The online version of this article can be found at:
2012 40: 186 originally published online 5 January 2012Toxicol Pathol
Richard A. Peterson
Regulatory T-Cells: Diverse Phenotypes Integral to Immune Homeostasis and Suppression
On behalf of:
Society of Toxicologic Pathology
can be found at:
Additional services and information for
What is This?
- Jan 5, 2012OnlineFirst Version of Record
- Mar 19, 2012 Version of Record >>
by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013 by guest on October 11, 2013tpx.sagepub.com tpx.sagepub.comtpx.sagepub.comtpx.sagepub.com tpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.comtpx.sagepub.com tpx.sagepub.com tpx.sagepub.comtpx.sagepub.comtpx.sagepub.com tpx.sagepub.comDownloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from
Regulatory T-Cells: Diverse Phenotypes Integral to
Immune Homeostasis and Suppression
RICHARD A. PETERSON
GlaxoSmithKline, Research Triangle Park, North Carolina, USA
Regulatory T-cells (TREG) are diverse populations of lymphocytes that regulate the adaptive immune response in higher vertebrates. TREGdelete
autoreactive T-cells, induce tolerance, and dampen inflammation. TREGcell deficiency in humans (i.e., IPEX [Immunodysregulation, Polyendocri-
nopathy and Enteropathy, X-linked syndrome]) and animal models (e.g., ‘‘Scurfy’’ mouse) is associated with multisystemic autoimmune disease.
TREGin humans and laboratory animal species are similar in type and regulatory function. A molecular marker of and the cell lineage specification
factor for TREG is FOXP3, a forkhead box transcription factor. CD4þTREG are either natural (nTREG), which are thymus-derived
CD4þCD25þFOXP3þT-cells, or inducible (i.e., Tr1 cells that secrete IL-10, Th3 cells that secrete TGF-b and IL-10, and Foxp3þTreg). The proin-
flammatory Th17 subset has been a major focus of research. TH17 CD4þeffector T-cells secrete IL-17, IL-21, and IL-22 in autoimmune and inflam-
matory disease, and are dynamically balanced with TREGcell development. Other lymphocyte subsets with regulatory function include: inducible
CD8þTREG, CD3þCD4?CD8?TREG(double-negative), CD4þVa14þ(NKTREG), and gd T-cells. TREGhave four regulatory modes of action: secre-
tion of inhibitory cytokines (e.g., IL-10 and TGF-b), granzyme-perforin-induced apoptosis of effector lymphocytes, depriving effector T-cells of
cytokines leading to apoptosis, or inhibition of dendritic cell function. The role of TREGin mucosal sites, inflammation/infection, pregnancy, and
cancer as well as a review of TREGas a modulatory target in drug development will be covered.
immunopathology; regulatory T-cells; lymphocytes; flow cytometry; Scurfy mouse; FOXP3; IPEX.
INTRODUCTION TO REGULATORY T-CELLS
This article will review the scientific literature and provide an
overview of Regulatory T-cell (TREG) characterization, function,
role in the immune response, at mucosal sites, during pregnancy
in the field of immunology, which translates into a prominent
TREGare diverse populations of lymphocytes that regulate
the adaptive immune response in higher vertebrates. TREG
delete autoreactive T-cells, induce tolerance, and dampen
inflammation (Akdis 2009; Ozdemir et al. 2009; Dittel 2008;
Orihara et al. 2008; Sakaguchi et al. 2008; Simpson 2008;
Bluestone and Tang 2005; Coutinho et al. 2005).
Over the past 40 years, the field of immunology has slowly
evolved, becoming much more complex than was initially
hypothesized in the 1960s. Between 1966 and 1968, several
groups identified B-lymphocytes as the source of the antibody
response and T-lymphocytes as critical in the delayed-type
hypersensitivity (DTH) response (Claman et al. 1966; Mosier
1967; Mitchell et al. 1968). In 1971, Gershon and Kondo iden-
tified negative interference on positively acting lymphocytes
during inflammation and referred to it as ‘‘infectious immuno-
logical tolerance’’ (Gershon and Kondo 1971). This was the
first indication that a ‘‘suppressive’’ process was active during
inflammation. The ‘‘suppressor T-cell’’ hypothesis was put
forth by two groups in the later years of the 1970s (Vadas et
al. 1976; Okumura et al. 1977). A series of negative results
from experiments conducted by several groups led to a loss
of confidence in the ‘‘suppressor T-cell’’ hypothesis, rejection
of the hypothesis by the field of immunology, and more
recently reacceptance (Kapp and Bucy 2008; Germain 2008;
Lehner 2008). The rejection of the ‘‘suppressor T-cell’’ hypoth-
esis did not explain away the presence of a dampening of
immune responses commonly seen in the inflammatory milieu.
In 1989, Mosmann and Coffman identified the T-helper (TH)-1
and (TH)-2 CD4þT-cell subsets based upon the profiles of
cytokines secreted after activation. TH1 cells primarily secrete
IFN-g, IL-15, and TNF-a, while TH2 cells secrete IL-4, IL-5,
IL-6, and IL-13. Powrie et al. showed in severe combined
immunodeficient (SCID) mice that subsets of CD4þT-cells
The author declared no potential conflicts of interests with respect to the
authorship and/or publication of this article. The author received no financial
support for the research and/or authorship of this article.
Address correspondence to: Richard A. Peterson II, DVM, PhD, DACVP,
Safety Assessment, 9.3009.2D, GlaxoSmithKline, 5 Moore Drive, Research
Triangle Park, NC 27709; phone: (919) 315-2450; fax: (919) 483-6858;
Abbreviations: A2A, adenosine receptor; APC, antigen presenting cell; BREG,
regulatoryB-cell; CD, cluster ofdifferentiation;CMV,cytomegalovirus;CTLA-
4, cytotoxic T-lymphocyte antigen-4; DN, double negative; DTH, delayed-type
immunodeficiency virus; IBD, inflammatory bowel disease; IFNg, interferon
pathy X-linked syndrome; LAG3, lymphocyte-activating gene-3; MHC, major
histocompatability complex; miR, microRNA; MoAb, monoclonal antibody;
mTOR, mammalian target of rapamycin; NK cell, natural killer cell; NKTREG,
natural killer regulatory T-cell; SCID, severe combined immunodeficient; TCR,
T-cell receptor; TGF-b, transforming growth factor beta; TH0, T-helper 0 cell;
TH1, T-helper 1 cell; TH2, T-helper 2 cell; TH3, T-helper 3 cell; TH17, T-helper
17 cell; TREG, regulatory T-cell; Tr1, type 1 regulatory T-cells.
