Hematopoietic Chimerism Following Non-Myeoloablative Stem Cell Transplantation Is Dependent on NK Cell Tolerance.

Abstract

NK cells are able to eliminate major mismatched hematopoietic cells and, in addition to T cells, represent a second barrier that prevents engraftment following transplantation over MHC barriers. Here, we studied the role of NK cells in the elimination of major mismatched hematopoietic cells in a Balb/c into B6 transplantation model after anti-CD40L (MR1) treatment. In this model, survival of donor cells was determined using an in-vivo cytotoxicity assay based on the infusion of differentially CFSE-labeled syngeneic and donor splenocytes. Likewise, the efficacy of treatment was determined directly by assessing the level of chimerism three months after transplantation. Four weeks after the infusion of 106 bone marrow cells (BMC), B6 mice rejected donor spleen cells within 3 hours. A combination with anti-CD40L treatment prevented this early elimination and these mice showed the same elimination kinetics as observed in untreated mice (97.0 ± 1.3% vs. 96.6 ± 1.3% in 3 days). These results indicate that anti-CD40L treatment prevents the induction of a memory T cell response after infusion of donor BMC. Nonetheless, elimination of donor cells was still present within 3 days, therefore we hypothesized that the elimination of donor cells was mediated by NK cells, rather than by T cells. A detailed analysis of the elimination kinetics in untreated B6 mice showed that the CFSE-labeled Balb/c splenocytes were gradually eliminated starting from the moment of infusion. Similar elimination kinetics were observed in T cell-deficient B6 nu/nu mice. In addition, in-vivo treatment with a depleting anti-NK cell antibody (PK136) prolonged the survival of donor splenocytes in both B6 and B6 nu/nu mice (54.7 ± 2.8% vs. 8.4 ± 5.9% in 2 days in B6 mice and 72.4 ± 7.2% vs. 19.4 ± 8.6% in 1 day in B6 nu/nu mice). A similarly prolonged survival of donor spleen cells was observed in NK cell-depleted mice that had received 106 BMC 4 weeks earlier in combination with anti-CD40L (77.4 ± 15.6% with NK cell depletion vs. 5.3 ± 0.6% without NK cell depletion in 2 days), while no effect was observed after in-vivo treatment with a depleting anti-CD8 antibody. Infusion of increasing numbers of Balb/c BMC (106, 107, 108) after treatment with anti-CD40L resulted in a dose-dependent prolongation of the survival of donor splenocytes, but up to 108 BMC were needed for complete non-responsiveness. This indicated that transplantation of 108 BMC resulted in tolerization of NK cells, which was also associated with stable chimerism (36.1 ± 14.0% of the GR-1+ fraction). In-vivo depletion of NK cells before transplantation allowed stable chimerism in mice treated with anti-CD40L and only 30 x 106 BMC (5/5 vs. 1/5 without NK cell depletion). These data demonstrate that: 1) The elimination of Balb/c donor splenocytes in untreated B6 recipient mice is mediated by NK cells. 2) In mice treated with donor BMC and anti-CD40L the elimination of donor splenocytes can be delayed by NK cell depletion or by increasing the initial dose of donor BMC. 3) NK cell tolerance over MHC barriers can be induced by transplantation of a high number of BMC (108) and results in sustained engraftment and chimerism. 4) Additional NK cell depletion allows sustained chimerism following transplantation of a lower number (30 x 106) of BMC. We conclude that the induction of NK cell tolerance is dependent on the dose of donor BMC injected. This may explain the high numbers of BMC required for engraftment over MHC barriers.