Natural CD25+CD4+ regulatory T cells (Treg) are essential for self-tolerance and for the control of T cell-mediated immune pathologies. However, the identification of Tregs in an ongoing immune response or in inflamed tissues remains elusive. Our experiments indicate that TIRC7, T cell immune response cDNA 7, a novel membrane molecule involved in the regulation of T lymphocyte activation, identifies two Treg subsets (CD25lowTIRC7+ and CD25highTIRC7−) that are characterized by the expression of Foxp3 and a suppressive activity in vitro and in vivo. We also showed that the CD25lowTIRC7+ subset represents IL-10-secreting Tregs in steady state, which is accumulated intratumorally in a tumor-bearing mice model. Blockade of the effect of IL-10 reversed the suppression imposed by the CD25lowTIRC7+ subset. Interestingly, these IL-10-secreting cells derived from the CD25highTIRC7− subset, both in vitro and in vivo, in response to tumoral Ags. Our present results strongly support the notion that, in the pool of natural Tregs, some cells can recognize foreign Ags and that this recognition is an essential step in their expansion and suppressive activity in vivo.
Natural CD4+CD25+ regulatory T cells (Treg)3 represent an important cellular component of the normal immune system (1, 2). The thymic origin of Tregs has been well established in mice from several observations. First, thymectomy of 3-day-old neonate mice induced long-term depletion of Tregs and resulted in a severe autoimmune syndrome (3). The Tregs are potent suppressors or regulators of effector and helper T cell responses in the setting of organ-specific autoimmunity, allograft rejection, and microbial immunity (1, 4, 5). More recently, two groups independently demonstrated that the forkhead transcription factor Foxp3 is a master regulator of Treg differentiation. Transfer of Foxp3 cDNA into naive CD4+ T cells was sufficient to induce their differentiation into phenotypically and functionally CD4+CD25+ Treg resembling cells (6, 7, 8). Furthermore, analysis of the relationship between Foxp3 and CD25 expression in the CD4+ T cell population derived from mice expressing a GFP-Foxp3 fusion protein revealed the presence of two Foxp3-expressing regulatory subsets, Foxp3+CD25low and Foxp3+CD25high as well as nonregulatory Foxp3−CD25low subset (8), and suggested that Foxp3 expression is highly restricted to Treg function (9). However, the fact that Foxp3 is expressed exclusively intracellularly excludes its use in the isolation of Treg subsets for functional studies. In addition to the ability of Tregs to block the initiation of immune responses in the secondary lymphoid tissues, recent studies indicate that Foxp3+ Tregs also migrate to nonlymphoid sites such as inflamed and tumor tissues (10, 11). However, the identification of Tregs in an ongoing immune response or in inflamed tissues is complicated by the fact that CD25 is also expressed on activated T cells. A description of new surface markers, in addition to CD25, that correlate with Foxp3 expression and capable of distinguishing the different subsets of Tregs would be highly informative.
The membrane protein T cell immune response cDNA 7 (TIRC7) was identified in human T lymphocytes and was shown to play an important role in T cell activation (12). Tcirg1 (T cell immune regulator 1), the mouse gene encoding the homologous TIRC7, also encodes another protein, the a3 subunit of the vacuolar proton pump expressed in osteoclasts and necessary for bone resorption (13, 14).
In this study, we show that in normal naive mice, an anti-TIRC7 Ab identifies a new subset within CD4+ T cells isolated from secondary lymphoid organs. Among the three subpopulations of CD4+ T cells defined by the expression of CD25 (negative, low, and high subsets), the marker TIRC7 exclusively labeled the CD25lowCD4+ subset. The CD25highTIRC7− and CD25lowTIRC7+ subsets represent two Treg subsets that are characterized by the expression of Foxp3 and suppressive activity in vitro. In contrast, the CD25lowTIRC7− subset expresses high levels of IL-2 and is devoided of suppressive function. Furthermore, the cotransfer of CD25lowTIRC7+ cells together with CD45RBhigh T cells into immunodeficient RAG1−/− mice prevented the development of colitis. The CD25lowTIRC7+ subset is capable of controlling intestinal inflammation and provides a suppressive activity in vivo. Interestingly, we showed that the CD25lowTIRC7+ subset derives from CD25highTIRC7− subset and represents IL-10-secreting Tregs in vitro and in vivo. These data highlight TIRC7 as a novel surface marker capable of distinguishing different stages of Treg activation.
