- Email: [email protected]
Harnessing CD8+ Regulatory T Cells: Therapy for Type 1 Diabetes?
Immunity
Previews monocyte recruitment, transient cell ablation, and modulation of the monocyte subset prevalence. An important question though remains: arguably the mere increase in production rate and shortening of bone marrow residency will modify the monocytes; however, does the inflammatory condition have a defined qualitative impact on the blood monocyte compartment? Or do circulating cells remain naive in the chronically inflamed organism and undistinguishable from the steady state until they reach the inflamed tissue? These and other questions should spur efforts to extend the study of mouse models of chronic disease and inflammation to the blood, as—to finish with Goethe—‘‘Blood
is a quite special kind of juice’’ (Faust I, 1808).
Schulz, O., Jaensson, E., Persson, E.K., Liu, X., Worbs, T., Agace, W.W., and Pabst, O. (2009). J. Exp. Med. 206, 3101–3114.
REFERENCES
Serbina, N.V., Salazar-Mather, T.P., Biron, C.A., Kuziel, W.A., and Pamer, E.G. (2003). Immunity 19, 59–70.
Bogunovic, M., Ginhoux, F., Helft, J., Shang, L., Hashimoto, D., Greter, M., Liu, K., Jakubzick, C., Ingersoll, M.A., Leboeuf, M., et al. (2009). Immunity 31, 513–525.
Shortman, K., and Naik, S.H. (2007). Nat. Rev. Immunol. 7, 19–30.
Geissmann, F., Jung, S., and Littman, D.R. (2003). Immunity 19, 71–82.
Siddiqui, K.R.R., Laffont, S., and Powrie, F. (2010). Immunity 32, this issue, 557–567.
Liu, K., Victora, G.D., Schwickert, T.A., Guermonprez, P., Meredith, M.M., Yao, K., Chu, F.F., Randolph, G.J., Rudensky, A.Y., and Nussenzweig, M. (2009). Science 324, 392–397.
Varol, C., Vallon-Eberhard, A., Elinav, E., Aychek, T., Shapira, Y., Luche, H., Fehling, H.J., Hardt, W.D., Shakhar, G., and Jung, S. (2009a). Immunity 31, 502–512.
Nakano, H., Lin, K.L., Yanagita, M., Charbonneau, C., Cook, D.N., Kakiuchi, T., and Gunn, M.D. (2009). Nat. Immunol. 10, 394–402.
Varol, C., Yona, S., and Jung, S. (2009b). Immunol. Cell Biol. 87, 30–38.
Harnessing CD8+ Regulatory T Cells: Therapy for Type 1 Diabetes? Paola Zaccone1 and Anne Cooke1,* 1Department of Pathology, University of Cambridge, Tennis Court Rd, Cambridge CB2 1QP, UK *Correspondence: [email protected] DOI 10.1016/j.immuni.2010.04.009
In this issue of Immunity, Tsai et al. (2010) demonstrate that low-avidity autoantigen-specific regulatory CD8+ T cells can reverse ongoing autoimmune disease and provide insight into the mechanism by which this is achieved. Type 1 diabetes (T1D) is an autoimmune disease in which the pancreatic beta cells are selectively destroyed by the immune system. It is a disease of juvenile onset and affected individuals rely on exogenous insulin for their survival. An enormous amount of effort has been expended to develop strategies to prevent onset of this autoimmune condition as well as to cure established disease. One of the major goals in developing therapies for autoimmune diseases such as T1D has been to target autoreactive T cells and generate specific immune tolerance while sparing the ability to respond to exogenous antigens. Although antigen-specific tolerogenic approaches have been shown to prevent onset of autoimmune pathology in animal models, they have proved to be relatively ineffec-
tive in the treatment of ongoing disease. It had been hoped that although epitope and antigen spread inevitably occurred with ongoing tissue destruction, antigenspecific approaches might be successful through inducing ‘‘infectious’’ or ‘‘bystander’’ suppression. In terms of clinical application targeting antigen specific autoreactive cells has thus far proved disappointing. Current therapeutic approaches with efficacy on established autoimmune disease include the use of monoclonal antibodies such as anti-CD3 and antiCD20 (Chatenoud, 2010; Pescovitz et al., 2009). The availability of an antigenspecific therapy would represent a major advance. By using dextran-coated iron oxide nanoparticles bearing diabetes-related peptide MHC complexes (pMHC-NP),
