A Murine Model of Bone Marrow Failure Mediated by Disparity in Minor Histocompatibility Antigens.

Destruction of hematopoietic cells in aplastic anemia and other hematologic diseases is mediated in most patients by effector cells of the immune system. We have developed a mouse model of immune-mediated bone marrow failure which employs parental lymph node cell infusion into F1 recipients, resulting in bone marrow hypercellularity, pancytopenia, and death from infection and bleeding, without other organ involvement by graft-versus-host disease (Bloom, ML et al. Exp. Hematol. 32:1163, 2004; Chen, J et al. Blood104:1671, 2004). Because major histocompatibility antigens are not the targets in acquired aplastic anemia, we now have developed a model of murine marrow failure based on disparity of minor histocompatibility antigens. Lymph node cells from C57BL/6 mice were infused into sublethally-irradiated, MHC-matched, C.B10-H2b/LilMce recipients. Animals developed severe pancytopenia and marrow hypoplasia within two-three weeks. CD8+ T lymphocytes were expanded in the blood and infiltrated bone marrow, becoming detectable at day 7 and reaching peak levels at days 10–12. There were no inflammatory responses observed in the skin, intestines, or other visceral organs by gross or microscopic pathological examination. In our experiments we observed a time-dependent expansion followed by contraction of CD8+ T cells specific for a minor histocompatibility antigen H60 as measured by flowcytometry using an H60-specific tetramer. H60 is an antigen peptide derived from a glycoprotein, a known ligand for stimulatory NKG2D receptors, and is immunodominant over other minor antigens in stem cell transplantation. The proportion of H60-specific CD8+ T cells was strongly negatively correlated with peripheral blood white cell, neutrophil, and platelet counts. Isolated H60-specific T-cells from bone marrow of affected animals induced apoptosis in vitro of normal C57BL/6 bone marrow cells in co-culture. The degree of apoptosis was further increased by addition of CD4 T-cells from same affected donors, suggesting a helper lymphocyte effect. The role of H60-specific T cells was demonstrated in further transplant experiments. Infusion of 5 × 106 C57BL/6 lymph node cells that had been depleted of H60-specific T cells was unable to induce marrow failure in C.B10 recipients, while the same number of cells with the addition of 20–90 × 103 H60-specific CD8+ T cells led to thrombocytopenia and leucopenia in recipients. H60-specific T cells thus appear to be key effectors, responsible at least for the initiation of marrow cell destruction. Mice could be treated with cyclosporine at 50 mg/g/day for five days, if treatments were begun at the time of lymph node cell infusion. Immunosuppressive therapy abolished H60-specific CD8+ T cell expansion and attenuated the development of peripheral pancytopenia, effectively rescuing animals. In conclusion, we demonstrate in this model of immune-mediated bone marrow failure that a single clone of peptide-specific T-cells is capable of efficient marrow destruction and the production of aplastic anemia.