specificity based on structural elements (Hou and Yu,
2010; Krishnan et al., 2011; Suganuma and Workman, 2010; Yang et al.,
2010). It was shown that KDM7A was able to demethylate H3K9me2
and H3K27me2 where K4me3 marks are absent (Krishnan et al., 2011).
The presence of K4me3 virtually abolishes K9me2 demethylation and stim-
ulates K27me2 demethylation (Krishnan et al., 2011). This is a result of how
the protein folds and the distance between the PHD aromatic cage and the
JmjC domain active site (Horton et al., 2010; Krishnan et al., 2011; Yang
et al., 2010; Yu et al., 2010; Yue et al., 2010).
The demethylase activity of KDM7A was first demonstrated by Tsukada
et al. (2010). They showed that the dual specificity of KDM7A was impor-
tant in the transcription of genes specific to the nervous system. Using ortho-
logs from Danio rerio (zebra fish) they showed that KDM7A localized to the
brain and tail bud. This led them to look at KDM7A in mice and they
showed that as in D. rerio, KDM7A is found in the brain and neuronal cells
(Tsukada et al., 2010). The methylation marks H3K9 and H3K27 are linked
to heterochromatin. The ability of KDM7A to antagonize these marks sug-
gests that it will have an active role in the induction of genes required for
neuronal and brain formation.
The role of KDM7A in brain tissue was further elucidated by Yokoyama
et al. (2010). They showed that the loss of H3K9me2 and H3K27me2 leads
to a transcriptionally active state and that KDM7A interacts with transcrip-

tion factors such as KAP1 in brain tissue to promote transcription
(Yokoyama et al., 2010). This was then expanded by Huang et al. (2010),
who showed that KDM7A is important in regulating neuronal cell differen-
tiation in mice (Huang et al., 2010). This role is mediated by affecting the
histone methylation state and thus the activity of the FGF4 (a signal mole-
cule involved in neural differentiation) promoter. The FGF4 signaling
pathway is required for correct neuronal fate determination (Huang et al.,
2010).
The KDM7 subgroup plays key roles in the regulation of neural path-
ways. This is highlighted by the fact that all three members have been shown
to have either direct or indirect effects leading to phenotypes such as mental
retardation. Whether this is directly through the demethylation of histone
marks regulating transcription of key factors for neural pathways and brain
formation or whether these proteins also have as yet unidentified roles in
complexes that regulate these pathways is still to be determined.
3.2.1.7 JmjC-domain-only
An increasing number of JmjC proteins contain a JmjC domain as the
only recognizable sequence homology domain. However, their functions
are largely unknown.
3.2.1.7.1 JMJD6, is it the first true arginine demethylase? Histone

methylation is not limited to lysine residues. Arginine residues on histones
3 and 4 have also been shown to be able to be methylated (Agger et al.,
2008; Klose and Zhang, 2007). These are usually found on H3R3,
H3R8, H3R17, H3R26, and H4R3 (Agger et al., 2008; Klose and Zhang,
2007). Unlike lysine residues, the nature of arginine allows not only for
mono- or dimethylation to occur but also for dimethylation, which can
occur in either di-symmetrical or di-asymmetrical orientations (Klose and
Zhang, 2007). Like lysine methylation, histone arginine methylation status
has been linked to both transcriptional activation and repression. However
to date, no true demethylase has been identified that can antagonize these
marks (Agger et al., 2008; Klose and Zhang, 2007).
JMJD6 was the first member of the JmjC-domain-only subgroup to be
described as a histone demethylase. As the name of the group suggests, the
only identifiable domain is the JmjC domain. JMJD6 was initially identified
as a phosphatidyl-serine receptor involved in apoptosis and the engulfment
of dead cells in zebra fish. However, sequence analysis revealed the presence
of the JmjC domain which was inconsistent with the protein being a recep-
tor (Fadok et al., 2000; Fadok and Henson, 2003; Hong et al., 2010; Li et al.,
2003; Somersan and Bhardwaj, 2001). Furthermore in knockdown mice,
the clearance rate of apoptotic cells remained the same even though the
levels of JMJD6 were severely decreased (Bose et al., 2004). JMJD6 has
been shown to localize to the nucleus of cells and this eliminated it as a
possible membrane receptor protein (Cikala et al., 2004; Cui et al., 2004;
Wolf et al., 2007a).
In the knowledge that JMJD6 is localized to the nucleus and contains a
JmjC domain, work was carried out to determine if, like other members of
this superfamily, it was able to demethylate histone residues. Chang et al.
(2007) were the first group to assign a demethylase activity to JMJD6,
however, unlike the other members of this family, they showed that it
was able to antagonize arginine rather than lysine methylation (Chang
et al., 2007). They showed that JMJD6 was able to demethylate
H4R3me2 (symmetrical) and to a lesser extent H4R3me1 (Chang et al.,
2007). They also showed that JMJD6 has some affinity for H3R2me2 but
had no effect on any of the other methylated arginine residues tested (Chang
et al., 2007). Unfortunately other groups have been unable to reproduce
these results (Hong et al., 2010).
There have been reports that instead of being an arginine demethylase,
JMJD6 is able to hydroxylate lysine residues. Webby et al. (2009) and
Hong et al. (2010) showed that JMJD6 was able to hydroxylate lysine resi-
dues specifically on the tail of U2AF65 (a protein associated with RNA
splicing), suggesting a potential role in the regulation of mRNA splicing
(Hahn et al., 2010; Hong et al., 2010; Webby et al., 2009). Structural anal-
ysis of JMJD6 has strengthened the argument that JMJD6 is a hydroxylase
rather than a demethylasedthe position of the JmjC domain and the pres-
ence of a helix-turn-helix motif separated by an inflexible region containing
two proline residues suggests that mobility to allow substrate access is limited
(Hong et al., 2010; Mantri et al., 2010). The current available evidence thus
indicates that JMJD6 may not be an arginine demethylase as first reported.
Consequently the search continues for a genuine arginine demethylase.
3.2.1.7.2 KDM8 KDM8 (previously identified as JMJD5) was the

second member of the JmjC-domain-only subgroup to be identified with
putative histone demethylase activity (Hsia et al., 2010). It was shown that
KDM8 has H3K36me2 demethylase activity and by this means is able to
regulate cyclin A1 transcription in MCF7 breast cancer cells. Hsia et al.
(2010) used two strains expressing wild-type KDM8 and a point mutant
H321A, respectively. They showed that 60% of cells expressing the mutated
protein exhibited a substantial H3K36me2 increase compared to cells with
the wild-type protein. They went on to show that KDM8 is critical for
MCF7 cell proliferation. In healthy cells KDM8 expression is relatively
low, however, in tumor cells, there is a significant increase in expression.
This increased expression compared to normal tissue control was not limited
to breast tumors but was also shown in thyroid, adrenal, bladder, uterine,
and liver tumors (Hsia et al., 2010).
KDM8 demethylase activity has been linked to cell proliferation through
its ability to affect cell cycle progression at the G2/M checkpoint. This is
mediated by KDM8 being recruited to the promoter of cyclin A1 and deme-
thylating H3K36me2 to allow cyclin A1 transcription. Cyclin A1 is required
for the progression of cells through G2/M phase (Hsia et al., 2010). KDM8 is
the first JmjC-domain-only subgroup member to be shown to have deme-
thylase activity, opening the way for further functional studies on this group.
4. PLANT JMJC HISTONE DEMETHYLATION

4.1 Roles of Nondemethylating JmjC Domain-
Containing Proteins
Among this diverse family of proteins, there are members that have
been shown not to be involved with the demethylation process, although
they contain the conserved JmjC domain with the key functional elements
to allow for enzymatic activity. This is the largest of the JmjC domain fam-
ilies. Unlike the rest of the family, the majority of the members of this sub-
group still have unknown functions. The only discernible domain in these
proteins is the JmjC domain, which is usually located C-terminally in
conjunction with a large N-terminal region. This arrangement is, however,
not universal. Some members of this group still have a significant C-terminal
region following the JmjC domain. Members of this subgroup include FIH
(factor inducing hypoxic inducible factor 1a), NO66 (a nucleolar protein),
Mina53 (MYC-induced nuclear antigen 53), and Phospholipase A2b to
name a few. Most functional studies on these proteins do not relate to the
presence of the JmjC domain. Consequently histone demethylation remains
the only biochemical function known for the JmjC domain.
FIH is one of the better studied of the JmjC-domain-only proteins. It has
been shown to be important in responses to hypoxia. Like the histone
demethylases, the JmjC domain of FIH has the characteristic double-

stranded b barrel structure as well as the key conserved amino acids shown
to be required for catalytic activity (Elkins et al., 2003). However, as this
protein is not a demethylase, the function of its JmjC domain is unknown.
4.2 Plant JmjC Histone Demethylases

Phylogenetic analysis has been carried out to study the evolutionary history
of animal and fungal JmjC domain-containing proteins. This led to identi-
fication of seven different subgroups already discussed. However, like many
eukaryotes, plants also pack their DNA around histones and use many of the
same proteins as other eukaryotes. Due to the presence or absence of
different chromosomes, there are some clear differences between plants
and animals and fungi (i.e., the lack of the metazoan X and Y chromosomes
in Viridiplantae removes the need for the sex-specific demethylases). An
initial phylogenetic study was carried out using 20 rice and 21 Arabidopsis se-
quences (Sun and Zhou, 2008). It showed that the JmjC domain proteins of
plants align with animal and fungal proteins in three of the already identified
subgroups (KDM3/JHDM2, KDM5/JARID and JmjC only) (Sun and
Zhou, 2008; Zhou et al., 2010). However, there are also plant-specific
groups of JmjC domain-containing proteins (Figure 18) (Sun and Zhou,
2008; Zhou and Ma, 2008).
A more systematic phylogenetic analysis of JmjC domain-containing
proteins in Arabidopsis and rice was conducted by Lu et al. (2008) and
reviewed by Chen et al. (2011). Their results showed that plants contain
putative histone demethylases for H3K4, H3K9, H3K9/H3K36, and
H3K36 (Chen et al., 2011; Lu et al., 2008). Like the demethylases in mam-
mals that regulate cellular differentiation and the cell cycle, many of these
demethylases regulate key pathways in floral organ development, circadian
regulation, flowering time, and signals related to specific tissue expression
(Chen et al., 2011). One example is JMJ14, which is involved in the silencing
of RNA signals, cell to cell movement of RNA silencing signals, and control
of flowering time (Lu et al., 2010a; Searle et al., 2010). Other examples are
ELF6 and REF6, both of which are involved in brassinosteroid regulation
of plant growth and development (Noh et al., 2004; Yu et al., 2008).
Some of the plant proteins have also been shown to contain two plant-
specific domains (FYRN and FRYC). These are located in proteins which,
in other organisms would normally have PHDs. It is suggested that these
play a similar role in chromatin binding and targeting to PHD target sites
Figure 18 Phylogenetic relationships of JmjC domain-containing proteins from

humans, rice and Arabidopsis. (A) A neighbor joining tree using the JmjC region.
Following Klose et al. (2006), Zhou and Ma (2008) used domain architecture and phylo-
genetic alignment to define the 12 subgroups shown here. Representative domain ar-
chitecture is shown next to each subgroup. (Modified from Figure 6 of Zhou and Ma
(2008).). (B) A phylogenetic tree showing the five plant-specific subgroups in relation
to the seven subgroups found in other organisms. (Modified from Figure 1 of Sun and
Zhou (2008).).
in mammals (Lu et al., 2008). Thus it appears that plants have adapted to the
specific needs of their biology by the loss and gain of particular domains.
5. CONCLUSIONS
JmjC domain-containing proteins are widespread throughout all or-
ganisms excluding the Archaea (Zhou and Ma, 2008). This shows the
conserved and important nature of this domain. The JmjC domain appears
to be responsible for the catalytic function of the proteins mentioned here,

but they depend on other domains to target them to specific sites or sub-
strates to be able to carry out their function. In some cases, they also require
additional protein binding partners to allow the correct targeting of these
proteins to their substrate. However, the function of the JmjC domain
outside the role of histone demethylation is still poorly understood. There
is no obvious evidence so far to suggest why specific JmjC domain proteins
cannot act on targets outside the nucleus. The JmjC domain has been shown
to exhibit functional activities other than histone demethylation, however,
further studies will need to be conducted to understand the role of this
domain in proteins in these and other potential pathways in eukaryotic cells.
REFERENCES
Abidi, F.E., et al., 2008. Mutations in JARID1C are associated with X-linked mental retar-
dation, short stature and hyperreflexia. J. Med. Genet. 45, 787e793.
Abidi, F.E., et al., 2007. A novel mutation in the PHF8 gene is associated with X-linked
mental retardation with cleft lip/cleft palate. Clin. Genet. 72, 19e22.
Accari, S.L., 2012. Characterisation of Kame, a Novel JmjC domain-containing protein in
Dictyostelium discoideum., Department of Microbiology, (Ph.D.), La Trobe University,
Melbourne.
Adegbola, A., et al., 2008. A novel mutation in JARID1C/SMCX in a patient with autism
spectrum disorder (ASD). Am. J. Med. Genet. A 146A, 505e511.
Agger, K., et al., 2008. The emerging functions of histone demethylases. Curr. Opin. Genet.
Dev. 18, 159e168.
Agger, K., et al., 2009. The H3K27me3 demethylase JMJD3 contributes to the activation of
the INK4A-ARF locus in response to oncogene- and stress-induced senescence. Genes
Dev. 23, 1171e1176.
Agger, K., et al., 2007. UTX and JMJD3 are histone H3K27 demethylases involved in HOX
gene regulation and development. Nature 449, 731e734.
Anand, R., Marmorstein, R., 2007. Structure and mechanism of lysine-specific demethylase
enzymes. J. Biol. Chem. 282, 35425e35429.
Ayoub, N., et al., 2003. A novel jmjC domain protein modulates heterochromatization in
fission yeast. Mol. Cell Biol. 23, 4356e4370.
Barradas, M., et al., 2009. Histone demethylase JMJD3 contributes to epigenetic control of
INK4a/ARF by oncogenic RAS. Genes Dev. 23, 1177e1182.
Barrett, A., et al., 2002. PLU-1 nuclear protein, which is upregulated in breast cancer, shows
restricted expression in normal human adult tissues: a new cancer/testis antigen? Int. J.
Cancer 101, 581e588.
Barrett, A., et al., 2007. Breast cancer associated transcriptional repressor PLU-1/JARID1B
interacts directly with histone deacetylases. Int. J. Cancer 121, 265e275.
Benevolenskaya, E.V., 2007. Histone H3K4 demethylases are essential in development and
differentiation. Biochem. Cell Biol. Biochim. Biol. Cell. 85, 435e443.
Beyer, S., et al., 2008. The histone demethylases JMJD1A and JMJD2B are transcriptional
targets of hypoxia-inducible factor HIF. J. Biol. Chem. 283, 36542e36552.
Black, J.C., et al., 2010. Conserved antagonism between JMJD2A/KDM4A and HP1gamma
during cell cycle progression. Mol. Cell 40, 736e748.
Blair, L.P., et al., 2011. Epigenetic regulation by lysine demethylase 5 (KDM5) enzymes in
cancer. Cancers 3, 1383e1404.
Bose, J., et al., 2004. The phosphatidylserine receptor has essential functions during embryo-
genesis but not in apoptotic cell removal. J. Biol. 3, 15.
Chang, B.S., et al., 2007. JMJD6 is a histone arginine demethylase. Science 318, 444e447.
Chen, X., et al., 2011. Epigenetic gene regulation by plant Jumonji group of histone
demethylase. Biochim. Biophys. Acta 1809, 421e426.
Chen, Z.Z., et al., 2007. Structural basis of the recognition of a methylated histone tail by
JMJD2A. Proc. Natl. Acad. Sci. U.S.A. 104, 10818e10823.
Cho, Y.W., et al., 2007. PTIP associates with MLL3-and MLL4-containing histone H3
lysine 4 methyltransferase complex. J. Biol. Chem. 282, 20395e20406.
Christensen, J., et al., 2007. RBP2 belongs to a family of demethylases, specific for tri- and
dimethylated lysine 4 on histone 3. Cell 128, 1063e1076.
Cikala, M., et al., 2004. The phosphatidylserine receptor from Hydra is a nuclear protein with
potential Fe(II) dependent oxygenase activity. BMC Cell Biol. 5, 26.
Clausen, R.P., et al., 2009. Histone Demethylases. Epigenetic Targets in Drug Discovery.
Wiley-VCH Verlag GmbH & Co. KGaA, pp. 269e290.
Clifton, I.J., et al., 2006. Structural studies on 2-oxoglutarate oxygenases and related double-
stranded beta-helix fold proteins. J. Inorg. Biochem. 100, 644e669.
Cloos, P.A., et al., 2006. The putative oncogene GASC1 demethylates tri- and dimethylated
lysine 9 on histone H3. Nature 442, 307e311.
Cloos, P.A.C., et al., 2008. Erasing the methyl mark: histone demethylases at the center of
cellular differentiation and disease. Genes Dev. 22, 1115e1140.
Couture, J.F., et al., 2007. Specificity and mechanism of JMJD2A, a trimethyllysine-specific
histone demethylase. Nat. Struct. Mol. Biol. 14, 689e695.
Cui, P., et al., 2004. Nuclear localization of the phosphatidylserine receptor protein via mul-
tiple nuclear localization signals. Exp. Cell Res. 293, 154e163.
Culhane, J.C., Cole, P.A., 2007. LSD1 and the chemistry of histone demethylation. Curr.
Opin. Chem. Biol. 11, 561e568.
Cuthbert, G.L., et al., 2004. Histone deimination antagonizes arginine methylation. Cell 118,
545e553.
De Santa, F., et al., 2009. Jmjd3 contributes to the control of gene expression in LPS-
activated macrophages. EMBO J. 28, 3341e3352.
De Santa, F., et al., 2007. The histone H3 lysine-27 demethylase Jmjd3 links inflammation to
inhibition of polycomb-mediated gene silencing. Cell 130, 1083e1094.
Dey, B.K., et al., 2008. The histone demethylase KDM5b/JARID1b plays a role in cell fate
decisions by blocking terminal differentiation. Mol. Cell Biol. 28, 5312e5327.
Dunwell, J.M., et al., 2001. Evolution of functional diversity in the cupin superfamily. Trends
Biochem. Sci. 26, 740e746.
Eissenberg, J.C., et al., 2007. The trithorax-group gene in Drosophila little imaginal discs
encodes a trimethylated histone H3 Lys4 demethylase. Nat. Struct. Mol. Biol. 14,
344e346.
Elkins, J.M., et al., 2003. Structure of factor-inhibiting hypoxia-inducible factor (HIF) reveals
mechanism of oxidative modification of HIF-1 alpha. J. Biol. Chem. 278, 1802e1806.
Fadok, V.A., et al., 2000. A receptor for phosphatidylserine-specific clearance of apoptotic
cells. Nature 405, 85e90.
Fadok, V.A., Henson, P.M., 2003. Apoptosis: giving phosphatidylserine recognition an
assistewith a twist. Curr. Biol. 13, R655eR657.
Fang, J., et al., 2007. The Saccharomyces cerevisiae histone demethylase Jhd1 fine-tunes the
distribution of H3K36me2. Mol. Cell. Biol. 27, 5055e5065.
Feng, W., et al., 2010. PHF8 activates transcription of rRNA genes through H3K4me3 bind-
ing and H3K9me1/2 demethylation. Nat. Struct. Mol. Biol. 17, 445e450.
Fodor, B.D., et al., 2006. Jmjd2b antagonizes H3K9 trimethylation at pericentric heterochro-
matin in mammalian cells. Genes Dev. 20, 1557e1562.
Forneris, F., et al., 2008. LSD1: oxidative chemistry for multifaceted functions in chromatin
regulation. Trends Biochem. Sci. 33, 181e189.
Fortschegger, K., Shiekhattar, R., 2011. Plant homeodomain fingers form a helping hand for
transcription. Epigenetics 6.
Frescas, D., et al., 2007. JHDM1B/FBXL10 is a nucleolar protein that represses transcription
of ribosomal RNA genes. Nature 450, 309e313.
Frescas, D., et al., 2008. KDM2A represses transcription of centromeric satellite repeats and
maintains the heterochromatic state. Cell Cycle 7, 3539e3547.
Gao, Y., et al., 2010. Regulation of H3K27me3 and H3K4me3 during early porcine embry-
onic development. Mol. Reprod. Dev. 77, 540e549.
Gray, S.G., et al., 2005. Functional characterization of JMJD2A, a histone deacetylase- and
retinoblastoma-binding protein. J. Biol. Chem. 280, 28507e28518.
Hahn, P., et al., 2008. Genomic structure and expression of Jmjd6 and evolutionary analysis
in the context of related JmjC domain containing proteins. BMC Genomics 9, 293.
Hahn, P., et al., 2010. Analysis of jmjd6 cellular localization and testing for its involvement in
histone demethylation. PLoS One 5, e13769.
Hasenpusch-Theil, K., et al., 1999. PHF2, a novel PHD finger gene located on human chro-
mosome 9q22. Mamm. Genome 10, 294e298.
He, J., et al., 2011. KDM2b/JHDM1b, an H3K36me2-specific demethylase, is required for
initiation and maintenance of acute myeloid leukemia. Blood 117, 3869e3880.
Heery, D.M., et al., 1997. A signature motif in transcriptional co-activators mediates binding
to nuclear receptors. Nature 387, 733e736.
Herz, H.M., et al., 2010. The H3K27me3 demethylase dUTX is a suppressor of Notch- and
Rb-dependent tumors in Drosophila. Mol. Cell Biol. 30, 2485e2497.
Hong, S.H., et al., 2007. Identification of JmjC domain-containing UTX and JMJD3 as his-
tone H3 lysine 27 demethylases. Proc. Natl. Acad. Sci. U.S.A. 104, 18439e18444.
Hong, X., et al., 2010. Interaction of JMJD6 with single-stranded RNA. Proc. Natl. Acad.
Sci. U.S.A. 107, 14568e14572.
Hoog, C., et al., 1991. Analysis of a murine male germ cell-specific transcript that encodes a
putative zinc finger protein. Mol. Reprod. Dev. 30, 173e181.
Horton, J.R., et al., 2010. Enzymatic and structural insights for substrate specificity of a family
of jumonji histone lysine demethylases. Nat. Struct. Mol. Biol. 17, 38e43.
Hosey, A.M., et al., 2010. Crosstalk between histone modifications maintains the develop-
mental pattern of gene expression on a tissue-specific locus. Epigenetics 5, 273e281.
Hou, H., Yu, H., 2010. Structural insights into histone lysine demethylation. Curr. Opin.
Struct. Biol. 20, 739e748.
Hsia, D.A., et al., 2010. KDM8, a H3K36me2 histone demethylase that acts in the cyclin A1
coding region to regulate cancer cell proliferation. Proc. Natl. Acad. Sci. U.S.A. 107,
9671e9676.
Hu, Z., et al., 2001. A novel nuclear protein, 5qNCA (LOC51780) is a candidate for the
myeloid leukemia tumor suppressor gene on chromosome 5 band q31. Oncogene 20,
6946e6954.
Huang, C., et al., 2010. Dual-specificity histone demethylase KIAA1718 (KDM7A) regulates
neural differentiation through FGF4. Cell Res. 20, 154e165.
Huang, Y., et al., 2006. Recognition of histone H3 lysine-4 methylation by the double tudor
domain of JMJD2A. Science 312, 748e751.
Huarte, M., et al., 2007. The fission yeast Jmj2 reverses histone H3 lysine 4 trimethylation.
J. Biol. Chem. 282, 21662e21670.
Isaac, S., et al., 2007. Interaction of Epe1 with the heterochromatin assembly pathway in
Schizosaccharomyces pombe. Genetics 175, 1549e1560.
Iwase, S., et al., 2007. The X-linked mental retardation gene SMCX/JARID1C defines a
family of histone H3 lysine 4 demethylases. Cell 128, 1077e1088.
Jang, Y.K., et al., 1999. RPH1 and GIS1 are damage-responsive repressors of PHR1. Mol.
Cell Biol. 19, 7630e7638.
Katoh, M., Katoh, M., 2003. Identification and characterization of TRIP8 gene in silico. Int.
J. Mol. Med. 12, 817e821.
Katoh, M., Katoh, M., 2004. Identification and characterization of JMJD2 family genes in
silico. Int. J. Oncol. 24, 1623e1628.
Katoh, M., Katoh, M., 2007. Comparative integromics on JMJD1C gene encoding histone
demethylase: conserved POU5F1 binding site elucidating mechanism of JMJD1C
expression in undifferentiated ES cells and diffuse-type gastric cancer. Int. J. Oncol.
31, 219e223.
Kim, S.M., et al., 2010. Regulation of mouse steroidogenesis by WHISTLE and JMJD1C
through histone methylation balance. Nucleic Acids Res. 38, 6389e6403.
Kim, T., Buratowski, S., 2007. Two Saccharomyces cerevisiae JmjC domain proteins demeth-
ylate histone H3 Lys(36) in transcribed regions to promote elongation. J. Biol. Chem.
282, 20827e20835.
Kim, T.D., et al., 2008. Repression of Smad3 activity by histone demethylase SMCX/
JARID1C. Biochem. Biophys. Res. Commun. 366, 563e567.
Klose, R.J., et al., 2007a. Demethylation of histone H3K36 and H3K9 by Rph1: a vestige of
an H3K9 methylation system in Saccharomyces cerevisiae? Mol. Cell Biol. 27, 3951e3961.
Klose, R.J., et al., 2006. JmjC-domain-containing proteins and histone demethylation. Nat.
Rev. Genet. 7, 715e727.
Klose, R.J., et al., 2007b. The retinoblastoma binding protein RBP2 is an H3K4
demethylase. Cell 128, 889e900.
Klose, R.J., Zhang, Y., 2007. Regulation of histone methylation by demethylimination and
demethylation. Nat. Rev. Mol. Cell Biol. 8, 307e318.
Knebel, J., et al., 2006. Repression of transcription by TSGA/Jmjd1a, a novel interaction
partner of the ETS protein ER71. J. Cell Biochem. 99, 319e329.
Koivisto, A.M., et al., 2007. Screening of mutations in the PHF8 gene and identification of a
novel mutation in a Finnish family with XLMR and cleft lip/cleft palate. Clin. Genet.
72, 145e149.
Koyama-Nasu, R., et al., 2007. The F-box protein Fbl10 is a novel transcriptional repressor
of c-Jun. Nat. Cell Biol. 9, 1074e1080.
Kravarusic, J., et al., 2004. The leukemia gene 5qNCA is a tissue-specific co-repressor of
retinoic acid receptor which interacts with c-Myb. In: AACR Meeting Abstracts,
972-a-.
Krieg, A.J., et al., 2010. Regulation of the histone demethylase JMJD1A by hypoxia-
inducible factor 1 alpha enhances hypoxic gene expression and tumor growth. Mol.
Cell Biol. 30, 344e353.
Krishnan, S., et al., 2011. Structure and function of histone H3 lysine 9 methyltransferases
and demethylases. Chembiochem 12, 254e263.
Lan, F., et al., 2007. A histone H3 lysine 27 demethylase regulates animal posterior
development. Nature 449, 689e694.
Lan, F., et al., 2008. Mechanisms involved in the regulation of histone lysine demethylases.
Curr. Opin. Cell Biol. 20, 316e325.
Laumonnier, F., et al., 2005. Mutations in PHF8 are associated with X linked mental retar-
dation and cleft lip/cleft palate. J. Med. Genet. 42, 780e786.
Lee, M.G., et al., 2007a. Physical and functional association of a trimethyl H3K4 demethylase
and Ring6a/MBLR, a polycomb-like protein. Cell 128, 877e887.
Lee, M.G., et al., 2007b. Demethylation of H3K27 regulates polycomb recruitment and
H2A ubiquitination. Science 318, 447e450.
Letunic, I., et al., 2006. SMART 5: domains in the context of genomes and networks.
Nucleic Acids Res. 34, D257eD260.
Li, M.O., et al., 2003. Phosphatidylserine receptor is required for clearance of apoptotic cells.
Science 302, 1560e1563.
Liefke, R., et al., 2010. Histone demethylase KDM5A is an integral part of the core Notch-
RBP-J repressor complex. Genes Dev. 24, 590e601.
Lim, S., et al., 2010. Epigenetic regulation of cancer growth by histone demethylases. Int. J.
Cancer 127, 1991e1998.
Lin, C.H., et al., 2008. Heterochromatin protein 1a stimulates histone H3 lysine 36 deme-
thylation by the Drosophila KDM4A demethylase. Mol. Cell 32, 696e706.
Liu, W., et al., 2010. PHF8 mediates histone H4 lysine 20 demethylation events involved in
cell cycle progression. Nature 466, 508e512.
Lockman, K., et al., 2007. The histone demethylase, Jmjd1a, interacts with the myocardin
factors to regulate SMC differentiation marker gene expression. Circ. Res. 101,
e115e23.
Loh, Y.H., et al., 2007. Jmjd1a and Jmjd2c histone H3 Lys 9 demethylases regulate self-
renewal in embryonic stem cells. Genes Dev. 21, 2545e2557.
Lopez-Bigas, N., et al., 2008. Genome-wide analysis of the H3K4 histone demethylase
RBP2 reveals a transcriptional program controlling differentiation. Mol. Cell 31,
520e530.
Lu, F., et al., 2010a. JMJ14 is an H3K4 demethylase regulating flowering time in Arabidopsis.
Cell Res. 20, 387e390.
Lu, F., et al., 2008. Comparative analysis of JmjC domain-containing proteins reveals
the potential histone demethylases in Arabidopsis and rice. J. Integr. Plant Biol. 50,
886e896.
Lu, P.J., et al., 1999. A novel gene (PLU-1) containing highly conserved putative DNA/
chromatin binding motifs is specifically up-regulated in breast cancer. J. Biol. Chem.
274, 15633e15645.
Lu, T., et al., 2009. Validation-based insertional mutagenesis identifies lysine demethylase
FBXL11 as a negative regulator of NFkappaB. Proc. Natl. Acad. Sci. U.S.A. 106,
16339e16344.
Lu, T., et al., 2010b. Regulation of NF-kappaB by NSD1/FBXL11-dependent reversible
lysine methylation of p65. Proc. Natl. Acad. Sci. U.S.A. 107, 46e51.
Mantri, M., et al., 2010. Crystal structure of the 2-oxoglutarate- and Fe(II)-dependent lysyl
hydroxylase JMJD6. J. Mol. Biol. 401, 211e222.
Marmorstein, R., Trievel, R.C., 2009. Histone modifying enzymes: structures, mechanisms,
and specificities. Biochim. Biophys. Acta 1789, 58e68.
Martin, C., Zhang, Y., 2005. The diverse functions of histone lysine methylation. Nat. Rev.
Mol. Cell Biol. 6, 838e849.
Metzger, E., Schule, R., 2007. The expanding world of histone lysine demethylases. Nat.
Struct. Mol. Biol. 14, 252e254.
Mikhaleva, I.I., et al., 2011. JmjC-domain-containing histone demethylases of the JMJD1B
type as putative precursors of endogenous DSIP. Peptides 32, 826e831.
Mosammaparast, N., Shi, Y., 2010. Reversal of histone methylation: biochemical and
molecular mechanisms of histone demethylases. Annu. Rev. Biochem. 79, 155e179.
Ng, S.S., et al., 2007. Crystal structures of histone demethylase JMJD2A reveal basis for
substrate specificity. Nature 448, 87e91.
Noh, B., et al., 2004. Divergent roles of a pair of homologous jumonji/zinc-finger-class tran-
scription factor proteins in the regulation of Arabidopsis flowering time. Plant Cell 16,
2601e2613.
Okada, Y., et al., 2007. Histone demethylase JHDM2A is critical for Tnp1 and Prm1 tran-
scription and spermatogenesis. Nature 450, 119e123.
Pedersen, M.T., Helin, K., 2010. Histone demethylases in development and disease. Trends
Cell Biol. 20, 662e671.
Pfau, R., et al., 2008. Members of a family of JmjC domain-containing oncoproteins

