The role of T cells in rheumatoid arthritis.

The most striking arguments in favor of a T cell dependent nature of RA are the strong association of the disease with selected class II HLA haplotypes (the "shared epitope" hypothesis) and the fact that, in experimental animal models such as adjuvant arthritis, the disease can be transferred by isolated T cell lines. It is true that T cell activation at the site of inflammation is not excessive. However, there is now unequivocal evidence for focal synthesis of IL-2 and IFN-gamma in the RA synovial membrane and one may realise that a limited but specific T cell activation may be sufficient to induce or perpetuate the immune process. This same argument may explain the lack of clear TCR restriction at the sites of inflammation. Until now, no antigen has been demonstrated to initiate and/or perpetuate RA. Different antigens though have been incriminated in the pathogenesis of RA, including cartilage antigens (collagen, proteoglycans, chondrocyte antigens), heat shock proteins or exogenous (viral/bacterial) antigens. Unless one can pick up the right antigen and clone the relevant T cells, it will be very hard to directly prove a T cell-dependent nature of the disease.