Lymph Node Occupancy Is Required for the Peripheral Development of Alloantigen-Specific Foxp3+ Regulatory T Cells
http://www.100md.com
免疫学杂志 2005年第11期
We previously demonstrated that L-selectin (CD62L)-dependent T cell homing to lymph nodes (LN) is required for tolerance induction to alloantigen. To explore the mechanisms of this observation, we analyzed the development and distribution of regulatory T cells (Treg), which play an important protective role against allograft rejection in transplantation tolerance. Alloantigen-specific tolerance was induced using either anti-CD2 plus anti-CD3 mAbs, or anti-CD40L mAbs plus donor-specific transfusion, in fully mismatched (BALB/c donor, C57BL/6 recipient) vascularized cardiac allografts. An expansion of CD4+CD25+CD62Lhigh T cells was observed specifically within the LN of tolerant animals, but not in other anatomic sites or under nontolerizing conditions. These cells exhibited a substantial up-regulation of Foxp3 expression as measured by real-time PCR and by fluorescent immunohistochemistry, and possessed alloantigen-specific suppressor activity. Neither LN nor other lymphoid cells expressed the regulatory phenotype if recipients were treated with anti-CD62L mAbs, which both prevented LN homing and caused early allograft rejection. However, administration of FTY720, a sphingosine 1-phosphate receptor modulator that induces CD62L-independent T cell accumulation in the LNs, restored CD4+CD25+ Treg in the LNs along with graft survival. These data suggest that alloantigen-specific Foxp3+CD4+CD25+ Treg develop and are required within the LNs during tolerization, and provide compelling evidence that distinct lymphoid compartments play critical roles in.
Introduction
The major goal of transplantation is the creation of clinically applicable protocols to induce alloantigen-specific tolerance. Long-term graft survival and donor-specific transplantation tolerance, in the absence of chronic immunosuppression, can be induced by a number of different regimens (1, 2, 3, 4). It is still not entirely clear by which mechanisms these regimens induce tolerance (5), although there is increasing evidence that tolerization protocols favor the development of regulatory or suppressor T cells (6, 7), and that tolerization strategies must be directed to induce, expand, or manipulate regulatory suppressive T cells (8, 9). The most extensively studied regulatory cells are the naturally occurring CD4+CD25+ T regulatory cells (Treg)4 generated in the thymus, constituting 5–10% of peripheral CD4+ T cells in mice (10). The transcription factor Foxp3 represents a unique marker involved in the development and function of Treg and is specifically expressed in CD4+CD25+ suppressor T cells (11, 12, 13). This population of suppressor T cells is thought to play an essential protective role during tolerization to alloantigen, and presence of Treg in transplanted recipients is tightly associated with indefinite graft survival (14), whereas removal of CD4+CD25+ Treg favors the production of alloantibodies (15) and graft rejection (16).
Although many studies have been conducted to define the cellular and molecular mechanisms by which Treg develop (17, 18, 19), there is little knowledge about the anatomic compartments where Foxp3-expressing Treg are activated, expanded, or display suppressor function for tolerization to ensue. Distinct lymphoid and nonlymphoid compartments may be differentially involved in the development of Treg, because the lymphoid environment plays essential and diverse roles in either rejection or tolerance (20). Lymphoid architecture also plays an important role in Treg development and differentiation, because differences in lymphoid origin have been used to classify two distinct subsets of Treg cells: natural (thymus) vs adaptive (periphery), according to structural site of generation (21). Naturally occurring CD4+CD25+ Treg are generated as a result of multiple selection events during T cell development within the thymus (18, 22). Evidence for extrathymic CD4+CD25+ T cell development supports an alternative de novo pathway for the origin of Treg in the periphery (23), and indicates that CD4+CD25+ Treg generated in the periphery are probably different from thymically produced CD25+CD4+ Treg (21). Therefore, it is likely that particular anatomic sites provide specific milieus that allow tolerization to occur by permitting suppressor Treg to be activated, expanded, or function in the periphery. Thus, defining the Foxp3-expressing CD4+CD25+ T cell population in distinct anatomic compartments may elucidate the precise sites where tolerization takes place.