Materials and Methods
6–8-wk-old Balb/cJ and C57BL/6J mice were purchased from the Janvier Laboratory. C57BL/6 RAG 1-deficient (RAG 1−/−) mice were obtained from the Charles River Laboratories. Animals were maintained in our central animal facility in accordance with the general guidelines edicted by the veterinary supervision.
Abs, cell purification, and flow cytometric analysis
Cells from mesenteric lymph nodes (MLN), Peyer’s patches (PP), and spleens of Balb/cJ and C57BL/6J mice were depleted by treatment for 30 min at 4°C with a mAb mixture containing anti-CD11b, anti-CD19, anti-CD49b, and anti-CD8 and separation with anti-rat Ig-coated magnetic beads (Dynal Biotech). FACS analysis was performed using a FACScan or FACS Aria (BD Biosciences) with Cellquest or Diva software respectively.
Purification of CD4+CD25+ T cell subsets
CD4+ T cells enriched from spleens, MLNs, and PP as described above, were stained with PE-Cy5 anti-CD4, PE anti-CD25, and anti-TIRC7 followed by FITC anti-rabbit IgG labeling. Subpopulations of CD4+ cells were sorted by three-color analysis on a FACS Vantage and FACS Aria (BD Biosciences). All populations were >98% pure on reanalysis.
Real time PCR analysis
Total RNA was extracted and reverse transcribed for real-time PCR analysis as described previously (15-Δcycle threshold method (16). At the end of the PCR assay, the specificity of amplification was controlled by generating a melting curve of the PCR product and analysis by gel electrophoresis. The following primers were used: IL2, 5′-CCCAGGATGCTCACCTTCAA-3′ and 5′-CCGCAGAGGTCCAAGTTCAT-3′; TGF-β, 5′-AACTGTGATCGCTCTGCACAAG-3′ and 5′-CGGCGGTAGGCGTGAAC-3′; Foxp3, 5′-GGTGGCTCTCCTTGTCATTTTC-3′ and 5′-CGTGGTATTCTCGCCGATGT-3′; IL-10, 5′-TTTGAATTCCCTGGGTGAGAAG-3′ and 5′-TGCTCCACTGCCTTGCTCTT-3′; and 36B4, 5′-TCCAGGCTTTGGGCATCA-3′ and 5′-CTTTATCAGCTGCACATCACTCAGA-3′.
In vitro Treg functional assay
The suppressive assay was performed as previously described (17). Sorted CD25−CD4+ T cells were labeled with CFSE (Molecular Probes) by incubating in PBS1X (Cambrex) with 2 μg/ml CFSE for 10 min at 37°C. CD3+-depleted spleen cells were used as APCs. To assess suppression activity, CFSE-labeled CD25−CD4+ T cells (5 × 104 per well in a U-bottom 96-well plate) in RPMI 1640 (Cambrex) supplemented with 5% of FCS (HyClone), and 5 × 104 irradiated APCs were cocultured in the presence of 1 μg/ml anti-CD3 (145–2C11) with different subsets of CD4+CD25+ T cell population. Four days later, FACS analysis was performed using a FACScan or FACSAria (BD Biosciences).
Transwell experiments were performed as described previously (18).
