The seminal observation made 30 years ago that T cells do not discriminate between native and denatured proteins, whereas B cells generally do, can now be explained by the fact that T cells never see antigens in their native conformation and that intact proteins cannot associate simultaneously with MHC molecules and the TCR. This difference in the ability to recognize antigen based on conformational specificity appears to be a consequence of the fact that the T cell sees antigen not free in solution, but on the surface of an APC in association with MHC molecules. The metabolic events that protein antigens undergo within APC, prior to their presentation in an appropriately processed form to T cells, are called antigen processing. The end-product of antigen processing for CD4+ T cells is a relatively short peptide fragment bound to class II MHC molecules on the surface of an APC that can be recognized by the TCR on the T cells. Because this event is difficult to monitor directly, antigen processing can only be assayed in conjunction with the temporally distal event of T-cell activation, manifested ultimately as proliferative responses or lymphokine secretion. In addition to occupancy of the TCR by the peptide/class II complex, several other antigen-nonspecific receptor-ligand interactions between APC and T cells are required for optimal T-cell activation. M phi, B cells, and LC/DC comprise the principal APC for CD4+ T cells. M phi and B cells have been studied extensively with respect to their antigen processing and presenting capacities. Only recently, however, have such capacities been investigated in LC and DC; these studies now indicate freshly isolated LC (but not cultured LC and DC) to possess efficient antigen processing capabilities. In this respect, LC have been proposed to represent evolving (or "maturing") forms of DC: Freshly isolated LC (which retain morphologic and functional properties of epidermal LC in situ) are the equivalent of tissue forms of DC, whereas cultured LC resemble lymphoid or circulating DC. Cultured LC and DC appear to be the sole effective APC for inducing primary T-cell responses in vitro. Possibly underlying this property is the ability of cultured LC and DC (but not M phi, B cells, or freshly isolated LC) to induce formation of T-cell clusters during the course of such responses. The capacity of accessory cells to function as APC varies depending upon the type of APC and T cells examined.(ABSTRACT TRUNCATED AT 400 WORDS)