Glucocorticoid-Induced Tumour Necrosis Factor Receptor-Related Protein: A Key Marker of Functional Regulatory T Cells
Glucocorticoid-induced tumour necrosis factor receptor-related protein (GITR, TNFRSF18, and CD357) is expressed at high levels in activated T cells and regulatory T cells (Tregs). In this review, we present data from mouse and human studies suggesting that GITR is a crucial player in the differentiation of thymic Tregs (tTregs), and expansion of both tTregs and peripheral Tregs (pTregs). The role of GITR in Treg expansion is confirmed by the association of GITR expression with markers of memory T cells. In this context, it is not surprising that GITR appears to be a marker of active Tregs, as suggested by the association of GITR expression with other markers of Treg activation or cytokines with suppressive activity (e.g., IL-10 and TGF-), the presence of GITR+ cells in tissues where Tregs are active (e.g., solid tumours), or functional studies on Tregs. Furthermore, some Treg subsets including Tr1 cells express either low or no classical Treg markers (e.g., FoxP3 and CD25) and do express GITR. Therefore, when evaluating changes in the number of Tregs in human diseases, GITR expression must be evaluated. Moreover, GITR should be considered as a marker for isolating Tregs.
Regulatory T cells (Tregs) are specialized cells that control immune responses to pathogens and mediate immunological self-tolerance and homeostasis, as initially described by Sakaguchi et al. . Multiple subsets of Treg populations have been described in the literature.
Naturally occurring Tregs are derived from the thymus [2, 3]. Thymus-derived Tregs (tTregs) are characterized by the expression of the transcription factor forkhead box P3 (FoxP3) and the interleukin- (IL-) 2 receptor α-chain CD25 . Most mouse tTregs also express glucocorticoid-induced tumour necrosis factor (TNF) receptor-related protein (GITR, TNFRSF18, CD357) and OX40 (TNFRSF14, CD134), both of which are members of the TNF receptor superfamily . Although tTregs are considered to be naïve T cells, several human tTregs are derived from expanded populations of tTregs and are CD45R0+ (e.g., memory T cells) . tTregs are anergic in vitro (e.g., do not proliferate in response to TCR triggering) but appear to be a proliferating population in vivo.
Identification of tTregs is more complex in humans than in mice, because FoxP3 is transiently expressed even by activated T cells. Further, FoxP3 immunostaining requires permeabilization of cells, preventing functional experiments. CD25 is a useful surrogate marker for Tregs. However, CD25 is highly expressed in activated T cells [7–10]. The notion that tTregs exhibit higher expression levels of CD25 ( or ) is theoretically interesting but difficult to demonstrate in a standardized manner due to differences in the affinities of anti-CD25 antibodies (Abs) and flow cytometers. Furthermore, immunohistochemistry cannot discriminate between CD25+ and cells. In addition, some Tregs are not CD25+FoxP3+ . Thus, additional markers have been examined with CD25. For example, tTregs have been identified based on CD4+CD25+CD127−, CD4+CD25+CD49d−CD127−, CD4+CD45RA+, or CD4+CD45RA− status [12–14]. Despite some evidence of plasticity, tTreg phenotype and function remain fairly stable .
Peripherally derived Tregs (pTregs) form a more heterogeneous Treg subset and are derived from CD4+ effector T cells that become activated in response to microenvironmental signals . Transforming growth factor- (TGF-) β-producing Th3 cells and IL-10-producing Tr1 cells are examples of pTreg cells [2, 3, 16]. Several studies demonstrate that some pTreg subsets express low to undetectable levels of FoxP3 and/or CD25. For example, Bacchetta et al. showed that a peripheral CD4+ T cell subset expressed FoxP3 and low levels of CD25 while having regulatory activity . pTregs exhibit a high degree of plasticity and can expand in response to specific needs [18, 19].
Although αβ CD4+ Tregs are the best characterized Tregs, many immune cells demonstrate regulatory functions. Among the T lymphocytes, there are CD8+ Tregs, CD3+CD4−CD8− double-negative Tregs, and γδ CD4+ Tregs.
Tregs are characterized by the expression of specific surface markers, some of which mediate immune suppression (Table 1). Tregs produce factors, such as IL-10, IL-35, granzyme B, and TGF-β, which also mediate immune suppression (Table 1). Suppression of effector T cells by tTregs occurs via cell-to-cell contact, whereas soluble factors mediate T cell suppression by pTregs. However, Tregs are highly complex cells. For example, CTLA-4, one of the molecules involved in cell-to-cell inhibition, is expressed by several pTregs. In addition, differences exist between the expression of Treg markers and mechanisms of regulation by mouse and human Tregs.
Expression levels of Treg markers differ among Treg subsets; many markers are not exclusively expressed in Tregs and are also markers of activated effector T cells. This complexity and the differences between human and mouse Tregs make it difficult to evaluate the roles of Tregs in human diseases and can yield contrasting results. The best molecular marker(s) to study active Tregs in autoimmune diseases and tumours remain unclear. In this review, we present data suggesting that GITR is a crucial marker of Tregs, especially functional Tregs, and can be used as a marker to purify and evaluate the number of active Tregs.
2. Expression of GITR in T Cells
Schaer et al., Clouthier and Watts, and our group suggested that GITR is expressed at intermediate levels in murine and human CD4+CD25− naïve effector T cells [20–22]; similar levels were found in CD8+ cells. However, reinterpretation of published data suggests that this is not completely true.
