Propofol is known to cause vasorelaxation of several systemic vascular beds. However, its effect on the pulmonary vasculature remains controversial. In the present study, we investigated the effects of propofol on human pulmonary arteries obtained from patients who had undergone surgery. Arterial rings were mounted in a Multi-Myograph system for measurement of isometric forces. U46619 was used to induce sustained contraction of the intrapulmonary arteries, and propofol was then applied (in increments from 10-300 μM). Arteries denuded of endothelium, preincubated or not with indomethacin, were used to investigate the effects of propofol on isolated arteries. Propofol exhibited a bifunctional effect on isolated human pulmonary arteries contracted by U46619, evoking constriction at low concentrations (10-100 μM) followed by secondary relaxation (at 100-300 μM). The extent of constriction induced by propofol was higher in an endothelium-denuded group than in an endothelium-intact group. Preincubation with indomethacin abolished constriction and potentiated relaxation. The maximal relaxation was greater in the endothelium-intact than the endothelium-denuded group. Propofol also suppressed CaCl2-induced constriction in the 60 mM K+-containing Ca2+-free solution in a dose-dependent manner. Fluorescent imaging of Ca2+ using fluo-4 showed that a 10 min incubation with propofol (10-300 μM) inhibited the Ca2+ influx into human pulmonary arterial smooth muscle cells induced by a 60 mM K+-containing Ca2+-free solution. In conclusion, propofol-induced arterial constriction appears to involve prostaglandin production by cyclooxygenase in pulmonary artery smooth muscle cells and the relaxation depends in part on endothelial function, principally on the inhibition of calcium influx through L-type voltage-operated calcium channels.