immortalize embryonic fibroblasts via a JmjC domain-dependent process. Proc. Natl.
Acad. Sci. U.S.A. 105, 1907e1912.
Pollard, P.J., et al., 2008. Regulation of Jumonji-domain-containing histone demethylases by
hypoxia-inducible factor (HIF)-1alpha. Biochem. J. 416, 387e394.
Polytarchou, C., et al., 2008. The JmjC domain histone demethylase Ndy1 regulates redox
homeostasis and protects cells from oxidative stress. Mol. Cell Biol. 28, 7451e7464.
Ponnaluri, V.K., et al., 2009. Identification of non-histone substrates for JMJD2A-C histone
demethylases. Biochem. Biophys. Res. Commun. 390, 280e284.
Qi, H.H., et al., 2010. Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain
and craniofacial development. Nature 466, 503e507.
Qiu, J., et al., 2010. The X-linked mental retardation gene PHF8 is a histone demethylase
involved in neuronal differentiation. Cell Res. 20, 908e918.
Sar, A., et al., 2009. Identification and characterization of demethylase JMJD1A as a gene
upregulated in the human cellular response to hypoxia. Cell Tissue Res. 337, 223e234.
Schneider, J., Shilatifard, A., 2006. Histone demethylation by hydroxylation: chemistry in
action. ACS Chem. Biol. 1, 75e81.
Schultz, J., et al., 1998. SMART, a simple modular architecture research tool: identification
of signaling domains. Proc. Natl. Acad. Sci. U.S.A. 95, 5857e5864.
Scibetta, A.G., et al., 2007. Functional analysis of the transcription repressor PLU-1/
JARID1B. Mol. Cell Biol. 27, 7220e7235.
Searle, I.R., et al., 2010. JMJ14, a JmjC domain protein, is required for RNA silencing
and cell-to-cell movement of an RNA silencing signal in Arabidopsis. Genes Dev. 24,
986e991.
Secombe, J., et al., 2007. The Trithorax group protein Lid is a trimethyl histone H3K4 deme-
thylase required for dMyc-induced cell growth [Review]. Genes Dev. 21, 537e551.
Seenundun, S., et al., 2010. UTX mediates demethylation of H3K27me3 at muscle-specific
genes during myogenesis. EMBO J. 29, 1401e1411.
Sen, G.L., et al., 2008. Control of differentiation in a self-renewing mammalian tissue by the
histone demethylase JMJD3. Genes Dev. 22, 1865e1870.
Shi, Y., 2007. Histone lysine demethylases: emerging roles in development, physiology and
disease. Nat. Rev. Genet. 8, 829e833.
Shi, Y., et al., 2004. Histone demethylation mediated by the nuclear amine oxidase homolog
LSD1. Cell 119, 941e953.
Shin, S., Janknecht, R., 2007. Diversity within the JMJD2 histone demethylase family.
Biochem. Biophys. Res. Commun. 353, 973e977.
Shirato, H., et al., 2008. A jumonji ( Jarid2) protein complex represses cyclin D1 expression
by methylation of histone H3-K9. J. Biol. Chem. 284, 733e739.
Somersan, S., Bhardwaj, N., 2001. Tethering and tickling: a new role for the phosphatidyl-
serine receptor. J. Cell Biol. 155, 501e504.
Suganuma, T., Workman, J.L., 2010. Features of the PHF8/KIAA1718 histone demethylase.
Cell Res. 20, 861e862.
Sun, Q., Zhou, D.X., 2008. Rice jmjC domain-containing gene JMJ706 encodes H3K9
demethylase required for floral organ development. Proc. Natl. Acad. Sci. U.S.A. 105,
13679e13684.
Swigut, T., Wysocka, J., 2007. H3K27 demethylases, at long last. Cell 131, 29e32.
Tahiliani, M., et al., 2007. The histone H3K4 demethylase SMCX links REST target genes
to X-linked mental retardation. Nature 447, 601e605.
Takeuchi, T., et al., 1995. Gene trap capture of a novel mouse gene, jumonji, required for
neural tube formation. Genes Dev. 9, 1211e1222.
Tanaka, Y., et al., 2010. JmjC enzyme KDM2A is a regulator of rRNA transcription in
response to starvation. EMBO J. 29, 1510e1522.
Tausendschon, M., et al., 2011. Hypoxia causes epigenetic gene regulation in macrophages
by attenuating Jumonji histone demethylase activity. Cytokine 53, 256e262.
Tian, X., Fang, J., 2007. Current perspectives on histone demethylases. Acta Biochim.
Biophys. Sin. (Shanghai) 39, 81e88.
Toyoda, M., et al., 2000. Jumonji is a nuclear protein that participates in the negative
regulation of cell growth. Biochem. Biophys. Res. Commun. 274, 332e336.
Trewick, S.C., et al., 2002. Oxidative demethylation by Escherichia coli AlkB directly reverts
DNA base damage. Nature 419, 174e178.
Trewick, S.C., et al., 2005. Methylation: lost in hydroxylation? EMBO Rep. 6, 315e320.
Tronnersjo, S., et al., 2007. The jmjN and jmjC domains of the yeast zinc finger protein Gis1
interact with 19 proteins involved in transcription, sumoylation and DNA repair. Mol.
Genet. Genomics 277, 57e70.
Tsai, M.C., et al., 2010. Tumor suppression by the histone demethylase UTX. Cell Cycle 9.
Tsukada, Y., et al., 2006. Histone demethylation by a family of JmjC domain-containing
proteins. Nature 439, 811e816.
Tsukada, Y., et al., 2010. KDM7 is a dual demethylase for histone H3 Lys 9 and Lys 27 and
functions in brain development. Genes Dev. 24, 432e437.
Tu, S.J., et al., 2007. Identification of histone demethylases in Saccharomyces cerevisiae. J. Biol.
Chem. 282, 14262e14271.
Wang, J.K., et al., 2010. The histone demethylase UTX enables RB-dependent cell fate
control. Genes Dev. 24, 327e332.
Wang, Y., et al., 2004. Human PAD4 regulates histone arginine methylation levels via
demethylimination. Science 306, 279e283.
Webby, C.J., et al., 2009. Jmjd6 catalyses lysyl-hydroxylation of U2AF65, a protein associ-
ated with RNA splicing. Science 325, 90e93.
Wellmann, S., et al., 2008. Hypoxia upregulates the histone demethylase JMJD1A via HIF-1.
Wen, H., et al., 2010. Recognition of histone H3K4 trimethylation by the plant homeodo-
main of PHF2 modulates histone demethylation. J. Biol. Chem. 285, 9322e9326.
Westbrook, C.A., et al., 2000. Novel nuclear receptor coactivator is a candidate for the
del(5q) leukemia tumor suppressor gene. Exp. Hematol. 28, 1492.
Whetstine, J.R., et al., 2006. Reversal of histone lysine trimethylation by the JMJD2 family of
histone demethylases. Cell 125, 467e481.
Wissmann, M., et al., 2007. Cooperative demethylation by JMJD2C and LSD1 promotes
androgen receptor-dependent gene expression. Nat. Cell Biol. 9, 347e353.
Wolf, A., et al., 2007a. Changing story of the receptor for phosphatidylserine-dependent
clearance of apoptotic cells. EMBO Rep. 8, 465e469.
Wolf, S.S., et al., 2007b. A novel variant of the putative demethylase gene, s-JMJD1C, is a
coactivator of the AR. Arch. Biochem. Biophys. 460, 56e66.
Xiang, Y., et al., 2007a. JARID1B is a histone H3 lysine 4 demethylase up-regulated in pros-
tate cancer. Proc. Natl. Acad. Sci. U.S.A. 104, 19226e19231.
Xiang, Y., et al., 2007b. JMJD3 is a histone H3K27 demethylase. Cell Res. 17, 850e857.
Xu, J., Andreassi, M., 2010. Reversible histone methylation regulates brain gene expression
and behavior. Horm. Behav. 59, 383e392.
Xu, J., et al., 2008a. Sex-specific expression of the X-linked histone demethylase gene Jarid1c
in brain. PLoS One 3, e2553.
Xu, J., et al., 2008b. Sex-specific differences in expression of histone demethylases Utx and
Uty in mouse brain and neurons. J. Neurosci. 28, 4521e4527.
Yamane, K., et al., 2007. PLU-1 is an H3K4 dernethylase involved in transcriptional repres-
sion and breast cancer cell proliferation. Mol. Cell 25, 801e812.
Yamane, K., et al., 2006. JHDM2A, a JmjC-containing H3K9 demethylase, facilitates tran-
scription activation by androgen receptor. Cell 125, 483e495.
Yang, Y., et al., 2010. Structural insights into a dual-specificity histone demethylase
ceKDM7A from Caenorhabditis elegans. Cell Res. 20, 886e898.
Yang, Z.Q., et al., 2000. Identification of a novel gene, GASC1, within an amplicon at 9p23-
24 frequently detected in esophageal cancer cell lines. Cancer Res. 60, 4735e4739.
Yokoyama, A., et al., 2010. KIAA1718 is a histone demethylase that erases repressive histone
methyl marks. Genes Cells 15, 867e873.
Yu, L., et al., 2010. Structural insights into a novel histone demethylase PHF8. Cell Res. 20,
166e173.
Yu, X., et al., 2008. Modulation of brassinosteroid-regulated gene expression by Jumonji
domain-containing proteins ELF6 and REF6 in Arabidopsis. Proc. Natl. Acad. Sci.
U.S.A. 105, 7618e7623.
Yue, W.W., et al., 2010. Crystal structure of the PHF8 Jumonji domain, an Nepsilon-methyl
lysine demethylase. FEBS Lett. 584, 825e830.
Zhang, D., et al., 2005. JMJD2A is a novel N-CoR-interacting protein and is involved in
repression of the human transcription factor achaete scute-like homologue 2 (ASCL2/
Hash2). Mol. Cell Biol. 25, 6404e6414.
Zhou, X., et al., 2010. Phylogenetic detection of numerous gene duplications shared by
animals, fungi and plants. Genome Biol. 11, R38.
Zhou, X., Ma, H., 2008. Evolutionary history of histone demethylase families: distinct evolu-
tionary patterns suggest functional divergence. BMC Evol. Biol. 8, 294.
Zhu, Z., et al., 2010. PHF8 is a histone H3K9me2 demethylase regulating rRNA synthesis.
Cell Res. 20, 794e801.
Zofall, M., Grewal, S.I., 2006. Swi6/HP1 recruits a JmjC domain protein to facilitate
transcription of heterochromatic repeats. Mol. Cell 22, 681e692.
CHAPTER SIX
New Insight into the Role of

Reactive Oxygen Species (ROS) in
Cellular Signal-Transduction
Processes
Eileen G. Russell and Thomas G. Cotter*
Tumour Biology Laboratory, School of Biochemistry and Cell Biology, Bioscience Research Institute,
University College Cork, Cork, Ireland
*Corresponding author: E-mail: t.cotter@ucc.ie
Contents
1. Introduction 222
2. Background 224
2.1 Source of ROS 224
2.2 Antioxidant Systems 226
2.3 ROS Signaling 226
2.3.1 ROS as signaling molecules 226
2.3.2 Mechanism of ROS signaling 227
3. Targets of Redox Regulation 228
3.1 Protein Tyrosine Phosphatases 228
3.2 Receptor Tyrosine Kinases 228
3.2.1 Platelet-derived growth factor receptor 228
3.2.2 Epidermal growth factor receptor 230
3.2.3 Vascular endothelial growth factor receptor 230
3.2.4 Insulin receptor kinase 230
3.2.5 Fibroblast growth factor receptor 231
3.3 Nonreceptor Kinases 231
3.3.1 Akt 231
3.3.2 cAMP-dependent protein kinase 232
3.3.3 Src family kinases 232
3.3.4 MAPK 233
3.3.5 ATM protein kinase 234
3.3.6 Inhibitory kB kinase (IkB) 234
3.3.7 Ca2þ/calmodulin-dependent protein kinase II (CaMKII) 234
3.3.8 cGMP-dependent protein kinase 235
4. Others 235
4.1 Forkhead BoxO Transcription Factors 235
4.2 Nuclear Factor-Like 2 (Nrf2) 235

j
ISSN 1937-6448
222 Eileen G. Russell and Thomas G. Cotter
5. Pathophysiological Significance 236

5.1 Atherosclerosis 236
5.2 Inflammation 237
5.3 Neuroinflammation and Neurodegenerative Diseases 238
5.4 Type 2 Diabetes 238
5.5 Hypertension 239
5.6 Preeclampsia 240
5.7 Obesity 240
5.8 Aging 241
5.9 Cancer 241
6. Treatment Strategies Involving ROS Modulation 242
7. Measurement of ROS 245
8. Conclusions and Future Perspectives 246
References 246
Abstract
Reactive oxygen species (ROS) were once considered to be deleterious agents,
contributing to a vast range of pathologies. But, now their protective effects are being
appreciated. Both their damaging and beneficial effects are initiated when they target
distinct molecules and consequently begin functioning as part of complex signal-trans-
duction pathways. The recognition of ROS as signaling mediators has driven a wealth of
research into their roles in both normal and pathophysiological states. The present
review assesses the relevant recent literature to outline the current perspectives on
redox-signaling mechanisms, physiological implications, and therapeutic strategies.
This study highlights that a more fundamental knowledge about many aspects of
redox signaling will allow better targeting of ROS, which would in turn improve
prophylactic and pharmacotherapy for redox-associated diseases.
1. INTRODUCTION
All highly reactive oxygen derivatives can be referred to as reactive ox-
ygen species (ROS). This group includes oxygen radicals such as superoxide
ðO2$ Þ, hydroxyl ðOH$ Þ, peroxyl ðRO2$ Þ, and alkoxyl ðRO$ Þ as well as non-
radicals that are either oxidizing agents or are easily converted into radicals,
such as hypochlorous acid (HOCl), singlet oxygen (1O2), and hydrogen
peroxide (H2O2) (Woolley et al., 2013). Classically ROS were considered
deleterious agents, contributing to a vast range of pathologies, however,
more recently their protective effects are being appreciated. In the past two de-
cades, much focus has been placed on the concept that oxidants can function as
part of signal-transduction pathways (Sundaresan et al., 1995) (Figure 1).
Various ROS have different physical and chemical properties that allow
ROS in Cellular Signaling Processes 223
Endogenous sources Exogenous sources
Mitochondria Ultraviolet light

NADPH oxidase ROS Ionizing radiation
Peroxisomes Pollutants
Cycooxygenase
Cytochrome c oxidase
Antioxidantt
Activities
Modulation of
Molecular
Molecular damage
Signaling Pathways
Figure 1 Cellular response to reactive oxygen species (ROS). ROS can be produced
endo- or exogenously. The level of ROS is regulated by antioxidants defense mecha-
nisms. In this way, the cell encases an antioxidant/pro-oxidant balance.
them to modify distinct target molecules. The reductioneoxidation (redox)--

dependent signaling system is highly conserved and based on the oxidation and
reduction of cysteine residues (Figure 2). Experimental evidence suggests that
ROS generation causes reversible posttranslational modification, not only of
cysteine residues, but also of selenocysteine, methionine, and histidine residues
(Woolley et al., 2013). Although the mechanisms underlying ROS signaling
are not fully understood, increasing numbers of ROS-related observations
S-S
Disulfide
+H2O2 +H2O2 +H2O2 SO3H

S- SOH SO2H
Thiolate anion Sulfenic Acid Sulfinic Acid Sulfonic acid
S-Glutathiolation
SSG
Figure 2 Oxidative modification of cysteine residues by H2O2. The initial reaction prod-
uct of a thiolate with H2O2 produces sulfenic acid. Depending on conditions, sulfenics
can then participate in a variety of reactions. Sulfenics can be further oxidized to sulfinic
and sulfonic acids, can form intermolecular or intramolecular disulfide bonds, or can be
glutathionylated. The sulfenic form is readily reversible, higher states of oxidation
generally, but not always, lead to irreversible modification.
coupled with advancements in methodologies for ROS measurement are

contributing to increasing knowledge of redox signaling and consequent
cascades.
2. BACKGROUND
2.1 Source of ROS
ROS are generated in numerous cellular compartments and by mul-
tiple enzymes within the cell (Finkel, 2011). It is estimated that approxi-
mately 90% of ROS can be traced back to the mitochondrion. Aerobic
respiration results in the generation of ROS as a by-product of the electron
transport chain. Mitochondria generate ATP utilizing oxygen. During this
process, the flow of electrons down the respiratory chain gathers at complex
IV. In this scenario, O2$ is formed when a single electron prematurely
reduces O2. The major sites of superoxide production are thought to be
in complex I and complex III of the electron transport chain
(Figure 3(A)). The result is release of ROS from mitochondria resulting in
an intercellular environment of oxidative stress. Mitochondrial ROS have
been classically considered as toxic, but their importance in intracellular
signaling pathways is now acknowledged. It has become apparent that
many cellular organelles and enzymes produce ROS not only as a by-
product, like the mitochondria but also as a primary function. In the last
decade, much research has focused on the phagocyte NAPDH oxidases
(Nox) generating ROS as its preeminent role (Figure 3(B)). These enzymes
produce large amounts of ROS to execute their role in host defense. The
Nox family consists of 7 isoforms, Nox1e5 and the Dual oxidase (Duox)
1 and 2. The major source of ROS generation is a flavin- and heme-
containing protein complex that removes electrons from cytosolic NADPH
that are used to reduce molecular O2 and intentionally producing superox-
ide. Extracellular superoxide can reenter the cell or become converted to
hydrogen peroxide. Nox2 is the catalytic subunit of this complex, however,
it does not stimulate superoxide on its own. To do so, it recruits a number of
cytosolic factors including, p40phox, p47phox, p67phox, and Rac1 GTPase
as well as membrane-bound p22phox (Woolley et al., 2013). The physio-
logical functions of Nox-dependent ROS generation are ongoing.
While mitochondria and Noxs are the best-characterized sources of
ROS, a host of other enzymes can produce these reactive molecules. Perox-
isomes, cytochrome P-450 enzymes, lipoxygenases, cyclooxygenases,
Figure 3 Intracellular sources of reactive oxygen species (ROS). (A) Generation of mito-
chondrial ROS. This process mainly takes place at the electron transport chain located
on the inner mitochondrial membrane during the process of oxidative phosphorylation
(OXPHOS). Superoxide is formed due to partial reduction of oxygen caused by leakage
of electrons at complex I and complex III from electron transport chains. This superox-
ide is quickly dismutated to membrane permeable hydrogen peroxide by two dismu-
tases including superoxide dismutase 2 (SOD2) in mitochondrial matrix and SOD1 in
mitochondrial intermembrane space. (B) Generation of NADPH ROS. The catalytic sub-
unit is a transmembrane protein containing electron-transferring FAD and heme
groups capable of removing an electron from cytosolic NADPH. Superoxide is gener-
ated when this electron is used to reduce molecular O2. Extracellular O2$ can pass
through the plasma membranes via anion channels so it can reenter the cell or become
converted to hydrogen peroxide. Nox1e4 are localized at the membrane associated to
the p22phox protein. The activity of Nox1e3 is regulated by several factors including
GTPase Rac p47phox, p67phox, and p40phox. Nox4 is not regulated by any of the
known regulatory subunits while Nox5 is known to be activated by calcium.
xanthine oxidase, and nitric oxide synthase are all capable of producing
ROS. Just like the mitochondria and Noxs, these enzymes can be grouped
by those who generate ROS as a side effect and those who produce it as a
primary function. ROS are generated as a by-product of biological reactions
involving peroxisomes (Schrader and Fahimi, 2004) and cytochrome P-450
(Gottlieb, 2003) while the lipoxygenase and cyclooxygenase families of en-
zymes produce ROS required for fatty acid metabolism and the biosynthesis
of hormones and inflammatory mediators.
2.2 Antioxidant Systems

The balance between oxidation and antioxidation is critical in maintaining
healthy biological systems. The human antioxidative defense system
including superoxide dismutase (SOD), catalase (CAT), Peroxiredoxins
(Prx), glutathione peroxidase (GPx), glutathione (GSH) facilitates the elim-
ination of excess ROS. In addition to endogenous antioxidant defense sys-
tems, exogenous originating reducing compounds such as vitamin C,
vitamin E, carotenoids, and polyphenols are also required (Bouayed and
Bohn, 2010). SOD were the first ROS-metabolizing enzymes discovered.
O2$ can be metabolized to hydrogen peroxide by two metal-containing
SOD isoenzymes, Mn-SOD, present in mitochondria, and the cytosolic
dimeric Cu/Zn-SOD. Catalase function by catalyzing the dismutation of
hydrogen peroxide to water and molecular oxygen. Catalase also has func-
tions in detoxifying different substrates as well as having an antioxidative
role. Similar to catalase, Prx are capable of directly reducing peroxides while
GPx can catalyze the reduction of H2O2 and also reduce other peroxides to
alcohol. GSH functions mainly as a sulfhydryl buffer but also serves to
detoxify compounds (Nordberg and Arner, 2001). Vitamin E suppresses
the generation of lipid peroxidation and together with vitamin C it inhibits
hydroperoxide formation. It is also capable of scavenging free radicals, as is
vitamin A (Maritim et al., 2003). Vitamin E and carotenoids quench singlet
oxygen and as such contribute to the first defense line against oxidative stress
(Krinsky, 2001). Polyphenolic compounds, such as flavonoids are recog-
nized as potent antioxidants due to their ability to scavenge free radicals
by single-electron transfer (Hirano et al., 2001).
2.3 ROS Signaling

2.3.1 ROS as signaling molecules
Continuous creation and removal of ROS in a system is not only capable
of causing damage, but also of conveying important information.
Redox-signaling is achieved when the oxidants produced modulate a

signal-transduction pathway. These ROS must then trigger the oxida-
tion-induced modification of specific target molecules. Bartosz summa-
rized the requirements for a signaling molecule as follows: (1)
regulating its concentration with regard to synthesis and removal, (2) ex-
istence of specific receptors, and (3) reversibility of the signaling effect
(Bartosz, 2009). Some ROS are superior signaling molecules to others.
H2O2 is an excellent signaling molecule, as it is highly ubiquitous, can
diffuse easily and is relatively stable allowing it time to encounter specific
targets. It is perhaps for these reasons that H2O2 is the prevailing intracel-
lular redox-signaling molecule. In comparison, O2 $ becomes rapidly dis-
muted, lacks diffusibility, and its targets are limited to those within the
immediate vicinity of its source. Hydroxyl is an inadequate signaling mole-
cule due to its high oxidation rate constant, which leads to highly nonspe-
cific oxidation. Peroxyl and alkoxyl possess aggressive reactivity and lack
enzymatic removal resulting in the outcome of their reactions being mainly
irreversible oxidation events, ultimately leading to damage. Singlet oxygen
rarely occurs intracellularly and so it is doubtful that it contributes to signal
transduction. In contrast, HOCl has been suggested to function as a
signaling mediator in immune cells (Corcoran and Cotter, 2013).
2.3.2 Mechanism of ROS signaling

