PDF Regulatory T Cells in Health and Disease 1St Edition Liston Ebook Full Chapter

Regulatory T cells in health and disease
1st Edition Liston

Visit to download the full and correct content document:
https://textbookfull.com/product/regulatory-t-cells-in-health-and-disease-1st-edition-list
on/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Co signal Molecules in T Cell Activation Immune

Regulation in Health and Disease Miyuki Azuma
https://textbookfull.com/product/co-signal-molecules-in-t-cell-
activation-immune-regulation-in-health-and-disease-miyuki-azuma/
Micronutrients in health and disease Second Edition

Prasad
https://textbookfull.com/product/micronutrients-in-health-and-
disease-second-edition-prasad/
Manganese in Health and Disease 1st Edition Lucio G

Costa
https://textbookfull.com/product/manganese-in-health-and-
disease-1st-edition-lucio-g-costa/
Hydrogen Peroxide Metabolism in Health and Disease 1st

Edition Hampton
https://textbookfull.com/product/hydrogen-peroxide-metabolism-in-
health-and-disease-1st-edition-hampton/
Growth hormone in health and disease First Edition
Casanueva
https://textbookfull.com/product/growth-hormone-in-health-and-
disease-first-edition-casanueva/
Ion Channels in Health and Disease 1st Edition Geoffrey

S. Pitt
https://textbookfull.com/product/ion-channels-in-health-and-
disease-1st-edition-geoffrey-s-pitt/
Human Retrotransposons in Health and Disease 1st

Edition Gael Cristofari (Eds.)
https://textbookfull.com/product/human-retrotransposons-in-
health-and-disease-1st-edition-gael-cristofari-eds/
Live cell assays from research to health and regulatory

applications 1st Edition Furger
https://textbookfull.com/product/live-cell-assays-from-research-
to-health-and-regulatory-applications-1st-edition-furger/
Circadian Clocks Role in Health and Disease 1st Edition

Michelle L. Gumz
https://textbookfull.com/product/circadian-clocks-role-in-health-
and-disease-1st-edition-michelle-l-gumz/
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
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
125 London Wall, London, EC2Y 5AS, 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: 978-0-12-803415-6
ISSN: 1877-1173
For information on all Academic Press publications

visit our website at http://store.elsevier.com
CONTRIBUTORS
Jakub Abramson
Faculty of Biology, Department of Immunology, Weizmann Institute of Science, Rehovot,
Israel
Michael Delacher
Immune Tolerance, Tumor Immunology Program, German Cancer Research Center
(DKFZ), Heidelberg, Germany
Maxime Dhainaut
Laboratory of Immunobiology, Department of Molecular Biology, Université Libre de
Bruxelles, Brussel, Belgium
Darcy Ellis
Infection and Immunity Program, Monash Biomedicine Discovery Institute and Department
of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria, Australia
Markus Feuerer
Thomas S. Fulford
Steve Gerondakis
Yael Goldfarb
Faculty of Biology, Department of Immunology, Weizmann Institute of Science, Rehovot,
Israel
Ann-Cathrin Hofer
Jochen Huehn
Department of Experimental Immunology, Helmholtz Centre for Infection Research,
Braunschweig, Germany
Axel Kallies
Walter and Eliza Hall Institute of Medical Research, and Department of Medical Biology,
University of Melbourne, Parkville, Victoria, Australia
Danny Kägebein
ix
x Contributors
Yohko Kitagawa
Department of Experimental Immunology, Immunology Frontier Research Center, Osaka
University, Suita, Osaka, Japan
Noriko Komatsu
Department of Immunology, Graduate School of Medicine and Faculty of Medicine,
The University of Tokyo, Bunkyo-ku, Tokyo, Japan
Adrian Liston
Translational Immunology Laboratory, VIB, and Department of Microbiology and
Immunology, University of Leuven, Leuven, Belgium
Matthias Lochner
Institute of Infection Immunology, TWINCORE, Centre for Experimental and Clinical
Infection Research: A Joint Venture Between the Medical School Hannover (MHH) and the
Helmholtz Centre for Infection Research (HZI), Hannover, Germany
Jennifer M. Lund
Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, and
Department of Global Health, University of Washington, Seattle, Washington, USA
Muriel Moser
Laboratory of Immunobiology, Department of Molecular Biology, Université Libre de
Bruxelles, Brussel, Belgium
Vitalijs Ovcinnikovs
Institute of Immunity & Transplantation, Division of Infection & Immunity, University
College London, London, United Kingdom
Maria Pasztoi
Joern Pezoldt
David M. Richards
Laura E. Richert-Spuhler
Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle,
Washington, USA
Shimon Sakaguchi
Tim Sparwasser
Contributors xi
Hiroshi Takayanagi
Department of Immunology, Graduate School of Medicine and Faculty of Medicine,
The University of Tokyo, and Japan Science and Technology Agency, Exploratory Research
for Advanced Technology Program, Takayanagi Osteonetwork Project, Bunkyo-ku,
Tokyo, Japan
Peggy P. Teh
Annemarie van Nieuwenhuijze
Translational Immunology Laboratory, VIB, and Department of Microbiology and
Immunology, University of Leuven, Leuven, Belgium
Ajithkumar Vasanthakumar
Lucy S.K. Walker
Institute of Immunity & Transplantation, Division of Infection & Immunity, University
College London, London, United Kingdom
Zuobai Wang
James Badger Wing
PREFACE
Regulatory T cells (or Tregs) are a unique subpopulation of T cells with sup-
pressive properties, acting to counter the immunogenic function of other
T cells. This function is critical for the prevention of autoimmune disease
and also has profound impacts on other aspects of the mammalian immune
system, leading to an intensive effort to harness the power of Tregs as a novel
therapeutic strategy across multiple immune diseases.
This volume takes a broad and comprehensive look at Tregs in health
and disease states. We have expert chapters on the generation of Tregs, with
contributions by Sakaguchi, Huehn, Feuerer, and Abramson on the pro-
cesses by which Tregs are generated in the thymus and peripheral organs
such as the gut. Complementing these chapters, we have articles by Geron-
dakis, van Nieuwenhuijze, and Kallies, which dissect the molecular path-
ways that control the induction and differentiation of Tregs. Sparwasser
and Moser discuss the cellular dynamics Tregs share with Th17 cells and
dendritic cells. Finally, we have an assessment of the physiological impact
on Tregs in disease, with expert chapters by Takayanagi, Lund, and Walker
on the role of Tregs in arthritis, infection, and diabetes.
ADRIAN LISTON
xiii
CHAPTER ONE
Transcriptional and Epigenetic

Control of Regulatory T Cell
Development
Yohko Kitagawa, James Badger Wing, Shimon Sakaguchi1
Department of Experimental Immunology, Immunology Frontier Research Center, Osaka University, Suita,
Osaka, Japan
1
Corresponding author: e-mail address: [email protected]
Contents
1. Introduction 2
2. Transcriptional Regulation in Treg Cells 4
2.1 Foxp3-Dependent Transcriptional Regulation 4
2.2 Foxp3-Independent Transcriptional Regulation 9
3. Epigenetic Regulation in Treg Cells 10
3.1 Stability of the Treg Cell Lineage 10
3.2 cis-Regulatory Elements of the Foxp3 Gene 12
3.3 DNA Demethylation 13
3.4 Histone Modification 14
3.5 Nucleosome Positioning 16
4. Cross talk Between Foxp3-Dependent Gene Regulation and Treg Cell-Type
Epigenetic Modifications 17
5. Treg Cell Development 18
5.1 Signals Involved in Treg Cell Development 20
5.2 Transcription Factors Involved in Foxp3 Induction 21
5.3 Induction of Epigenetic Modification During Treg Cell Development 24
5.4 Coordination of Transcriptional and Epigenetic Changes During Treg Cell
Development 25
6. Conclusion 27
Acknowledgment 27
References 27
Abstract
The control of immune responses against self and nonharmful environmental antigens
is of critical importance to the immune homeostasis. Regulatory T (Treg) cells are the key
players of such immune regulation and their deficiency and dysfunction are associated
with various immune disorders, such as autoimmunity and allergy. It is therefore essen-
tial to understand the molecular mechanisms that make up Treg cell characteristics; that
is, how their unique gene expression profile is regulated at transcriptional and
Progress in Molecular Biology and Translational Science, Volume 136 # 2015 Elsevier Inc. 1
ISSN 1877-1173 All rights reserved.
http://dx.doi.org/10.1016/bs.pmbts.2015.07.011
2 Yohko Kitagawa et al.
epigenetic levels. In this chapter, we focus on the components of molecular features of

Treg cells and discuss how they are introduced during their development.
1. INTRODUCTION
Treg cells are a subset of CD4+ T cells, specialized in the maintenance
of immune tolerance and prevention of autoimmunity. Treg cells are unique
in that their primary function is to suppress aberrant or excessive immune
responses harmful to the host by counteracting the effects of conventional
T cells. This property of Treg cells is particularly important in the establish-
ment of self-tolerance. Discrimination between self and nonself is required
for the immune system to avoid attacking self-tissues and organs and causing
autoimmune diseases. Along with deletion of self-reactive T cells during
their development and induction of an anergic state in self-reactive
T cells in peripheral lymphoid organs, thymic production of Treg cells,
and their immune suppression in the periphery are a critical mechanism
of self-tolerance. In addition, conventional T cells can give rise to Treg cells
under certain conditions, contributing to immune homeostasis in the
periphery.
The production of suppressive cells in the thymus was initially noted in
experiments where the removal of thymus from neonatal mice led to severe
autoimmunity.1 However, it was not until 1995 that Treg cells were defin-
itively identified by specific expression of the alpha chain of the IL-2 recep-
tor (CD25),2 which enabled the finding that Treg cells constituted around
10% of CD4+ T cells and clearly demonstrating that they have a critical role
in self-tolerance. This was then further confirmed with the discovery of the
lineage defining transcription factor Foxp3.3,4 Foxp3 is essential for the
function of Treg cells, as loss-of-function mutations of Foxp3 in either
the scurfy mouse strain or IPEX (immunodysregulation, poly-
endocrinopathy, enteropathy, X-linked) syndrome leads to severe autoim-
munity including Type-1 diabetes (T1D), immunopathology such as
inflammatory bowel disease, and allergy accompanying hyperproduction
of IgE.5–7 Furthermore, depletion of Treg cells in adults also leads to similar
autoimmune pathology, demonstrating that Treg cells are needed not just
for the establishment, but also the lifelong maintenance, of immune self-
tolerance and homeostasis.8
In addition to severe acute autoimmunity seen in the complete absence
of Treg cells, more subtle defects in Treg cell function have been implicated
Transcriptional and Epigenetic Control 3
in the development of a wide range of chronic autoimmune diseases. Partial