504 Immunity 32, April 23, 2010 ª2010 Elsevier Inc.
Tsai et al. (2010) in this issue of Immunity have shown that this may be achievable. The composition of the nanoparticle is key to its efficacy. Different nanoparticles have previously been used to expand antigen-specific effector cells in vaccine design and to target tumors in cancer therapy. This study highlights the potential to use nanoparticles to downregulate immune responses and provides a potential therapeutic approach for the treatment of human autoimmune disease. The NOD mouse spontaneously develops type 1 diabetes and has been used not only to explore the etiology of disease development but also to develop therapeutic strategies (Kikutani and Makino, 1992). Studies in this model have shown that there is a need for both CD4+ T cells and CD8+ T cells in disease
Immunity
Previews Disease-relevant peptide
Low-avidity CD8+ T cell
pMHC-nanoparticle
Regulatory CD8+ T cell
Proliferation
Killing
Inhibition of proliferation
IFN-γ IFN-γ IDO
Autoreactive CD8+ T cell B cell Dendritic cell (Self-antigen loaded)
Autoreactive CD4+ T cell
IDO
Pancreatic β cell
Tolerogenic DC
Figure 1. Generation and Expansion of Autoantigen-Specific Regulatory CD8+ T Cells Disease-relevant peptide MHC nanoparticle complexes mediate in vivo expansion of low-avidity CD8+CD122+CXCR3+ T cell expressing high amounts of CD44 and CD62L (surface markers not depicted) in the NOD mouse. These memory-regulatory CD8+ T cells do not express Foxp3 and are negative for CCR7, CD69, and CD25, contain granzyme B, and secrete large amounts of IFN-g. Autoantigen-specific regulatory CD8+ T cells can kill antigen presenting cells (APCs), presenting self-peptide in the presence of costimulation. This results in an inhibition of autoreactive effector T cell proliferation. Perforin, but not FasFas-L interactions, plays a role in APC deletion. Pancreatic b cells would not be killed because they lack costimulatory molecule expression. The important mechanistic role of IFN-g and IDO, involved in diabetes prevention by these autoantigen-specific regulatory CD8+ T cells, will require further study. It is possible that IFN-g induces IDO in the b cell or in DCs, making them tolerogenic.
development (Phillips et al., 2009). Key T cell activation events appear to occur in the pancreatic draining lymph nodes (PLNs) in which dendritic cells (DCs) present b cell antigens to both T cell subsets (Turley et al., 2003). CD8+ regulatory T cells have had a checkered history, but more recent studies have highlighted the potential for this population to downregulate immune responses. In this landmark study, Tsai and colleagues have shown that nanoparticles coated with the Kd containing a mimotope peptide (NRP-V7) of an isletspecific glucose-6-phosphatase catalytic subunit-related protein (IGRP) peptide (IGRP206-214) can be used to expand lowavidity CD8+ T cells that not only inhibit the development of type 1 diabetes in NOD mice but also reverse established disease (Figure 1). The effect was not restricted to these NRP-V7-Kd-NPs given that comparable results were obtained with NPs coated with a Db-presenting mimotope of another islet antigen. These studies suggested that there were antigen-experienced pools of CD8+ T cells that were expanded by pMHC-NP treatment. The requirement for the presence of the autoantigenic epitope to induce the CD8+ T cells was nicely demonstrated
with gene-targeted NOD mice that failed to express the IGRP epitope and that could not be rescued from diabetes development by NRP-V7-Kd-NPs. Diabetes could, however, be reversed by using NPs coated with another isletrelated peptide-MHC conjugate. Analyses of mechanism revealed that there seemed to be an expansion of low-avidity CD8+CD44hiCD122+CCR7 CXCR3+CD62Lhi regulatory cells in the diabetes-protected mice (Figure 1). Apart from the expression of CD62L, these cells bear some similarity with an effector and memory CD8+ T cell population, identified by Guarda et al., that downregulated immune responses through selective killing of antigen-carrying dendritic cells (Guarda et al., 2007). Indeed, these lowavidity autoantigen-specific CD8+ T cells identified in this current study were also shown to target antigen-expressing DCs in the PLN. These low-avidity autoantigen-specific CD8+ regulatory T cells were able to suppress the proliferation of effector CD8+ T cells in vitro and in the PLN in vivo. Importantly, the response to any antigen being copresented with islet antigen by DCs was affected by these T cells, providing a clue to the in vivo efficacy of the pMHC-NPs at late stages of
