BACKGROUND Hyperglycemia, via peroxynitrite-mediated endothelial nitric oxide synthase (eNOS) enzymatic uncoupling, induced endothelial dysfunction. Propofol has been reported to improve high glucose-induced endothelial dysfunction. However, its mechanisms of action remain unclear. We hypothesized that propofol could improve hyperglycemia-induced endothelial dysfunction by decreasing the peroxynitrite level and thus restoring eNOS coupling. METHODS At the end of 3 days of incubation in medium with 30 mM glucose, human umbilical vein endothelial cells were treated with different concentrations (0.2, 1, 5, and 25 μM) of propofol for different times (0.5, 1, 2, and 4 hours). In parallel experiments, cells were cultured in 5 mM glucose for 3 days as a control. Nitric oxide (NO) production was measured with a nitrate reductase assay. Superoxide anion (O(2)(·-)) accumulation was measured with the reduction of ferricytochrome c and dihydroethidine fluorescence assay. The treatment that had maximal effect on 30 mM glucose-induced NO production and O(2)(·-) accumulation was applied in the following studies to examine the underlying signaling pathways. eNOS total protein, eNOS dimer and monomer expression, eNOS phosphorylation at Ser(1177), inducible NO synthase total protein, inducible NO synthase dimer and monomer expression, peroxynitrite, and guanosine triphosphate cyclohydrolase I expression were measured by Western blot. Tetrahydrobiopterin (BH(4)) level was measured with liquid chromatography-mass spectrometry. RESULTS Compared with 5 mM glucose treatment, 30 mM glucose significantly decreased NO production by 60% (P < 0.001) and increased O(2)(·-) accumulation by 175% (P = 0.0026), which were both attenuated by propofol in a concentration- and time-dependent manner. Compared with 5 mM glucose treatment, total eNOS protein expression was increased by 30 mM glucose (P < 0.001), whereas the ratio of eNOS dimer/monomer (P = 0.0001) and eNOS phosphorylation (P < 0.001) were decreased by 30 mM glucose. Propofol did not affect 30 mM glucose-induced total eNOS protein expression, but restored the ratio of eNOS dimer/monomer (P = 0.0005) and increased eNOS phosphorylation (P < 0.001). 30 mM glucose-induced O(2)(·-) accumulation was inhibited by the eNOS inhibitor hydrochloride. Furthermore, compared with 5 mM glucose treatment, 30 mM glucose decreased the BH(4) level (P = 0.0001) and guanosine triphosphate cyclohydrolase I expression (P < 0.001), whereas it increased peroxynitrite level (P = 0.0003), which could all be reversed by propofol (P = 0.0045, P < 0.001, P = 0.0001 vs 30 mM glucose treatment, respectively). CONCLUSIONS Propofol has beneficial effects on 30 mM glucose-induced NO reduction and O(2)(·-) accumulation in human umbilical vein endothelial cells. This may be mediated through inhibiting peroxynitrite-mediated BH(4) reduction, and restoring eNOS coupling.