The precise mechanisms of ROS signaling are poorly understood. However,
the concept that Cys residues can function as redox-dependent switches is
thought to be the mechanism that underlies most ROS-dependent signaling
(Woolley et al., 2013). The main targets of H2O2 are a thiol group of protein
cysteine residues (Figure 2). Hydrogen peroxide oxidizes cysteines to
disulfides. The disulfide bonds can be reduced by glutaredoxins (Grxs),
thioredoxins (Trxs), or GSH. Therefore, this modification is reversible.
Formation of a mixed proteineglutathione disulfide, known as S-glutathio-
nylation, is thought to prevent further, irreversible oxidation of cysteine
sulfur. Glutathionylation and deglutathionylation are catalyzed by glutare-
doxins. Additional oxidation of a thiol group leads to sulfinic acids
SO2H or sulfonic acids SO3H. Thiol groups are reactive primarily in
their deprotonated form. The pKa of most cysteine thiols in proteins is
approximately 8.5 so these groups are mainly protonated. It is for this reason
that Cys residues of low pKa, being ionized at physiological pH, may be
selectively oxidized, even in the presence of excess of other Cys residues
(Bartosz, 2009).
3. TARGETS OF REDOX REGULATION

Proteins are the main targets for ROS in the cells. Oxidation-induced
alterations in the charge, size, hydrophobicity, and polarity of the amino
acids in a polypeptide can affect its secondary and tertiary structure, in
turn impacting on the stability and activity of the entire protein. ROS regu-
late distinct signaling pathways through interaction with key targets,
affecting a variety of mechanisms.
3.1 Protein Tyrosine Phosphatases

Oxidation has been implicated in the signaling processes involves in the
regulation of GTPases, transcription factors, proteases, and protein disulfide
isomerases. However, it is protein tyrosine phosphatases (PTPs), which are
the most widely studied redox protein targets. Oxidation of the cysteine
in the active site of PTP leads to a sulfenic acid-containing enzyme
(Figure 4). This enzyme is nonfunctional and allows phosphorylation by
phosphatase activity, reducing and regenerating the PTP. Many PTPs that
are sensitive to redox activity are also regulated by phosphorylation. These
two processes of enzyme regulation work in harmony with each other to
tightly control the phosphorylation signal (Corcoran and Cotter, 2013).
The nature of the oxidant species produced determines the type of signaling
effect, which ensues. For example, PTP1B is inhibited by H2O2 but not by
O2$ (Juarez et al., 2008).
3.2 Receptor Tyrosine Kinases

It is now clear that as well as targeting phosphatases, ROS can also directly
affect kinase signaling. It appears that RTK stimulation may utilize redox-
based mechanisms in tandem with protein phosphorylation during signal
transduction. RTKs are enzymes consisting of an extracellular ligand-
binding domain, a transmembrane domain, and an intracellular domain con-
taining a conserved tyrosine core and additional regulatory sequences. Upon
activation, the RTK transfers the signal downstream through nonreceptor
kinases to mediate a number of biological processes.
3.2.1 Platelet-derived growth factor receptor

Platelet-derived growth factor receptor (PDGFR) initiates signal transduc-
tion by binding to its receptors, which has alpha and beta forms. Activation
of PDGFR engages signaling pathways such as Ras/MAPK and PI3K/Akt.
In 1995, Sundaresan et al., showed that endogenous ROS levels peaked
RTK
ligand
RTK
PI3K
Nox
P P
AKT
T PDK1
K1
ROS
P P Active Inhibited
GSK-3β
3β
β P70S6K
K
PTP PTP
SH SH S S
Downstream Signaling
Figure 4 Schematic illustration of a typical redox signaling pathway. ROS modulate
receptor tyrosine kinase (RTK) signaling by regulating the redox state of protein
tyrosine phosphatases (PTPs). When a peptide ligand such as PDGF binds to its receptor
RTK, the signal can involve activation of PI3K and other downstream target proteins of
this kinase-driven pathway, which results in activation of membrane NOX and genera-
tion of ROS. This ROS, acts on a redox-sensitive cysteine residue in the active site of PTPs
and transform the eSH group into the oxidized SeS group, thus reversibly inactivating
PTPs. Oxidation of cysteine residues leading to PTP inhibition results in a stronger
signaling flux through the kinase arm of the pathway.
upon PDGF stimulation and were inhibited by H2O2 scavengers. Research

efforts have focused on elucidating the mechanisms that produce and regu-
late H2O2 during cell signaling. Results suggest that PIP3, the product of
activated PI3K, is essential for redox-based PDGF signaling. Furthermore,
expression of a dominant-negative mutant of the small GTPase Rac1
(Rac1N17) blocks increases in H2O2 (Sundaresan et al., 1996). Nox com-
plexes are the main oxidant source during PDGF signaling (Chiarugi,
2001; Lassegue et al., 2001).
PDGFR activation can also occur through a mechanism known as trans-
activation (Heenaman et al., 2000). In this system, angiotensin II binds to
GPCRs to trigger H2O2 production and mobilization of Ca2þ from intra-
cellular stores. Consequently, H2O2 activates kinases such as cytoplasmic
Src (c-Src) and protein kinase C (PKC) to stimulate RTK phosphorylation.
The mechanisms discussed here thus far are “indirect.” There is currently
very little data to support direct modification of PDGFR through cysteine
oxidation (Truong and Carroll, 2013).
3.2.2 Epidermal growth factor receptor

Upon stimulation epidermal growth factor receptor (EGFR) forms a
homo- or heterodimer followed by autophosphorylation of key residues
(Schlessinger, 2002). Following this, EGFR communicates the signal
through either Ras/MAPK or PI3K/Akt. EGFR employs H2O2 as a sec-
ondary messenger during cell signaling (Truong and Carroll, 2012). EGFR
is the only other RTK besides PDGFR known to undergo transactivation
by Ang II, and proceeds through H2O2-dependent c-Src activation (Chen
et al., 2001). Redox-based modulation of EGFR activity is similar to that
of PDGFR as it also appears to be controlled by indirect mechanisms.
Phosphorylation and sulfenylation work together to regulate receptor
kinase activity during EGF signaling (Troung and Carroll, 2013).
3.2.3 Vascular endothelial growth factor receptor

VEGF is an angiogenic growth factor. Vascular endothelial growth factor
receptor (VEGFR) activation initiates downstream signaling through Ras/
MAPK, PI3K/Akt, PLC-GAMMA, and the Src family of kinases to pro-
mote growth and regulate presurvival effects. There are three main
subtypes of VEGFR numbered 1, 2, and 3. From a redox-signaling
perspective, most research has focused on VEGFR-2. Colavitti et al.
(2002) concluded from previous studies in the area that upregulation of
VEGR and VEGFR-2 protects endothelial cells (ECs) against oxidative
injury. Furthermore, VEGF-mediated signaling is accompanied by bursts
of intracellular ROS. Similar to PDGFR and EGFR, VEGF-induced
H2O2 production increases receptor autophosphorylation and activation
of downstream targets. However, VEGFR differs from these other
RTKs in that it can undergo modulation through cysteine oxidation. In
this scenario, Nox2-derived H2O2 selectively exerts its effects on the
PI3K/Akt pathway by inducing c-Src activation (Abid et al., 2007).
3.2.4 Insulin receptor kinase

Insulin receptor kinase (IRK) is a heterotetrameric receptor composed of
two extracellular a-subunits and two transmembrane b-subunits. Insulin is
a hormone responsible for glucose and lipid metabolism. Binding of this
hormone to the IRK extracellular domain induces a conformational change
that facilitates ATP binding and leads to increased receptor autophosphory-

lation. Following this, IRK elicits Tyr phosphorylation of various cytosolic
docking proteins including SHC and members of the insulin receptor sub-
strate (IRS) protein family (Kido et al., 2001). Insulin’s effects are mainly
associated with the Ras/MAPK pathway through SHC (Avruch, 1998).
One of the main differences between IRK and other redox-mediated
RTKs is that IRK utilizes IRS docking proteins to initiate signal transduc-
tion. Activated IRS proteins recruit effectors to relay downstream signaling
through PI3K/Akt pathway and GLUT4. Unlike PDGF, insulin-induced
H2O2 production is independent of PI3K activation (Bae et al., 2000).
H2O2 acts as a secondary messenger to mediate the effects of insulin.
H2O2 can enhance or curtail insulin responsiveness, depending on the con-
centration and duration of exposure (Houstis et al., 2006).
3.2.5 Fibroblast growth factor receptor

The FGF family is composed of 18 distinct ligands and five fibroblast growth
factor receptors (FGFRs) have been identified. FGF exerts its actions
through four of these; FGFR1, FGFR2, FGFR3, and FGFR4. When stim-
ulated, FGFR undergoes receptor dimerization and TRY autophosphoryla-
tion within its intracellular kinase domain. This activation leads to triggering
of the Ras/MAPK, PI3K/Akt, STAT, and PLC-g pathways. FGF stimula-
tion may generate H2O2 through mechanisms dependent and independent
of NOX complexes (Nose et al., 1991; Thannickal et al., 2000). There is
also evidence identifying direct oxidation of cysteine residues in FGFR
(Kemble and Sun, 2009).
3.3 Nonreceptor Kinases

3.3.1 Akt
Akt is a serine/threonine-specific protein kinase of which there are three
isoforms; Akt1, Akt2, and Akt3. Endogenous H2O2 has been shown to in-
crease Akt activation in numerous cell types (Ushio-Fukai et al., 1999). This
activation is dependent on upstream production of lipid products by PI3K
(Kwon et al., 2004). H2O2 activation of Akt is also mediated by Src kinase
(Esposito et al., 2003). Furthermore, EGFR-dependent activation of Akt
enhances cell survival during H2O2-induced apoptosis (Wang et al.,
2000). More recent studies suggest that the effects of oxidation on Akt are
isoform specific. Wani et al. identified, in response to PDGF stimulation,
the direct oxidation of Cys124 in the linker domain and of Cys297 and
Cys311 in the kinase domain activation loop of the Akt2 isoform. This
oxidative modification correlated with a decrease in Akt2 kinase activity

while in contrast Akt1 and Akt3 remained active. Unlike Akt2, Akt1, and
Akt3 do not have cysteine analogs corresponding to Cys124. This is a valu-
able example of the degree of target specificity that can be achieved through
redox-signaling.
3.3.2 cAMP-dependent protein kinase

cAMP-dependent protein kinase (PKA) is a ubiquitously expressed signaling
molecule that is essential for numerous metabolic processes and has been
implicated in a wide range of cellular activities (Shabb, 2001). It is composed
of two types of subunits, the catalytic subunit and the regulatory subunit.
The catalytic subunit performs the phosphate-adding reaction while the
regulatory subunit senses the level of cyclic AMP and turns the catalytic sub-
units on or off based on that level. When cAMP levels are low, a dimer of
the regulatory subunits binds to two copies of the catalytic subunit, forming
an inactive complex. When cAMP levels rise, it binds to the regulatory sub-
unit, releasing the catalytic subunit in an active form (Goodsell, 2015). As
with many of the other kinases discussed in this review, PKA is regulated
by both phosphorylation and oxidation. Upon oxidation of Cys199, the
free C-subunit can be inactivated. This cysteine is capable of forming a
mixed disulfide with glutathione or an internal disulfide with Cys in the
C terminus, essentially inactivating the kinase in a redox-sensitive manner
(Humphries et al., 2005).
3.3.3 Src family kinases

The Src kinase family are nonreceptor tyrosine kinases that consist of 11
members; Blk, Brk, Fgr, Frk, Fyn, Hck, Lck, Lyn, c-Src, Srm, and Yes
(Roskoski, 2004, 2005). The domain organization of Src kinases contains
a myristoyl group attached to an SH4 domain, a unique domain, an SH3
domain, an SH2 domain, a linker between the SH2 and the kinase domain,
a tyrosine kinase domain, and a C-terminal regulatory segment (Brown and
Cooper, 1996; Boggon and Eck, 2004). Src kinases are regulated by phos-
phorylation and dephosphorylation but can also be activated by H2O2
and peroxynitrite (Mallozzi et al., 1999) as well as several stimuli leading
to an increase in ROS production (Giannoni et al., 2005).
H2O2-induced Src activation is generally accompanied by increased
Tyr416 phosphorylation (Li et al., 2008), while inactivation is accompanied
by increased Tyr527 phosphorylation (Tang et al., 2005; Troung and
Carroll, 2013). In addition to this, oxidation of the residues Cys 245 and
487, respectively, localized within SH2 and kinase domain, leads to a

conformational change which facilitates the activation of Src kinases
(Giannoni et al., 2005). A disulfide bridge between the two cysteine residues
allows clustering of different c-Src molecules and protects the activating res-
idue against dephosphorylation (Hebert-Chatelain, 2013).
3.3.4 MAPK
The MAPKs comprise a family of ubiquitous proline-directed, protein-
serine/threonine kinases that play an essential role in relaying extracellular
signals from the cell membrane to the nucleus (Boutros et al., 2008). In
mammalian cells, there are three well-defined subgroups of MAPKs:
Erks, JNKs, and p38 MAPKs. These three subgroups are involved in
both cell growth and cell death (Winter-vann and Johnson, 2007). Each
subgroup of MAPKs is activated through a three-tiered cascade that begins
the activation of MAPK kinase kinases (MAP3Ks). The MAP3Ks phos-
phorylate and activate a downstream MAPK kinases (MAP2Ks), which
in turn stimulate MAPK (Boutros et al., 2008). Although the process(es)
by which ROS can activate the MAPK pathways is not well defined, there
appears to be three main mechanisms (Son et al., 2011). The first of these is
direct redox regulation of the MAPKs themselves (Greene et al., 2000).
Data generated by Galli et al. suggest that efficient Erk2 binding to
MEK is achieved by oxidation of two of the five Erk2 cysteine thiols to
sulfinic and sulfonic acid, which occur at low but not high H2O2 concen-
trations (Galli et al., 2008). The second hypothesis is that ROS activate
MAPK pathways through the oxidative modification of intracellular ki-
nases. ASK-1 is a member of the MAP3K superfamily for JNK and p38
that binds to reduced thioredoxin in nonstressed cells. Upon oxidative
stress, thioredoxin becomes oxidized and disassociates from ASK-1, leading
to activation of JNK and p38 pathways through oligomerization of ASK-1
(Kamata et al., 2005). The final idea that is currently being investigated is
the inactivation and degradation of the MKPs that maintain the pathway in
an inactive state. Kamata et al. demonstrated that intracellular H2O2 accu-
mulation inactivates MKPs by oxidation of their catalytic cysteine, which
leads to sustained activation of JNK pathway. Lornejad-Sch€afer et al. inves-
tigated the regulation of MKP-1 expression and JNK activation by light
damage that has shown to enhance ROS production in retinal pigment
epithelial, ARPE-19 cells. In this study, doses of light lower than 2 J/
cm2 upregulated MKP-1 expression which was accompanied by inactiva-
tion of JNK pathway. However, doses that exceeded 3 J/cm2 led to a
decrease in the expression of MKP-1, resulting in continuous activation of

JNK pathway (Lornejad-Sch€afer et al., 2009). This study again highlights
the dual role of ROS in cellular signaling.
3.3.5 ATM protein kinase

ATM is a member of the phosphatidylinositol 3 kinase-like kinase (PIKK)
family of Ser/Thr-protein kinases that is best known for its role in the
DNA damage response. Recent findings suggest that it also functions as a
redox sensor that controls the levels of ROS (Ciccia and Elledge, 2010).
Upon activation by DSBs, ATM undergoes monomerization and requires
free DNA ends and MRN. In contrast, oxidized ATM is an active dimer
in which the two monomers are covalently linked by intermolecular disul-
fide bonds. In vitro experiments showed direct ATM activation in the pres-
ence of H2O2 independently from both DNA and MRN (Guo et al., 2010).
3.3.6 Inhibitory kB kinase (IkB)

IKK is a complex serine/threonine kinase composed of two catalytic
subunits, IKK-a and IKK-b, as well as a regulatory subunit, IKK-g.
Upon activation, IKK phosphorylates IkB-a, which triggers ubiquitination
and degradation of IkB-a. The degradation of IkB-a releases NF-kB, facil-
itating NF-kB to translocate to the nucleus and to promote transcription of
survival genes (Ghosh and Karin, 2002). IKK becomes inactivated by stim-
ulation with NO or H2O2 through Cys179 (Reynaert et al., 2004, 2006).
TNFa signaling also activates NF-kB. In this scenario, NF-kB can be
directly activated by ROS (Pantano et al., 2006).
3.3.7 Ca2þ/calmodulin-dependent protein kinase II (CaMKII)

CaMKII is a serine/threonine kinase that is a Ca2þ-activated enzyme that is
highly abundant in the brain (Lisman et al., 2002). Unlike the previously dis-
cussed kinases, CaMKII appears to become oxidized on a methionine resi-
due rather than a cysteine. In 2008, Erickson et al. showed that oxidation of
paired regulatory domain methionine residues sustains CaMKII activity in
the absence of Ca2þ/CaM. CaMKII is activated by angiotensin II (Ang
II)-induced oxidation, which leads to apoptosis in cardiomyocytes. Further-
more, this study showed that CaMKII oxidation is reversed by methionine
sulfoxide reductase A. He et al. explored this pathway further and found that
aldosterone can also induce oxidation of CaMKII by recruiting Nox (He
et al., 2011). It is also possible to inactivate CaMKII phosphatase through
cysteine oxidation (Baba et al., 2012).
3.3.8 cGMP-dependent protein kinase

cGMP-dependent protein kinase (PKG) is a serine/threonine-specific
protein kinase that is activated by cGMP and phosphorylates a number of
biologically important targets. Three types of PKG exist in mammalian cells.
PKGIa and Ib are splice variants of the type I gene and are found mostly in
the cytosol, whereas PKGII is transcribed from a separate gene and is asso-
ciated with the membrane at the N terminus. All PKGs share the same
domain organization as follows: an N-terminal regulatory domain, consist-
ing of a leucine zipper domain; an autoinhibitory sequence; two tandem
cyclic nucleotide-binding domains; and a C-terminal catalytic domain
(Hoffman et al., 2009). Burgoyne et al. reported that cGMPedependent
PKG, specifically the Ia isoform, is redox-sensitive and that oxidation
directly activates the kinase. Oxidative stress causes interprotein disulfide
bond formation between two cysteine 42 residues on adjacent chains in
the PKGIa homodimer complex, rendering the kinase catalytically active,
independently of cGMP (Burgoyne et al., 2007).
4. OTHERS
4.1 Forkhead BoxO Transcription Factors
Forkhead boxO (FOXO) transcription factors are activated by various
cellular stresses. The family which includes FOXO1, FOXO3, and FOXO4
are critical mediators of oxidative stress. Oxidative stress regulates FOXO
activity through various posttranslational modifications including phosphor-
ylation, acetylation, and ubiquitination. Their activity is regulated by
hydrogen peroxide and is usually associated with inducing apoptosis (Storz,
2011). An increase in intracellular ROS prompts the localization of FOXO
to the nucleus where it is transcriptionally active. FOXO proteins play an
important role in protecting cells against oxidative stress. For example, cells
activate FOXO transcription factors to reduce the level of oxidative stress by
inducing antioxidant enzymes. FOXO proteins are becoming of increasing
interest to researchers due to their roles in cardiovascular cell development
(Milkiewitcz et al., 2011), and carbohydrate and fatty acid metabolism
(Nunn et al., 2010).
4.2 Nuclear Factor-Like 2 (Nrf2)

Nrf 2 is referred to as the “master regulator” of the antioxidant response
(Rochette et al., 2014). It modulates the expression of hundreds of genes
involved in antioxidation, immune and inflammatory responses (Hybertson

et al., 2011). Under nonstressful conditions, concentrations of Nrf2 are low,
and the protein is retained in the cytoplasm due to its association with
Keap1. In response to oxidative stress, Nrf2 is stabilized and translocated
to the nucleus. Here, it upregulates the expression of several genes that
are implicated in the antioxidant response. These genes have a promoter re-
gion, which contains an antioxidant response element (ARE) (Magesh et al.,
2012). Oxidative stress is known to activate the Keap1eNrf2eARE
pathway as upon stimulation by oxidative stress the system expresses a broad
range of cytoprotective enzymes (Negi et al., 2011).
5. PATHOPHYSIOLOGICAL SIGNIFICANCE
5.1 Atherosclerosis
Atherosclerosis is the hardening and narrowing of the arteries due to
invasion and accumulation of white blood cells. As is the case in numerous
pathological states of the cardiovascular system, Nox are the principal source
of the superoxide anion. Nox1 controls proliferation and motility of smooth
muscle cells (SMC) and plays a role in blood pressure regulation. Increased
levels of Nox1 in SMC result in an elevated level of superoxide anion,
uncoupling of endothelian nitric oxide synthases (eNOS), and production
of xanthine oxidase (McNally et al., 2003). Nox2 generates superoxide an-
ions in professional phagocytes among other cells. Nox1/2 encourages the
development of endothelial dysfunction, hypertension, and inflammation
(Konior et al., 2014). Nox4 generates H2O2 (Martyn et al., 2006) and is
found in ECs, SMC, osteoclasts, hemopoietic stem cells, adipocytes, and car-
diomyocytes (Kuroda et al., 2005). Noticeable changes in the expression of
Noxs in atherosclerotic vessels can be observed. Regarding lesion progres-
sion, Nox1 protein expression was increased early and then decreased, while
activation of Nox4 was a late event. Contrastingly, Nox2 was elevated
throughout lesion progression (Xu et al., 2014). Noxs have a dual role in
the development of atherosclerosis as Nox4 is thought to protect the vascu-
lature while Nox5 has also been implicated in oxidative damage in athero-
sclerosis (Konior et al., 2014).
Although Nox appears to be the main source of ROS in atherosclerosis,
ROS generated by mitochondrial respiration and other enzymatic sources
also play a part. Like Nox-derived ROS, CYP also have contrasting effects
on the development of atherosclerosis. They can induce proinflammatory
effects but also cause vascular SMC to relax (Thum and Borlak, 2004).
While LOX and COX cannot directly act as a source of pathogenic
ROS, they generate arachidonic acid metabolites which can produce
ROS by stimulating NOX (Cho et al., 2011). In ECs and lesional macro-
phages, mitochondrial ROS contribute to induction of inflammatory reac-
tions via NFkB (Pamukcu et al., 2011). In hypoxic conditions, red blood
cells can independently generate ROS due to oxidation of iron. It is clear
that ROS generation is involved in atherosclerosis. As highlighted by Gon-
charov et al. in their recent review, a better understanding of feedback regu-
lation of ROS in both normal physiological state and atherosclerosis could
lead to more effective prophylactic strategies (Goncharov et al., 2015).
5.2 Inflammation
ROS are key signaling molecules in the progression of inflammation.
Elevated ROS generation by polymorphonuclear neutrophils (PMNs) at
the site of inflammation causes endothelial dysfunction and tissue damage.
Under inflammatory conditions, these PMNs are stimulated by oxidative
stress. The vascular endothelium is involved in the passage of macromole-
cules and inflammatory cells from the blood to tissue. This promotes the
migration of inflammatory cells across the endothelial barrier. These
migrated inflammatory cells have conflicting roles as they aid in the clear-
ance of pathogens but can also lead to tissue injury (Mittal et al., 2014).
It is not just PMNs that are regulated by oxidative stress. Tight junctions,
adherens junctions, and the actin cytoskeleton are all regulated by oxidative
stress. Oxidative stress produced by leukocytes at the site of inflammation
plays a crucial role in initiating junctional disassembly. Here, several medi-
ators that are released from inflammatory cells, including ROS, disrupt junc-
tions. The result is gap formation between cells. ROS disassemble the
endothelial barrier by activating Ca2þ signaling and influencing different
cellular events, which trigger inflammation. ROS-mediated regulation of
intracellular free Ca2þ concentration is a major mechanism of increased
vascular permeability which is the hallmark of inflammation (Ludwig
et al., 2011). An increase in intracellular Ca2þ leads to activation of Ca2þ/
calmodulin-dependent myosin light chain kinase (MLCK), which leads to
reorganization of actin cytoskeleton (Dudek and Garcia, 2001). Further-
more, ROS are capable of activating PKC (Gopalakrishna and Anderson,
1989), toll-like receptors (Ryan et al., 2004), and members of the GTPase
family (van Wetering et al., 2002) to mediate inflammation.
Nox-derived (Pendyala et al., 2009) and mitochondrial-derived

(Zinovkin et al., 2014) ROS which have previously been discussed in this
review both play a role in inflammation-induced injury. Two other sources
of ROS which play are implicated in inflammatory conditions are NOS
(Hink et al., 2001; Mollnau et al., 2003) and Xanthine oxidoreductase
(Jankov et al., 2008). NOSs are a family of enzymes that catalyze the produc-
tion of nitric oxide (NO) from L-arginine. XO is an enzyme that catalyzes
the oxidation of hypoxanthine to xanthine and uric acid.
5.3 Neuroinflammation and Neurodegenerative Diseases

The brain inflammatory diseases, including Alzheimer’s disease (AD) and
Parkinson’s disease (PD), are characterized by “redox state” imbalance and
chronic inflammation. Oxidative stress by mitochondria and Nox (von
Bernhandi and Eugenin, 2012) has been shown to mediate the pathogenesis
of neurodegenerative diseases, including PD (Halliwell, 2006), AD (Shi and
Gibson, 2007), and cerebrovascular disorders such as stroke (Chrissobolis and
Faraci, 2008).
Proinflammatory factors stimulate ROS generation. The ROS then
target proteins, including matrix metalloproteinase-9 (MMP-9), cytosolic
phospholipase A (cPLA), cyclooxygenase-2 (COX-2), inducible nitric oxide
synthase (iNOS), and adhesion molecules, which results in inflammation.
Furthermore, ROS act as a signaling molecule to trigger inflammatory
responses in central nervous systems (CNS) through the activation of
MAPK, RTK, the PI3K/AKT cascade and redox-sensitive transcription
factors, NF-kB and AP-1 (Hsieh and Yang, 2013).
5.4 Type 2 Diabetes