disease development, given that in the PLN of a diabetic mouse, many islet antigens would be presented. There was no evidence of involvement of Foxp3expressing CD4+ or CD8+ T cells in this regulation (Figure 1). The Santamaria group generated TCR transgenics with either high-avidity CD8+ T cells or low-avidity CD8+ T cells specific for the IGRP peptide and showed that the low-avidity cells behaved exactly like the cells expanded by the NRP-V7-Kd-NPs. These cells were able to prevent diabetes induction in NOD.scid recipient mice by effector T cells from diabetic donors. It would seem likely that this was through an effect on autoantigen presentation to both CD4+ T cells as well as CD8+ T cells although this was not formally shown. With regard to molecular mechanism, the involvement of perforin, IFN-g, and IDO was identified by using blocking antibodies, inhibitors, and gene-targeted mice with both in vitro and in vivo approaches. Whereas the low-avidity regulatory CD8+ T cells were shown to express granzyme B and IFN-g, the source of IDO remains to be clarified. Work of others has clearly identified roles for IDO in immune tolerance (Mellor and Munn, 2004; Saxena et al., 2007). IFN-g has been shown to induce IDO in a range of cell types including dendritic cells and macrophages, but it is also induced in islets where it may serve to protect b cells from destruction (Sarkar et al., 2007). Any cell bearing IGRP or related peptides and expressing costimulatory molecules such as CD86 could be targeted by these CD8+ T cells. Thus, B cells pulsed with these peptides were eliminated in vivo by these autoantigen-specific regulatory CD8+ T cells. This is particularly interesting in the context of the current focus on B cell involvement in the pathogenesis of T1D and the equivalent efficacy of CD3 or CD20 directed therapy in this autoimmune condition. Could this approach be used to treat humans with T1D? This was addressed experimentally with NOD mice transgenically expressing HLA-A*0201 on a murine MHC class I deficient background with only A*0201-restricted CD8+ T cells in the periphery. Treatment of either prediabetic or diabetic HLA-A*0201-expressing mice with NPs coated with relevant islet antigen epitopes complexed with the transgene-encoded MHC molecule
Immunity 32, April 23, 2010 ª2010 Elsevier Inc. 505
Immunity
Previews prevented or reversed diabetes onset, respectively. This study strongly suggests that such a nanoparticle approach might have efficacy in T1D patients. Given that the authors found that there was recurrence of diabetes in 40%–50% of treated diabetic mice within 20 weeks of the last treatment, however, this might suggest that repeated treatment protocols might be needed. In conclusion, this study highlights a therapeutic strategy for the treatment of autoimmune pathology involving the recruitment and expansion of low-avidity autoantigen-specific regulatory CD8+ T cells.
REFERENCES Chatenoud, L. (2010). Nat. Rev. Endocrinol. 6, 149–157. Guarda, G., Hons, M., Soriano, S.F., Huang, A.Y., Polley, R., Martı´n-Fontecha, A., Stein, J.V., Germain, R.N., Lanzavecchia, A., and Sallusto, F. (2007). Nat. Immunol. 8, 743–752.
Study Group. (2009). N. Engl. J. Med. 361, 2143–2152. Phillips, J.M., Parish, N.M., Bland, C., Sawyer, Y., De La Pen˜a, H., and Cooke, A. (2009). Rev. Diabet. Stud. 6, 97–103. Sarkar, S.A., Wong, R., Hackl, S.I., Moua, O., Gill, R.G., Wiseman, A., Davidson, H.W., and Hutton, J.C. (2007). Diabetes 56, 72–79.
Kikutani, H., and Makino, S. (1992). Adv. Immunol. 51, 285–322.
Saxena, V., Ondr, J.K., Magnusen, A.F., Munn, D.H., and Katz, J.D. (2007). J. Immunol. 179, 5041–5053.
Mellor, A.L., and Munn, D.H. (2004). Nat. Rev. Immunol. 4, 762–774.
Tsai, S., Shameli, A., Yamanouchi, J., ClementeCasares, X., Wang, J., Serra, P., Yang, Y., Medarova, Z., Moore, A., and Santamaria, P. (2010). Immunity 32, this issue, 568–580.
Pescovitz, M.D., Greenbaum, C.J., Krause-Steinrauf, H., Becker, D.J., Gitelman, S.E., Goland, R., Gottlieb, P.A., Marks, J.B., McGee, P.F., Moran, A.M., et al. Type 1 Diabetes TrialNet Anti-CD20
506 Immunity 32, April 23, 2010 ª2010 Elsevier Inc.
Turley, S., Poirot, L., Hattori, M., Benoist, C., and Mathis, D. (2003). J. Exp. Med. 198, 1527–1537.