We previously demonstrated that CD62L-mediated lymph node (LN) homing is necessary for the induction of anti-CD2 plus anti-CD3 mAb-induced tolerance, because coadministration of anti-CD62L mAbs prevents both LN homing and tolerance, in both nonvascularized and vascularized cardiac transplant models (24). Adoptive transfer of CD62L–/– T cells and the use of CD62L–/– recipients provided complementary genetic evidence to confirm the importance of CD62L and LN occupancy by T cells in tolerization. In this report, we further characterized these observations by studying the CD4+CD25+ Treg population in distinct anatomic domains after treatment with two different tolerization protocols (3, 4), and after manipulating T cell LN homing with anti-CD62L mAb and the sphingosine 1-phosophate receptor modulator FTY720 (reviewed in Ref.25). Our results demonstrate that alloantigen-specific Treg occupancy in the LNs of tolerant animals is necessary for immunological tolerance to vascularized cardiac allografts. Alloantigen-specific Treg were expanded in recipient LNs, but not other peripheral sites, under both tolerization protocols. Anti-CD62L mAbs prevented T cell LN homing, resulting in abrogation of both Treg expansion in the LNs or other lymphoid areas and graft survival, whereas FTY720-driven T cell LN sequestration restored both LN Treg expansion and graft survival. Together, the results suggest an essential role of the LN for the development and function of CD4+CD25+ Treg in tolerant animals.
Materials and Methods
Statistical analysis
For graft survival, one-way ANOVA was performed. For CD4+CD25+ T cell growth, t test and F test were performed to compare variances. For cell proliferation, one-way ANOVA was performed at each time point. The group effects were all significant at p 0.05. To examine individual differences, comparison between every pair of groups was performed. For Foxp3+ T cell counts and chemokine receptor gene expression, one-way ANOVA and Dunnett test were performed, to examine individual differences compared with the naive control.
Results
LN occupancy is required to induce tolerance to vascularized cardiac transplants
We have previously shown that lymphocyte LN homing and localization are required for anti-CD2 plus anti-CD3 mAb-induced tolerance to nonvascularized and vascularized heart allografts (24). We next investigated whether this finding held true using another well-defined tolerizing regimen, DST plus anti-CD40L mAb (4). BALB/c donor vascularized cardiac grafts were transplanted into C57BL/6 recipients, which received either anti-CD2 plus anti-CD3 mAbs, DST plus anti-CD40L mAbs, or were left untreated. Both tolerance protocols induce long-term graft survival with normal histology in vascularized allografts (3, 28) (Fig. 1A). Third-party CBA (H-2k), but not donor strain BALB/c (H-2d), second-set allografts placed >60 days following the initial transplant were rejected (Refs.29 and 30 , and data not shown), demonstrating donor alloantigen-specific tolerance. Administration of anti-CD62L mAb prevented allograft survival in animals given DST plus anti-CD40L mAb (Fig. 1B). Coadministration of anti-CD62L mAb decreased mean survival time from >70.6 ± 13.2 to 28.0 ± 6.8 days in the anti-CD2 plus anti-CD3 mAb-treated group, and from >88.2 ± 1.8 to 39.6 ± 0.9 days in the DST plus anti-CD40L mAb-treated group (Fig. 1B). These results demonstrate that both the anti-CD2 plus anti-CD3 mAb and the DST plus anti-CD40L mAb tolerogenic regimens require CD62L-dependent LN homing to induce prolonged survival and tolerance in a vascularized cardiac allograft model. Furthermore, administration of FTY720 to anti-CD62L mAb plus tolerogen-treated mice (Fig. 1C) restored graft survival with normal allograft histology, and supports the hypothesis that LN occupancy is required for tolerance.
Discussion
We report for the first time that alloantigen-specific Foxp3+ Treg accumulate in vivo in the LNs of tolerant animals as a result of either expansion of naturally occurring Treg and/or development of de novo Treg. Treg accumulating in the LN under two distinct tolerogenic treatments (anti-CD2 plus anti-CD3 mAbs, DST plus anti-CD40L mAb) meet many of the criteria established for their identification. They are CD4+CD25+CD44highCD45RBlowCD62LhighCD69low, Ag experienced, partially activated, anergic, suppressive, alloantigen specific, and Foxp3+. LN accumulation is tightly linked to the induction and maintenance of tolerance, because interfering with T cell LN occupancy with anti-CD62L mAb prevents localization of Treg to the LN and tolerance, whereas sphingosine 1-phosphate receptor modulation restores both Treg presence in lymphoid tissues and tolerance. These observations suggest that Treg occupancy of the LN, but not the spleen, is a generalized phenomenon in the establishment of tolerance, and reveals the peripheral sites where Treg develop in vivo in response to alloantigen, and where they likely exercise functional effector activities.
