The objective of this study was to design a methodology of gene transfer into a resistance vascular bed and to show if such a method can be used to examine the physiological function of a given gene product in vivo. We developed such a method and validated it by defining the role in vivo of endothelial nitric oxide synthase (eNOS). In a constant flow perfused rat hindlimb, gene transfer to the vascular endothelium was accomplished by incubating a "first-generation" serotype 5, replication-deficient, adenoviral vector (1.2 X 10(9) plaque-forming units/ml) containing cDNA encoding either the eNOS or the beta-galactosidase (beta-Gal) gene in the hindlimb vasculature for 30 min. Five days after infection, immunohistochemical staining for eNOS localized recombinant gene expression to vascular endothelial cells and eNOS protein levels were increased fourfold (11.9 +/- 6.6 vs. 2.9 +/- 1.3 intensity units/microg protein, n = 4, p < 0.05). Perfusion pressures were measured at different flow rates (10-50 ml/min). In addition, basal and acetylcholine (ACh)-stimulated vascular resistance (VR) in phenylephrine (PE)-precontracted (100 microM) hindlimb was measured at constant flow. There were flow-dependent increases (p < 0.05) in perfusion pressure. Overexpression of eNOS shifted the pressure-flow curve downward and administration of N(G)-nitro-L-arginine methyl ester (L-NAME) shifted the curve upward. Compared with beta-Gal-transfected rats, PE-induced VR decreased (p < 0.05) in eNOS-transfected rats (100 +/- 27 vs. 164 +/- 49 mmHg, n = 5). Addition of 100 microM L-NAME increased (p < 0.05) PE-induced VR in both eNOS-transfected and control rats (145 +/- 50 and 232 +/- 38 mmHg, n = 5, p < 0.05), respectively, which was partially abolished by L-arginine pretreatment. ACh-induced vasorelaxation was increased 45% (p < 0.05) in eNOS-transfected hindlimbs. L-NAME decreased (p < 0.05) ACh-induced vasorelaxation by 58% in eNOS-transfected hindlimbs versus 25% in beta-Gal-transfected hindlimbs (p < 0.05). We used this gene transfer method to examine the physiological function of a gene product in vivo and showed that (1) the flow-pressure relationship in the hindlimb vascular bed is NO dependent and (2) the eNOS enzyme modulates NO-mediated vasorelaxation in the rat hindlimb resistance arteries in vivo.