Alexander Y.

Rudensky Regulatory T cells and Foxp3

Author’s address Summary: Regulatory T (Treg) cells play central role in regulation of
Alexander Y. Rudensky1 immune responses to self-antigens, allergens, and commensal microbiota
1
Howard Hughes Medical Institute and Immunology Program, as well as immune responses to infectious agents and tumors. Transcrip-
Memorial Sloan-Kettering Cancer Center, New York, NY, tional factor Foxp3 serves as a lineage specification factor of Treg cells.
USA. Paucity of Treg cells due to loss-of-function mutations of the Foxp3 gene
is responsible for highly aggressive, fatal, systemic immune-mediated
Correspondence to: inflammatory lesions in mice and humans. Recent studies of Foxp3
Alexander Y. Rudensky expression and function provided critical novel insights into biology of
Howard Hughes Medical Institute and Immunology Program, Treg cells and into cellular mechanisms of the immune homeostasis.
Memorial Sloan-Kettering Cancer Center
1275 York Avenue, New York, NY 10065, USA Keywords: regulatory T cells, Foxp3, tolerance, inflammation
Tel.: +1 646 888 3109
Fax: +1 646 422 0453
e-mail: rudenska@mskcc.org
Introduction
Acknowledgements
The discovery that a subset of CD4+ T cells expressing inter-
I would like to thank former and current members of our
laboratory for discussions. This work was supported by a leukin-2 (IL-2) receptor a-chain (CD25) exhibits potent sup-
grant from the National Institutes of Health. A. Y. R. is an pressor activity launched its intense investigation (1). In
investigator of the Howard Hughes Medical Institute. The
addition to high amounts of CD25, these cells, dubbed regula-
author declares no conflicts of interest.
tory T cells (Treg cells), display increased levels of CD5 and
cytotoxic T-lymphocyte antigen 4 (CTLA4) in resemblance of
activated T cells (2, 3). These characteristic features led to
proposition that these cells, rather than being a T-cell sub-
lineage with dedicated suppressor function, are but a meta-
stable state of chronically activated T cells depriving other cells
of the immune system of resources, i.e. growth factors. How-
ever, the ability of transferred CD25+ Treg cells to protect
mice subjected to day 3 thymectomy from chronic inflamma-
tory lesions suggested that either the effect of their suppressor
function is long lasting or that at least some of these cells are
long lived or their progeny is persistent (4). Although initial
in vitro studies suggested that CD25+ Treg cells are ‘anergic’,
i.e. unable to proliferate and produce IL-2 upon T-cell recep-
tor (TCR) cross-linking in vitro, CD25+ CD4+ Treg cells were
shown to robustly proliferate in unmanipulated lymphore-
plete mice (5). Moreover, adoptive transfer of CD25+ Treg
Immunological Reviews 2011
Vol. 241: 260–268
cells into lymphopenic mice showed that they proliferate
Printed in Singapore. All rights reserved robustly in an major histocompatibility complex (MHC) class
II-dependent manner and their suppressor capacity increases
 2011 John Wiley & Sons A/S
Immunological Reviews despite a decline in expression of CD25 (6). In contrast to
0105-2896 CD25+ CD4 Treg cells, TCR transgenic T cells of defined spec-

260  2011 John Wiley & Sons A/S • Immunological Reviews 241/2011
 Rudensky Æ Regulatory T cells and Foxp3

ificity with activation induced expression of CD25 failed to rescent reporter proteins under control of the endogenous
suppress autoimmunity upon adoptive transfer into thymec- Foxp3 regulatory elements showed that Foxp3 protein expres-
tomized mice (7). These studies suggested that CD25+ CD4+ sion is largely restricted to a subset of T cells with suppressor
Treg cells represent a distinct long-lived thymus-derived cell function (17, 18). The majority of Foxp3+ cells were found
subset with suppressor function. within CD4+ T-cell subset. However, relatively small, but
readily detectable numbers of peripheral CD8+, CD4+ CD8+
The discovery of Foxp3 gene and its role in Treg cell and CD4)CD8) TCRab+ T cells expressed Foxp3 and corre-
differentiation sponding subsets of Foxp3-positive cells were present in the
The aforementioned findings prompted exploration of genetic thymus (17). Although the majority of Foxp3+ T cells
mechanisms underlying differentiation and function of Treg expressed high amounts of CD25, their CD25-negative coun-
cells. These studies were facilitated by the discovery of X chro- terparts were readily detectable in secondary lymphoid organs
mosome encoded transcription factor Foxp3 and its loss- and non-lymphoid tissues. Importantly, Foxp3+ cells charac-
of-function mutations in humans leading to a severe terized by either high and low CD25 or even lacking CD25
multi-organ autoimmune and inflammatory disorder immu- expression exhibited common transcriptional signature and
nodysregulation polyendocrinopathy enteropathy X-linked potent suppressor function (17).
syndrome (IPEX) and similarly devastating widespread lesions Although these results were consistent with the idea that the
in a mouse mutant strain scurfy (8–11). Early studies of scurfy paucity Treg cells is responsible for the disease in Foxp3
mice revealed an essential role of T cells in the observed mutant animals, it remained possible that in addition to its
pathologies (12, 13). The fact that in both humans and mice abundant presence in Treg cells low level or transient Foxp3
only Foxp3 mutant males but not heterozygote female carriers expression in immune cells other than T cells or non-immune
were affected and the systemic nature of immune mediated cells is equally essential for the immune homeostasis. How-
lesions were consistent with an idea that Foxp3 mutations ever, the latter possibility was effectively refuted by the obser-
might impair differentiation or function of CD25+ Treg cells. vation that mice subjected to ablation of a conditional Foxp3
Indeed, high amounts of Foxp3 mRNA and protein were allele in the T-cell lineage and in the germ-line were pheno-
found in CD25+ Treg cells (14–16). Forced expression of typically indistinguishable, i.e. both strains of mice exhibited
Foxp3 in CD25)CD4+ T cells using retroviral vectors resulted T-cell-dependent autoimmune disease with identical onset,
in acquisition of suppressor function and Treg phenotype, progression and severity (17).
whereas in Foxp3 transgenic mice, CD8+ T cells exhibited Furthermore, deletion of self antigen-specific thymocytes as
suppressor function (14–16). Furthermore, transfer of alleli- well as activation and clonal expansion of, and cytokine pro-
cally marked bone marrow cells from Foxp3 knockout and duction by peripheral antigen-specific T cells were not
wildtype mice mixed at a 1:1 ratio showed that CD25+ Treg affected by the presence or absence of Foxp3 gene (14, 17,
cells originated only from Foxp3-suffcieint but not Foxp3- 19, 20). In these experiments, healthy (Ly5.2 Foxp3ko · Ly5.1
deficient hematopoietic precursor cells in the resulting (Ly5.2 Foxp3wt) fi Rag2ko bone marrow chimeras were challenged
Foxp3ko · Ly5.1 Foxp3wt) fi Rag2ko chimeric mice which with the virus or bacteria or with a superantigen staphylococ-
remained as healthy as heterozygote female carriers of the cal enterotoxin B. In these mice, allelically marked thymocytes
Foxp3 null allele (14). These results showed that in vivo Foxp3 and peripheral T cells containing or lacking functional Foxp3
is essential for differentiation of Treg cells and raised a ques- gene showed identical responses. Likewise, thymocytes in
tion as to whether lack of Treg cells can fully account for the wildtype mice were deleted to a similar degree by a trans-
observed pathology in Foxp3-deficient mice and humans or gene-encoded cognate ligand expressed in the thymus
putative lesions in other tissues and organs resulting from whereas peripheral T cells from these mice showed identical
Foxp3 deficiency can contribute to the disease in combination dose-dependent response to cognate ligand stimulation and
with a Treg deficiency. dependence on CD28 costimulation (20). Thus, Foxp3 gene
expression was dispensable for cell-intrinsic mechanisms of
thymic and peripheral tolerance and of negative regulation of
Treg cell deficiency fully accounts for inflammatory the peripheral T-cell responses. Finally, adoptive transfers of
lesions associated with Foxp3 deficiency Treg cells into 1–2 days old mutant recipients rescued lym-
This question was addressed in a series of genetic studies. pho- and myeloproliferative syndrome (14). Together, these
First, generation and analysis of knockin mice expressing fluo- results provided a definitive proof that Treg cell paucity