After murine GITR was cloned, our group used a goat IgG polyclonal anti-GITR antibody (AF524, R&D System) to evaluate GITR expression in murine T cells [23, 24]. The antibody was specific for GITR, as demonstrated by the absence of staining in GITR knockout cells. All CD4+ cells expressed GITR. Similar results were obtained by another group with a rat IgG2a monoclonal anti-GITR antibody (DTA-1) [25, 26]. Conversely, Uraushihara et al., assessing GITR expression on murine CD4+ T with the DTA-1 antibody, found that most CD4+CD25− T cells did not express GITR; however, a small percentage of GITR-expressing cells were found in a CD25− subpopulation . The basis for the contradictory results remains unclear but may be due to preincubation of cells with FcγR-blocking antibody, although theoretically this should not affect T cell staining. However, the latter study may be more reliable, as the results were similar to what is observed in humans and mice. Thus, a small population of CD4+CD25− cells appears to express GITR at levels that are sufficient to facilitate isolation of this specific cellular subset [27, 28].
Our group obtained human data consistent with those described by Uraushihara et al. Among populations of human peripheral CD4+CD25− cells, GITR+ cells possess regulatory activity and exhibit a phenotype similar to memory and Treg cells [10, 29]. Consistent with our results, a previous study reported a small subpopulation of CD4+CD25−GITR+ cells among human peripheral blood lymphocytes . Therefore, it is reasonable to conclude that GITR is expressed in such a small population of CD4+ effector T cells and that depletion of GITR+ cells will not eliminate most CD4+ cells [27–29]. Indeed, Snell and colleagues have described distinct subsets of and cells . Furthermore, real-time PCR experiments in sorted human subpopulations demonstrated that CD4+CD25−GITR− cells express GITR mRNA at levels that are 10-fold lower than those in CD4+CD25−GITR+ cells . Therefore, we propose that naïve CD4+ cells are cells even if some antibodies at some experimental conditions suggest that most CD4+ and CD8+ cells are GITR+.
CD4+ T cells express high levels of GITR following activation. Studies suggest that GITR upregulation occurs rapidly following CD4+ T cell activation and peaks after one day to three days [25, 32, 33]. However, GITR does not appear to be a marker of long-term activation [10, 34]. CD8+ T cells express high levels of GITR following activation too . As demonstrated for the first time by Shimizu et al. and McHugh et al., GITR is expressed at high levels and provides regulatory functions in peripheral and thymic CD4+CD25+CD8− Tregs [26, 35] and several other Treg subsets, as discussed below.
3. GITR Participates in Costimulation of Effector T Cells
GITR is triggered by the ligand GITRL, which is mainly expressed in antigen-presenting cells (APCs) and endothelial cells [36–38]. GITR is also activated by a newly described GITR ligand called SECTM1A . GITR costimulation activates T cell receptor- (TCR-) triggered CD4+ and CD8+ T cells, promoting proliferation (Figure 1) [24, 25, 40–42]. GITR activation can be obtained by agonist anti-GITR Abs, soluble GITRL, or transfection of GITRL [24, 25, 40, 41, 43]. The costimulatory effect of GITR activation in T cells increases T cell expansion and cytokine production [24, 25, 40, 42], exacerbates autoimmune/inflammatory diseases [44–46], favours tumour rejection, performs viral and parasite clearance, and potentiates immune/inflammatory responses [21, 22, 47–52]. A peculiar effect of GITR costimulation is increased IL-10 production, such that neutralizing anti-IL-10 antibodies increase CD4+ proliferation following GITR activation .
GITR may have a role in CD8+ T cells different from CD4+ T cells, as initially suggested by the observation that GITR triggering exerts a different effect in alloreactive CD4+ and CD8+ T cells in GvHD . One difference refers to the reciprocal interaction between GITR and CD28. During activation of CD4+CD25− cells, GITR upregulation depends on CD28 stimulation [41, 102]. On the contrary, CD8+ cells cannot be stimulated by CD28 in the absence of GITR if suboptimal doses of anti-CD3 Ab are used; however, GITR can coactivate downstream functions in the absence of CD28 [103, 104]. Thus, in CD8+ cells, GITR is necessary for CD28 costimulatory activity. Expression of 4-1BB also depends on GITR expression in CD8+ memory T cells  and GITR promotes survival of memory bone marrow CD8+ T cells . A specific role for GITR activation in the stimulation of CD8+ T cells is well-defined during chronic viral infection [34, 104, 107].
Interestingly, the number of CD8+ T cells is not affected when GITR is activated by a supraphysiological level of ligand in GITRL-transgenic mice [108, 109]; thus, physiological GITR activation is sufficient to fully stimulate CD8+ T cells. Conversely, the number and phenotype of CD4+ T cells are dramatically altered in two different transgenic mice that constitutively express GITRL in B cells  or in most APCs (i.e., majority of B cells, DCs, NK cells, and a fraction of macrophages) . The most impressive phenotypic change is CD4+ Treg expansion, as discussed in Section 5. However, CD4+ effector T cell expansion and maturation are favoured as well. The number of CD4+ T cells with an effector memory-like (CD44+CD62L−) and central memory-like (CD44+CD62L+) phenotype increased by twofold in GITRL-B-cell transgenic mice compared to that of wild-type control mice. Robust activation of GITR in GITRL-APC-transgenic mice resulted in 10-fold activation of CD44+CD62L− compared to that in wild-type control (at 10 weeks). Conversely, CD4+ naïve T cells (CD44−CD62L+) decreased by two- to threefold in transgenic mice, suggesting that GITR-triggered naïve T cells tend to be more reactive against antigens and differentiate towards the memory phenotype.
4. GITR Is a Crucial Player in tTreg Differentiation
Thymic development of Tregs is a two-step process [110, 111]. First, TCR and CD28 signalling induce IL-2 and chromatin remodelling at the FoxP3 locus (Treg progenitors). The second signalling event is cytokine-dependent and leads to FoxP3 expression. Mahmud et al. recently demonstrated that this step involves three TNFRSF members (GITR, OX40, and TNFR2) (Figure 1) . In fact, Treg progenitors express high levels of GITR, OX40, and TNFR2. Combined neutralization of their ligands (GITRL, OX40L, and TNF) abrogates the development of Tregs and costimulation of GITR and OX40 results in a tenfold increased sensitivity to low doses of IL-2. Treg progenitors with TCRs of the highest affinity and the highest expression of GITR, OX40, and TNFR2 compete more effectively for the respective ligands and are more likely to differentiate into mature Tregs. As a consequence, GITR expression is high in mature tTregs.
