Michel Félétou

EDHF: bringing the concepts together.

Endothelial cells synthesize and release vasoactive mediators in response to various neurohumoural substances (e.g. bradykinin or acetylcholine) and physical stimuli (e.g. cyclic stretch or fluid shear stress). The best-characterized endothelium-derived relaxing factors are nitric oxide and prostacyclin. However, an additional relaxant pathway associated with smooth muscle hyperpolarization also exists. This hyperpolarization was originally attributed to the release of an endothelium-derived hyperpolarizing factor (EDHF) that diffuses to and activates smooth muscle K(+) channels. More recent evidence suggests that endothelial cell receptor activation by these neurohumoural substances opens endothelial cell K(+) channels. Several mechanisms have been proposed to link this pivotal step to the subsequent smooth muscle hyperpolarization. The main concepts are considered in detail in this review.

Endothelium-dependent hyperpolarization of canine coronary smooth muscle

1 Experiments were designed to determine whether endothelium‐dependent relaxing factor(s) released by acetylcholine from the canine femoral artery influences the membrane potential of coronary arterial smooth muscle. 2 The membrane potential was recorded in small canine coronary arteries (internal diameter ≤ 500 μm; without endothelium) by means of intracellular microelectrodes. The organ bath also contained a strip of left descending coronary artery without endothelium in which isometric force was measured to bioassay relaxing factors(s) as well as segments of femoral artery with endothelium, which served as the source of endothelium‐derived relaxing factor(s). 3 Acetylcholine induced endothelium‐dependent, transient hyperpolarizations and relaxations that were not affected by indomethacin. 4 Inhibition of the sodium‐potassium pump by ouabain or potassium‐free solution did not inhibit the relaxation to acetylcholine but prevented the corresponding hyperpolarization. 5 Activation of the sodium‐potassium pump of the smooth muscle cells by readmission of potassium ions after incubation in potassium‐free solution caused relaxation and marked hyperpolarization. 6 These results suggest that endothelium‐derived relaxing factor(s) induces hyperpolarization of vascular smooth muscle of the canine coronary artery, possibly by activation of sodium‐potassium pumping, but that this effect on the cell membrane may only partially explain endothelium‐dependent relaxations evoked by acetylcholine.

Endothelial dysfunction and vascular disease.

The endothelium can evoke relaxations (dilatations) of the underlying vascular smooth muscle, by releasing vasodilator substances. The best characterized endothelium‐derived relaxing factor (EDRF) is nitric oxide (NO). The endothelial cells also evoke hyperpolarization of the cell membrane of vascular smooth muscle (endothelium‐dependent hyperpolarizations, EDHF‐mediated responses). Endothelium‐dependent relaxations involve both pertussis toxin‐sensitive Gi (e.g. responses to serotonin and thrombin) and pertussis toxin‐insensitive Gq (e.g. adenosine diphosphate and bradykinin) coupling proteins. The release of NO by the endothelial cell can be up‐regulated (e.g. by oestrogens, exercise and dietary factors) and down‐regulated (e.g. oxidative stress, smoking and oxidized low‐density lipoproteins). It is reduced in the course of vascular disease (e.g. diabetes and hypertension). Arteries covered with regenerated endothelium (e.g. following angioplasty) selectively loose the pertussis toxin‐sensitive pathway for NO release which favours vasospasm, thrombosis, penetration of macrophages, cellular growth and the inflammatory reaction leading to atherosclerosis. In addition to the release of NO (and causing endothelium‐dependent hyperpolarizations), endothelial cells also can evoke contraction (constriction) of the underlying vascular smooth muscle cells by releasing endothelium‐derived contracting factor (EDCF). Most endothelium‐dependent acute increases in contractile force are due to the formation of vasoconstrictor prostanoids (endoperoxides and prostacyclin) which activate TP receptors of the vascular smooth muscle cells. EDCF‐mediated responses are exacerbated when the production of NO is impaired (e.g. by oxidative stress, ageing, spontaneous hypertension and diabetes). They contribute to the blunting of endothelium‐dependent vasodilatations in aged subjects and essential hypertensive patients.

