1. In addition to nitric oxide (NO) and prostacyclin (PGJ2) an as yet unidentified endothelium-derived hyperpolarizing factor (EDHF) contributes to the dilator effect of bradykinin in different vascular beds. We have investigated the nature and mechanism of action of this factor in freshly isolated bovine and porcine coronary artery segments which were preconstricted with the thromboxane mimetic U46619 (9,11-dideoxy-lla, 9ac-epoxymethano-prostaglandin F2a, 10-30 nM). 2. The concentration-response curve of bradykinin was significantly shifted to the right after inhibition of NO synthesis with NG-nitro-L-arginine (L-NNA, 30 FM), whereas cyclo-oxygenase blockade with diclofenac (1 AM) had no effect. Preconstriction of the segments with potassium chloride (40-60 mM) completely abrogated the NO/PGJ2-independent dilator response to bradykinin. In sandwich bioassay experiments, both the luminal and abluminal release of NO, but not that of EDHF, was readily detectable.3. Inhibitors of Ca2+-activated K+ channels (K+.), such as apamin (1 /M) and tetrabutylammonium (TBA, 3 mm), strongly attenuated the EDHF-mediated bradykinininduced relaxation, while glibenclamide (3/1M), an inhibitor of KATP channels, had no effect. 4. These relaxations were also significantly inhibited by the phospholipase A2 inhibitor, quinacrine (30 #M), and the cytochrome P450 inhibitors, SKF525a (30-100 SM) and clotrimazole (100 #M). Moreover, incubation of endothelium-denuded coronary artery rings with a cytochrome P450-derived arachidonic acid metabolite, 11,12-epoxyeicosatetraenoic acid, elicited a concentration-dependent (1-10 /IM) dilatation which was abolished both in the presence of TBA (3 mM) and following preconstriction of the segments with potassium chloride instead of U46619. 5. These findings suggest that EDHF released by bradykinin is a cytochrome P450-derived arachidonic acid metabolite, presumably an epoxide. This factor seems to hyperpolarize the underlying smooth muscle cell layers by opening Kca channels.

The angiotensin I (AI) metabolite, A(1-7), elicited a concentration-dependent dilator response (ED50 > or = 2 microM) in porcine coronary artery rings which was markedly attenuated by the nitric oxide (NO) synthase inhibitor, NG-nitro-L-arginine, and abolished after removal of the endothelium. This effect of the heptapeptide was not mimicked by AII, AIII or A(3-8) at comparable concentrations. The A(1-7)-induced relaxation was not affected by AT1 or AT2 receptor blockade or cyclo-oxygenase inhibition, but was attenuated by the B2 receptor antagonist, Hoe 140, and augmented by the angiotensin-converting enzyme (ACE) inhibitor, quinaprilat. These findings suggest that the relaxation to A(1-7) was mediated by the release of NO from the coronary endothelium through activation of an, as yet unidentified, AT receptor, the occupation of which also seems to stimulate the release of vasoactive kinins. Since A(1-7) accumulates during ACE inhibition, this mechanism may contribute to the coronary dilator effect of ACE inhibitors in vivo.

1 Angiotensin-converting enzyme (ACE) inhibitors exert their cardiovascular e ects not only by preventing the formation of angiotensin II (AII), but also by promoting the accumulation of bradykinin in or at the vessel wall. In addition, certain ACE inhibitors have been shown to augment the vasodilator response to bradykinin, presumably by an interaction at the level of the B 2 receptor. We have investigated whether this is a speci®c e ect of the ACE inhibitor class of compounds in isolated endothelium-denuded segments of the rabbit jugular vein where bradykinin elicits a constrictor response which is exclusively mediated by activation of the B 2 receptor. 2 Moexiprilat and ramiprilat (4 3 nM) enhanced the constrictor response to bradykinin three to four fold. Captopril and enalaprilat were less active by approximately one and quinaprilat by two orders of magnitude. Moexiprilat and ramiprilat, on the other hand, had no e ect on the constrictor response to AII or the dilator response to acetylcholine. 3 The bradykinin-potentiating e ect of the ACE inhibitors was not mimicked by inhibitors of amino-, carboxy-, metallo-or serine peptidases or the synthetic ACE substrate, hippuryl-L-histidyl-L-leucine, at a concentration which almost abolished the residual ACE activity in the vessel wall. In contrast, angiotensin-(1 ± 7) (10 mM), an angiotensin I metabolite, signi®cantly enhanced the constrictor response to bradykinin. 4 Ramiprilat did not alter the binding of [ 3 H]-bradykinin to a membrane fraction prepared from endothelium-denuded rabbit jugular veins or to cultured ®broblasts, and there was no ACE inhibitorsensitive, bradykinin-induced cleavage of the B 2 receptor in cultured endothelial cells. 5 These ®ndings demonstrate that ACE inhibitors selectively potentiate the B 2 receptor-mediated vascular e ects of bradykinin. Their relative e cacy appears to be independent of their ACE-inhibiting properties and might be related to di erences in molecule structure. Moreover, the potentiation of the biological activity of bradykinin by this class of compounds does not seem to be mediated by a shift in a nity of the B 2 receptor or a prevention of its desensitization, but may involve an increase in the intrinsic activity of unoccupied B 2 receptor molecules.