5. GITR Is a Crucial Player in Treg Expansion
The role of GITR activation in the expansion of Tregs is supported by studies in mice (Figure 1). The number of Tregs is lower in GITR-knockout mice [23, 41, 113]. GITR activation by an anti-GITR Ab costimulates Tregs and promotes loss of anergy in the absence of IL-2 [24, 25]. In the presence of IL-2 and anti-CD3 Ab, GITR activation by a GITRL-Fc fusion protein promotes proliferation of both effector and Treg cells; however, FoxP3+ cells exhibit a stronger response to ligation of GITR compared to that in FoxP3− cells . Indeed, treatment of mice with GITRL-Fc favors a preferential proliferation of FoxP3+ cells .
Other in vivo models confirm the in vitro findings. As expected, the number of FoxP3+ Tregs is higher in transgenic mice that overexpress GITRL in B cells or APCs, compared to that of wild-type mice [108, 109]. Tregs from GITRL-transgenic mice are phenotypically activated and retain suppressive abilities. The costimulatory effect towards Tregs is stronger than that towards effector T cells.
Tr1 cells, which are CD4+ pTregs derived from activated effector T cells in peculiar conditions, do not express FoxP3 but do produce IL-10 . In 12-week-old wild-type mice, 1% of splenic CD4+ T cells are FoxP3−IL-10+. In transgenic mice with APCs that constitutively express GITRL, the FoxP3−IL-10+ subset is 6%-7% of all CD4+ T cells. These findings suggest that chronic GITR/GITRL signalling favours not only expansion of FoxP3+ tTregs but also the expansion and/or generation of FoxP3− Tr1-like cells . Suppression by these Tr1-like cells is mediated by IL-10, as indicated by experiments using a neutralizing anti-IL-10 monoclonal antibody. The Tr1-like cells in transgenic mice are GITR+, suggesting that GITR is a marker of murine Tr1 cells and participates in their expansion. Interestingly, GITR activation of CD4+ effector T cells stimulates IL-10 production . Generation of in vitro induced Tregs (iTregs) is induced in cocultures of CD4+ effector T cells and malignant plasma cells, and iTregs show high levels of GITR expression . However, the effects of blocking GITR have not been tested.
Studies on GITRL−/− mice confirm the role of GITR activation in Treg expansion and/or generation . In fact, the expansion or generation of Tregs is impaired in GITRL−/− mice after injection of the dendritic cell-inducing factor Flt3 ligand. Furthermore, the expansion or generation of OVA-specific Tregs is impaired after gene transfer of ovalbumin (using the adenoassociated virus AAV8-OVA). Considering that GITR activation is elicited even by the newly discovered GITR ligand, SECTM1A , it is possible that these results would be even more relevant if GITR activation was abolished by loss of both GITR ligands (GITRL and SECTM1A).
In multiple sclerosis, treatment with IFN-β increases the number of CD4+CD25+FoxP3+ Tregs following increased expression of GITRL in CD14+ monocytes ; this may suggest that GITR activation favours Treg expansion in humans too. However, so far no study demonstrates the role of GITR in human Treg expansion.
6. Pharmacological Activation of GITR Transiently Inhibits Treg Activity
GITR stimulation enhances T cell proliferation/activation not only through costimulation of effector T cells but also by the inhibition of Tregs (Figure 1), as originally demonstrated by the Sakaguchi and Shevach groups using anti-GITR Abs [26, 35]. Other studies have demonstrated the same effect when GITR was activated by GITRL overexpressed in APCs [24, 40, 117, 118]. GITR stimulation abolishes the activity of other suppressor cells, such as retinal pigment epithelial cells  or CD4+CD25− T cells in aged mice . The same effect is elicited when human Tregs are exposed to anti-GITR Abs [29, 121].
Some studies indicate that increased proliferation of effector cells in response to GITR activation of effector T cells renders them more resistant to Treg suppression . In this context, two studies suggest that GITR stimulation activates an unknown pathway in effector T cells distinct from that which is activated by CD28, blocking immunosuppression [41, 119]. However, other studies demonstrate that GITR stimulation activates transduction pathways in Treg cells that are responsible for Treg suppression. In particular, GITR signalling downregulates granzyme B , degrades FoxP3 protein [124, 125], phosphorylates c-Jun N-terminal kinase (JNK), and activates NF-κB . Furthermore, in vitro and in vivo experiments demonstrate that stimulation of GITR on Tregs underlies the increased activation of effector T cells during inflammatory responses or tumour rejection [24, 26, 45, 124–127].
The full activity of Tregs in GITRL-transgenic mice [108, 109] suggests that inhibition of Treg suppression is transient and may be due to overstimulation of GITR in a nonphysiological condition (pharmacological effects). Indeed, GITR-dependent decrease in FoxP3 protein expression is not due to changes in the levels of FoxP3 mRNA, confirming that this is a transient effect.
Some studies demonstrate that the effect of anti-GITR Ab on the suppressive activity of Tregs depends on Treg depletion [128, 129], which in turn depends on the binding of anti-GITR Ab with activating Fc γ receptors (FcγR) . The increased susceptibility of Tregs compared to that of activated CD4+ and CD8+ cells may be due to higher levels of GITR expression on Tregs. Studies suggest that the effect of anti-GITR Ab is tumour-specific and may depend on myeloid cells and natural killer cells present in tumours but not in draining lymph nodes .
In conclusion, GITR activation has four distinct effects on Treg/effector cell interplay: (1) transient inhibition of Treg regulatory activity, (2) decreased sensitivity of effector T cells to Treg suppression, (3) killing of Tregs (at least within solid tumours), and (4) increased proliferation and expansion of the Treg compartment.