Adoptive T cell transfer experiments
To induce inflammatory bowel disease, C57BL/6 RAG 1−/− mice were injected i.p. with 4 × 105 sorted syngenic CD4+CD45RBhigh T cells alone, or in combination with sorted CD25highTIRC7− Tregs, or with sorted CD25lowTIRC7+ T cells, at a ratio 1:1 or 4:1. Mice were observed and weighed daily. Mice injected with CD4+CD45RBlow T cells alone developed clinical signs of colitis within 4–6-wk post-transfer. Mice showing clinical signs of severe disease were sacrificed immediately.
After sacrifice, colons were removed from mice, fixed in Bouin’s fixative, embedded in paraffin, sectioned at 5 μm, and stained with H&E.
Tumorigenicity assay in syngeneic immunocompetent mice
Syngenic C26 colon carcinoma cells were kindly provided by Mario Colombo (Milan, Italy) and maintained as previously described (19). Tumorigenic activity of C26 cells was assayed in mice injected s.c. in the left flank of BALB/c with 5.105 cells in 0.1 ml. Tumor growth and size were recorded twice each week.
To isolate CD25+TIRC7+ T cells from the tumor tissue, the tumor tissues were collected, washed in PBS, cut into pieces, and resuspended in DMEM supplemented with 1% of FCS and 1 mg/ml collagenase D for 20 min in a 37°C. The cell suspension was collected after 20 min, and the single-cell suspension was stained with polyclonal TIRC7 Ab followed by goat anti-rabbit IgG magnetic beads (Miltenyi Biotech).
Dendritic cells (DCs) derived from bone marrow (BMDCs)
BMDCs were generated from BM cells, as described (18). BMDCs were harvested on day 6, were purified using CD11c microbeads (Miltenyi Biotech), and incubated with chicken OVA (fraction V) (Sigma-Aldrich) or with lysate from C26 cells by sonication.
Purification of splenic DCs
DCs, enriched from spleens of tumor bearing mice (TBM) and tumor free mice (TFM) as described previously (18), were purified using CD11c microbeads (Miltenyi Biotech).
Sandwich ELISA were used to measure IL-10 and IL-2 as described (18
Identification of two subsets within peripheral CD4+CD25+ Treg using the TIRC7 marker in normal naive mice
CD25 provides the classical cell surface marker used to identify the natural suppressor CD4+ T cells. We therefore examined the expression of TIRC7 on peripheral CD4+CD25+ Tregs isolated from secondary lymphoid organs. We observed that the expression of TIRC7 identified two subsets of CD4+CD25+ Tregs isolated from MLNs, PPs, and spleens (Fig. 1⇓A). Interestingly, a marked increase in the TIRC7 expressing Tregs was observed in MLNs and PPs (30 and 67%, respectively) compared with the spleens of normal mice (5%) (Fig. 1⇓A). Further analysis of the TIRC7 and CD25 expression on CD4+ T cells isolated from MLNs revealed a new subset of Tregs expressing both low levels of CD25 and TIRC7 (CD25lowTIRC7+ subset) (Fig. 1⇓B). Similar results were found in PPs and spleens (data not shown). FACS analysis revealed that this subset was enriched for cells expressing low levels of CD62L and intermediate levels of CD45RB compared with the CD25+TIRC7− and CD25−TIRC7− subsets (Fig. 1⇓C). Previous studies have shown that CTLA-4 plays an essential role in the function of CD25+CD4+ Tregs (20, 21). FACS analysis on permeabilized cells indicated that CTLA-4 is expressed both in the CD25lowTIRC7+ cells (26% ± 2.1) and in the CD25+TIRC7− cells (45% ± 1.4), but not in the CD25−TIRC7− subset (Fig. 1⇓C). Moreover, we also examined the expression of the CD103 integrin, described as a marker subdividing the classical CD4+CD25+ Tregs into two subsets. Our results showed that, in contrast to the CD25−TIRC7− subset, both CD25lowTIRC7+ and CD25+TIRC7− subsets displayed a bimodal pattern of expression of CD103 (Fig. 1⇓C).
