AIMS/HYPOTHESIS: Endogenous NO inhibits insulin release in isolated beta cells and insulin-degrading enzyme activity in hepatocytes, while NO release from endothelial cells has been suggested to enhance insulin action. We assessed the overall effect of systemic inhibition of endogenous NO synthesis on glucose homeostasis in humans. METHODS: Twenty-four non-diabetic volunteers underwent two hyperglycaemic (+7 mmol/l) clamps with either saline or L-NG-nitroarginine methyl ester (L-NAME, at rates of 2.5, 5, 10 and 20 μg min(-1) kg(-1)) infusion. Another five volunteers underwent an OGTT with either saline or L-NAME (20 μg min(-1) kg(-1)) infusion. Blood pressure and heart rate were measured to monitor NO blockade; during the OGTT, endothelial function was assessed by peripheral arterial tonometry and insulin secretion by C-peptide deconvolution and insulin secretion modelling. RESULTS: Compared with saline, L-NAME at the highest dose raised mean blood pressure (+20 ± 2 mmHg), depressed heart rate (-12 ± 2 bpm) and increased insulin clearance (+50%). First-phase insulin secretion was impaired, but insulin sensitivity (M/I index) was unchanged. During the OGTT, L-NAME raised 2 h plasma glucose by 1.8 mmol/l (p < 0.01), doubled insulin clearance and impaired beta cell glucose sensitivity while depressing endothelial function. CONCLUSIONS/INTERPRETATION: In humans, systemic NO blockade titrated to increase blood pressure and induce endothelial dysfunction does not affect insulin action but significantly impairs glucose tolerance by increasing plasma insulin clearance and depressing insulin secretion, namely first-phase and beta cell glucose sensitivity.