The most relevant of these four remains a matter of debate. The relative importance of each of these mechanisms depends on the context and the disease [21, 130]. Nonetheless, the balanced effects on Tregs may reduce the chances of adverse effects. Indeed, no overt autoimmunity was observed in adult animals treated with anti-GITR Abs [131, 132].
7. GITR Is a Marker of Murine Tregs
The evidence supporting a crucial role for GITR in the maturation of tTregs and expansion of tTregs and pTregs includes the fact that tTregs and expanded pTregs constitutively express GITR. Indeed, GITR is expressed in many (if not all) subsets of Tregs.
Constitutive expression of GITR in murine tTregs was first described by McHugh et al.  and Shimizu et al.  and was confirmed by Zelenika et al. . Cells with suppressive activity express high levels of GITR in FoxP3 transgenic mice . Indeed, gene expression in the Gitr locus is regulated by NF-κB and FoxP3 through an enhancer . In FoxP3 transgenic mice, GITR expression is observed surprisingly in CD25− cells, demonstrating that GITR represents a Treg marker independent of CD25 . Uraushihara et al. demonstrated that cotransfer of the CD4+GITR+ population prevents the development of CD4+ T cell-transferred colitis . Interestingly, CD4+GITR+ T cells prevent wasting disease and colitis independently of CD25 expression. In fact, both CD4+CD25+GITR+ and CD4+CD25−GITR+ T cells express CTLA-4, show anergy, suppress T cell proliferation, and produce IL-10 and TGFβ. In GITRL-transgenic mice, the expanded CD4+FoxP3−IL-10+ Tr1-like cells are GITR+ , suggesting that GITR expression in Tregs is FoxP3-independent.
In summary, several subsets of murine Tregs appear to express GITR. GITR expression was observed in CD4+CD25+ Tregs [20, 36, 38] and some CD4+ Treg subsets (e.g., CD25+CD4+CD103+ cells  and CD25+CD4+CD83+ cells ), which are collectively called tTregs. GITR is expressed by Tr1 Tregs too, which are defined by their capacity to produce high levels of IL-10 and lack of FoxP3 [109, 138]. Moreover, GITR is expressed in CD8+CD25+ Tregs [139–141], γδ CD25+ Tregs , and TCR+CD4−CD8−CD25+/−PD-FoxP3− cells (double-negative Tregs) .
Functional studies confirm that GITR is a highly specific marker for identifying and isolating Tregs. Ono et al., for example, demonstrated that the transfer of GITR+-depleted T cell populations caused death in 90% of nude mice due to autoimmune diseases . In contrast, injection of CD25+-depleted cells did not cause death. The same study demonstrates that the transfer of GITR+-depleted T cells from NOD mice to NOD-SCID mice promotes the accelerated development of diabetes as compared to those transferred with CD25+-depleted T cells (one month after cell transfer versus more than two months). The former mice died or had to be sacrificed because of debilitation before or at seven weeks after transfer, whereas only 20% of the latter (transferred with CD25-depleted T cells) died at 12 weeks after transfer.
These studies demonstrate that GITR is expressed in many (if not all) Treg subsets and at much lower levels in effector T cells. Because it also allows CD25− Tregs to be isolated, GITR appears to be a more useful marker than CD25. The big difference among the diseases observed in GITR+ cell-depleted and CD25+ cell-depleted mice suggests that either known CD25− Tregs (e.g., Tr1 and double-negative cells) are crucial to immune homeostasis or the number of CD25−GITR+ cells is big and other CD25−GITR+ cell subsets with regulatory activities exist in mice.
8. GITR Is a Marker of Human Tregs
Soon after the discovery of GITR as a marker of murine Tregs [26, 35], studies reported that GITR is coexpressed with CD25 and FoxP3 in human CD4+ tTregs [121, 144, 145] and some of their subsets (e.g., FoxP3+Tim+ ). CD4+GITR+ cells isolated from healthy donors express FoxP3 and show regulatory activity . These cells are likely to be pTregs, because they express CD45RO, IL-10, and TGF-β. Thus, CD4+GITR+ cells are similar to human Tr1 clones  and FoxP3−IL-10-producing Tr1-like cells that have expanded in GITRL-transgenic mice . Finally, GITR is found also in CD8+CD25+ and CD8+FoxP3+ human Tregs [70, 90, 147, 148].
Therefore, GITR is used as a human Treg marker in several studies. A PubMed search revealed 49 studies in the past three years that investigated the roles of Tregs in autoimmune/allergic diseases, tumours, and infections and reported GITR as a Treg marker (Table 2). More interestingly, GITR is found in memory/activated Tregs, as reported below.
9. GITR Is Frequently Found in Memory and/or Active Tregs
Most Tregs, including tTregs, are memory Tregs and GITR expression occurs more frequently in these cells. Uraushihara et al. assessed correlations among GITR, CD25, and CD45RB expression in freshly isolated murine splenic CD4+ T cells . Both CD25+ and CD25− Tregs expressed GITR; GITR was exclusively expressed on CD4+ cells, suggesting that CD4+GITR+ cells are memory cells. We recently described a human CD4+GITR+ Treg subset [10, 29, 58]. The majority of these cells are CD45RA− and CD45RO+, demonstrating that CD4+GITR+ cells are memory Tregs and most likely pTregs. The association of GITR with a memory phenotype is not surprising considering that GITR participates in Treg expansion. Indeed, an expanded population of CD4+GITR+ Tregs is found in some patients with Sjogren’s syndrome and SLE [10, 58]; expansion is observed in patients with inactive disease, suggesting that CD4+GITR+ Tregs participate in disease control.