These data showed that the use of both TIRC7 and CD25 markers allow to define three subsets within CD4+CD25+ Treg cells: CD25lowTIRC7−, CD25highTIRC7−, and CD25low TIRC7+ (Fig. 2⇓A). Therefore, we have investigated the relationship between TIRC7 expression and Treg function. The four indicated subsets of CD4+ T cells (Fig. 2⇓A) isolated from MLNs and PPs were purified up to 98% purity by FACS sorting after enrichment of CD4+ T cells. These purified cell populations were analyzed by real-time PCR for the expression of Foxp3, IL-10, and TGF-β mRNA, three characteristic genes of the ≪ Treg cell signature ≫, as well as IL-2 mRNA (Fig. 2⇓A). As shown in Fig. 2⇓B, the CD25lowTIRC7− subset displayed a very low level of Foxp3 and a high level of IL-2 mRNA compared with the CD25highTIRC7− cells. In contrast, the CD25lowTIRC7+ subset has 80% of Foxp3 and 40% of TGF-β mRNA expression and a similar expression of IL-10 mRNA compared with the CD25highTIRC7− subset (Fig. 2⇓B). The CD25negTIRC7− subset displayed very low levels of IL-2 expression and a TGF-β mRNA level similar to the CD25lowTIRC7− subset, it did not express Foxp3 and IL-10 mRNA (Fig. 2⇓B). The analysis of intracellular staining of Foxp3 confirmed that the CD25lowTIRC7+ and CD25highTIRC7− cells expressed the Foxp3 protein at an equivalent level (Fig. 2⇓C).
We next undertook the determination of the suppressive activity of these subsets in vitro in a coculture system (Fig. 3⇓A). In contrast to sorted CD25lowTIRC7− cells, both sorted CD25lowTIRC7+ and CD25highTIRC7− cells were able to suppress the APC- and anti-CD3-driven proliferation of CD25−CD4+ T cells in vitro. We also evaluated the relative proliferative capacity of each of the Foxp-3+ Treg subsets. Compared with control non-Tregs (CD25−CD4+ T cells), both Treg subsets were anergic or unresponsive to stimulation with anti-CD3 and APCs (Fig. 3⇓A and data not shown). Moreover, both Treg subsets suppressed the APC- and anti-CD3-driven proliferation of CD25−CD4+ T cells in a dose-dependent manner with a similar efficiency (Fig. 3⇓B). Regardless of the difference in the expression of some cell surface markers, our results demonstrated that both CD25highTIRC7− and CD25lowTIRC7+ subsets share key characteristics of Tregs. Interestingly, the addition of a neutralizing anti-IL-10R mAb reverted only the suppressive effect of the CD25lowTIRC7+ cells. This result suggested that the suppressive effect of these cells is mediated by the secretion of IL-10, an immunosuppressive cytokine.
To confirm the role of IL-10 in the suppressive effect of the CD25lowTIRC7+ cells, we have used IL-10-deficient mice. As shown in Fig. 3⇑C, FACS analysis revealed the presence of the CD25highTIRC7− and CD25lowTIRC7+ subsets in the MLNs. Sorted CD25lowTIRC7+ cells from IL-10-deficient mice were unable to suppress the APC- and anti-CD3-driven proliferation of CD25−CD4+ T cells in vitro in a coculture system. Moreover, transwell experiments indicated that sorted CD25lowTIRC7+ cells from normal mice inhibit the proliferation of CD25−CD4+ T cells in the bottom well (Fig. 3⇑D), whereas CD25lowTIRC7+ cells from IL-10-deficient mice were unable to inhibit this proliferation (Fig. 3⇑D). These results confirmed that the suppressive effect of the CD25lowTIRC7+ cells is IL-10 dependent and cell contact independent. In contrast, CD25highTIRC7− cells sorted from both IL-10-deficient mice and normal mice suppress the response of CD25−CD4+ T cells, in vitro in a coculture system (Fig. 3⇑C), as previously shown (22), but not in a transwell system (Fig. 3⇑D). This result demonstrated that the suppressive effect of the CD25highTIRC7− cells is cytokine independent and cell contact dependent.