Toxicologic Pathology, 40: 186-204, 2012
Copyright # 2012 by The Author(s)
ISSN: 0192-6233 print / 1533-1601 online
induce and protect from intestinal inflammation (Powrie et al.
1994). This once again defined a population of lymphocytes
with immunosuppressive actions, which was identified as a
population expressing CD45RBlowwhile the inducing cells
were CD45RBhigh. In 1995, Sakaguchi et al. found that immu-
nologic self-tolerance was maintained by activated CD4þT-
cells expressing CD25 (IL-2R alpha chain). Depletion of this
found to be thymically derived, natural TREGcells (Hori et al.
2003; Fontenot et al. 2003, 2005). FOXP3 has been shown to
be critical for regulatory action of most subtypes of TREG.
TYPES OF REGULATORY T-CELLS (TREG, NATURAL,
AND ADAPTIVE/INDUCED) CD4þCD25þTREG
Natural TREGare thymus-derived CD4þCD25þFOXP3þT-
cells (Figure 2). They arise in the thymus during early stages of
human fetal development (gestational week 14) (Cupedo et al.
2005) and are resistant to thymic deletion (Lim et al. 2006).
nTREGdifferentiate from thymocytes that express TCRs with
an increased affinity for self-peptide-MHC complexes (Lim
et al. 2006; Maggi et al. 2005; Schwartz 2005). nTREGcan sup-
press the following cell types: CD4þ T-cells, dendritic cells,
CD8þT-cells, NKT cells, NK cells, monocytes/macrophages,
B-cells, mast cells, basophils, eosinophils, and osteoblasts
(Shevach et al. 2009; Lim et al. 2006). FOXP3þexpression
is necessary for thymocytes to commit to the TREGlineage
(Lim et al. 2006). Resting nTREGare CD45RAþFOXP3low,
while activated nTREG are CD45ROþFOXP3high(Beissert
et al. 2006; Lim et al. 2006). Natural TREGalso express the
following markers: T cell activation/differentiation markers
(CD45RO [memory phenotype; activated], CD45RB [resting],
CD25 [both]), adhesion molecules (CD62L [both], CD44
[both] and Integrin a4b7[resting]), cytotoxic T lymphocyte-
associated protein-4 (CTLA-4; both), co-stimulatory molecule
CD28 (both), chemokine receptors (CCR7 [both], CXCR4
[both],CCR9 [resting]), glucocorticoid-induced
related protein (GITR, both), OX40 (CD134, both), and folate
tial for nTREGdevelopment, function, and homeostasis (i.e.,
CD25 is the a-chain of the high affinity IL2R) (Malek et al.
2008). Severe autoimmunity is seen in IL-2-, IL2Ra-, and IL-
2Rb-deficient mice, which also lack a normal number of TREG
conserved. Nonhuman primates have the same nTREGprotein
FOXP3 isoforms (Roncarolo and Battaglia 2007; Bloom 2006),
whileinrodents thereare onlysubtledifferences primarilyinthe
type of granzyme expressed (i.e., B versus A) and only a single
FOXP3 isoform (Roncarolo and Battaglia 2007).
The two most common types of adaptive or induced TREG
are the T regulatory type 1 cells (Tr1) and TH3 TREGcells (Fig-
ure 2). Tr1 cells secrete IL-10 and also have an IL-10-
dependent induction process (Fujio et al. 2010; Beissert et al.
2006; Roncarolo et al. 2006). They are capable of secreting
high levels of IL-10 and TGF-b in the human and mouse and
also secrete low levels of IL-2, IL-5, and IFN-g (Fujio et al.
2010; Roncarolo et al. 2006). An important growth factor for
Tr1 cells is IL-15, which can support Tr1 cell proliferation even
without TCR activation (Fujio et al. 2010; Roncarolo et al.
2006). They are CD4þ, anergic, and proliferate poorly upon
antigen-specific activation which is likely due to the autocrine
production of IL-10 leading to suppression of proliferation
(Fujio et al. 2010; Roncarolo et al. 2006). The mechanism of
suppressive effects with Tr1 cells is soluble factor-based (i.e.,
IL-10), and the suppressive effects of Tr1 cells are negated
by anti-IL-10 neutralizing antibody in vitro (Fujio et al.
2010; Roncarolo et al. 2006). There is no specific marker for
Tr1 cells, although repressor of GATA-3 (ROG) shows poten-
tial utility (Fujio et al. 2010; Roncarolo et al. 2006), but it is not
specific for this cell population.
TH3 cells secrete TGF-b and IL-10 and express FOXP3
(Beissert et al. 2006). They are induced from naı ¨ve CD4þ
T-cells by TGF-b and have an important role in oral tolerance
to non-self antigens and negating autoimmune reactions (Wan
and Flavell 2007). TH3 cells secrete TGF-b, which has immu-
nosuppressive effects acting through a soluble-factor mechan-
ism. TH3 cells have a reciprocal relationship with TH17 cells
(Korn et al. 2009; Nistala and Wedderburn 2009). TH17 cells
are highly proinflammatory and are thought to be important in
autoimmune disease (Abraham and Cho 2009; Crome et al.
2009; Korn et al. 2009; Nistala and Wedderburn 2009;
Romagnani et al. 2009); this will be discussed in detail later
in the article. There has been no specific surface marker
identified for TH3 cells, although expression of FOXP3 is
induced in TH3 cells (i.e., cannot differentiate from nTREG)
(Korn et al. 2009).
FIGURE 1.—There has been a steady increase in the number of regula-
tory T-cell (TREG) citations in the scientific literature from the mid-
1990s to the present (from Scopus). The key milestones in TREG
research are highlighted on the horizontal axis.