Several studies suggest that GITR is a marker of Tregs that are actively suppressing effector cells. This is supported by the association of GITR expression with other Treg activation markers (e.g., CTLA-4), the expression of GITR in antigen-specific Tregs and/or T cells that produce cytokines with suppressive activity (e.g., IL-10 and TGF-β), the presence of GITR+ cells in tissues with active Tregs (e.g., solid tumours), and functional studies on Tregs. Our purpose is to illustrate the relevance of GITR as a marker of human Tregs; thus, we focus mainly on studies performed in humans.
Wang et al. characterized CD4+ and CD8+ T cell clones derived from the peripheral blood of patients infected with human herpes virus 6 (HHV-6) . They found that HHV-6-specific T cell clones with suppressive activities expressed not only CD25 and FoxP3, as expected, but also GITR; furthermore, nonsuppressive HHV-6-specific CD4+ or CD8+ T cells were negative for all three markers. Similar data were found in CD8+ Tregs from mice, in which FoxP3+ Tregs expressed CD25, GITR, and IL-10 . Consistent with the Wang study, our group recently observed elevated expression of FoxP3, TGF-β, and CTLA-4 in CD4+CD25+GITR+ PBMCs compared to that of CD4+CD25+GITR− PBMCs (unpublished data). In human decidua, expression of GITR is higher in CD25+CTLA-4+ cells compared to that in CD25+CTLA-4−  and suppressive activity correlates with the intensity of GITR and CTLA-4 expression in human CD4+CD25+ clones . Similar findings have been described in human Tr1 clones (GITR+), in which FoxP3 downregulation results in the loss of GITR surface expression and Treg suppressive activity but does not alter CD25 expression .
Treg dysregulation (decreased number or function) has been demonstrated in several autoimmune diseases. GITR expression appears to be lower in patients with type 1 diabetes (T1D) compared to that of controls, according to two studies. In the first study, the percentages of CD4+ cells did not differ between children with T1D and controls, but mRNA levels of several Treg markers, including GITR, were lower in Tregs from children with T1D . In the second study, the number of cells, especially CD4+GITR+ cells, was lower in T1D patients than in controls . Even within the CD4+ population, the number of GITR+ cells was lower in T1D patients, indicating that GITR+ Tregs (independent of FoxP3 or CD25 expression) are less abundant in T1D patients. These data suggest that GITR+ Tregs are crucial for the control of T1D.
In several solid tumour types, a subset of tumour-infiltrating lymphocytes is formed by antigen-specific Tregs that inhibit other immune cells and prevent tumour rejection. Data from animal tumour models suggest that GITR+ Tregs tend to infiltrate tumours. For example, Sacchetti et al. demonstrated that GITR expression in Tregs (CD4+FoxP3+ cells) residing in a B16 melanoma is approximately 10-fold higher than that of Tregs in the spleens of the same animals . In Colon26 tumours, Tregs found in tumour infiltrating lymphocytes (TILs) show approximately fourfold stronger GITR signals than those of the equivalent lymph node population . The role of GITR as a TIL Treg marker has also been confirmed in humans. In head and neck squamous cell carcinoma (HNSCC), cells were enriched and represented 3% of CD3+CD4+ TILs compared to circulating CD3+CD4+ T cells of the same patients, which comprised 1% of CD3+CD4+ cells; circulating CD3+CD4+ T cells of normal controls comprised 0.4% of CD3+CD4+ cells . Analysis of markers expressed on cells among PBMCs and TILs from the same patients shows that some markers (e.g., FoxP3 and CTLA-4) are found in both Tregs at the same levels but others are not. In particular, a lower percentage of TIL Tregs express CCR7 (77% in TIL Tregs versus 49% in PBMC Tregs) and CD62L (28% versus 47%). A higher percentage of TIL Tregs express CD132 (20% versus 40%). TIL Tregs are fully active, as suggested by expression of FasL (57% versus 0%), intracellular IL-10 (71% versus 0%), and intracellular TGFβ (97% versus 0%) and as demonstrated by a much higher suppressive activity compared to that of PBMC Tregs from the same patients. Importantly, IL-10/TGFβ-producing TIL Tregs are best detected by GITR, which is expressed in 83% of TIL Tregs and about 5% of PBMC Tregs of patients. Thus, GITR is the main marker of active Tregs, at least in HNSCC tumour. Similar findings were found in glioblastoma tumours, where all tTregs (CD4+FoxP3+Helios+) were GITR+ .
A study on patients with invasive breast cancer confirms the role of GITR, even if data were less impressive . FoxP3 and CTLA-4 are expressed in about 20% of cells among PBMCs and about 30% of the CD25+ cells infiltrating the tumours, resulting in an increase of about 1.5-fold in TILs. In contrast, GITR is expressed in about 5% of the cells of PBMCs and about 30% of the CD25+ cells infiltrating the tumours, resulting an increase of about sixfold in TILs. In PBMCs of the patients, the mean frequency of and cells was more than twofold higher than in healthy controls and expression of GITR, CTLA-4, and CCR4 was significantly higher in both subsets . Comparing T-lymphocyte phenotypes among the lymph nodes of three breast cancer patients, Krausz et al. demonstrated that CD4+CD127+GITR+ cells are more abundant in tumour-positive lymph nodes than in tumour-negative lymph nodes of the same patient . These findings suggest that tumour cells stimulate expansion of this Treg subset or that GITR+ cells tend to accumulate within the tumour.
