Tolerance to discordant xenografts. I. Sharing of human natural antibody determinants on miniature swine bone marrow cells and endothelial cells.

Xenotransplantation could potentially overcome the organ shortage that currently limits the field of transplantation. Because of their breeding characteristics as well as their size and physiologic similarities to humans, we have chosen miniature swine as possible xenograft donors, and are currently attempting to develop a means of using mixed xenogeneic chimerism as an approach to tolerance induction in swine-to-primate species combinations. One major barrier to organ grafting from pig to man is the presence in human serum of preformed natural antibodies (NAb) reacting with antigens expressed on porcine endothelial cells and causing hyperacute rejection. Previous experiments performed in our laboratory have shown that both humoral and cellular tolerance can be induced in a concordant xenogeneic species combination (rat-->mouse) using donor bone marrow infusion following conditioning with a nonmyeloablative regimen. Induction of chimerism in these animals was associated with a marked reduction in the level of IgM natural antibodies that recognize rat bone marrow cells. A similar approach could also lead to humoral and cellular tolerance induction in the swine-->human species combination, permitting transplantation of vascularized organs from the swine donor. To determine the potential of bone marrow transplantation to induce a state of "natural antibody tolerance," it was essential to determine whether or not all human NAb target antigens expressed on swine EC are also expressed on cells derived from swine bone marrow. We have addressed this question by evaluating the ability of various swine bone marrow-derived cells to absorb human IgM and IgG NAb that bind to swine EC. Our results demonstrate that swine bone marrow cells and their progeny can absorb almost all IgM NAb that bind to swine EC, as detected by flow cytometric and ELISA assays. Specificity of absorption was demonstrated, as total serum IgM levels declined only minimally after absorption on swine BMC and to an extent comparable to that observed following absorption with human cells, which did not deplete swine EC-binding NAb. Human IgG binding to swine EC was also completely absorbed by swine BMC. These results suggest that a state of "NAb tolerance" could be induced by successful swine marrow engraftment in man. Furthermore, swine PBL, platelets, and EC were able to absorb most IgM NAb that bound to swine BMC, suggesting that absorption using antigen from any of these tissues might facilitate marrow engraftment, and hence tolerance induction, in this species combination.

Induction of tolerance to renal allografts across single-haplotype MHC disparities in miniature swine.

We have previously demonstrated that a 12-day course of cyclosporine A (CsA) leads to the induction of tolerance to renal allografts in 100% of recipients selectively mismatched at class I for both haplotypes, and in 71% of recipients selectively mismatched at class II for both haplotypes, but in 0% of recipients mismatched for two haplotypes at both class I and class II. We have postulated that the mechanism by which tolerance is induced may therefore require matching for either class I or class II antigens. One might predict from this hypothesis that tolerance would also be induced in donor-recipient combinations sharing one full haplotype (e.g., AC-->AD), which mimics the clinically relevant transplant combination of parent to offspring. We have therefore investigated the effects of the CsA regimen on renal transplants in this combination. Without immunosuppression, such kidney allografts were uniformly rejected (n = 12; 10.6 +/- 2.4 days). In contrast, a course of CsA (10-13 mg/kg/day) during the first 12 postoperative days induced long-term acceptance of the allograft in 67% (4/6) of recipients. Some acceptor animals also showed specific unresponsiveness to donor antigens as measured by in vitro assays and by failure to develop anti-donor antibodies. Tolerance was confirmed in four of these animals by failure to reject a second transplant SLA-matched to the first kidney donor without additional immunosuppression. These results suggest the feasibility of inducing specific tolerance across a single-haplotype mismatch in the majority of the cases, which could have clinical implications for living-related transplants.

Tolerance to class I-disparate renal allografts in miniature swine. Maintenance of tolerance despite induction of specific antidonor CTL responses.

Miniature swine that become tolerant to renal allografts across an MHC class I barrier following a short course of cyclosporine are unresponsive to donor class I antigens in cell-mediated lymphocytotoxicity. However, skin grafts bearing donor class I plus third-party class II antigens are promptly rejected, and the animals then develop marked cell-mediated lymphocytoxic reactivity to donor class I antigens in vitro, but do not reject the kidney transplants. We show here that CTL generation is directed toward the same donor class I antigens as are expressed by the kidney donor, and is not the result of recognition in vitro of the tolerated class I antigen plus peptides of minor antigens shared between the skin graft donor and the stimulator/target cells. We also show that detection by CTLs of peptides expressed by skin but not by kidney is also not a sufficient explantation of the results, since the survival of skin grafts from the kidney donor is also prolonged, even after precursor CTL can be detected in vitro. The data are most consistent with suppression in vivo in tolerant animals of the helper pathways necessary for activation of precursor CTLs. Differences in patterns of cytokine expression by graft infiltrating cells may provide a mechanism for local suppression of help in this model. Finally, we have examined antibody production after sensitizing by skin grafts in long-term tolerant animals and have found that anti-donor class I antibodies are not produced, even though the same animals produce both anti-class II and anti-third-party class I antibodies.