Vol. 40, No. 2, 2012REGULATORY T-CELLS187
OTHER T-CELLS WITH REGULATORY FUNCTION
Other T-cell populations that have been shown to have reg-
(double-negative [DN]). CD8þTREG(Kapp and Bucy 2008;
Beissert et al. 2006; Nakamura et al. 2003; Ke and Kapp
1996; Jiang et al. 1992) are CD8þT-cells that have regulatory
activity on CD4þTH1 cells. This TREGpopulation inhibits
priming of CD8þT-cells, CD4þT-cells, and antibody
responses against oral antigens. They are induced by manipula-
tion of co-stimulatory molecule interactions (i.e., CD137 and
CD40) primarily, although antigens introduced into immune
privileged sites (i.e., anterior chamber of eye leading to anterior
chamber-associated autoimmune deviation [ACAID]) elicit
CD8þTREGdue to high levels of TGF-b and IL-10 at this site
(Biros 2008; Niederkorn 2008). They have also been shown to
arise spontaneously in Experimental Allergic Encephalitis
(EAE) and murine Ovalbumin-induced oral tolerance models
and in tumor-bearing hosts due to IL-10 secretion by APCs
in the neoplasm. They are well characterized in rodents, but
their importance in humans is uncertain. CD8þCD28-TREG
(Cortesini et al. 2001; Chang et al. 2002) are FOXP3?and are
induced by inhibitory immunoglobulin-like transcript (ILT)-3
and ILT-4 receptor expression on dendritic cells and down-
regulation of CD80 and CD86. They interfere with the
CD28/B7 pathway with allogeneic antigens and CD154/
CD40 pathway with xenogeneic antigens. CD8þCD28?TREG
regulate immune responses by direct cell-to-cell interactions
with dendritic cells; therefore they are directly tolerogenic.
This cell population has been described in rodents and humans.
CD3þCD4?CD8?(DN) TREGare TCRabþ, CD4?, and CD8?
(therefore DN). They promote the development of tolerance
and regulate autoimmunity (Strober et al. 1996). This cell pop-
ulation has been defined and characterized in heart xenotrans-
plantation experiments (Chen et al. 2003, 2005; Zhang et al.
2006; Ma et al. 2008). DN TREGare poorly suppressive in naı ¨ve
mice but are actively suppressive in recipients of heart xeno-
grafts co-transferred with donor DN Tregs. They have been
shown to prevent heart xenograft rejection (i.e., transplant tol-
erance). It has been shown that antigen presenting cells (APC)
from tolerant mice had down-regulation of MHCII, CD40, and
B7 molecules, and likely were in an immature state. The
FIGURE 2.—TREGare either formed naturally by thymic differentiation (nTREG) or are induced in the periphery (iTREG) from naı ¨ve T-cells (TH0).
Types of iTREGinclude TH3, Tr1, which are CD4þCD25þFOXP3þ, and CD4þCD25-FOXP3þiTREG. The main effector CD4þ subsets are TH1,
TH2, and TH17. The cytokines that are important in inducing these cells from TH0 cells and the cytokines that these cells secrete are listed.
significance of DN TREGin nontransplant scenarios is therefore
uncertain and more data is needed.
Natural killer T-cells coexpress NK receptors (i.e., NK1.1 or
CD161) as well as a conserved T-cell receptor (TCR) molecule
(Beissert et al. 2006), Va14. Va14 pairs with Vb8 in the mouse
andVb11inthe humanandare restricted byCD1d,aconserved
MHCI-type molecule that presents glycolipids. NKTREGcells
are generated in the thymus and can be CD4þ, CD8þ, DN, or
double positive (DP). In mice, DN NKTREGare the dominant
phenotype (Zeng et al. 1999); while in humans, CD4þNKTREG
marily in rodent xenogeneic transplantation models (Ikehara et
al. 2000). They secrete IL-4, IL-10, TGF-b (paracrine), and
induce cytotoxicity by cell-to-cell contact, and the targets of
NKTREGcells are T-cells and APCs (Ikehara et al. 2000; Zeng
et al. 1999). The significance of NKTREGcells in immune regu-
lation outside of xenotransplantation is uncertain.
gd T-CELLS (MUCOSAL/INTRAEPITHELIAL)
Gamma delta (gd) T-cells, which express TCRgd and have a
mucosal distribution, are primarily suppressive and are associ-
ated with mucosal tolerance (Kapp and Ke 1997; Ke et al.
1997; Mowat 1994), but can also function to regulate autoim-
munity (Weiner et al. 1994), elicit ACAID in the eye (Biros
2008), and function in tumor immunity (Seo et al. 1999). These
cells do not recognize peptides associated with MHC receptors
(Kronenberg 1994); alternatively they recognize cellular pro-
teins including heat shock proteins (Munk et al. 1989) and non-
classic MHC molecules like CD1 (Brossay et al. 1998) and Qa-
1 (Vidovic et al. 1989). gd T-cells that infiltrate tumors have
FIGURE 3.—TREGcan cause suppression of effector T-cells and dendritic cells through secretion of soluble factors including IL-10, TGF-b,
fibrinogen-like protein-2 (FLG-2), granzyme A or B and perforin, and adenosine. The figure shows the signaling pathways and the suppressive
effects on effector T-cells and dendritic cells.
Vol. 40, No. 2, 2012REGULATORY T-CELLS189
cytokine profiles that are similar to Tr1 cells (i.e., IL-10 and
TGF-b) and inhibit the immune response against neoplasia
(Kapp et al. 2004; Seo and Tokura 1999; Seo et al. 1999; Kapp
and Ke 1997; Ke et al. 1997). Therefore, gd T-cells are a minor
population of mucosal and intratumoral T-cells with regulatory
activity and major roles in mucosal tolerance and tumor
A NON-T LYMPHOCYTE SUBSET WITH REGULATORY FUNCTION
Another cell type exhibiting regulatory activity is the regu-
latory B-cell (BREG) which has been recently reviewed by Li et
al. (2011). BREGcells are beyond the scope of this review, but
are mentioned here briefly to showcase the diversity of lym-
phocytes with regulatory function in the immune system and
for completeness-sake. BREGcells have been shown in various
animal disease models, including collagen-induced arthritis
(Trentham et al. 1977), experimental autoimmune encephalitis
(EAE) (Wolf et al. 1996), nonobese diabetes mellitus (Tian et
al. 2001), inflammatory bowel disease (Mizoguchi et al. 2000),
and systemic lupus erythematosus (Lenert et al. 2005), and
function primarily to regulate T-cells, B-cells, and NKT cells.
