To determine whether a disparity at a single mill genetic loci are sufficient to generate GVHD in mice, we focused on well-known genetic alleleic differences at the mill gene loci, H3 and H4. For H3 congenic GVHD studies, C57BL/10 (H2b) mice were used as recipients of mill-disparate B10.LP-H3b donor cells. For H4 congenic GVHD studies, C57BL/10 were used as recipients for mill-disparate B10.129 (21M)-H4b. To overcome the low frequency of mill-reactive CTLs in naive mice, multiple immunizations of the donor strains with host lymphohematopoietic cells were used. Peripheral blood cells from immunized mice were shown to have potent CTL activity against their respective host-type stimulator cells when analyzed 1 week prior to obtaining donor splenocytes for GVHD induction. Lethally irradiated C57BL/6 recipients of either 50 x 106 donor B10.LP-H3b or B10.129 (21M)-H4b splenocytes did not develop acute or chronic GVHD as assessed by monitoring the animals for survival, weight loss, splenic flow cytometry, and histological examination of skin, liver, colon, and lung in long-term survivors. Engraftment was documented in long-term chimeras in both strain combinations by using the post-BMT cells as alloantigen targets for cloned CTL lines specific for donor and not host-type mill antigens (H3b or H4b). On day 6 post-BMT, donor antihost CTL activity could not be detected in the spleen, although third-party responses were intact. These results suggest a rapid downregulation or disappearance of mill antigen-reactive CTL after BMT. These data have implications for the use of in vitro assays to predict GVHD risk in recipients of mill loci-disparate donor grafts.