Obesity increases the risk of type 2 diabetes. Beta-cell dysfunction, insulin
resistance (IR), and free fatty acids (FFA) are the main characteristics of
this disease. Beta-cell dysfunction is the result of either chronic exposure
to hyperglycemia or FFA. In IR syndrome, there is an increased FFA flux
from adipocytes into arterial ECs that may result in increased FFA oxidation
by the mitochondria. In cells exposed to sustained high glucose concentra-
tions, more glucose is being oxidized, which leads to an increase in ROS
generation. Under diabetic conditions, chronic hyperglycemia induces an
alteration in pancreas function and aggravates IR; these processes are associ-
ated with oxidative stress. Beta cells are sensitive to oxidative stress due to
their relatively low expression of antioxidant enzymes. In diabetes, hyper-
glycemia and the subsequent production of oxidative stress decrease insulin
gene expression and secretion. Activation of the JNK pathway is associated

with a reduction in insulin gene expression induced by oxidative stress
(Rochette et al., 2014).
Oxidative stress has also been implicated in diabetic complications,
particularly accelerated atherosclerosis and microvascular damage of the
retina, kidney, and nerves. The polyol pathway is involved in these adverse
effects. This pathway is facilitated during a hyperglycemic state. Aldose
reductase utilizes and depletes NADPH to convert excess glucose to sorbitol.
The depletion of NADPH results in reduced levels of GSH, which contrib-
utes to oxidative stress. Sorbitol is then converted to fructose, simultaneously
producing NADH, which leads to increased generation of ROS. Hypergly-
cemia facilitates the nonenzymatic glycation of proteins and lipids. This pro-
cess leads to the formation of advanced glycation end products (AGEs).
AGEs are capable of modifying lipoproteins, inducing abnormalities in the
extracellular matrix and generating ROS through interaction with its recep-
tor. Two targets of ROS, NF-kB and PKC, which were previously
described in this review, also come into play in this physiological state as
they further increase oxidative stress (Rochette et al., 2014).
A number of studies have strengthened the hypothesis that increased
oxidative stress leads to impaired insulin secretion. Kawamori et al. showed
that there was suppression of insulin gene expression in B-cells exposed to
oxidative stress. Furthermore, oxidative stress has been shown to suppress
the activity of pancreatic and duodenal homeobox factor-1 (PDX1) and
v-maf musculoaponeurotic fibrosarcoma oncogene homolog A (MafA),
which are 2 important transcription factors of the insulin gene (Kawamori
et al., 2003; Matsuoka et al., 2003).
5.5 Hypertension
The role of ROS in the regulation of blood pressure under normal physio-
logical condition remains unclear, however, the notion that ROS genera-
tion and signaling in the brain stem are involved in the pathogenesis of
hypertension is undisputed (Hirooka, 2008). NADPH oxidase is a major
source of ROS in hypertension and plays a pivotal role in generating
ROS in the brain (Zimmerman et al., 2004). In the vascular system, ROS
production from Nox is prompted by vasoconstrictor agents such as angio-
tensin II (Ang II), endothelin-1 (ET-1), and norepinephrine (NE). Activa-
tion of angiotensin type 1 (AT1) receptors by Ang II triggers a number of
ROS-producing events. The key cardiovascular effects of angiotensin II
involve superoxide generation. Nox-derived superoxide anion mediates
angiotensin II-induced pressor effect via activation of p38 mitogen-activated

protein kinase in the rostral ventrolateral medulla (Chan et al., 2005). The
involvement of PI3-kinase-protein kinase B-dependent signaling pathway
in this response is now appreciated (Yang and Raidaza, 1999). The full
role of ROS in hypertension remains unknown and complicated.
5.6 Preeclampsia
Preeclampsia (PE) is a pregnancy-induced hypertensive disorder that affects
7e10% of pregnancies. This disorder is highly associated with perinatal
morbidity and mortality (Chappel et al., 2008). The etiology of PE is not
well defined but vascular dysfunction resulting in poor placentation is
thought to be the main cause (Roberts et al., 1989). Neutrophils in periph-
eral circulation of patients with PE are activated (Abe et al., 2003). Activated
neutrophils produce ROS through Nox, XO, and uncoupled eNOS (Gielis
et al., 2011). ROS are produced by both the ischemic placenta and systemic
vasculature during this disorder. Scavenging of NO by ROS leads to the for-
mation of ONOO, which seems to be involved in hypertension (Tschudi
et al., 1996). High levels of ONOO oxidize and damage DNA, proteins,
and lipids while low levels interfere with vascular signaling. ONOO can
lead to irreversible nitration of tyrosine residues on other proteins causing
defective phosphorylation and enzymatic dysfunction (Matsubara et al.,
2015). More research is needed to establish the role of ROS in PE.
5.7 Obesity
Associations have been made between obesity and oxidant stress (Fenster
et al., 2002). One of the main observations is that the oxidizability of
non-HDL lipids in vitro is elevated in obese women (Van Gaal et al.,
1998). Furthermore, obesity has been associated with increased myocardial
oxidative stress (Vincent et al., 1999). Although the mechanisms that are
responsible for these associations are unclear, several hypotheses have been
proposed. One such hypothesis was put forward by Bakker et al. They
suggest that oxidant stress in obesity may result from the accumulation of
intracellular triglycerides. Intracellular triglycerides are expected to elevate
O2$ production within the electron transport chain by inhibiting the mito-
chondrial adenosine nucleotide transporter. This leads to a decrease in intra-
mitochondrial adenosine diphosphate, which results in a reduction in the
flux of protons through the adenosine triphosphateesynthase reaction.
Consequently, electrons build up within the electron transport chain that
can produce O2$ (Bakker et al., 2000). Adiposity may also contribute to
oxidant stress. Adipocytes are recognized as sources of inflammatory cyto-

kines. Lipopolysaccharides, intracellular triglycerides, and catecholamines
are examples of stimuli that are capable of inducing cytokine release from
adipocytes (Coppack, 2001). In turn, cytokines stimulate the production
of ROS by macrophages and monocytes (Podrez et al., 2000).
The evidence presented here is not conclusive. However, if the accumu-
lation of intracellular triglycerides and/or adiposity increases oxidant stress,
then reduction of total body fat should reduce oxidant stress. Several studies
have produced results consistent with this prediction (Dandona et al., 2001;
Lund et al., 2015).
5.8 Aging
The mitochondrial free radical theory of aging was proposed over 50 years
ago by Denham Harman (1956). This theory suggests mitochondrial-
derived ROS damage cellular macromolecules ultimately leading to the
dysfunction and failure that characterizes aging. While this theory was
largely accepted, with many studies supporting the theory, many studies,
particularly from the last decade, have produced contradictory data. The lat-
est twist in this saga arises from the notion of hormesis. This phenomenon
proposes that a slight stress protects the cell from a subsequent larger stress.
It is now recognized that mitochondrial ROS might not have a completely
harmful role in regulating life span (Holmstrom and Finkel, 2014). For
example, Ristow et al. (2009) showed that the beneficial effects of physical
exercise were eliminated in subjects who were given antioxidant supple-
ments. It is likely that these supplements inhibited an ROS-dependent her-
metic response. In conclusion, there is currently no consistent relationship
between mitochondrial ROS and longevity (Stuart et al., 2014).
5.9 Cancer
Cancer initiation and progression have been linked to oxidative stress by
increasing DNA mutations or inducing DNA damage, genome instability,
and cell proliferation (Visconti and Grieco, 2009). One of the key character-
istics of cancer cells compared to the normal cells is a persistent pro-oxidative
state that can lead to intrinsic oxidative stress (Toyokuni et al., 1995). The
Warburg effect refers to cancer cells having higher rates of glycolysis.
Glucose normally produces ATP and lactate but can be redirected to the
pentose phosphate pathway (PPP), yielding NADPH which maintains
GSH in its reduced state. Whether glucose produces ATP and lactate or
generates NAPDH is largely determined by pyruvate kinase. Increased levels
of ROS target a unique Cys residue in pyruvate kinase M2 (PKM2), an iso-

form of pyruvate kinase expressed in tumor cells. The oxidized PKM2 di-
verts glucose toward the PPP leading to redox buffering by GSH
(Anastasiou et al., 2011). The increased oxidative stress in cancer cells results
not only from their increased metabolism but also inactivation of antioxidant
mechanisms. Cancer cells have evolved mechanisms to protect themselves
from intrinsic oxidative stress such as rearranging antioxidant functions
and upregulating prosurvival molecules (Farber and Rubin, 1991).
ROS may also contribute to the initiation of cancer through accelerating
protumorigenic signaling pathways. ROS can oxidize disulfide bonds of
cysteine residues, changing the activity of certain proteins, most notably
the tyrosine phosphatases superfamily (Tonks, 2006). For example, the inhi-
bition of PTEN by ROS hyperactivates the PI3K/Akt signaling pathway,
which is potentially the most frequently activated signaling pathway in can-
cer cells (Hay, 2005).
In contrast to its antitumorigenic actions, oxidative stress has also been
linked to senescence and apoptosis, two major mechanisms that can act as
a barrier to tumor development (Visconti and Grieco, 2009). For example,
alterations in ROS levels can activate p53, resulting in the induction of
senescence (Vigneron and Vousden, 2010). This means that the high levels
of ROS that cancer cells exhibit may be their Achilles’ heel as it makes them
more susceptible to cell death. The precise connection between oxidative
stress and cancer has yet to be elucidated. As is the case for all of the diseases
discussed in this review, a better understanding of the ROS mechanisms
could undoubtedly improve prophylaxis and treatment.
6. TREATMENT STRATEGIES INVOLVING ROS

MODULATION
From the examples discussed in this review, it is clear that oxidative
stress is involved in the initiation and progression of many diseases and
disorders. Therefore, it is logical to think that inhibiting ROS would be a
successful way of treating ROS-related diseases.
Antioxidative agents include small ROS scavengers, inhibitors of ROS-
generating enzymes, as well as antioxidative enzymes. Studies using
vitamin C and vitamin E as antioxidants, have yielded inconsistent results.
More recent antioxidant strategies have included non-vitaminic antioxidants
such as alpha-lipoic acid (ALA) and coenzyme Q10 (CoQ10). ALA can
scavenge hydroxyl radicals, hypochlorous acid, and singlet oxygen while
CoQ10 can lead to an increase in mitochondrial respiration (Rochette et al.,

2014). However, antioxidant therapy has a number of limitations, which are
discussed in a review by Drummond et al. (2011). Here, they suggest among
other reasons that the reaction of the targeted ROS with the antioxidant
may produce another type of ROS that does not react with the original anti-
oxidant. Therefore, it now seems that prevention of ROS formation by tar-
geting the enzymes responsible for their generation could be a more
effective strategy for treating these diseases.
Nox are a well-established source of superoxide. Apocynin and dipheny-
leneiodonium are the most widely studied Nox inhibitors. Although these
compounds have proven to be invaluable in Nox research, their lack of
selectivity for Noxs over other enzymes and/or their potential to act as
pro-oxidants under certain conditions is likely to limit their use in a clinical
setting. Triazolopyrimidines, including VAS2870 and VAS3947, have
emerged as promising inhibitors of Nox activity. These compounds
inhibited NADPH oxidase-derived ROS in several cell lines expressing
Noxs and in primary endothelial and VSMC cultures and had no effect
on ROS generated by XO or on eNOS activity (Wind et al., 2010). GK-
136901 and ML171 also show promise with respect to inhibiting Nox1
and Nox4. However, none of these compounds are capable of inhibiting
Nox2. Therefore, there is a need to identify Nox2 inhibitors as well as tar-
geting the p47phox subunit, which acts as the “organizer” protein for the
Nox2 oxidase complex (Drummond et al., 2011).
XO is one of the major sources of O2 generation in vascular systems and
in most inflammatory lesions (Miyomoto et al., 1996). Therefore, XO
inhibitors are also of therapeutic interest. Allopurinol is a well-known
XO inhibitor, which is used to treat gout that is caused by high levels of
uric acid in the body. Unfortunately the therapeutic effect of allopurinol
is not dose-dependent. Fang et al. discovered a more potent XO inhibitor,
4-amino-6-hydroxypyrazolo[3,4-d]pyrimidine (AHPP) (Fang et al., 2009).
Studies by Miyomoto et al. highlight the therapeutic potential of this com-
pound. They showed that it has an antihypertensive effect in rats (Miyomoto
et al., 1996). Modulating ROS still shows promise with respect to treating
inflammatory diseases and warrants further investigation.
Another approach for modulating oxidative stress in humans is simply
using drugs that are already being applied to treat the disease at hand.
Many of these drugs have pleiotropic properties and so can act as indirect
antioxidants. The optimal approach to antioxidant therapy requires stimula-
tion of NO production and simultaneous inhibition of vascular superoxide
production. Drugs such as statins, angiotensin-converting enzyme inhibitors

and AT1-receptor blockers, possess such properties (Rochette et al., 2014).
In addition to inflammation, oncology is the other main area where
ROS therapeutics are making great strives. The elevated ROS levels of
cancer cells can be exploited for cancer therapy. The principals underlying
this type of therapy are triggering ROS accumulation and/or inhibiting
ROS-scavenging systems. Reagents capable of enhancing ROS generation
in this way include mitochondrial electron transport chain modulators e.g.,
doxorubicin, redox-cycling compounds e.g., motexafin gadolinium, and
compounds that disrupt the antioxidant defense mechanism e.g., beta-
phenylethyl isothiocyanates (PEITC) (Nogueira and Hay, 2013).
A major obstacle for anticancer strategy by ROS is targeting the ROS
production in cancer tissues selectively to avoid side effects. This problem
could be effectively solved by enhanced permeability and retention
(EPR)-effect using macromolecular drugs, polymeric micelles, or nanopar-
ticles (Fang et al., 2003). EPR effect is now becoming a gold standard for
anticancer drug design. Another challenge for cancer therapy is the problem
of chemoresistance. AKT is known to promote resistance to agents that
induce apoptosis. AKT lowers the threshold of oxidative stress needed to
induce cell death. Rapamycin inhibits Akt activity. Thus, rapamycin analogs
are currently in clinical trial with some already having been approved for
certain types of cancer. A combination therapy using rapamycin with
PEITC was efficient in selectively eradicating tumors with hyperactivated
Akt in preclinical studies. This is not the only example of ROS-modulating
agents being successful when used in combination with other chemothera-
peutics. Combination of exogenous ROS inducers and inhibitors of AMPK,
a master switch of metabolic activation could be another promising strategy
for cancer therapy (Nogueira and Hay, 2013).
Of course, not all methods of modulating ROS are pharmacological.
Exercise produces a short-term inflammatory response that increases oxida-
tive stress. This proinflammatory response is followed by a long-term anti-
inflammatory effect. It has been demonstrated that regular exercise increases
anti-inflammatory responses and it is believed that even short-term endur-
ance exercise training results in rapid increases in Mn-SOD and GPx activity
in different tissues (Yamashiti et al., 1999). Watson stated in his report in
“The Lancet” last year that exercise-stressed skeletal muscle generates
ROS and further postulates that diabetes, cardiovascular disease, and some
cancers are accelerated by failure of the endoplasmic reticulum to generate
sufficient redox potential for disulfide bonds to be formed (Watson, 2014).
7. MEASUREMENT OF ROS
It is clear that ROS play a vital role in many essential biological
processes. It is therefore crucial to have adequate tools to measure ROS
generation to further investigate the fundamental role of ROS. There are
many approaches to measure the generation and accumulation of different
ROS.
Fluorescent dyes for ROS are based on the oxidationereduction
processes between the ROS being targeted and the reduced probe. Fluores-
cence occurs upon oxidation of the probe. The two most commonly used
dyes are 20 ,70 -dichlorodihydrofluorescein (DCF) and dihydroethidium
(DHE) (Woolley et al., 2013). The generally accepted mechanism of
DCF acting as a probe is as follows: the nonfluorescent DCFH2-DA diffuses
and crosses the cell membrane, DCFH2-DA then deacetylates to form
DCFH2 which is now membrane impermeable. DCFH2 then reacts with
intracellular ROS to give DCF, which fluoresces (Brandt and Keston,
1965; Chen et al., 2010). While DCF detects hydrogen peroxide, DHE is
specific for superoxide. The reaction between superoxide and HE generates
a highly specific red fluorescent product, 2-hydroxyethidium.
The main problems associated with ROS fluorescent probes in cells are
lack of specificity, cytotoxicity, reversibility, and reaction rate as well as
diffusion in tissue. If a probe reacts irreversibly with an ROS, results can
be skewed by the redox state of the compartment that it is found in.
Regarding the reaction rate, dyes are in competition with the antioxidant
enzymes in the cell. Probes cannot compete with the catalytic efficiencies
of antioxidant systems, therefore, the redox state of the cellular compart-
ments again comes into play (Woolley et al., 2013).
Major advances have been made to overcome these limitations. For
example, to overcome the problems associated with the irreversibility of
fluorescent probes, genetically encoded reporters have been employed.
This technique involves genetically modifying cells to express a redox-
sensitive fluorescent protein. This method allows for the reversible detection
of ROS inside the cell (Ostergaard et al., 2001). Encapsulating the dye of
interest in a nanoparticle has proved successful in overcoming the nonspe-
cific interactions and cytotoxicity associated with traditional dyes. This
encapsulation protects the probes from nonspecific interactions while simul-
taneously protecting the cells from potential cytotoxic effects (Koo et al.,
2007). Poor probe diffusion in tissue and organ samples can hinder imaging.
To tackle this problem, a combination of near-infrared detection of the

cyanine-7 (Cy-7) with the chemoselectivity of phenylboronic acid was
developed. This tool allows deep penetration and low background fluores-
cence (Van de Bittner et al., 2010).
The methods discussed here provide valuable insights into redox
dynamics. However, as highlighted by Woolley et al., the information
obtained from experiments using fluorescent probes should be carefully
analyzed, and all variables should be considered (Woolley et al., 2013).
8. CONCLUSIONS AND FUTURE PERSPECTIVES

It is clear that ROS are now appreciated as important signaling mol-
ecules that are capable of inducing protective as well as deleterious effects.
The vast number of signal-transduction pathways that ROS are involved
in and the consequent physiological or pathophysiological impacts of these
processes have justified extensive research in this field. While progress has
been made in developing therapies by manipulating our knowledge of these
signaling molecules, it is evident that lack of our fundamental knowledge
about many aspects of ROS is hampering this process. There is an urgency
to establish the role of these reactive oxygen derivates in a normal physio-
logical state before truly effective therapies can take advantage of their prop-
erties in various diseases. Indeed, the development of better experimental
tools coupled with the diversity of the disciplines which ROS intrigue is
ensuring constant interest and research into revealing their full potential.
Based on the results of preclinical experiments thus far, it seems that
ROS-modulating therapies will be used in conjunction with existing
pharmaceutics rather than as a stand-alone therapy. In conclusion, while
the ultimate goal of ROS-orientated research is to produce successful treat-
ments for the diseases that ROS play a role in; a tremendous amount of
investigation into the fundamental role of ROS is warranted.
REFERENCES
Abe, E., Matsubara, K., Ochi, H., Ito, M., Oka, K., Kameda, K., 2003. Elevated levels of
adhesion molecules derived from leukocytes and endothelial cells in patients with preg-
nancy-induced hypertension. Hypertens. Pregnancy 22, 31e43.
Abid, M.R., Spokes, K.C., Shih, S.C., Aird, W.C., 2007. NADPH oxidase activity selec-
tively modulates vascular endothelial growth factor signaling pathways. J. Biol. Chem.
282, 35373e35385.
Anastasiou, D., Poulogiannis, G., Asara, J.M., Boxer, M.B., Jiang, J.K., Shen, M.,
Bellinger, G., Sasaki, A.T., Locasale, J.W., Auld, D.S., Thomas, C.J., Vander
Heiden, M.G., Cantley, L.C., 2011. Inhibition of pyruvate kinase M2 by reactive oxy-
gen species contributes to cellular antioxidant responses. Science 334, 1278e1283.
Avruch, J., 1998. Insulin signal transduction through protein kinase cascades. Mol. Cell Bio-
chem. 182, 31e48.
Baba, H., Sueyoshi, N., Shigeri, Y., Ishida, A., Kameshita, I., 2012. Regulation of Ca(2þ)/
calmodulin-dependent protein kinase phosphatase (CaMKP) by oxidation/reduction at
Cys-359. Arch. Biochem. Biophys. 526, 9e15.
Bae, Y.S., Sung, J.Y., Kim, O.S., Kim, Y.J., Hur, K.C., Kazlauskas, A., Rhee, S.G., 2000.
Platelet-derived growth factor-induced H(2)O(2) production requires the activation of
phosphatidylinositol 3-kinase. J. Biol. Chem. 275, 10527e10531.
Bakker, S.J., IJzerman, R.G., Teerlink, T., Westerhoff, H.V., Gans, R.O., Heine, R.J., 2000.
Cytosolic triglycerides and oxidative stress in central obesity: the missing link between
excessive atherosclerosis, endothelial dysfunction, and beta-cell failure? Atherosclerosis
148, 17e21.
Bartosz, G., 2009. Reactive oxygen species: destroyers or messengers? Biochem. Pharmacol.
77, 1303e1315.
von Bernhardi, R., Eugenín, J., 2012. Alzheimer’s disease: redox dysregulation as a common
denominator for diverse pathogenic mechanisms. Antioxid. Redox Signal 16, 974e1031.
Boggon, T.J., Eck, M.J., 2004. Structure and regulation of Src family kinases. Oncogene 23,
7918e7927.
Bouayed, J., Bohn, T., 2010. Exogenous antioxidants e double-edged swords in cellular
redox state: health beneficial effects at physiologic doses versus deleterious effects at
high doses. Oxid. Med. Cell Longev. 3, 228e237.
Boutros, T., Chevet, E., Metrakos, P., 2008. Mitogen-activated protein (MAP) kinase/MAP
kinase phosphatase regulation: roles in cell growth, death, and cancer. Pharmacol. Rev.
60, 261e310.
Brandt, R., Keston, A.S., 1965. Synthesis of diacetyldichlorofluorescin: a stable reagent for
fluorometric analysis. Anal. Biochem. 11, 6e9.
Brown, M.T., Cooper, J.A., 1996. Regulation, substrates and functions of src. Biochim. Bio-
phys. Acta 1287, 121e149.
Burgoyne, J.R., Madhani, M., Cuello, F., Charles, R.L., Brennan, J.P., Schr€ oder, E.,
Browning, D.D., Eaton, P., 2007. Cysteine redox sensor in PKGIa enables oxidant-
induced activation. Science 317, 1393e1397.
Chan, S.H., Hsu, K.S., Huang, C.C., Wang, L.L., Ou, C.C., Chan, J.Y., 2005. NADPH ox-
idase-derived superoxide anion mediates angiotensin II-induced pressor effect via activa-
tion of p38 mitogen-activated protein kinase in the rostral ventrolateral medulla. Circ.
Res. 97, 772e780.
Chappell, L.C., Enye, S., Seed, P., Briley, A.L., Poston, L., Shennan, A.H., 2008. Adverse
perinatal outcomes and risk factors for preeclampsia in women with chronic hyperten-
sion: a prospective study. Hypertension 51, 1002e1009.
Chen, K., Vita, J.A., Berk, B.C., Keaney, J.F., 2001. c-Jun N-terminal kinase activation by
hydrogen peroxide in endothelial cells involves SRC-dependent epidermal growth fac-
tor receptor transactivation. J. Biol. Chem. 276, 16045e16050.
Chen, X., Zhong, Z., Xu, Z., Chen, L., Wang, Y., 2010. 20 ,70 -Dichlorodihydrofluorescein
as a fluorescent probe for reactive oxygen species measurement: forty years of application
and controversy. Free Radic. Res. 44, 587e604.
Chiarugi, P., 2001. The redox regulation of LMW-PTP during cell proliferation or growth
inhibition. IUBMB Life 52, 55e59.
Cho, K.J., Seo, J.M., Kim, J.H., 2011. Bioactive lipoxygenase metabolites stimulation of
NADPH oxidases and reactive oxygen species. Mol. Cells 32, 1e5.
Chrissobolis, S., Faraci, F.M., 2008. The role of oxidative stress and NADPH oxidase in
cerebrovascular disease. Trends Mol. Med. 14, 495e502.
Ciccia, A., Elledge, S.J., 2010. The DNA damage response: making it safe to play with knives.
Mol. Cell 40, 179e204.
Colavitti, R., Pani, G., Bedogni, B., Anzevino, R., Borrello, S., Waltenberger, J.,
Galeotti, T., 2002. Reactive oxygen species as downstream mediators of angiogenic
signaling by vascular endothelial growth factor receptor-2/KDR. J. Biol. Chem. 277,
3101e3108.
Coppack, S.W., 2001. Pro-inflammatory cytokines and adipose tissue. Proc. Nutr. Soc. 60,
349e356.
Corcoran, A., Cotter, T.G., 2013. Redox regulation of protein kinases. FEBS J. 280,
1944e1965.
Dandona, P., Mohanty, P., Ghanim, H., Aljada, A., Browne, R., Hamouda, W.,
Prabhala, A., Afzal, A., Garg, R., 2001. The suppressive effect of dietary restriction
and weight loss in the obese on the generation of reactive oxygen species by leuko-
cytes, lipid peroxidation, and protein carbonylation. J. Clin. Endocrinol. Metab. 86,
355e362.
Drummond, G.R., Selemidis, S., Griendling, K.K., Sobey, C.G., 2011. Combating oxidative
stress in vascular disease: NADPH oxidases as therapeutic targets. Nat. Rev. Drug Discov.
10, 453e471.
Dudek, S.M., Garcia, J.G., 2001. Cytoskeletal regulation of pulmonary vascular permeability.
J. Appl. Physiol. (1985) 91, 1487e1500.
Esposito, F., Chirico, G., Montesano Gesualdi, N., Posadas, I., Ammendola, R., Russo, T.,
Cirino, G., Cimino, F., 2003. Protein kinase B activation by reactive oxygen species is
independent of tyrosine kinase receptor phosphorylation and requires SRC activity.
J. Biol. Chem. 278, 20828e20834.
Fang, J., Iyer, A.K., Seki, T., Nakamura, H., Greish, K., Maeda, H., 2009. SMA-copolymer
conjugate of AHPP: a polymeric inhibitor of xanthine oxidase with potential antihyper-
tensive effect. J. Control Release 135, 211e217.
Fang, J., Sawa, T., Maeda, H., 2003. Factors and mechanism of “EPR” effect and the
enhanced antitumor effects of macromolecular drugs including SMANCS. Adv. Exp.
Med. Biol. 519, 29e49.
Farber, E., Rubin, H., 1991. Cellular adaptation in the origin and development of cancer.
Cancer Res. 51, 2751e2761.
Fenster, C.P., Weinsier, R.L., Darley-Usmar, V.M., Patel, R.P., 2002. Obesity, aerobic
exercise, and vascular disease: the role of oxidant stress. Obes. Res. 10, 964e968.
Finkel, T., 2011. Signal transduction by reactive oxygen species. J. Cell Biol. 194, 7e15.
Galli, S., Antico Arciuch, V.G., Poderoso, C., Converso, D.P., Zhou, Q., Bal de Kier
Joffé, E., Cadenas, E., Boczkowski, J., Carreras, M.C., Poderoso, J.J., 2008. Tumor
cell phenotype is sustained by selective MAPK oxidation in mitochondria. PLoS One
3, e2379.
Ghosh, S., Karin, M., 2002. Missing pieces in the NF-kappaB puzzle. Cell 109 (Suppl.),
S81eS96.
Giannoni, E., Buricchi, F., Raugei, G., Ramponi, G., Chiarugi, P., 2005. Intracellular
reactive oxygen species activate Src tyrosine kinase during cell adhesion and
anchorage-dependent cell growth. Mol. Cell Biol. 25, 6391e6403.
Gielis, J.F., Lin, J.Y., Wingler, K., Van Schil, P.E., Schmidt, H.H., Moens, A.L., 2011. Path-
ogenetic role of eNOS uncoupling in cardiopulmonary disorders. Free Radic. Biol.
Med. 50, 765e776.
Goncharov, N.V., Avdonin, P.V., Nadeev, A.D., Zharkikh, I.L., Jenkins, R.O., 2015. Reac-
tive oxygen species in pathogenesis of atherosclerosis. Curr. Pharm. Des. 21, 1134e1146.
Goodsell, D., 2015. cAMP-dependent protein kinase (PKA). Mol. Mon. http://dx.doi.org/
10.2210/rcsb_pdb/mom_2012_8.
Gopalakrishna, R., Anderson, W.B., 1989. Ca2þ- and phospholipid-independent activation

