We have developed a model of transforming growth factor (TGF)beta1 gene transfer into mouse vascularized cardiac allografts to study the use of gene transfer as an immunosuppressive therapy in transplantation. Donor hearts were perfused with either DNA-liposome complexes or adenoviral vectors that encode the active form of human TGFbeta1. DNA-liposome mediated transfection prolonged allograft survival in approximately two-thirds of transplant recipients, while adenoviral delivery of TGFbeta1 was not protective. Protective TGFbeta1 gene transfer was associated with reduced Th1 responses and an inhibition of the alloantibody isotype switch. The protective effects of TGFbeta1 gene transfer were overridden by exogenous interleukin-12 administration. Interestingly, alloreactive CD4+ and CD8+ cells exhibited distinct sensitivities to TGFbeta1 gene transfer: CD4+ Th1 function was abrogated by this modality, although CD8+ Th1 function was not. Transient depletion of recipient CD8+ cells markedly prolonged the survival of grafts transfected with either DNA-liposome complexes or adenoviral vectors. Transgene expression persisted for at least 60 days, and Th1 responses were not detectable until CD8+ T cells repopulated the periphery. However, long-term transfected allografts appeared to exhibit exacerbated fibrosis and neointimal development. These manifestations of chronic rejection were absent in long-term transfected isografts, suggesting that long-term expression of active TGFbeta1 alone is not sufficient to induce fibrosis of the grafts. Collectively, these data illustrate the utility of immunosuppressive gene therapy as a treatment for transplantation when combined with additional conditioning regimens. Further, they illustrate that alloreactive CD4+ and CD8+ cells may be differentially influenced by cytokine manipulation strategies.