Endothelium-derived hyperpolarising factors and associated pathways: a synopsis

The term endothelium-derived hyperpolarising factor (EDHF) was introduced in 1987 to describe the hypothetical factor responsible for myocyte hyperpolarisations not associated with nitric oxide (EDRF) or prostacyclin. Two broad categories of EDHF response exist. The classical EDHF pathway is blocked by apamin plus TRAM-34 but not by apamin plus iberiotoxin and is associated with endothelial cell hyperpolarisation. This follows an increase in intracellular [Ca2+] and the opening of endothelial SKCa and IKCa channels preferentially located in caveolae and in endothelial cell projections through the internal elastic lamina, respectively. In some vessels, endothelial hyperpolarisations are transmitted to myocytes through myoendothelial gap junctions without involving any EDHF. In others, the K+ that effluxes through SKCa activates myocytic and endothelial Ba2+-sensitive KIR channels leading to myocyte hyperpolarisation. K+ effluxing through IKCa activates ouabain-sensitive Na+/K+-ATPases generating further myocyte hyperpolarisation. For the classical pathway, the hyperpolarising “factor” involved is the K+ that effluxes through endothelial KCa channels. During vessel contraction, K+ efflux through activated myocyte BKCa channels generates intravascular K+ clouds. These compromise activation of Na+/K+-ATPases and KIR channels by endothelium-derived K+ and increase the importance of gap junctional electrical coupling in myocyte hyperpolarisations. The second category of EDHF pathway does not require endothelial hyperpolarisation. It involves the endothelial release of factors that include NO, HNO, H2O2 and vasoactive peptides as well as prostacyclin and epoxyeicosatrienoic acids. These hyperpolarise myocytes by opening various populations of myocyte potassium channels, but predominantly BKCa and/or KATP, which are sensitive to blockade by iberiotoxin or glibenclamide, respectively.

Endothelium‐dependent contractions in hypertension

1 Endothelial cells, under given circumstances, can initiate contraction (constriction) of the vascular smooth muscle cells that surround them. Such endothelium‐dependent, acute increases in contractile tone can be due to the withdrawal of the production of nitric oxide, to the production of vasoconstrictor peptides (angiotensin II, endothelin‐1), to the formation of oxygen‐derived free radicals (superoxide anions) and/or the release of vasoconstrictor metabolites of arachidonic acid. The latter have been termed endothelium‐derived contracting factor (EDCF) as they can contribute to moment‐to‐moment changes in contractile activity of the underlying vascular smooth muscle cells. 2 To judge from animal experiments, EDCF‐mediated responses are exacerbated by aging, spontaneous hypertension and diabetes. 3 To judge from human studies, they contribute to the blunting of endothelium‐dependent vasodilatations in aged subjects and essential hypertensive patients. 4 Since EDCF causes vasoconstriction by activation of the TP‐receptors on the vascular smooth muscle cells, selective antagonists at these receptors prevent endothelium‐dependent contractions, and curtail the endothelial dysfunction in hypertension and diabetes.

Characterization of an apamin-sensitive small-conductance Ca2+-activated K+ channel in porcine coronary artery endothelium: relevance to EDHF

The apamin‐sensitive small‐conductance Ca2+‐activated K+ channel (SKCa) was characterized in porcine coronary arteries. In intact arteries, 100 nM substance P and 600 μM 1‐ethyl‐2‐benzimidazolinone (1‐EBIO) produced endothelial cell hyperpolarizations (27.8±0.8 mV and 24.1±1.0 mV, respectively). Charybdotoxin (100 nM) abolished the 1‐EBIO response but substance P continued to induce a hyperpolarization (25.8±0.3 mV). In freshly‐isolated endothelial cells, outside‐out patch recordings revealed a unitary K+ conductance of 6.8±0.04 pS. The open‐probability was increased by Ca2+ and reduced by apamin (100 nM). Substance P activated an outward current under whole‐cell perforated‐patch conditions and a component of this current (38%) was inhibited by apamin. A second conductance of 2.7±0.03 pS inhibited by d‐tubocurarine was observed infrequently. Messenger RNA encoding the SK2 and SK3, but not the SK1, subunits of SKCa was detected by RT – PCR in samples of endothelium. Western blotting indicated that SK3 protein was abundant in samples of endothelium compared to whole arteries. SK2 protein was present in whole artery nuclear fractions. Immunofluorescent labelling confirmed that SK3 was highly expressed at the plasmalemma of endothelial cells and was not expressed in smooth muscle. SK2 was restricted to the peri‐nuclear regions of both endothelial and smooth muscle cells. In conclusion, the porcine coronary artery endothelium expresses an apamin‐sensitive SKCa containing the SK3 subunit. These channels are likely to confer all or part of the apamin‐sensitive component of the endothelium‐derived hyperpolarizing factor (EDHF) response.