loss of Treg cell function or reduction in Treg cell numbers has been asso-
ciated with a range of human autoimmune disorders such as T1D, rheuma-
toid arthritis, systemic lupus erythematosus, thyroiditis, hepatic disease, and
dermatitis (reviewed in Ref. 9). These finding are confirmed in a number of
mouse models of autoimmunity. In nonobese diabetes mice, a model of
T1D, defective IL-2 signaling is associated with low Treg cell numbers in
the pancreas and the development of diabetes. Conversely, treatment with
IL-2 expands Treg cells and prevents the development of diabetes.10 In the
case of colitis, transfer of naı̈ve (CD45RBhigh) CD4+ T cells into T cell-
deficient mice leads to the development of colitis; while cotransfer of Treg
cells is able to prevent the disease.11 Treg cells also play a critical role in the
regulation of humoral immunity and prevention of allergy, as evidenced by
the characteristically high levels of IgE production seen in scurfy mice and
IPEX patients.12 Another aspect of Treg cell-mediated suppression of self-
reactive T cells is that Treg cells are able to suppress antitumor immune
responses. The presence of Treg cells in tumors is often inversely correlated
with survival in both mice and humans. This indicates that depletion of Treg
cells and targeting of their suppressive functions can be an important tool in
antitumor immunotherapy.13
A wide range of Treg cell-mediated suppressive mechanisms have been
described, suggesting that they may have context-specific roles at different
sites.14 To date, CTLA4, IL-10, TGFβ, ITGβ8, micro-RNA containing
exosomes, IL-35, granzyme, perforin, CD39, CD73, and TIGIT have all
been demonstrated to have a role in Treg suppressive function. In particular,
CTLA4 expression by Treg cells is crucial for Treg cell-mediated immune
suppression. CTLA4 downregulates the expression of the costimulatory
molecules CD80 and CD86 on the surface of antigen presenting cells,
thereby influencing their ability to activate conventional T cells.15 Treg
cell-specific loss of CTLA4 leads to the development of fatal autoimmunity
and dysregulated humoral immunity, similar to that seen in scurfy or Treg-
depleted mice.16–18 Further information on the critical role of CTLA4 in
humans has been revealed by the finding that haploinsufficiency of CTLA4
leads to a severe autoimmune syndrome, similar to that seen in IPEX, albeit
with variable penetrance and age of onset.19,20
Another key feature of Treg cells is their inability to produce IL-2,
despite their high dependency on IL-2 for survival and proliferation. IL-2
is also a driving factor for conventional T cell proliferation and some effector
T cell differentiation. In this competition for IL-2, high expression of the
high-affinity IL-2 receptor even at the resting state gives Treg cells an advan-
tage and IL-2 deprivation by Treg cells from other T cells is one mechanism
of immune suppression. Indeed, overexpression of CTLA4 and repression of
IL-2 in conventional T cells enable them to behave like Treg cells.21 Con-
versely, failure to repress IL-2 in Treg cells is associated with the develop-
ment of autoimmunity.22
These molecular features are regulated at both the transcriptional and epi-
genetic levels. Foxp3-dependent transcriptional programs, which often
involve interaction with other transcription factors, control some Treg cell-
type gene expression, while Foxp3-independent epigenetic modifications also
contribute to the generation of Treg cell characteristics. There is dynamic
cross talk between transcriptional and epigenetic regulation in a cooperative
manner, which enables stable maintenance of Treg cell characteristics
throughout multiple divisions, regardless of environmental changes. Given
the critical and wide-ranging roles of Treg cells in autoimmunity, allergy,
infection, and tumor immunology, it is vital to understand the molecular
mechanisms underlying the development and maintenance of Treg cells in
order to develop more sophisticated strategies to either enhance or dampen
the function of Treg cells in clinical settings. Here, we review the current
understanding of transcriptional and epigenetic regulation in Treg cells and
discuss how these molecular changes occur during Treg cell development.
2. TRANSCRIPTIONAL REGULATION IN TREG CELLS

Treg cells have a distinct gene expression profile. Foxp3 regulates
some gene expression directly and others in cooperation with its cofactors,
while there is also a set of gene expression that is controlled independently
of Foxp3.
2.1 Foxp3-Dependent Transcriptional Regulation

2.1.1 Foxp3 as a Master Regulator
Foxp3 is a transcription factor that is specifically expressed by Treg cells. As
its deletion impairs the suppressive function of Treg cells and causes similar
autoimmune diseases to Treg cell depletion, Foxp3 is indispensable for Treg
cell function and is considered as the master regulator of Treg cells. Indeed,
Foxp3 is able to upregulate or downregulate about half of the genes that are
overexpressed or underexpressed, respectively, in Treg cells, compared to
conventional T cells.23 Importantly, such transcriptional changes induced
by overexpression of Foxp3 in conventional CD4+ T cells are sufficient
to provide suppressive function similar to that of Treg cells.4 Moreover,

overexpression of Foxp3 and certain transcription factors, such as Irf4,
Eos, and Gata1, generates almost complete Treg cell-type transcription pro-
file in conventional CD4+ T cells.24 Taken together, these findings demon-
strate that Foxp3, solely or cooperatively with other transcription factors,
regulates the majority of gene transcription in Treg cells.
At the molecular level, Foxp3 mainly functions as a transcriptional
repressor and contributes to some of the key characteristics of Treg cells.25,26
The direct targets of Foxp3 are predominantly those that are normally
upregulated by TCR stimulation in conventional CD4+ T cells. A large
fraction of them are involved in signaling pathways, such as Zap70, Ptpn22,
and Itk.27 Foxp3 also represses the expression of IL-2.28 This repression and
high dependence on paracrine IL-2 enable Treg cells to suppress conven-
tional T cell proliferation by IL-2 deprivation. Furthermore, Foxp3 directly
represses Satb1 by binding to its promoter and inducing microRNAs that
target Satb1, to prevent the expression of proinflammatory cytokines that
are normally produced by effector T helper cells.29 Thus, one function of
Foxp3 is to repress genes that are activated by T cell activation, and Foxp3
targets genes that serve as regulators of many other genes, thereby efficiently
maintaining Treg cell characteristics.
Foxp3 is also involved in upregulation some genes. Hallmarks of Treg
cells such as Il2ra, Ctla4, and Tnfrsf18 are all bound by Foxp3 and positively
regulated.27 However, Foxp3-null Treg cells, analyzed using mouse models
that express a fluorescent marker instead of Foxp3, still express these genes,
as well as most of the genes upregulated in Treg cells, but at a lower level
than in wild-type Treg cells.30 These findings illustrate the role of Foxp3
in amplification of pre-existing molecular features.
In terms of the regions that Foxp3 binds to, only a subset of Foxp3-bound
genes showed differential expression between Foxp3+ and Foxp3 T cell
hybridomas, suggesting that Foxp3 requires cofactors for its transcription.27
Consistently, many of the Foxp3-binding sites overlap with other transcrip-
tion factor binding sites.31 Therefore, Foxp3, as a master regulator of Treg
cells, is able to directly regulate some characteristics of Treg cells, but is
insufficient for the generation of full Treg cell-type gene expression, which
may require other transcription factors and epigenetic regulation.
2.1.2 Foxp3 and Its Cofactors

As with most transcription factors, Foxp3 interacts with a number of other
transcription factors: some being general transcriptional regulators and
others being T cell or Treg cell-specific ones. Some of the proteins currently
reported to be capable of interacting with Foxp3 are NFκB,32 NFAT,22
Runx1,28 Eos, CtBP1,33 CBFb, Gata3, Ash2l, Bcl11b, Ikzf3, Foxp1,
Smarcc1, Smarce1, Smarca4, Smarca5, Chd4, Hdac2, Rcor1, Lsd1,34
HIF-1α, IRF-4,35 p300, TIP60,36 and Ezh2.26 Though Foxp3 is likely to
exist in a large protein complex, not all these cofactors are always found
in the same complex. There are two features determined by the interaction
with particular cofactors: effects of binding on target gene transcription and
location of Foxp3 binding.
First, Foxp3 can serve as both transcriptional activator and repressor and
these modes of action are determined by the recruitment of coactivators or
corepressors. For example, human FOXP3 protein is capable of interacting
with the coactivators p300 and TIP60 and such interaction promotes the
transcriptional activity of FOXP3, while Treg cell-specific deletion of
p300 and TIP60 results in loss of Treg function.36 In contrast, Foxp3 recruits
Eos and the corepressor CtBP1 to repress the expression of genes such as Il2.
Since IL-2 repression is critical for Treg cell-mediated immune regulation,
silencing Eos in Treg cells abrogates their suppressive function.33 Notably,
some of the factors that Foxp3 interacts with, such as Smarca4, Hdac2, and
Ezh2 are known as epigenetic regulators, suggesting that Foxp3 recruits
these factors to modulate epigenetic features for long-term control of gene
expression (discussed in Section 4). Thus, Foxp3 interacts with appropriate
cofactors in a locus-specific manner in order to generate Treg cell-type gene
expression (Fig. 1).
Second, Foxp3 is dependent on other transcription factors for binding
guidance in some loci, meaning that cofactors alter the targets of its gene
regulation. Some interactions are fundamentally required for generating
Treg cell phenotypes in physiological conditions. For example, NFκB
and NFAT transcription factors have been shown to interact with Foxp3
and cooperatively repress the expression of proinflammatory cytokine genes
such as Il2, Il4, and Ifng.22,32 Mutations at the interface of Foxp3 and NFAT
interaction resulted in the production of IL-2 by Treg cells and failure to
prevent the manifestation of type I diabetes.22 Other interactions are utilized
for particular purposes, such as regulation of specific effector T cell subsets
during inflammation. For example, during Th2-type inflammation, Foxp3
interacts with IRF4, which is a transcription factor essential for Th2 cell dif-
ferentiation program, and enables Treg cells to efficiently control Th2-type
inflammation.37 Importantly, in addition to the variety of Foxp3 complexes
at different genomic loci, the repertoire of Foxp3–cofactor complexes
Figure 1 Foxp3-dependent gene expression. Some Foxp3-dependent gene regulation

is mediated by the interaction of Foxp3 with transcription factors downstream of TCR/
costimulation and IL-2, which are also required for the induction of Foxp3 expression.
Others involve the interaction of Foxp3 with T cell-specific or Treg cell-specific transcrip-
tion factors, such as Runx and Eos.
within a cell may vary depending on the differentiation stage of Treg cells
and the environmental conditions they are exposed to. In this sense, the bal-
ance among Foxp3 cofactors may be an important determinant of what
Foxp3 interacts with. When a fluorescent marker is fused to the
N-terminus of Foxp3, it impaired the interaction of Foxp3 with HIF-1α
and instead recruited IRF4.35 Consequently, some gene regulation is altered
with particular upregulation of IRF4 signature genes, and these mutant Treg
cells alleviated rheumatoid arthritis, but exacerbated type I diabetes.35 The
cause of cofactor change may be due to the competition between HIF-1α
and IRF4, or due to the alteration in posttranslational modification of
Foxp3. Nevertheless, selection of partners for Foxp3 can serve as a molecular
switch for Foxp3-dependent transcription and consequent Treg cell
function.
The requirement of Foxp3 to interact with its cofactors indicates that

these cofactors also need to be expressed in Treg cells for Foxp3-dependent
transcription. Interestingly, a large proportion of these cofactors are direct
targets of Foxp3.34 This notion indicates that Foxp3 directly upregulates
the minimum targets by itself, and then regulate the rest of the gene expres-
sion in cooperation with these Foxp3 targets that now serve as cofactors.
Furthermore, some cofactors such as Runx1, NFAT, and Bcl11b are known
to promote Foxp3 transcription, suggesting that Foxp3 and some cofactors
positively regulate each other to achieve stable gene regulation.38–40 There
are also cofactors that are independently expressed from Foxp3. For exam-
ple, NFκB and NFAT are transcription factors activated upon TCR/
costimulation. The requirement of these factors for Foxp3-dependent tran-
scriptional regulation suggests that Treg cell specification and maintenance
requires TCR signaling in addition to Foxp3 expression. In fact, a large part
of Foxp3 targets are coregulated by TCR/costimulation and the number of
genes regulated by Foxp3 increase dramatically, as Treg cells become acti-
vated.23,25 Consistent with this, genetic ablation of TCR in mature Treg
cells results in a loss of 25% of activated Treg cell signature.41 Therefore,
while some cofactors are upregulated by Foxp3, others are independently
expressed, possibly under limited conditions in which Treg cell lineage spec-
ification occurs.
Finally, there are “quintet” factors that have been shown to redundantly
cooperate with Foxp3 to generate most of the Treg-type gene expression:
Eos, Gata1, IRF4, Satb1, and Lef1.24 Notably, these factors and Foxp3 were
retrovirally transduced in conventional CD4+ T cells in this experimental
setting, suggesting that TCR stimulation required for retroviral transduction
may contribute to some of the Treg cell-type transcriptional regulation.
However, even so, coexpression of at least one of the quintet factors with
Foxp3 enabled the much more efficient induction of the Treg up- and
downregulated gene expression profile than the overexpression of Foxp3
alone. Not all of these quintet factors have been shown to physically interact
with Foxp3 protein yet, but they are certainly the coregulators of Foxp3-
dependent transcription. How they maximize the transcriptional capacity
of Foxp3 remains to be elucidated and it is particularly puzzling that two
of the quintet factors, Satb1 and Lef1, are downregulated in Treg cells.
One speculation is that coexpression of quintet factors and Foxp3 turns
on the molecular switch to build and activate the protein complex around
Foxp3. The redundancy among quintet factors, despite belonging to differ-
ent families and having different functions, may be a mechanism to allow the
generation of Treg cell-type gene expression, once Foxp3 is expressed, in

various settings where only one of the quintet factors may be expressed.
2.1.3 Foxp3 Posttranslational Modification

