Abstract-Flow-induced dilation (FID) is dependent largely on hyperpolarization of vascular smooth muscle cells (VSMCs) in human coronary arterioles (HCA) from patients with coronary disease. Animal studies show that shear stress induces endothelial generation of hydrogen peroxide (H 2 O 2 ), which is proposed as an endothelium-derived hyperpolarizing factor (EDHF). We tested the hypothesis that H 2 O 2 contributes to FID in HCA. Arterioles (135Ϯ7 m, nϭ71) were dissected from human right atrial appendages at the time of cardiac surgery and cannulated with glass micropipettes. Changes in internal diameter and membrane potential of VSMCs to shear stress, H 2 O 2 , or to papaverine were recorded with videomicroscopy. In some vessels, endothelial H 2 O 2 generation to shear stress was monitored directly using confocal microscopy with 2Ј,7Ј-dichlorofluorescin diacetate (DCFH) or using electron microscopy with cerium chloride. Catalase inhibited FID (%max dilation; 66Ϯ8 versus 25Ϯ7%; PϽ0.05, nϭ6), whereas dilation to papaverine was unchanged. Shear stress immediately increased DCFH fluorescence in the endothelial cell layer, whereas treatment with catalase abolished the increase in fluorescence. Electron microscopy with cerium chloride revealed shear stress-induced increase in cerium deposition in intimal area surrounding endothelial cells. Exogenous H 2 O 2 dilated (%max dilation; 97Ϯ1%, ED 50 ; 3.0Ϯ0.7ϫ10 Ϫ5 mol/L) and hyperpolarized HCA. Dilation to H 2 O 2 was reduced by catalase, 40 mmol/L KCl, or charybdotoxin plus apamin, whereas endothelial denudation, deferoxamine, 1H-1,2,4 -oxadiazole-[4,3-a]quinoxalin-1-one, or glibenclamide had no effect. These data provide evidence that shear stress induces endothelial release of H 2 O 2 and are consistent with the idea that H 2 O 2 is an EDHF that contributes to FID in HCA from patients with heart disease. The full text of this article is available at http://www.circresaha.org. (Circ Res. 2003;92:e31-e40.)Key Words: human Ⅲ coronary microcirculation Ⅲ flow-induced dilation Ⅲ hydrogen peroxides P hysiologically, shear stress plays a critical role in the regulation of vascular tonus and vascular homeostasis, contributing to the maintenance of tissue perfusion and vascular integrity. Shear stress-induced release of nitric oxide (NO) from endothelial cells is widely recognized as one of the most important and common mechanisms for shear-induced vasomotion. For example, flow-induced release of NO is responsible for the mediation of flow-induced vasodilation (FID). [1][2][3] Animal studies have reported that the contribution of NO to FID is reduced as oxidative stress increases in the presence of risk factors for cardiovascular disease such as hypercholesterolemia 4 and hypertension. 5 In humans, in vivo and in vitro studies have demonstrated that relaxant factor(s) other than NO compensate to maintain FID when NO availability is reduced. 6,7 We recently reported that FID is mediated largely by endothelium-derived hyperpolarizing factor (EDHF) in human coronary arterioles (HCAs...

Thrombin (Thromb), activated as part of the clotting cascade, dilates conduit arteries through an endothelial pertussis toxin (PTX)-sensitive G-protein receptor and releases nitric oxide (NO). Thromb also acts on downstream microvessels. Therefore, we examined whether Thromb dilates human coronary arterioles (HCA). HCA from right atrial appendages were constricted by 30–50% with endothelin-1. Dilation to Thromb (10−4–1 U/ml) was assessed before and after inhibitors with videomicroscopy. There was no tachyphylaxis to Thromb dilation (maximum dilation = 87.0%, ED50 = 1.49 × 10−2). Dilation to Thromb was abolished with either hirudin or denudation but was not affected by PTX. Neither N ω-nitro-l-arginine methyl ester ( n = 7), indomethacin ( n = 9),1H-[1,2,4] oxadiazolo-[4,3-a]quinoxalin-1-one ( n = 6), tetraethylammonium chloride ( n = 5), nor iberiotoxin ( n = 4) reduced dilation to Thromb. However, KCl (maximum dilation = 89 ± 5 vs. 20 ± 10%; P < 0.05; n = 7), tetrabutylammonium chloride (maximum dilation = 79 ± 7 vs. 21 ± 4%; P < 0.05; n = 5), and charybdotoxin (maximum dilation = 89 ± 4 vs. 10 ± 2%; P < 0.05; n = 4) attenuated dilation to Thromb. In contrast to animal models, Thromb-induced dilation in human arterioles is independent of Gi-protein activation and NO release. However, Thromb dilation is endothelium dependent, is maintained on consecutive applications, and involves activation of K+ channels. We speculate that an endothelium-derived hyperpolarizing factor contributes to Thromb-induced dilation in HCA.