The ultimate goal of prophylactic gene therapy is to confer permanent protection against ischemia. Although gene therapy with inducible nitric oxide synthase (iNOS) is known to protect against myocardial infarction at 3 days and up to 2 months, the long-term effects on myocardial ischemic injury and function are unknown. To address this issue, we created a recombinant adeno-associated viral vector carrying the iNOS gene (rAAV/iNOS), which enables long-lasting transgene expression. The ability of rAAV/iNOS to direct the expression of functional iNOS protein was confirmed in COS-7 cells before in vivo gene transfer. Mice received injections in the anterior LV wall of rAAV/LacZ or rAAV/iNOS; 1 year later, they underwent a 30-min coronary occlusion (O) and 4 h of reperfusion (R). iNOS gene transfer resulted in elevated iNOS protein expression (+3-fold vs. the LacZ group, n = 6; P < 0.05) and iNOS activity (+4.4-fold vs. the LacZ group, n = 6; P < 0.05) 1 year later. Infarct size (% of risk region) was dramatically reduced at 1 year after iNOS gene transfer (13.5 ± 2.2%, n = 12, vs. 41.7 ± 2.9%, n = 10, in the LacZ group; P < 0.05). The infarct-sparing effect of iNOS gene therapy at 1 year was as powerful as that observed 24 h after ischemic preconditioning (six 4-min O/4-min R cycles) (19.3 ± 2.3%, n = 11; P < 0.05). Importantly, compared with the LacZ group (n = 11), iNOS gene transfer (n = 10) had no effect on LV dimensions or function for up to 1 year (at 1 year: FS 34.5 ± 2.0 vs. 34.6 ± 2.6%, EF 57.0 ± 2.0 vs. 59.7 ± 2.9%, LVEDD 4.3 ± 0.1 vs. 4.2 ± 0.2 mm, LVESD 2.8 ± 0.1 vs. 2.9 ± 0.2 mm) (echocardiography). These data demonstrate, for the first time, that rAAV-mediated iNOS gene transfer affords long-term, probably permanent (1 year), cardioprotection without adverse functional consequences, providing a strong rationale for further preclinical testing of prophylactic gene therapy.