BREGcells act by soluble factor-based means by secreting IL-
10 (Reigert and Bar-Or 2008), TGF-b (Takenoshita et al. 2002)
and producing anti-inflammatory antibodies (Diamond et al.
2003). Cell-cell interactions are an important mode of BREG
regulatory function and include CD1d binding (Yanaba et al.
2008) by CD1dhiCD5þcells and CD40/CD40 L interactions
(Quezada et al. 2004). More research into specific roles for
BREGis needed to more completely define this cell type.
TREGcells function at sites of inflammation in close spatial
proximity to effector T-cells, thus allowing direct interactions
between the two cell types (Askenasy et al. 2008). TREGcells
FIGURE 4.—TREGcan cause suppression of effector T-cells and dendritic cells through cell contact-based mechanisms, including galectin-1 bind-
ing, CTLA-4 binding to CD80/86, lymphocyte activation gene-3 (LAG-3) binding to MHCII molecules, and neuropilin-1 binding. The figure
shows the ligands, receptors, and the suppressive effects on effector T-cells and dendritic cells.
190 PETERSONTOXICOLOGIC PATHOLOGY
are attracted by inflammatory signals, although they are less
chemotactic than effector T-cells (Siegmund et al. 2005).
When arriving at the site of inflammation, TREG down-
regulate chemotactic receptors and adhesive interactions to
arrest further migration (Siegmund et al. 2005). nTREGare
mitotically inactive under basal conditions (Kuniyasu et al.
2000). Activated TREG-mediated immune suppression is
antigen-nonspecific (i.e., not MHCII-restricted) (Bienvenu et
al. 2005), although induction of suppressor function in TREG
requires antigen-specific stimulation (i.e., via TCR) leading
to tissue-specific suppressive effects (‘‘bystander’’ suppres-
sion) (Shevach 2009; Weiner 1997). TREGhave also been asso-
ciated with regulating lymphopenia-induced lymphocyte
proliferation, which is a normal homeostatic event that occurs
after lymphopenia of numerous causes including those of a
physiologic and pathologic nature. Lymphopenia-induced lym-
phocyte proliferation does not increase the number of naı ¨ve
lymphocytes, but instead the number of memory T-cells
increase, which might result in potential self-reactive clones
and subsequent autoimmunity. TREGhave an important role
in regulating this process (Datta and Sarvetnick 2009).
EFFECTS OF ACUTE AND CHRONIC STRESS ON HUMAN TREG
beta 1-adrenergic and glucocorticoid areceptors are overex-
pressed in TREGversus effector T-cells (Freier et al. 2010; Ata-
nackovic et al. 2006). These receptors likely mediate TREG
responses to stress (Freier et al. 2010; Atanackovic et al.
2006). TREGnumbers have been shown to decrease with acute
physiologic stress (Freier et al. 2010; Atanackovic et al. 2006;
Atanackovic et al. 2002). Based on these findings, chronic
stress could exacerbate autoimmune disease and other inflam-
matory conditions (Freier et al. 2010).
MODES OF ACTION OF TREGREGULATION
TREGhave primary effect on T-cells and/or dendritic cells
by three main regulatory modes of action: (1) soluble factors
(Figure 3), (2) cell-to-cell contact (Figure 4) and (3) competi-
tion for growth factors (local effect) (Figure 5). Soluble factors
include IL-10 and TGF-b with direct suppressive effects on
effector T-cells (Joetham et al. 2007; Annacker et al. 2003),
fibrinogen-like protein-2 (FLG-2) with apoptotic effects on
effector T-cells and prevention of maturation of dendritic cells
(Shalev et al. 2009), granzyme A/B and perforin apoptotic
effects on effector T-cells (Grossman et al. 2004), and produc-
tion of adenosine by CD39/73 cleavage of ATP, which causes
cell cycle arrest in effector T-cells and prevention of matura-
tion and decreased antigen presenting capability in dendritic
cells by binding tothe A2Areceptor on these celltypes (Deaglio
FIGURE 5.—TREGcan cause suppression of effector T-cells by providing competition for growth factors (primarily IL-2), which is essential for
their survival in the periphery. TREGexpress multiple copies of IL-2R/CD25, which bind up local IL-2 competing with effector T-cells for
IL-2 stimulation. The figure shows the ligands, receptors, and the suppressive effects on effector T-cells.
FIGURE 6.—FOXP3 and CD3 dual immunofluorescence on tissue from
the red cytoplasmic labelling (CD3þ). The image shows a group of T-
cells with several FOXP3þTREGcells in the field. 400? magnification.
Vol. 40, No. 2, 2012REGULATORY T-CELLS191
et al. 2007). Galectin-1 binding to effector T-cells and dendritic
cells resulting in cell cycle arrest and/or apoptosis (Garin et al.
2007), CTLA4 and CD80/86 binding to dendritic cells causing
decreased costimulation and decreased antigen presentation
(Read et al. 2000), lymphocyte activating gene-3 (LAG3 or
CD223, a CD4 homolog) binding to MHCII molecules on den-
antigen presentation capability (Workman and Vignali 2004),
antigen presentation (Sarris et al. 2008) are examples of cell
contact-based mediated effects of TREG. TREGcan also deny
al. 2007). Therefore, TREGhave a variety of mechanisms that
can be used to suppress an immune response and are able to use
all or a subset of these mechanisms simultaneously.