of protein kinase C by selective oxidative modification of the regulatory domain. Proc.
Natl. Acad. Sci. U.S.A. 86, 6758e6762.
Gottlieb, R.A., 2003. Cytochrome P450: major player in reperfusion injury. Arch. Biochem.
Biophys. 420, 262e267.
Greene, E.L., Houghton, O., Collinsworth, G., Garnovskaya, M.N., Nagai, T., Sajjad, T.,
Bheemanathini, V., Grewal, J.S., Paul, R.V., Raymond, J.R., 2000. 5-HT(2A) receptors
stimulate mitogen-activated protein kinase via H(2)O(2) generation in rat renal mesan-
gial cells. Am. J. Physiol. Renal Physiol. 278, F650eF658.
Guo, Z., Kozlov, S., Lavin, M.F., Person, M.D., Paull, T.T., 2010. ATM activation by
oxidative stress. Science 330, 517e521.
Halliwell, B., 2006. Oxidative stress and neurodegeneration: where are we now? J. Neuro-
chem. 97, 1634e1658.
Harman, D., 1956. Aging: a theory based on free radical and radiation chemistry. J. Gerontol.
11, 298e300.
Hay, N., 2005. The Akt-mTOR tango and its relevance to cancer. Cancer Cell 8, 179e183.
He, B.J., Joiner, M.L., Singh, M.V., Luczak, E.D., Swaminathan, P.D., Koval, O.M.,
Kutschke, W., Allamargot, C., Yang, J., Guan, X., Zimmerman, K., Grumbach, I.M.,
Weiss, R.M., Spitz, D.R., Sigmund, C.D., Blankesteijn, W.M., Heymans, S.,
Mohler, P.J., Anderson, M.E., 2011. Oxidation of CaMKII determines the cardiotoxic
effects of aldosterone. Nat. Med. 17, 1610e1618.
Hebert-Chatelain, E., 2013. Src kinases are important regulators of mitochondrial functions.
Int. J. Biochem. Cell Biol. 45, 90e98.
Heeneman, S., Haendeler, J., Saito, Y., Ishida, M., Berk, B.C., 2000. Angiotensin II induces
transactivation of two different populations of the platelet-derived growth factor beta
receptor. Key role for the p66 adaptor protein Shc. J. Biol. Chem. 275, 15926e15932.
Hink, U., Li, H., Mollnau, H., Oelze, M., Matheis, E., Hartmann, M., Skatchkov, M.,
Thaiss, F., Stahl, R.A., Warnholtz, A., Meinertz, T., Griendling, K., Harrison, D.G.,
Forstermann, U., Munzel, T., 2001. Mechanisms underlying endothelial dysfunction
in diabetes mellitus. Circ. Res. 88, E14eE22.
Hirano, R., Sasamoto, W., Matsumoto, A., Itakura, H., Igarashi, O., Kondo, K., 2001. Anti-
oxidant ability of various flavonoids against DPPH radicals and LDL oxidation. J. Nutr.
Sci. Vitaminol. (Tokyo) 47, 357e362.
Hirooka, Y., 2008. Role of reactive oxygen species in brainstem in neural mechanisms of
hypertension. Auton. Neurosci. 142, 20e24.
Hofmann, F., Bernhard, D., Lukowski, R., Weinmeister, P., 2009. cGMP regulated protein
kinases (cGK). Handb. Exp. Pharmacol. 137e162.
Holmstr€ om, K.M., Finkel, T., 2014. Cellular mechanisms and physiological consequences of
redox-dependent signalling. Nat. Rev. Mol. Cell Biol. 15, 411e421.
Houstis, N., Rosen, E.D., Lander, E.S., 2006. Reactive oxygen species have a causal role in
multiple forms of insulin resistance. Nature 440, 944e948.
Hsieh, H.L., Yang, C.M., 2013. Role of redox signaling in neuroinflammation and neuro-
degenerative diseases. Biomed. Res. Int. 2013, 484613.
Humphries, K.M., Deal, M.S., Taylor, S.S., 2005. Enhanced dephosphorylation of cAMP-
dependent protein kinase by oxidation and thiol modification. J. Biol. Chem. 280,
2750e2758.
Hybertson, B.M., Gao, B., Bose, S.K., McCord, J.M., 2011. Oxidative stress in health and
disease: the therapeutic potential of Nrf 2 activation. Mol. Asp. Med. 32, 234e246.
Jankov, R.P., Kantores, C., Pan, J., Belik, J., 2008. Contribution of xanthine oxidase-derived
superoxide to chronic hypoxic pulmonary hypertension in neonatal rats. Am. J. Physiol.
Lung Cell Mol. Physiol. 294, L233eL245.
Juarez, J.C., Manuia, M., Burnett, M.E., Betancourt, O., Boivin, B., Shaw, D.E.,
Tonks, N.K., Mazar, A.P., Do~ nate, F., 2008. Superoxide dismutase 1 (SOD1) is essential
for H2O2-mediated oxidation and inactivation of phosphatases in growth factor
signaling. Proc. Natl. Acad. Sci. U.S.A. 105, 7147e7152.
Kamata, H., Honda, S., Maeda, S., Chang, L., Hirata, H., Karin, M., 2005. Reactive oxygen
species promote TNFalpha-induced death and sustained JNK activation by inhibiting
MAP kinase phosphatases. Cell 120, 649e661.
Kawamori, D., Kajimoto, Y., Kaneto, H., Umayahara, Y., Fujitani, Y., Miyatsuka, T.,
Watada, H., Leibiger, I.B., Yamasaki, Y., Hori, M., 2003. Oxidative stress induces
nucleo-cytoplasmic translocation of pancreatic transcription factor PDX-1 through acti-
vation of c-Jun NH(2)-terminal kinase. Diabetes 52, 2896e2904.
Kemble, D.J., Sun, G., 2009. Direct and specific inactivation of protein tyrosine kinases in the
Src and FGFR families by reversible cysteine oxidation. Proc. Natl. Acad. Sci. U.S.A.
106, 5070e5075.
Kido, Y., Nakae, J., Accili, D., 2001. Clinical review 125: the insulin receptor and its cellular
targets. J. Clin. Endocrinol. Metab. 86, 972e979.
Konior, A., Schramm, A., Czesnikiewicz-Guzik, M., Guzik, T.J., 2014. NADPH oxidases in
vascular pathology. Antioxid. Redox Signal 20, 2794e2814.
Koo, Y.E., Fan, W., Hah, H., Xu, H., Orringer, D., Ross, B., Rehemtulla, A.,
Philbert, M.A., Kopelman, R., 2007. Photonic explorers based on multifunctional nano-
platforms for biosensing and photodynamic therapy. Appl. Opt. 46, 1924e1930.
Krinsky, N.I., 2001. Carotenoids as antioxidants. Nutrition 17, 815e817.
Kuroda, J., Nakagawa, K., Yamasaki, T., Nakamura, K., Takeya, R., Kuribayashi, F.,
Imajoh-Ohmi, S., Igarashi, K., Shibata, Y., Sueishi, K., Sumimoto, H., 2005. The super-
oxide-producing NAD(P)H oxidase Nox4 in the nucleus of human vascular endothelial
cells. Genes Cells 10, 1139e1151.
Kwon, J., Lee, S.R., Yang, K.S., Ahn, Y., Kim, Y.J., Stadtman, E.R., Rhee, S.G., 2004.
Reversible oxidation and inactivation of the tumor suppressor PTEN in cells stimulated
with peptide growth factors. Proc. Natl. Acad. Sci. U.S.A. 101, 16419e16424.
Lassegue, B., Sorescu, D., Sz€ ocs, K., Yin, Q., Akers, M., Zhang, Y., Grant, S.L.,
Lambeth, J.D., Griendling, K.K., 2001. Novel gp91(phox) homologues in vascular
smooth muscle cells: nox1 mediates angiotensin II-induced superoxide formation and
redox-sensitive signaling pathways. Circ. Res. 88, 888e894.
Li, Q., Zhang, Y., Marden, J.J., Banfi, B., Engelhardt, J.F., 2008. Endosomal NADPH ox-
idase regulates c-Src activation following hypoxia/reoxygenation injury. Biochem. J.
411, 531e541.
Lisman, J., Schulman, H., Cline, H., 2002. The molecular basis of CaMKII function in syn-
aptic and behavioural memory. Nat. Rev. Neurosci. 3, 175e190.
Lornejad-Sch€afer, M.R., Sch€afer, C., Sch€ offl, H., Frank, J., 2009. Cytoprotective role of
mitogen-activated protein kinase phosphatase-1 in light-damaged human retinal
pigment epithelial cells. Photochem. Photobiol. 85, 834e842.
Ludwig, A., Sommer, A., Uhlig, S., 2011. Assessment of endothelial permeability and leukocyte
transmigration in human endothelial cell monolayers. Methods Mol. Biol. 763, 319e332.
Lund, J., Hafstad, A.D., Boardman, N.T., Rossvoll, L., Rolim, N.P., Ahmed, M.S.,
Florholmen, G., Attramadal, H., Wisløff, U., Larsen, T.S., Aasum, E., 2015. Exercise
training promotes cardioprotection through oxygen-sparing action in high-fat fed mice.
Am. J. Physiol. Heart Circ. Physiol. http://dx.doi.org/10.1152/ajpheart.00734.02014.
Magesh, S., Chen, Y., Hu, L., 2012. Small molecule modulators of Keap1-Nrf 2-ARE
pathway as potential preventive and therapeutic agents. Med. Res. Rev. 32, 687e726.
Mallozzi, C., Di Stasi, A.M., Minetti, M., 1999. Activation of src tyrosine kinases by
peroxynitrite. FEBS Lett. 456, 201e206.
Maritim, A.C., Sanders, R.A., Watkins, J.B., 2003. Diabetes, oxidative stress, and antioxi-
dants: a review. J. Biochem. Mol. Toxicol. 17, 24e38.
Martyn, K.D., Frederick, L.M., von Loehneysen, K., Dinauer, M.C., Knaus, U.G., 2006.
Functional analysis of Nox4 reveals unique characteristics compared to other NADPH
oxidases. Cell Signal 18, 69e82.
Matsubara, K., Higaki, T., Matsubara, Y., Nawa, A., 2015. Nitric oxide and reactive oxygen
species in the pathogenesis of preeclampsia. Int. J. Mol. Sci. 16, 4600e4614.
Matsuoka, T.A., Zhao, L., Artner, I., Jarrett, H.W., Friedman, D., Means, A., Stein, R.,
2003. Members of the large Maf transcription family regulate insulin gene transcription
in islet beta cells. Mol. Cell Biol. 23, 6049e6062.
McNally, J.S., Davis, M.E., Giddens, D.P., Saha, A., Hwang, J., Dikalov, S., Jo, H.,
Harrison, D.G., 2003. Role of xanthine oxidoreductase and NAD(P)H oxidase in endo-
thelial superoxide production in response to oscillatory shear stress. Am. J. Physiol. Heart
Circ. Physiol. 285, H2290eH2297.
Milkiewicz, M., Roudier, E., Doyle, J.L., Trifonova, A., Birot, O., Haas, T.L., 2011. Iden-
tification of a mechanism underlying regulation of the anti-angiogenic forkhead tran-
scription factor FoxO1 in cultured endothelial cells and ischemic muscle. Am. J.
Pathol. 178, 935e944.
Mittal, M., Siddiqui, M.R., Tran, K., Reddy, S.P., Malik, A.B., 2014. Reactive
oxygen species in inflammation and tissue injury. Antioxid. Redox Signal 20,
1126e1167.
Miyamoto, Y., Akaike, T., Yoshida, M., Goto, S., Horie, H., Maeda, H., 1996. Potentiation
of nitric oxide-mediated vasorelaxation by xanthine oxidase inhibitors. Proc. Soc. Exp.
Biol. Med. 211, 366e373.
Mollnau, H., Schulz, E., Daiber, A., Baldus, S., Oelze, M., August, M., Wendt, M.,
Walter, U., Geiger, C., Agrawal, R., Kleschyov, A.L., Meinertz, T., M€ unzel, T.,
2003. Nebivolol prevents vascular NOS III uncoupling in experimental hyperlipidemia
and inhibits NADPH oxidase activity in inflammatory cells. Arterioscler. Thromb. Vasc.
Biol. 23, 615e621.
Negi, G., Kumar, A., Joshi, R.P., Sharma, S.S., 2011. Oxidative stress and Nrf2 in the path-
ophysiology of diabetic neuropathy: old perspective with a new angle. Biochem.
Biophys. Res. Commun. 408, 1e5.
Nogueira, V., Hay, N., 2013. Molecular pathways: reactive oxygen species homeostasis in
cancer cells and implications for cancer therapy. Clin. Cancer Res. 19, 4309e4314.
Nordberg, J., Arnér, E.S., 2001. Reactive oxygen species, antioxidants, and the mammalian
thioredoxin system. Free Radic. Biol. Med. 31, 1287e1312.
Nose, K., Shibanuma, M., Kikuchi, K., Kageyama, H., Sakiyama, S., Kuroki, T., 1991.
Transcriptional activation of early-response genes by hydrogen peroxide in a mouse oste-
oblastic cell line. Eur. J. Biochem. 201, 99e106.
Nunn, A.V., Guy, G.W., Bell, J.D., 2010. Endocannabinoids, FOXO and the metabolic
syndrome: redox, function and tipping pointethe view from two systems. Immunobi-
ology 215, 617e628.
Ostergaard, H., Henriksen, A., Hansen, F.G., Winther, J.R., 2001. Shedding light on disul-
fide bond formation: engineering a redox switch in green fluorescent protein. EMBO J.
20, 5853e5862.
Pamukcu, B., Lip, G.Y., Shantsila, E., 2011. The nuclear factorekappa B pathway in athero-
sclerosis: a potential therapeutic target for atherothrombotic vascular disease. Thromb.
Res. 128, 117e123.
Pantano, C., Reynaert, N.L., van der Vliet, A., Janssen-Heininger, Y.M., 2006. Redox-sen-
sitive kinases of the nuclear factor-kappaB signaling pathway. Antioxid. Redox Signal 8,
1791e1806.
Pendyala, S., Usatyuk, P.V., Gorshkova, I.A., Garcia, J.G., Natarajan, V., 2009. Regulation
of NADPH oxidase in vascular endothelium: the role of phospholipases, protein kinases,
and cytoskeletal proteins. Antioxid. Redox Signal 11, 841e860.
Podrez, E.A., Abu-Soud, H.M., Hazen, S.L., 2000. Myeloperoxidase-generated oxidants
and atherosclerosis. Free Radic. Biol. Med. 28, 1717e1725.
Reynaert, N.L., Ckless, K., Korn, S.H., Vos, N., Guala, A.S., Wouters, E.F., van der
Vliet, A., Janssen-Heininger, Y.M., 2004. Nitric oxide represses inhibitory kappaB
kinase through S-nitrosylation. Proc. Natl. Acad. Sci. U.S.A. 101, 8945e8950.
Reynaert, N.L., van der Vliet, A., Guala, A.S., McGovern, T., Hristova, M., Pantano, C.,
Heintz, N.H., Heim, J., Ho, Y.S., Matthews, D.E., Wouters, E.F., Janssen-
Heininger, Y.M., 2006. Dynamic redox control of NF-kappaB through glutaredoxin-
regulated S-glutathionylation of inhibitory kappaB kinase beta. Proc. Natl. Acad. Sci.
U.S.A. 103, 13086e13091.
Ristow, M., Zarse, K., Oberbach, A., Kl€ oting, N., Birringer, M., Kiehntopf, M.,
Stumvoll, M., Kahn, C.R., Bl€ uher, M., 2009. Antioxidants prevent health-promoting
effects of physical exercise in humans. Proc. Natl. Acad. Sci. U.S.A. 106, 8665e8670.
Roberts, J.M., Taylor, R.N., Musci, T.J., Rodgers, G.M., Hubel, C.A., McLaughlin, M.K.,
1989. Preeclampsia: an endothelial cell disorder. Am. J. Obstet. Gynecol. 161, 1200e1204.
Rochette, L., Zeller, M., Cottin, Y., Vergely, C., 2014. Diabetes, oxidative stress and ther-
apeutic strategies. Biochim. Biophys. Acta 1840, 2709e2729.
Roskoski, R., 2004. Src protein-tyrosine kinase structure and regulation. Biochem. Biophys.
Res. Commun. 324, 1155e1164.
Roskoski, R., 2005. Src kinase regulation by phosphorylation and dephosphorylation.
Ryan, K.A., Smith, M.F., Sanders, M.K., Ernst, P.B., 2004. Reactive oxygen and nitrogen
species differentially regulate Toll-like receptor 4-mediated activation of NF-kappa B
and interleukin-8 expression. Infect. Immun. 72, 2123e2130.
Schlessinger, J., 2002. Ligand-induced, receptor-mediated dimerization and activation of
EGF receptor. Cell 110, 669e672.
Schrader, M., Fahimi, H.D., 2004. Mammalian peroxisomes and reactive oxygen species.
Histochem. Cell Biol. 122, 383e393.
Shabb, J.B., 2001. Physiological substrates of cAMP-dependent protein kinase. Chem. Rev.
101, 2381e2411.
Shi, Q., Gibson, G.E., 2007. Oxidative stress and transcriptional regulation in Alzheimer
disease. Alzheimer Dis. Assoc. Disord. 21, 276e291.
Son, Y., Cheong, Y.K., Kim, N.H., Chung, H.T., Kang, D.G., Pae, H.O., 2011. Mitogen-
activated protein kinases and reactive oxygen species: how can ROS activate MAPK
pathways? J. Signal Transduct. 2011, 792639.
Storz, P., 2011. Forkhead homeobox type O transcription factors in the responses to oxida-
tive stress. Antioxid. Redox Signal 14, 593e605.
Stuart, J.A., Maddalena, L.A., Merilovich, M., Robb, E.L., 2014. A midlife crisis for the
mitochondrial free radical theory of aging. Longev. Heal. 3, 4.
Sundaresan, M., Yu, Z.X., Ferrans, V.J., Irani, K., Finkel, T., 1995. Requirement for gen-
eration of H2O2 for platelet-derived growth factor signal transduction. Science 270,
296e299.
Sundaresan, M., Yu, Z.X., Ferrans, V.J., Sulciner, D.J., Gutkind, J.S., Irani, K., Goldschmidt-
Clermont, P.J., Finkel, T., 1996. Regulation of reactive-oxygen-species generation in
fibroblasts by Rac1. Biochem. J. 318 (Pt 2), 379e382.
Tang, H., Hao, Q., Rutherford, S.A., Low, B., Zhao, Z.J., 2005. Inactivation of SRC
family tyrosine kinases by reactive oxygen species in vivo. J. Biol. Chem. 280,
23918e23925.
Thannickal, V.J., Day, R.M., Klinz, S.G., Bastien, M.C., Larios, J.M., Fanburg, B.L., 2000.
Ras-dependent and -independent regulation of reactive oxygen species by mitogenic
growth factors and TGF-beta1. FASEB J. 14, 1741e1748.
Thum, T., Borlak, J., 2004. Mechanistic role of cytochrome P450 monooxygenases in
oxidized low-density lipoprotein-induced vascular injury: therapy through LOX-1
receptor antagonism? Circ. Res. 94, e1e13.
Tonks, N.K., 2006. Protein tyrosine phosphatases: from genes, to function, to disease. Nat.
Rev. Mol. Cell Biol. 7, 833e846.
Toyokuni, S., Okamoto, K., Yodoi, J., Hiai, H., 1995. Persistent oxidative stress in cancer.
FEBS Lett. 358, 1e3.
Truong, T.H., Carroll, K.S., 2012. Redox regulation of epidermal growth factor receptor
signaling through cysteine oxidation. Biochemistry 51, 9954e9965.
Truong, T.H., Carroll, K.S., 2013. Redox regulation of protein kinases. Crit. Rev. Biochem.
Mol. Biol. 48, 332e356.
Tschudi, M.R., Mesaros, S., L€ uscher, T.F., Malinski, T., 1996. Direct in situ measurement of
nitric oxide in mesenteric resistance arteries. Increased decomposition by superoxide in
hypertension. Hypertension 27, 32e35.
Ushio-Fukai, M., Alexander, R.W., Akers, M., Yin, Q., Fujio, Y., Walsh, K.,
Griendling, K.K., 1999. Reactive oxygen species mediate the activation of Akt/
protein kinase B by angiotensin II in vascular smooth muscle cells. J. Biol. Chem.
274, 22699e22704.
Van de Bittner, G.C., Dubikovskaya, E.A., Bertozzi, C.R., Chang, C.J., 2010. In vivo
imaging of hydrogen peroxide production in a murine tumor model with a chemoselec-
tive bioluminescent reporter. Proc. Natl. Acad. Sci. U.S.A. 107, 21316e21321.
Van Gaal, L.F., Vertommen, J., De Leeuw, I.H., 1998. The in vitro oxidizability of
lipoprotein particles in obese and non-obese subjects. Atherosclerosis 137 (Suppl.),
S39eS44.
Vigneron, A., Vousden, K.H., 2010. p53, ROS and senescence in the control of aging. Aging
(Albany NY) 2, 471e474.
Vincent, H.K., Powers, S.K., Stewart, D.J., Shanely, R.A., Demirel, H., Naito, H., 1999.
Obesity is associated with increased myocardial oxidative stress. Int. J. Obes. Relat.
Metab. Disord. 23, 67e74.
Visconti, R., Grieco, D., 2009. New insights on oxidative stress in cancer. Curr. Opin. Drug
Discov. Devel. 12, 240e245.
Wang, X., McCullough, K.D., Franke, T.F., Holbrook, N.J., 2000. Epidermal growth factor
receptor-dependent Akt activation by oxidative stress enhances cell survival. J. Biol.
Chem. 275, 14624e14631.
Watson, J.D., 2014. Type 2 diabetes as a redox disease. Lancet 383, 841e843.
van Wetering, S., van Buul, J.D., Quik, S., Mul, F.P., Anthony, E.C., ten Klooster, J.P.,
Collard, J.G., Hordijk, P.L., 2002. Reactive oxygen species mediate Rac-induced loss
of cell-cell adhesion in primary human endothelial cells. J. Cell Sci. 115, 1837e1846.
Wind, S., Beuerlein, K., Eucker, T., M€ uller, H., Scheurer, P., Armitage, M.E., Ho, H.,
Schmidt, H.H., Wingler, K., 2010. Comparative pharmacology of chemically distinct
NADPH oxidase inhibitors. Br. J. Pharmacol. 161, 885e898.
Winter-Vann, A.M., Johnson, G.L., 2007. Integrated activation of MAP3Ks balances cell fate
in response to stress. J. Cell Biochem. 102, 848e858.
Woolley, J.F., Stanicka, J., Cotter, T.G., 2013. Recent advances in reactive oxygen species
measurement in biological systems. Trends Biochem. Sci. 38, 556e565.
Xu, S., Chamseddine, A.H., Carrell, S., Miller, F.J., 2014. Nox4 NADPH oxidase contrib-
utes to smooth muscle cell phenotypes associated with unstable atherosclerotic plaques.
Redox Biol. 2, 642e650.
Yamashita, N., Hoshida, S., Otsu, K., Asahi, M., Kuzuya, T., Hori, M., 1999. Exercise pro-
vides direct biphasic cardioprotection via manganese superoxide dismutase activation.
J. Exp. Med. 189, 1699e1706.
Yang, H., Raizada, M.K., 1999. Role of phosphatidylinositol 3-kinase in angiotensin II regu-
lation of norepinephrine neuromodulation in brain neurons of the spontaneously hyper-
tensive rat. J. Neurosci. 19, 2413e2423.
Zimmerman, M.C., Dunlay, R.P., Lazartigues, E., Zhang, Y., Sharma, R.V.,
Engelhardt, J.F., Davisson, R.L., 2004. Requirement for Rac1-dependent NADPH
oxidase in the cardiovascular and dipsogenic actions of angiotensin II in the brain.
Circ. Res. 95, 532e539.
Zinovkin, R.A., Romaschenko, V.P., Galkin, I.I., Zakharova, V.V., Pletjushkina, O.Y.,
Chernyak, B.V., Popova, E.N., 2014. Role of mitochondrial reactive oxygen species
in age-related inflammatory activation of endothelium. Aging (Albany NY) 6, 661e674.
CHAPTER SEVEN
Regeneration, Stem Cells, and