For protein interaction and activity of each protein, posttranslational mod-
ifications are crucial. Foxp3 is also subjected to such modification. In partic-
ular, acetylation of lysine residues is a key determinant of Foxp3 stability and
transcriptional activity. Histone acetyltransferases p300 and TIP60, acetylate
Foxp3, whereas histone deacetylases SIRT1, HDAC7, and HDAC9 reverse
this process.42 When acetylated, Foxp3 has higher DNA-binding capacity,
thereby enhancing transcriptional activity and becomes more resistant to
polyubiquitination and proteasomal degradation.43 This accords with the
result that deleting SIRT1 does not have much effect on conventional
T cell function and proliferation, but increases Foxp3 expression and Treg
cell suppressive activity. These positive effects on Foxp3 function make
SIRT1 a promising target for the induction of transplantation tolerance.
Indeed, T cell-specific deletion of SIRT1 or administration of pharmaco-
logical SIRT1 inhibitors in mice prevented allograft rejection.44
Another posttranslational modification that regulates Foxp3 transcrip-
tional activity is the phosphorylation of a serine residue (Ser418 in humans).
Lack of this modification results in impaired Foxp3 function as indicated by
the failure to repress IL-2 production.45 Ser418 can be dephosphorylated by
protein phosphatase 1 (PP1), and during rheumatoid arthritis, induction of
PP1 by the proinflammatory cytokine, TNFα, in inflamed synovium
dephosphorylates Foxp3 protein, impairs Treg cell function and contributes
to disease pathogenesis. This demonstrates that posttranslational modifica-
tions of Foxp3 serve as a key regulator of Treg cell-mediated immune
suppression.
2.2 Foxp3-Independent Transcriptional Regulation

Though Foxp3 is the master regulator of Treg cells, Treg cell-type gene reg-
ulation also includes Foxp3-independent features.30,46 This is evident from
the fact that Foxp3-null Treg cells retain a large portion of Treg-type gene
expression.30,47,48 This finding can be partly explained by the fact that TCR,
IL-2, and TGFβ signaling also regulate the majority of Foxp3 target genes
and the number of genes that are solely controlled by Foxp3 is limited.23
However, there is still a significant fraction (more than 25%) of Treg-type
gene expression that is not regulated by Foxp3, TCR, IL-2, or TGFβ sig-
naling.30,46 Some are regulated by other transcription factors coexpressed in
Treg cells. For example, Foxo1, which is highly expressed and activated by
phosphorylation in Treg cells, controls a subset of Treg cell-type gene
expression, independently of Foxp3.49 Others, such as Eos and Helios,
are associated with Treg cell-type epigenetic modifications. This suggests
that the permissive chromatin status of these genes enables constitutively
expressed transcription factors to induce their expression, rather than specif-
ically expressed transcription factors being responsible for their expression.48
3. EPIGENETIC REGULATION IN TREG CELLS

To understand the mechanisms of cell type-specific transcriptional
regulation, in addition to the activity of transcription factors, the status of
target gene loci is another factor that needs to be considered. That is, there
are two requirements for the activation of gene transcription: (1) the respon-
sible transcription factors (trans-regulatory factors) are expressed and (2) the
chromatin configuration of the target gene locus (cis-regulatory elements) is
permissive so that the transcription factors can bind. The latter is regulated
by various epigenetic modifications of chromatin, such as DNA methyla-
tion, histone modification, and nucleosome positioning (Fig. 2). These basic
criteria need to be met at least at the gene promoters. In addition, such
requirements extend to enhancers for stabilizing high gene expression.
Epigenetic modifications of cis-regulatory elements have been implicated
in lineage determination. There is a close association among cell differenti-
ation, permissive epigenetic modifications at gene loci associated with the
cell lineage, and repressive epigenetic modifications at gene loci related to
the alternative cell fate. For example, as multipotent progenitors differentiate
into common lymphoid progenitors, they show DNA demethylation in
lymphoid lineage-specific genes, while undergoing DNA methylation at
myeloid lineage-specific genes.50 These lineage-specific epigenetic modifi-
cations are thought to assist irreversible lineage specification by ensuring the
stable expression of key regulator genes. This concept is also applicable to
Treg cells, which are indeed characterized by distinct epigenetic
modifications.
3.1 Stability of the Treg Cell Lineage

The gene expression regulation by Foxp3 and its cofactors is required not
only during the Treg cell development but also for their functional mainte-
nance. Ablation of Foxp3 in mature Treg cells resulted in the reversal of
Foxp3-dependent gene expression program and consequently these cells lost
Figure 2 Alteration in transcription factor accessibility by epigenetic modifications.

Accessibility of transcription factor to target regions can be determined by various epi-
genetic modifications: the removal of methyl group on CpG residues by DNA demeth-
ylation, loosening of chromatin around histone octamer by permissive histone
modification, and detachment or sliding of nucleosome that facilitates transcription fac-
tor binding to DNA.
suppressive function.47 Thus, maintaining Treg cell identity requires contin-

uous expression of Foxp3. This raises an important question regarding the
stability of Treg cells. If Treg cells lose Foxp3 under certain conditions, such
as during inflammation where effector T cell polarizing stimuli are abundant,
they can lose suppressive function and behave like effector T cells. Because
Treg cells possess a relatively self-reactive TCR repertoire, Treg cell plastic-
ity indicates a potential hazard of eliciting immune responses against self.
Several studies showed Treg cells could be plastic when they receive chronic
antigenic stimulation.51,52 Moreover, once they lose Foxp3, they secrete
proinflammatory cytokines such as IFNγ, possibly contributing to the
amplification of inflammation. In contrast, there are also reports demonstrat-
ing that Treg cells are stable regardless of environmental conditions.53,54
This controversy remains unsolved, but may be explained by the definition
of function-competent Treg cells and different experimental systems
utilized.
There are some Foxp3+ T cells that are not committed to Treg cell lin-
eage. TCR stimulation induces Foxp3 expression in some murine conven-
tional T cells and these activation-induced Foxp3+ T cells do not possess
Treg cell characteristics except low expression of Foxp3.54 In humans,
Foxp3 is more loosely regulated and can be transiently upregulated by
in vitro stimulation of CD4+CD25 non-Treg cells, while there is a clear
fraction of Foxp3+ T cells with no suppressive function in vivo.55,56 Foxp3
expression can also be induced in vitro by stimulating both murine and
human conventional T cells in the presence of TGFβ and IL-2.57 These cells
have some suppressive function but are unable to maintain prolonged Foxp3
expression upon removal of such stimulation, indicating that they are not
fully committed to Treg cell lineage. Therefore, maintaining Foxp3 expres-
sion involves additional mechanisms to those that activate Foxp3 promoter
activity. This suggests that whatever that ensures the stable expression of
Foxp3 is the true indicator of Treg cell lineage.
3.2 cis-Regulatory Elements of the Foxp3 Gene

There are four cis-regulatory elements in the Foxp3 locus, important for the
regulation of gene expression: the promoter and three enhancers. The pro-
moter contains response elements for transcription factors downstream of
TCR/costimulation, such as NFAT, AP-1, and Nr4a family members
and for STAT5, which is activated by IL-2 signaling. This explains the
induction of low, transient Foxp3 expression by these signals. In order to
achieve stable Foxp3 expression at high level, however, appropriate
enhancers need to be activated and looped to Foxp3 promoter. The Foxp3
gene has three conserved noncoding regions that serve as enhancers. These
are referred to as conserved noncoding sequence (CNS) 1, CNS2, and
CNS3.58
CNS1 is an enhancer within intron 1 of the Foxp3 locus, shown to be
required for peripheral Treg cell differentiation. It contains a TGFβ signaling
response element with Smad3-binding site. CNS2 is another enhancer
located in the intron 1 with binding sites for Ets1, Foxp3, and CREB
and is associated with Foxp3 expression stability. Its deletion results in grad-
ual loss of Foxp3 expression as cells divide.58 CNS3 is considered as an
enhancer required for efficient thymic Treg cell development, as its deletion
leads to a severe reduction in Treg cells in the thymus. In this way, these
enhancers are activated at different stages of Treg cell development and
maintenance.
3.3 DNA Demethylation

Of currently known Treg cell-specific epigenetic modifications, DNA
demethylation of Foxp3 CNS2 region most highly correlates with the
stability of Treg cells. When CpG residues are methylated, the methyl group
interferes with transcription factor binding, whereas demethylation increases
the accessibility for transcription factors by revealing their consensus
sequence. Indeed, the transcription factors CREB, Ets1, and Foxp3 bind
to CNS2 in a demethylation-dependent manner.58–60 Treg cell-specific
DNA demethylated regions (TSDRs) are present not only at Foxp3 locus
but also at other Treg signature gene loci, such as Ctla4, Ikzf4, Tnfrsf18,
and Ikzf2 and their demethylated status is highly stable and specific in Treg
cells, suggesting that TSDR DNA demethylation ensures the stable expres-
sion of genes closely associated with Treg cell function.48 DNA demethyl-
ation at lineage-specific gene loci has also been observed in many other cell
types,50 indicating that this is a common mechanism of lineage specification.
One possible mechanism for the key role of CNS2 demethylation in
determining stable expression of Foxp3 is the constitutive expression of tran-
scription factors bound to CNS2. Unlike Smad3 that is activated upon TGFβ
signaling, CNS2-bound transcription factors such as Runx, Gata3, and Ets1
are constitutively present in T cells, enabling stable enhancer activation
regardless of changes to the cell environment. Indeed, Runx/Cbfb deletion
results in gradual loss of Foxp3 expression in mature Treg cells.61 Similarly,
Gata3 binds to CNS2 in Treg cells and Treg cell-specific Gata3 deletion leads
to failure to maintain Foxp3 expression.62 Furthermore, Foxp3 binds to
CNS2 by interacting with CNS2-bound Runx1/Cbfb complex and amplifies
its own expression, forming a positive feedback loop.58 These mechanisms
may contribute to the stable inheritance of Foxp3 expression as cells divide.
However, in order for Treg cells to stably exert their suppressive function
even during inflammation, they also need mechanisms to counteract the
effects of helper T cell polarizing stimuli. Recent evidence demonstrates that
TCR activation in Treg cells facilitates the binding of downstream transcrip-
tion factors, NFAT and NFκB to promoter and CNS2, which are looped to
ensure stable expression of Foxp3.63 STAT5 activated by IL-2 signaling also
binds to CNS2 and protects Treg cells from losing Foxp3 expression.64 These
results suggest that DNA demethylation at Foxp3 CNS2 is a key determinant
of Treg cell lineage stability and TCR stimulation and IL-2 signaling further
lock their identity under inflammatory conditions.
DNA demethylation of TSDRs may also contribute to the regulation of
Foxp3-dependent and -independent gene expression in Treg cells. First,
TSDR demethylation is observed at Ikzf4 and Ikzf2 gene loci and these
genes are not upregulated by Foxp3, TCR stimulation, or TGFβ signal-
ing.23,48 With no signals or transcription factors identified to induce and
maintain these genes in Treg cells, it is conceivable that TSDR demethyla-
tion, followed by binding of constitutively expressed transcription factors
induces the expression of some Foxp3-independent genes. Second, TSDR
demethylation is also observed at genes upregulated by Foxp3 and TCR
stimulation, such as Ctla4 and Tnfrsf18.48 This may explain how Treg cells
maintain high level of these molecules, even at a resting state. Taken together,
these findings suggest that TSDR demethylation facilitates Foxp3-dependent
transcriptional regulation by stabilizing Foxp3 expression as well as assisting
with both Foxp3-dependent and -independent gene regulation.
3.4 Histone Modification