TREGMARKERS (IMMUNOHISTOCHEMISTRY, FLOW CYTOMETRY,
INTRAVITAL MICROSCOPY, AND MIRNA ANALYSIS)
Immunofluorescent and chromogenic IHC and flow cyto-
metry (de Boer et al. 2007; Roncador et al. 2005) for the
following markers has been shown to be useful in characteriz-
ing TREGin tissue sections and isolated cells: FOXP3 (Figures
6, 7, 8, and 9), CD25 (alpha chain of IL2R) (Figures 7, 8, and
9), and CD62L (L-Selectin) (Figures 6 and 7). Intercellular
interactions of TREGcells and effector T-cells have been visua-
lized in rodent models using intravital microscopy (Tang and
Krummel 2006). Differences in micro-RNA (miRNA) in
human TREGversus effector T-cells have been shown (Hezova
et al. 2010). There was a significant increase in miR-146a and
decreases in miR-20b, 31, 99a, 100, 125b, 151, 335, and 365 in
TREGwhen compared with effector T-cells. The pattern of
increased expression of miR-146a in combination with
decreases in the other miRNAs is a possible marker of TREG.
FOXP3: FUNCTION AND BIOLOGY
FOXP3, a forkhead box transcription factor, is a molecular
marker and cell lineage specification factor for TREG(Fontenot
et al. 2003; Hori et al. 2003). FOXP3 is critical for the thymic
differentiation of ab TCRþthymocytes into nTREG(Sakaguchi
et al. 2008; Fontenot et al. 2003; Hori et al. 2003). The FOXP3
gene product is referred to as scurfin and functions by
FIGURE 7.—Flow cytometry was performed on blood from male and female CD-IGS rats to evaluate the percentage of
CD4þCD25þCD62LþFOXP3þ(nTREGand iTREG) cells. The image highlights the gating strategy and steps used in this evaluation.
regulating a set of genes required for (1) the suppressor activity
of TREG, proliferative, and metabolic fitness of TREGand (2)
repressing alternative T cell differentiation pathways (Brun-
kow et al. 2001). It has been shown that a higher level of
expression of FOXP3 is sufficient for suppressive activity of
non-TREGT-cells in rodents (Sakaguchi et al. 2008; Lourenco
and La Cava 2011). Several pathways are important in the
induction of FOXP3 expression including TGF-b, IL-2, reti-
sphingosine-1-phosphate receptor 1 (S1PR1) (Figure 10).
TGF-b is essential for induction of FOXP3 expression in naı ¨ve
T-cells (i.e., iTREGincluding TH3 cells), but in the human it
does not necessarily confer regulatory function (Lourenco and
La Cava 2011).
and signaling throughthe
The ‘‘Scurfy’’ Mouse
In mice (B.Cg-Foxp3sf),a frameshift mutation inthe FOXP3
gene results in a truncated protein lacking the forkhead domain
and is responsible for the ‘‘Scurfy’’ phenotype (Clark et al.
1999; Ochs et al. 2002; Patel 2001). Scurfy is an X-linked
recessive mutation in the mouse that is lethal in hemizygous
males 16 to 25 days after birth (Ochs et al. 2002; Godfrey
et al. 1991). Scurfy mice have an overproliferation of CD4þ
T-lymphocytes and other cell types due to uncontrolled
cytokine secretion with extensive multiorgan infiltration due
to lack of functional TREG(Ochs et al. 2002; Godfrey et al.
1991). Lymphadenopathy, parenchymal organ infiltration, and
subcutaneous inflammatory cell infiltrates (e.g., ear pinnae) are
prominent (Figures 11 and 12) (Godfrey et al. 1991). The phe-
notype of Scurfy mice is similar to mouse models that lack
CTLA-4 or TGF-b, and in humans with immunodysregulation,
polyendocrinopathy and enteropathy, X-linked syndrome
(IPEX) (Clark et al. 1999; Ochs et al. 2002; Patel 2001).
IPEX in Humans
IPEX is a rare human multisystemic autoimmune disease
(X-linked: females are carriers, males have the disease) that
is linked to the dysfunction of the transcriptional activator
FOXP3 which results in dysfunction of regulatory T-cells with
subsequent autoimmunity (Ochs et al. 2002). Mutations in the
forkhead domain of FOXP3 disrupt FOXP3 protein: DNA
interactions (Ochs et al. 2002; Patel 2001). The disorder man-
ifests with enteropathy, food allergy, massive lymphoprolifera-
tion, psoriasiform or eczematous dermatitis, nail dystrophy,
autoimmune endocrinopathies (primarily insulin-dependent
diabetes mellitus and autoimmune thyroid disease), Coombs-
positive anemia, autoimmune thrombocytopenia, autoimmune
neutropenia, tubular nephropathy, and autoimmune skin condi-
tions such as alopecia universalis and bullous pemphigoid
(Ochs et al. 2002). Treatment for IPEX includes immunosup-
pressive agents (e.g., cyclosporin A, FK506) alone or in com-
bination with steroids (Ochs et al. 2002). There has been
limited success in treating the syndrome with bone marrow
transplantation (Ochs et al. 2002).
FIGURE 8.—The graphs of flow cytometry data using markers useful in
characterizing TREG show the percentage of CD3þCD4þcells,
CD62Lþcells in the CD3þCD4þpopulation, CD25þcells in the
CD3þCD4þpopulation, and FOXP3þcells in the CD3þCD4þpopu-
lation cell subsets in male and female rats. There was no definitive dif-
ference in cell percentages between males and females in this
FIGURE 9.—The graphs of flow cytometry data show the percentages of
FOXP3þcells in the CD4þCD25þpopulation and CD25þcells in the
CD4þFOXP3þpopulation. The majority (approximately 80%) of
CD4þCD25þcells are also FOXP3þ, which would include nTREGand
TH3 cells. Only a minority (approximately 30%) of CD4þFOXP3þ
cells are also CD25þ, which means approximately 70% of the
CD4þFOXP3þ population likely represents an iTREGcell population.
Vol. 40, No. 2, 2012 REGULATORY T-CELLS193
TH17 CELLS AND TREG: A RECIPROCAL BALANCE
TH17 cells are a subpopulation of effector T cells that are
very pro-inflammatory and implicated in allergic/autoimmune
disease (Abraham and Cho 2009; Crome et al. 2009; Korn et al.