 2011 John Wiley & Sons A/S • Immunological Reviews 241/2011 261
Rudensky Æ Regulatory T cells and Foxp3

accounts for devastating disease associated with Foxp3 defi- self and that in Foxp3-deficient animals activated effector T
ciency in humans and mice. cells frequently utilize TCRs expressed by Treg cells in
However, this notion was challenged by Liu et al. (21), who wildtype mice (20). This finding raised a possibility that self-
employed staining with Foxp3 polyclonal antibodies and PCR reactive T cells, which failed to differentiate into Treg cells in
analysis to suggest that thymic, mammary and prostate epithe- the absence of Foxp3, are the primary cause of the disease in
lium widely expresses Foxp3. Furthermore, they proposed Foxp3 mutant mice. On the other hand, the documented
that in the absence of Foxp3 in the thymic epithelium early expression of self-reactive TCR by non-Treg cells and an over-
T-cell differentiation is impaired and that the lack of Foxp3 in lap between self-reactive TCR repertoire displayed by Treg
the thymic epithelium can account for or largely contribute and non-Treg cells left open a possibility that Treg cell abla-
to the disease in Foxp3 mutant animals (22). In mammary tion in adults may be as severe as a congenital Treg deficiency
gland and prostate, Foxp3 was proposed to act as a tumor in Foxp3-deficient mice (20). These alternative scenarios were
repressor essential to prevent cancer development in these tis- explored through generation and analysis of knockin mice
sues (23, 24). In spite of this, several groups using staining of harboring human diphtheria toxin receptor (DTR) expressed
these and other tissues from Foxp3-sufficient and -deficient under control of the Foxp3 locus. Ablation of Treg cells in
mice with monoclonal Foxp3 antibodies failed to reveal Foxp3DTR mice induced by chronic administration of diphtheria
detectable Foxp3 protein expression in epithelial cells (25, toxin (DT) starting from birth resulted in a similar disease
26). Furthermore, deletion of a conditional Foxp3 allele in thy- found in Foxp3null mice and death by 4 weeks of age (27). Fur-
mic epithelium using Cre recombinase driven by Foxn1 regu- thermore, adult healthy Foxp3DTR mice on a B6 genetic back-
latory elements did not result in any measurable effect on ground resistant to autoimmunity succumb to vast lympho-
thymic differentiation or lymphocyte activation and immune- and myeloproliferative disease within 2–3 weeks upon elimi-
mediated pathology (26). Likewise, deletion of a conditional nation of Treg cells (27). In the affected mice, massive expan-
Foxp3 gene in mammary epithelial cells using mammary gland sion and activation of diverse immune cell types including
epithelium-specific MMTV-Cre did not lead to tumorigenesis granulocytes, dendritic cells (DCs), NK cells, CD4+ and CD8+
or even mammary gland hyperplasia in mice up to 18 months T cells, and B cells was observed (27). In contrast to Foxp3DTR
of age (L. Henninghausen and A. Rudensky, unpublished knockin mice, DT mediated depletion of Treg cells in adult
observations). mice harboring a Foxp3 BAC transgene encoded DTR (DEREG)
Although more studies employing stringent genetic models did not result in a spontaneous autoimmune inflammatory
are needed to explore whether Foxp3 is expressed in cell types lesions, whereas Treg ablation in DEREG neonates led to a
other than T cells and its potential role in normal and cancer- severe disease (28). The latter discrepancy was likely a conse-
ous tissues, the overwhelming majority of experimental evi- quence of incomplete Treg cell elimination in adult DEREG
dence indicates that Foxp3 facilitates differentiation of Treg mice due to a BAC transgene inactivation. Within 24–48 h
cells and their paucity is responsible for the disease observed after Treg cell elimination in adult Foxp3DTR mice, both DCs
in Foxp3-deficient mice and most likely in IPEX patients. The and CD4+ T cells showed signs of activation. Importantly,
aforementioned genetic studies provided a definitive proof CD4+ T cells of a known specificity for a foreign antigen did
that the suppressive function of Foxp3-dependent Treg cells is not undergo activation upon Treg cell depletion, suggesting
vital for the immune homeostasis. that T-cell activation in Treg ablated mice is limited to T cells
specific for ‘self’, which includes genome encoded self, envi-
A critical role for Treg cells in the immune homeostasis ronmental and food antigens, and commensal micobiota
throughout the lifespan of normal animals (27). Furthermore, antibody-mediated depletion of CD4+ T
Since the immune system develops in the absence of Treg cells cells when combined with Treg cell ablation in Foxp3DTR mice
in Foxp3-deficient individuals, it remained unknown as to prevented massive lympho- and myeloproliferative disease.
whether Treg cell function in adults with a fully developed However, DC activation manifested in upregulation of costim-
immune system is not as critical as in newborns. According to ulatory molecules CD80 and CD86 and increased surface MHC
the latter scenario, elimination of Treg cells in adults would class II expression remained. These findings indicate that in
result in relatively mild immune mediated lesions in compari- the absence of Treg cells activation of ‘self’-reactive CD4+ T
son to those associated with a congenital Treg deficiency. cells, likely by DCs, fuels massive expansion of cells of the
Additional impetus to this question was given by an observa- innate and adaptive immune systems. Furthermore, recent
tion that Treg cells exhibit TCRs with an increased affinity for detailed analysis of DC differentiation showed that in normal

