This Article Extract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cohen, R. A. Articles by Vanhoutte, P. M. Search for Related Content PubMed PubMed Citation Articles by Cohen, R. A. Articles by Vanhoutte, P. M. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB NITRIC OXIDE

(Circulation. 1995;92:3337-3349.)
© 1995 American Heart Association, Inc.

Articles

Endothelium-Dependent Hyperpolarization

Beyond Nitric Oxide and Cyclic GMP

Richard A. Cohen; Paul M. Vanhoutte

From the Vascular Biology Unit (R.A.C.), Vascular Medicine Section, Robert Dawson Evans Department of Clinical Research, Boston University School of Medicine, Boston, Mass; and Institut de Recherche Servier (P.M.V.), Suresnes, France.

Correspondence to Dr Richard A. Cohen, Vascular Biology Unit, Vascular Medicine Section, Robert Dawson Evans Department of Clinical Research, Boston University School of Medicine, Boston, MA.

Key Words: endothelium-dependent hyperpolarization • nitric oxide • vascular smooth muscle • ion channels • membrane potential

	Introduction

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
Blood flow and blood pressure are determined by an integration of reflex, humoral, and local vascular control mechanisms. Knowledge of these mechanisms has mushroomed over the past 15 years, particularly in the area of local endothelium-dependent vasomotor control. This has stemmed from the pioneering report in 1980 by Robert Furchgott¹ demonstrating that the endothelium releases a vasodilator substance in response to acetylcholine. This concept has been expanded with knowledge that the endothelium releases a variety of relaxing and contracting factors that regulate the underlying smooth muscle.² The most widely known endothelium-derived relaxing factor, nitric oxide, is released from endothelial cells in response to shear stress or stimulation of different receptors for a variety of neurohumoral mediators on the endothelial cell surface.³ The increase in endothelial cell calcium initiated by these stimuli increases the activity of a constitutively expressed enzyme, nitric oxide synthase, which converts L-arginine to nitric oxide and citrulline.^{4 5 6 7} Nitric oxide thus formed diffuses to and inhibits contraction of the underlying vascular smooth muscle.³ The physiological significance of the production of endothelial nitric oxide is suggested by the vasoconstriction observed in most vascular beds^{8 9 10 11 12 13} and the increase in systemic arterial blood pressure,^{14 15 16} which occurs on infusion of inhibitors of nitric oxide synthase. This observation further implies that under normal conditions, endothelial cells are locally liberating nitric oxide which effectively inhibits vasoconstriction arising by other mechanisms. Thus, normal vascular homeostasis depends in the periphery on a balance between neurally and humorally mediated vasoconstriction in skeletal muscle, mesentery, and the kidney, and local endothelium-dependent vasodilatation. In the truly vital areas of the heart, brain, and genitalia, vasodilator neural mechanisms reinforce the vasodilator influence of the endothelium.¹⁷

All is not known regarding the vasodilator function of the endothelium. In the present article, we elaborate that endothelium-dependent vasodilatation can best be explained by the participation of at least two vasodilator substances, nitric oxide and a substance whose identity is unknown. The identification of nitric oxide as an endothelium-derived relaxing factor rests on the similarities in physical, chemical, and physiological characteristics between the endogenous substance and authentic nitric oxide, as well as evidence derived from the use of arginine analogues as competitive inhibitors of nitric oxide synthesis.^{18 19} Thus, although the precise chemical nature of the nitric oxide–like endothelium-derived vasodilator remains debatable,²⁰ the evidence appears incontrovertible that nitric oxide, in one chemical form or another, participates in endothelium-dependent relaxation. On the other hand, it has long been known that acetylcholine, the prototypical endothelium-dependent vasodilator, causes hyperpolarization of smooth muscle,^{21 22 23 24 25} and the endothelium dependence of the response was first described by Bolton and colleagues.²⁶ Hyperpolarization of smooth muscle is an action that has not been generally associated with nitric oxide.^{27 28 29} In addition, in many studies of isolated blood vessels and the intact circulation, the action of endothelium-dependent vasodilators is, at least in part, resistant to inhibitors of nitric oxide.^{9 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44} Although the number of observations in human blood vessels is limited, endothelium-dependent hyperpolarization of human coronary arteries in response to bradykinin also is resistant to nitric oxide synthase inhibitors.⁴³ Therefore, the existence of two endothelium-derived mediators has been proposed to account for endothelium-dependent relaxation—one nitric oxide, and the other an endothelium-derived hyperpolarizing factor (EDHF).^{45 46 47 48 49 50} It is important to note that hyperpolarization is, thus far, a phenomenon that has been measured only in vitro. Thus, the existence of a hyperpolarizing factor in vivo has been deduced to play a role in vasodilatation by the noted failure of nitric oxide synthase inhibitors to completely block endothelium-dependent vasodilation.