2009 and 2007; Nistala and Wedderburn 2009; Romagnani et
al. 2009). TH17 cells secrete IL-17, IL-22, IL-21 and IL-17F
(Korn et al. 2009, 2007; Veldhoen et al. 2006; Bettelli et al.
2006; Mangan et al. 2006). TGF-b is critical in induced/adap-
tive TREGcell (i.e., TH3 TREG) differentiation as well as devel-
opment of TH17 effector cells (Wan and Flavell 2007) (Figure
13). Naı ¨ve T-cells develop into a ‘‘transitional’’ cell-type that
expresses FOXP3 (TH3 TREGcell marker) and ROR-gt and
RORa (TH17 effector cell markers) in the presence of TGF-b
(Figure 14) (Nistala et al. 2009; Korn et al. 2007; Veldhoen
et al. 2006; Bettelli et al. 2006; Mangan et al. 2006). Further
exposure to TGF-b or IL-6/IL-21 in the inflammatory milieu
leads to TH3 cell differentiation or TH17 effector cell differen-
tiation (Zhou et al. 2008; Korn et al. 2007; Veldhoen et al.
2006; Bettelli et al. 2006; Mangan et al. 2006). Therefore, a
reciprocal balance between TH17 cells and TH3 TREGcells
exists (Bettelli et al. 2006) where IL-6 and IL-21 inhibit TH3
cell differentiation to the benefit of TH17 cells (Korn et al.
2009, 2007) and TGF-b elicits TH3 cell differentiation to the
disadvantage of TH17 cells (Zhou et al. 2008).
ROLE OF TREGIN MUCOSAL SITES
Oral tolerance is an important mechanism in the mucosal
immune system whereby oral administration of a specific anti-
gen induces unresponsiveness of systemic immune responses
with the same or different antigens (Chen et al. 2007; Weiner
et al. 1994). TGF-b is critical for the induction of oral tolerance
(Chen et al. 2007; Weiner et al. 1994). Low-dose of antigen
cells leads to ‘‘dampening,’’ a paracrine effect, of the immune
response due to TGF-b derived from nTREG, TH3, and Tr1 cells.
High-dose of an antigen is associated with mucosal T-cell
anergy and apoptosis (Chen et al. 2007; Weiner 1997; Weiner
FIGURE 10.—The figure highlights several pathways of induction of FOXP3 expression including the ligands, receptors, and/or intermediary steps
in induction of gene expression. A schematic of the FOXP3 gene shows the promoters and important transcription factors that bind to the pro-
moters. There are 11 coding exons in both the mouse and human FOXP3 genes.
et al. 1994). At mucosal sites, TREG are responsible for
‘‘bystander’’ suppression, where TGF-b produced at mucosal
sites during development of oral tolerance regulates immune
responses on T-cells, B-cells, NK cells, macrophages, and den-
dritic cells in an antigen nonspecific manner (Weiner 1997).
Animal models of inflammatory bowel disease (IBD) have been
used to study TREGand mucosal tolerance (Blumberg 2009).
Researchers have adoptively transferred CD4þCD45RBhigh
T-cells into SCID or Rag-1 mice (Powrie et al. 1994) or have
colitis rodent model (Te Velde et al. 2006). A negative effect of
TREGin mucosal immunity is the secretion of TGF-b from TREG
in the face of IL-6 and IL-21 in the inflammatory milieu. TGF-b
of TREGdifferentiation, resulting in more severe inflammation
‘‘bystander’’ suppression of mucosal sites with resident gd
ROLE OF TREGIN INFLAMMATORY/INFECTIOUS DISEASE
Both natural and induced TREGare important in suppressing
inflammation in infectious disease (Figure 15; Abraham and
Medhitzov 2011; Liu et al. 2010; Jager and Kuchroo 2010;
Scholzen et al. 2009; de Boer et al. 2007; Belkaid and Rouse
2005). nTREGare activated by microbial infections and prod-
ucts of the resulting inflammatory milieu. TREGexpress toll-
like receptors 4, 5, 7, and 8, allowing the cells to respond to
bacterial products such as lipopolysaccharide leading to
increased TREGsurvival and proliferation (Caramahlo et al.
2003). Probiotics have been shown to elicit activation of TREG,
which then suppress TH1-mediated colitis in the TNBS mouse
colitis model (Di Giacinto et al. 2005). Microbes are also able
to manipulate antigen-presenting cells leading to priming/acti-
vation of TREGand elicit secretion of chemokines and cyto-
kines that favor induction, priming, recruitment, and survival
of TREG, thus using TREGfor their own benefit (Cabrera et al.
2004; Hesse et al. 2004; Caramahlo et al. 2003; Belkaid et al.
2002). These microbial organisms include virus (HIV [Aandahl
et al. 2004], HCV [Cabrera et al. 2004], CMV [Aandahl et al.
2004], FIV [Vahlenkamp et al. 2004], Herpes simplex virus
[Suvas et al. 2004]), bacteria (Helicobacter spp. [Lundgren et
al. 2003; Kullberg et al. 2002] and Listeria monocytogenes
[Kursar et al. 2002]), fungi (Candida albicans [Montagnoli et
al. 2002] and Pneumocystis carinii [Hori et al. 2002]), and pro-
tozoal (Leishmania major [Belkaid et al. 2002], Schistosoma
mansoni [Hesse et al. 2004] and Plasmodium spp. [Scholzen
et al. 2009]). TREGcan become infected with viral pathogens
(e.g., HIV and FIV), and it is uncertain if they remain suppres-
sive when infected (Joshi et al. 2004; Oswald-Richter et al.
2004). Animal infection models and infected humans show
accumulation of TREGat the site of infection with leishmania-
sis, herpes simplex virus, and schistosomiasis (Hesse et al.
