Characterization of cytotoxic cells in mice rendered neonatally tolerant of MHC alloantigens: evidence for repertoire modification.

A common prediction of clonal deletion/inactivation hypotheses is that cells with high avidity for tolerogen are preferentially eliminated, with low avidity cells being most likely to escape the tolerance induction mechanism. Thus it would be expected that the tolerogen-specific cells in tolerant mice would have a different repertoire than those in normal mice. To find evidence in favor of this prediction, neonatal B10.A mice were rendered tolerant to B10 by the injection of 15 X 10(6) (B10.A X B10)F1 spleen and bone marrow cells, and tolerance was assessed by the acceptance of B10 skin grafts for greater than 50 days. Mice rendered tolerant in this manner contain severely reduced (to less than 10% of normal) but detectable numbers of tolerogen-specific cytotoxic cell precursors that can be activated in the presence, but not absence, of exogenous interleukin 2. Spleen cells from the tolerant animals were compared with those of normal B10.A mice with respect to the expression of differentiation markers on the surface of B10-specific cytotoxic cells and their precursors, and the relative strength of the anti-B10 response toward Kb and Db as a measure of the repertoire of the cytotoxic cell populations. The T cell nature of the tolerogen-specific cytotoxic cells in both normal and tolerant mice was confirmed by their susceptibility to lysis by anti-Thy-1 or Lyt-2 antibody and complement. More importantly, cold target inhibition experiments showed that cytotoxic T cells from tolerant mice were inhibited to a greater degree by B10.A(2R) (KkDb) cold targets than B10.A(5R) (KbDd), suggesting that the response was preferentially directed at the D end of H-2, in direct contrast to normal B10.A spleen cells, which show a preferential response against Kb. Measurement of the frequency of anti-Kb and anti-Db cytotoxic T cell precursors in the spleens of normal and tolerant mice confirmed the differential specificities seen in the cold target experiments. The data suggest that neonatal tolerance induction results in repertoire modification of the anti-tolerogen response rather than a uniform decrease in anti-tolerogen reactivity. Possible mechanisms to explain the alteration in the repertoire of tolerant mice are discussed.