Are primary alloresponses truly primary?

Proliferative T cell responses against major histocompatibility complex (MHC) incompatible stimulator cells in the mixed lymphocyte reaction are conventionally regarded as primary. However, it is generally accepted that the recognition of allogeneic MHC products results from a cross-reaction by self-MHC-restricted cells. These two assumptions were tested by examining the contribution of previously primed and naive T cells to 'primary' alloresponses. Peripheral blood T cells were separated into LFA-3+, memory, and LFA-3-, naive, populations by fluorescence-activated cell sorting. In contrast, to recall antigen responses to Candida albicans which were almost entirely confined to the LFA-3+, memory, population, the proliferative response to MHC incompatible stimulator cells, including HLA-DR-expressing mouse L cell transfectants, was equally distributed between the two T cell subsets in 5 day assays. Furthermore, limiting dilution analysis showed that the frequency of alloreactive T cells did not differ significantly between the two populations. The kinetics of proliferation in the two populations differed but were consistent with their naive and memory phenotype, in that after 3 days of culture the LFA-3+ cells proliferated more strongly to MHC alloantigens. These results show that a substantial proportion of 'primary' alloresponses are contributed by previously primed cells. In addition, the evidence for the cross-reactive hypothesis is supported and extended from the clonal to the population level.