Histone modification is another relatively well-studied epigenetic feature.
Histones form an octomeric nucleosome core, consisting of two copies each
of the core histones H2A, H2B, H3, and H4, and together with DNA wrap-
ped around it, make up the nucleosome. Posttranslation modifications of his-
tones have a critical role in the control of transcription as they alter the
accessibility of a particular genomic region to transcription factors. There
are various modification types, each associated with a permissive or repressive
effect on transcription. Well-studied modifications are monomethylation,
dimethylation, and trimethylation of histone H3 at Lys4 (H3K4me1,
H3K4me2, and H3K4me3, respectively), acetylation and trimethylation of
histone H3 at Lys27 (H3K27ac, H3K27me3), and acetylation of histone
H3 at Lys9 (H3K9ac). Studying these histone codes can reveal the status of
gene transcription and enhancer activity. For example, H3K4me3 and
H3K9ac are found in actively transcribed promoters, whereas H3K4me1
and H3K27ac indicate poised and active enhancers, respectively.65
In Treg cells, promoters of Treg cell-associated genes, such as Foxp3, are
marked with permissive H3K4me3 modification, strongly correlating with
gene expression.66 Moreover, DNA demethylation at TSDRs was found to
correlate with increased H3K4me3 modification, suggesting that Treg cell-
specific DNA demethylation and H3K4me3 modification have similar roles
in the maintenance of Treg cell lineage.67 However, H3K4me3 modifica-
tion at the Foxp3 promoter is more easily induced, compared to DNA
demethylation of CNS2 region; whereas DNA demethylation does not
occur after TCR/costimulation, IL-2 signaling or TGFβ signaling,
H3K4me3 modification at transcription start site increases when naı̈ve

CD4+ T cells are stimulated.48 This correlates with the temporary induction
of Foxp3 expression, but as Foxp3 expression is lost, H3K4me3 modifica-
tion also decreases.68 Thus, this particular type of histone modification may
be merely an indicator of active transcription.
Treg cells are also characterized by unique patterns of H3K4me1 and
H3K27ac modifications. With these modifications serving as the markers
for poised and active enhancers, their comparison in human conventional
T cells and Treg cells revealed the presence of Treg cell-specific enhancers
and there was a correlation between the activation of cell-specific enhancers
and neighboring gene expression.69 Moreover, Treg cell-specific enhancers
were enriched with STAT5 binding, whereas conventional T cell-specific
enhancers were frequently bound by Runx and Ets1, indicating that Treg
cells are more dependent on the IL-2-STAT5 pathway for their enhancer
activation and/or gene regulation at activated enhancers. It will be impor-
tant to address whether enhancer activation is required for binding of these
transcription factors, or vice versa.
While permissive histone modifications are found near genes
upregulated in Treg cells, repressive histone modification, H3K27me3, is
found near genes downregulated in Treg cells and is controlled at least par-
tially in a Foxp3-dependent manner.26 Foxp3 itself does not possess the abil-
ity to directly modify epigenetic features, but it interacts with a number of
components of nucleosome remodeling deacetylase complex and SWI/SNF
complex.34 These complexes are known to modulate chromatin organiza-
tion and histone modification, suggesting that Foxp3 utilizes these com-
plexes to stably control gene expression. A recent study has revealed that
many Foxp3 target genes are characterized by H3K27me3 modification
in activated Treg cells and their expression is epigenetically repressed.26
The loss of H3K27me3 pattern at some of these locations in Foxp3-deficient
Treg cells demonstrates that Foxp3 and its partner proteins induce epigenetic
repression. PRC2 (polycomb repressive complex 2) is one of the partner
protein complexes recruited by Foxp3 for this purpose. Foxp3 has been
shown to interact with a PRC2 component, Ezh2 in activated Treg cells
and when Ezh2 was specifically ablated in Treg cells, certain genes were
upregulated in a similar manner to that found when Foxp3 was deleted in
Treg cells, and a large fraction of them were characterized by H3K27me3
modification in wild-type activated Treg cells.70 These findings suggest that
Foxp3 and Ezh2 cooperatively repress some genes by induction of repressive
histone modifications (Fig. 3).
Figure 3 Foxp3-mediated induction of repressive histone modification. Foxp3 mainly

serves as a transcription repressor, targeting genes that are normally upregulated by
TCR/costimulation. One mechanism of Foxp3-dependent gene repression in activated
Treg cells is the induction of repressive histone modification, H3K27me3 by recruiting
Ezh2-containing polycomb complex PRC2.
3.5 Nucleosome Positioning

Nucleosomes are a subunit of chromatin made up of a histone octamer and
DNA wrapped around it. Chromatin remodeling enzymes can slide nucle-
osomes, remove the histone octamer or loosen the DNA around histones, in
an ATP-dependent manner.71 The consequence of these epigenetic events
is the alteration in the exposed region of the genome, which changes the
accessibility to transcription factors. One method to assess nucleosome posi-
tioning is the examination of DNase I hypersensitivity (DHS) sites, taking
advantage of the fact that nucleosome-free regions can be cleaved by DNase
I. For example, active enhancers are bound by a number of transcription fac-
tors and are characterized by high DHS.
The nucleosome positioning in naı̈ve CD4+ T cells and Treg cells are
largely similar, but there are some limited DHS regions specific to Treg cells
(less than 1% of all DHS sites).31 These regions are located near Treg signa-
ture genes, such as Foxp3, Ctla4, and Ikzf2, suggesting that the key mole-
cules that define Treg cell lineage are marked with Treg cell-specific
epigenetic modifications to ensure their stable expression. The overlap of
gene sets with TSDRs and Treg cell-specific DHS regions suggest that a
common mechanism may regulate these two epigenetic processes. Given

that these genes are highly associated with Treg cell function and identity,
these epigenetic modifications may ensure their stable expression by pro-
moting the binding of transcription factors.
In terms of the interaction between transcription factors and nucleosome
positioning, Foxp3 does not have a profound effect. Foxp3 binding sites
mostly show an open chromatin structure both in naı̈ve CD4+ T cells
and Treg cells, indicating that Foxp3 does not have to modulate chromatin
structure in order to bind to its targets.31 Instead, Foxp3 binds to regions that
are already bound by other transcription factors; in some regions, Foxp3
binds to where Runx1 is bound and cooperatively regulate the gene expres-
sion, while in other regions, it replaces Foxo1 and initiates Treg cell-type
gene regulation.
4. CROSS TALK BETWEEN FOXP3-DEPENDENT GENE

REGULATION AND TREG CELL-TYPE EPIGENETIC
MODIFICATIONS
Treg cell-type gene regulation involves both Foxp3-dependent tran-
scriptional programs and epigenetic modifications. Both factors contribute
cooperatively to the regulation of some genes, while in other cases Foxp3
is required but epigenetic modifications are not, and vice versa. As examples
of the former, Foxp3 and Ctla4 gene upregulation is ensured by DNA
demethylation and open chromatin structure of their enhancers, to which
Foxp3 binds and promotes transcription. In contrast, as examples of the latter
scenario, Ikzf4 and Ikzf2 gene upregulation in Treg cells occurs indepen-
dently of Foxp3 expression but are associated with DNA demethylation
and DHS, and the repression of Il2 is dependent on Foxp3 expression but
not associated with Foxp3-independent epigenetic modification.28,31,48,59
Genome-wide analyses of gene expression, Foxp3-binding sites, and
TSDR demethylation further revealed that indeed there is a division of labor
between Foxp3 and TSDR demethylation. Foxp3 acts predominantly as a
transcriptional repressor after TCR stimulation, whereas TSDR demethyl-
ation is associated with gene upregulation before activation.25,26 This is con-
sistent with the observation that Foxp3-deficient Treg cells express most of
the Treg hallmark genes at steady state, yet they cannot repress the expres-
sion of proinflammatory cytokines such as IFNγ and IL-17, especially under
inflammatory conditions.30,48 This suggests that in a simplified model, Treg
cell-type chromatin landscape sets the environment in which general or
Figure 4 Cross talk between transcriptional and epigenetic regulation for the genera-
tion of Treg cell-type gene expression. Foxp3 expression is induced and maintained by
both transcription factors and epigenetic modifications. Foxp3 then generates some
Treg cell-type gene expression, which includes the upregulation of Foxp3 cofactors.
There are also genes expressed independently of Foxp3, but associated with Treg
cell-specific epigenetic modifications. Some of these genes are further upregulated
by Foxp3, and serve as Foxp3 cofactors. Foxp3 and its cofactors then cooperatively reg-
ulate more gene expression.
T cell-specific transcription factors can induce the expression of genes

upregulated in Treg cells, while Foxp3 acts later, predominantly preventing
the activation of effector T cell differentiation programs and maintains Treg
cell identity (Fig. 4).
5. TREG CELL DEVELOPMENT

The transcriptional and epigenetic features of Treg cells described
above are introduced during the development of Treg cells. The highly spe-
cific and stable nature of Treg cell-type epigenetic features, interacting with
the transcriptional networks allows the irreversible commitment of progen-
itor cells into the Treg cell lineage. Then the important questions are what
kind of signals trigger Treg cell development and what kind of molecular
mechanisms are involved in interpreting such signals and coordinating tran-
scriptional and epigenetic changes?
Treg cells are broadly divided into two subpopulations, based on their
site of origin. The majority develops in the thymus and is referred to as
thymus-derived Treg (tTreg) cells, while some Treg cells also differentiate
from conventional CD4+ T cells in the peripheral lymphoid organs as
peripherally induced Treg (pTreg) cells (Fig. 5). tTreg cells, particularly
those that develop during neonatal period, are nonredundantly required
for the establishment of self-tolerance.72 In contrast, pTreg cells are predom-
inantly found in mucosa-associated lymphoid tissues such as Peyers’ patches
and lamina propria of small and large intestines, and are involved in the
induction of immune tolerance to commensal microbes and nonpathogenic
environmental antigens, such as food antigens. Moreover, pTreg cells,
which exist only in placental mammals, appear to play roles in the establish-
ment of maternal–fetal tolerance.73 Therefore, tTreg and pTreg cells have a
division of labor in some scenarios. However, in terms of their
Figure 5 Thymic and peripheral development of Treg cells. Thymus-derived Treg

(tTreg) cells develop from progenitors to tTreg cells, through tTreg precursor cells in
the thymus, dependently on IL-2 availability. In contrast, peripherally induced Treg
(pTreg) cells differentiate from conventional T cells at mucosa-associated lymphoid tis-
sues, and this is facilitated by TGFβ and butyrate produced by tolerogenic DC and com-
mensal microbes, respectively.
transcriptional and epigenetic profiles, these two subpopulations of Treg

cells show high levels of similarity, with some exceptions such as Nrp1, Itgb8,
and Ikzf2, which are expressed primarily in tTreg cells.74–77 Both of them
are stably committed to the Treg cell lineage with similar suppressive capac-
ity, yet they may exert immune regulatory function at different locations
and/or timing, which could explain the need for two distinct developmental
pathways. As expected from the different environments that they develop in,
different, but overlapping molecular mechanisms are involved in these two
developmental systems.
5.1 Signals Involved in Treg Cell Development