Aging in the Tunicate Ciona:
Insights from the Oral Siphon
William R. Jeffery1, 2
1
Eugene Bell Center for Regenerative Biology and Tissue Engineering, Marine Biological Laboratory,
Woods Hole, MA, USA
2
Department of Biology, University of Maryland, College Park, MD, USA
E-mail: Jeffery@umd.edu
Contents
1. Introduction 256
2. Background 257
2.1 Life Cycle, Adult Organization, and Growth 257
2.2 Partial Body Regeneration 260
2.3 OS Model 260
3. OS Regeneration 262
3.1 Siphon Tip and Tube Regeneration 264
3.1.1 OPO replacement 264
3.1.2 Short-distance regeneration 265
3.1.3 Long-distance regeneration 267
3.2 Siphon Base Regeneration 268
4. Adult Stem Cells 269
4.1 Multiple Stem Cells 269
4.2 Branchial Sac Stem Cells 271
5. Stem and Progenitor Cell Mobilization and Deployment 271
6. Aging and OS Regeneration 275
7. Concluding Remarks and Perspectives 278
Acknowledgments 279
References 279
Abstract
Regeneration studies in the tunicate Ciona intestinalis have recently been focused on the
potential of adult stem cells to replace injured tissues and organs during the adult life
cycle using the oral siphon (OS) as a model. The OS has oral siphon pigment organs
(OPOs) along its rim and an underlying network of muscle fibers in its tube. Different
regeneration processes are triggered by OS amputation at the tip, along the tube, or
at the base. One process involves the replacement of OPOs without new cell division
by direct differentiation of locally deployed stem cells or stem cells that migrate from
j
ISSN 1937-6448
256 William R. Jeffery
the branchial sac. Another process involves blastema formation by the migration of pro-
genitor cells produced from branchial sac stem cells. The capacity for complete and ac-
curate OS regeneration declines continuously during the adult life cycle. Finally, after an
age threshold is reached, OS regeneration ceases in old animals. The loss of regeneration
capacity in old animals involves the depletion of stem cells in the branchial sac, the
inability of branchial sac progenitor cells to migrate to the sites of regeneration, and
defective oral pigment organ replacement. The significance of the OS model for studying
regeneration, stem cells, and aging will be enhanced by the application of molecular
methods.
1. INTRODUCTION
The ability to replace injured tissues and organs is a key feature of many
different animals (Brockes and Kumar, 2008). However, regenerative poten-
tial has been repeatedly lost during the evolution of some animal groups (Bely,
2010; Poss, 2010). For example, although certain vertebrates, such as teleost
fishes and urodele amphibians, are able to replace severed appendages
completely, anuran amphibians, birds, and mammals do not have this capacity,
at least as mature adults. The loss of regenerative capacity has a developmental
basis. Anuran tadpoles can replace severed limbs prior to metamorphosis, but
not as adult frogs (Givan et al., 2002). Young opossums can replace their
slowly developing hind limbs but not their rapidly developing forelimbs
(Mizell, 1968). Some human children also have the capacity to replace severed
digits but this ability disappears later during childhood (Illingworth and
Barker, 1980). Even species with powerful regeneration and tissue repair ca-
pacities as young adults show subsequent reduction in this ability during aging
(Reed et al., 2003; Poss, 2010; Seifert and Voss, 2013; Sousounis et al., 2014).
Important clues about how and why the ability to regenerate has been lost
during evolution might be obtained by studying the decay of regenerative po-
tential during aging, a process termed “regenerative aging.”
To study regenerative aging, we need to understand the general principles
of regeneration in an animal that has extensive tissue and organ replacement
capacities. Questions to be answered include: (1) what steps are necessary for
regeneration, (2) what cells and molecules are involved in replacing tissues
and organs, and (3) how are regenerative processes regulated? Then, we
need to understand how aging affects regeneration. Ultimately, it would
be desirable to obtain enough information to reverse regenerative aging.
This knowledge then might also be used to rescue regeneration in animals
that have evolved restricted regenerative capacity, such as mammals.
Ciona Regeneration, Stem Cells, and Aging 257
Achieving these goals requires a model animal with favorable attributes

for studying regenerative aging. These attributes include powerful regener-
ative capacities during youth and more restricted capacities during aging.
Other favorable attributes would be: the ability to study aging in the labo-
ratory over a reasonably short time period, a structurally simple organism
capable of rapid and accurate regeneration, anddsince a long-term goal is
the reversal of lost regenerative processes in vertebratesdthe model animal
should be closely related to vertebrates.
One of the most suitable organisms for this type of study is the ascidian
tunicate Ciona intestinalis (Kourakis and Smith, 2015). First, it can be grown
in laboratory culture from eggs to adults (Cirino et al., 2002; Joly et al.,
2007), offering the opportunity of testing regenerative potential in individ-
ual animals at any stage of the life cycle. Second, Ciona has a relatively short
life cycle of about 1.5 years (Berrill, 1947; Millar, 1952; Dybern, 1965;
Peterson et al., 1995; Kourakis and Smith, 2015). Third, it has a fairly simple
body plan and contains fewer different types of tissues and organs than ver-
tebrates (Millar, 1953). Fourth, it has a long history of use as a model in
developmental biology (Satoh, 1994, 2013), and a large molecular toolkit
is available for studying this organism (Stolfi and Christiaen, 2012). Fifth,
Ciona, like other tunicates (Berrill, 1951; Rinkevich et al., 1995; Tiozzo
et al., 2008), has powerful regenerative capacities and displays regenerative
aging (Jeffery, 2015a). Lastly and most importantly, Ciona belongs to a chor-
date group (tunicates) that has been inferred as the closest living relative of
the vertebrates (Bourlat et al., 2006; Desuc et al., 2006).
There is a long history of regeneration research in adult Ciona (Jeffery,
2015b) but few studies have considered regenerative phenomena during
the life cycle or in the context of stem cell function and aging. Here, we
review recent studies on regeneration, stem cells, and aging in Ciona using
the oral siphon (OS) as a model.
2. BACKGROUND
2.1 Life Cycle, Adult Organization, and Growth
Ciona is a solitary ascidian with a life cycle consisting of larval and adult
stages (Figure 1(A)). The motile nonfeeding larva, which consists of a trunk
(or head) and a tail, is formed from an egg within about a day after fertiliza-
tion by rapid stereotypical cleavages and post-gastrulation cell movements
(Satoh, 1994). Like other ascidians, embryonic development in Ciona is
Figure 1 Ciona intestinalis. (A) Life cycle. (B) Body organization. (C) OS structure
showing the approximate locations (red (gray in print versions) horizontal lines) of
amputations at the tip, along the tube, and at the base in regeneration studies. OS,
oral siphon; AS, atrial siphon; NC, neural complex; BS, branchial sac; V, Viscera; CM,
circular muscle band; LM, longitudinal muscle band; PB, Inter-OPO pigment band;
OPO, oral siphon pigment organ; TB, tentacle band. C. Modified from Auger et al.
(2010).
strictly determinate: ablation or removal of blastomeres results in missing

parts of the embryo or larva (Jeffery, 2001). Likewise, excision of larval parts,
such as the tail (Takamura et al., 2002), does not result their replacement.
Therefore, regenerative potential is not apparent during the embryonic
and larval stages of the Ciona life cycle.
After a free swimming period, the larva settles head first on a substrate
and undergoes metamorphosis, a process in which the larval tail is retracted
into the trunk, and the latter then undergoes reorganization and differenti-
ation of new body parts to form a sessile filter feeding adult (Figure 1(A) and
(B)). The newly formed distal body tissues and organs first become capable
of regeneration after metamorphosis (Jeffery, 2015b). The mechanisms
responsible for the conversion of a nonregenerating larva into a regenerating
adult are unknown in Ciona and other tunicates.
Adult Ciona have an elongate vase-like body consisting of proximal (to-
ward the attached base) viscera, a large pharynx containing a branchial sac,
and two distal siphons (Figure 1(B)). The tunic, a protective layer containing
cellulose-like material that is a defining feature of the tunicates, covers the
entire body. The perforated branchial sac is suspended in the pharyngeal
cavity leading into the OS, the orifice through which food particles are
driven into the underlying branchial sac for entrapment and passing to the
stomach for digestion (Figure 1(B)). The endostyle, an elongate tube-shaped
organ, which is considered to be the ascidian homolog to the vertebrate thy-
roid gland (Ogasawara and Satoh, 1998), runs along the ventral side of the
branchial sac, and the dorsal strand, a neural tissue, runs along its dorsal
side. The viscera include the stomach, intestine, heart, and hermaphroditic
gonad. The gonad and intestine are connected to gonoducts and a rectum
respectively, which empty into the atrial cavity. The atrial siphon is used
to expel gametes and fecal materials (Figure 1(B)). The neural complex
(NC), which consists of a single ganglion (brain) with an associated glandular
organ (neural gland), lies in the apex between the two siphons and tapers
into the dorsal strand on its posterior side. The brain is quite small, contain-
ing only a few hundred neurons and possibly some associated supporting
cells (Millar, 1953; Bollner et al., 1992, 1993, 1995, 1997; Dahlberg et al.,
2009). Paired nerve tracts exit the anterior and posterior ends of the brain
and extend into the siphons, branchial sac, and viscera, dividing into multi-
ple tracts and fibers along the way (Millar, 1953; Markman, 1958; Mackie
et al., 2006).
Ciona adults grow rapidly and isometrically, eventually developing go-
nads and becoming gravid in only a few months (Berrill, 1947; Millar,
1952; Dybern, 1965; Peterson et al., 1995). During the growth phase, the
NC and siphons show remarkable regenerative potential. However, the
rate of regeneration declines continuously during the life cycle (Dahlberg
et al., 2009; Auger et al., 2010). After about 1.5 years, gamete production
ceases, the tunic thickens and wrinkles, and the old animals eventually die
(Figure 1(A)). Regeneration capacity is severely compromised or can cease
entirely during old age (Jeffery, 2012, 2015c).
In summary, the Ciona life cycle exhibits three different periods with re-
gard to regenerative capacity: (1) the embryonic/larval phase in which
regeneration is absent, (2) the young and middle-age adult growth phase
in which regeneration is extensive, and (3) old age in which regeneration
is weak or nonexistent (Figure 1(A)). In this article, we focus on the second
and third periods of the life cycle.
2.2 Partial Body Regeneration

Mingazzini (1891) and Loeb (1892) independently discovered the remark-
able regenerative capacities of the Ciona NC and siphons. Subsequently, a
detailed analysis showed that the NC could be completely replaced within
about a month after ablation (Schultze, 1899). Many different investigators
have now confirmed NC and OS regeneration and contributed additional
details about the processes (Sutton, 1953; Whittaker, 1975; Bollner et al.,
1992, 1993, 1995, 1997; Dahlberg et al., 2009; Auger et al., 2010). The ca-
pacity for complete regeneration from body fragments has also been inves-
tigated. Hirschler (1914) separated adult body parts and recorded their
capacity to regenerate an entire animal. He bisected animals into two parts
by cutting across a horizontal plane immediately distal to the viscera and
found that the proximal (basal) portion was capable of regenerating the distal
portion, and thus a complete animal, but the distal portion could not regen-
erate the basal part and eventually died. Furthermore, if body pieces were
separated at positions more distal to the border between the branchial sac
and viscera, the basal parts, but not the distal parts, could regenerate com-
plete animals. Using a similar approach, Jeffery (2015c) separated the body
into three parts across the distaleproximal axis and showed that the middle
part, which lacked the viscera, NC, or the siphons, but contained part of the
branchial sac, also has the potential to replace the OS. Therefore, in Ciona,
the branchial sac appears to be a general regenerative center that is involved
in the replacement of injured distal body parts, including the OS.
The phenomenon of distal regeneration from proximal parts of the adult
body was termed “partial body regeneration” (Jeffery, 2015b) to distinguish
it from other types of regeneration displayed by ascidians: bipolar regenera-
tion in Clavellina (Brien, 1968) and whole-body regeneration in colonial
ascidians (Freeman, 1964; Tiozzo et al., 2008; Brown and Swalla, 2012).
2.3 OS Model
Recent studies of Ciona regeneration have focused on the OS as a model
(Auger et al., 2010; Jeffery, 2012, 2015c). The OS is a thin muscular tube
covered by tunic on its outer side (Figures 1(C) and 2(A)). The siphon tube is
bounded by inner and outer epidermal epithelia, which cover a mesen-
chymal layer containing crisscrossing longitudinal and circular muscle bands
(Figure 2(B)), nerve tracts, and loosely organized cells embedded in a dense
extracellular matrix. Different sensory organs are located at the distal and
proximal ends of the OS. The proximal end contains a ring of sensory ten-
tacles, which can interdigitate and close off the siphon aperture at its base.
(A) (B)
(C) (D)
(E) (F)
Figure 2 Oral siphon (OS) pigment organ structure. (A) The distal portion of an
adult showing the OS, atrial siphon (AS), and a longitudinal muscle band (LMB). (B) A
phalloidin-stained OS showing bands of circular muscle (CM) and longitudinal muscle
(LM). (C) The OS rim with an oral siphon pigment organ (OPO) and the inter-OPO
pigment band (PB). (D) A drawing showing an OPO in section. The inner side of the
OS is on the right. PC, yellow (light gray in print versions) pigment cells; OC, orange
(dark gray in print versions) pigment cup; RC, neuronal receptor cells in the epidermal
crypt. (E) A section through the distal rim of an OS showing the structure of an OPO. (F)
A fluorescent micrograph of an OS from a transgenic animal with a green fluorescent
protein-labeled neural ganglion (NG) located at the base of the OPO (outlined). From
Auger et al. (2010).
The distal rim contains pigmented sensory organs (OS sensory organs or
OPOs) on its inner side, which lie in niches between extended lobes of
siphon wall tissue (Dilly and Wolken, 1973) (Figures 1(C) and 2(A)).
Each OPO has three parts: a crypt of elongated epidermal cells, which are
ciliated and rich in actin filaments, a cup-shaped layer of orange mesen-
chymal pigment cells, and a ganglion-like neural structure at its base
(Figure 2(CeF)). A yellow stripe of pigment cells, the inter-OPO pigment
band, extends laterally along the siphon rim between each oral siphon
pigment organ (OPO). A ring of OPOs connected by an inter-OPO pigment
band also rims the atrial siphon (Figure 2(A)), although there are no tentacles
at its base. The OPOs are thought to be sensory organs (Dilly and Wolken,
1973; Auger et al., 2010), but their precise function is not understood.
The siphons are highly contractile organs (Mackie et al., 2006; Dahlberg
et al., 2009). They have the capacity to shorten along their distal to proximal
axis by contracting the longitudinal muscle bands and their lateral axis by
contracting the circular muscle bands. Contractions can be elicited by water
movements or by objects touching the surface of the body.
3. OS REGENERATION
OS regeneration has been separated into three steps (Auger et al.,
2010). The first step is wound epidermis formation, which involves
epidermal cell division to produce a new epithelium closing over the wound
site (Figure 3(A)). The wound epidermis forms within the first day after
amputation. After the wound epidermis is formed, the OPOs are replaced
(Figure 3(BeD)). The formation of the wound epidermis and the early steps
in OPO replacement (see Section 5) occur before the beginning of siphon
regrowth (Figure 3(AeC)). Several days after amputation, a blastema of
proliferating cells is formed proximal to the level of OPO replacement,
the siphon grows outward (first replacing only the lobes and then showing
isometric growth along the tube) (Figure 3(DeG)), and new muscle bands
and nerve tracts differentiate in the extending tube. A new ring of basal ten-
tacles is also formed if the original ring was removed by siphon amputation.
During the growth process, the neural ganglion at the base of each OPO is
replaced, but the process involved is not completely understood. An inter-
OPO pigment band is gradually re-formed during the period of siphon
growth by differentiation of yellow pigment cells extending along the
siphon rim from the lateral sides of each new OPO. Depending on the level
(A) (AA) (H) (I)
(B)
(BB)
(C)
(D)
(CC)
(E)
(F)
(DD)
(G)
(J) (K) (L)
(M)
Figure 3 Oral siphon (OS) regeneration. AeG (including AAeDD). OS regeneration and
OPO regeneration showing the first two steps in the absence of siphon growth (AeC)
and showing the third step involving growth (DeG). The time span from A to G is about
a month. AAeDD are magnifications of the siphon rim in AeD respectively showing the
wound epidermis (WE) and orange (gray in print versions) pigment cell differentiation.
OPO, oral siphon pigment organ; PB, inter-OPO pigment band; OPL, line of orange
pigment cells; OPS, orange pigment spot. (H, I) Diagram showing oral pigment organ
replacement after siphon amputation at the tip/tube (H) or base (I). (J, K) Siphon
tube regeneration showing (J) neural ganglion (NG) replacement and (K) the blastema
of proliferating cells. The diagonal dashed lines represent the approximate location of
amputations. (L, M) Siphon base regeneration showing multiple OPO replacement. M.
Enlargement of the siphon rim showing duplicated and triplicated oral siphon pigment
organs. From Auger et al. (2010).
of siphon amputation (at the distal tip, within the muscular tube, or at the
base) and the age of the animal, there are differences in the rate and details
of OS regeneration (Figure 1(C)). The differences in regeneration of siphon
parts after amputation at different levels is described below and summarized
in Table 1.
Table 1 Characteristics of regeneration after amputation at different levels of the

oral siphon
Amputation Regeneration Replacement Cell
level rate type Source Blastema division
Tip Fast Short distance SSC No No

Tube Fast Short distance SSC No No
Long distance BCSC Yes No
Long distance BSPC Yes Yes
Base Slow Long distance BCSC Yes No
Long distance BSPC Yes Yes
SSC: local siphon stem cells. BCSC: branchial sac stem cells. BSPC: branchial sac progenitor cells.
Data compiled from Auger et al. (2010) and Jeffery (2015c).
3.1 Siphon Tip and Tube Regeneration

When the OS is amputated at the tip (removing OPOs and the inter-OPO
pigment band) or along the muscular tube (above a level approximating the
tentacle bands) the OPOs are replaced rapidly and with a high degree of pre-
cision (Table 1).
3.1.1 OPO replacement

The stages of OPO replacement during siphon tip or tube regeneration
is shown in Figure 3(AAeDD) and (H). First, orange pigment cells diff-
erentiate in a band of mesenchyme below the wound epidermis
(Figure 3(AAeBB) and (H-1)). This region, called the OPO regeneration
band, is a vertical stripe of loosely organized mesenchymal cells located be-
tween each longitudinal band of muscle fibers along the entire length of the
siphon (Figure 1(C)). There are as many OPO regeneration bands as OPO
in the OS. Second, the differentiated orange pigment cells migrate distally
through the OPO regeneration band, line up along the wound epidermis,
and condense into pigment spots (Figure 3(CCeDD) and (H-2,3)). Third,
the surface epithelium invaginates to form the ciliated epidermal crypt, and
orange pigment spots surround the crypt, which becomes the orange cup of
the OPO (Figure 3(H-4,5)). At about the same time, the neural ganglion is
re-formed at the base of the epidermal crypt, and the inter-OPO pigment
band begins to be replaced.
Most animals in a natural population have a ring of eight OPO surround-
ing the siphon rim, although a small number of animals have rings of seven
or nine OPOs (Millar, 1953; Auger et al., 2010). There is a high degree of
fidelity in repatterning the number of OPOs during both siphon tip and
tube regeneration, as evidenced by their accurate replacement during mul-

tiple cycles of OS regeneration. Multiple cycles of regeneration are carried
out by: (1) amputating the OS along its tube, (2) allowing the siphon to
regrow, (3) amputating the regenerated siphon along its tube again, (4)
allowing siphon to regrow again, and so forth. The number of replaced
OPOs has been determined through four consecutive regeneration cycles
(Auger et al., 2010). After three consecutive cycles, animals with eight orig-
inal OPOs always formed exactly eight new OPO, whereas animals with
seven or nine original OPOs always formed exactly seven or nine new
OPOs respectively. The response of the siphon to multiple amputations in-
dicates that OPO replacement is a precisely controlled process. After the
fourth consecutive amputation cycle, however, there is no similarity
between the numbers of original and new OPOs, implying that pattern
regulation has been affected.
The siphon tip and tube regeneration processes are similar, both begin-
ning with the formation of a wound epidermis and subsequent replacement
of the OPOs and inter-OPO pigment band. However, since part of the
underlying muscle, nerve, and OPO regeneration band are also removed,
the tube regeneration process includes these tissues.
3.1.2 Short-distance regeneration

The source of progenitor cells has been a major question in Ciona
regenerative biology (Jeffery, 2015a,b). In principle, regeneration could be
a short-distance (local) process involving precursors from within the siphon
tube itself, a long-distance process involving precursors from another part
of the body, or a combination of both processes. Current evidence suggests
that siphon tip regeneration involves a short-distance process, whereas siphon
tube regeneration involves both short- and long-distance processes (Table 1)
(Auger et al., 2010; Jeffery, 2015c).
Two lines of evidence support a short-distance process in siphon tube
regeneration: UV irradiation and siphon explants (Auger et al., 2010)
(Figure (4)). First, UV irradiation of the siphon tube, which is carried out
while shielding the remainder of the body, blocks OPO replacement and
siphon regeneration following siphon amputation: if one side of the siphon
is irradiated, OPO replacement occurs only on the non-UV irradiated side,
whereas if both sides are irradiated, OPO replacement is blocked on both
sides (Figure 4(B) and (C)). The UV irradiation results imply that siphon
tube regeneration has a high degree of local autonomy. Autonomous
replacement of the OPOs by short-distance regeneration is also supported
Figure 4 Local source of progenitor cells for oral siphon pigment organ (OPO) regen-
eration demonstrated by oral siphon (OS) ultraviolet irradiation and explant experi-
ments. Top. Ultraviolet (UV) irradiation. (A, D) Normal OPO replacement after UV
irradiation of a shielded control animal prior to siphon tube amputation. (B, E) No
OPO replacement after UV irradiation of both sides of the OS prior to siphon tube
amputation. (C, F) No OPO regeneration on the left side of an animal after UV irradiation
of the left side of the OS prior to siphon tube amputation. Black rectangles indicate the
shielded areas during UV irradiation (downward pointing arrows). Vertical lines indicate
the amputation planes. Bottom. Explant cultures. (G) Diagram of the two-step proce-
dure used to obtain explants of regenerating siphons. The first amputation (1) yields
an OS tip explant and the second amputation (2) an OS mid-section explant. (H) A
siphon mid-section explant after 6 days in culture showing orange (dark gray in print
versions) pigment cell accumulation in OPO regeneration bands (arrowhead). Arrow
shows the exposed siphon tentacles. (I) Explants after 9 days in culture showing an
OS distal explant (top) containing the original OPOs on its distal margin and an OS
mid-section explant (bottom) with oral siphon pigment organ formation on its distal
(arrowhead) but not its proximal side. From Auger et al. (2010).
by the destruction of a single OPO by microcautery, which quickly induces

a new OPO to re-form in its place without affecting any other part of the
siphon (Auger et al., 2010). Second, if the siphon is amputated at its distal
tip and the tip and stump are cultured as explants, they show contrasting fates
(Figure 4(GeI)). The stump explants form OPO with fidelity in structure
and number on their distal but not their proximal ends. The tip explants,
which already have OPOs on their distal ends, do not form new OPO on
their proximal ends. Thus, OPO regeneration in culture respects the original
polarity of the siphon. The UV irradiation and explant experiments suggest
that the OS responds to injury by activating a local source of stem or pro-
genitors cells for OPO replacement.
Short-distance regeneration is independent of cell proliferation (Table 1).
This has been demonstrated by the inability of OPOs to be labeled with the
cell proliferation marker EdU and the inability of the cell division inhibitors
colchicine and nocodazole to affect OPO replacement (Jeffery, 2015c).
Accordingly, local stem or progenitor cells must differentiate directly into
new OPOs, which explains their rapid replacement during siphon tube
regeneration.
3.1.3 Long-distance regeneration

Long-distance regeneration, which occurs during siphon tube and base (see
below) regeneration, but not during siphon tip regeneration (Table 1),
involves the activity of progenitor cells originating from outside of the
siphon. As OPO replacement is occurring at the distal end of the regenerat-
ing siphon, a blastema appears in the proximal region of the siphon stump
(Figure 3(J) and (K)). The presence of a blastema indicates that long-distance
regeneration occurs by epimorphosis. Blastema formation involves the pro-
liferation and migration of progenitor cells from the branchial sac (see Sec-
tion 7), which are involved in the replacement of muscle and nerves in the
regenerating siphon tube.
During the first few days following amputation, there are no muscle cells
in the regenerating siphon and it is not able to contract. Muscle differenti-
ation begins after blastema formation. The first muscle cells to be replaced
are circular muscle fibers, which begin to differentiate and extend from a
germinal area in the OPO regeneration band. The circular muscle bands
are formed de novo by the differentiation of progenitor cells in the blastema
since the original muscle cells (and any satellite cells associated with them)
are completely removed when the siphon is amputated. In contrast, the lon-
gitudinal muscle bands are replaced by the outgrowth of preexisting
longitudinal muscle fibers that were severed during the amputation process.
As a consequence of new muscle differentiation, the regenerating siphon
eventually regains the ability to contract. Considering the large number of
muscle fibers in the siphon, a significant part of siphon regenerative activity
must be expended on their replacement after injury.
The OS is rich in nerve fibers, which ramify and radiate distally from a
large nerve tract beginning in the NC. When the OS is amputated these
siphon nerve fibers are also severed. As observed in transgenic Ciona with
green fluorescent protein staining throughout the nervous system, new neu-
rons extend distally into the blastema from the severed tips of the preexisting
fibers, eventually contacting the rim of the regenerating siphon (Auger et al.,
2010). The replacement of siphon nerves and OPO is independent of the
NC: siphon nerves invade the blastema and OPOs are replaced after NC
ablation or severing its neural connections, although regeneration occurs
more slowly without the NC (Auger et al., 2010). Replacement of siphon
axons by distal extension is possible because cell bodies are present in the
nerve tracks of the OS (Dahlberg et al., 2009). Presumably nerves are
directly involved the Ciona siphon regeneration, as they are in the regener-
ating vertebrate limb (Brockes, 1987), but this is yet to be demonstrated
experimentally.
3.2 Siphon Base Regeneration

When the OS is severed at its base other parts of the body must be involved
in its replacement, implying a long-distance regeneration process. Accord-
ingly, siphon base regeneration occurs more slowly than siphon tip or
tube regeneration (Table 1). Moreover, OPO replacement does not show
fidelity: multiple OPOs are reformed even after a single cycle of base ampu-
tation, suggesting that patterning mechanisms are changed (Auger et al.,
2010). The ring of tentacles is also removed and subsequently replaced dur-
ing siphon base regeneration. Little is known about tentacle regeneration
except that it involves cell division (Jeffery, 2015c).
During the longer interval required for base regeneration, large numbers
of orange precursor cells differentiate in the OPO regeneration band and form
thick lines along the wound epidermis (Figure 3(I-1,2)). The thick lines of
pigment cells then condense into pigment spots; however, multiple spots
and OPOs develop in the place of a single original OPO (Figure 3(I-3,5)).
For each original OPO, two or three OPOs can be formed (Figure 3(L)
and (M)). Thus, the regenerating siphon of animals subjected to base ampu-
tation can show two (16) or three (24) times the normal number of OPOs.
The multiple OPOs are packed very close along the margin of the siphon
without increasing its girth (Auger et al., 2010).
Interestingly, if animals with multiple OPOs produced by base regener-
ation are amputated a second time, but along the siphon tube rather than the
base, there is a tendency to reproduce the same number of (multiple) OPOs
developed in the previous cycle, rather than the original number of OPOs
(Jeffery, unpublished). This suggests that there is a “memory” of pattern
involved in OPO replacement. The basis for this “memory” of preexisting
pattern is currently unknown, but could involve structural cues in the siphon
tube, possibly the number and/or width of new OPO regeneration bands,
which is under the influence of siphon base amputation.
4. ADULT STEM CELLS

The progenitor cells involved in animal regeneration are produced by
pluripotent stem cells (Weissman, 2000; Voog and Jones, 2010; Sanchez
Alvarado and Yamanaka, 2014). In some cases, these stem cells are located
near the sites of regenerative activities. Examples are the stem cells involved
in replacement of the hair follicles in the epidermis of mammals, stem cells
that replenish cells lining the stomach, and the satellite stem cells that replace
injured vertebrate striated muscle cells. In other cases, stem cells are located
in niches distant from the sites where the progenitor cells derived from them
ultimately function. Examples are the hematopoietic stem cell niches in the
pancreas and the bone marrow of mammals. The Ciona regeneration studies
discussed above suggest that stem cells located both within (short-distance
regeneration) and outside (long-distance regeneration) the siphon may be
involved in the replacement of the OS.
4.1 Multiple Stem Cells

The location of adult stem cells has been addressed in Ciona using the
stem cell markers alkaline phosphatase (AP) and PIWI (Jeffery, 2015c).
Pluripotent stem cells display active AP on their surfaces, which can be
detected using appropriate substrates of this enzyme. AP has been used as
a stem cell marker in a variety of animals (Riekstina et al., 2009), including
ascidians (Akhmanieva et al., 2007). The piwi gene is expressed specifically in
totipotent stem cells (germ cells and gametes) and pluripotent adult stem
cells throughout the animal kingdom (Cox et al., 1998; Seipel et al.,
2004; Palakodeti et al., 2008) and has also been used to detect stem cells
in ascidians (Brown et al., 2009; Rinkevich et al., 2010; Suganaga et al.,

2010; Kawamura and Sunanaga, 2011).
Adult stem cell niches have been detected in four locations in the Ciona
body according to the expression of the AP and PIWI markers (Figure 5(A)
and (C)). First, stem cells are present in groove-like structures in the stomach
and intestinal epithelium (Figure 5(A)). These stem cells replenish the gut
epithelial cells, which undergo continuous turnover during the digestion
process (Ermak, 1975a, 1976a, 1981). Second, stem cells are detected in
the basal stalk or stolon, the structure that attaches the body to the substrate
(Figure 5(A)). The function of basal stalk stem cells is unknown, however,
they may be involved in forming a body growth zone. Third, stem cells
are present in the transverse vessels of the branchial sac where they are
concentrated in lymph nodes (Figure 5(A) and (C)). Lymph nodes contain-
ing stem cells are also localized in the vessels lining the pharynx, the
(A) (B) (C)
(D)
Figure 5 Adult stem cell localization. (A) A young adult showing the distribution of
AP-labeled stem cells. (B) A portion of an oral siphon (OS) showing clusters of AP-
labeled stem cells (arrows). (C) A section through the transverse vessel of the branchial
sac showing PIWI-labeled stem cells (arrows). (D) EdU pulse labeling following OS
amputation shows cell proliferation primarily in the transverse vessels (TV) of the bran-
chial sac (BS). The OS stump is outlined by dashed lines. Insert: higher magnification of
D showing proliferating cells in a lymph node (LN). NC, central ganglion (neural com-
plex); AS, atrial siphon; E, endostyle; R, rectum; I, intestine; S, stomach; H, heart; Sk, basal
stalk; OPO, oral siphon pigment organ. From Jeffery (2015c).
endostyle, and the atrial cavity. Stem cells in lymph nodes are involved in the
renewal of coelomic hemoblast cells (Ermak, 1975b, 1976b), which are
dispersed throughout the animal and have many different functions (De
Leo et al., 1987; Satoh, 1994). In colonial ascidians, coelomic hemoblasts
differentiate into body wall muscle (Sugino et al., 2007). Fourth, small clus-
ters of stem cells are present in the siphon walls, where they are localized
within or near the OPO regeneration bands (Figure 5(B)). As mentioned
above, orange pigment cells of the OPO and circular muscle fibers begin
to differentiate in the OPO regeneration bands. Thus, the Ciona body con-
tains multiple stem cell niches, and those of the branchial sac are especially
prominent (Figure 5(A)).
4.2 Branchial Sac Stem Cells

The branchial sac stem cells respond to siphon amputation by dividing to
produce progenitor cells. The response of branchial sac stem cells to siphon
injury was determined by subjecting Ciona adults to short pulses of EdU after
OS amputation (Jeffery, 2015c). The results showed that cell proliferation
occurred almost exclusively in the transverse vessels of the branchial sac
(Figure 5(D)). Furthermore, double labeling with EdU and the AP stem
cell marker demonstrated that stem cells in the lymph nodes were induced
to proliferate as a response to siphon injury (Figure 5(D) inset). Lower levels
of cell proliferation were also noted in the stomach and intestine during the
short EdU pulses (Figure 5(D)), however, their levels were unchanged after
amputation, suggesting that they represent a constitutive level of cell
renewal not associated with siphon injury. EdU pulse labeling also
confirmed that cell proliferation is not involved in OPO replacement after
the siphon tube is amputated. Thus, the branchial sac appears to be the major
adult stem cell niche involved in long-distance regeneration.
5. STEM AND PROGENITOR CELL MOBILIZATION AND

DEPLOYMENT
The branchial sac contains two different cell types that are deployed in
long-distance OS regeneration: AP/PIWI-stained stem cells and EdU-
labeled progenitor cells. After a delay of a few days, progenitor cells are
mobilized in the branchial sac and migrate into the siphon stump, where
they form a regeneration blastema. This has been demonstrated by EdU
pulse-chase experiments (Jeffery, 2015c). In these experiments, branchial
sac cells labeled with EdU during a short pulse beginning after OS
(A) (B)
(C) (D) (E)
(F)
Figure 6 Stem and progenitor cell contributions to the OS regeneration blastema.

