Role of non-nitric oxide non-prostaglandin endothelium-derived relaxing factor(s) in bradykinin vasodilation

The most conspicuous effect of bradykinin following its administration into the systemic circulation is a transient hypotension due to vasodilation. In the present study most of the available evidence regarding the mechanisms involved in bradykinin-induced arterial vasodilation is reviewed. It has become firmly established that in most species vasodilation in response to bradykinin is mediated by the release of endothelial relaxing factors following the activation of B2-receptors. Although in some cases the action of bradykinin is entirely mediated by the endothelial release of nitric oxide (NO) and/or prostacyclin (PGI2), a large amount of evidence has been accumulated during the last 10 years indicating that a non-NO/PGI2 factor accounts for bradykinin-induced vasodilation in a wide variety of perfused vascular beds and isolated small arteries from several species including humans. Since the effect of the non-NO/PGI2 endothelium-derived relaxing factor is practically abolished by disrupting the K+ electrochemical gradient together with the fact that bradykinin causes endothelium-dependent hyperpolarization of vascular smooth muscle cells, the action of such factor has been attributed to the opening of K+ channels in these cells. The pharmacological characteristics of these channels are not uniform among the different blood vessels in which they have been examined. Although there is some evidence indicating a role for KCa or KV channels, our findings in the mesenteric bed together with other reports indicate that the K+ channels involved do not correspond exactly to any of those already described. In addition, the chemical identity of such hyperpolarizing factor is still a matter of controversy. The postulated main contenders are epoxyeicosatrienoic acids or endocannabinoid agonists for the CB1-receptors. Based on the available reports and on data from our laboratory in the rat mesenteric bed, we conclude that the NO/PGI2-independent endothelium-dependent vasodilation induced by BK is unlikely to involve a cytochrome P450 arachidonic acid metabolite or an endocannabinoid agonist.