Endothelium-mediated control of vascular tone: COX-1 and COX-2 products

Endothelium‐dependent contractions contribute to endothelial dysfunction in various animal models of aging, diabetes and cardiovascular diseases. In the spontaneously hypertensive rat, the archetypal model for endothelium‐dependent contractions, the production of the endothelium‐derived contractile factors (EDCF) involves an increase in endothelial intracellular calcium concentration, the production of reactive oxygen species, the predominant activation of cyclooxygenase‐1 (COX‐1) and to a lesser extent that of COX‐2, the diffusion of EDCF towards the smooth muscle cells and the subsequent stimulation of their thromboxane A2‐endoperoxide TP receptors. Endothelium‐dependent contractions are also observed in various models of hypertension, aging and diabetes. They generally also involve the generation of COX‐1‐ and/or COX‐2‐derived products and the activation of smooth muscle TP receptors. Depending on the model, thromboxane A2, PGH2, PGF2α, PGE2 and paradoxically PGI2 can all act as EDCFs. In human, the production of COX‐derived EDCF is a characteristic of the aging and diseased blood vessels, with essential hypertension causing an earlier onset and an acceleration of this endothelial dysfunction. As it has been observed in animal models, COX‐1, COX‐2 or both isoforms can contribute to these endothelial dysfunctions. Since in most cases, the activation of TP receptors is the common downstream effector, selective antagonists of this receptor should curtail endothelial dysfunction and be of therapeutic interest in the treatment of cardiovascular disorders.

Acetylcholine‐induced relaxation in blood vessels from endothelial nitric oxide synthase knockout mice

Isometric tension was recorded in isolated rings of aorta, carotid, coronary and mesenteric arteries taken from endothelial nitric oxide synthase knockout mice (eNOS(−/−) mice) and the corresponding wild‐type strain (eNOS(+/+) mice). The membrane potential of smooth muscle cells was measured in coronary arteries with intracellular microelectrodes. In the isolated aorta, carotid and coronary arteries from the eNOS(+/+) mice, acetylcholine induced an endothelium‐dependent relaxation which was inhibited by Nω‐L‐nitro‐arginine. In contrast, in the mesenteric arteries, the inhibition of the cholinergic relaxation required the combination of Nω‐L‐nitro‐arginine and indomethacin. The isolated aorta, carotid and coronary arteries from the eNOS(−/−) mice did not relax in response to acetylcholine. However, acetylcholine produced an indomethacin‐sensitive relaxation in the mesenteric artery from eNOS(−/−) mice. The resting membrane potential of smooth muscle cells from isolated coronary arteries was significantly less negative in the eNOS(−/−) mice (−64.8±1.8 mV, n=20 and −58.4±1.9 mV, n=17, for eNOS(+/+) and eNOS(−/−) mice, respectively). In both strains, acetylcholine, bradykinin and substance P did not induce endothelium‐dependent hyperpolarizations whereas cromakalim consistently produced hyperpolarizations (−7.9±1.1 mV, n=8 and −13.8±2.6 mV, n=4, for eNOS(+/+) and eNOS(−/−) mice, respectively). These findings demonstrate that in the blood vessels studied: (1) in the eNOS(+/+) mice, the endothelium‐dependent relaxations to acetylcholine involve either NO or the combination of NO plus a product of cyclo‐oxygenase but not EDHF; (2) in the eNOS(−/−) mice, NO‐dependent responses and EDHF‐like responses were not observed. In the mesenteric arteries acetylcholine releases a cyclo‐oxygenase derivative.

Calcium-activated potassium channels and endothelial dysfunction: therapeutic options?

The three subtypes of calcium‐activated potassium channels (KCa) of large, intermediate and small conductance (BKCa, IKCa and SKCa) are present in the vascular wall. In healthy arteries, BKCa channels are preferentially expressed in vascular smooth muscle cells, while IKCa and SKCa are preferentially located in endothelial cells. The activation of endothelial IKCa and SKCa contributes to nitric oxide (NO) generation and is required to elicit endothelium‐dependent hyperpolarizations. In the latter responses, the hyperpolarization of the smooth muscle cells is evoked either via electrical coupling through myo‐endothelial gap junctions or by potassium ions, which by accumulating in the intercellular space activate the inwardly rectifying potassium channel Kir2.1 and/or the Na+/K+‐ATPase. Additionally, endothelium‐derived factors such as cytochrome P450‐derived epoxyeicosatrienoic acids and under some circumstances NO, prostacyclin, lipoxygenase products and hydrogen peroxide (H2O2) hyperpolarize and relax the underlying smooth muscle cells by activating BKCa. In contrast, cytochrome P450‐derived 20‐hydroxyeicosatetraenoic acid and various endothelium‐derived contracting factors inhibit BKCa. Aging and cardiovascular diseases are associated with endothelial dysfunctions that can involve a decrease in NO bioavailability, alterations of EDHF‐mediated responses and/or enhanced production of endothelium‐derived contracting factors. Because potassium channels are involved in these endothelium‐dependent responses, activation of endothelial and/or smooth muscle KCa could prevent the occurrence of endothelial dysfunction. Therefore, direct activators of these potassium channels or compounds that regulate their activity or their expression may be of some therapeutic interest. Conversely, blockers of IKCa may prevent restenosis and that of BKCa channels sepsis‐dependent hypotension.