Several laboratories, including our own (38, 39, 40, 41, 42), have reported the induction of transplantation tolerance using a variety of distinct approaches that interfere with signals 1, 2, and/or 3. The mechanisms that prevent graft rejection following these therapies have not been fully characterized, although evidence for anergy, ignorance, deletion, immune deviation, and immune regulation has been reported (43, 44, 45, 46, 47). Recent studies have also implicated Treg in tolerance induction (7, 48) and have shown that CD4+CD25+ Treg can also develop in the periphery through thymus-independent pathways (48, 49). To further address the role of Treg in tolerance, we induced tolerance to vascularized cardiac transplants by using two distinct tolerant protocols, and characterized the CD4+CD25+ T cell population in distinct anatomic compartments at different time points. In the anti-CD40L plus DST model, other investigators have provided evidence that anergy, activation-induced cell death, immune deviation, and Treg contribute to the induction and maintenance of tolerance (29). In the anti-CD2 plus anti-CD3 mAb model, graft survival and tolerance were related to mechanisms that included partial T cell activation, anergy, immune deviation, and partial T cell depletion (3, 30). In this report, we now provide evidence that both of these tolerance protocols also require the activity of Foxp3+CD4+CD25+ Treg and that these cells must be localized to the LNs but not the spleen.
Evidence from a variety of autoimmune studies suggests that the LN is a critical site for Ag presentation that determines either priming or tolerization (21). These models suggest that natural Treg expand in the pancreatic LN as a consequence of exposure to self-Ag, and failure to generate CD4+CD25+ T cells correlates with accelerated diabetes progression in the nonobese diabetic mouse (50, 51). The increase in CD4+CD25+ T cells occurs in the draining LN, but not the spleen or other peripheral LNs (6), supporting the idea that distinct anatomic sites play different roles in autoimmunity. Transplant studies are significantly different from the autoimmune model in many respects. With regard to the role of secondary lymphoid organs, recent reports demonstrated that alloantigen is presented not in a localized fashion, but systemically in all recipient lymphoid organs very soon after allografting (52, 53). As a result, multiple lymphoid tissues participate simultaneously in alloantigen presentation, which makes it more difficult to define the role of each secondary lymphoid organ and its microdomains in transplant rejection or tolerance. Our results here show at the structural level, that the LN architecture in the tolerized animals is preserved, and is similar to that of naive mice, whereas in the spleen the architecture is significantly altered (Fig. 2). Conversely, the architecture in the spleen of the acutely rejecting untreated mice is preserved, but is disorganized in the LNs. Thus, as in autoimmune models, the lymphoid organs and their anatomic structure where lymphocytes encounter and interact with alloantigen under the influence of systemic immunosuppression are critical determinants for the development of either immunity or tolerance. In particular, LNs are important for actively supporting tolerization.
Secondary lymphoid tissues have a highly organized architecture, regulating complex T cell-APC interactions (54, 55) that determine the development of either rejection or tolerance (20). Lakkis et al. (20) reported that asplenic Hox11–/– mice reject cardiac and skin allografts, whereas splenectomized aly/aly mice that lack all secondary lymphoid organs do not reject either skin or vascularized heart allografts. These results were interpreted as showing that secondary lymphoid organs are required for priming and initiation of an alloresponse, and indicate that the spleen is not absolutely necessary for rejection, as long as other lymphoid tissues are intact. Aly/aly mice reject heart but not skin grafts, arguing that it is the draining lymphoid tissues that are responsible for proper Ag presentation and T cell priming; and that the designation of "draining" is dependent on the type of allograft, its anatomic location, and the surgical manipulation required for its placement. In contrast, Zhou et al. (56) showed that splenectomized lymphotoxin- and lymphotoxin- receptor gene knockout mice, that also lack secondary lymphoid organs, reject skin and heart allografts, demonstrating that secondary lymphoid organs are not always necessary for initiating the alloimmune response. These contradictory results on the role of secondary lymphoid organs in transplant rejection may relate to the use of genetically altered mice, which have immunologic and developmental differences due to both gene knockout and background strain variations. We also suggest that experimental manipulations not only cause differences in T cell priming that lead to graft rejection, but also cause differences in T cell priming that lead to Treg development, with subsequent effects on suppressive and tolerogenic responses to alloantigen. Therefore, the study of lymphoid organs and domains that are responsible for priming and rejection must now be coupled to studies that analyze these same domains for the induction or maintenance of tolerance (57). This conclusion is supported by a recent report that shows that autoimmune diabetes is the result of the balance between effector and regulatory T cells in the pancreatic LN, and that evaluation of only a single T cell subset in the LN may lead to incomplete conclusions (58).