Current understanding of tTreg cell development is based on the two-step
model where the first step generates tTreg precursor cells
(CD4+CD8 CD25+Foxp3 thymocytes) by agonistic TCR/costimulation
and the second step converts them to tTreg cells by IL-2 stimulation.78 Dur-
ing thymocyte selection, interaction with self-antigens presented in the
medulla measures the self-reactivity of TCRs and determines the fate of
individual thymocytes. In general, weak interaction with self-antigens
selects conventional thymocytes, strong interaction, indicative of thymo-
cytes being highly self-reactive, causes apoptosis, and tTreg cell develop-
ment comes in between. The requirement of agonistic TCR/
costimulation for tTreg cell generation is clearly demonstrated by the lack
of tTreg cells in foreign antigen-specific TCR transgenic mice, but the
enhanced tTreg cell development in double transgenic mice where cognate
antigens and corresponding TCRs are transgenically expressed (reviewed in
Ref. 79). These findings indicate that development of tTreg cells requires
stronger TCR/costimulation than that positively selects conventional thy-
mocytes. In support of this notion, blockade of CD28-mediated
costimulation by CD28 deletion severely reduces tTreg numbers but not
so much conventional CD4SP thymocytes.80,81 Consequently, tTreg cells
possess relatively self-reactive TCR repertoire, which enables them to effi-
ciently suppress immune response against self-antigens. However, given the
relatively high level of TCR/costimulation that developing Treg cells are
subject to and the proapoptotic nature of Foxp3 protein, tTreg cell devel-
opment occurs at the verge of death and requires other signals to diverge
from apoptosis.82 Particularly, IL-2 signaling is critical for survival and
expansion of Treg cells and the expression of high-affinity IL-2 receptor,
CD25, prior to Foxp3 expression in tTreg precursors is necessary to be
protected from the proapoptotic effect of Foxp3.82
In contrast, pTreg cell differentiation is more dependent on the exposure

to nonpathogenic environmental antigens. First, persistent low dose-
antigenic stimulation, instead of strong TCR/costimulation required for
tTreg cell development, is favorable for pTreg cell generation, as evident
from the induction of pTreg cells when cognate antigens were orally admin-
istered in corresponding TCR transgenic mice but not when antigens were
injected with adjuvant.83 This oral tolerance is a pivotal mechanism to pre-
vent food allergy. Second, pTreg cell generation in the gut-associated lym-
phoid tissues is dependent on the interaction of immune system with
commensal microbes. This is evident from the fact that maintaining mice
under germ-free conditions reduces colonic Treg cell numbers and inocu-
lation of a particular species of microbiota increases them.84 At mucosa-
associated lymphoid tissues, the immune system requires appropriate
immune regulation, so that immune responses are elicited toward patho-
genic microbes but are suppressed to tolerate commensal microbes and non-
pathogenic environmental antigens. As a mechanism of the latter,
commensal bacteria, such as Clostridia, generate a fermentation product,
butyrate, which induces pTreg cells.85,86 Furthermore, antigens found in
the mucosa modulate the function of dendritic cells (DCs) to tolerogenic
type and this then facilitates pTreg cell development. Tolerogenic DCs
express a lower level of costimulatory molecules such as CD80 and
CD86 compared to mature DCs, making it less likely to drive effector
T cell differentiation programs. They also express immunosuppressive cyto-
kines such as TGFβ, IL-10, and retinoic acids, which potentiate pTreg cell
conversion.87 While there are distinct molecular pathways involved in tTreg
and pTreg cell development, the common signals include IL-2 signaling and
TCR/costimulation.38,88 Thus, Treg cell development occurs by combina-
tion of signals that are available in different settings.
5.2 Transcription Factors Involved in Foxp3 Induction

Given the indispensable roles of Foxp3 in Treg cell function, Foxp3 induc-
tion is the central event during Treg cell development and therefore what-
ever induces Foxp3 expression holds the key for Treg cell lineage
determination. However, if there should be one transcription factor specif-
ically appointed to induce Foxp3 expression, the next question is what
induces this transcription factor and then what upstream factors induce this
further factor and so on, a question that might continue until identifying the
initial lineage-specifying factor. Given the highly specific expression of
Foxp3 in Treg cells and the close association of Foxp3 with Treg cell
function, it is plausible that Foxp3 is the ultimate lineage-specifying factor

for Treg cells. In this case, multiple transcription factors may be involved in
Foxp3 transcription in a nonredundant manner, reflecting the need for var-
ious requirements, such as appropriate TCR/costimulation and IL-2 avail-
ability, to be met for Treg cell development. In this section, transcription
factors currently known to be involved in Foxp3 transcription are discussed.
First, several transcription factors downstream of TCR/costimulation
are required for Foxp3 induction. This is expected as agonistic TCR stim-
ulation is essential for Treg cell development and the deletion of CD28 also
severely diminishes the efficiency of thymic Treg cell development.79 cRel
is a transcription factor of NFκB family, downstream of TCR/
costimulation. It is involved in the initiation of Foxp3 transcription and
often referred to as the pioneer factor for tTreg cell development. cRel
knockout mice show a severely reduced frequency of Treg cells in the thy-
mus, whereas enhanced NFκB signaling pathway in mice that transgenically
express constitutively active IKK-β kinase have an increased number of
Foxp3+ cells among both CD4SP and CD8SP thymocytes.89,90 Mechanis-
tically, one study shows that cRel is capable of binding to the Foxp3
enhancer, CNS3. CNS3 shows permissive histone modification,
H3K4me1, indicating that the enhancer is at a poised state, at least from
the double positive (DP) thymocyte stage. A possible scenario is that cRel
activation and nuclear translocation upon agonistic TCR/costimulation
leads to activation of CNS3.58 In fact, CNS3 knockout mice shows impaired
tTreg cell development, similar to cRel knockout mice, supporting this sce-
nario. Another study observed cRel binding to CNS2, even when its CpG
residues are methylated, in T cell lines and primary CD4+ T cells, but not in
non-T cell lines or DP thymocyte cell lines, suggesting that cRel may also be
involved in DNA demethylation of CNS2 and enhancement of Foxp3 tran-
scription.90 However, cRel is not exclusively expressed in tTreg precursor
cells and tTreg cells, and it is unclear at present whether cRel binding to
Foxp3 enhancers occurs specifically in developing tTreg cells, and if so, what
directs cRel to these regions and whether cRel binding is sufficient to acti-
vate Foxp3 transcription. Another report has demonstrated the ability of
cRel and another subunit of the NFκB family, p65, to bind to the Foxp3
promoter during in vitro TGFβ-dependent Foxp3 induction in peripheral
naı̈ve CD4+ T cells.89 cRel ablation also hampers in vitro induction of Foxp3
expression in this system, partly by impairing endogenous IL-2 production
and by preventing the formation of Foxp3 enhanceosome at the Foxp3 pro-
moter.89,91 This Foxp3 enhanceosome contains transcription factors
involved in Foxp3 transcription such as Smad3, CREB, and NFAT, in addi-

tion to cRel/p65 and because some of these factors first bind to enhancers
CNS1 and CNS2, cRel/p65 may serve as the base of enhanceosome forma-
tion and local chromatin looping to connect enhancers and promoter.89
Orphan nuclear receptors, Nr4a family members, are also involved in
Foxp3 induction. These transcription factors are upregulated upon TCR
stimulation and Nr4a1 and Nr4a3 are redundantly essential for Foxp3 induc-
tion, as clearly shown by the complete lack of Foxp3+ cells both in the thy-
mus and periphery of Nr4a1/3 double-knockout mice.92 There is a binding
site for Nr4a family members in the Foxp3 core promoter region and at least
Nr4a2 has been shown to bind to Foxp3 promoter and CNS1.93 In vitro
Nr4a2 overexpression alone is sufficient to induce Foxp3 expression in con-
ventional CD4+ T cells, indicating it has the ability to activate Foxp3 pro-
moter. However, Nr4a family members are expressed immediately after
general TCR stimulation in conventional CD4+ T cells and such expression
is insufficient to activate Foxp3 transcription. Interestingly, Nr4a1 is not only
required for Foxp3 expression but also involved in promoting negative
selection and its expression level reflects TCR stimulation strength.94,95
Thus, as a sensitive interpreter of TCR stimulation during thymocyte selec-
tion, the expression level and the activity of Nr4a family members may con-
tribute to the fate decision among conventional T cell selection, Treg cell
development, and negative selection.
Foxo1 and Foxo3 are another set of transcription factors regulated by
TCR/costimulation and redundantly required for Foxp3 expression.
Foxo1/3 double-knockout mice have severely reduced Treg cell percent-
ages and the remaining Treg cells are not functional.96 These factors bind
to Foxp3 promoter, CNS2 and CNS3, regulating Foxp3 transcription.
Notably, Foxo proteins are exported out of the nucleus when phosphory-
lated by PI3K–Akt signal, which is activated by prolonged TCR/
costimulation.96 This suggests that Foxo proteins act as another sensor of
TCR/costimulation, setting a window of TCR/costimulation duration that
allows Treg cell differentiation.
IL-2 signaling is also required for both tTreg and pTreg cell develop-
ment, and the responsible downstream transcription factor is STAT5. Stat5
deletion results in severely impaired development of Foxp3+ Treg cells both
in the thymus and periphery, while transgenic expression of a constitutively
active form of STAT5 increases Treg cell percentages.97 Furthermore, the
ability of IL-2 signaling to induce Foxp3 expression from tTreg precursor
cells in vitro indicates the direct involvement of IL-2–STAT5 pathway in
Foxp3 transcription.78 Mechanistically, STAT5 binds to Foxp3 promoter

and CNS2 and likely activates these cis-regulatory elements.97,98
Furthermore, there is a group of transcription factors that are expressed in
the T cell lineage and are involved in Foxp3 transcription. Ets1, Bcl11b, and
Bach2 fall into this category; these are not specifically induced during Treg
cell development but binds to cis-regulatory elements of Foxp3 gene and
their genetic ablation results in impaired generation of Foxp3+ Treg
cells.40,99,100 Given their roles in the development and maintenance of Treg
progenitors, it is possible that these factors are involved in the pre-
establishment of chromatin landscape in which transcription factors down-
stream of TCR and IL-2 signaling can work.
In addition, some transcription factors are required only for pTreg cell
development. Reflecting the involvement of TGFβ signaling in pTreg cell
development, deletion of either Smad2, or Smad3, transcription factors
downstream of TGFβ signaling, does not affect tTreg cell development effi-
ciency, but impairs TGFβ-dependent Foxp3 induction from peripheral
naı̈ve CD4+ T cells.101 In concordance with this finding, the deletion of
Foxp3 CNS1 region or Smad3 response element within CNS1 results in
the specific reduction in Treg cells within gut-associated lymphoid tissues,
which are enriched with pTreg cells.101,102 Although TGFβ signaling itself
is also required for tTreg cell development, these results indicate that the
activation of CNS1 enhancer by Smad3 is required for pTreg, but not tTreg
cell development.103
In summary, there are a number of transcription factors reported to be
required for Foxp3 induction. Most of them fit with the required signals,
such as TCR/costimulation and IL-2 signaling, suggesting that only when
all the required signals are provided, the downstream transcription factors
cooperatively activate Foxp3 transcription.
5.3 Induction of Epigenetic Modification During Treg Cell