Endothelium‐derived factors and hyperpolarization of the carotid artery of the guinea‐pig

1 Transmembrane potentials were recorded from isolated carotid arteries of the guinea‐pig supervised with modified Krebs‐Ringer bicarbonate solution. Smooth muscle cells were impaled from the adventitial side with intracellular glass microelectrodes filled with KCl (30–80 MΩ). 2 Acetylcholine (1 μm) in the presence of inhibitors of nitric oxide synthase, (Nω‐nitro‐L‐arginine (L‐NOARG) 100 μm) and cyclo‐oxygenase, (indomethacin 5 μm) induced an endothelium‐dependent hyperpolarization (−18.9±1.6 mV, n = 15). 3 In the presence of these two inhibitors, S‐nitroso‐L‐glutathione (10 μm), sodium nitroprusside (10 μm), 3‐morphohnosydnonimine (SIN‐1, 10 μm) and iloprost (0.1 μm) induced endothelium‐independent hyperpolarizations of the smooth muscle cells (respectively: −16.0±2.3, −16.3±3.4, −12.8±2.0 and −14.5±1.5 mV, n = 4–6). 4 The addition of glibenclamide (1 μm) did not influence the acetylcholine‐induced L‐NOARG/indomethacin‐resistant hyperpolarization (−18.0 ± 1.8 mV, n = 10). In contrast, the responses induced by S‐nitroso‐L‐glutathione, sodium nitroprusside, SIN‐1 and iloprost were abolished (changes in membrane potential: −0.8 ± 1.1, 1.3 ± 3.9, 4.5 ± 4.6 and 0.3 ± 0.8 mV respectively, n = 4–5). 5 In the presence of NO synthase and cyclo‐oxygenase inhibitors, charybdotoxin (0.1 μm) or apamin (0.5 μm) did not influence the hyperpolarization produced by acetylcholine. However, in the presence of the combination of charybdotoxin and apamin, the acetylcholine‐induced L‐NOARG/indomethacin‐resistant hyperpolarization was converted to a depolarization (4.4 ± 1.2mV, n = 20) while the endothelium‐independent hyperpolarizations induced by S‐nitroso‐L‐glutathione, sodium nitroprusside, SIN‐1 and iloprost were not affected significantly (respectively: −20.4 ± 3.4, −22.5 ± 4.9, −14.5 ± 4.7 and −14.5 ± 0.5mV, n = 4–5). 6 In the presence of the combination of charybdotoxin and apamin and in the absence of L‐NOARG and indomethacin, acetylcholine induced a hyperpolarization (−19.5 ± 3.7 mV, n = 4). This hyperpolarization induced by acetylcholine was not affected by the addition of indomethacin (−18.3 ± 4.6 mV, n = 3). In the presence of the combination of charybdotoxin, apamin and L‐NOARG (in the absence of indomethacin), acetylcholine, in 5 out of 7 vessels, still produced hyperpolarization which was not significantly smaller (−9.1 ± 5.6 mV, n = 7) than the one observed in the absence of L‐NOARG. 7 These findings suggest that, in the guinea‐pig isolated carotid artery, the endothelium‐independent hyperpolarizations induced by NO donors and iloprost involve the opening of KATP channels while the acetylcholine‐induced endothelium‐dependent hyperpolarization (resistant to the inhibition of NO‐synthase and cyclo‐oxygenase) involves the opening of Ca2+‐activated potassium channel(s). Furthermore, in this tissue, acetylcholine induces the simultaneous release of various factors from endothelial origin: hyperpolarizing factors (NO, endothelium derived hyperpolarizing factor (EDHF) and prostaglandins) and possibly a depolarizing factor.