Incubation of murine bone marrow and splenocytes with the dipeptide methyl ester, L-leucyl-L-leucine methyl ester (Leu-Leu-OMe), which results in the selective depletion of cytotoxic T cells and their precursors, natural killer cells, and monocytes, completely protected 30 recipients of fully allogeneic donor grafts from lethal graft-versus- host disease (GVHD). These results were comparable with those obtained in 30 recipients of anti-Thy 1.2 plus complement (C')-treated donor marrow. However, in contrast to antibody- and C'-dependent T-cell depletion, which reduces the level of donor cell engraftment in our model system, we did not observe such effects using Leu-Leu-OMe marrow pretreatment. As compared with the 24 H-2 typed recipients of anti-Thy 1.2 + C'-treated donor grafts, the 29 H-2 typed recipients of Leu-Leu- OMe-treated donor grafts had significantly (P less than .001) higher percentages of donor cells (mean = 93% v 74%) and significantly (P less than .001) lower percentages of host cells (mean = 6% v 15%) posttransplantation. In vitro limiting dilution assay (LDA) was performed to assess the comparative efficacy of cytolytic T-lymphocyte (CTL) precursor depletion by Leu-Leu-OMe or anti-Thy 1.2 + C' pretreatment. We observed greater levels of CTL precursor depletion in Leu-Leu-OMe treated as compared with anti-Thy 1.2 + C'-treated bone marrow plus spleen cells (BMS) obtained from nontransplanted mice. This suggests that the in vivo results cannot simply be attributed to a less efficacious functional inactivation of cytolytic T-cell precursors by Leu-Leu-OMe treatment as compared with anti-Thy 1.2 + C' treatment. Immunoreconstitution was similar in recipients of Leu-Leu-OMe-treated grafts and anti-Thy 1.2 + C'-treated grafts 100 days posttransplant. In our opinion, Leu-Leu-OMe marrow pretreatment deserves further investigation as a methodology to achieve GVHD prevention without significantly reducing the propensity toward host cell repopulation.