This study evaluated the efficacy and mode of action of rapamycin (RPM) in a model of accelerated (24-hr) rejection of LBNF1 cardiac allografts in specifically sensitized LEW rats. RPM treatment (0.25 mg/kg/day i.p.) between the day of sensitizing skin grafts (day -7) and subsequent heart (day 0) transplantation (Tx), abrogated fulminant rejection and prolonged cardiac allograft survival to 46 +/- 22 days (mean +/- SD, P < 0.0001). The delayed introduction of RPM until day -2 or day -1 was equally effective, whereas treatment initiated after cardiac Tx was ineffectual. Untreated accelerated rejection was associated with strong production of circulating IgM, whereas an IgG alloantibody response was not detected until after rejection was complete. RPM therapy (day -7 to -1) diminished this systemic IgM response and prevented the switch from IgM to IgG alloantibody production. Immunohistologic evaluation at 24 hr after cardiac Tx showed that compared with untreated hosts RPM treatment largely abolished intragraft cellularity, and was associated with decreased mononuclear and endothelial cell activation. Specifically, Ia and ICAM-1 upregulation was abolished, and no cells elaborating IL-2 or IFN-gamma were detected. In addition, RPM treatment prevented intragraft production of the proinflammatory cytokines IL-1 beta, IL-6, and IL-8. The effects of RPM therapy on recipient cellular responses were evaluated in vitro by mixed lymphocyte reaction. Surprisingly, the donor-specific proliferative response of cells from RPM-treated hosts at 1 or 7 days after Tx was markedly increased, compared with cells from rejecting, untreated controls, and bioassay of IL-2 within supernatants of MLR cultures showed comparable levels of IL-2 in both groups. The effects of RPM upon adhesion properties of lymph node lymphocytes were also tested in an in vitro binding assay. The binding of naive cells to sections of cardiac allografts collected from RPM-treated hosts at 24 hr post-Tx was decreased compared with that in untreated recipients. Interestingly, the binding of mononuclear cells to high endothelial venules of peripheral lymph nodes in RPM-treated hosts remained relatively high. Thus, treatment with RPM prevents and/or erases the sensitization, which otherwise leads to accelerated allograft rejection. Abrogation of allograft injury by RPM was associated with profound and long-lasting depression of host IgM and IgG alloantibody responses in the circulation, and selective downregulation of host cellular immunity and endothelial activation at the graft site. In contrast, antigen alloreactivity and endothelial adhesivity in peripheral lymphoid tissues were spared, indicating novel and potent selective effects of RPM therapy in allograft recipients.