Intrathymic and extrathymic tolerance in bone marrow chimeras.

Parent-->F1 bone marrow (BM) chimeras provide a useful model for studying self tolerance induction. When prepared with supralethal irradiation (1300 cGy) and conditioned with anti-T cell antibodies, parent-->F1 BM chimeras are devoid of host BM-derived cells; host H-2 expression is apparent in both the intrathymic and extrathymic environments but is limited to non BM-derived cells. When parent-->F1 chimeras are injected with T cells from normal parental strain mice, the expression of host H-2 antigens on nonprofessional APC might be expected to induce tolerance through induction of clonal anergy. In practice, this does not occur. Instead, a small proportion of the injected T cells is induced to proliferate and differentiate into effector cells. Tolerance is not seen. Similarly, tolerance is not apparent when thymectomized parent-->F1 chimeras are given parental strain thymus grafts. These findings suggest that the expression of host H-2 antigens in the post-thymic environment of chimeras is not intrinsically tolerogenic for mature T cells or recent thymic emigrants. Interestingly, post-thymic tolerance does occur when parental strain T cells differentiate in the endogenous thymus of chimeras. Thus, when mature CD8+ cells are prepared from thymus vs lymph nodes (LN) of parent-->F1 chimeras, tolerance to host class I antigens is more marked in LN than thymus; this applies to cytotoxic T lymphocyte (CTL) precursors, generated by limiting dilution analysis. It would appear therefore that many of the host-reactive CTL precursors generated in the thymus of chimeras undergo tolerance induction (deletion or irreversible inactivation) in the post-thymic environment. We suggest that such tolerance is a reflection of a covert form of tolerance induced in the thymus: intrathymic contact with host antigens on thymic epithelial cells (TEC) in chimeras does not delete typical CTL precursors, but these cells are rendered "semi-tolerant". When cultured in vitro in the presence of lymphokines, the cells are able to recover and differentiate into CTL. In vivo, however, the cells recognize antigen in the periphery in the relative absence of lymphokines and the cells die. Although host class I expression on TEC in chimeras deletes only a small proportion of CTL precursors, contact with TEC induces strong tolerance of CD8+ cells in terms of helper-independent proliferative responses in vitro and induction of lethal graft-versus-host disease in vivo. We postulate that these latter responses are controlled by high-affinity T cells, whereas typical CTL generated in LDA are predominantly low-affinity cells.(ABSTRACT TRUNCATED AT 400 WORDS)