Studies on other tumours confirm the overall picture. Padovani et al. studied cervical tissue of healthy women in comparison with tissue from women infected with papillomavirus (HPV) or women with carcinoma . Most (85%) of carcinoma samples showed high GITR expression, whereas only 41% of normal and low-grade cervical intraepithelial neoplasia samples showed high GITR expression. Moreover, most (78%) of the samples highly positive for HPV showed high GITR expression, whereas only 43% of the samples negative for HPV showed high GITR expression. No differences in CD25 staining were observed, suggesting that the presence of GITR+ Tregs correlates with malignancy and HPV infection of the cervical stroma and possibly with Treg activity. von Rahden et al. demonstrated that GITR expression in TILs is associated with oesophageal adenocarcinomas in the absence of Barrett’s mucosa . In patients with hepatocellular carcinoma and liver metastases from colon cancer, the number of CD25+FoxP3+ Tregs is significantly higher in tumours than that in tumour-free liver tissue. The MFI of GITR and ICOS are significantly higher in Tregs from tumours than in those from tumour-free liver tissue .
In conclusion, GITR is associated with activation and increased suppressive activity of Tregs in both mice and humans.
GITR is a crucial player in differentiation of tTreg and expansion of Tregs, including both tTregs and pTregs. Indeed, some data suggest that GITR is associated with markers of memory cells. Approximately 70–80% of Tregs are memory T cells, and expanded Tregs are active, at least in some diseases (e.g., several solid tumour types); thus, it is not surprising that several studies support GITR as a marker of active Tregs. Moreover, some Treg subsets, including Tr1 cells, express low levels of classical Treg markers (e.g., FoxP3 and CD25) or do not express them at all despite exhibiting expression of GITR. For this reason, we believe that GITR+ Tregs must be evaluated when considering increased or decreased numbers of Tregs in human diseases. Finally, the use of GITR as a marker to isolate Tregs should be considered as we have recently proposed .
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Simona Ronchetti and Erika Ricci are joint first authors.
This work was supported by Associazione Italiana per la Ricerca sul Cancro IG-14291 to Carlo Riccardi and the “Fondazione Cassa di Risparmio di Perugia.”
S. Sakaguchi, N. Sakaguchi, M. Asano, M. Itoh, and M. Toda, “Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor α-chains (CD25): breakdown of a single mechanism of self-tolerance causes various autoimmune diseases,” Journal of Immunology, vol. 155, no. 3, pp. 1151–1164, 1995.View at: Google Scholar
G. Nocentini, A. Alunno, M. Petrillo et al., “Expansion of regulatory GITR+ CD25Low/-CD4+ T cells in systemic lupus erythematosus patients,” Arthritis Research & Therapy, vol. 16, article 444, 2014.View at: Google Scholar
M. G. Roncarolo, S. Gregori, R. Bacchetta, and M. Battaglia, “Tr1 cells and the counter-regulation of immunity: natural mechanisms and therapeutic applications,” in Interleukin-10 in Health and Disease, vol. 380 of Current Topics in Microbiology and Immunology, pp. 39–68, Springer, Berlin, Germany, 2014.View at: Publisher Site | Google Scholar
M. Ono, J. Shimizu, Y. Miyachi, and S. Sakaguchi, “Control of autoimmune myocarditis and multiorgan inflammation by glucocorticoid-induced TNF receptor family-related proteinhigh, Foxp3-expressing CD25+ and CD25− regulatory T cells,” Journal of Immunology, vol. 176, no. 8, pp. 4748–4756, 2006.View at: Publisher Site | Google Scholar
Z. Li, S. P. Mahesh, B. J. Kim, R. R. Buggage, and R. B. Nussenblatt, “Expression of Glucocorticoid Induced TNF Receptor Family Related Protein (GITR) on peripheral T cells from normal human donors and patients with non-infectious uveitis,” Journal of Autoimmunity, vol. 21, no. 1, pp. 83–92, 2003.View at: Publisher Site | Google Scholar
G. Nocentini, L. Giunchi, S. Ronchetti et al., “A new member of the tumor necrosis factor/nerve growth factor receptor family inhibits T cell receptor-induced apoptosis,” Proceedings of the National Academy of Sciences of the United States of America, vol. 94, no. 12, pp. 6216–6221, 1997.View at: Publisher Site | Google Scholar
P. M. Lacal, M. G. Petrillo, F. Ruffini et al., “Glucocorticoid-induced tumor necrosis factor receptor family-related ligand triggering upregulates vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 and promotes leukocyte adhesion,” Journal of Pharmacology and Experimental Therapeutics, vol. 347, no. 1, pp. 164–172, 2013.View at: Publisher Site | Google Scholar
G. L. Stephens, R. S. McHugh, M. J. Whitters et al., “Engagement of glucocorticoid-induced TNFR family-related receptor on effector T cells by its ligand mediates resistance to suppression by CD4+CD25+ T cells,” The Journal of Immunology, vol. 173, no. 8, pp. 5008–5020, 2004.View at: Publisher Site | Google Scholar
A. P. Kohm, J. S. Williams, and S. D. Miller, “Cutting edge: ligation of the glucocorticoid-induced TNF receptor enhances autoreactive CD4+ T cell activation and experimental autoimmune encephalomyelitis,” Journal of Immunology, vol. 172, no. 8, pp. 4686–4690, 2004.View at: Publisher Site | Google Scholar