Development
Along with Foxp3 induction, Treg cell-specific epigenetic modifications
occur during Treg cell development. CD25+Foxp3+ CD4SP thymocytes
are already fully committed to Treg cell lineage and are capable of
maintaining stable expression of Foxp3.54 This is at least partially due to
the DNA demethylation of CNS2, taking place at this stage.104 pTreg cells
in vivo are similarly demethylated at the CNS2 region.74 In terms of the mol-
ecules that induce these changes, it is not as well understood as transcription
factors that regulate Foxp3 expression. It was only in 2009 that enzymes
Another random document with
no related content on Scribd:
magnificently staged; livelier pieces such as Clorise, by Baro, a very
popular play-writer; other fashionable plays; ballets in which young
Royalties danced—Mademoiselle, Gaston’s daughter, Mademoiselle
de Bourbon, Mademoiselle de Longueville, Mademoiselle de
Vendôme, the Duc d’Enghien; his future wife Mademoiselle de
Maillé-Brézé, and other nieces and cousins of the Cardinal. These
gay fantastic ballets, even more than regular plays, were the delight
of society, young and old. All the courtiers and great ladies joined in
them; Louis XIII. himself often composed both the words and the
music of lutes, spinets, violins, and forgot his gloomy stiffness in
dancing.
In the intervals of the performances the Cardinal’s guests enjoyed
rare fruits and dainty sweetmeats, handed round by his pages in
baskets tied with English ribbons of gold and silver tissue. When
comedy and dance were over the company was offered a gorgeous
supper on the great service of plate which the Cardinal left to the
King.
The entertainments at the Palais-Cardinal reached their zenith in
January 1641, with the representation of Mirame. Richelieu, to quote
a contemporary, “témoigna des tendresses de père pour cette
pièce”; and it seems actually to have been in great part his work, in
collaboration with the academician Desmarets. The larger of his two
theatres, holding three thousand persons, was used for the first time
and decorated with special magnificence. It was rather a vast saloon
than a theatre, with gilded galleries for the most distinguished
guests; the ordinary admiring crowd finding place on the floor. His
Eminence, happy and triumphant, was near the Queen: the Abbé de
Marolles, once a timid student, now a critical spectator, describes
him as dressed in a long mantle of flame-coloured taffeta over a
black soutane, with collar and facings of ermine.
The scenery of the play, with the new machinery which astonished
all eyes, had been ordered from Italy by Cardinal Mazarin, now a
familiar figure in Paris and Richelieu’s right hand. There was a long
perspective of palaces and gardens, with terraces, grottos, fountains,
statues, all looking out over the sea, “with agitations,” says the
Gazette, “which seemed natural to the waves of that vast element,
and two large fleets, one appearing two leagues distant, both of
which passed in sight of the spectators.”
Over this lovely scene night gradually fell, and all was lit up by the
moon. Then, just as naturally, day dawned and the sun rose, taking
his turn in this “agréable tromperie.”
The majority of the guests were amazed and transported beyond
measure. A few critics, among whom was the Abbé de Marolles, did
not particularly care for all this “fine machinery and grand
perspective.” He found it fatiguing to the eyes and the mind: in his
opinion a comedy should depend for success on story, poetry, and
fine acting. “Le reste n’est qu’un embarras inutile.”
There were other more malicious critics who saw in the story of
the play—the love of Princess Mirame, daughter of the King of
Bithynia, for the daring sailor Arimant, commanding the fleet of
Colchos, with all the tragical events which at last brought about a
happy ending—a veiled allusion to the old romance of Queen Anne
and the Duke of Buckingham. It is very improbable, to say the least,
that Richelieu, who had at this time ceased to persecute the Queen,
should choose to offend her afresh by stirring up grievances fifteen
years old. His object, never indeed attained, was to live at peace
among princes and nobles who had learnt their lesson. What really
annoyed him in connection with this performance of Mirame was the
discovery by his watchful enemies of various disreputable persons
among the invited guests. The King was displeased; Monsieur
enjoyed the incident; and the Cardinal could only revenge himself on
an unlucky official who had been too free with his cards of
admittance.
In spite of fault-finders Mirame was a triumph. Standing up in his
place, the Cardinal joyfully acknowledged the constant thunders of
applause, then waving his hand for silence, that none of his fine lines
might be missed. When the play was over, and the Queen had
passed on a golden bridge drawn by peacocks to a silver throne
prepared for her beyond the lifted curtain of the stage, to preside
over a grand ball that ended the evening, there was no prouder man
in Europe than her host—the weary, sickly statesman who had
already given provinces to France and made her paramount in Italy
and Spain.
CHAPTER IX
1633-1637
Conquests in Lorraine—The return of Monsieur—The fate

of Puylaurens—France involved in the Thirty Years’ War—
Last adventures of the Duc de Rohan—Defeat, invasion, and
panic—The turn of the tide—Narrow escape of the Cardinal—
The flight of the Princes.
From the year 1630, Richelieu had employed historians and

