To determine if endogenous local levels of nitric oxide (NO) modulate atherogenesis, we studied the effect of inhibiting NO with A fO -nitro-L-arginine methyl ester (L-NAME) on early neointima formation in cholesterol-fed rabbits. Male rabbits were fed for 5 weeks with a 0.5% cholesterol diet alone or treated in addition during the last 4 weeks with L-NAME (12 mg/kg per day SC) via osmotic minipump. Endothelial cell function was assessed in isolated aortic rings by vascular reactivity and levels of cyclic GMP. In L-NAME-treated rabbits there was inhibition of endotheliumdependent relaxations to acetylcholine and the calcium ionophore A23187 as well as impaired cyclic GMP accumulation in response to acetylcholine. Neointima formation in the ascending thoracic aorta was assessed by determining media and intima cross-sectional areas with computerized image analysis. N itric oxide (NO), a potent physiological vasodilator that accounts for the biological activities of endothelium-derived relaxing factor, 1 is synthesized in endothelial cells from the terminal guanidino nitrogen of L-arginine by the constitutive calciumcalmodulin-NADPH-dependent enzyme NO synthase (NOS). 2 In humans and animals with hypercholesterolemia-induced atherosclerosis, endothelium-dependent vasodilation is impaired, 35 suggesting reduced NO synthesis or action. NO has antiproliferative actions on vascular cells, 6 -7 supporting the hypothesis that the endothelial cell dysfunction observed in hypercholesterolemia could contribute to the initiation and progression of the atherosclerotic neointima. When administered in the diet of cholesterol-fed rabbits, the NO precursor L-arginine limits development of aortic atherosclerosis and improves endothelial cell function. 8 - 9 We recently reported 10 a technique for the chronic administration to rabbits of the NOS inhibitor iV G -nitro-L-arginine methyl ester ( L -N A M E ) and demonstrated its persistent systemic effects on endothelial cells and Received November 22, 1993; revision accepted February 11, 1994.From the Robert Dawson Evans Department of Clinical Research, Vascular Biology Unit, Boston University School of Medicine, Boston, Mass.Presented in part at the 66th Scientific Sessions of the American Heart Association, Atlanta, Ga, November 6-11, 1993, and published in abstract form in Circulation. 1993;88(pt 2):I-366.Correspondence to Antonio J. Cayatte, MD, Vascular Biology Unit, E-401, Department of Medicine, Boston University School of Medicine, Boston, MA 02118.Compared with rabbits that consumed the cholesterol diet alone, L-NAME-treated rabbits had significant increases in lesion area (0.29+0.04 versus 0.15±0.03 mm 2 ) and in lesion/ media ratio (0.06±0.01 versus 0.03±0.01). Plasma levels of cholesterol and fluorescent lipid peroxide products were unchanged, suggesting no difference in cholesterol metabolism or oxidation. Because arterial blood pressure was not altered by L-NAME treatment, the increased atherogenesis could not be attributed to an increase in blood pressure. mi...