Recipient natural killer cell allorecognition of passenger donor lymphocytes and its effect on adaptive alloimmunity after transplantation.

BACKGROUND: Memory T cells are known to reside in peripheral non-lymphoid tissue, but how their presence within solid organ allografts affects transplant outcomes is not known. We have previously described how graft-versus-host (GVH) allorecognition by passenger CD4 T cells within MHC class II-mismatched bm12 heart grafts provokes antinuclear humoral autoimmunity in C57BL/6 recipient mice. Here we aimed to examine how such GVH recognition affects the alloresponse to allografts with greater mismatching. METHODS: A MHC class I and II mismatched murine model of cardiac transplantation was developed (bm12.Kd.IE to C57BL/6). After transplantation, cellular and humoral responses against mismatched antigens were measured with ELISPOT and ELISA, and the effect of GVH recognition assessed by depletion of donor CD4 T cells before graft procurement. Antinuclear autoantibody development was assessedwith HeP-2 indirect immunofluorescence. The role of recipient natural killer (NK) cells was examined by depletion with anti-NK1.1 antibody. FINDINGS: Bm12.Kd.IE heart grafts provoked strong germinal centre alloantibody and autoantibody responses in C57BL/6 recipients and developed allograft vasculopathy. By contrast, heart grafts from CD4 T-cell-depleted donors developed only minimal vasculopathy, and the alloantibody responses were weaker, without observable autoantibody. Bm12.Kd.IE CD4 T cells survived long term when transferred to RAG hosts suggesting that avoidance of killing by host NK cells might be essential for autoantibody development. In support, in a model of alloantibody-mediated vasculopathy, depletion of NK cells from a C57BL/6 recipient of a BALB/c heart graft resulted in the development of autoantibody, amplification of the alloantibody response, and rapid allograft rejection. This amplification was abrogated by depletion of donor CD4 T cells. INTERPRETATION: Although host adaptive immunity is expected to bring about destruction of passenger lymphocytes within heart allografts, this process occurs too slowly to prevent GVH-mediated augmentation of the alloresponse to the graft. Rather, rapid killing of donor lymphocytes by host alloreactive NK cells is essential. Passenger CD4 lymphocytes might therefore contribute to chronic rejection in recipients receiving an allograft that does not prompt innate NK cell recognition. FUNDING: Wellcome Trust Clinical Research Training Fellowship.