Despite improvement in posttransplant immunosuppressive regimens, GHVD persists as a major complication after hematopoietic cell transplantations, especially when the donor is not an HLA-identical family member. Attempts to prevent GVHD by removing all T cells from the graft have been complicated by increased risks of graft rejection and by impaired immune reconstitution. GVHD is known to be initiated by the relatively small subset of donor T cells that recognize recipient alloantigens. This subset of T cells cannot easily by distinguished from the majority of T cells that do not recognize recipient alloantigens. Transgenic T cells that express the herpes simplex virus enzyme thymidine kinase (TK) are killed when they proliferate during exposure to ganciclovir, while nonproliferating cells remain unaffected. Marrow transplant experiments have shown that TK-transgenic donor T cells have a strikingly reduced ability to cause GVHD in mice treated with ganciclovir. Although ganciclovir clearly depleted donor T cells that proliferated after recognizing recipient alloantigens, these studies did not determine whether ganciclovir also depleted the TK-transgenic donor T cells that did not recognize recipient alloantigens. Experiments in this project will measure the proliferation of T cells that cause GVHD as compared to the proliferation of T cells that do not cause GVHD. T cells that recognize recipient alloantigens are expected to proliferate more rapidly than those that do not recognize recipient alloantigens. Information from these experiments will be used to develop regimens of ganciclovir administration, which achieve optimal depletion of TK- transgenic T cells that recognize recipient alloantigens with optimal sparing of cells that do not recognize recipient alloantigens. The long term goals of this project are to test the hypothesis that the TK- transgenic T cells surviving an optimized regimen of ganciclovir 1) will be sufficient to prevent allogeneic marrow graft rejection in sublethally irradiated recipients and 2) will be sufficient to reconstitute immune function in histoincompatible thymus-deficient recipients. Results of these preclinical studies will help to assess the potential risks and benefits of using similar approaches to prevent GVHD in human marrow transplantation.