Inhibition of endothelium-dependent smooth muscle relaxation by calmodulin antagonists

Involvement of Ca2+/calmodulin-dependent protein kinase II in endothelial NO production and endothelium-dependent relaxation

Nitric oxide (NO) is synthesized from l-arginine by the Ca(2+)/calmodulin-sensitive endothelial NO synthase (NOS) isoform (eNOS). The present study assesses the role of Ca(2+)/calmodulin-dependent protein kinase II (CaMK II) in endothelium-dependent relaxation and NO synthesis. The effects of three CaMK II inhibitors were investigated in endothelium-intact aortic rings of normotensive rats. NO synthesis was assessed by a NO sensor and chemiluminescence in culture medium of cultured porcine aortic endothelial cells stimulated with the Ca(2+) ionophore A23187 and thapsigargin. Rat aortic endothelial NOS activity was measured by the conversion of l-[(3)H]arginine to l-[(3)H]citrulline. Three CaMK II inhibitors, polypeptide 281-302, KN-93, and lavendustin C, attenuated the endothelium-dependent relaxation of endothelium-intact rat aortic rings in response to acetylcholine, A23187, and thapsigargin. None of the CaMK II inhibitors affected the relaxation induced by NO donors. In a porcine aortic endothelial cell line, KN-93 decreased NO synthesis and caused a rightward shift of the concentration-response curves to A23187 and thapsigargin. In rat aortic endothelial cells, KN-93 significantly decreased bradykinin-induced eNOS activity. These results suggest that CaMK II was involved in NO synthesis as a result of Ca(2+)-dependent activation of eNOS.

Vascular nitric oxide production during the development of two experimental models of portal hypertension

BACKGROUND/AIMS

The aim of this study was to determine the relative roles of constitutive NOS (NOS3) and inducible NOS (NOS2) isoforms during the development of two models of portal hypertension in rats.

METHODS

Vascular reactivity of aortic rings for norepinephrine was performed in control, sham-operated, portal-vein-stenosed and secondary biliary cirrhotic rats 1, 4, 7, 14 and 28 days after surgery. NOS activity and nitrate plasma levels were also measured.

RESULTS

An impaired response to norepinephrine observed in sham-operated, portal-vein-stenosed and cirrhotic rats at days 1 and 4 compared with controls was reversed after L-NNA and aminoguanidine. Portal hypertensive rats remained hyporeactive at days 7, 14 and 28 compared with sham-operated rats. At days 7 and 14 in portal-vein-stenosed rats, vascular hyporeactivity was reversed by L-NNA and W7. At days 14 and 28 in cirrhotic rats, vascular hyporeactivity was reversed by L-NNA and W7. Nitrate levels increased at day 1 in the 3 groups, and increased at days 14 and 28 in portal hypertensive rats. Total NOS-activity increased in cirrhotic rats at day 28, in portal-vein-stenosed rats at day 14, and in sham-operated rats at day 1 compared to controls. NOS2 activity increased only in sham-operated rats at day 1.

CONCLUSIONS

This study shows that for two models of portal hypertension, increased NO production in the first days is related to NOS2 induction secondary to surgery. On the other hand, when portal hypertension has fully developed, the NOS3 isoform appears to play the major role.

Endothelial calcium-calmodulin dependent nitric oxide synthase in the in vitro vascular hyporeactivity of portal hypertensive rats

BACKGROUND/AIMS

Increased nitric oxide production has been implicated in impaired vascular responsiveness to vasoconstrictors in portal hypertension. However, there is no firm evidence concerning the involved nitric oxide synthase isoform. The present study investigated the possible contribution of one nitric oxide synthase isoform, the endothelial constitutive Ca2+-calmodulin dependent, in the overproduction of nitric oxide in portal hypertension.

METHODS

Vascular responses to norepinephrine and acetylcholine were evaluated in isolated thoracic aortic rings from normal and portal vein stenosed rats.

