We hypothesized that physiological increments in plasma homocysteine after low-dose oral methionine or dietary animal protein induce vascular endothelial dysfunction and that there is a graded, inverse relationship between homocysteine concentration and endothelial function. We studied 18 healthy volunteers aged 18 to 59 years. Brachial artery flow-mediated and glyceryltrinitrate-induced dilatation were measured after 1) oral L-methionine (10, 25, and 100 mg/kg), 2) dietary animal protein (lean chicken 551+/-30 g, comprising 3.2+/-0.2 g methionine), and 3) methionine-free amino acid mix (100 mg/kg). Methionine (10, 25, and 100 mg/kg) induced a dose-related increase in homocysteine (9.4+/-1.3 to 12.2+/-2.1, 17. 6+/-2.6, and 26.1+/-4.2 micromol/L, respectively; P<0.001) and a reduction in flow-mediated dilatation (4.1+/-0.8 to 2.1+/-0.8, 0. 3+/-0.8, and -0.7+/-0.8%, respectively; P<0.001) at 4 hours. Compared with usual meal, animal protein increased plasma homocysteine (9.6+/-0.8 to 11.2+/-0.9 micromol/L, P=0.005) and reduced flow-mediated dilatation (4.5+/-0.7% to 0.9+/-0.6%, P=0.003). Methionine-free amino acid mix did not induce any changes. Glyceryltrinitrate-induced dilatation was unchanged throughout. In this study, small physiological increments in plasma homocysteine after low-dose methionine and dietary animal protein induced vascular endothelial dysfunction. We propose that protein intake-induced increments in plasma homocysteine may have deleterious effects on vascular function and contribute to the development and progression of atherosclerosis.