Finally, the mechanisms by which Treg accumulate or develop in the LN of tolerant animals may be related to chemokine compartmentalization. Table I shows that CCR7 is up-regulated in LN CD4+CD25+ T cells during tolerance, but not during rejection, which suggests that CD4+ T cells may migrate to the LNs where they further develop during tolerance. Therefore, it is possible that under the cover of immunosuppressive therapy such as anti-CD2 plus anti-CD3 mAb or DST plus anti-CD40L mAb, CD4+ T cells home to the LNs, where altered TCR signal transduction, partial T cell activation, and/or low-affinity Ag binding in the context of persistent allo- and tissue-specific Ag exposure, favor the development of Treg. Furthermore, chemokine-dependent LN T cell migration may represent a mechanism directly responsible for preventing CD4+ T cell-mediated rejection. H?pken et al. (59) recently described a key role for CCR7 in alloreactive T cell priming within the LNs, and specific localization of Treg within the lymphoid compartment may interfere with successful effector T cell priming that leads to graft rejection. This implies that anti-CD62L-treated T cells become primed to alloantigen, yet are not subject to tolerogenic influences because of their inability to home to the LNs, whereas FTY720 induces homing and colocalization of both effector and Treg cells to the LN.
We conclude that, whereas alloantigen can prime effector T cells either in the LNs or the spleen to initiate an immune response that leads to graft rejection, Treg can only be activated and/or expanded in an alloantigen-dependent manner in the LNs under systemic tolerization protocols. Our results reveal the location where thymic-derived CD4+CD25+ Treg may expand, and/or where CD4+ T cells may develop de novo into alloantigen-specific Treg in the peripheral lymphoid environment. CD4+CD25+ T cells localized near the high endothelial venule conduits within the LN microenvironment may help to maintain a tolerant environment by regulating multiple interactions between naive T cells with APC. This finding may help to unmask the factors that promote the differentiation and activation of regulatory suppressor cells and allow the generation of extrathymic Treg for the control of immune responses.
Acknowledgments
We acknowledge the technical contributions of Minwei Mao, Dan Chen, Italas George, and Patricia Rebollo, and the helpful discussion with Dr. Gwendalyn Randolph.
Footnotes
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 National Institutes of Health Grants R01 AI41428, AI62765, and AI44929 (to J.S.B.).
2 Address correspondence and reprint requests to Dr. Jonathan S. Bromberg or Dr. Jordi C. Ochando, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1104, New York, NY 10029-6574. E-mail address: jon.bromberg{at}msnyuhealth.org or jordi.ochando{at}mssm.edu
3 Current address: Bristol-Myers Squibb, Princeton, NJ 08543.
4 Abbreviations used in this paper: Treg, T regulatory cell; LN, lymph node; DST, donor-specific transfusion; SI, stimulation index; TMN, total mononuclear cells.
Received for publication November 15, 2004. Accepted for publication March 22, 2005.
References
Cobbold, S. P., A. Jayasuriya, A. Nash, T. D. Prospero, H. Waldmann. 1984. Therapy with monoclonal antibodies by elimination of T-cell subsets in vivo. Nature 312: 548-551.
Isobe, M., H. Yagita, K. Okumura, A. Ihara. 1992. Specific acceptance of cardiac allograft after treatment with antibodies to ICAM-1 and LFA-1. Science 255: 1125-1127.
Chavin, K. D., L. Qin, J. Lin, H. Yagita, J. S. Bromberg. 1993. Combined anti-CD2 and anti-CD3 receptor monoclonal antibodies induce donor-specific tolerance in a cardiac transplant model. J. Immunol. 151: 7249-7259.[
Parker, D. C., D. L. Greiner, N. E. Phillips, M. C. Appel, A. W. Steele, F. H. Durie, R. J. Noelle, J. P. Mordes, A. A. Rossini. 1995. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc. Natl. Acad. Sci. USA 92: 9560-9564.
Shevach, E. M.. 2002. CD4+CD25+ suppressor T cells: more questions than answers. Nat. Rev. Immunol. 2: 389-400.
Belghith, M., J. A. Bluestone, S. Barriot, J. Megret, J. F. Bach, L. Chatenoud. 2003. TGF--dependent mechanisms mediate restoration of self-tolerance induced by antibodies to CD3 in overt autoimmune diabetes. Nat. Med. 9: 1202-1208.