RESULTS

An impaired concentration-dependent contraction to norepinephrine was observed in intact rings from portal hypertensive rats compared to controls. The hyporeactivity to norepinephrine was reversed after endothelium denudation, the inhibition of nitric oxide synthase with L-NOARG or the inhibition of calmodulin with W-7, but not after pre-incubation with indomethacin. Stimulation of intact rings with norepinephrine after the inhibition of calmodulin with calmidazolium was followed by a decreased vascular response in vessels from normal rats but not in those from portal hypertensive rats. Stimulation of intact rings with norepinephrine in a Ca2+-free medium was followed by a decreased vascular response in vessels from both portal hypertensive and normal rats. No difference in vasoconstrictive responses was observed between the two groups after calmidazolium or in a Ca2+-free medium. Relaxation induced by acetylcholine in norepinephrine-precontracted rings was more marked in rings from portal hypertensive rats than in controls. No differences in the vasodilator responses were observed after relaxations had been inhibited by the removal of the endothelium, pre-incubation with L-NOARG, indomethacin, W-7 or calmidazolium and in a Ca2+-free medium.

CONCLUSIONS

This study demonstrates the involvement of the endothelial constitutive Ca2+-calmodulin dependent nitric oxide synthase isoform in the overproduction of nitric oxide in portal hypertension.

Inhibition by calmodulin antagonists of the neurogenic relaxation in cerebral arteries

The present study was aimed to determine the effect of calmodulin inhibitors on the relaxant response of isolated dog and monkey cerebral arteries to vasodilator nerve stimulation, which is hypothesized to be mediated by nitric oxide (NO) from nerve endings. The relaxations caused by nerve stimulation by electrical pulses in endothelium-denuded arteries were attenuated by treatment with calmidazolium and W-7 (N-(6-aminohexyl)-5-chloro-1-naphthalene sulfonamide hydrochloride) and were abolished by NG-nitro-L-arginine, an inhibitor of nitric oxide synthase, and tetrodotoxin. The calmodulin inhibitors also attenuated the relaxations caused by nicotine and substance P, which were endothelium-independent and -dependent, respectively, but did not influence the relaxant response to NO. It is concluded that calmodulin is required for activation of the NO synthase present in the vasodilator nerve as well as that in the endothelium.

Relaxation of rat thoracic aorta induced by the Ca(2+)-ATPase inhibitor, cyclopiazonic acid, possibly through nitric oxide formation

1 The effect of the Ca(2+)-ATPase inhibitor, cyclopiazonic acid (CPA), was studied on rat thoracic aortic ring preparations. 2 At concentrations above 0.3 microM, CPA induced relaxation in the arteries precontracted with phenylephrine. Removal of the endothelium abolished CPA-induced relaxation. 3 The nitric oxide (NO) synthase inhibitor NG-nitro L-arginine (3-300 microM), the free radical scavenger haemoglobin (0.1-3 microM), the soluble guanylate cyclase inhibitor, LY83583 (0.1-10 microM), each inhibited the endothelium-dependent relaxation to CPA. The potassium channel blocker, glibenclamide (10 microM) and cyclo-oxygenase inhibitor, indomethacin (100 microM for 60 min and then washed out) did not alter the action of CPA. 4 The calmodulin inhibitors calmidazolium (3-10 microM) and W-7 (100 microM) also abolished CPA-induced relaxation. 5 CPA (10 microM) increased guanosine 3':5'-cyclic monophosphate (cyclic GMP) levels in arteries with an intact endothelium, without affecting adenosine 3':5'-cyclic monophosphate (cyclic AMP) levels. 6 The inhibitors of NO synthesis and actions, the calmodulin inhibitor and removal of the endothelium abolished the CPA-stimulated increase in the levels of cyclic GMP. 7 In Ca(2+)-free solution, CPA failed to induce relaxation or to stimulate cyclic GMP production. Relaxation to nitroprusside was not affected under these conditions. 8 These results suggest that CPA can stimulate NO synthesis, possibly by inhibiting a Ca(2+)-ATPase, which replenishes Ca2+ in the intracellular storage sites in endothelial cells. Depletion of the Ca2+ store in the endothelium may then trigger influx of extracellular Ca2+, contributing to an increase in free Ca2+ in the endothelial cells, which activates NO synthase and NO formation.

Endothelin-3-induced relaxation of rat thoracic aorta: a role for nitric oxide formation