2004; Suvas et al. 2004; Belkaid et al. 2002), in the peripheral
blood with HIV and HCV (Oswald-Richter et al. 2004) and in
lymphoid organs of HIV patients (Andersson et al. 2005). A
good example of the role of TREGin infection would be Plas-
modium spp. that causes malaria in human patients and rodent
models and is associated with increased numbers of TREG(i.e.,
natural and induced), which have a direct association with the
parasitic load in the host. TREGaccumulate in areas of inflam-
mation induced by the organisms and inhibit TH1 responses
(i.e., both protective and pathologic) primarily by the suppres-
sive effects of IL-10 and TGF-b, and ultimately result in either
clearance of the organism and resolution or clinical cases of
malaria depending on the level of TREGcell induction and the
organism’s manipulation of this lymphocyte population
(Scholzen et al. 2009). TREGhas also been shown to be a factor
FIGURE 11.—The ‘‘Scurfy’’ mouse has a frameshift mutation in the
forkhead box transcription factor, FOXP3. This results in early lethal-
ity due to unregulated inflammation (parenchymal organs, subcutis,
and lymphoid organs are the main sites of inflammatory cell accumu-
lation) due to the lack of functional TREGcells. Note the gross pathol-
ogy image of a Scurfy mouse with marked hepato- and splenomegaly
and evidence of subcutaneous lesions at the tail base.
FIGURE 12.—Note the prominent mononuclear inflammatory cell accu-
mulation in the dermis and upper subcutis from a ‘‘Scurfy’’ mouse.
Hematoxylin and eosin, 200? magnification.
Vol. 40, No. 2, 2012 REGULATORY T-CELLS195
in the development of silica-induced pulmonary fibrosis in a
mouse model. Depletion of TREGleads to the maintenance of
a TH1 response and subsequent attenuation of fibrosis versus
the shift to a TH2 response that normally predisposes to devel-
opment of fibrosis in this model (Liu et al. 2010). The balance
between effector T-cells and TREGis critical in determining the
immune response to pathogens (Figure 16).
ROLE OF TREGIN ORGAN TRANSPLANTATION
TREGhave been shown to have a pivotal role in transplant
tolerance leading to graft acceptance and prevention of rejec-
tion in xenotransplantation (Muller et al. 2009; Zhang et al.
2009; Le and Chao 2007; Yong et al. 2007). Adoptive transfer
of TREGinto mice has been shown to inhibit, delay, and prevent
development of graft versus host disease (GVHD) (Taylor et al.
2002; Cohen et al. 2002) as well as to prevent allograft (i.e.,
pancreatic islet cell transplant) rejection (Battaglia et al.
2006; Gregori et al. 2001). Post-transplant immunosuppressive
drugs have been shown to have variable effects on TREG.
Mycophenolate mofetil increases TREGnumbers and induces
transplant tolerance when given with vitamin D (Han et al.
2005; Huang et al. 2003; Gregori et al. 2001), calcineurin inhi-
bitors (e.g., cyclosporine, tacrolimus, sirolimus) decrease TREG
and abrogate transplant tolerance (Shoji et al. 2005; Larsen et
al. 1996), and Rapamycin and methylprednisolone do not seem
to have effects on TREGor transplant tolerance (Blaha et al.
ROLE OF TREGIN PREGNANCY
Natural TREG, CD8þTREG, TH3 cells, and Tr1 cells have
been shown to be important in immune tolerance during preg-
nancy (Guerin et al. 2009; Trowsdale and Betz 2006; Aluvihare
FIGURE 13.—There is a reciprocal relationship between TH17 (proinflammatory effector T-cell subset that are associated with autoimmune dis-
ease) and TH3 cells (iTREG that are CD4þCD25þFOXP3þ), which is elicited during differentiation of naı ¨ve TH0 cells, but final differentiation
depends on the cytokine milieu. TGF-b is critical in differentiation of both cell types but depends on the cytokine milieu. The schematic shows the
cytokine pathways in the differentiation of TH17 and TH3 cells from TH0 effector cells.
196 PETERSONTOXICOLOGIC PATHOLOGY
et al. 2004). IL-2/STAT5 signaling and 17-b-estradiol (E2)
during pregnancy cause induction of FOXP3 and increased
TREGnumbers (i.e., humans and animal models) by induction
of FOXP3þ iTREGand TREGcell expansion (Fainboim and
Arruvito 2011). Fetal alloantigen (Zhao et al. 2007) and placen-
tal trophoblast antigens such as heat shock protein-60, carci-
noembryonic antigen, CD274 (i.e., ligand for programmed
cell death-1 receptor), and human leukocyte antigen-G
(Taglauer et al. 2008; Yang et al. 2006; Shao et al. 2005;
LeMaoult et al. 2004) are also important in TREGcell induc-
tion/activation and expansion during early pregnancy. Paternal
alloantigens and TGF-b in seminal fluid can also induce/acti-
vate TREG (Robertson et al. 2009; Robertson and Sharkey
2001). During the course of pregnancy, there are increased
numbers of CD4þCD25þcells in the circulation in early preg-
nancy, reaching anapex duringthe second trimester anddeclin-
ing until the postpartum period where they reach a number that
is nearly at the prepregnancy level (Heikkinen et al. 2004).
TREGare significantly decreased in deciduas from spontaneous
vaginal deliveries when compared with Caesarean section,
indicating a potential role of declining TREGin fetal delivery
(Xhao et al. 2007; Sindram-Trujillo et al. 2004). The ovary nor-
mally contains a population of thymic-derived nTREG(Samy et
al. 2005), and the testis has been shown to have a locally
induced iTREGpopulation that are specifically targeted to testi-
cular antigens and contributes to testicular immune privilege
(Nasr et al. 2005). TREGhave been shown to be necessary for
immune tolerance of the conceptus, oocytes, and spermatozoa
(Trowsdale and Betz 2006). There are increased numbers of
TREGin the blood, decidual tissue, and lymph nodes draining
the uterus (Tilburgs et al. 2008; Thuere et al. 2007). TREGcell
depletion and/or decreased function have been shown to lead to
loss of pregnancy in rodent models (Aluvihare et al. 2004) and
in patients who have miscarried (Yang et al. 2008; Sasaki et al.
2004). Infertility, miscarriage, and pre-eclampsia have been
linked with decreased TREGnumber and/or function (Guerin
et al. 2009; Tilburgs et al. 2008; Thuere et al. 2007; Trowsdale
and Betz 2006; Aluvihare et al. 2004).