The CD25lowTIRC7+ subset inhibits the development of CD4+CD45RBhigh-transferred colitis
The CD25lowTIRC7+ subset is highly enriched in MLNs and PPs and may provide a mechanism capable of controlling intestinal inflammation. Thus, we tested in vivo the regulatory function of the CD25lowTIRC7+ subset using the well-characterized model of inflammatory bowel disease induced by the transfer of naive CD4+CD45RBhigh T cells into syngenic immunodeficient RAG1−/− mice (23, 24). i.p. injection of CD45RBhigh T cells into RAG1−/− mice induced, after 4 wk, the development of clinical signs of colitis, including piloerection, hunching, diarrhea, and weight loss (Fig. 4⇓A), as previously described (24). Colitis was characterized by a moderate epithelial cell hyperplasia, a leukocytic infiltrate, and a diminution of the number of mucin-secreting cells (Fig. 4⇓B). Interestingly, cotransfer of the CD25lowTIRC7+ subset together with the CD45RBhigh T cells into RAG1−/− mice inhibited the development of colitis, as revealed by the absence of weight loss (Fig. 4⇓A) and other clinical symptoms. Furthermore, colon histological examination showed a reduction of the epithelial cell hyperplasia, a rare leukocytic infiltrate, and the reappearance of mucin-secreting cells (Fig. 4⇓B). This result was similar to the one observed when the CD25highTIRC7− positive control subset was cotransferred: recipient mice typically preserved their body weight (Fig. 4⇓B) and displayed no clinical symptom of colitis. Our results demonstrated that the CD4+CD25+ cells are constituted of two regulatory subsets (CD25highTIRC7− and CD25lowTIRC7+), both displaying a suppressive activity in vivo consistent with their Foxp3+ expression.
To compare the regulatory potency in vivo between the CD25highTIRC7− and the CD25lowTIRC7+ Tregs subsets, their ability to inhibit the development of colitis after transfer of CD45RBhigh T cells into RAG1−/− mice was tested using two ratios of Tregs and CD45RBhigh T cells. Based on our in vitro analysis (Fig. 3⇑B), we have used a ratio of 1:1 and 1:4 corresponding to the more suppressive activity and an intermediate suppressive activity respectively. Analysis of the body weights, the clinical and histological scores of mice reconstituted with naive plus protective Tregs, revealed that both CD25lowTIRC7+ and CD25highTIRC7− cells have the same suppressive effect at a ratio of 1:1, and that both lost this suppressive function in a similar manner at a ratio of 1:4 (Fig. 4⇑, C and D). These data clearly show that the CD25highTIRC7− and CD25lowTIRC7+ Treg subsets have an equivalent regulatory potency in vivo.
The CD25lowTIRC7+ subset is an IL-10-producing subset and derive from CD25highTIRC7−
We next addressed the relationship between CD25highTIRC7− and CD25lowTIRC7+ regulatory subsets. We have purified these two subsets from MLNs of normal mice by FACS sorting (Fig. 5⇓A). The sorted subsets were then stimulated with coated anti-CD3 and soluble anti-CD28 mAbs in the presence of 10 ng/ml IL-2. Our results showed that the sorted CD25lowTIRC7+ cells produce large amounts of IL-10 measured by ELISA after 48 h of stimulation (Fig. 5⇓C), but no mature TGF-β or IL-2 (data not shown). Furthermore, in contrast to the sorted CD25lowTIRC7+ cells that maintained a stable phenotype after stimulation, the sorted CD25highTIRC7− cells became TIRC7 positive after 5 days of stimulation and shifted to a typical CD25lowTIRC7+ phenotype after 14 days of stimulation (Fig. 5⇓B). The stimulation of the whole sorted CD4+CD25+ cells (referred as all CD25+ in the figure) gave rise to two populations, CD25highTIRC7− and CD25lowTIRC7+ cells (Fig. 5⇓B). More importantly, these cells maintained both Foxp3 expression (Fig. 5⇓D) and their suppressive activity (Fig. 5⇓E).