262  2011 John Wiley & Sons A/S • Immunological Reviews 241/2011
 Rudensky Æ Regulatory T cells and Foxp3

mice classical DCs proliferate in secondary lymphoid organs lung (34). Thus, the Treg cell subset in GF mice is fully com-
and that their differentiation and proliferation is kept in check petent and conditioning by commensal microbiota is dispens-
by Treg cells (29). Consistent with this notion, intravital able for the ability of Treg cells to suppress systemic
imaging of Treg cells expressing a transgene-encoded TCR autoimmunity. Moreover, ablation of Treg cells in GF mice
showed stable contacts between Treg cells and antigen-bear- led to comparable systemic inflammatory responses including
ing DCs but not with antigen-specific effector T cells (30, 31). lymphadenopathy, splenomegaly, T-cell and DC activation
Although Treg ablation studies failed to definitively identify and expansion, augmented Ig production, increases in CD80
early cellular targets of Treg cells in vivo, they strongly impli- and CD86 expression, and cytokine production including
cated Treg dependent restraint of dendritic cell activation as TNFa, IFNc, and IL-17 (34). These results suggest that the
one of the mechanisms of control of the immune homeostasis self-MHC-restricted T-cell recognition combined with basal
by Treg cells. On the other hand, manipulation of dendritic activity of innate immune sensors poses a threat of inflamma-
cell numbers through provision of Flt3L or inactivation of tory responses against self that cannot be met by conventional
Flt3L or Flt3 genes suggested that the size of the dendritic cell means of negative regulation and requires a dominant mecha-
population and of IL-2-producing activated effector T-cell nism of Treg-mediated suppression acting in trans.
subsets control the size of Treg cell population (32, 33). Thus,
autoregulatory feedback loops maintain a fine balance
between suppressive Treg cells, dendritic cell, and activated Role of Foxp3 in Treg cells
effector T cells. Studies discussed above established an essential non-redundant
role for Foxp3 in differentiation of Treg cells. However, spe-
A requirement for Treg cell function is independent on cific contribution of Foxp3 to differentiation and function of
the presence of commensal microbiota Treg cells remained unclear. This question was addressed
A critical role of Foxp3-dependent differentiation and func- through comparison of functional and transcriptional features
tion of Treg cells in immune homeostasis raised a question as of cells, which received all necessary signals required to pro-
to why numerous cell-intrinsic mechanisms preventing differ- mote Foxp3 expression yet lacked Foxp3 protein, and with
entiation of self-reactive T cells in the thymus and restraining those of Treg cells containing functional Foxp3 protein (35).
their numbers and activity in the periphery fail to ensure tol- In these studies, analysis of mice harboring a GFP coding DNA
erance to self. Considering that recognition of microorganisms sequence knocked-in into the Foxp3 locus with concomitant
and their products by innate immune sensors displayed by disruption of the Foxp3 protein expression revealed that
DCs endows them with the ability to activate T cells and that Foxp3 was not necessary for survival of Treg precursors, but
Treg cells control DC activation, it is reasonable to assume that was essential for Treg suppressor function (35). Similar results
the encounter of cells of the innate and adaptive immune sys- were obtained in studies of Foxp3 reporter mice harboring a
tems with commensal microbiota underlies the need of the truncated Foxp3 protein lacking DNA binding domain (36). It
dominant suppression in trans afforded by Treg cells. Alterna- is noteworthy that the amount of Foxp3 protein is critical for
tively, it is possible that regardless of the presence of microbi- Treg suppressor function as indicated by inability of Treg cells
ota ‘self-referential’ T-cell recognition, i.e. a requirement for to suppress spontaneous autoimmunity in mice with approxi-
certain affinity for self peptide–MHC complexes for thymocyte mately 10-fold decrease in Foxp3 protein expression due to
maturation and for T-cell survival in the periphery, poses a alteration in 3¢ untranslated region (UTR) of the Foxp3 gene
continuous threat of pathogenic self-reactivity that cannot be (37). Analysis of transcriptional signatures and functional
constrained by ‘conventional’ cell-intrinsic mechanisms of characteristics of cells expressing Foxp3 reporter null and func-
negative regulation. To explore these possibilities, suppressive tional alleles further showed that some of the features of Treg
capacity of Treg cells differentiating in the presence or absence cells, including increased CD25 and CTLA4, and diminished
of commensal microbiota was compared and their role in IL7Ra and immune response-promoting cytokine expression,
these two environments was assessed (34). Unexpectedly, are conferred to their precursors prior to Foxp3 expression
Treg cells present in germ-free (GF) and specific pathogen- likely by TCR and cytokine signaling, but Foxp3 exaggerates
free (SPF) mice were equally potent at suppressing lympho- this pre-existent pattern and makes it permanent (35). Thus,
and myeloproliferative disease and inflammatory lesions Foxp3 to a large extent amplifies and stabilizes molecular fea-
caused by effector T cells isolated from Foxp3-deficient mice tures of Treg precursor cells, beneficial to their function and
in a variety of tissues including barrier organs such as skin and maintenance, and reverses features deleterious to Treg cell