1 To examine the possibility that angiotensin-converting enzyme (ACE) inhibitors modulate the action of bradykinin at the receptor level, their effect on the dilator response to bradykinin was studied in the isolated saline-perfused heart of the rabbit. 2 Continuous infusion of bradykinin (10 nM were unable to elicit a significant change in CPP or PGI2 release while ramiprilat and another ACE inhibitor, quinaprilat, were still active in the presence of these substrates. 6 To reveal the potential B2-receptor action of ramiprilat, its effect on the constrictor response to bradykinin was studied in the rabbit isolated jugular vein. Ramiprilat (0.1 M), but not RA-octyl (1 pM), potentiated the endothelium-independent, B2-receptor-mediated constrictor response to bradykinin, but not that to the thromboxane-mimetic U46619 (9,11-dideoxy-lla,9a-epoxymethano-prostaglandin F2.).Moreover, ramiprilat but not RA-octyl caused a concentration-dependent, B2-receptor antagonistsensitive increase in tone when administered alone. 7 These findings suggest that an interaction of ACE inhibitors with the B2-receptor or its signal transduction pathway rather than an accumulation of bradykinin within the vascular wall is responsible for the restoration of the endothelial response to bradykinin (dilatation, PGI2 release) in the coronary vascular bed of the rabbit.

The vasodilator effects of C92-4609 (4-hydroxymethyl-furoxan-3-carboxamide, CAS 1609), C92-4678(4-phenyl-furoxan-3-carboxylic acid (pyridyl-3-yl-methyl)-amide), C92-4679 (3-phenyl-furoxan-4-carboxylic acid (pyridyl-3-yl-methyl)-amide) and C93-4759 (3-hydroxymethyl-furoxan-4-carboxamide) were studied in the isolated rabbit femoral artery and jugular vein. All furoxans were potent vasodilators in the femoral artery (EC50 0.1-50 microM), while they were less potent in the jugular vein by at least one order of magnitude. Apart from C92-4679, the vasodilatory potency of the furoxans correlated well with their nitric oxide (NO)-releasing capacity which was estimated both by stimulation of purified soluble guanylyl cyclase activity and electron spin resonance spectroscopy with a trapping agent for NO. The hypothesis that furoxans stimulate soluble guanylyl cyclase in the smooth muscle by spontaneously releasing NO was supported by the marked attenuation of their vasodilator effect in the presence of oxyhaemoglobin (10 microM) or following treatment with methylene blue (30 microM). In contrast to earlier findings, NO release from these furoxans was not thiol-dependent, as demonstrated for C92-4609, the relaxant effect of which in the femoral artery was not altered in the presence of N-acetyl-L-cysteine (1 mM). Moreover the KCa+ channel inhibitor, tetrabutylammonium (3 mM), but not the KATP+ channel inhibitor, glibenclamide (3 microM), significantly attenuated the dilator response to C92-4679 in the femoral artery. Pretreatment of these segments with the cytochrome P450 inhibitor, SKF525a (30 microM), also reduced the C92-4679-induced relaxation in this vascular bed.(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract

The vascular endothelium is the source of a number of vasodilator and vasoconstrictor autacoids and is thus a key regulator of vascular homeostasis. We studied the effects of altering the balance between protein tyrosine kinase and phosphatase activity on Ca²⁺ signalling and phosphotyrosine levels in cultured human endothelial cells, as well as on autacoid production in native endothelial cells. In isolated segments of rabbit aorta and carotid artery, as well as in bovine coronary arteries, the tyrosine phosphatase inhibitors phenylarsine oxide (PAO) and sodium orthovanadate initiated endothelium-dependent relaxations which could be attributed to the release of nitric oxide and the endothelium-derived hyperpolarizing factor. In cultured endothelial cells incubation with PAO resulted in a time-dependent accumulation in 6-keto prostaglandin F_1α, the stable metabolite of prostacyclin, as well as in an increase in the intracellular concentration of free Ca²⁺ ([Ca²⁺]i). Inhibition of tyrosine kinases attenuated both the PAO-induced relaxation and the increase in endothelial [Ca²⁺]i. Western blot analysis of endothelial cells treated with the tyrosine phosphatase inhibitors revealed a time-dependent increase in the tyrosine phosphorylation of a series of bands in both the Triton X-100-soluble and Triton X-100-insoluble (cytoskeletal) fractions. These observations suggest that alterations in cellular levels of phosphotyrosine may have profound effects on vascular homeostasis by modulating Ca²⁺ signalling and autacoid production in endothelial cells.

In the presence of indomethacin (10 µM) and without previous exposure to bradykinin, two angiotensin-converting enzyme (ACE) inhibitors, moexiprilat and ramiprilat (0.1 µM), elicited distinct relaxation responses from preconstricted endothelium-intact but not from denuded bovine coronary artery rings, and enhanced the relaxation response to bradykinin (3 nM). All of these responses were strongly reduced by the selective B2-kinin receptor antagonist Hoe 140 (0.1 µM). Bradykinin (30 nM), moexiprilat or ramiprilat (0.3 µM) also significantly increased the cyclic GMP content of these coronary segments, an effect which was abolished by the nitric oxide (NO) synthase inhibitor N^G-nitro-L-arginine (L-NNA; 30 µM) or by removal of the endothelium. L-NNA also strongly reduced the relaxation response to moexiprilat but only partially inhibited that to bradykinin, demonstrating that the ACE inhibitor-induced relaxation was predominantly mediated by endothelial NO release, whereas bradykinin acted in part through another endothelium-dependent mechanism. These findings suggest that ACE inhibitors induce endothelium-dependent relaxation presumably by facilitating the accumulation of endothelium-derived vasoactive kinins in or at the vessel wall. This local mechanism may significantly contribute to the dilator action of these compounds in vivo.