The current model of lymphocyte activation suggests that at least two types of signals are required for optimal stimulation of helper T cell activity. First, T cell antigen receptor (TcR) recognition of a complex of MHC and antigenic peptide on the surface of antigen presenting cells (APC) activates multiple early biochemical signalling pathways. However, important late events such as lymphokine secretion and clonal expansion do not occur in the absence of second 'costimulatory signal' provided by APC. It has been hypothesized that restriction of costimulatory activity to APC would protect the tissue cells of a given organ from an autoimmune attack, as T cells would not be activated by antigen-expressing tissue cells which do not also express costimulatory activity. Furthermore, T cells which are exposed to antigen in the absence of costimulatory activity may become clonally inactivated, a mechanism by which tolerance to tissue-specific antigens would be maintained. Recent investigations demonstrated that binding of an APC surface protein, B7, to a T cell surface ligand, CD28, results in signalling events which synergize with TcR signals to activate transcription of multiple lymphokine genes and stimulate T cell proliferation. Interruption of the CD28-B7 interaction prevents T cell activation by APC capable of presenting relevant antigens. Thus, it appears that CD28 transduces a costimulatory signal upon interaction with B7. Recognition of specific costimulatory signalling molecules allows us to design a strategy to test predictions of the two signal model of T cell activation in vivo. Although previous models suggested that tissue- specific expression of exogenous antigens would elicit autoimmune responses, transgenic expression of exogenous antigens in a tissue- specific manner has resulted in peripheral tolerance. The two signal model predicts that an autoimmune response would occur only if both antigen and costimulatory activity were expressed by the same tissue cells. The ability to express a costimulatory ligand in a tissue- specific manner would allow us to test this prediction. We propose to construct a transgenic model which expresses the costimulatory ligand, B7, under control of the tissue-specific insulin promoter. We have initiated collaborative experiments with the Hanahan laboratory, which has used the rat insulin promoter to direct transgenic pancreas-specific expression of the SV40 T antigen, and has analyzed T antigen-expressing mice for tolerance versus autoimmunity against T antigen-expressing self tissues. Expression of B7 on Beta-cells may result in a histologically identifiable immune response against Beta-cell antigens presented in the context of endogenous MHC molecules. However, either MHC or antigen expression may not be sufficient to elicit an immune response even in the presence of B7. We plan to breed B7 transgenic mice with mice expressing SV40 T antigen to determine whether co-expression of B7 and T antigen results in an immune response to T antigen rather than the previously described tolerance. We may also be able to test the clonal anergy model in vivo by transplanting B7-expressing SV40 T antigen-transformed pancreatic islet cell tumors into tolerant SV40 transgenics and looking for evidence of graft rejection. If the experiment described confirm the prediction that tissue-specific expression of costimulatory activity results in autoimmunity, we can then construct other models of organ-specific autoimmune disease. The availability of tissue-specific promoters which allow expression of exogenous molecules on muscle, thyroid, and glial cells, make it possible to construct animal models of immune-mediated human diseases such as polymyositis, thyroiditis, and multiple sclerosis. In the future, synovial-specific promoters may allow construction of a model of inflammatory arthritis as well.
Showing the most recent 10 out of 111 publications