(A, B) The results of a pulse-chase EdU labeling experiment. After a 2-day EdU pulse fol-
lowed by a 4-day chase, animals show strong labeling in the OS after amputation of the
tube (B) but not if there is no amputation (A). (C) AP-labeled stem cells in the OS of a
regenerating animal. Horizontal line indicates the original plane of amputation. (D, E)
Oral siphon pigment organ (OPO) of a regenerating animal showing pigment cells
labeled with alkaline phosphatase (D) but not EdU (E). OS, oral siphon; AS, atrial siphon;
E, endostyle; BS, branchial sac; TV, transverse vessels. (F) Diagram illustrating temporal
differences in appearance of nonproliferating branchial sac stem cells (filled circles) and
proliferating (open circles) branchial sac progenitor cells in the siphon regeneration
blastema. Left. A nonregenerating animal is shown with nonproliferating stem cells
amputation were subsequently detected in the blastema during the chase

(Figure 6(B)). If the same pulse-chase regime is applied, but the siphon is
not amputated, the EdU-labeled cells are not chased into the siphon and
remain in the branchial sac, demonstrating that siphon injury triggers the
mobilization and deployment of branchial sac progenitor cells (Figure 6(A)).
It remains to be determined whether cells from the branchial sac are solely
responsible for blastema formation or if they join other progenitor cells that
are already present in the siphon stump. However, it is likely that the local
contribution of proliferating cells to the blastema is minimal. Two lines of
evidence support the possibility that long-distance migration is responsible
for most, if not all, of the progenitor cells in the regeneration blastema (Jeff-
ery, 2015c). First, dividing cells are not revealed in the siphon (with the
exception of the wound epidermis) by EdU labeling during the first few
days after amputation. Second, differentiation of orange pigment cells and
their incorporation into nascent OPOs occurs without cell proliferation.
Techniques are not yet available to directly follow cell migration from the
branchial sac into the siphon. Accordingly, the evidence for long-distance
stem cell migration has been obtained by branchial sac transplantation exper-
iments (Figure 7) (Jeffery, 2015c). In these experiments, branchial sacs of small
EdU-labeled animals are isolated and transplanted into the branchial sacs of
larger unlabeled animals, and then the distribution of EdU-labeled donor cells
is determined after OS amputation of the unlabeled hosts (Figure 7(A)). As a
response to siphon amputation, EdU-labeled cells in the donor branchial sac
migrate into the host regeneration blastema (Figure 7(C)). This migration is a
specific response to amputation: if the host siphon is not amputated, then
EdU-labeled donor cells do not appear in the siphon. These experiments sug-
gest that progenitor cells originating in the branchial sac migrate distally to
form the regeneration blastema.
Unexpectedly, stem cells of the branchial sac also appear in the regener-
ating siphon (Jeffery, 2015c). In contrast to progenitor cell migration, the
:
localized in the TV of the branchial sac. Middle. A regenerating animal (1e3 h postam-
putation) is shown with nonproliferating stem cells in the distal regeneration blastema.
Right. A regenerating animal (4 þ days postamputation) is shown with branchial sac
stem cells in the transverse vessels and the regeneration blastema and proliferation
of progenitor cells in the transverse vessels. Some of the branchial sac stem cells
have migrated into the regenerating siphon and differentiated into oral siphon
pigment organs (OPOs). Dashed line, site of amputation; open star-shaped structures,
original oral siphon pigment organs; closed star-shaped structures, new oral siphon
pigment organs. From Jeffery (2015c).
Figure 7 Branchial sac transplantation. (A) Diagram showing the transplantation of a

donor branchial sac from a small EdU-labeled animal into the host branchial sac (BS) of
a large unlabeled animal. The horizontal lines indicate the sites of oral siphon (OS) ampu-
tation in the donor prior to BS transplantation and the host after transplantation. T, trans-
plant. (B) A chimeric host animal containing the (red (dark gray in print versions)-stained)
EdU-labeled donor branchial sac. (C) EdU labeling (circle) in the regenerating OS of a host
animal containing a transplanted donor branchial sac. CNS, neural complex. From Jeffery
(2015c).
stem cells migrate into the siphon early after amputation, particularly when
siphon amputation occurs at the base (Figure 6(F)). After base regeneration,
branchial sac stem cells, which continue to be detectable by AP labeling,
invade the regenerating siphon and differentiate directly into orange
pigment cells of the OPOs without prior cell division (Figure 6(CeF)).
The branchial sac stem cells can be detected in the new OPOs because
they still express stem cell markers for a short time after their differentiation
into orange pigment cells. This is in contrast to the situation for siphon tip
and tube regeneration, in which small clusters of stem cells located in the
OPO regeneration bands (Figure 5(B)), and not the branchial sac stem cells,
give rise to OPO components (Table 1).
In summary, two sources of stem cells are involved in OS regeneration
(Figure 6(F)). When the siphon is amputated at its tip, a source within the
siphon is mobilized and deployed for short-distance regeneration. When
the siphon is amputated along its tube, both the local and long-distance
(branchial sac-based) sources are involved in regeneration: OPOs are
derived from the local source and the blastema is derived from the long-
distance source. When the siphon is amputated at its base only the long-
distance source is involved in regeneration: OPOs are directly formed
from branchial sac stem cells and the blastema is derived from progenitor
cells that also originate in the branchial sac. The local clusters of stem cells
normally present in the OPO regeneration bands are likely to be replen-
ished by the invasion of branchial sac stem cells during long-distance
regeneration.
6. AGING AND OS REGENERATION

The regenerative abilities of Ciona decline with age. Regenerative aging
of the NC and OS has been observed in both natural marine environments
and laboratory cultures (Dahlberg et al., 2009; Auger et al., 2010; Jeffery,
2012, 2015c). In wild collected adults, regeneration capacity during the life
cycle was evaluated using size (e.g., distal to proximal length) as a proxy
for age. Under favorable conditions, wild Ciona adults grow continuously
during a life span of about 1e1.5 years (Berrill, 1947; Millar, 1952; Dybern,
1965; Peterson et al., 1995). NC ablation or OS amputation of different sized
animals resulted in the discovery of an inverse relationship between body
length (age) and the rate of regeneration (Dahlberg et al., 2009; Auger
et al., 2010): larger (older) animals regenerate much more slowly than smaller
(younger) animals (Figure 8(A)). A similar decline in regeneration rate as a
function of age also occurs in laboratory cultures in which animals of precisely
known age are grown from fertilized eggs (Jeffery, 2015c).
Despite a decrease in the rate of regeneration during the adult life cycle,
the regenerative processes involved, for example OPO replacement, are still
completed with accuracy. This situation changes as animals reach an age
threshold, when the capacity for regeneration disappears (Jeffery, 2012).
OS regeneration is compromised in the oldest animals collected in the
wild or cultured in the laboratory (Jeffery, 2012, 2015c). The old animals
show morphological and reproductive abnormalities relative to their
younger counterparts. They exhibit a thickened and withered tunic, a
reduction in gamete production, and have malformed and larger OPOs.
Moreover, the pharynx and siphons of these animals appear to be inflamed
due to the overproduction of orange pigment cells, a condition that is
possibly related to stress (Parrinello et al., 2010). When the siphons of old
animals are amputated at any position, the rate of regeneration is either
very slow or it does not occur at all, even after a month or more of obser-
vation (Figure 8(B)).
Even though there is no net siphon growth after amputation in old an-
imals, OPO replacement still occurs, although it shows several differences
from normal replacement in younger animals (Jeffery, 2012). First, when
the oral siphons of old animals are amputated in the tube, OPO replacement
occurs, but multiple OPOs are formed in the place of the original single
OPO. Thus, siphon tube regeneration in old animals resembles siphon
base regeneration in younger animals. This difference could be due to the
(A) (B)
(C)
(D)
(E) (F)
Figure 8 Aging and oral siphon (OS) regeneration. (A) The time required for oral siphon
pigment organ (OPO) replacement increases during the life cycle. A: From Auger et al.
(2010). (B) Lack of OS growth in an old animal subjected to two consecutive cycles of
amputation. (C) Oral siphon pigment organ (OPO) replacement has been arrested at
an intermediate stage as a line of mixed orange (black in print versions) and yellow
(white in print versions) pigment cells (arrows). (D) Cell proliferation determined by
phosphohistone H3 antibody staining in the regeneration blastema of middle-aged an-
imals and in the siphon stump of old animals. The number of proliferating cells is signif-
icantly higher in the region above the siphon amputation plane (AeA) compared to the
region below it in middle-aged animals but not in old animals. Asterisks represent sig-
nificant differences. N is shown below each bar. BeD: From Jeffery (2012). (EF) Lymph
node cells stained with PIWI stem cell marker in transverse vessels (TV) in the branchial
sac of middle age (E) but not old (F) animals. E, F. From Jeffery (2015c).
abundance of differentiated orange pigment cells both in the siphons and

elsewhere in the body of old animals. Second, OPO replacement is arrested
when old animals are subjected to an additional cycle of siphon amputa-
tion. After two consecutive cycles of amputation, large numbers of orange
pigment cells differentiate in the siphon stump, and the pigment cell masses
line up along the wound epidermis, which seems to be formed normally,
but the orange pigment cells do not condense into distinct pigment
spots or form OPOs. Instead, they arrest in lines along the siphon rim
and later mix with similar lines of yellow pigment responsible for inter-
OPO pigment band formation (Figure 8(B) and (C)). Lastly, when the
amputated siphon stumps of old animals are excised and cultured as ex-
plants in vitro, the OPO regeneration bands overproduce orange pigment
cells, but they do not move distally to form OPOs, as occurs in siphon
explants derived from young animals (Figure 4(I)). Therefore, although or-
ange pigment cells still differentiate, and in fact are overproduced, OPO
replacement is defective, suggesting that aging disrupts morphogenetic
processes in old animals.
The formation of a blastema of proliferating cells plays an important role
in OS regeneration, contributing the precursors of new muscle and nerve
cells during siphon growth (Auger et al., 2010). As described above, the blas-
tema is formed by stem cells that migrate from the branchial sac early during
regeneration and reform the OPO without undergoing cell division and
progenitor cells that proliferate in the branchial sac and migrate into the blas-
tema later during regeneration (Jeffery, 2015c). Although OPO replacement
can occur (albeit defectively) in old animals, they do not develop a blastema
of proliferating cells (Figure 8(D)) (Jeffery, 2012). Accordingly, old animals
do not replace muscle cells and regain the ability for siphon contraction.
There is no information available about the effects of age on nerve cell
replacement.
The absence of a blastema in old animals suggests that defects might
occur in the production or migration of proliferating cells. Branchial sac
stem cells have been compared in young and old animals following OS
amputation by EdU pulse labeling and expression of AP and PIWI stem
cell markers (Jeffery, 2015c). The structure of the branchial sac and distribu-
tion of proliferating cells appear to be abnormal in old animals (Millar, 1952;
Jeffery, 2015c). Furthermore, the branchial sac of old animals shows much
lower levels of AP- and PIWI-labeled adult stem cells compared to their
younger counterparts. Thus, a reduction in stem cell number, the failure
of stem cells to produce progenitor cells, and/or the inability of progenitor
cells to migrate into the injured siphons may be responsible for regenerative
aging. It is also possible that the injured siphon stumps of old animals do not
produce a normal signal for the attraction of progenitor cells to the regen-
eration blastema.
Based on all the information currently available about siphon regenera-
tion, stem cells, and aging, a model has been suggested to explain the gradual
decline and eventual abrupt cessation of regenerative capacity of the Ciona
OS (Jeffery, 2015c). The model states (1) that all of the branchial sac stem
cells that are used in growth (and potentially in regeneration) are produced
early during the adult life cycle, (2) after a maximal number of stem cells is
reached, they slowly lose potency or disappear as animals age, and (3) during
old age, no additional stem cells are available for further growth or, if neces-
sary, regeneration. A test of this model would entail the experimental reduc-
tion and elevation of stem cell number, and approaches to accomplish this
are currently under development.
7. CONCLUDING REMARKS AND PERSPECTIVES

We suggest that new animal models are needed to study the relation-
ship between regeneration and aging. As well as being able to reveal the
principles underlying regenerative aging, these models should also help us
understand how and why regeneration has been lost during the evolution
of some animals, including ourselves. Ultimately, using information gleaned
from these models, we might find ways to reverse the loss of regenerative
potential that is imposed during aging and evolution. Ciona is an animal
that fills the need for a model in regenerative aging.
Although many avenues of regenerative biology research need to be
explored further in Ciona, in recent years there has been progress in deci-
phering the underpinnings of regeneration. In Ciona, it is possible to study
regeneration in a small part of the animal while also observing related
changes in the entire animal, thus allowing both short- and long-distance
regeneration processes to be analyzed. This approach led to the discovery
of the Ciona branchial sac stem cell niche and its long-distance contribution
to the regenerating siphon (Jeffery, 2015c).
Most of the insights described in this article are derived from experiments
on OS regeneration, however, Ciona also promises to be an excellent model
to study the regeneration capacity of other organs, such as the brain and
heart, and these studies should be vigorously pursued in the future. Although
some vertebrates, namely amphibians, show partial brain regeneration (Endo

et al., 2007), no chordate group other than the tunicates can regenerate an
entire brain after its complete removal. Thus, insights about vertebrate brain
regeneration may be gained from studying the capacity for complete brain
regeneration in Ciona.
We have described recent results showing that adult stem cells are instru-
mental in Ciona regeneration and aging. Similar properties of stem cells have
been observed in other animals (e.g., Conboy and Rando, 2005; Sharpless
and Schatten, 2009; Waterstrat and Van Zant, 2009). It is remarkable that
all stages in the life and function of adult stem cells, from their initial forma-
tion during metamorphosis (or perhaps even during embryogenesis) to their
possible depletion or decay during old age, are accessible for study in Ciona.
This attribute is likely to have important impacts on understanding the life
history of adult stem cells.
Molecular analysis of Ciona regeneration is still being developed. This
approach will be fostered by the existence of molecular and genomic tools
that have been pioneered for studying the development of this organism
(Stolfi and Christiaen, 2012). Thus, it will be imperative to take advantage
of these tools in future studies to reveal the molecular basis of regenerative
aging.
ACKNOWLEDGMENTS
This article was prepared under the auspices of a grant from the National Institutes of Heath
(AG037918) and Frederick and Betsy Bang and Laura and Arthur Colwin Fellowships from
the Marine Biological Laboratory, Woods Hole, MA.
REFERENCES
Akhmadieva, A.V., Shukalyuk, A.I., Aleksandrova, Y.N., Isaeva, V.V., 2007. Stem cells in
asexual reproduction of the colonial ascidian Botryllus tubaratus (Tunica: Ascidiacea).
Russ. J. Mar. Biol. 33, 181e186.
Auger, H., Sasakura, Y., Joly, J.-S., Jeffery, W.R., 2010. Regeneration of oral siphon
pigment organs in the ascidian Ciona intestinalis. Dev. Biol. 339, 374e389.
Bely, A.E., 2010. Evolutionary loss of animal regeneration: pattern and process. Integr.
Comp. Biol. 50, 515e527.
Berrill, N.J., 1947. The development and growth of Ciona. J. Mar. Biol. Assoc. U.K. 26,
616e625.
Berrill, N.J., 1951. Regeneration and budding in tunicates. Biol. Rev. 26, 456e475.
Bollner, T., Beesley, P.W., Thorndike, M., 1992. Pattern of substance P- and cholecysto-
kinin-like immunoreactivity during regeneration of the neural complex in the ascidian
Ciona intestinalis. J. Comp. Neurol. 325, 572e580.
Bollner, T., Beesley, P.W., Thorndike, M.C., 1993. Distribution of GABA-like immunore-
activity during postmetamorphic development and regeneration of the nervous system in
the ascidian Ciona intestinalis. Cell Tissue Res. 272, 553e561.
Bollner, T., Howalt, S., Thorndyke, M.C., Beesley, P.W., 1995. Regeneration and post-
metamorphic development of the central nervous system in the protochordate Ciona
intestinalis: a study with monoclonal antibodies. Cell Tissue Res. 279, 421e432.
Bollner, T., Beesley, P.W., Thorndike, M.C., 1997. Investigation of the contribution from
peripheral GnRH-like immunoreactive ‘neuroblasts’ to the regenerating central nervous
system in the protochordate Ciona intestinalis. Proc. R. Soc. Lond. B 264, 1117e1123.
Bourlat, S.J., Juliusdottir, T., Lowe, C.J., Freeman, R., Aronowicz, J., Kirschner, M.,
Lander, E.S., Thorndyke, M., Nakano, M., Kohn, A.B., Heyland, A., Moroz, L.L.,
Copley, R.R., Telford, M.J., 2006. Deuterostome phylogeny reveals monophyletic
chordates and the new phylum Xenoturbellida. Nature 444, 85e88.
Brien, P., 1968. Blastogenesis and morphogenesis. Adv. Morphog. 7, 151e203.
Brockes, J.P., 1987. The nerve dependence of amphibian limb regeneration. J. Exp. Biol.
132, 79e91.
Brockes, J.P., Kumar, A., 2008. Comparative aspects of animal regeneration. Annu. Rev.
Cell Dev. Biol. 24, 525e549.
Brown, F.D., Swalla, B.J., 2012. Evolution and development of budding by stem cells:
ascidian coloniality as a case study. Dev. Biol. 369, 151e162.
Brown, F.D., Keeling, E.L., Le, A.D., Swalla, B.J., 2009. Whole body regeneration in a colo-
nial ascidian, Botrylloides violaceus. J. Exp. Zool. Mol. Dev. Evol. 312B, 885e900.
Cirino, P.A., Toscano, D., Caramiello, A., Macina, A., Miraglia, V., Monte, A., 2002.
Laboratory culture of the ascidian Ciona intestinalis (L.): “a model system for molecular
developmental biology research.” Mar. Mod. Elec. Rec. http://www.mbl.edu/
BiologicalBulletin/MMER/cirino/CirTit.html.
Conboy, I.M., Rando, T.A., 2005. Aging, stem cells, and tissue regeneration. Lessons from
muscle. Cell Cycle 4, 407e410.
Cox, D.N., Chao, A., Baker, J., Chang, L., Qiao, D., Lin, H., 1998. A novel class of evolu-
tionarily conserved genes defined by piwi are essential for stem cell self-renewal. Genes
Dev. 12, 3715e3727.
Dahlberg, C., Auger, H., Dupont, S., Sasakura, Y., Thorndyke, M., Joly, J.-S., 2009.
Refining the model system of central nervous system regeneration in Ciona intestinalis.
PLoS One 4, e4458.
Delsuc, F., Brinkmann, H., Chourrout, D., Phillipe, H., 2006. Tunicates and not cephalo-
chordates are the closest living relatives of vertebrates. Nature 439, 965e968.
De Leo, G., Parrinello, N., Di Bella, M.A., 1987. Fine structure of blood system in Ciona
intestinalis (Tunicata). Vessel and hemocytes in pharyngeal wall. Arch. Biol. 98, 35e52.
Dilly, P.N., Wolken, J.J., 1973. Studies on the receptors of Ciona intestinalis: IV. The ocellus
in the adult. Micron 4, 11e29.
Dybern, B.I., 1965. The life cycle of Ciona intestinalis (L.) f. typical; in relation to environ-
mental temperature. Oikos 16, 109e131.
Endo, T., Yoshino, J., Kado, K., Tochinai, S., 2007. Brain regeneration in anuran
amphibians. Dev. Growth Differ. 49, 121e129.
Ermak, T.H., 1975a. Cell proliferation in the digestive tract of Styela clava (Urochordata): as
revealed by autoradiography with tritiated thymidine. J. Exp. Zool. 194, 449e466.
Ermak, T.H., 1975b. An autoradiographic demonstration of blood cell renewal in Styela clava
(Urochordata: Ascidiacea). Experientia 31, 837e838.
Ermak, T.H., 1976a. Cell migration kinetics in the stomach of Styela clava (Urochordata;
Ascidiacea). J. Exp. Zool. 197, 339e346.
Ermak, T.H., 1976b. The hematogenic tissues of tunicates. In: Wright, R.K., Cooper, E.L.
(Eds.), Phylogeny of Thymus and Bone Marrowebursa Cells. Elsevier/North Holland,
Amsterdam, The Netherlands, pp. 45e56.
Ermak, T.H., 1981. A comparison of cell proliferation patterns in the digestive tract of
ascidians. J. Exp. Zool. 217, 325e339.
Freeman, G., 1964. The role of blood cells in the process of asexual reproduction in the tuni-
cate Perophora. J. Exp. Zool. 156, 157e184.
Givan, J.E., Olson, W.N., Hall, B.K., 2002. Hind-limb regeneration in the dwarf African
clawed frog, Hymenochirus boettgeri (Anura: Pipidae). J. Herpetol. 36, 537e543.
Hirschler, J., 1914. Uber die Restitutions-und Involutionsvorlange bei operierten Exempla-
ren von Ciona intestinalis Flem. (Teil I) nebst Bemurkungen uber den Wert des Nega-
tiven fur das Potenzproblem. Arch. Mikr. Anat. 85, 205e227.
Illingworth, C.M., Barker, A.T., 1980. Measurement of electrical currents emerging
during the regeneration of amputated finger tips in children. Clin. Phys. Physiol.
Meas. 1, 87e89.
Jeffery, W.R., 2001. Determinants of cell and positional fate in ascidian embryos. Int. Rev.
Cytol. 203, 3e62.
Jeffery, W.R., 2012. Siphon regeneration capacity is compromised during aging in the
ascidian Ciona intestinalis. Mech. Ageing Dev. 133, 629e636.
Jeffery, W.R., 2015a. The tunicate Ciona: a model system for understanding the relationship
between regeneration and aging. Invertbr. Reprod. Dev. 59, 17e22.
Jeffery, W.R., 2015b. Closing the wounds. One hundred and twenty five years of regener-
ative biology in the ascidian Ciona intestinalis. Genesis 53, 48e65.
Jeffery, W.R., 2015c. Distal regeneration involves the age dependent activity of branchial sac
stem cells in the ascidian Ciona intestinalis. Regeneration 1, 1e18.
Joly, J.-S., Kano, S., Matsuoka, T., Auger, H., Hirayama, K., Satoh, N., Awazu, S.,
Legendre, L., Sasakura, Y., 2007. Culture of Ciona intestinalis in closed systems. Dev.
Dyn. 236, 1832e1840.
Kawamura, K., Sunanaga, T., 2011. Role of Vasa, Piwi, and Myc-expressing coelomic cells in
gonad regeneration of the colonial tunicate, Botryllus primigenus. Mech. Dev. 128, 457e470.
Kourakis, M.J., Smith, W.C., 2015. An organismal perspective on C. intestinalis development,
origins and diversification. eLife 4, e06024. http://dx.doi.org/10.7554/eLife.06024.
Loeb, J., 1892. Unter such ungen zur physiologischen morphologie. II. Wurzburg.
Mackie, G.O., Burighel, P., Caicci, F., Manni, L., 2006. Innervation of ascidian siphons and
their responses to stimulation. Can. J. Zool. 84, 1146e1162.
Markman, B., 1958. On the peripheral nervous system of ascidians. Acta Zool. 34, 13e18.
Millar, R.H., 1952. The annual growth and reproductive cycle in four ascidians. J. Mar. Biol.
Assoc. U.K. 31, 41e61.
Millar, R.H., 1953. Ciona. In: Colman, J.S. (Ed.), LMBC Memoirs on Typical British
Marine Plants and Animals. Liverpool University, Liverpool, U.K, pp. 1e123.
Mingazzini, P., 1891. Sulla rigenerazione nei Tunicati. Boll. Dell. Soc. Natur. Napoli 5, 76e79.
Mizell, M., 1968. Limb regeneration: Induction in the newborn opossum. Science 161,
283e286.
Ogasawara, M., Satoh, N., 1998. Isolation and characterization of endostyle-specific genes in
the ascidian Ciona intestinalis. Biol. Bull. 195, 60e69.
Palakodeti, D., Smielewska, M., Lu, Y.C., Yeo, G.W., Graveley, B.R., 2008. The PIWI
proteins SMEDWI-2 and SMEDWI-3 are required for stem cell function and piRNA
expression in planarians. RNA 14, 1174e1186.
Parrinello, N., Vizzini, A., Salerno, G., Sanfratello, M.A., Cammarata, M., Arizza, V.,
Vazzana, M., Parrinello, D., 2010. Inflamed adult pharynx tissues and swimming larva
of Ciona intestinalis share CiTNFalpha-producing cells. Cell Tissue Res. 341, 299e311.
Peterson, J.K., Chou, O., Thor, P., 1995. Growth and energetics in the ascidian Ciona
intestinalis. Mar. Ecol. Prog. Ser. 120, 175e184.
Poss, K.D., 2010. Advances in understanding tissue regenerative capacity and mechanisms in
animals. Nat. Rev. Genet. 11, 710e722.
Reed, M.J., Koike, T., Puolakkainenen, P., 2003. Wound repair in aging. Methods Mol.
Med. 78, 217e237.
Riekstina, U., Cakstina, I., Parfejevs, V., Hoogdujn, M., Jankovskis, G., Mulznieks, I.,
Muceniece, R., Ancans, J., 2009. Embryonic stem cell marker expression pattern in
human mesenchymal stem cells derived from bone marrow, adipose tissue, heart and
dermis. Stem Cell Rev. 5, 378e786.
Rinkevich, B., Shlemberg, Z., Fishelson, L., 1995. Whole body protochordate regeneration
from totipotent blood cells. Proc. Natl. Acad. Sci. U.S.A. 92, 7695e7699.
Rinkevich, Y., Rosner, A., Rabinowitz, C., Lapidot, Z., Moiseeva, E., Rinkevich, B., 2010.
Piwi positive cells that line the vasculature epithelium underlie whole body regeneration
in a basal chordate. Dev. Biol. 345, 94e104.
Sanchez Alvarado, A., Yamanaka, S., 2014. Rethinking differentiation: stem cells, regener-
ation, and plasticity. Cell 157, 110e119.
Satoh, N., 1994. Developmental Biology of Ascidians. Cambridge University Press, Cam-
bridge U.K, pp 1e234.
Satoh, N., 2013. Developmental Genomics of Ascidians. Wiley-Blackwell, Hoboken, NJ,
pp. 1e216.
Schultze, L.S., 1899. Die Regeneration des Ganglions von Ciona intestinalis L und uber das
Verhaltneiss der Regeneration und Knospung zur Keimenblatterlehre. Jena Z. F. Natur-
wiss. 33, 263e344.
Seifert, A.W., Voss, S.R., 2013. Revisiting the relationship between regenerative ability and
aging. BMC Biol. 11 (2) http://dx.doi.org/10.1186/1741-7007-11-2.
Seipel, K., Yanze, N., Schmid, V., 2004. The germ cell line and somatic stem cell gene Cniwi
in the jellyfish Podocoryne carnea. Int. J. Dev. Biol. 48, 1e7.
Sharpless, N.E., Schatten, G., 2009. Stem cell aging. J. Gerontol. A. Biol. Sci. Med. Sci. 64A,
202e204.
Sousounis, K., Badour, J.A., Tsonis, P.A., 2014. Aging and regeneration in vertebrates. Curr.
Top. Dev. Biol. 108, 217e246.
Stolfi, A., Christiaen, L., 2012. Genetic and genomic toolkit of the chordate Ciona intestinalis.
Genetics 192, 55e66.
Sugino, Y., Matsumura, M., Kawamura, K., 2007. Body muscle-cell differentiation from
coelomic stem cells in colonial tunicates. Zool. Sci. 24, 542e546.
Sunanaga, T., Inubushi, H., Kawamura, K., 2010. Piwi-expressing hemoblasts serve as germ-
line stem cells during postembryonic germ cell specification in colonial ascidian, Botryllus
primigenus. Dev. Growth Differ. 52, 603e614.
Sutton, M.F., 1953. The regeneration of the siphons of Ciona intestinalis L. J. Mar. Biol.
Assoc. U.K. 32, 249e286.
Takamura, K., Fujimura, M., Yamaguchi, Y., 2002. Primordial germ cells originate from the
endodermal strand cells in the ascidian Ciona intestinalis. Dev. Genes Evol. 212, 11e18.
Tiozzo, S., Brown, F.D., De Tomaso, A.W., 2008. Regeneration and stem cells in ascidians.
In: Bosch, T.C.G. (Ed.), Stem Cells: From Hydra to Man, Chapter 6. Springer, New
York, pp. 95e112.
Voog, J., Jones, D.L., 2010. Stem cells and the niche: a dynamic duo. Cell Stem Cell 6, 103e115.
Waterstrat, A., Van Zant, G., 2009. Effects of aging on hematopoietic stem and progenitor
cells. Curr. Opin. Immunol. 21, 408e413.
Weissman, I.L., 2000. Stem cells: units of development, units of regeneration, and units in
evolution. Cell 100, 157e168.
Whittaker, J.R., 1975. Siphon regeneration in Ciona. Nature 255, 224e225.
INDEX
Note: Page numbers with “f ” and “t” denote figures and tables, respectively.
A Calu-3 cells, 13–14

