A large number of studies have documented the effects of CsA on immunologic tolerance. Initially, these studies concentrated on the ability of CsA to induce tolerance to allografts, particularly in rats. However, it has become apparent that CsA can sometimes block the induction of tolerance, and provoke or aggravate specific autoimmune diseases in several species (Table 1). CsA can inhibit at least three processes thought to contribute to tolerance (Table 2). In the thymus, CsA has been reported by some investigators to block negative selection, but surprisingly, large numbers of undeleted or" forbidden" T-cells are rarely observed in the periphery. In mature T-cells (extrathymic), CsA can block the induction of anergy which occurs after immunization with superantigens, and in vitro this drug can block anergy induction in Th1 clones which occurs when antigenic peptides are presented by metabolically inactive (treated with a fixative) antigen-presenting cells. However, CsA can also enhance the deletion of peripheral superantigen-reactive T-cells, when the drug is administered at high doses. Thus, responses to superantigens can be either enhanced or inhibited depending on the protocol of CsA administration. Whether or not these phenomena apply to conventional antigens has not been determined. Numerous studies demonstrate that CsA treatment can either enhance or depress suppressor T-cell function in various experimental models. The interpretation of data on suppressor cells is complicated by the current poor understanding of the function of the cells. In CsA-induced syngeneic (or autologous) GVHD, CsA may inhibit both T-cell anergy induction and the generation and/or function of suppressor cells. CsA can alter Th1/Th2 antagonism, such that DTH responses are enhanced. Enhanced Th1 activity could explain the deterioration that CsA sometimes provokes in autoimmune diseases where DTH is important, eg, collagen-induced arthritis and EAE. CsA can prevent the development of neonatal tolerance. The latter effect may account, at least in part, for the induction of organ-specific autoimmunity observed after treating mice with CsA during the neonatal period. However, in all the situations mentioned the effects of CsA on tolerance cannot be easily explained by a single mechanism. There is redundancy in the immune tolerance mechanism, such that clonal deletion, clonal anergy, and suppressor cells may all be capable of maintaining tolerance to similar antigens. Thus, it is likely that CsA must inhibit more than one of these mechanisms for autoimmunity to occur.(ABSTRACT TRUNCATED AT 400 WORDS)