Cobbold, S. P., R. Castejon, E. Adams, D. Zelenika, L. Graca, S. Humm, H. Waldmann. 2004. Induction of foxP3+ regulatory T cells in the periphery of T cell receptor transgenic mice tolerized to transplants. J. Immunol. 172: 6003-6010.
Waldmann, H., S. Cobbold. 2001. Regulating the immune response to transplants: a role for CD4+ regulatory cells?. Immunity 14: 399-406.
Wood, K. J., S. Sakaguchi. 2003. Regulatory T cells in transplantation tolerance. Nat. Rev. Immunol. 3: 199-210.(Jordi C. Ochando2,*, Adam)
Introduction
The major goal of transplantation is the creation of clinically applicable protocols to induce alloantigen-specific tolerance. Long-term graft survival and donor-specific transplantation tolerance, in the absence of chronic immunosuppression, can be induced by a number of different regimens (1, 2, 3, 4). It is still not entirely clear by which mechanisms these regimens induce tolerance (5), although there is increasing evidence that tolerization protocols favor the development of regulatory or suppressor T cells (6, 7), and that tolerization strategies must be directed to induce, expand, or manipulate regulatory suppressive T cells (8, 9). The most extensively studied regulatory cells are the naturally occurring CD4+CD25+ T regulatory cells (Treg)4 generated in the thymus, constituting 5–10% of peripheral CD4+ T cells in mice (10). The transcription factor Foxp3 represents a unique marker involved in the development and function of Treg and is specifically expressed in CD4+CD25+ suppressor T cells (11, 12, 13). This population of suppressor T cells is thought to play an essential protective role during tolerization to alloantigen, and presence of Treg in transplanted recipients is tightly associated with indefinite graft survival (14), whereas removal of CD4+CD25+ Treg favors the production of alloantibodies (15) and graft rejection (16).
Although many studies have been conducted to define the cellular and molecular mechanisms by which Treg develop (17, 18, 19), there is little knowledge about the anatomic compartments where Foxp3-expressing Treg are activated, expanded, or display suppressor function for tolerization to ensue. Distinct lymphoid and nonlymphoid compartments may be differentially involved in the development of Treg, because the lymphoid environment plays essential and diverse roles in either rejection or tolerance (20). Lymphoid architecture also plays an important role in Treg development and differentiation, because differences in lymphoid origin have been used to classify two distinct subsets of Treg cells: natural (thymus) vs adaptive (periphery), according to structural site of generation (21). Naturally occurring CD4+CD25+ Treg are generated as a result of multiple selection events during T cell development within the thymus (18, 22). Evidence for extrathymic CD4+CD25+ T cell development supports an alternative de novo pathway for the origin of Treg in the periphery (23), and indicates that CD4+CD25+ Treg generated in the periphery are probably different from thymically produced CD25+CD4+ Treg (21). Therefore, it is likely that particular anatomic sites provide specific milieus that allow tolerization to occur by permitting suppressor Treg to be activated, expanded, or function in the periphery. Thus, defining the Foxp3-expressing CD4+CD25+ T cell population in distinct anatomic compartments may elucidate the precise sites where tolerization takes place.
We previously demonstrated that CD62L-mediated lymph node (LN) homing is necessary for the induction of anti-CD2 plus anti-CD3 mAb-induced tolerance, because coadministration of anti-CD62L mAbs prevents both LN homing and tolerance, in both nonvascularized and vascularized cardiac transplant models (24). Adoptive transfer of CD62L–/– T cells and the use of CD62L–/– recipients provided complementary genetic evidence to confirm the importance of CD62L and LN occupancy by T cells in tolerization. In this report, we further characterized these observations by studying the CD4+CD25+ Treg population in distinct anatomic domains after treatment with two different tolerization protocols (3, 4), and after manipulating T cell LN homing with anti-CD62L mAb and the sphingosine 1-phosophate receptor modulator FTY720 (reviewed in Ref.25). Our results demonstrate that alloantigen-specific Treg occupancy in the LNs of tolerant animals is necessary for immunological tolerance to vascularized cardiac allografts. Alloantigen-specific Treg were expanded in recipient LNs, but not other peripheral sites, under both tolerization protocols. Anti-CD62L mAbs prevented T cell LN homing, resulting in abrogation of both Treg expansion in the LNs or other lymphoid areas and graft survival, whereas FTY720-driven T cell LN sequestration restored both LN Treg expansion and graft survival. Together, the results suggest an essential role of the LN for the development and function of CD4+CD25+ Treg in tolerant animals.