Furthermore, the addition of a neutralizing anti-IL-10R mAb reverted the suppressive effect of activated CD25highTIRC7− cells in coculture, confirming that these activated cells produce IL-10 necessary for their suppressive activity (Fig. 5⇑E). Like the CD25lowTIRC7+ cells, the activated CD25highTIRC7− cells are cell contact independent as attested by transwell experiments (data not shown). Altogether, these results showed that the CD25lowTIRC7+ IL-10-secreting cells derive from CD25highTIRC7− precursors.
CD25lowTIRC7+ IL-10-secreting cells accumulate inside the tumor
Previous studies have shown that natural CD4+CD25+ Tregs preferentially move to and accumulate into tumor sites to comprise >70% of CD4+ T cells. This selective accumulation of Tregs induces the suppression of local immune response leading to tumor progression (10, 25, 26). We have used the mouse-bearing C26 tumor model (27) to confirm the presence of CD25lowTIRC7+ IL-10-secreting cells in vivo as effector Tregs. Indeed, we examined the tumor-infiltrating lymphocytes (TIL) in our animal model by FACS staining. Although few lymphocytes infiltrate the tumor (<1% of the entire tumor cellularity), we found that the percentage of CD4+CD25+ T cells increased dramatically and reached 90% of the CD4+ T cells present inside the tumor (Fig. 6⇓A). Among the TIL, 80% were CD25lowTIRC7+ (Fig. 6⇓A). More importantly, the tumor-infiltrating CD25lowTIRC7+ T cells maintained Foxp3 expression (Fig. 6⇓B). To examine whether these tumor-infiltrating CD25lowTIRC7+ T cells suppress T cell responses in vitro, we isolated these cells from the tumor tissues and cocultured them with purified CFSE-labeled CD4+CD25− cells from the spleen of naive mice. We observed that these tumor-infiltrating CD25lowTIRC7+ T cells did significantly suppress the proliferation of CD4+CD25− cells measured by CFSE dilution, and that IL-10 was involved in this suppression mechanism (Fig. 6⇓C).
In an additional series of experiments, we purified DCs from the spleen of TBM and control mice (tumor-free mice:TFM) and evaluated their capacity to induce the proliferation of tumor-infiltrating CD25lowTIRC7+ T cells. Splenic DCs from TBM induced the proliferation of tumor-infiltrating CD25lowTIRC7+ T cells, whereas splenic DCs from control mice failed to stimulate tumor-infiltrating CD25lowTIRC7+ T cell proliferation (data not shown). Moreover, tumor-infiltrating CD25lowTIRC7+ T cells produced high levels of IL-10 only in response to DCs from TBM (Fig. 6⇑E). To formally assess the antigenic specificity of the CD25lowTIRC7+ cells, we addressed the capacity of DCs generated in vitro from the BM to present tumor Ags (C26 cell lysate) to purified CD25lowTIRC7+ TIL. As shown in Fig. 6⇑D, in contrast to BMDCs loaded by OVA, the BMDCs loaded by C26 cell lysate were capable to induce the proliferation of CD25lowTIRC7+ TIL with IL-10 secretion (Fig. 6⇑E). These results suggested that tumor-infiltrating CD25lowTIRC7+ T cells are able to respond to DCs in a tumor-specific environment.
Together, these data confirm that the CD25lowTIRC7+ cells represent IL-10-producing CD4+CD25+ effector Tregs in vivo.
