Granzyme B (GZB) has been implicated as an effector mechanism in regulatory T cells (T(reg)) suppression. In a model of T(reg)-dependent graft tolerance, it is shown that GZB- deficient mice are unable to establish long-term tolerance. Moreover, mice overexpressing the inhibitor of GZB, serine protease inhibitor 6, are also resistant to tolerization to alloantigen. Graft survival was shorter in bone marrow-mixed chimeras reconstituted with GZB-deficient T(reg) as compared with wild-type T(reg). Whereas there was no difference in graft survival in mixed chimeras reconstituted with wild-type, perforin-deficient, or Fas ligand-deficient T(reg). Finally, data also show that if alloreactive effectors cannot express FoxP3 and be induced to convert in the presence of competent T(reg), then graft tolerance is lost. Our data are the first in vivo data to implicate GZB expression by T(reg) in sustaining long-lived graft survival.

A protocol consisting of a single donor-specific transfusion (DST) plus a brief course of anti-CD154 monoclonal antibody (anti-CD40 ligand mAb) induces permanent islet allograft survival in chemically diabetic mice, but its efficacy in mice with autoimmune diabetes is unknown. Confirming a previous report, we first observed that treatment of young female NOD mice with anti-CD154 mAb reduced the frequency of diabetes through 1 year of age to 43%, compared with 73% in untreated controls. We also confirmed that spontaneously diabetic NOD mice transplanted with syngeneic (NOD-Prkdc(scid)/Prkdc(scid)) or allogeneic (BALB/c) islets rapidly reject their grafts. Graft survival was not prolonged, however, by pretreatment with either anti-CD154 mAb alone or anti-CD154 mAb plus DST. In addition, allograft rejection in NOD mice was not restricted to islet grafts. Anti-CD154 mAb plus DST treatment failed to prolong skin allograft survival in nondiabetic male NOD mice. The inability to induce transplantation tolerance in NOD (H2g7) mice was associated with non-major histocompatibility complex (MHC) genes. Treatment with DST and anti-CD154 mAb prolonged skin allograft survival in both C57BL/6 (H2b) and C57BL/6.NOD-H2g7 mice, but it was ineffective in NOD, NOD.SWR-H2q, and NOR (H2g7) mice. Mitogen-stimulated interleukin-1beta production by antigen-presenting cells was greater in strains susceptible to tolerance induction than in the strains resistant to tolerance induction. The results suggest the existence of a general defect in tolerance mechanisms in NOD mice. This genetic defect involves defective antigen-presenting cell maturation, leads to spontaneous autoimmune diabetes in the presence of the H2g7 MHC, and precludes the induction of transplantation tolerance irrespective of MHC haplotype. Promising islet transplantation methods based on overcoming the alloimmune response by interference with costimulation may require modification or amplification for use in the setting of autoimmune diabetes.

Treatment of C57BL/6 mice with one transfusion of BALB/c spleen cells and anti-CD154 (anti-CD40-ligand) antibody permits BALB/c islet grafts to survive indefinitely and BALB/c skin grafts to survive for approximately 50 d without further intervention. The protocol induces long-term allograft survival, but the mechanism is unknown. We now report: (a) addition of thymectomy to the protocol permitted skin allografts to survive for > 100 d, suggesting that graft rejection in euthymic mice results from thymic export of alloreactive T cells. (b) Clonal deletion is not the mechanism of underlying long-term graft survival, as recipient thymectomized mice were immunocompetent and harbor alloreactive T cells. (c) Induction of skin allograft acceptance initially depended on the presence of IFN-gamma, CTLA4, and CD4(+) T cells. Addition of anti-CTLA4 or anti-IFN-gamma mAb to the protocol was associated with prompt graft rejection, whereas anti-IL-4 mAb had no effect. The role of IFN-gamma was confirmed using knockout mice. (d) Graft survival was associated with the absence of IFN-gamma in the graft. (e) Long-term graft maintenance required the continued presence of CD4(+) T cells. The results suggest that, with modification, our short-term protocol may yield a procedure for the induction of long-term graft survival without prolonged immunosuppression.

Antibodies to the ligand for CD40 (CD154) have been shown to exert profound effects on the development of cell-mediated immune responses in mice. The present study shows that an antibody to human CD154 (hCD40L) inhibits in vivo Tetanus toxoid (TT) specific secondary antibody responses in hu-PBL-scid mice, as well as the expansion of xenoreactive human T cells in the scid mice. A possible cause for the reduced expansion of xenoreactive, human T cells, was the decreased expression of murine B7.1 and B7.2 caused by the administration of anti-hCD40L. Therefore, it may be that defective maturation of murine antigen-presenting cells impeded the priming and expansion of human xenoreactive T cells.

Combined treatment with allogeneic small lymphocytes or T-depleted small lymphocytes plus a blocking antibody to CD40 ligand (CD40L) permitted indefinite pancreatic islet allograft survival in 37 of 40 recipients that differed from islet donors at major and minor histocompatibility loci. The effect of the allogeneic small lymphocytes was donor antigen-specific. Neither treatment alone was as effective as combined treatment, although anti-CD40L by itself allowed indefinite islet allograft survival in 40% of recipients. Our interpretation is that small lymphocytes expressing donor antigens in the absence of appropriate costimulatory signals are tolerogenic for alloreactive host cells. Anti-CD40L antibody may prevent host T cells from inducing costimulatory signals in donor lymphocytes or islet grafts.