Previews monocyte recruitment, transient cell ablation, and modulation of the monocyte subset prevalence. An important question though remains: arguably the mere increase in production rate and shortening of bone marrow residency will modify the monocytes; however, does the inflammatory condition have a defined qualitative impact on the blood monocyte compartment? Or do circulating cells remain naive in the chronically inflamed organism and undistinguishable from the steady state until they reach the inflamed tissue? These and other questions should spur efforts to extend the study of mouse models of chronic disease and inflammation to the blood, as—to finish with Goethe—‘‘Blood
is a quite special kind of juice’’ (Faust I, 1808).
Schulz, O., Jaensson, E., Persson, E.K., Liu, X., Worbs, T., Agace, W.W., and Pabst, O. (2009). J. Exp. Med. 206, 3101–3114.
REFERENCES
Serbina, N.V., Salazar-Mather, T.P., Biron, C.A., Kuziel, W.A., and Pamer, E.G. (2003). Immunity 19, 59–70.
Bogunovic, M., Ginhoux, F., Helft, J., Shang, L., Hashimoto, D., Greter, M., Liu, K., Jakubzick, C., Ingersoll, M.A., Leboeuf, M., et al. (2009). Immunity 31, 513–525.
Shortman, K., and Naik, S.H. (2007). Nat. Rev. Immunol. 7, 19–30.
Geissmann, F., Jung, S., and Littman, D.R. (2003). Immunity 19, 71–82.
Siddiqui, K.R.R., Laffont, S., and Powrie, F. (2010). Immunity 32, this issue, 557–567.
Liu, K., Victora, G.D., Schwickert, T.A., Guermonprez, P., Meredith, M.M., Yao, K., Chu, F.F., Randolph, G.J., Rudensky, A.Y., and Nussenzweig, M. (2009). Science 324, 392–397.
Varol, C., Vallon-Eberhard, A., Elinav, E., Aychek, T., Shapira, Y., Luche, H., Fehling, H.J., Hardt, W.D., Shakhar, G., and Jung, S. (2009a). Immunity 31, 502–512.
Nakano, H., Lin, K.L., Yanagita, M., Charbonneau, C., Cook, D.N., Kakiuchi, T., and Gunn, M.D. (2009). Nat. Immunol. 10, 394–402.
Varol, C., Yona, S., and Jung, S. (2009b). Immunol. Cell Biol. 87, 30–38.
Harnessing CD8+ Regulatory T Cells: Therapy for Type 1 Diabetes? Paola Zaccone1 and Anne Cooke1,* 1Department of Pathology, University of Cambridge, Tennis Court Rd, Cambridge CB2 1QP, UK *Correspondence: [email protected] DOI 10.1016/j.immuni.2010.04.009
In this issue of Immunity, Tsai et al. (2010) demonstrate that low-avidity autoantigen-specific regulatory CD8+ T cells can reverse ongoing autoimmune disease and provide insight into the mechanism by which this is achieved. Type 1 diabetes (T1D) is an autoimmune disease in which the pancreatic beta cells are selectively destroyed by the immune system. It is a disease of juvenile onset and affected individuals rely on exogenous insulin for their survival. An enormous amount of effort has been expended to develop strategies to prevent onset of this autoimmune condition as well as to cure established disease. One of the major goals in developing therapies for autoimmune diseases such as T1D has been to target autoreactive T cells and generate specific immune tolerance while sparing the ability to respond to exogenous antigens. Although antigen-specific tolerogenic approaches have been shown to prevent onset of autoimmune pathology in animal models, they have proved to be relatively ineffec-
tive in the treatment of ongoing disease. It had been hoped that although epitope and antigen spread inevitably occurred with ongoing tissue destruction, antigenspecific approaches might be successful through inducing ‘‘infectious’’ or ‘‘bystander’’ suppression. In terms of clinical application targeting antigen specific autoreactive cells has thus far proved disappointing. Current therapeutic approaches with efficacy on established autoimmune disease include the use of monoclonal antibodies such as anti-CD3 and antiCD20 (Chatenoud, 2010; Pescovitz et al., 2009). The availability of an antigenspecific therapy would represent a major advance. By using dextran-coated iron oxide nanoparticles bearing diabetes-related peptide MHC complexes (pMHC-NP),