Emerging evidences suggest that CD4+ Tregs are key mediators of peripheral tolerance. Two major Treg populations have been described so far, and are designated as naturally occurring CD4+CD25+ Tregs (28) and inductible Tregs or IL-10-secreting Treg type 1 (Tr1) cells (29, 30). Many studies have indicated that inductible Tregs mediate their inhibitory activities by producing immunosuppressive cytokine, such as IL-10 in vitro and in vivo (5, 18). In contrast, according to the majority of studies, the in vitro suppression by natural Tregs was shown to be independent of IL-10 but required cell-cell contact (31), whereas their in vivo suppression favors the production of immunosuppressive cytokines, such as IL-10 (32, 33).
In this study, we show that in normal naive mice, the CD25highTIRC7− and CD25lowTIRC7+ subsets represent two Treg subsets that are characterized by the expression of Foxp3 and suppressive activity in vitro and in vivo. Interestingly, we showed that the CD25lowTIRC7+ subset derived from CD25highTIRC7− subset and represent IL-10-secreting Tregs in vitro.
The ex vivo analysis of TIL in the C26-bearing tumors model revealed that the CD4+CD25+ cells selectively accumulate inside the tumor and comprise the majority of the total TIL (Fig. 6⇑A), as previously shown by P. Yu et al. (10). More interestingly, among these tumor infiltrating CD4+CD25+ cells, 80% were CD25lowTIRC7+. Furthermore, these cells proliferate and produce IL-10 in response to DCs purified solely from mice bearing tumors. This report strongly supports the concept describing the existence of a pool of activated Tregs derived from the pool of resting Tregs in draining lymph nodes and at the site of inflammation (11, 34, 35) and points TIRC7 as a new surface marker, capable of distinguishing different stages of Treg activation.
In a healthy gastrointestinal tract, the mucosal immune response protects against pathogens while maintaining tolerance to organisms that compose the normal enteric microbiota. In this report, we showed that IL-10-secreting CD25lowTIRC7+ cells are selectively accumulated in PP, which are believed to be the principal sites for induction of tolerance. Similarly, studies from Levine’s group have demonstrated that the majority of T cells from PP display an activated memory phenotype and that PP favor differentiation of an IL-10-producing T cells with suppressive activity (36). The identification of IL-10-secreting CD25lowTIRC7+ cells in normal PPs and MLNs suggests a constitutive role for those cells in maintaining the tolerance to organisms that compose the normal enteric microbiota. Our data are supported by the recent work from Powrie’s group that shows the presence of Foxp3+ IL-10-secreting cells in both the secondary lymphoid organs and the colon of experimental colitic mice (35).
Selective accumulation of Tregs in the tumor environment can be one important local factor for suppression of immune responses against a strong tumor Ag leading to progressive growth of cancer in the immune-competent hosts (26). A closer examination of TIL in our animal model revealed that the CD25lowTIRC7+ cells accumulate inside the tumor and comprise the majority of total TIL. We also showed that CD25lowTIRC7+ cells mediate a suppressive activity via their IL-10 secretion. Supporting this observation, a recent study has demonstrated that the local depletion of CD4+ T cells inside the tumor, in a murine fibrosarcoma model, induced a drastic decrease of IL-10 level, and led to the eradication of tumors and the development of a long-term antitumor memory (10). Taken together, these results suggest that IL-10-secreting CD25lowTIRC7+ cells maintained an anti-inflammatory environment that inhibited antitumor immunity inside the tumor.
Similarly, recent studies have demonstrated the presence of IL-10-secreting CD4+CD25+ cells in the colon of normal and colitic mice and in prediabetic and Leishmania major lesions (11, 35, 37). In our C26-bearing tumor model, the CD25+TIRC7+ cells proliferate and produce IL-10 in response to DCs purified solely from mice bearing tumors. Therefore, we speculate that IL-10-secreting CD25lowTIRC7+ cells are derived from natural CD4+CD25+ Tregs in response to tumoral Ag. Supporting evidences for this observation that natural Tregs can respond to foreign Ags come from studies on Schistosoma and Leishmania infections. Natural Tregs taken from chronically infected mice can secrete IL-10 in response to parasite Ags (11, 38). Our present observations support the notion that, in the pool of natural Tregs, some can recognize tumor Ags and that this recognition is an essential step in their expansion and regulatory function. Furthermore, the identification of IL-10-secreting CD25lowTIRC7+ cells in secondary lymphoid organs and at tumor site suggests that the suppressive action of Tregs is functional in both case.
