Four weeks after balloon denudation, porcine coronary arteries with regenerated endothelium selectively lose endothelium-dependent relaxations to agonists such as serotonin while those to bradykinin or adenosine diphosphate (ADP) are maintained. The aim of the present study was to investigate the relative contribution of endothelium-derived hyperpolarizing factor (EDHF) to the relaxation induced by bradykinin in arteries with regenerated endothelium. Transmembrane potential, isometric tension and levels of cyclic-Guanosine Monophosphate and cyclic-Adeno sine Monophosphate (cyclic GMP and cyclic AMP) were measured simultaneously in each strip with regenerated endothelium and compared with those of the corresponding control coronary artery. Under basal conditions, in coronary arteries with regenerated endothelium, a depolarization of vascular smooth muscle cells was associated with a decreased level of cyclic GMP without alteration in that of cyclic AMP. Exogenous nitric oxide (NO) was added in order to compensate for the reduced level of cyclic GMP. The resulting repolarization of the injured coronary arteries demonstrated the involvement of the NO pathway in the control of resting membrane potential. This effect was cyclic GMP-dependent, as it was blocked by an inhibitor of soluble guanylate cyclase, oxadiazoloquinoxalin (ODQ). When contracted with prostaglandin F2alpha, arteries with normal or regenerated endothelium depolarized whatever the conditions studied (control, inhibition of cycloxygenase with or without that of nitric oxide synthase). In the presence of regenerated endothelium, spikes and phasic contraction were observed. In all blood vessels, both in the presence or the absence of indomethacin, bradykinin evoked nearly maximal relaxations suggesting no involvement of prostacyclin in the response. Additional blockade of nitric oxide synthase by N-omega-nitro-L-arginine, reduced the relaxation, demonstrating a small contribution of NO. In both coronary arteries, the relaxation induced by bradykinin was mediated essentially by EDHF. The membrane potential reached during exposure to bradykinin was always less negative in the presence of regenerated endothelium, suggesting that the NO pathway participated to the hyperpolarization in repolarizing the blood vessel. Thus, the unaltered relaxation to bradykinin despite the reduced production of NO, suggests that the endothelium-dependent hyperpolarization is sufficient to maintain a normal relaxation in coronary arteries with regenerated endothelium. Abnormal endothelium-dependent relaxation is an early event in the development of vascular disease both in animal models (Yamamoto et al., 1987; Shimokawa et al., 1987; Cohen et al., 1988) and in humans (Forstermann et al., 1988; Creager et al., 1990). In porcine coronary arteries with regenerated endothelium after balloon denudation, endothelium-dependent relaxations are normalized 8 days after the procedure (Shimokawa et al., 1987). However, 4 weeks after such denudation, the pertussis-toxin sensitive G-protein coupled relaxation to serotonin is reduced while that induced by bradykinin is normal (Shimokawa et al., 1987, 1989; Borg-Capra et al., 1997). In the porcine coronary artery, it is unlikely that NO fully explains the endothelium-dependent relaxations induced by bradykinin (Richard et al., 1990; Cowan and Cohen, 1991; Nagao and Vanhoutte, 1992), suggesting an important contribution of endothelium-derived hyperpolarizing factor (EDHF) in the response (Beny and Haefliger, 1999). Four weeks after balloon denudation, in porcine coronary arteries with regenerated endothelium, the smooth muscle cells are depolarized and the hyperpolarization induced by bradykinin, in the presence of inhibitors of nitric oxide synthase and cyclooxygenase, is correlated with the value of membrane potential before the administration of the kinin. As a result, an increase in the hyperpolarization in response to bradykinin occurs in the most depolarized cells, suggesting a greater contribution of EDHF in these coronary arteries (Thollon et al., 1999a). During contractions to prostaglandin F-2alpha, in the presence of a blocker of cyclooxygenase, the hyperpolarization induced by bradykinin was normal in previously denuded coronary arteries (Thollon et al., 1999b), suggesting that nitric oxide may curtail the EDHF-mediated component (Olmos et al., 1995). Thus the aim of the present study was to measure the endothelium-dependent hyperpolarization in response to bradykinin in porcine coronary ateries with regenerated endothelium contracted with prostaglandin F-2alpha and to investigate its relative contribution in the relaxation evoked by the peptide.