Materials and Methods
Statistical analysis
For graft survival, one-way ANOVA was performed. For CD4+CD25+ T cell growth, t test and F test were performed to compare variances. For cell proliferation, one-way ANOVA was performed at each time point. The group effects were all significant at p 0.05. To examine individual differences, comparison between every pair of groups was performed. For Foxp3+ T cell counts and chemokine receptor gene expression, one-way ANOVA and Dunnett test were performed, to examine individual differences compared with the naive control.
Results
LN occupancy is required to induce tolerance to vascularized cardiac transplants
We have previously shown that lymphocyte LN homing and localization are required for anti-CD2 plus anti-CD3 mAb-induced tolerance to nonvascularized and vascularized heart allografts (24). We next investigated whether this finding held true using another well-defined tolerizing regimen, DST plus anti-CD40L mAb (4). BALB/c donor vascularized cardiac grafts were transplanted into C57BL/6 recipients, which received either anti-CD2 plus anti-CD3 mAbs, DST plus anti-CD40L mAbs, or were left untreated. Both tolerance protocols induce long-term graft survival with normal histology in vascularized allografts (3, 28) (Fig. 1A). Third-party CBA (H-2k), but not donor strain BALB/c (H-2d), second-set allografts placed >60 days following the initial transplant were rejected (Refs.29 and 30 , and data not shown), demonstrating donor alloantigen-specific tolerance. Administration of anti-CD62L mAb prevented allograft survival in animals given DST plus anti-CD40L mAb (Fig. 1B). Coadministration of anti-CD62L mAb decreased mean survival time from >70.6 ± 13.2 to 28.0 ± 6.8 days in the anti-CD2 plus anti-CD3 mAb-treated group, and from >88.2 ± 1.8 to 39.6 ± 0.9 days in the DST plus anti-CD40L mAb-treated group (Fig. 1B). These results demonstrate that both the anti-CD2 plus anti-CD3 mAb and the DST plus anti-CD40L mAb tolerogenic regimens require CD62L-dependent LN homing to induce prolonged survival and tolerance in a vascularized cardiac allograft model. Furthermore, administration of FTY720 to anti-CD62L mAb plus tolerogen-treated mice (Fig. 1C) restored graft survival with normal allograft histology, and supports the hypothesis that LN occupancy is required for tolerance.
Discussion
We report for the first time that alloantigen-specific Foxp3+ Treg accumulate in vivo in the LNs of tolerant animals as a result of either expansion of naturally occurring Treg and/or development of de novo Treg. Treg accumulating in the LN under two distinct tolerogenic treatments (anti-CD2 plus anti-CD3 mAbs, DST plus anti-CD40L mAb) meet many of the criteria established for their identification. They are CD4+CD25+CD44highCD45RBlowCD62LhighCD69low, Ag experienced, partially activated, anergic, suppressive, alloantigen specific, and Foxp3+. LN accumulation is tightly linked to the induction and maintenance of tolerance, because interfering with T cell LN occupancy with anti-CD62L mAb prevents localization of Treg to the LN and tolerance, whereas sphingosine 1-phosphate receptor modulation restores both Treg presence in lymphoid tissues and tolerance. These observations suggest that Treg occupancy of the LN, but not the spleen, is a generalized phenomenon in the establishment of tolerance, and reveals the peripheral sites where Treg develop in vivo in response to alloantigen, and where they likely exercise functional effector activities.
Several laboratories, including our own (38, 39, 40, 41, 42), have reported the induction of transplantation tolerance using a variety of distinct approaches that interfere with signals 1, 2, and/or 3. The mechanisms that prevent graft rejection following these therapies have not been fully characterized, although evidence for anergy, ignorance, deletion, immune deviation, and immune regulation has been reported (43, 44, 45, 46, 47). Recent studies have also implicated Treg in tolerance induction (7, 48) and have shown that CD4+CD25+ Treg can also develop in the periphery through thymus-independent pathways (48, 49). To further address the role of Treg in tolerance, we induced tolerance to vascularized cardiac transplants by using two distinct tolerant protocols, and characterized the CD4+CD25+ T cell population in distinct anatomic compartments at different time points. In the anti-CD40L plus DST model, other investigators have provided evidence that anergy, activation-induced cell death, immune deviation, and Treg contribute to the induction and maintenance of tolerance (29). In the anti-CD2 plus anti-CD3 mAb model, graft survival and tolerance were related to mechanisms that included partial T cell activation, anergy, immune deviation, and partial T cell depletion (3, 30). In this report, we now provide evidence that both of these tolerance protocols also require the activity of Foxp3+CD4+CD25+ Treg and that these cells must be localized to the LNs but not the spleen.