504 Immunity 32, April 23, 2010 ª2010 Elsevier Inc.
Tsai et al. (2010) in this issue of Immunity have shown that this may be achievable. The composition of the nanoparticle is key to its efficacy. Different nanoparticles have previously been used to expand antigen-specific effector cells in vaccine design and to target tumors in cancer therapy. This study highlights the potential to use nanoparticles to downregulate immune responses and provides a potential therapeutic approach for the treatment of human autoimmune disease. The NOD mouse spontaneously develops type 1 diabetes and has been used not only to explore the etiology of disease development but also to develop therapeutic strategies (Kikutani and Makino, 1992). Studies in this model have shown that there is a need for both CD4+ T cells and CD8+ T cells in disease
Immunity
Previews Disease-relevant peptide
Low-avidity CD8+ T cell
pMHC-nanoparticle
Regulatory CD8+ T cell
Proliferation
Killing
Inhibition of proliferation
IFN-γ IFN-γ IDO
Autoreactive CD8+ T cell B cell Dendritic cell (Self-antigen loaded)
Autoreactive CD4+ T cell
IDO
Pancreatic β cell
Tolerogenic DC
Figure 1. Generation and Expansion of Autoantigen-Specific Regulatory CD8+ T Cells Disease-relevant peptide MHC nanoparticle complexes mediate in vivo expansion of low-avidity CD8+CD122+CXCR3+ T cell expressing high amounts of CD44 and CD62L (surface markers not depicted) in the NOD mouse. These memory-regulatory CD8+ T cells do not express Foxp3 and are negative for CCR7, CD69, and CD25, contain granzyme B, and secrete large amounts of IFN-g. Autoantigen-specific regulatory CD8+ T cells can kill antigen presenting cells (APCs), presenting self-peptide in the presence of costimulation. This results in an inhibition of autoreactive effector T cell proliferation. Perforin, but not FasFas-L interactions, plays a role in APC deletion. Pancreatic b cells would not be killed because they lack costimulatory molecule expression. The important mechanistic role of IFN-g and IDO, involved in diabetes prevention by these autoantigen-specific regulatory CD8+ T cells, will require further study. It is possible that IFN-g induces IDO in the b cell or in DCs, making them tolerogenic.
development (Phillips et al., 2009). Key T cell activation events appear to occur in the pancreatic draining lymph nodes (PLNs) in which dendritic cells (DCs) present b cell antigens to both T cell subsets (Turley et al., 2003). CD8+ regulatory T cells have had a checkered history, but more recent studies have highlighted the potential for this population to downregulate immune responses. In this landmark study, Tsai and colleagues have shown that nanoparticles coated with the Kd containing a mimotope peptide (NRP-V7) of an isletspecific glucose-6-phosphatase catalytic subunit-related protein (IGRP) peptide (IGRP206-214) can be used to expand lowavidity CD8+ T cells that not only inhibit the development of type 1 diabetes in NOD mice but also reverse established disease (Figure 1). The effect was not restricted to these NRP-V7-Kd-NPs given that comparable results were obtained with NPs coated with a Db-presenting mimotope of another islet antigen. These studies suggested that there were antigen-experienced pools of CD8+ T cells that were expanded by pMHC-NP treatment. The requirement for the presence of the autoantigenic epitope to induce the CD8+ T cells was nicely demonstrated
with gene-targeted NOD mice that failed to express the IGRP epitope and that could not be rescued from diabetes development by NRP-V7-Kd-NPs. Diabetes could, however, be reversed by using NPs coated with another isletrelated peptide-MHC conjugate. Analyses of mechanism revealed that there seemed to be an expansion of low-avidity CD8+CD44hiCD122+CCR7 CXCR3+CD62Lhi regulatory cells in the diabetes-protected mice (Figure 1). Apart from the expression of CD62L, these cells bear some similarity with an effector and memory CD8+ T cell population, identified by Guarda et al., that downregulated immune responses through selective killing of antigen-carrying dendritic cells (Guarda et al., 2007). Indeed, these lowavidity autoantigen-specific CD8+ T cells identified in this current study were also shown to target antigen-expressing DCs in the PLN. These low-avidity autoantigen-specific CD8+ regulatory T cells were able to suppress the proliferation of effector CD8+ T cells in vitro and in the PLN in vivo. Importantly, the response to any antigen being copresented with islet antigen by DCs was affected by these T cells, providing a clue to the in vivo efficacy of the pMHC-NPs at late stages of
