Objective The present study was designed to analyse whether hypertension alters the involvement of cyclooxygenase-2-derived mediators in phenylephrineinduced vasoconstrictor responses. Methods Vascular reactivity experiments were performed in aortic segments from normotensive, Wistar-Kyoto, and spontaneously hypertensive rats (SHR); protein expression was measured by western blot and/or immunohistochemistry, and prostaglandin F-2 alpha. (PGF(2 alpha).), 8-isoprostane and prostacyclin release were determined by enzyme immunoassay commercial kits. Results The protein synthesis inhibitor dexamethasone 1 mu mol/I), the non-selective cyclooxygenase inhibitor indomethacin (10 mu mol/I), the selective cyclooxygenase-2 inhibitor NS 398 (11 mu mol/I), and the thromboxane A(2)/ prostaglandin H-2 (TIP) receptor antagonist SQ 29,548 (1 mu mol/I reduced the concentration -response curves to phenylephrine more in segments from hypertensive than from normotensive rats; however, the thromboxane A(2) (TxA(2)) synthase inhibitors furegrelate (10 mu mol/I and OKY 046 (1 and 10 mu mol/I) had no effect in either strain. Removing endothelium or adding dexamethasone almost abolished the NS 398 effect. Cyclooxygenase-2 protein expression, which was reduced by dexamethasone, was higher in aorta from hypertensive animals. In both strains cyclooxygenase-2 was localizedmainly in endothelial cells and adventitial fibroblasts. 13, 14-Dihydro-15-keto-PGF(2 alpha) 6-keto-PGF(1 alpha) and 8-isoprostane levels were greater in the medium from hypertensive than from normotensive rats; NS 398 decreased levels of the three metabolites studied only in the medium from SHR. Conclusions PGF(2 alpha) and 8-isoprostane seem to be involved in the response to phenylephrine in rat aorta; this involvement is greater in hypertensive rats, probably due to a higher endothelial induction of cyclooxygenase-2. (c) 2005 Lippincott Williams & Wilkins.