1. Endothelin-3 (ET-3) at concentrations below those which caused contraction (30 nM) elicited endothelium-dependent relaxation followed by rebound contraction in rat isolated thoracic aorta. 2. Endothelin-1 also relaxed the rat aorta with a similar potency. 3. The nitric oxide (NO) synthase inhibitor, NG-nitro L-arginine, the radical scavenger, haemoglobin and the soluble guanylate cyclase inhibitor, methylene blue, each inhibited the ET-3-induced relaxation. 4. The calmodulin inhibitor, calmidazolium, considerably attenuated the relaxation caused by ET-3 without affecting that to nitroprusside. 5. Concentrations of ET-3 that were necessary to induce the relaxation also caused concentration-dependent elevation of guanosine 3':5'-cyclic monophosphate (cyclic GMP) levels. 6. NG-nitro L-arginine, haemoglobin, methylene blue, calmidazolium and removal of the endothelium completely abolished ET-3-stimulated cyclic GMP production. 7. These results suggest that ET-3 triggers NO formation possibly via ETB receptors on the endothelium to activate soluble guanylate cyclase, which in turn stimulates cyclic GMP production and smooth muscle relaxation. The enzyme contributing to the NO formation may be of the calcium/calmodulin-dependent, constitutive type.

Calmidazolium, a calmodulin inhibitor, inhibits endothelium-dependent relaxations resistant to nitro-L-arginine in the canine coronary artery

1. The role of calmodulin in endothelium-dependent relaxations in the canine coronary artery, was investigated by use of the inhibitor of calmodulin, calmidazolium. 2. The endothelium-dependent relaxations to adenosine diphosphate (ADP) and nebivolol, a beta-adrenoceptor antagonist, in control solution, and to bradykinin in high potassium solution (to inhibit endothelium-dependent hyperpolarization), were abolished by nitro-L-arginine (30 microM), an inhibitor of nitro oxide-synthase. Calmidazolium (10 microM) did not inhibit these relaxations. 3. Calmidazolium did not affect the endothelium-independent relaxations to SIN-1, an exogenous donor of nitric oxide (NO). 4. The relaxations to bradykinin and to the calcium ionophore A23187 in control solution were inhibited to a small extent by calmidazolium (10 microM). 5. Bradykinin and A23187 induced relaxations in the presence of nitro-L-arginine (30 microM) that were abolished by calmidazolium (10 microM) but not affected by glibenclamide (10 microM), an inhibitor of ATP-sensitive K+ channels. 6. The endothelium-independent relaxations to lemakalim, an ATP-sensitive K+ channel opener, were not affected by calmidazolium (10 microM) but were inhibited by glibenclamide (10 microM). 7. These results suggest that calmidazolium does not inhibit the endothelium-dependent relaxations due to endothelium-derived NO in the canine coronary artery but inhibits either the production of endothelium-derived hyperpolarizing factor (EDHF) from endothelial cells or its effects on vascular smooth muscle cells. Furthermore these results suggest that EDHF contributes to endothelium-dependent relaxations in the canine coronary artery.

Calmodulin antagonists inhibit endothelium-dependent hyperpolarization in the canine coronary artery

1. The effects of the calmodulin antagonists, calmidazolium and fendiline were investigated on endothelium-dependent hyperpolarization in the canine coronary artery. The membrane potential of vascular smooth muscle cells was measured with the microelectrode technique. 2. Smooth muscle cells of the canine coronary artery had a resting membrane potential of -50 mV. Bradykinin and the Ca(2+)-ionophore, A23187, induced concentration- and endothelium-dependent hyperpolarization. The hyperpolarization induced by a supramaximal concentration of bradykinin (10(-6) M) reached approximately 20 mV. 3. Calmidazolium (10(-5) M) and fendiline (10(-4) M) inhibited hyperpolarization induced by bradykinin and A23187. By contrast, calmidazolium did not affect the hyperpolarization induced by lemakalim, an opener of ATP-sensitive K(+)-channels. 4. These observations suggest that calmodulin is involved in the generation of endothelium-dependent membrane hyperpolarization of vascular smooth muscle.

Putative therapeutic applications of calmodulin antagonists

Calmodulin is the most important intracellular receptor protein for the second messenger calcium. The calcium-calmodulin complex regulates a number of physiological processes. An increasing number of pharmaceutical products is reported to interfere with the calcium-calmodulin complex. Despite the fact that the precise mechanisms of action of these so-called calmodulin antagonists await further clarification, reports accumulate in the literature indicating a broadening spectrum of putative therapeutic applications of calmodulin antagonists. Some of these applications, such as in cell proliferation, hypertension, congestive heart failure, arrhythmia and gastro-intestinal disorders, are discussed in the present review.