N. van der Werf, S. A. Redpath, A. T. Phythian-Adams et al., “Th2 responses to helminth parasites can be therapeutically enhanced by, but are not dependent upon, GITR-GITR ligand costimulation in vivo,” Journal of Immunology, vol. 187, no. 3, pp. 1411–1420, 2011.View at: Publisher Site | Google Scholar
G. Nocentini, S. Cuzzocrea, T. Genovese et al., “Glucocorticoid-induced tumor necrosis factor receptor-related (GITR)-Fc fusion protein inhibits GITR triggering and protects from the inflammatory response after spinal cord injury,” Molecular Pharmacology, vol. 73, no. 6, pp. 1610–1621, 2008.View at: Publisher Site | Google Scholar
B. Moradi, P. Schnatzer, S. Hagmann et al., “CD4+CD25+/highCD127low/- regulatory T cells are enriched in rheumatoid arthritis and osteoarthritis joints-analysis of frequency and phenotype in synovial membrane, synovial fluid and peripheral blood,” Arthritis Research and Therapy, vol. 16, no. 2, article R97, 2014.View at: Publisher Site | Google Scholar
D. Mesquita Jr., W. M. Cruvinel, J. A. Araujo, K. C. Salmazi, E. G. Kallas, and L. E. Andrade, “Imbalanced expression of functional surface molecules in regulatory and effector T cells in systemic lupus erythematosus,” Brazilian Journal of Medical and Biological Research, vol. 47, pp. 662–669, 2014.View at: Google Scholar
A. Alunno, G. Nocentini, O. Bistoni et al., “Expansion of CD4+CD25-GITR+ regulatory t-cell subset in the peripheral blood of patients with primary Sjögren's syndrome: correlation with disease activity,” Reumatismo, vol. 64, no. 5, pp. 293–298, 2012.View at: Google Scholar
M. Chen, G. Chen, S. Deng, X. Liu, G. J. Hutton, and J. Hong, “IFN-beta induces the proliferation of CD4+CD25+Foxp3+ regulatory T cells through upregulation of GITRL on dendritic cells in the treatment of multiple sclerosis,” Journal of Neuroimmunology, vol. 242, no. 1-2, pp. 39–46, 2012.View at: Publisher Site | Google Scholar
M. Pietruczuk, M.-O. Eusebio, L. Kraszula, M. Kupczyk, and P. Kuna, “Phenotypic characterization of ex vivo CD4+CD25highCD127low immune regulatory T cells in allergic asthma: pathogenesis relevance of their FoxP3, GITR, CTLA-4 and FAS expressions,” Journal of Biological Regulators and Homeostatic Agents, vol. 26, no. 4, pp. 627–639, 2012.View at: Google Scholar
F. F. Ni, C. R. Li, Q. Li, Y. Xia, G. B. Wang, and J. Yang, “Regulatory T cell microRNA expression changes in children with acute Kawasaki disease,” Clinical & Experimental Immunology, vol. 178, pp. 384–393, 2014.View at: Google Scholar
N. Arandi, A. Mirshafiey, M. Jeddi-Tehrani et al., “Alteration in frequency and function of CD4+CD25+FOXP3+ regulatory T cells in patients with immune thrombocytopenic purpura,” Iranian Journal of Allergy, Asthma, and Immunology, vol. 13, no. 2, pp. 85–92, 2014.View at: Google Scholar
K. Giannopoulos, W. Kaminska, I. Hus, and A. Dmoszynska, “The frequency of T regulatory cells modulates the survival of multiple myeloma patients: detailed characterisation of immune status in multiple myeloma,” British Journal of Cancer, vol. 106, no. 3, pp. 546–552, 2012.View at: Publisher Site | Google Scholar
L. T. Krausz, Z. Z. Major, D. F. Muresanu, E. Chelaru, G. Nocentini, and C. Riccardi, “Characterization of CD4+ and CD8+ tregs in a Hodgkin's lymphoma patient presenting with myasthenia-like symptoms,” Ideggyogyaszati Szemle, vol. 66, no. 9-10, pp. 343–348, 2013.View at: Google Scholar
L. T. Krausz, E. Fischer-Fodor, Z. Z. Major, and B. Fetica, “GITR-expressing regulatory T-cell subsets are increased in tumor-positive lymph nodes from advanced breast cancer patients as compared to tumor-negative lymph nodes,” International Journal of Immunopathology and Pharmacology, vol. 25, no. 1, pp. 59–66, 2012.View at: Google Scholar
L. Benevides, C. R. B. Cardoso, D. G. Tiezzi, H. R. C. Marana, J. M. Andrade, and J. S. Silva, “Enrichment of regulatory T cells in invasive breast tumor correlates with the upregulation of IL-17A expression and invasiveness of the tumor,” European Journal of Immunology, vol. 43, no. 6, pp. 1518–1528, 2013.View at: Publisher Site | Google Scholar
C. T. J. Padovani, C. M. Bonin, I. A. Tozetti, A. M. T. Ferreira, C. E. Dos Santos Fernandes, and I. P. Da Costa, “Glucocorticoid-induced tumor necrosis factor receptor expression in patients with cervical human papillomavirus infection,” Revista da Sociedade Brasileira de Medicina Tropical, vol. 46, no. 3, pp. 288–292, 2013.View at: Publisher Site | Google Scholar
G. Méndez-Lagares, M. M. Pozo-Balado, M. Genebat, A. García-Pergañeda, M. Leal, and Y. M. Pacheco, “Severe immune dysregulation affects CD4+CD25 hiFoxP3+ regulatory T cells in HIV-infected patients with low-level CD4 T-Cell repopulation despite suppressive highly active antiretroviral therapy,” Journal of Infectious Diseases, vol. 205, no. 10, pp. 1501–1509, 2012.View at: Publisher Site | Google Scholar
S. B. Rathod, R. Das, S. Thanapati, V. A. Arankalle, and A. S. Tripathy, “Suppressive activity and altered conventional phenotype markers/mediators of regulatory T cells in patients with self-limiting hepatitis E,” Journal of Viral Hepatitis, vol. 21, no. 2, pp. 141–151, 2014.View at: Publisher Site | Google Scholar
F. Koukouikila-Koussounda, F. Ntoumi, M. Ndounga, H. V. Tong, A.-A. Abena, and T. P. Velavan, “Genetic evidence of regulatory gene variants of the STAT6, IL10R and FOXP3 locus as a susceptibility factor in uncomplicated malaria and parasitaemia in Congolese children,” Malaria Journal, vol. 12, no. 1, article 9, 2013.View at: Publisher Site | Google Scholar
