Macrophages infiltrating the allografts are heterogeneous, consisting of proinflammatory (M1 cells) as well as antiinflammatory and fibrogenic phenotypes (M2 cells); they affect transplantation outcomes via diverse mechanisms. Here we found that macrophage polarization into M1 and M2 subsets was critically dependent on tumor necrosis factor receptor-associated factor 6 (TRAF6) and mammalian target of rapamycin (mTOR), respectively. In a heart transplant model we showed that macrophage-specific deletion of TRAF6 (LysM Traf6 ) or mTOR (LysM Mtor ) did not affect acute allograft rejection. However, treatment of LysM Mtor recipients with CTLA4-Ig induced long-term allograft survival (>100 days) without histological signs of chronic rejection, whereas the similarly treated LysM Traf6 recipients developed severe transplant vasculopathy (chronic rejection). The presentation of chronic rejection in CTLA4-Ig-treated LysM Traf6 mice was similar to that of CTLA4-Ig-treated wild-type B6 recipients. Mechanistically, we found that the graft-infiltrating macrophages in LysM Mtor recipients expressed high levels of PD-L1, and that PD-L1 blockade readily induced rejection of otherwise survival grafts in the LysM Mtor recipients. Our findings demonstrate that targeting mTOR-dependent M2 cells is critical for preventing chronic allograft rejection, and that graft survival under such conditions is dependent on the PD-1/PD-L1 coinhibitory pathway.