 2011 John Wiley & Sons A/S • Immunological Reviews 241/2011 263
Rudensky Æ Regulatory T cells and Foxp3

function. An example of the latter activity is Foxp3-mediated

repression of IL-17 expression (35). The latter is accomplished Stability of Foxp3 expression and of the Treg cell
through Foxp3 binding to RORc and abrogation of its ability function
to facilitate IL-17 transcription (38). Comprehensive analysis These results raised an important issue of phenotypic and
of transcriptional signatures of Foxp3-expressing and non- functional stability of Treg cells, i.e. whether under physio-
expressing cells generated in vivo and in vitro was fully consistent logic or inflammatory conditions Treg cells lose Foxp3 expres-
with these conclusions (39). Furthermore, Foxp3 solidifies sion and convert into potentially pathogenic (or even
Treg cell lineage stability through modification of cell surface beneficial) effector T cells. A number of recent studies pro-
and signaling molecules, resulting in adaptation to the signals vided varying answers to this question. A notion of plasticity
required to induce and maintain Treg cells. The latter notion is of Treg cell suppressor program came from an in vitro study
supported by the observation that Foxp3 dependent repression utilizing sorted Treg cells expressing a fluorescent reporter
of phosphodiesterase 3B expression results in adaption of Treg which showed that the exposure to pro-inflammatory cyto-
cells in chronic TCR and IL-2 signaling and their superior kines IL-6 and IL-1 leads to Foxp3 down-modulation and pro-
maintenance (35). These results raise a possibility that late act- duction of IL-17 (43). Along the same lines, a recent study
ing differentiation factors like Foxp3 direct cell fate decisions suggested that Treg cells suppressor activity is neutralized by
by acting in an ‘opportunistic’ manner by exploiting chroma- IL-6 produced in response to CpG stimulation. Upon CpG
tin landscapes prepared by developmental history of precursor stimulation, according to this study, Treg cells become the
cells rather than drastically changing these landscapes. predominant producers of immune response promoting cyto-
This idea is not inconsistent with the results of the genome- kines and serve a critical role in facilitating cross-presentation
wide analyses of Foxp3 target genes in Treg cells or in Foxp3 by DCs. In this study, Treg cells exposed to inflammatory
transfected hybridoma T-cell line using chromatin immuno- environment continue to express a GFP reporter, yet Foxp3
precipitation (ChIP) combined with whole genome tiling or protein was undetectable by FACS, yet was detectable by Wes-
custom promoter-enhancer arrays, respectively (40, 41). tern blot analysis (44). In contrast to these studies, a relatively
These studies suggested that Foxp3 acts as both transcriptional small proportion of Foxp3+ Treg cells, characterized by some-
activator and repressor and that Foxp3 direct targets make up what lower levels of CD25, lost Foxp3 expression upon adop-
approximately 10–20% of Foxp3-dependent genes. Foxp3 tive transfer together with allelically marked ‘non-Treg’ cells
binding to repressed and activated genes correlated with non- into lymphopenic recipients, whereas CD25hi Treg cells lar-
permissive H3K27me3 and permissive H3K4me3 marks, gely maintained Foxp3 expression (45).
respectively, indicative of epigenetically controlled Foxp3- One caveat associated with studies relying on cell sorting is
dependent transcriptional program (40). However, it is possi- that a minor contamination of Foxp3-negative cells might lead
ble that relatively few of these marks, if any, are imparted to their expansion in vivo or in vitro and cell stress associated
upon Foxp3 binding. Instead, we propose that it is likely that with the isolation procedure might facilitate apparent Foxp3
the bulk of Foxp3 target genes are ‘prepared’ for Foxp3 arrival loss or sequestration. These issues were addressed in studies
by other transcriptional factors acting as ‘place holders’. employing irreversible genetic tagging of Treg cells in vivo
Foxp3 binding-site analysis combined with the assessment using Cre recombinase expressed under the control of Foxp3
of the histone modification status seemed to suggest that Foxp3 regulatory elements combined with a recombination reporter
expression was necessary to establish Foxp3-dependent tran- allele. To explore induction and loss of Foxp3 expression one
scriptional program during Treg differentiation but was dis- such study employed Foxp3 BAC transgene driving expression
pensable for its maintenance. Contrary to these expectations, of a GFP-Cre fusion protein and a recombination reporter
Cre-mediated deletion of a conditional Foxp3 allele in fully allele, a ubiquitously expressed Rosa26 locus harboring YFP
differentiated peripheral Treg cells resulted in a cell division- coding sequence preceded by a loxP site flanked STOP cassette
dependent loss of developmentally established characteristic (R26Y) (46). Studies in these mice showed that some YFP-
gene expression pattern and of suppressor function (42). Fur- tagged T cells lacked Foxp3, exhibited features of effector
thermore, upon loss of Foxp3 ‘ex-Treg’ cells acquired ability to T cells and were able to promote autoimmune inflammation
produce immune effector cytokines, i.e. IFNc, IL-4, and IL-17 in diabetes-prone NOD mice (46). These results were inter-
and pathogenic potential. Indeed, the ‘ex-Treg’ cells, upon preted as a proof of plasticity of Treg cells and raised concerns
transfer into lymphopenic recipients without functional Treg with the use of Treg cell transfers in clinic for therapeutic pur-
cells, caused severe tissue lesions and wasting disease (42). poses. One caveat with this conclusion was that Foxp3 BAC

