The nonobese diabetic (NOD) mouse has been developed as a model for insulin-dependent diabetes. One gene required for the development of diabetes is associated with the major histocompatibility complex. This gene possibly could be linked to class II genes, which show a unique pattern in NOD mice. To evaluate the role of the I-A class II antigen expressed in NOD mice, we studied the effect of anti-I-A monoclonal antibodies on disease onset in vivo. Long-term treatment with anti-class II IgG2a antibodies specific for NOD I-A antigen prevented the spontaneous development of diabetes, as opposed to control antibodies shown not to react with NOD I-A antigen. Anti-class II antibodies apparently elicited active immune suppression, requiring a fully immunocompetent host, rather than passive blockade of class II antigen. Treatment with anti-class II antibody effectively prevented the adoptive transfer of diabetes produced by splenocytes from diabetic NOD mice into newborn mice but failed to prevent adoptive transfer into irradiated adult NOD recipients. Direct evidence for the induction of suppressor cells was obtained from the passive transfer of spleen cells from anti-class II antibody-treated NOD donors. The injection of anti-class II antibody-treated spleen cells collected from NOD donors prevented the development of diabetes, which normally follows transfer of diabetogenic spleen cells into irradiated 8-week-old male NOD recipients. Depletion experiments indicate that CD41 cells are responsible for anti-class II-induced protection transferred by spleen cells.

The nonobese diabetic (NOD) mouse is a relevant model for studying human insulin-dependent diabetes mellitus (IDDM). The selective destruction of insulin-secreting cells in this model is subsequent to an autoimmune reaction directed towards the beta cells inside the islets of Langerhans of the pancreas. Given the key role played by T cells in the development of IDDM, we investigated a model of IDDM prevention in NOD mice by administration of a monoclonal antibody to the alpha/beta dimer of the T cell receptor for antigen. Our data provide evidence that aiming at the T cell receptor protects against both spontaneous and cyclophosphamide-induced diabetes in the NOD mouse. Interestingly, potential clinical application is suggested by the efficient and durable reversal of recent onset diabetes in mice treated with anti-alpha/beta monoclonal antibody within 1 week following the clinical discovery of IDDM.

SUMMARYAccumulating evidence suggests that Th1 T cells play a pivotal role in the development of autoimmune diabetes. Conversely, promoting a Th2 response inhibits disease progression. However, it has not been determined whether Th2 cells are regulatory T cells that fail at the time of diabetes development in naive non-diabetic NOD mice. Therefore, in order to evaluate cytokine secretion by spleen and islet infiltrating T cells in NOD mice at different stages of the autoimmune process, we developed an ELISPOT assay that detects IL-2, IL-4, and interferon-gamma (IFN-g) secretion in vitro at the singlecell level. We showed that, whatever the age considered, IFN-g is predominantly secreted, and that no IL-4-secreting cells are detected in the islets of male and female NOD mice. Spleen cells from 8-weekold female NOD mice, which include regulatory suppressor T cells, do not secrete IL-4, either upon presentation of islet cell antigens in vitro, or after transfer in vivo, but do secrete IFN-g. IFN-g secretion by T cells from diabetic mice results from CD4 but not CD8 T cells in transfer experiments into NOD/ severe combined immunodeficient (SCID) recipients. These results suggest that (i) detection of regulatory CD4 T cells in NOD mice is not paralleled by a Th2 response; (ii) b cell destruction does not depend on a switch from a Th2 to a Th1-type response; and (iii) CD8 T cells do not participate in induction of diabetes by secreting IFN-g.

The non-obese diabetic (NOD) mouse spontaneously develops a T cell-mediated autoimmune disease, sharing many features with human insulin-dependent diabetes mellitus (IDDM), leading to insulin-secreting beta cell destruction. The role of CD4+ T cells has been evidenced at two levels. First, CD4+ T cells from diabetic animals are required to transfer diabetes to non-diabetic recipients in conjunction with CD8+ effector T cells. Second, suppressive CD4+ T cells have been characterized in non-diabetic NOD mice. T cells with different functions can thus share the CD4+ phenotype. Since CD4+ T cells can be divided into at least two subgroups on the basis of CD45 isoform expression, we evaluated the distribution of CD4+ T cells expressing the CD45RA isoform on NOD mouse thymocytes and peripheral T cells. The percentage of CD45RA+ cells was dramatically increased among the most mature CD3bright thymocytes and among CD4+ T cells in lymph nodes of the NOD mouse as compared with control strains. This increase was related to the development of insulitis. Interestingly, the CD45RA isoform was expressed on most CD4+ T cells invading the islets. In vivo treatment with an anti-CD45RA mAb prevented the development of insulitis and spontaneous diabetes in female animals but not the transfer of diabetes by T cells collected from diabetic NOD donors. These results indicate that anti-CD45RA mAb is only effective if given before the full commitment of effector T cells to the destruction of islet beta cells. Thus CD4+CD45RA+ T cells play a key role in early activation steps of anti-islet immunity.