Evidence from a variety of autoimmune studies suggests that the LN is a critical site for Ag presentation that determines either priming or tolerization (21). These models suggest that natural Treg expand in the pancreatic LN as a consequence of exposure to self-Ag, and failure to generate CD4+CD25+ T cells correlates with accelerated diabetes progression in the nonobese diabetic mouse (50, 51). The increase in CD4+CD25+ T cells occurs in the draining LN, but not the spleen or other peripheral LNs (6), supporting the idea that distinct anatomic sites play different roles in autoimmunity. Transplant studies are significantly different from the autoimmune model in many respects. With regard to the role of secondary lymphoid organs, recent reports demonstrated that alloantigen is presented not in a localized fashion, but systemically in all recipient lymphoid organs very soon after allografting (52, 53). As a result, multiple lymphoid tissues participate simultaneously in alloantigen presentation, which makes it more difficult to define the role of each secondary lymphoid organ and its microdomains in transplant rejection or tolerance. Our results here show at the structural level, that the LN architecture in the tolerized animals is preserved, and is similar to that of naive mice, whereas in the spleen the architecture is significantly altered (Fig. 2). Conversely, the architecture in the spleen of the acutely rejecting untreated mice is preserved, but is disorganized in the LNs. Thus, as in autoimmune models, the lymphoid organs and their anatomic structure where lymphocytes encounter and interact with alloantigen under the influence of systemic immunosuppression are critical determinants for the development of either immunity or tolerance. In particular, LNs are important for actively supporting tolerization.
Secondary lymphoid tissues have a highly organized architecture, regulating complex T cell-APC interactions (54, 55) that determine the development of either rejection or tolerance (20). Lakkis et al. (20) reported that asplenic Hox11–/– mice reject cardiac and skin allografts, whereas splenectomized aly/aly mice that lack all secondary lymphoid organs do not reject either skin or vascularized heart allografts. These results were interpreted as showing that secondary lymphoid organs are required for priming and initiation of an alloresponse, and indicate that the spleen is not absolutely necessary for rejection, as long as other lymphoid tissues are intact. Aly/aly mice reject heart but not skin grafts, arguing that it is the draining lymphoid tissues that are responsible for proper Ag presentation and T cell priming; and that the designation of "draining" is dependent on the type of allograft, its anatomic location, and the surgical manipulation required for its placement. In contrast, Zhou et al. (56) showed that splenectomized lymphotoxin- and lymphotoxin- receptor gene knockout mice, that also lack secondary lymphoid organs, reject skin and heart allografts, demonstrating that secondary lymphoid organs are not always necessary for initiating the alloimmune response. These contradictory results on the role of secondary lymphoid organs in transplant rejection may relate to the use of genetically altered mice, which have immunologic and developmental differences due to both gene knockout and background strain variations. We also suggest that experimental manipulations not only cause differences in T cell priming that lead to graft rejection, but also cause differences in T cell priming that lead to Treg development, with subsequent effects on suppressive and tolerogenic responses to alloantigen. Therefore, the study of lymphoid organs and domains that are responsible for priming and rejection must now be coupled to studies that analyze these same domains for the induction or maintenance of tolerance (57). This conclusion is supported by a recent report that shows that autoimmune diabetes is the result of the balance between effector and regulatory T cells in the pancreatic LN, and that evaluation of only a single T cell subset in the LN may lead to incomplete conclusions (58).
Finally, the mechanisms by which Treg accumulate or develop in the LN of tolerant animals may be related to chemokine compartmentalization. Table I shows that CCR7 is up-regulated in LN CD4+CD25+ T cells during tolerance, but not during rejection, which suggests that CD4+ T cells may migrate to the LNs where they further develop during tolerance. Therefore, it is possible that under the cover of immunosuppressive therapy such as anti-CD2 plus anti-CD3 mAb or DST plus anti-CD40L mAb, CD4+ T cells home to the LNs, where altered TCR signal transduction, partial T cell activation, and/or low-affinity Ag binding in the context of persistent allo- and tissue-specific Ag exposure, favor the development of Treg. Furthermore, chemokine-dependent LN T cell migration may represent a mechanism directly responsible for preventing CD4+ T cell-mediated rejection. H?pken et al. (59) recently described a key role for CCR7 in alloreactive T cell priming within the LNs, and specific localization of Treg within the lymphoid compartment may interfere with successful effector T cell priming that leads to graft rejection. This implies that anti-CD62L-treated T cells become primed to alloantigen, yet are not subject to tolerogenic influences because of their inability to home to the LNs, whereas FTY720 induces homing and colocalization of both effector and Treg cells to the LN.