264  2011 John Wiley & Sons A/S • Immunological Reviews 241/2011
 Rudensky Æ Regulatory T cells and Foxp3

transgene driven constitutive expression of GFP-Cre affords

continuous labeling of Foxp3 expressing cells including those Mechanisms of Foxp3 induction and maintenance of
Treg stability
with low levels of Foxp3 like the CD25lo cells discussed above,
as well as cells with transient Foxp3 expression and recently Considering a central role for Foxp3 in establishing Treg cell
generated Treg cells prior to Foxp3 stabilization (see below). function during differentiation and its maintenance in the
Another recent study employed inducible labeling of a cohort periphery, recent exploration of regulatory elements within
of Treg cells upon tamoxifen treatment of knock-in mice the Foxp3 locus offered insights into the molecular mecha-
expressing a GFP–ER–Cre fusion protein (containing GFP, Cre nisms of regulatory T-cell differentiation and maintenance.
recombinase and the ligand binding domain of estrogen Analysis of proximal conserved non-coding sequences (CNS)
receptor) under the control of Foxp3 locus. The Foxp3GFP–ER–Cre in the Foxp3 locus revealed several intronic elements playing
allele combined with R26Y reporter allows for YFP tagging of prominent roles in control of Foxp3 expression. One of these
a cohort of differentiated Treg cells and monitor maintenance elements, CNS3, contains a DNase I hypersensitive site and is
of Foxp3 expression in these cells. Analysis of these mice enriched in histone H3 monomethylated at position K4
revealed remarkably stable maintenance of Foxp3 expression (H3K4me1, a mark associated with enhancer elements) not
in the progeny of tagged Treg cells under the basal conditions only in Foxp3 expressing Treg cells, but also in their Foxp3-
over the lifespan of a mouse, during Listeria infection and negative thymic and peripheral precursors (50). NF-jB family
under irradiation-induced lymphopenia. In addition, in this member c-Rel binds to this element and likely facilitates the
study Treg cells maintained Foxp3 expression and failed to activity of the Foxp3 promoter (50, 51). Gene targeting studies
produce pro-inflammatory cytokines in the context of autoim- revealed an essential role for CNS3 in induction of Foxp3
mune inflammation. In these experiments, double FACS expression in the thymus and in the periphery (50). However,
sorted Treg cells expressing diabetogenic BDC2.5 or arthrito- CNS3 deficient mice remain healthy despite approximately
genic K ⁄ BxN TCRs were transferred into prediabetic or pre- fivefold decrease in repertoire of Treg cells in comparison to
arthritic recipients, respectively, and their ability to produce CNS3-suffcient mice. Consistent with the documented c-Rel
IL-2, IFNc, and IL-17 and maintain Foxp3 expression was binding to CNS3, c-Rel deficiency in thymic precursor cells
assessed after the onset of disease in the target tissue or the resulted in a profound impairment in Treg differentiation
related lymphoid sites. Although these experiments showed (50–55). Although the mechanism of CNS3 mediated potenti-
remarkable stability of Foxp3 expression, antibody-mediated ation of Foxp3 induction remains unclear, it was suggested
neutralization of IL-2 resulted in a moderate decrease in that c-Rel binding to CNS3 in cooperation with other tran-
Foxp3 expression in the entire Treg population and the emer- scription factors, e.g. NFAT, CREB, p65, and Smad, facilitates
gence of a minor subset of YFP-tagged cells which completely formation of c-Rel containing enhanceosome at the Foxp3 pro-
lost Foxp3 expression. At the same time the entire peripheral moter. The experimental evidence in support of this model is
Treg population size decreased by more than a third (47). limited to temporal changes in c-Rel binding sites within the
These results were in agreement with the observation that Foxp3 locus (51). In addition to c-Rel and other NF-jB family
Foxp3+ Treg cells in mice lacking IL-2 (or IL-2Ra chain) were members, CREB ⁄ ATF, Ets-1, Foxo1 ⁄ 3, and Stat5 facilitate
found in half the numbers of those in wildtype mice and Foxp3 transcription and bind to its promoter and other regu-
exhibited a similarly decreased amount of Foxp3 protein on latory elements within the Foxp3 gene (56–61). The involve-
per cell basis. Expression of Foxp3 was recovered to wildtype ment of transcription factors activated downstream of TCR
levels upon provision of IL-2 to IL-2-deficient mice (48). and IL2R signaling pathways is consistent with an elegant
Interestingly, a decrease in Treg cell numbers was recently two-stage model of generation of Treg cells in response to
reported during fatal Toxoplasma infection in mice due to a heightened strengths of TCR engagement and IL-2 signaling
sharp drop in IL-2 amounts. In addition, some Treg cells in (62–65).
infected mice produced IFNc. Provision of IL-2-anti-IL2 Another intronic element dubbed CNS2 [also known as
immune complexes rescued a loss of Treg cells (49). Although Treg-specific demethylated region (TSDR)] contains a CpG
more experiments are needed to evaluate stability of Foxp3 island, whose methylation status was correlated with the
expression and suppressor function of Treg cells in diverse sit- Foxp3 expression and its stability (66, 67). The CNS2 CpGs
uations of infection and autoimmunity, it is reasonable to were demethylated in ex vivo isolated Treg cells, whereas in
assume that there must be a mechanism that ensures stability Foxp3-negative peripheral T cells or thymocytes CNS2 was
of Treg function in inflammatory and basal settings. fully methylated and remained so upon induction of Foxp3 in