antiquaries in hunting up documents to justify his plans for the
greater glory of France. Amazing were the pretensions that these
learned persons encouraged him to make for his King. According to
them, Louis XIII. might claim sovereign rights over England, Spain,
Milan, Naples, and Sicily, not to mention Flanders, Artois, Franche-
Comté, Lorraine, and other frontier provinces. How far Richelieu’s
dreams of conquest really extended, it is difficult to say. But the year
1633 found him resolved at least, in his own words, to “re-establish
the monarchy in its original greatness” by asserting “the ancient
rights of the Crown”; and Duke Charles of Lorraine soon gave him
his desired opportunity of annexing a large part of the old Austrasian
province.
Relying on imperial support and on his sister’s marriage with the
heir-presumptive of France, the Duke had broken treaties and had
neglected to pay homage for his French fief, the duchy of Bar. In the
summer of 1633 the Parliament of Paris was directed by Richelieu to
declare that duchy confiscated to France. In August a French army,
led by the King and the Cardinal, marched once more upon the
frontier of Lorraine.
The Duke tried to gain time, hoping for the help of a Spanish army
under the Duke of Feria, which was advancing from Italy. He sent his
brother, Cardinal Nicolas-François, to negotiate with the French,
offering not only to consent to the dissolution of his sister’s marriage,
but that the Cardinal, who had taken only minor orders, should ally
himself with Richelieu by marrying Madame de Combalet. This
proposal was coolly put aside by Richelieu, who observed that he
had not advised the King to enter Lorraine with a powerful army for
his private family ends. He insisted that Nancy, the capital, with
Princess Marguerite in person, should be placed in the King’s hands
as a pledge of submission.
As to his sister, Duke Charles was willing enough, being painfully
aware that the alliance with Gaston was a mistake which might ruin
him; but he would not consent to surrender his capital, protesting,
with oaths, that he would rather burn it down. Nevertheless, the city
did not stand a long siege; but when Louis XIII. and Richelieu made
their entry, their promised captive had escaped. By the help of her
brother the Cardinal, and with great spirit and courage on her own
part, Madame Marguerite had slipped out of Nancy at the beginning
of the blockade, and in a page’s disguise had joined her husband at
Brussels. There she was formally received as Duchess of Orléans by
the Queen-mother and the Infanta, and the marriage was confirmed
by the Archbishop of Malines.
Richelieu was not altogether displeased. Well convinced of his
power to separate Monsieur from his new wife as soon as the Prince
himself should return to France and his duty, he was not sorry to
have an honourable excuse for going to extremes with the Duke of
Lorraine. No hostage, no capital. Duke Charles was helpless; his
sister was no longer in his hands; his Spanish allies, checked on
their way by a Protestant army, failed to come to his aid. He had to
see a parliament established in Metz and almost the whole of his
province garrisoned by French troops. When the King returned to
Paris the lilies of France were flying over Lorraine. Town after town
submitted, fortress after fortress. In January 1634 Charles abdicated
for the time in favour of his brother the Cardinal, and with the small
remains of his army took service under the Emperor.
Then Cardinal de Richelieu bent all his energies to forcing on
Gaston’s return to France and reconciliation with his brother. He
regarded this as a necessity of State, and he was equally resolved
that the Queen-mother, who had made some overtures on her own
account, should never again set foot in France. Both Marie and
Gaston, while quarrelling between themselves, played the Minister’s
game by their own foolishness. A murderer, caught at Metz, was
suspected with reason of being sent from Brussels by Chanteloube,
Marie’s unwise counsellor, to attempt the life of Richelieu: he lost his
own. The same fate befell others, in Lorraine and elsewhere,
charged with the same designs; and while this secret campaign went
on, Gaston and his favourite Puylaurens made an independent treaty
with Spain, promising to invade France with a foreign army to be
supplied by the Imperial generals in the Low Countries.
Well served by spies, Richelieu knew all this. He replied to
Monsieur’s treason by representing to the King that such a prince,
who could promise French fortresses to the enemy, was not fit to
wear the crown; and with a bold decision before which, at such a
crisis, not even the hereditary monarchy was sacred, he proposed a
league of nobles and princes of the blood who should pledge
themselves, in case of Louis’ death, against the unconditional
succession of his brother. France after all, in the eyes of Richelieu,
was greater than her kings.
By the autumn of 1634 Puylaurens and his master knew that they
had made a huge mistake in allying themselves with Spain. No
troops were forthcoming, and it began to be evident that the prospect
was not one of triumph and revenge, but of ruin and perpetual exile.
All through September M. de Puylaurens was negotiating secretly
with Cardinal de Richelieu, promising for Monsieur, among other
things, the renunciation of his marriage, and also making a good
bargain for himself.
Gaston left Brussels one day in October, and galloped hard to the
frontier. He had been an exile for two years, and was enchanted to
see France again. His little daughter, Mademoiselle, now seven
years old, met him at Limours, and flew joyfully into the arms of a
gay and fascinating father.
As to Madame, left behind in Flanders, her marriage was solemnly
declared null and void by an assembly of French clergy, as having
been contracted against the civil law. In this decision, however, the
clergy acted on Gallican lines, independently of the Pope, who was
of a different opinion; and although, after long resistance, Monsieur
formally submitted, he had protected himself in advance by a letter to
Urban VIII. refusing to be bound by any extorted promise. The
consequence was, that Richelieu’s apparent triumph in this affair of
the Lorraine marriage only lasted his life. Gaston and Marguerite
remained faithful to each other; and the stiff Madame who reigned in
after years at Blois and at the Luxembourg was the same Princess,
the heroine, in her adventurous girlhood, of a secret marriage and a
romantic escape.
It was that private letter of Gaston’s to the Pope which brought
about the ruin of the unlucky Puylaurens. He had gained high favour
with Richelieu, who had purchased his faithful service, as he
thought, by making him a duke and a peer of France and by
marrying him to his own first cousin, Mademoiselle Philippe de
Pontchâteau, younger daughter of his aunt, Louise du Plessis, his
father’s sister. The marriage took place in Paris at the end of
November 1634, and on the same day the Duc de la Valette, son of
the Duc d’Épernon and widower of Henry IV.’s daughter,
Mademoiselle de Verneuil, was married to the elder sister, Marie de
Pontchâteau, and the Comte, afterwards Maréchal, de Guiche to
another cousin, Mademoiselle du Plessis de Chivray. The Cardinal
celebrated the triple wedding by a magnificent fête. At this time the
first nobles in France found it politic to quarrel for the honour of his
alliance, and it was matter of general talk in society that he meant to
marry Monsieur to Madame de Combalet, the Lorraine marriage
being set aside. This report even reached the ears of Monsieur’s
little daughter, and filled her with just indignation.
A few weeks after the wedding the Cardinal’s spies brought him
not only the secret, well kept by Puylaurens, of Monsieur’s letter to
Rome, but proofs of a fresh treasonable correspondence carried on
by the new Duke with Spain. Swiftly fell Richelieu’s vengeance.
Puylaurens, with several of his friends, was arrested at the Louvre
on February 14, and carried off by royal order to Vincennes. The
entreaties of Monsieur, newly reconciled at Court, delayed his trial,
but he died after four months of prison. “His good fortune,” says
Richelieu, “withdrew him from this world, and saved him from the
infamy of a shameful death, which he could not have escaped.”
Whether the fatal atmosphere of the dungeons of Vincennes was
assisted by poison of a more active kind, will never be known. That
suspicion hung about the deaths of many of the Cardinal’s prisoners.
Richelieu consoled the young widow of Puylaurens by marrying her
to the Comte d’Harcourt, of the House of Lorraine, younger brother
of the Duc d’Elbeuf, a queer personage, but a fine soldier. He had
fought a successful duel with Bouteville, in itself a distinction. He
proved himself worthy of the Cardinal’s favour by serving His
Eminence faithfully for the rest of his life.
But for Richelieu, the Thirty Years’ War might have ended with the
death of Wallenstein and the imperial victories which followed it.
Even the Protestant princes of Germany were ready for a
compromise with the Emperor. But Richelieu had no intention of
accepting a general peace which would leave his Swedish friends
weak and dissatisfied, his own conquests incomplete, Spain and
Austria easily predominant in Italy and the Low Countries. He
resolved that France, as an ally of Sweden, Holland, and the
German Protestants, should now take an active part in the war, and
he prepared for the actual declaration by a treaty with the Dutch for
the partition of the Spanish Netherlands, to be followed by one with
the Dukes of Savoy, Parma, and Mantua, for the conquest and
division of the Milanese.
In May 1635, after some military provocation on the part of Spain,
Louis XIII. sent his herald-at-arms to Brussels—a noble Gascon,
Jean Gratiollet, Captain of Abbeville—and solemnly declared war
against his brother-in-law, Philip IV., while publicly inviting the Low
Countries to rebel against Spain. “Europe was amazed,” says a
modern French writer, “to see Richelieu suddenly take up arms for
those same Huguenots whom he had crushed with such good will at
La Rochelle.”
Europe was amazed: and what of the French nation, flung
unconsulted into the struggle with Catholic Europe which might
easily have become a fight for its own existence? The three Estates
of the realm had each its own separate point of view. The princes
and nobles loved war; but the majority, Catholic and hating
Richelieu, were rebels at heart. However, each man had his orders:
content or malcontent, each governor found himself dispatched to
his own province, each commander to his post, while generals
dashed hither and thither in pursuit of armies which had to be hired,
recruited, disciplined, poured in half-a-dozen directions over the
frontier—Germany, Flanders, Lorraine, Switzerland, Italy. Richelieu,
the directing brain, at this moment of high energy, moved the
members even against their will.
To most of the clergy, again, the war was of the nature of
sacrilege; and still more so, later on, the demand of an enormous
payment of arrears for lands held under the Crown, which had been
suffered to go free for nearly a hundred years. But at a time when
the taxes of France had rolled up to more than a hundred million
francs a year, a gigantic and as yet unheard-of sum, Richelieu could
no longer grant the clergy the privilege of paying no tax but their
prayers, which he had himself claimed for them at the States-
General of 1613.
“The people give their goods, the nobles their blood, the clergy
their prayers.” As ever, the patience of the most heavily taxed
seemed almost inexhaustible; and it was not till France was deeply
engaged in the war, her middle class and her peasantry crushed by
Richelieu’s intendants and financiers under burdens every week
more enormous, that in the south and the north populations made
some effort to save themselves; made it by rioting, their only
resource, and found themselves—Croquants in Guienne, Va-nu-
pieds in Normandy—in a last state worse than the first.
In spite of all these discontents there were ways in which
Frenchmen now realized the national unity which was Richelieu’s
dream. The famous leader, Duke Henry de Rohan, was again in
arms, not now as a Huguenot chief, but commanding an army
against the Duke of Lorraine, fighting for his duchy with imperial
troops behind him. In the spring of 1635, it was to Rohan that
Richelieu committed the task of preparing for his designs on Milan by
a new occupation of the Valtelline, thus once more playing the old
game of blocking the chief military road between Austria and Spain.
All went well at first, the Duke proving himself a loyal subject and a
good general. The cause that finally discomfited him and drove him
at last to throw up his command and to retire to Geneva was the
failure of Richelieu’s government to pay a promised indemnity to the
Grisons, rightful possessors of the valley, who after two years’
French occupation, secretly encouraged by Spain, rebelled suddenly
against Rohan and insisted on the evacuation of their territory.
Blamed by Richelieu for a failure which was no fault of his, and
broken by severe illness, the Huguenot hero was still ready to bear
arms for France. In the spring of 1638 he volunteered to serve under
Duke Bernard of Saxe-Weimar—the great soldier who, if actually
fighting for his own hand, nevertheless gave Alsace to France—and
died of his wounds after the siege of Rheinfeld, having lived long
enough to know with what swift brilliance Bernard had turned defeat
into victory.
For many months, as readers of history know, the fortune of war
went against Richelieu. The ravages of the French and the Dutch
armies in the Netherlands, under the Prince of Orange and the
Marshals de Châtillon and de Brézé, did not incline the population to
change masters. In Germany, one town after another fell into
imperialist hands, and it was only with difficulty that the French held
their own in Lorraine. The invasion of the Milanese failed; and later
on the deaths of the Dukes of Savoy and of Mantua deprived France
of two important allies.
The French fleet, though making a fine show for those days—
forty-seven men-of-war—wasted its strength in vainly flourishing
about the coast; and owing to the quarrels of its commanders, the
Comte d’Harcourt and the Archbishop of Bordeaux, with M. de Vitry,
governor of Provence—the slayer of Concini—did not for a long time
succeed in even recovering the Isles of Lérins, seized by Spain at
the opening of the war.
And then, in July 1636, a terrible disaster threatened France.
Imperial troops crossed the frontier, and had taken two strong places
in Picardy, La Capelle and Le Catelet, before the French
commanders were ready to oppose them. Imperial cavalry crossed
the Somme and advanced to the Oise, the Comte de Soissons
retreating before them, and spread a very natural terror throughout
the country. They were mostly Croats and Hungarians, fierce and
savage men, whose road was marked by robbery, fire, and
slaughter. Their leader was the Bavarian, John of Werth, a name of
fear in the campaigns of his day.
Paris was in a state of terror and fury. The black shadows of the
streets, in the sweltering heat of late July and early August, were
loud with raging men and women, whose voices taught the Cardinal-
Duc his unpopularity. Paris was ill fortified, ill defended, and part of
her strong old walls had been destroyed by him for the sake of his
Palais-Cardinal. They cried against him because of that; because of
his ingratitude to the Queen-mother, his failure, so far, in the war he
had undertaken, his alliance with heretics. And Richelieu knew that
their fear, if not their hatred, was too well justified. The Comte de
Soissons, whose army, camping in the forests and holding the fords
of the Oise, protected Paris, was not above suspicion as to his
loyalty; the Duc de Chaulnes, governor of Picardy, was lazy and
negligent; money and men were lacking for the defence of a divided,
discontented, panic-stricken country.
The first news of the invasion found the King and the Cardinal
absent from Paris as usual in the heat of summer. They returned at
once to the stifling, frantic city.
Then “the great Armand” showed the stuff he was made of.
“Remember, I pray you,” he wrote to the Comte de Soissons, “on
such occasions as these, moments are worth years.” Paris being
always and before all things a Catholic city, he appealed to her
religion. All the bishops in the kingdom were commanded to hold
processions within and without their cathedrals, with the special
devotions of the Forty Hours. From every church in Paris and in the
whole of France, with every chapel of convent or monastery, the
bells clanged out, calling the faithful to pray for their country. In his
own person, the Cardinal vowed to the Paris convent of the Filles du
Calvaire, in the Marais, Père Joseph’s favourite foundation, a large
sum of money and a silver lamp to burn perpetually before Our
Lady’s altar.
Whatever his own personal faith may have been, he knew the
spiritual needs of the people. That he did not fear their angry voices
he proved by driving alone, “at a foot’s pace, without suite and
without guards,” through the wild crowds in the streets, from the
Palais-Cardinal to the Hôtel de Ville, bearing the royal order that the
city trades and companies should assemble for the purpose of giving
their help to the King. His courage triumphed. The people, says
Montglat, “dared not say a word to him.”
Royal decrees followed thick and fast; their succession was like
the sending round of the Fiery Cross, summoning men to serve their
country. Those Parisians who had planned to escape John of Werth
and his pillaging horde by flying with all their movable goods to
Orléans or some other city of the west, found the gates of Paris shut
against them. All privileges and exemptions were abolished in the
city. All men capable of bearing arms were ordered to present
themselves for enrolment, either at the Hôtel de Ville, where the old
Maréchal de la Force sat on the steps to receive them, or mounted
and armed at Saint-Denis. All the workshops of Paris were closed; all
building stopped; no master of a trade, excepting bakers, butchers,
armourers, gun-makers, saddlers, and the like, might keep more
than one apprentice; the rest, with masons, stone-cutters,
carpenters, artisans of every sort, must serve the King. From each
owner of a coach, a horse was demanded; and every house in Paris
was expected to furnish a man with belt and sword. The peasants of
the surrounding villages were set to work on new fortifications at
Saint-Denis.
A day sufficed to change terror into enthusiasm. On August 5
representatives of all the trade guilds and syndicates were received
by Louis XIII. in the great gallery of the Louvre, “and offered him their
persons and their goods with so great gaiety and affection, that most
of them embraced and kissed his knees.” Louis rose to the occasion
and kissed them all, not excepting the chief of the cobblers, whose
guild made the noble gift of 5,000 francs. The Parliament—not
without grudging conditions—the municipality, the colleges,
monasteries, and other bodies, poured money at the King’s feet:
there was enough to pay and keep, for three months at least, twelve
thousand foot and three thousand horse.
In the meanwhile, the news that the enemy had taken Corbie on
the Somme, thus drawing alarmingly near to Amiens, on the direct
road to Paris, fanned the flame so fiercely that “tout le jeune
bourgeois,” says Montglat, “à toute force, vouloit aller à la guerre.”
Not many days later, the King and the Cardinal advanced to Amiens,
and a strong army, commanded by Monsieur and the Comte de
Soissons, held the enemy in effectual check along the banks of the
Somme. By the middle of September, all the actual danger of
invasion was past, though the Imperialists still held Corbie. John of
Werth and his merry men, loaded with booty, had galloped back
across the frontier of Artois.
Corbie was not retaken till November, but the Cardinal Infant, his
aunt’s successor as ruler of the Netherlands, with the other Spanish
and Imperialist generals, discouraged by the advance of the French
army, had already withdrawn from French territory; and it seemed, as
the autumn advanced, as if the fortune of war was changing in
Richelieu’s favour. The enemy was repulsed everywhere: in
Burgundy, by Weimar, Condé, and the Cardinal de la Valette; on the
Spanish frontier, where St. Jean de Luz was taken, but further
advance was resisted by the old Duc d’Épernon and the Comte de
Grammont, governors of Guyenne and of Béarn; on the Morbihan
coast, where a Spanish force, disembarking near Vannes, attacked
the Abbey of Prières. The sturdy monks defended themselves so
gallantly that the country-side had time to rise against the invaders,
who fled back in disorder to their ships.
At this moment of danger, the two young men whom Richelieu had
called to the command of the King’s armies were busily plotting his
destruction. To them and their like the death of the Minister and the
anarchy that must follow were not only desirable for their own ends,
but the best medicines for the ills of France.
Monsieur and the Comte de Soissons were seldom friends, except
when they joined hands against Richelieu, and it happened that at
this time each was nursing special grievances: Monsieur, as to his
forbidden marriage and the death of Puylaurens; Soissons, because
the Cardinal had dared to offer him his niece in marriage, had
refused him the command of the army in Alsace, and more recently
had shown distrust by setting Monsieur over him as Commander-in-
Chief of the army on the Oise. There were not wanting faithful friends
who pointed out to both princes that now was the moment to
revenge themselves. The army was theirs; the Cardinal was at
Amiens; the King, staying at the Château de Demuin, a few miles
away, rode constantly into the city to hold council with his Ministers.
It was natural that the princes in command of the army should attend
the council. The rest was easily thought out, with the help of M. de
Montrésor, a follower of Monsieur, M. de Saint-Ibal, in M. le Comte’s
confidence, and two “solid men,” Varicarville and Bardouville. These
six conspirators fixed a day on which the Cardinal should be stabbed
to death after the King had left the council.
All went well for their purpose. On the appointed day, “the council
being ended, the King went away with all his guards, and the
Cardinal remained alone in the courtyard with Monsieur and the
Comte de Soissons. Immediately,” writes the Marquis de Montglat,
“Varicarville, who knew the secret, stationed himself behind the
Cardinal, expecting the signal which Monsieur was to give, while
Saint-Ibal and Bardouville took their stand, one on the right, the other
on the left. But instead of commanding that the projected deed
should be done, Monsieur, seized with fear, remounted the staircase
without a word; while Montrésor, surprised at the change, followed
him, telling him that his enemy was in his power, and that he had
only to speak.”
It was not the first time that Richelieu had owed his life to Gaston’s
temperament. So éperdu was the Prince, so utterly had his nerve
failed him, that he could only mutter something about “another time,”
and escaped as quickly as possible, leaving the Comte de Soissons,
“dans la dernière confusion,” face to face with Richelieu. Unaware of
his danger, and the King’s brother having disappeared, the Cardinal
bade his other enemy farewell and retired to his lodging. The fingers
of Saint-Ibal, Varicarville, Bardouville, relaxed on their dagger-hilts,
and one may imagine that these three gentlemen stared rather
blankly on each other as their doomed victim walked away.
When the story became known, which was not immediately, many
persons blamed the Comte de Soissons that he had not made up for
Monsieur’s weakness by finishing the affair. “He excused himself,”
says Montglat, “by the respect he owed Monsieur, so that he dared
undertake nothing in his presence without his command.” He was
too wise to act alone in such a matter: the position of Gaston’s cat’s-
paw, to be disclaimed and forsaken and left to the King’s justice, was
not attractive. The army might rally round the heir to the throne in
sudden rebellion; the Comte de Soissons was not equally secure.
Three days later there was another chance, for Richelieu visited
the camp; but he was attended by his own guards, and the
assassination was “judged impossible.” On this occasion a whisper
of the plot reached his ears, and with his usual fearlessness he
spoke of it to the Comte de Soissons, haughtily reprimanding him.
The princes were frightened, for their plots had gone beyond the
death of Richelieu. They had disloyally done their best to delay the
relief of Corbie; they had attempted to draw the Duc d’Épernon into
the project of a rising, already favoured by the Duc de Bouillon and
others, the object of which was to lay hold on the government, to
reinstate the Queen-mother, and to make peace with Spain. They
failed; the various successes of the autumn were against them; the
Duc d’Épernon, though two of his sons were on their side, refused to
listen to them. After the re-taking of Corbie, having returned from the
army to Paris, they were seized with a great fear of the Cardinal. He
was certain to know all; he was of a temper that never forgave; the
Court, they felt assured, was not a safe place for them. They took
counsel with each other and resolved to fly, at once, on a dark
November night, while Paris was singing and rejoicing over the good
news of victory.
Both princes, before leaving Paris, paid a separate visit to the
Tuileries. There, under the care of M. de Montglat’s mother, Madame
de Saint-Georges, lived Mademoiselle de Montpensier, Gaston’s
daughter, now nine years old, a person of decided character, and
one of Richelieu’s most hearty haters. The Comte de Soissons paid
great court to this little lady, the richest heiress in France if not in
Europe. Though four years older than her father and twenty-three
years older than herself, and having failed ten years earlier to run
away with her mother, he proposed to marry her, and Gaston was
ready to consent. This plan was one of the links that now united
them. Mademoiselle herself liked Monsieur le Comte, and accepted
his compliments and sugar-plums with satisfaction: but at this time
she did not understand his object.
It is doubtful if the royal consent would ever have been given to
this marriage. But a curious little passage in the Cardinal’s own
Memoirs shows how keenly he noticed every detail in the lives of the
princes, and on what slight if sure grounds he accused them of
conspiracy.
“The next day at evening, which was the night of the 19th to the
20th, Monsieur and he (M. le Comte) left Paris; and that it was
plotted between them is shown by this: Monsieur having arrived in
Paris, and visiting Mademoiselle his daughter, Madame de Saint-
Georges told him that M. le Comte had but just gone out. He leaned
his head against a chimney-piece, remained long thoughtful, then
said, and repeated several times, ‘What! Monsieur le Comte is here?
What! He has not gone to Champagne!’ Which showed plainly that
there was a plot between them.”
Disguised and almost alone, the princes retired in different
directions: Monsieur to his castle of Blois, the Comte de Soissons to
neutral ground at Sedan, held by its sovereigns of the House of
Bouillon for more than a hundred years. From these retreats they
sent their demands and remonstrances to Louis XIII., while on the
other hand they corresponded with the Queen-mother and with
Spain.
Richelieu seems to have treated the discontents of the Comte de
Soissons with some scorn. He allowed negotiations with him to drag
on for some months, and then advised the King not only to forgive
him, but to allow him to remain four years at Sedan unless he chose
to return to the Court: a leniency for which the Cardinal has been
blamed; dangerous to the State and fatal to Soissons himself.
As to Monsieur, a mixture of threats and entreaties, the advance of
royal troops to Orléans, the clever management of M. de Chavigny,
the Cardinal’s most trusted agent, soon brought about a change in
his weathercock mind. He met the King at Orléans in February 1637,
“with many demonstrations of friendship.” Indeed, “dissimulation
went so far, that there appeared to be a sincere reconciliation
between Monsieur and the Cardinal.”
CHAPTER X
1637-1639
Palace intrigues—Mademoiselle de Hautefort—