FIGURE14.—This schematic focuses on the differentiation pathways ofTH0 to TH17 and TH3 cells showing a transitional state (FOXP3þ,ROR-gtþ
and RORaþcell) where addition of certain cytokines/factors drives differentiation to TH3 cells (FOXP3þ) leading to suppression or TH17 cells
(ROR-gtþand RORaþ) resulting in inflammation/autoimmune disease.
Vol. 40, No. 2, 2012REGULATORY T-CELLS197
ROLE OF TREGIN CANCER
TREGhave been shown to dampen the cytotoxic immune
response (CD8þ/NK cell) as well as suppressing overall level
of host immune response to the neoplasm in cancer. Research-
ers have looked at numbers and location of TREGand the effects
on cancer patient outcome (Menetrier-Caux et al. 2009). They
showed that TREGare present in a large panel of solid tumors in
humans. TREGin pulmonary, pancreatic, gastric, hepatic, and
ovarian carcinomas have been correlated with a negative out-
come; while in B-cell lymphoma, head and neck cancer and
colonic carcinoma TREGhave been associated with a positive
patient outcome (Menetrier-Caux et al. 2009). TREGhad no
effect on survival with prostatic, renal, and squamous cell car-
cinomas (Menetrier-Caux et al.2009). In breast cancer, TREGin
peri-tumoral lymphoid aggregates were correlated with a
negative outcome, while intra-tumoral TREGwere associated
with a positive outcome. (Menetrier-Caux et al. 2009) The
results of vaccine-induced cancer T-cell mediated immu-
notherapy (i.e., cancer vaccine therapy) have proven to be dis-
appointing (Welters etal. 2008)due to the vaccine’s inabilityto
induce effector T-cells, TREGalready present in the tumor, and/
or induction of TREGby the vaccine itself. Based on these find-
ings, the success of cancer vaccine therapy will depend on the
balance between TREGand effector cells at the start of therapy
(Welters et al. 2008). Neoplastic TREGcells in Se ´zary Syn-
drome (leukemic variant of cutaneous T-cell lymphoma in the
human) express variant splices of FOXP3. The neoplastic cells
with variant FOXP3 are able to function as TREGand work to
dampen the host immune response against the neoplastic
T-lymphocytes (Krejsgaard et al. 2008).
FIGURE 15.—Duringan infection with a pathogenic organism complex networksdevelop where nTREGundergo antigen-specific expansion and Tr1
and iTREGcells undergo pathogen-specific differentiation in the presence of tolerogenic cytokines (IL-10 and TGF-b). All three types of TREG
cells then work to suppress effector T-cells in the inflammatory milieu by soluble factors, cell-to-cell contact, or competing for IL-2. nTREGcan
also induce infectious tolerance in dendritic cells. Together these result in suppression of immune responses to pathogens. Pathogens can also use
TREGfor their own benefit to evade the immune response.
TREGAS A MODULATORY TARGET IN DRUG DEVELOPMENT Download full-text
Pharmacological agents have been developed that control
TREGnumber and/or function (Ohkura et al. 2011; Nizar et
al. 2010; Thorburn and Hansbro 2010; Nandakumar et al.
2009; Tao et al. 2007; Yang et al. 2006). Molecules that
increase TREGnumber include rTGF-b, FTY720 (Fingolimod,
a S1P1 modulator), retinoic acids, histone deacetylase inhibi-
tors, and probiotics. TREGsuppressive function is augmented
with the following agents: recombinant IL-2, Rapamycin
(mTOR pathway inhibitor) in type-1-diabetes mellitus and
renal transplantation, and histone deacetylase inhibitors. His-
tone deacetylase inhibitors that are also known as chromatin
modifying agents (e.g., HDAC9) lead to increased TREGnum-
ber and function by a combination of FOXP3 protein acetyla-
tion, TREG chromatin remodeling, and promotion of TREG
development (Tao et al. 2007). While agents that block the
TGF-b and/or CTLA-4 signaling pathways decrease TREG
number and/or function. Aromatase inhibitors (e.g., Letrozole)
used for estrogen modulation in cancer chemotherapy regimens
and recombinant IL-2 lead to decreased TREGnumbers.
Biopharmaceutical agents have been developed to modulate
TREG(Khan et al. 2011; Ohkura et al. 2011; De Serres et al.
2011; Weber et al. 2009) including Tremelimumab (CTLA-4
modulation) in advanced melanoma which leads to increased
effector T-cell numbers with no change in TREG number,
CD25 modulation by Daclizumab in multiple sclerosis (MS)
and advanced breast cancer, Denileukin diftitox as a cancer
vaccine, LMB-2 and RFT5-SMPY-dgA in advanced melanoma
decrease TREGnumber, anti-GITR MoAb (DTA1) which leads
to decreased TREGnumber, agonist anti-OX40 MoAb (OX86)
and anti-FR4 MoAb which decrease TREGnumber, and Alem-
tuzumab (anti-CD52 MoAb) which increases TREGnumber.
Adoptive transfer of TREGhas been evaluated in clinical
trials of patients that have received human stem cell transplants
to induce transplant tolerance, patients with GVHD, and type 1
diabetes mellitus patients to ameliorate autoimmune disease
with a good patient safety profile and showed that the proce-
dure is feasible, yet efficacy still needs to be shown convin-
cingly (Riley et al. 2009; Trzonkowski et al. 2009).
TREG have both positive and negative functions in the
immune system. Positive aspects of TREGinclude suppression
of immune responses including autoimmunity, maintaining
immune homeostasis, as well as tolerance in xenotransplanta-
tion, at mucosal sites, and during pregnancy. Negative aspects
FIGURE 16.—The balance between effector cells and TREGcells in the inflammatory milieu dictate the course of and ultimate resolution or per-
sistence of the inflammatory response to an infectious agent. There are advantages and disadvantages for both the host and microbe if there is an
imbalance in the numbers/function of effector cells and TREG.
Vol. 40, No. 2, 2012 REGULATORY T-CELLS199