 2011 John Wiley & Sons A/S • Immunological Reviews 241/2011 265
Rudensky Æ Regulatory T cells and Foxp3

peripheral T cells stimulated in the presence of TGFb (68). RARa binding sites. Although initially CNS1 was proposed to
In vitro luciferase reporter studies suggested that CNS2 might play a role in Foxp3 induction in the thymus and in the
act as a cryptic promoter by recruiting CREB ⁄ ATF and Stat5 periphery (74, 75), analysis of mice deficient in CNS1
(66) as well as Ets-1 (69). Mutation of the corresponding sites revealed that thymic differentiation of Foxp3+ cells was unaf-
resulted in a loss or a marked decrease in transcriptional activ- fected, whereas peripheral Foxp3 induction was severely
ity in non-chromatinized reporter assays. Ablation of CNS2 impaired by CNS1 deficiency (50). Surprisingly, deficiency in
element, however, showed that its non-redundant function peripherally generated Treg cells did not result in detectable
lies in heritable maintenance of Foxp3 expression in the prog- tissue-specific autoimmunity in CNS1-deficient mice up to
eny of dividing Treg cells. While Foxp3 induction in the thy- 1 year of age. These results suggested that mechanistic
mus and in the periphery was unimpaired in CNS2-deficient requirements for thymic and peripheral induction of Foxp3
mice, Treg cells lacking CNS2 progressively lost Foxp3 expres- are distinct and that Treg cells generated in the periphery have
sion upon division (50). Furthermore, Foxp3 is recruited to a non-redundant function likely distinct from that of Treg
CNS2 with the assistance of Runx1 and its co-factor CBFb and cells differentiated in the thymus. Future studies of CNS1-defi-
the binding of Foxp3-Runx1-CBFb complex confers stability cient mice will likely reveal potential functions of peripherally
of Foxp3 expression via an unidentified epigenetic mechanism generated Treg cells.
(50). It must be noted that in addition to preserving the active The results of initial biochemical and genetic analyses of
state of the Foxp3 locus in dividing Treg cells, Runx1-CBFb regulation of Foxp3 expression illuminated molecular mecha-
complex is also involved in facilitating Foxp3 promoter activ- nisms governing generation and function of regulatory T cells
ity (70, 71). and future molecular studies will undoubtedly provide further
Thus, CNS2 appears to be the first mammalian ‘cellular insights into the biology of regulatory T cells. The uncovered
memory module’ (CMM) analogous to elements involved in salient features of Treg cell differentiation, including a
the maintenance of hedgehog or Abdominal B expression dur- requirement for Foxp3 expression in fully differentiated Treg
ing Drosophila development (50, 72, 73). Furthermore, pro- cells for their function and for the maintenance of Foxp3
posed role for Foxp3 recruitment to CNS2 and the resulting expression during cell division, are likely applicable to other
trans-generational maintenance of Foxp3 expression repre- cell lineages. This point is strongly supported by recent dem-
sents the first description of a feed-forward mechanism onstration of a need for Pax5 and E2-2 to maintain identity of
enforcing heritable cell lineage identity. differentiated B cells and plasmacytoid DCs, respectively
CNS1 (or enhancer 1), the last among the Foxp3 regulatory (76, 77).
elements characterized so far, contains NFAT, Smad3, and

References
1. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, 4. Asano M, Toda M, Sakaguchi N, Sakaguchi S. results in the fatal lymphoproliferative
Toda M. Immunologic self-tolerance main- Autoimmune disease as a consequence of disorder of the scurfy mouse. Nat Genet
tained by activated T cells expressing IL-2 developmental abnormality of a T cell sub- 2001;27:68–73.
receptor alpha-chains (CD25). Breakdown of population. J Exp Med 1996;184:387–396. 9. Bennett CL, et al. The immune dysregula-
a single mechanism of self-tolerance causes 5. Papiernik M, de Moraes ML, Pontoux C, tion, polyendocrinopathy, enteropathy,
various autoimmune diseases. J Immunol Vasseur F, Penit C. Regulatory CD4 T cells: X-linked syndrome (IPEX) is caused by
1995;155:1151–1164. expression of IL-2R alpha chain, resistance to mutations of FOXP3. Nat Genet
2. Takahashi T, et al. Immunologic self-toler- clonal deletion and IL-2 dependency. Int 2001;27:20–21.
ance maintained by CD25(+)CD4(+) regula- Immunol 1998;10:371–378. 10. Wildin RS, et al. X-linked neonatal diabetes
tory T cells constitutively expressing 6. Gavin MA, Clarke SR, Negrou E, Gallegos A, mellitus, enteropathy and endocrinopathy
cytotoxic T lymphocyte-associated antigen 4. Rudensky A. Homeostasis and anergy of syndrome is the human equivalent of mouse
J Exp Med 2000;192:303–310. CD4(+)CD25(+) suppressor T cells in vivo. scurfy. Nat Genet 2001;27:18–20.
3. Sakaguchi S, Fukuma K, Kuribayashi K, Nat Immunol 2002;3:33–41. 11. Chatila TA, et al. JM2, encoding a fork head-
Masuda T. Organ-specific autoimmune dis- 7. Suri-Payer E, Amar AZ, Thornton AM, Shev- related protein, is mutated in X-linked au-
eases induced in mice by elimination of T ach EM. CD4+ CD25+ T cells inhibit both toimmunity-allergic disregulation syndrome.
cell subset. I. Evidence for the active partici- the induction and effector function of J Clin Invest 2000;106:R75–R81.
pation of T cells in natural self-tolerance; autoreactive T cells and represent a unique 12. Blair PJ, Bultman SJ, Haas JC, Rouse BT,
deficit of a T cell subset as a possible cause of lineage of immunoregulatory cells. J Immu- Wilkinson JE, Godfrey VL. CD4+ CD8) T
autoimmune disease. J Exp Med nol 1998;160:1212–1218. cells are the effector cells in disease patho-
1985;161:72–87. 8. Brunkow ME, et al. Disruption of a new genesis in the scurfy (sf) mouse. J Immunol
forkhead ⁄ winged-helix protein, scurfin, 1994;153:3764–3774.

