Cyclooxygenase-1 and prostacyclin production by endothelial cells in the presence of mild oxidative stress

• Published in: PloS one 2013-02, Vol.8 (2), p.e56683
• Main Authors: Toniolo, Alice, Buccellati, Carola, Pinna, Christian, Gaion, Rosa Maria, Sala, Angelo, Bolego, Chiara
• Format: Article
• Language: English
• Subjects: Animals; Aorta; Aorta, Thoracic - drug effects; Aorta, Thoracic - metabolism; Arachidonic acid; Biology; Blood vessels; Catalase; Catechin; Catechin - analogs & derivatives; Catechin - pharmacology; Cell culture; Cells, Cultured; COX-2 inhibitors; Cyclooxygenase 1 - metabolism; Cyclooxygenase 2 - metabolism; Cyclooxygenase Inhibitors - pharmacology; Cyclooxygenase-1; Cyclooxygenase-2; Endothelial cells; Endothelial Cells - drug effects; Endothelial Cells - metabolism; Endothelium; Epigallocatechin-3-gallate; Epoprostenol - biosynthesis; Human Umbilical Vein Endothelial Cells - metabolism; Humans; Hydrogen; Hydrogen peroxide; Hydrogen Peroxide - metabolism; In Vitro Techniques; Kinases; Male; Neovascularization; Nitric oxide; Nitric-oxide synthase; Oxidation; Oxidative Stress; Polyphenols; Prostacyclin; Rats; Rodents; Umbilical vein; Vasodilation; Vasodilator Agents - pharmacology
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0056683

This study aimed at evaluating the relative contribution of endothelial cyclooxygenase-1 and -2 (COX-1 and COX-2) to prostacyclin (PGI(2)) production in the presence of mild oxidative stress resulting from autooxidation of polyphenols such as (-)-epigallocatechin 3-gallate (EGCG), using both endothelial cells in culture and isolated blood vessels. EGCG treatment resulted in an increase in hydrogen peroxide formation in human umbilical vein endothelial cells. In the presence of exogenous arachidonic acid and EGCG, PGI(2) production was preferentially inhibited by a selective COX-1 inhibitor. This effect of selective inhibition was also substantially reversed by catalase. In addition, EGCG caused vasorelaxation of rat aortic ring only partially abolished by a nitric oxide synthase inhibitor. Concomitant treatment with a selective COX-1 inhibitor completely prevented the vasorelaxation as well as the increase in PGI(2) accumulation in the perfusate observed in EGCG-treated aortic rings, while a selective COX-2 inhibitor was completely uneffective. Our data strongly support the notions that H(2)O(2) generation affects endothelial PGI(2) production, making COX-1, and not COX-2, the main source of endothelial PGI(2) under altered oxidative tone conditions. These results might be relevant to the reappraisal of the impact of COX inhibitors on vascular PGI(2) production in patients undergoing significant oxidative stress.