disease development, given that in the PLN of a diabetic mouse, many islet antigens would be presented. There was no evidence of involvement of Foxp3expressing CD4+ or CD8+ T cells in this regulation (Figure 1). The Santamaria group generated TCR transgenics with either high-avidity CD8+ T cells or low-avidity CD8+ T cells specific for the IGRP peptide and showed that the low-avidity cells behaved exactly like the cells expanded by the NRP-V7-Kd-NPs. These cells were able to prevent diabetes induction in NOD.scid recipient mice by effector T cells from diabetic donors. It would seem likely that this was through an effect on autoantigen presentation to both CD4+ T cells as well as CD8+ T cells although this was not formally shown. With regard to molecular mechanism, the involvement of perforin, IFN-g, and IDO was identified by using blocking antibodies, inhibitors, and gene-targeted mice with both in vitro and in vivo approaches. Whereas the low-avidity regulatory CD8+ T cells were shown to express granzyme B and IFN-g, the source of IDO remains to be clarified. Work of others has clearly identified roles for IDO in immune tolerance (Mellor and Munn, 2004; Saxena et al., 2007). IFN-g has been shown to induce IDO in a range of cell types including dendritic cells and macrophages, but it is also induced in islets where it may serve to protect b cells from destruction (Sarkar et al., 2007). Any cell bearing IGRP or related peptides and expressing costimulatory molecules such as CD86 could be targeted by these CD8+ T cells. Thus, B cells pulsed with these peptides were eliminated in vivo by these autoantigen-specific regulatory CD8+ T cells. This is particularly interesting in the context of the current focus on B cell involvement in the pathogenesis of T1D and the equivalent efficacy of CD3 or CD20 directed therapy in this autoimmune condition. Could this approach be used to treat humans with T1D? This was addressed experimentally with NOD mice transgenically expressing HLA-A*0201 on a murine MHC class I deficient background with only A*0201-restricted CD8+ T cells in the periphery. Treatment of either prediabetic or diabetic HLA-A*0201-expressing mice with NPs coated with relevant islet antigen epitopes complexed with the transgene-encoded MHC molecule
Immunity 32, April 23, 2010 ª2010 Elsevier Inc. 505
Immunity
Previews prevented or reversed diabetes onset, respectively. This study strongly suggests that such a nanoparticle approach might have efficacy in T1D patients. Given that the authors found that there was recurrence of diabetes in 40%–50% of treated diabetic mice within 20 weeks of the last treatment, however, this might suggest that repeated treatment protocols might be needed. In conclusion, this study highlights a therapeutic strategy for the treatment of autoimmune pathology involving the recruitment and expansion of low-avidity autoantigen-specific regulatory CD8+ T cells.
REFERENCES Chatenoud, L. (2010). Nat. Rev. Endocrinol. 6, 149–157. Guarda, G., Hons, M., Soriano, S.F., Huang, A.Y., Polley, R., Martı´n-Fontecha, A., Stein, J.V., Germain, R.N., Lanzavecchia, A., and Sallusto, F. (2007). Nat. Immunol. 8, 743–752.
Study Group. (2009). N. Engl. J. Med. 361, 2143–2152. Phillips, J.M., Parish, N.M., Bland, C., Sawyer, Y., De La Pen˜a, H., and Cooke, A. (2009). Rev. Diabet. Stud. 6, 97–103. Sarkar, S.A., Wong, R., Hackl, S.I., Moua, O., Gill, R.G., Wiseman, A., Davidson, H.W., and Hutton, J.C. (2007). Diabetes 56, 72–79.
Kikutani, H., and Makino, S. (1992). Adv. Immunol. 51, 285–322.
Saxena, V., Ondr, J.K., Magnusen, A.F., Munn, D.H., and Katz, J.D. (2007). J. Immunol. 179, 5041–5053.
Mellor, A.L., and Munn, D.H. (2004). Nat. Rev. Immunol. 4, 762–774.
Tsai, S., Shameli, A., Yamanouchi, J., ClementeCasares, X., Wang, J., Serra, P., Yang, Y., Medarova, Z., Moore, A., and Santamaria, P. (2010). Immunity 32, this issue, 568–580.
Pescovitz, M.D., Greenbaum, C.J., Krause-Steinrauf, H., Becker, D.J., Gitelman, S.E., Goland, R., Gottlieb, P.A., Marks, J.B., McGee, P.F., Moran, A.M., et al. Type 1 Diabetes TrialNet Anti-CD20
506 Immunity 32, April 23, 2010 ª2010 Elsevier Inc.
Turley, S., Poirot, L., Hattori, M., Benoist, C., and Mathis, D. (2003). J. Exp. Med. 198, 1527–1537.