266  2011 John Wiley & Sons A/S • Immunological Reviews 241/2011
 Rudensky Æ Regulatory T cells and Foxp3

13. Godfrey VL, Wilkinson JE, Rinchik EM, 27. Kim JM, Rasmussen JP, Rudensky AY. Regu- 43. Yang XO, et al. Molecular antagonism and
Russell LB. Fatal lymphoreticular disease in latory T cells prevent catastrophic autoimmu- plasticity of regulatory and inflammatory T
the scurfy (sf) mouse requires T cells that nity throughout the lifespan of mice. Nat cell programs. Immunity 2008;29:44–56.
mature in a sf thymic environment: potential Immunol 2007;8:191–197. 44. Sharma MD, et al. Reprogrammed foxp3(+)
model for thymic education. Proc Natl Acad 28. Lahl K, et al. Selective depletion of Foxp3+ regulatory T cells provide essential help to
Sci USA 1991;88:5528–5532. regulatory T cells induces a scurfy-like dis- support cross-presentation and CD8(+) T
14. Fontenot JD, Gavin MA, Rudensky AY. Foxp3 ease. J Exp Med 2007;204:57–63. cell priming in naive mice. Immunity
programs the development and function of 29. Liu K, et al. In vivo analysis of dendritic cell 2010;33:942–954.
CD4+ CD25+ regulatory T cells. Nat Immu- development and homeostasis. Science 45. Komatsu N, Mariotti-Ferrandiz ME, Wang Y,
nol 2003;4:330–336. 2009;324:392–397. Malissen B, Waldmann H, Hori S. Heteroge-
15. Hori S, Nomura T, Sakaguchi S. Control of 30. Tang Q, et al. Visualizing regulatory T cell neity of natural Foxp3+ T cells: a committed
regulatory T cell development by the control of autoimmune responses in nonob- regulatory T-cell lineage and an uncommit-
transcription factor Foxp3. Science 2003; ese diabetic mice. Nat Immunol 2006;7:83– ted minor population retaining plasticity.
299: 1057–1061. 92. Proc Natl Acad Sci USA 2009;106:1903–
16. Khattri R, Cox T, Yasayko SA, Ramsdell F. An 31. Tadokoro CE, et al. Regulatory T cells inhibit 1908.
essential role for Scurfin in CD4+ CD25+ T stable contacts between CD4+ T cells and 46. Zhou X, et al. Instability of the transcription
regulatory cells. Nat Immunol 2003;4:337– dendritic cells in vivo. J Exp Med 2006; 203: factor Foxp3 leads to the generation of path-
342. 505–511. ogenic memory T cells in vivo. Nat Immunol
17. Fontenot JD, Rasmussen JP, Williams LM, 32. Darrasse-Jeze G, et al. Feedback control of 2009;10:1000–1007.
Dooley JL, Farr AG, Rudensky AY. Regulatory regulatory T cell homeostasis by dendritic 47. Rubtsov YP, et al. Stability of the regulatory
T cell lineage specification by the forkhead cells in vivo. J Exp Med 2009;206:1853– T cell lineage in vivo. Science
transcription factor foxp3. Immunity 1862. 2010;329:1667–1671.
2005;22:329–341. 33. Swee LK, Bosco N, Malissen B, Ceredig R, 48. Fontenot JD, Rasmussen JP, Gavin MA,
18. Wan YY, Flavell RA. Identifying Foxp3- Rolink A. Expansion of peripheral naturally Rudensky AY. A function for interleukin 2 in
expressing suppressor T cells with a bicis- occurring T regulatory cells by Fms-like tyro- Foxp3-expressing regulatory T cells. Nat
tronic reporter. Proc Natl Acad Sci USA sine kinase 3 ligand treatment. Blood Immunol 2005;6:1142–1151.
2005;102:5126–5131. 2009;113:6277–6287. 49. Oldenhove G, et al. Decrease of Foxp3+ Treg
19. Chen Z, Benoist C, Mathis D. How defects in 34. Chinen T, Volchkov PY, Chervonsky AV, cell number and acquisition of effector cell
central tolerance impinge on a deficiency in Rudensky AY. A critical role for regulatory T phenotype during lethal infection. Immunity
regulatory T cells. Proc Natl Acad Sci USA cell-mediated control of inflammation in the 2009;31:772–786.
2005;102:14735–14740. absence of commensal microbiota. J Exp 50. Zheng Y, Josefowicz S, Chaudhry A, Peng
20. Hsieh CS, Zheng Y, Liang Y, Fontenot JD, Med 2010;207:2323–2330. XP, Forbush K, Rudensky AY Role of con-
Rudensky AY. An intersection between the 35. Gavin MA, et al. Foxp3-dependent pro- served non-coding DNA elements in the
self-reactive regulatory and nonregulatory T gramme of regulatory T-cell differentiation. Foxp3 gene in regulatory T-cell fate. Nature
cell receptor repertoires. Nat Immunol Nature 2007;445:771–775. 2010;463:808–812.
2006;7:401–410. 36. Lin W, et al. Regulatory T cell development 51. Ruan Q, et al. Development of Foxp3(+)
21. Chen GY, Chen C, Wang L, Chang X, Zheng in the absence of functional Foxp3. Nat regulatory t cells is driven by the c-Rel
P, Liu Y. Cutting edge: broad expression of Immunol 2007;8:359–368. enhanceosome. Immunity 2009;31:932–
the FoxP3 locus in epithelial cells: a caution 37. Wan YY, Flavell RA. Regulatory T-cell func- 940.
against early interpretation of fatal inflamma- tions are subverted and converted owing to 52. Visekruna A, et al. c-Rel is crucial for the
tory diseases following in vivo depletion of attenuated Foxp3 expression. Nature induction of Foxp3(+) regulatory CD4(+) T
FoxP3-expressing cells. J Immunol 2007;445:766–770. cells but not T(H)17 cells. Eur J Immunol
2008;180:5163–5166. 38. Zhou L, et al. TGF-beta-induced Foxp3 2010;40:671–676.
22. Chang X, et al. The Scurfy mutation of FoxP3 inhibits T(H)17 cell differentiation by antag- 53. Vang KB, et al. Cutting edge: CD28 and
in the thymus stroma leads to defective thy- onizing RORgammat function. Nature c-Rel-dependent pathways initiate regulatory
mopoiesis. J Exp Med 2005;202:1141– 2008;453:236–240. T cell development. J Immunol
1151. 39. Hill JA, et al. Foxp3 transcription-factor- 2010;184:4074–4077.
23. Zuo T, et al. FOXP3 is an X-linked breast dependent and -independent regulation of 54. Long M, Park SG, Strickland I, Hayden MS,
cancer suppressor gene and an important the regulatory T cell transcriptional signa- Ghosh S. Nuclear factor-kappaB modulates
repressor of the HER-2 ⁄ ErbB2 oncogene. ture. Immunity 2007;27:786–800. regulatory T cell development by directly
Cell 2007;129:1275–1286. 40. Zheng Y, Josefowicz SZ, Kas A, Chu TT, regulating expression of Foxp3 transcription
24. Wang L, et al. Somatic single hits inactivate Gavin MA, Rudensky AY. Genome-wide factor. Immunity 2009;31:921–931.
the X-linked tumor suppressor FOXP3 in the analysis of Foxp3 target genes in developing 55. Isomura I, et al. c-Rel is required for the
prostate. Cancer Cell 2009;16:336–346. and mature regulatory T cells. Nature development of thymic Foxp3+ CD4 regula-
25. Kim J, et al. Cutting edge: depletion of 2007;445:936–940. tory T cells. J Exp Med 2009;206:3001–
Foxp3+ cells leads to induction of autoim- 41. Marson A, et al. Foxp3 occupancy and regu- 3014.
munity by specific ablation of regulatory T lation of key target genes during T-cell stim- 56. Harada Y, Elly C, Ying G, Paik JH, DePinho
cells in genetically targeted mice. J Immunol ulation. Nature 2007;445:931–935. RA, Liu YC Transcription factors Foxo3a and
2009;183:7631–7634. 42. Williams LM, Rudensky AY. Maintenance of Foxo1 couple the E3 ligase Cbl-b to the
26. Liston A, et al. Lack of Foxp3 function and the Foxp3-dependent developmental pro- induction of Foxp3 expression in induced
expression in the thymic epithelium. J Exp gram in mature regulatory T cells requires regulatory T cells. J Exp Med
Med 2007;204:475–480. continued expression of Foxp3. Nat Immu- 2010;207:1381–1391.
nol 2007;8:277–284.