In contrast, we and others have described other IL-10-secreting Tregs or Tr1 cells, which can develop under different regimens of antigenic stimulation, both in vitro and in vivo (18, 39, 40). Furthermore, in contrast to natural CD4+CD25+ Tregs, the forkhead transcription factor Foxp3 expression is not a prerequisite for Tr1 cell activity and suppose that Tr1 and natural CD4+CD25+ Tregs are independent (41, 42). However, the relationship between Tr1 cells and the Ag-driven IL-10-secreting CD25+TIRC7+ Tregs in vivo remains to be determined. More recently, an elegant study using IL-10 and Foxp3 dual-transgenic reporter mice demonstrated that in the steady state in vivo, coordinate or differential expression of IL-10 and Foxp3 by CD4+ T cells defines three distinct subsets of Tregs, Foxp3+IL-10−, Foxp3+IL-10+, and Foxp3−IL-10+ referred to as Tr1 cells (44). Our present study confirms and extends these data, showing that TIRC7 is a marker for the Foxp3+IL-10-producing Tregs and that this population derives from the Foxp3+IL-10− T cells (natural Tregs).
TIRC7 was identified by differential display on mRNA obtained from human-activated T cells compared with resting T cells. Targeting of TIRC7 with specific Abs causes inhibition of Th1 cytokine secretion in human lymphocytes (12). In contrast, TIRC7-deficient mice show an enhanced T cell proliferation and Th1 cytokine secretion, suggesting a role for TIRC7 in regulating T cell response (43). However, the role of TIRC7 in IL-10-secreting CD25lowTIRC7+ cells remains to be determined.
Finally, the ability to use TIRC7 as a marker for Ag-driven IL-10-secreting CD25+ Tregs provides an opportunity to track the fate of these cells in vivo. Our results also open new therapeutic perspectives for the use TIRC7 protein as a target at the tumor site for locally depleting the activated Tregs and enhancing antitumor immunity.
We are grateful to Dr. A. Saoudi for critical comments on the manuscript, Dr. M. Rouleau for helpful discussions, and Dr. D. Burke for reading the manuscript. Histological analyses were kindly performed at the Department of anatomo-pathology (Hopital Pasteur, Nice). We thank V. Verhasselt and V. Julia for providing the IL-10-deficient mice and M. P. Colombo and A. Schmid-Alliana for the gift of the C26 colon carcinoma cells. We thank J. C. Scimeca for the purification of the anti-TIRC7 Ab, D. Quincey for technical help, and M. Topi for animal care.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
↵1 This work was supported by a fellowship of the Fondation pour la Recherche Médicale (to A.W.). This work was also funded by the Centre National de la Recherche Scientifique and the Association pour la Recherche sur le Cancer.
↵2 Address correspondence and reprint requests to Dr. Abdelilah Wakkach, GEPITOS, Université Nice Sophia-Antipolis, Centre National de la Recherche Scientifique, Unité de Formation et de Recherche de Médecine, 28 avenue de Valombrose, 06100 Nice, France. E-mail address:
↵3 Abbreviations used in this paper: Treg, T regulatory cell; TIRC7, T cell immune response cDNA 7; MLN, mesenteric lymph node; DC, dendritic cell; BMDC, bone marrow-derived DC; TBM, tumor bearing mice; TFM, tumor free mice; TIL, tumor-infiltrating lymphocyte; Tr1, Treg type 1; PP, Peyer’s patches.
- Received May 4, 2007.
- Accepted February 26, 2008.
- Copyright © 2008 by The American Association of Immunologists