A variety of delivery systems have been used to genetically modify vascular endothelial cells and smooth muscle cells (SMCs), but currently available systems suffer from either inefficient in vivo gene transfer, transient episomal vector expression, or significant immune responses and inflammation. In the present study, we evaluated an alternate vector system, recombinant adeno-associated virus (rAAV) for transduction of vascular cells in culture and in vivo. Primary cultures of rabbit, monkey, and human SMCs; macaque and human microvascular endothelial cells; and human umbilical vein endothelial cells were efficiently transduced at a dose of 100 to 1000 DNase-resistant particles per cell. rAAV-mediated transduction of the vasculature in vivo was observed after intraluminal gene delivery or after intra-adventitial injection in carotid arteries of atherosclerotic cynomolgus monkeys. Whether vector delivery was intraluminal or adventitial, transduction was observed in the adventitia, particularly within microvessels (vasa vasorum) but not in cells of the intima or media. Transduction of adventitial microvessels was enhanced by balloon injury 4 days before gene transfer. This was particularly true for adventitial delivery. We have previously shown that adventitial cell proliferation increases significantly 4 days after balloon injury (45%) in this animal model. Together, these data suggest that cell proliferation may enhance AAV transduction in vivo in the vasculature. AAV vectors exhibited a tropism in vivo for the microvascular endothelium at the doses used in the present study, which may provide the opportunity for targeting gene delivery. In summary, we have demonstrated the utility of rAAV vectors for ex vivo vascular cell gene delivery and present an initial experience with rAAV for in vivo vascular gene delivery. This alternate vector system may overcome some of the limitations hampering the development of gene therapy for vascular disorders.