We conclude that, whereas alloantigen can prime effector T cells either in the LNs or the spleen to initiate an immune response that leads to graft rejection, Treg can only be activated and/or expanded in an alloantigen-dependent manner in the LNs under systemic tolerization protocols. Our results reveal the location where thymic-derived CD4+CD25+ Treg may expand, and/or where CD4+ T cells may develop de novo into alloantigen-specific Treg in the peripheral lymphoid environment. CD4+CD25+ T cells localized near the high endothelial venule conduits within the LN microenvironment may help to maintain a tolerant environment by regulating multiple interactions between naive T cells with APC. This finding may help to unmask the factors that promote the differentiation and activation of regulatory suppressor cells and allow the generation of extrathymic Treg for the control of immune responses.
Acknowledgments
We acknowledge the technical contributions of Minwei Mao, Dan Chen, Italas George, and Patricia Rebollo, and the helpful discussion with Dr. Gwendalyn Randolph.
Footnotes
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 National Institutes of Health Grants R01 AI41428, AI62765, and AI44929 (to J.S.B.).
2 Address correspondence and reprint requests to Dr. Jonathan S. Bromberg or Dr. Jordi C. Ochando, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1104, New York, NY 10029-6574. E-mail address: jon.bromberg{at}msnyuhealth.org or jordi.ochando{at}mssm.edu
3 Current address: Bristol-Myers Squibb, Princeton, NJ 08543.
4 Abbreviations used in this paper: Treg, T regulatory cell; LN, lymph node; DST, donor-specific transfusion; SI, stimulation index; TMN, total mononuclear cells.
Received for publication November 15, 2004. Accepted for publication March 22, 2005.
References
Cobbold, S. P., A. Jayasuriya, A. Nash, T. D. Prospero, H. Waldmann. 1984. Therapy with monoclonal antibodies by elimination of T-cell subsets in vivo. Nature 312: 548-551.
Isobe, M., H. Yagita, K. Okumura, A. Ihara. 1992. Specific acceptance of cardiac allograft after treatment with antibodies to ICAM-1 and LFA-1. Science 255: 1125-1127.
Chavin, K. D., L. Qin, J. Lin, H. Yagita, J. S. Bromberg. 1993. Combined anti-CD2 and anti-CD3 receptor monoclonal antibodies induce donor-specific tolerance in a cardiac transplant model. J. Immunol. 151: 7249-7259.[
Parker, D. C., D. L. Greiner, N. E. Phillips, M. C. Appel, A. W. Steele, F. H. Durie, R. J. Noelle, J. P. Mordes, A. A. Rossini. 1995. Survival of mouse pancreatic islet allografts in recipients treated with allogeneic small lymphocytes and antibody to CD40 ligand. Proc. Natl. Acad. Sci. USA 92: 9560-9564.
Shevach, E. M.. 2002. CD4+CD25+ suppressor T cells: more questions than answers. Nat. Rev. Immunol. 2: 389-400.
Belghith, M., J. A. Bluestone, S. Barriot, J. Megret, J. F. Bach, L. Chatenoud. 2003. TGF--dependent mechanisms mediate restoration of self-tolerance induced by antibodies to CD3 in overt autoimmune diabetes. Nat. Med. 9: 1202-1208.
Cobbold, S. P., R. Castejon, E. Adams, D. Zelenika, L. Graca, S. Humm, H. Waldmann. 2004. Induction of foxP3+ regulatory T cells in the periphery of T cell receptor transgenic mice tolerized to transplants. J. Immunol. 172: 6003-6010.
Waldmann, H., S. Cobbold. 2001. Regulating the immune response to transplants: a role for CD4+ regulatory cells?. Immunity 14: 399-406.
Wood, K. J., S. Sakaguchi. 2003. Regulatory T cells in transplantation tolerance. Nat. Rev. Immunol. 3: 199-210.(Jordi C. Ochando2,*, Adam)