 2011 John Wiley & Sons A/S • Immunological Reviews 241/2011 267
Rudensky Æ Regulatory T cells and Foxp3

57. Ouyang W, Beckett O, Ma Q, Paik JH, DePin- cells in the thymus. Nat Immunol 2009; 72. Bantignies F, Grimaud C, Lavrov S, Gabut M,
ho RA, Li MO. Foxo proteins cooperatively 10:610–617. Cavalli G. Inheritance of Polycomb-depen-
control the differentiation of Foxp3+ regula- 65. Leung MW, Shen S, Lafaille JJ. TCR-depen- dent chromosomal interactions in Drosoph-
tory T cells. Nat Immunol 2010;11:618– dent differentiation of thymic Foxp3+ cells ila. Genes Dev 2003;17:2406–2420.
627. is limited to small clonal sizes. J Exp Med 73. Maurange C, Paro R. A cellular memory
58. Kerdiles YM, et al. Foxo transcription factors 2009;206:2121–2130. module conveys epigenetic inheritance of
control regulatory T cell development and 66. Kim HP, Leonard WJ. CREB ⁄ ATF-dependent hedgehog expression during Drosophila
function. Immunity 2010;33:890–904. T cell receptor-induced FoxP3 gene expres- wing imaginal disc development. Genes Dev
59. Burchill MA, Yang J, Vogtenhuber C, Blazar sion: a role for DNA methylation. J Exp Med 2002;16:2672–2683.
BR, Farrar MA. IL-2 receptor beta-dependent 2007;204:1543–1551. 74. Tone Y, Furuuchi K, Kojima Y, Tykocinski
STAT5 activation is required for the develop- 67. Floess S, et al. Epigenetic control of the ML, Greene MI, Tone M. Smad3 and NFAT
ment of Foxp3+ regulatory T cells. J Immu- foxp3 locus in regulatory T cells. PLoS Biol cooperate to induce Foxp3 expression
nol 2007;178:280–290. 2007;5:e38. through its enhancer. Nat Immunol
60. Yao Z, et al. Nonredundant roles for Stat5a ⁄ b 68. Polansky JK, et al. DNA methylation controls 2008;9:194–202.
in directly regulating Foxp3. Blood 2007; Foxp3 gene expression. Eur J Immunol 75. Xu L, Kitani A, Stuelten C, McGrady G, Fuss
109:4368–4375. 2008;38:1654–1663. I, Strober W. Positive and negative transcrip-
61. Mouly E, et al. The Ets-1 transcription factor 69. Polansky JK, et al. Methylation matters: bind- tional regulation of the Foxp3 gene is medi-
controls the development and function of ing of Ets-1 to the demethylated Foxp3 gene ated by access and binding of the Smad3
natural regulatory T cells. J Exp Med contributes to the stabilization of Foxp3 protein to enhancer I. Immunity
2010;207:2113–2125. expression in regulatory T cells. J Mol Med 2010;33:313–325.
62. Burchill MA, et al. Linked T cell receptor and 2010;88:1029–1040. 76. Cobaleda C, Jochum W, Busslinger M. Con-
cytokine signaling govern the development 70. Rudra D, Egawa T, Chong MM, Treuting P, version of mature B cells into T cells by
of the regulatory T cell repertoire. Immunity Littman DR, Rudensky AY. Runx-CBFbeta dedifferentiation to uncommitted progeni-
2008;28:112–121. complexes control expression of the tran- tors. Nature 2007;449:473–477.
63. Lio CW, Hsieh CS. A two-step process for scription factor Foxp3 in regulatory T cells. 77. Ghosh HS, Cisse B, Bunin A, Lewis KL, Reizis
thymic regulatory T cell development. Nat Immunol 2009;10:1170–1177. B. Continuous expression of the transcription
Immunity 2008;28:100–111. 71. Kitoh A, et al. Indispensable role of the factor e2-2 maintains the cell fate of mature
64. Bautista JL, et al. Intraclonal competition Runx1-Cbfbeta transcription complex for in plasmacytoid dendritic cells. Immunity
limits the fate determination of regulatory T vivo-suppressive function of FoxP3+ regula- 2010;33:905–916.
tory T cells. Immunity 2009;31:609–620.

268  2011 John Wiley & Sons A/S • Immunological Reviews 241/2011