Acanthamoeba castellanii, 9–10 cAMP-dependent protein kinase, 232
Acinetobacter baumannii, 23–24 Cancer, 241–242
Adult stem cells, 269–271 Candida albicans, 11
branchial sac stem cells, 270f, 271 CD63 (Tspan30), 149–150
multiple stem cells, 269–271, 270f CD151 (Tspan24), 148–149
Aging, 241 Cell-based therapies, 53–58
Akt, 231–232 embryonic stem cell (ESC), 54
Aldehyde dehydrogenase (ALDH), induced pluripotent stem cell (iPSC),
156–157 54–55
Alkaline phosphatase (AP), 269–270 mesenchymal stem cells.
AlkB protein demethylates, 174–175 See Mesenchymal stem cell (MSC)
Allogeneic transplantation, 80–81 Cell culture models, 3f
Amine oxidase (LSD1/KDM1), 167–170, key points, 2–3
172f organ equivalents and tissue explants,
Amniotic membrane transplantation 21–29
(AMT), 73 complex cell systems “En Miniature”,
Antioxidant systems, 226 22–24
ARX1, 119 piece of reality, 24–26
Atherosclerosis, 236–237 2D and 3D cell culture, microfluidic
ATM protein kinase, 234 systems in, 26–29
three-dimensional (3D) cell culture,
B 14–21
BBB. See Blood-brain barrier (BBB) air–liquid surface, requirements of,
Biological materials 18–19
collagen-based materials, 61–62 BBB, coculture-based reconstruction
fibrin, 60–61 of, 17–18
silk fibroin-based materials, 62–63 benefits and limitations, 15–19
Blastema, formation of, 277 microgravity-variations of, 19–21
Blood-brain barrier (BBB), 11 two-dimensional (2D) cell culture, 4–14
coculture-based reconstruction of, 17–18 alternative infection models, protozoa
complexity of, 12 as, 9–10
Blood-cerebrospinal fluid barrier bacterial pathogens, 4, 5t–6t
(BCSFB), 11 coculture infection models.
Bombyx mori (BM), 62–63 See Coculture infection models
Brain microvascular endothelial cells fluorescence staining, 7
(BMECs), 11 immortalized cell lines vs. primary cell
in vitro models, 12 culture, 7–9
Branchial sac stem cells, 270f, 271 Cellular organoids, 21–22
Cellular spheroids, 21–22
C cGMP-dependent protein kinase, 235
Ca2+/calmodulin-dependent protein Chlamydia trachomatis, 19–21
kinase II (CaMKII), 234 Chromatography, 176–177
283 j
284 Index
Ciona intestinalis D
adult stem cells, 269–271 Dictyostelium discoideum, 9–10
branchial sac stem cells, 270f, 271 Double-stranded b helix (DSBH) fold, 166
multiple stem cells, 269–271, 270f Drosophila melanogaster, 177
life cycle/adult organization and growth, Dual TUDOR domains, 190
257–260, 258f
oral siphon (OS) E
aging, 275–278, 276f ECM1, 119
model, 260–262, 261f Embryonic stem cell (ESC), 46–47, 54
regeneration. See Oral siphon (OS), Endostyle, 259
regeneration Engineering limbal niche, 67–68, 69t–72t
partial body regeneration, 260 Enterohemorrhagic Escherichia coli
stem/progenitor cell mobilization and (EHEC), 22–23
deployment, 271–274, Enteropathogenic Escherichia coli (EPEC),
272f–273f 22–23
Circular debridement wounds, 77–78 EpiAirwayTM. See Epithelial airway tissue
Circular muscle bands, 267–268 model (EpiAirwayTM)
Citrobacter freundii, 11 Epithelial airway tissue model
CLET. See Cultivated limbal epithelial (EpiAirwayTM), 18–19
transplantation (CLET) Epithelial cells culture, 15–17
Coculture infection models, 10–14 Epithelial–mesenchymal transition (EMT),
suspension, adherent cells and 142–150, 149f, 153–154, 156–157
neutrophiles in, 13–14 liver fibrosis. See Liver fibrosis
tissue barriers, coculture-based generation zebrafish, muscle cells in, 145–146
of, 11–13 Ethylenediamine tetra-acetic acid
Collagen, 15–17 (EDTA), 81–82
COMET. See Cultivated oral mucosal Eukaryotic ribosome
epithelial transplantation biogenesis, experimental approaches to,
(COMET) 109–110
Conjunctival basal cells, 49–50 cytoplasmic maturation, 124–130
Conjunctival epithelium, 49–50 pre-40S subunits, 127–130
Corneal cells, 57 pre-60S subunits, 124–127, 124f
Corneal epithelium, 48 export, 114–124
Corneal wound healing, 77–80, 79t pre-40S subunits, 121–124
Cross-linking collagen, 61–62 pre-60S subunits, 117–121, 118f
Cultivated limbal epithelial transplantation shared export factors, 115–117
(CLET), 82f nuclear maturation, 111–114
results, 83–84, 83f, 85t–87t pre-40S subunits, 113–114
surgical steps, 82 pre-60S subunits, 111–113
Cultivated oral mucosal epithelial 90S preribosome assembly, 110–111, 110f
transplantation (COMET) ribosomal RNAs (rRNAs), 108
results, 90, 90t synthesis, 108
surgical technique, 89–90 Extracellular matrix, 151
Cysteine residues, 222–224, 223f
Cytoplasmic maturation, 124–130 F
pre-40S subunits, 127–130 Factor-inhibiting HIF-1 alpha, 166, 167f
pre-60S subunits, 124–127, 124f Flow cytometry analysis, 7
Index 285
Fluorescence staining, 7 I
Forkhead boxO (FOXO) transcription induced pluripotent stem cell (iPSC),
factors, 235 30–31, 54–55
Four-point-one, ezrin, radixin and moesin Inflammation, 237–238
(FERM), 151 Inhibitory kB (IkB), 234
Fractionation, 176–177 Intracellular loop (ICL), 151
Francisella tularensis, 19–21 iPSC. See induced pluripotent stem cell
(iPSC)
G
Gle2-binding sequence (GLEBS), 120 J
Glycocalyx forms, 49 JARID subgroup, 193–199
Golgi complex, 15–17 JARID2-Jumonji, 199
KDM5A, 193–195, 194f
H KDM5B, 195–196, 197f
Hartmannella vermiformis, 9–10 KDM5C, 196–199
Henle’s crypt, 49–50 JHDM2 (JMJD1) subgroup, 182–187
Hepatocyte growth factor (HGF), 145 Jumonji C (JmjC) domain-containing
Herpes simplex virus (HSV), 80 proteins, 208–210
Histone demethylase subgroups, 176–210 double-stranded b helix (DSBH) fold,
JARID subgroup, 193–199 166
JHDM2 (JMJD1) subgroup, 182–187 factor-inhibiting HIF-1 alpha,
JmjC-domain-only, 208–210 166, 167f
Jumonji histone demethylase 1 (JHDM1) histone demethylases, 174–210, 175f
subgroup, 176–182, 177f demethylase subgroups, 176–210
Jumonji histone demethylase 2 (JHDM2) histone demethylation and demethylases
subgroup, 182–187 peptidylarginine deiminase 4 (PADI4/
Jumonji histone demethylase 3 (JHDM3) PAD4) and amine oxidase (LSD1/
subgroup, 187–193 KDM1), 167–170, 172f
PHF2/PHF8/KIAA1718 subgroup, histone modification and methylation,
203–208 167–171, 168t–169t
UTX/UTY/JMJD3 subgroup, 199–203 arginine residues, 170, 171f
H3K9 methylation mark, 183–185 JMJD6, 208–209
human amniotic membrane (hAM), 58–60 KDM8, 209–210
Human embryonic stem cell (hESC), plant histone demethylation
30–31 nondemethylating roles, 210–211
Human epidermal equivalent (HEE), 30–31 plant histone demethylases,
Human limbal epithelial cells, 68 211, 212f
Human malignant choroid plexus Jumonji histone demethylase 1 (JHDM1)
papilloma cells (HIBCPP), 13–14 subgroup, 176–182, 177f
Human Rio2 (hRio2), 122 Epe1, 182
Human umbilical vein endothelial cells Jhd1, 181
(HUVEC), 12–13 KDM2A/KDM2B, 178–181, 179f
HUVEC. See Human umbilical vein Jumonji histone demethylase 2 (JHDM2)
endothelial cells (HUVEC) subgroup, 182–187
Hydrophilic siloxane hydrogel contact KDM3A, 183–185, 184f
lenses, 63–64 KDM3B, 185–186
Hypertension, 239–240 KDM3C, 186–187
286 Index
Jumonji histone demethylase 3 (JHDM3) MIDAS interacting domains (MIDOs),

subgroup, 187–193 112–113
KDM4A, 189–191, 190f Motile nonfeeding larva, 257–258
KDM4B, 191 MSC. See Mesenchymal stem cell (MSC)
KDM4C, 191–192 Mucins, 49
yeast KDM4 members, 192–193 Mucus-secreting goblet cells, 49–50
Multiple stem cells, 269–271, 270f
K
Kap123, 111 N
Neisseria meningitides, 13–14
L Neural complex (NC), 259
Legionella pneumophila, 4–7, 24–25 Neural ganglion, 264
Leptomycin B (LMB), 115 Neurodegenerative diseases, 238
Limbal/corneal epithelial cells Neuroinflammation, 238
chemical and mechanical injury of, N-glycosylation, 156–157
74–75, 75f, 76t, 77f Nmd3, 118–119
genetic defects of, 73–74 Nonkeratinized stratified limbal
Limbal stem cell deficiency (LSCD), epithelium, 50–52
50–52, 56–58, 57f–58f, 73–80 Nonreceptor kinases
corneal wound healing, 77–80, 79t Akt, 231–232
limbal/corneal epithelial cells, chemical ATM protein kinase, 234
and mechanical injury of, 74–75, Ca2+/calmodulin-dependent protein
75f, 76t, 77f kinase II (CaMKII), 234
limbal/corneal epithelial cells, genetic cAMP-dependent protein kinase, 232
defects of, 73–74 cGMP-dependent protein kinase, 235
Limbus, 50–53, 51f–53f inhibitory kB (IkB), 234
Listeria monocytogenes, 11 MAPK, 233–234
Liver fibrosis Src family kinases, 232–233
cross-talk between tetraspanins, 148–150, Non-small cell lung cancer (NSCLC),
149f 153–154
TGFb1 signaling, 147–148 Nuclear export signal (NES), 115
Longitudinal muscle bands, 267–268 Nuclear factor-Like 2 (Nrf2), 235–236
LSCD. See Limbal stem cell deficiency Nuclear maturation, 111–114
(LSCD) pre-40S subunits, 113–114
Lymph nodes, 270–271, 270f pre-60S subunits, 111–113
Nuclear pore complex (NPC), 114–115
M Nucleolar ribosome assembly, 111
Manual superficial keratectomy (MSK),
77–78 O
Melanoma cells culture, 15–17 Obesity, 240–241
Mesenchymal stem cell (MSC) Ocular mucus, 49
corneal reconstruction, 56 Ocular surface reconstruction
LSCD, 56–58, 57f–58f anatomy and pathology, 47
Messenger ribonucleic acid (mRNA), conjunctival epithelium, 49–50
146 limbus, 50–53, 51f–53f
Mex67-Mtr2 heterodimer, 115–116 preocular tear film, 49
Microfluidic cell culture, 26–28 cell-based therapies for, 53–58
Index 287
embryonic stem cell (ESC), 54 short-distance regeneration, 264t,

induced pluripotent stem cell (iPSC), 265–267, 266f
54–55 siphon base, 268–269
mesenchymal stem cells. tip and tube, 264–268
See Mesenchymal stem cell (MSC) structure, 258f
embryonic stem cell (ESC), 46–47 Oral siphon pigment organ (OPO),
limbal stem cells (LSCs), 46–47 260–263, 261f, 263f
LSCD, animal models for, 73–80 blastema and siphon nerves, 268
corneal wound healing, 77–80, 79t multiple, 268–269
limbal/corneal epithelial cells, chemical regeneration band, 264, 270–271
and mechanical injury of, 74–75, OS. See Oral siphon (OS)
75f, 76t, 77f
limbal/corneal epithelial cells, genetic P
defects of, 73–74 Palpebral conjunctiva, 49–50
new material technologies, 58–68 Partial body regeneration, 260
biological materials. See Biological Pathophysiological significance
materials aging, 241
engineering limbal niche, 67–68, atherosclerosis, 236–237
69t–72t cancer, 241–242
human amniotic membrane (hAM), hypertension, 239–240
58–60 inflammation, 237–238
synthetic materials. See Synthetic neuroinflammation and
materialss neurodegenerative diseases, 238
ophthalmology, 47 obesity, 240–241
overview, 80–81 preeclampsia, 240
surgical techniques type 2 diabetes, 238–239
cultivated limbal epithelial PAX6 mutation, 73–74
transplantation. See Cultivated Peptidylarginine deiminase 4 (PADI4/
limbal epithelial transplantation PAD4), 167–170, 172f
(CLET) PHF2/PHF8/KIAA1718 subgroup,
cultivated oral mucosal epithelial 203–208
transplantation. See Cultivated oral KDM7A (KIAA1718), 207–208
mucosal epithelial transplantation KDM7B, 204–207
(COMET) Piwi gene, 269–270
simple limbal epithelial transplantation. Plant homeodomain (PHD), 178
See Simple limbal epithelial Plasma polymer coating, 63–64
transplantation (SLET) Plasmodium falciparum, 28–29
OPO. See Oral siphon pigment organ Poly(N-isopropylacrylamide) (PIPAAm),
(OPO) 66–67
Oral siphon (OS) Polycomb-group (PcG), 202
aging, 275–278, 276f Polydimethylsiloxane (PDMS), 26–28
model, 260–262, 261f Polyethylene glycol diacrylate (PEGDA),
regeneration, 261f, 262–269, 263f, 68
264t Polymorphnuclear granulocytes (PMNs),
capacity, 260 12–13
long-distance regeneration, 267–268 Preeclampsia, 240
OPO replacement, 264–265 Preocular tear film, 49
288 Index
Preribosomal subunits signaling molecules, 226–227

cytoplasmic maturation, 124–130 source of, 224–226, 225f
pre-40S subunits, 127–130 treatment strategies, 242–244
pre-60S subunits, 124–127, 124f Receptor tyrosine kinases, 228–230
export, 114–124 epidermal growth factor receptor
pre-40S subunits, 121–124 (EGFR), 230
pre-60S subunits, 117–121, 118f fibroblast growth factor receptors
shared export factors, 115–117 (FGFRs), 231
nuclear maturation, 111–114 insulin receptor kinase (IRK), 230–231
pre-40S subunits, 113–114 platelet-derived growth factor receptor
pre-60S subunits, 111–113 (PDGFR), 228–230
Pre-60S maturation, 125 vascular endothelial growth factor
Primary porcine choroid plexus epithelial receptor (VEGFR), 230
cells (PCPEC), 13–14 Recombinant human type III collagen
Progenitor cell mobilization, 271–274, (RHCIII), 61–62
272f–273f Redox regulation, 228–235
nonreceptor kinases. See Nonreceptor
R Kinases
Rea1, 112–113 protein tyrosine phosphatases, 228, 229f
Reactive oxygen species (ROS), 223f receptor tyrosine kinases. See Receptor
antioxidant systems, 226 tyrosine kinases
cysteine residues, 222–224, 223f Regenerative aging, 256–257
forkhead boxO (FOXO) transcription Ribosomal RNAs (rRNAs).
factors, 235 See Eukaryotic ribosome
future perspectives, 246 Rotating cell culture systems (RCCS),
measurement, 245–246 21–23
nuclear factor-Like 2 (Nrf2), 235–236 Rotating wall vessel (RWV), 19–21
pathophysiological significance RPMI 2650 epithelial cells, 19
aging, 241 Rrp12, 116–117
atherosclerosis, 236–237 RWV. See Rotating wall vessel (RWV)
cancer, 241–242
hypertension, 239–240 S
inflammation, 237–238 Saccharomyces cerevisiae, 177
neuroinflammation and Schizosaccharomyces pombe, 177
neurodegenerative diseases, 238 Signal recognition particle (SRP), 125
obesity, 240–241 Signal transducer and activator of
preeclampsia, 240 transcription (STAT), 152
type 2 diabetes, 238–239 Simple limbal epithelial transplantation
redox regulation, 228–235 (SLET)
nonreceptor kinases. See Nonreceptor results, 84–89, 88f
Kinases surgical technique, 84
protein tyrosine phosphatases, 228, SLET. See Simple limbal epithelial
229f transplantation (SLET)
receptor tyrosine kinases. See Receptor Src family kinases, 232–233
tyrosine kinases Staphylococcus aureus, 28–29
signaling Stem cell-based therapy, 80–81
mechanism, 227 Stevens–Johnson syndrome (SJS), 50–52
Index 289
Streptococcus pneumoniae, 11 invasion, c-Src regulation of, 152–153

Streptococcus pyogenes, 15–17 self-renewal capacity, 156–157
Surface-coated lenses, 63–64 tetraspanin enriched microdomains,
Symblepharon, 82 component of, 142–143, 144f
Synthetic materials, 63–67 Trithorax (TrxG), 202
contact lenses, 63–64 Trojan Horse mechanisms, 11
polylactic glycolic acid (PLGA), 64–66, Two-dimensional (2D) cell culture,
65f–66f 4–14
thermoresponsive substrate, 66–67 alternative infection models, protozoa as,
9–10
T bacterial pathogens, 4, 5t–6t
Tails, 167–170 coculture infection models. See Coculture
Tear film, 49 infection models
Tetraspanin CD151, 148–149, 149f fluorescence staining, 7
Tetratricopeptide repeats (TPR), 200 immortalized cell lines vs. primary cell
Three-dimensional (3D) cell culture, culture, 7–9
14–21 Type 2 diabetes, 238–239
air–liquid surface, requirements of, 18–19 Tyrosine kinase inhibitors (TKIs), 153
BBB, coculture-based reconstruction of,
17–18 U
benefits and limitations, 15–19 Ubiquitin-associated (UBA), 115–116
microgravity-variations of, 19–21 UTX/UTY/JMJD3 subgroup,
Transepithelial electrical resistance 199–203
(TEER), 12 KDM6A, 199–203
Transmembrane 4-L six family member 5 KDM6A (UTX), 200–202
(TM4SF5), 143f KDM6B (JMJD3), 202–203
drug resistance, 153–155
epithelial–mesenchymal transition, V
143–150 Visual acuity, 84–89
liver fibrosis. See Liver fibrosis
zebrafish, muscle cells in, 145–146 W
metastatic potential Weibel–Palade bodies, 7
direct migration, FAK activation for, Wound epidermis forms, 262–263
150–152

Cell and Molecular Biology: Nternational Eview of

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cell and Molecular Biology: Nternational Eview of

Uploaded by

Copyright:

Available Formats

VOLUME THREE HUNDRED AND NINETEEN

Editorial Advisory Board

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

First edition 2015

Copyright © 2015 Elsevier Inc. All rights reserved.

For information on all Academic Press publications

Dilip Kumar Mishra

From Single Cells to Engineered

The implementation of natural or synthetical scaffolds elevated the model complexity

disciplines approved that simpliﬁed models enhance the probability to eluci-

Table 1 Representative examples of bacterial pathogens analyzed in different cell

Table 1 Representative examples of bacterial pathogens analyzed in different cell

2.1 Culture of Immortalized Cell Lines versus Primary Cell

PAA laboratories GmbH, Germany (part of GE Healthcare, USA) are

used as a model to study cellular processes during different stages of mono-

2.2 Protozoa as Alternative Infection Models

2.3 Coculture Infection Models

2.3.1 Coculture-based generation of tissue barriers

The complexity of BBB models ranges from transwell-based models

cells). The cells were simultaneously cultured on opposite sides of transwell

2.3.2 Coculture of adherent cells and neutrophiles in suspension

enabled basolateral infection of host cells as well as investigation of transmi-

(extracellular matrix) interactions, involve cytoskeletal orientation and

3.1 Beneﬁts and Limitations of 3D Scaffold

which are speciﬁcally required for bacterial interaction, is also highlighted in

3.1.1 Coculture-based reconstruction of BBB with matrix scaffold

astrocytes or related neuroepithelial cells contribute to the induction of

3.1.2 Requirements of 3D tissue models generating aireliquid

simulates cell type-speciﬁc function including vigorous ciliary beating, al-

3.2 MicrogravitydVariations of 3D Cell Culture Models

the phenotypic expression of eukaryotic immortalized host cells have

nonpolarized cells revealed speciﬁc differences in the infection process

4. ORGAN EQUIVALENTS AND TISSUE EXPLANTS

Moreover, after homogenization of organoid tissues, enzyme activities such

4.1 Organoids and Tissue Equivalents Providing Complex

markers are better represented in 3D aggregates formed under microgravita-

4.2 Tissue ExplantsdPiece of Reality

Legionnaires’ disease L. pneumophila for the human lung. Until recently,

investigate bacterial adherence following the infection progress, immortal-

4.3 Integration of Microﬂuidic Systems in 2D and 3D Cell

2013). However, building 3D vascularized organs remains the main

microslides allowed a microscopic visualization of long vWF (von Willebrand

5. CONCLUDING REMARKS AND FUTURE

hESC/iPSC-derived keratinocytes expose a functional permeability barrier

human polymeric Immunoglobulin receptor (hpIgR) mediate invasion of Streptococcus

Science and Art of Cell-Based

International Review of Cell and Molecular Biology, Volume 319

unique functions of these LSCs with an unlimited self-renewal capacity. We

2. ANATOMY AND PATHOLOGY OF OCULAR SURFACE

2.1 Preocular Tear Film

2.2 Conjunctival Epithelium

cells. Compared to the corneal epithelium, while the superﬁcial epithelial

inﬂammation such as in cases of acute ocular surface burns and acute

3. CELL-BASED THERAPIES FOR OCULAR SURFACE

3.1 Embryonic Stem Cells

3.2 Induced Pluripotent Stem Cells

(2007), revolutionary changes took place in the ﬁeld of regenerative science.

3.3 Mesenchymal Stem Cells

(Haynesworth et al., 1992). Due to their linkage with the formation of

3.3.1 MSCs in corneal reconstruction

3.3.2 MSCs in LSCD

human limbal niche may participate in angiogenesis and regeneration during

Figure 5 Potential questions to be answered in context of successful mesenchymal