ONE MORE STEP TOWARD TRANSPLANTATION TOLERANCE

Tolerance in a Concordant Nonhuman Primate Model. Transplantation 1999; 68: 1708.

Bartholomew AM, Powelson J, Sachs DH, Bailin M, Boskovic S, Colvin R, Hong HZ, Johnson M, Kimikawa M, LeGuern A, Meehan S, Sablinski T, Wee SL, and Cosimi AB.

The article by Bartholomew et al. is the most recent update on the mechanisms required to achieve xenograft tolerance and is based on many years of efforts by the Boston group to achieve allograft tolerance ⁽¹⁾.

Using a complex conditioning regime consisting of antithymocyte globulin, non-lethal total body irradiation, thymic irradiation, a splenectomy followed by simultaneous bone marrow and renal transplants, and a 1-month course of cyclosporin A, mixed lymphohematopoietic chimerism and renal allograft tolerance has been achieved. Although this form of therapy is not currently clinically applicable in human allotransplantation, it has provided major insights into the mechanisms of allograft rejection, and it highlights the extent to which the immune system needs to be manipulated to achieve long term transplant survival. The potential benefits of this system are greatly reducing the long term mortality and morbidity associated with immunosuppressive therapy and reducing the cost to the healthcare system.

Using the established regime for achieving allograft tolerance, Bartholomew et al. pose the question of whether tolerance can be achieved using the same preconditioning methods in a concordant primate xenograft model. The answer clearly is no. The study immediately dispels several assumptions regarding xenotransplantation between concordant species combinations. First, concordant species combinations, which by definition do not have preformed xenoreactive antibodies, will be met with a cell-mediated form of rejection similar to allograft rejection. Second, the currently available immunosuppressive agents will prevent cell-mediated concordant xenograft rejection and prevent the emergence of specific anti-graft antibodies by inhibiting T-cell activation.

What then are the differences between allografts and concordant xenografts that permit tolerance induction between allografts but not xenografts? In allografts, the difference between donor and recipient resides largely in MHC allelic variation between the immune system of the recipient and the MHC status of the donor organ. The ensuing rejection response is mediated by antigen specific T-cells directed toward MHC antigens. Furthermore, in highly sensitized allograft recipients, the preformed antibodies are most commonly directed toward MHC antigens. The number of differences encountered after xenotransplants is much greater and extends beyond the MHC locus.

Transplanting organs between species has identified an ever expanding list of incompatibilities that may explain some of the differences between allograft and xenograft rejection. These differences result from structural variation between species of molecules, such as cytokines, cytokine receptors, and adhesion molecules, which play an integral role in the rejection response ⁽²⁾. These molecular incompatibilities, in addition to influencing the evolution of an immune response, may also add to the large pool of additional antigens encountered after transplantation across species.

Work by Auchincloss ⁽³⁾ and others has recently highlighted the relative importance of direct pathways of antigen recognition compared with indirect antigen recognition in xenotransplantation, and this may be the key to understanding the mechanisms of concordant xenograft rejection. Bartholomew et al. have demonstrated that the addition of deoxyspergualin (DSG) to the standard conditioning regime for the first 2 weeks significantly extended transplant survival. DSG is a potent immunosuppressive agent with only limited use in allotransplantation. Unlike cyclosporin A, which acts only on T-cells, DSG inhibits T-cells, monocytes/macrophages, and antibody generation ⁽⁴⁾. The function of DSG as an inhibitor of antigen presentation and monocyte activation may explain the prolongation in organ transplant survival observed in this model. The rejection of the transplanted kidney after the application of a donor-specific skin transplant lends support to the argument that the antigen-presenting cells in the skin initiate T-cell reactivity and the emergence of anti-graft antibody. This hypothesis could be confirmed by irradiation of skin before the transplant or an additional course of DSG after the skin transplant. Because of the logistic constraints of organ transplantation between primates, however, these experiments would be a large undertaking.

One remaining puzzle is the mechanism of the concordant xenograft rejection response and the cellular repertoire of the infiltrating cells. Organ transplant rejection is associated with the emergence of anti-donor antibody and the loss of donor-specific hyporesponsiveness, but the relative contribution of each to organ failure is yet to be defined. This is an area where the availability of reagents is limiting, so that detailed histological phenotyping of the infiltrate is difficult to perform. If anti-graft antibody is the prime initiator of graft damage, is the complement system contributing to cell damage or alternatively are natural killer (NK) cells and monocytes attracted to the organ by bound immunoglobulin? NK cells may also recognize xenogeneic tissue in the absence of antibody providing an additional mechanism of tissue damage. Studies performed in our laboratory and by Iverardi involving perfusion of hearts with xenogeneic NK cells demonstrated that small numbers of NK cells can initiate significant tissue damage ⁽⁵⁾. Furthermore, a histological examination of the perfused hearts revealed very few cells, suggesting that either only a few cells are required to initiate organ failure or, more likely, that the cells reside transiently on the endothelium where they damage the cell membrane and then recirculate, leaving the scene of the crime. With the limitations of reagents one can only speculate about immune deviation toward Th1 or Th2 type responses in this model.

The studies by Kirk et al. ⁽⁶⁾, using monoclonal antibody directed toward CD154 to prolong primate allograft survival without long term immunosuppressive therapy, provide a simple, well tolerated alternative to the conditioning regime used by Bartholomew et al. It would be exciting to test these reagents in concordant xenograft models, but, for the reasons outlined above, what may induce tolerance in MHC mismatched allografts may not extend xenograft survival.

The study by Bartholomew et al. opens the way to understanding the mechanisms of concordant xenograft rejection but raises as many questions as it answers. It is hoped that by continuing these resource-intensive studies, strategies for achieving long term tolerance between concordant species combinations will be identified that may also be applicable to human allotransplantation. With the current international decline in the availability of cadaveric organs and the increase in the number of living related and, more recently, unrelated organs transplanted, the achievement of transplant tolerance or even doubling the survival time of transplanted organs would have a major impact on the transplant community.

David J. Goodman

Anthony J. F. d'Apice

Immunology Research Centre; St Vincent's Hospital Melbourne; Australia 3065