Endothelium-derived relaxing factor and the pulmonary circulation

Endothelium-derived relaxing factor (EDRF) is probably identical to nitric oxide (NO) and is released by the vascular endothelium both in the basal unstimulated state and in response to a wide range of physical and chemical stimuli. Since it was first described 10 years ago, evidence is accumulating that it is an important modulator of vascular smooth muscle tone. EDRF acts on the pulmonary vascular bed as on the systemic circulation. EDRF release to pharmacologic stimuli is impaired in pulmonary arteries from patients with chronic hypoxemia. This impairment is associated with severity of respiratory failure and of structural change of vessel walls. Disturbance of EDRF activity may be important in the pathophysiology of pulmonary vascular disease. This brief review describes the current status of experimental studies concerning the possible role of EDRF on the pulmonary circulation in normal conditions and in the pathogenesis of pulmonary hypertension.

Pertussis toxin and N-ethylmaleimide inhibit histamine- but not calcium ionophore-induced endothelium-dependent relaxation

Endothelium-dependent relaxation of the guinea pig pulmonary artery induced by histamine was inhibited by preincubation of the tissue with 10 microM N-ethylmaleimide (NEM) for 10 min, whereas the endothelium-dependent relaxation induced by the calcium ionophore A 23187 was not affected by NEM. Pretreatment of the preparations with 0.2-1 microgram/ml pertussis toxin for 120 min inhibited concentration-dependently the histamine-induced relaxation. In contrast, endothelium-dependent relaxation in response to the calcium ionophore A 23187 was not affected by pertussis toxin. Since NEM and pertussis toxin are thought to interfere with membrane located GTP binding proteins, it is suggested that such a coupling protein is involved in the signal transduction of the histamine receptor leading to endothelium-dependent relaxation.

Use of cultured cells to study the relationship between arachidonic acid and endothelium-derived relaxing factor

We have used mixed- and co-cultures of endothelial and vascular smooth muscle cells to investigate the role of phospholipase activation and arachidonic acid metabolites in the production of endothelium-derived relaxing factor (EDRF). Inhibition of phospholipase A2 with para-bromophenacyl bromide, dexamethasone or quinacrine, alone or in combination, blocked arachidonate release by 50%-60% but had no effect on EDRF production as assessed by cyclic GMP accumulation in mixed- or co-cultures of endothelial and vascular smooth muscle cells. Inhibition of the phospholipase C-diacylglycerol (DAG) lipase pathway of arachidonate release by the DAG lipase inhibitor RHC-80267 also caused partial inhibition of arachidonate release and had no effect on EDRF. When both phospholipase A2 and phospholipase C pathways for arachidonate mobilization were inhibited (dexamethasone + RHC 80267), arachidonate release was totally inhibited while EDRF release remained intact. We conclude that neither phospholipase activation nor arachidonate mobilization is required for EDRF release from cultured bovine endothelial cells.

Interaction between Bay-K-8644 and various relaxant agents on K+-depolarized contracture of guinea pig taenia coli

1. Interaction between Bay-K-8644 and nifedipine, D600, diltiazem, papaverine or W-7 on a K+-contracture of the guinea pig taenia coli have been investigated. 2. These relaxing drugs except Bay-K-8644 inhibited the K+-contracture in a concentration-dependent manner. 3. Complete inhibition by nifedipine (1 x 10(-7) M) of the contracture was absolutely antagonized by Bay-K-8644 (1 x 10(-6) M), a Ca2+-agonist, an apparent pA2 value of Bay-K-8644 as an antagonist for nifedipine obtained from Schild plot being 7.94 and its slope 0.99. 4. Marked inhibitions by D600 (1 x 10(-5) M), diltiazem (1 x 10(-5) M), papaverine (1 x 10(-4) M) and W-7 (1 x 10(-4) M) were unaffected but mild inhibitions by around IC50 of these relaxing agents, except W-7, were fully antagonized by Bay-K-8644. 5. These results suggest that Bay-K-8644 is a competitive antagonist for nifedipine, but not for other agents, on common Ca2+ channel sites.

Interference of phorbolesters with endothelium-dependent vascular smooth muscle relaxation

Histamine-induced endothelium-dependent relaxation (EDR) in the pulmonary artery was inhibited in a concentration-dependent manner by the phorbolester phorbol 12,13-dibutyrate (PDBu) (IC50: 70 nM) whereas EDR occurring in response to ionophore A 23187 was not affected by PDBu. The phorbolester 4 alpha-phorbol 12,13-didecanoate (4 alpha-PDD), which does not activate protein kinase C (PKC), was without effect on receptor- or ionophore-induced EDR. The observed inhibition of signal transduction by PKC activation is suggested to reflect phosphorylation of the GTP binding protein Ni.