Mademoiselle de la Fayette—The affair of the Val-de-Grâce—
The birth of the Dauphin—The death of Père Joseph—
Difficulties in the Church.
In Richelieu’s own mind his worst enemies were to be found

among his nearest neighbours. “Les intrigues de cabinet,” says M.
de Montglat, “donnèrent plus de peine au Cardinal de Richelieu que
toute la guerre étrangère.” Not only mischievous great ladies like the
Duchesse de Chevreuse, but every man or woman who had
anything to do with the Court, were objects of his watchful suspicion,
and to most of them, while they begged his favour and flocked to his
entertainments, he seemed the cruel ogre, the mysterious sphinx, so
long represented in history.
He never really trusted the King. Louis was fond of gossip, easily
amused by small things, and often attracted by persons undesirable
from Richelieu’s point of view. And even at his height of power he
found it impossible to carry out the ideal arrangement which would
have hindered any one not bound to his own service from
approaching the King at times such as the petit coucher, when
intimate talk was allowed, and men might even dare to tell a story
against the Eminentissime himself. They would probably repent; for
though Louis might laugh and enjoy such jokes, he had a way of
repeating them to the Cardinal, if only with a half-childish notion of
teasing him. The consequences to a chattering courtier might be
serious.
The influence of these gentlemen with the King was seldom really
dangerous, and yet the Cardinal was justified in his distrust, for the
majority hated him, and he went about always with his life in his
hand, not because of ambitious princes alone. Men’s consciences
were no protection to him. For instance, the Abbé de Retz,
afterwards Cardinal and Coadjutor of the Archbishop of Paris, felt
little doubt that he would have done a right action, socially and
politically, had he carried out a plan for killing Richelieu in the chapel
of the Tuileries, at the long-deferred christening of Mademoiselle de
Montpensier.
Over and over again Richelieu tried to confine the King’s special
favour to persons chosen by himself, and over and over again he
failed. It was not so much that people played him false, as that he
found them—men and women—too proud, too independent, and too
faithful to their order for the place he meant them to fill—that of the
King’s favourites and his own spies. There was Mademoiselle de
Hautefort, with whom Louis fell in love when she was a beautiful girl
of fifteen, brought to Court from her native province by her
grandmother, Madame de la Flotte, and appointed one of the Queen-
mother’s maids-of-honour. After the “Day of Dupes,” when Marie de
Médicis left France and her household was broken up, Madame de
la Flotte became lady-in-waiting to the young Queen in the place of
Madame du Fargis, whom Richelieu sent into exile; and
Mademoiselle de Hautefort, transferred at the same time, was
specially recommended by Louis to his wife’s favour.
At first, very naturally, Queen Anne was not pleased. Marie de
Hautefort was in every way a dazzling person. Madame de Motteville
declares that she made a greater effect at Court than any other
beauty. “Her eyes were blue, large, and full of fire; her teeth white
and even; her complexion had the white and red suitable to a fair
beauty.” Added to this, she had a sharp tongue; she was high-
spirited, “railleuse,” and by no means soft-hearted.
Louis XIII.’s love-affairs contrast curiously with those of his father.
Nothing could be more innocent, more purely platonic, than his
devotion to Mademoiselle de Hautefort. He hardly dared approach
her; his talk was of dogs and of birds; and yet he showed the stormy
jealousy and the sulks and humours of a passionate lover, and spent
hours in writing songs and music for his lady. She disputed with him
freely and laughed at him unmercifully.
At the beginning Richelieu encouraged this singular affection. But
after about three years he saw reason to change his mind.
Mademoiselle de Hautefort was not inclined to act as his political
agent, and she had soon given the loyalty of a warm and generous
nature to her mistress, the Queen, whom she saw neglected by
Louis and subject to the tyranny of the Cardinal. This is to say that
the woman most admired by the King had joined the Spanish party
at Court and was rightly counted by Richelieu among his enemies.
It cost him little trouble to drive Mademoiselle de Hautefort out of
favour—at least for a time. When Louis had become slightly tired of
his quarrels with the fair beauty and slightly chilled by her friendship
with the Queen, it was made easy for him to find consolation in the
dark eyes of Louise de la Fayette, a cousin of Père Joseph, whose
family was supposed to be devoted to the Cardinal.
Mademoiselle de la Fayette was as good and gentle as she was
lovely; in the varied records of the French Court there exists no
sweeter figure. During two years she and the eccentric King adored
each other with a tender affection and mutual confidence quite
absent from the Hautefort affair; yet this, like the other, never passed
the bounds of friendship. It went so far, however, that the girl’s
conscience was alarmed, and she began to think of taking refuge in
a convent.
The idea was not unwelcome to Cardinal de Richelieu. The Court
was full of his spies, who warned him that Mademoiselle de la
Fayette’s intimate talk with the King was not to his advantage; that
she was inspired by Père Caussin, the royal confessor, to speak to
Louis in favour of his mother, his wife, his brother, and all the other
victims of a warlike, heretical policy; that she was encouraged by her
uncle the Bishop of Limoges and her brother the Chevalier de la
Fayette, to set him against the Cardinal; that the Bishop had even
been heard to say, “When the Cardinal is ruined, we will do this and
that. As for me, I shall inhabit the Hôtel de Richelieu.”
The Court was buzzing with intrigues all through 1636, the “year of
Corbie,” while the King still enjoyed, as far as possible, the society of
the only woman who had ever loved him for his own sake.

PDF Regulatory T Cells in Health and Disease 1St Edition Liston Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

PDF Regulatory T Cells in Health and Disease 1St Edition Liston Ebook Full Chapter

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

PDF Regulatory T Cells in Health and Disease 1St Edition Liston Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

Regulatory T cells in health and disease

1st Edition Liston

Co signal Molecules in T Cell Activation Immune

Micronutrients in health and disease Second Edition

Manganese in Health and Disease 1st Edition Lucio G

Hydrogen Peroxide Metabolism in Health and Disease 1st

Ion Channels in Health and Disease 1st Edition Geoffrey

Human Retrotransposons in Health and Disease 1st

Live cell assays from research to health and regulatory

Circadian Clocks Role in Health and Disease 1st Edition

First edition 2015

Copyright © 2015 Elsevier Inc. All rights reserved.

For information on all Academic Press publications

Transcriptional and Epigenetic

epigenetic levels. In this chapter, we focus on the components of molecular features of

in the development of a wide range of chronic autoimmune diseases. Partial

2. TRANSCRIPTIONAL REGULATION IN TREG CELLS

2.1 Foxp3-Dependent Transcriptional Regulation

to provide suppressive function similar to that of Treg cells.4 Moreover,

2.1.2 Foxp3 and Its Cofactors

Figure 1 Foxp3-dependent gene expression. Some Foxp3-dependent gene regulation

The requirement of Foxp3 to interact with its cofactors indicates that

generation of Treg cell-type gene expression, once Foxp3 is expressed, in

2.1.3 Foxp3 Posttranslational Modification

2.2 Foxp3-Independent Transcriptional Regulation

3. EPIGENETIC REGULATION IN TREG CELLS

3.1 Stability of the Treg Cell Lineage

Figure 2 Alteration in transcription factor accessibility by epigenetic modifications.

suppressive function.47 Thus, maintaining Treg cell identity requires contin-

3.2 cis-Regulatory Elements of the Foxp3 Gene

3.3 DNA Demethylation

3.4 Histone Modification

H3K4me3 modification at transcription start site increases when naı̈ve

Figure 3 Foxp3-mediated induction of repressive histone modification. Foxp3 mainly

3.5 Nucleosome Positioning

common mechanism may regulate these two epigenetic processes. Given

4. CROSS TALK BETWEEN FOXP3-DEPENDENT GENE

T cell-specific transcription factors can induce the expression of genes

5. TREG CELL DEVELOPMENT

Figure 5 Thymic and peripheral development of Treg cells. Thymus-derived Treg

transcriptional and epigenetic profiles, these two subpopulations of Treg

5.1 Signals Involved in Treg Cell Development

In contrast, pTreg cell differentiation is more dependent on the exposure

5.2 Transcription Factors Involved in Foxp3 Induction

function, it is plausible that Foxp3 is the ultimate lineage-specifying factor

involved in Foxp3 transcription such as Smad3, CREB, and NFAT, in addi-

Foxp3 transcription.78 Mechanistically, STAT5 binds to Foxp3 promoter

5.3 Induction of Epigenetic Modification During Treg Cell

Conquests in Lorraine—The return of Monsieur—The fate

From the year 1630, Richelieu had employed historians and

Palace intrigues—Mademoiselle de Hautefort—

In Richelieu’s own mind his worst enemies were to be found

You might also like