A. P. Kohm, J. R. Podojil, J. S. Williams, J. S. McMahon, and S. D. Miller, “CD28 regulates glucocorticoid-induced TNF receptor family-related gene expression on CD4+ T cells via IL-2-dependent mechanisms,” Cellular Immunology, vol. 235, no. 1, pp. 56–64, 2005.View at: Publisher Site | Google Scholar
L. M. Snell, G. H. Y. Lin, and T. H. Watts, “IL-15-dependent upregulation of GITR on CD8 memory phenotype T cells in the bone marrow relative to spleen and lymph node suggests the bone marrow as a site of superior bioavailability of IL-15,” The Journal of Immunology, vol. 188, no. 12, pp. 5915–5923, 2012.View at: Publisher Site | Google Scholar
D. L. Clouthier, A. C. Zhou, and T. H. Watts, “Anti-GITR agonist therapy intrinsically enhances CD8 T cell responses to chronic lymphocytic choriomeningitis virus (LCMV), thereby circumventing LCMV-induced downregulation of costimulatory GITR ligand on APC,” Journal of Immunology, vol. 193, no. 10, pp. 5033–5043, 2014.View at: Publisher Site | Google Scholar
S. P. Mahesh, Z. Li, B. Liu, R. N. Fariss, and R. B. Nussenblat, “Expression of GITR ligand abrogates immunosuppressive function of ocular tissue and differentially modulates inflammatory cytokines and chemokines,” European Journal of Immunology, vol. 36, no. 8, pp. 2128–2138, 2006.View at: Publisher Site | Google Scholar
D. C. Gondek, L. F. Lu, S. A. Quezada, S. Sakaguchi, and R. J. Noelle, “Cutting edge: contact-mediated suppression by CD4+CD25+ regulatory cells involves a granzyme B-dependent, perforin-independent mechanism,” Journal of Immunology, vol. 174, no. 4, pp. 1783–1786, 2005.View at: Publisher Site | Google Scholar
A. Joetham, H. Ohnishi, M. Okamoto et al., “Loss of T regulatory cell suppression following signaling through glucocorticoid-induced tumor necrosis receptor (GITR) is dependent on c-Jun N-terminal kinase activation,” The Journal of Biological Chemistry, vol. 287, no. 21, pp. 17100–17108, 2012.View at: Publisher Site | Google Scholar
B. Valzasina, C. Guiducci, H. Dislich, N. Killeen, A. D. Weinberg, and M. P. Colombo, “Triggering of OX40 (CD134) on CD4+CD25+ T cells blocks their inhibitory activity: a novel regulatory role for OX40 and its comparison with GITR,” Blood, vol. 105, no. 7, pp. 2845–2851, 2005.View at: Publisher Site | Google Scholar
T. Ramirez-Montagut, A. Chow, D. Hirschhorn-Cymerman et al., “Glucocorticoid-induced TNF receptor family related gene activation overcomes tolerance/ignorance to melanoma differentiation antigens and enhances antitumor immunity,” The Journal of Immunology, vol. 176, no. 11, pp. 6434–6442, 2006.View at: Publisher Site | Google Scholar
C. T. Mayer, S. Floess, A. M. Baru, K. Lahl, J. Huehn, and T. Sparwasser, “CD8+Foxp3+ T cells share developmental and phenotypic features with classical CD4+Foxp3+ regulatory T cells but lack potent suppressive activity,” European Journal of Immunology, vol. 41, no. 3, pp. 716–725, 2011.View at: Publisher Site | Google Scholar
I. M. De Kleer, L. R. Wedderburn, L. S. Taams et al., “CD4+CD25bright regulatory T cells actively regulate inflammation in the joints of patients with the remitting form of juvenile idiopathic arthritis,” Journal of Immunology, vol. 172, no. 10, pp. 6435–6443, 2004.View at: Publisher Site | Google Scholar
L. Scotto, A. J. Naiyer, S. Galluzzo et al., “Overlap between molecular markers expressed by naturally occurring CD4+CD25+ regulatory T cells and antigen specific CD4+CD25+ and CD8+CD28− T suppressor cells,” Human Immunology, vol. 65, no. 11, pp. 1297–1306, 2004.View at: Publisher Site | Google Scholar
M. K. Levings, R. Sangregorio, C. Sartirana et al., “Human CD25+CD4+ T suppressor cell clones produce transforming growth factor beta, but not interleukin 10, and are distinct from type 1 T regulatory cells,” Journal of Experimental Medicine, vol. 196, no. 10, pp. 1335–1346, 2002.View at: Publisher Site | Google Scholar
W. Łuczyński, N. Wawrusiewicz-Kurylonek, A. Stasiak-Barmuta et al., “Diminished expression of ICOS, GITR and CTLA-4 at the mRNA level in T regulatory cells of children with newly diagnosed type 1 diabetes,” Acta Biochimica Polonica, vol. 56, no. 2, pp. 361–370, 2009.View at: Google Scholar
L. Strauss, C. Bergmann, M. Szczepanski, W. Gooding, J. T. Johnson, and T. L. Whiteside, “A unique subset of CD4+CD25highFoxp3+T cells secreting interleukin-10 and transforming growth factor-beta1 mediates suppression in the tumor microenvironment,” Clinical Cancer Research, vol. 13, no. 15, pp. 4345–4354, 2007.View at: Publisher Site | Google Scholar
B. H. A. von Rahden, S. Kircher, M. Kafka et al., “Glucocorticoid-induced TNFR family-related receptor (GITR)-expression in tumor infiltrating leucocytes (TILs) is associated with the pathogenesis of esophageal adenocarcinomas with and without Barrett's mucosa,” Cancer Biomarkers, vol. 7, no. 6, pp. 285–294, 2010.View at: Publisher Site | Google Scholar
A. Alunno, G. Nocentini, O. Bistoni et al., “Expansion of CD4+CD25-GITR+ regulatory T-cell subset in the peripheral blood of patients with primary Sjögren's syndrome: correlation with disease activity,” Reumatismo, vol. 64, no. 5, pp. 293–298, 2012.View at: Google Scholar