Furthermore, the picture is complicated by the fact that at least two different modes of signal transduction may account for endothelium-dependent vasodilation, even for a single endothelium-derived relaxing factor. The first, described for nitric oxide, depends on the stimulation of guanylate cyclase and production of its product, cGMP. Reliance of endothelium-dependent vasodilation on this mechanism is based on the parallel drawn between the increase in cGMP content of arterial tissue caused by endothelium-dependent vasodilators and nitrovasodilators, whose action is based on releasing nitric oxide.^{51 52 53 54 55} In addition, endothelium-dependent relaxation to nitric oxide may be reduced by hemoglobin or methylene blue, which antagonize the rise in cGMP either by inhibiting guanylate cyclase or by scavenging nitric oxide and preventing its stimulation of the enzyme.^{53 54 56 57 58 59 60} The cyclic nucleotide causes smooth muscle relaxation chiefly via phosphorylation of, and multiple actions mediated by cGMP-dependent kinase(s), and thus to a decrease in intracellular calcium and a decrease in calcium sensitivity of the contractile apparatus.^{61 62 63 64 65 66 67}

The second mechanism is largely independent of cGMP and may be mediated by hyperpolarization of the smooth muscle cell membrane. Having demonstrated efficacious inhibition of the rise in arterial cGMP by arginine analogue inhibitors of nitric oxide synthesis or by methylene blue or hemoglobin, potent relaxation^{30 32 44} and hyperpolarization^{37 68} to bradykinin or acetylcholine persist in the porcine coronary artery,^{30 44} the carotid artery and abdominal aorta of the rabbit,³² and the aorta and pulmonary artery of the rat.³⁷ These observations attest to cGMP-independent mechanisms of endothelium-dependent relaxation. A cGMP-independent mechanism of relaxation may also apply to nitric oxide and nitrovasodilators.^{69 70} Thus, there is evidence for more than one endothelium-derived mediator, one or all of which may mediate relaxation by more than one mechanism. It is also possible that the two mediators are interrelated, either by sharing common mechanisms or by reinforcing or otherwise modulating the action of the other (Figs 1 and 2). The purpose of this review is to summarize the current understanding, gaps in knowledge, and potential directions for research on endothelium-dependent vasodilation as it relates to the question of the existence, identification, and mode of action of a second, as-yet-unidentified mediator that behaves principally as an EDHF.

View larger version (31K): [in this window] [in a new window]   Figure 1. Schematic demonstrating the release of putative endothelium-derived hyperpolarizing factors (EDHFs), including nitric oxide (NO), prostacylin (PGI2), eicosatrienoic acids (EETs), and unidentified EDHFs. Hormones and autacoids acting at endothelial membrane receptors (R) or shear stress cause G protein (G)-mediated increases in calcium (Ca2+) influx and phospholipase C (PLC)-mediated hydrolysis of phosphotidyl inositol bisphosphate, yielding inositol triphosphate (IP3), which releases calcium from intracellular stores. The resulting increase in calcium activates nitric oxide synthase to produce nitric oxide, activates phospholipase A2 to release arachidonic acid (AA), and releases unidentified EDHFs. The released arachidonic acid is converted to prostacyclin and eicosatrienoic acids, two putative EDHFs. The endothelial cell membrane potential (Vm) is hyperpolarized, possibly due to activation of calcium-dependent potassium channels, either by the rise in calcium or by the EDHFs acting in an autocrine manner. The resulting hyperpolarization accentuates calcium influx due to the increased electrochemical gradient for calcium and thereby by a positive feedback loop potentiates the release of endothelium-derived mediators of relaxation.

View larger version (40K): [in this window] [in a new window]   Figure 2. Schematic demonstrating mechanisms by which vascular smooth muscle cells respond to EDHFs and nitric oxide. EDHF may hyperpolarize smooth muscle cells by activating potassium channels or possibly Na+,K+-ATPase (ATP). The resulting hyperpolarization decreases smooth cell calcium levels by inhibiting influx through voltage-dependent calcium channels and favoring the re-uptake of calcium into intracellular stores and extrusion of calcium from the cell, resulting in a decrease in intracellular free calcium levels and relaxation. Nitric oxide also stimulates guanylate cyclase (GC), which converts GTP to cGMP. The cyclic nucleotide phosphorylates kinases, which favor reuptake of calcium, increase calcium extrusion from the cell and possibly inhibit calcium influx, all leading to a decrease in intracellular free calcium and relaxation. cGMP may also hyperpolarize the smooth muscle cell by activating potassium channels. Nitric oxide may also have direct calcium inhibitory effects, as well as direct hyperpolarizing actions via activation of potassium channels or the Na+,K+-ATPase. cAMP generated by adenylate cyclase (AC) from ATP in response to prostacyclin or other mediators may also regulate ion channels and intracellular calcium to mediate relaxation. (Other abbreviations are the same as in Fig 1.)

	Mechanism of Endothelium-Dependent Hyperpolarization

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
Endothelial Signal
Vascular smooth muscle cell hyperpolarization may be initiated by several autacoids and hormones in addition to acetylcholine, including histamine,⁷¹ bradykinin,^{43 68} substance P,^{72 73} calcitonin gene–related peptide,⁷³ ADP,⁷⁴ endothelin,^{75 76 77} and vasoactive intestinal polypeptide.⁷² The hyperpolarization of the smooth muscle usually precedes and is more transient than the accompanying relaxation induced by endothelium-dependent vasodilators.^{28 37 71 78} The mediator is believed to be a soluble transferable factor because of the success of some investigators in passing a hyperpolarizing substance between a vessel with endothelium to one that has been denuded of endothelium.^{79 80} The release of EDHF, like that of nitric oxide, is believed to be initiated by an increase in free calcium concentration in the endothelial cell. This conclusion is based on the ability of the calcium ionophore A23187 to cause endothelium-dependent hyperpolarization and of calcium-free solutions to make the hyperpolarizations to acetylcholine more transient and prevent the action of A23187.⁸¹ The transient response to acetylcholine in calcium-free solutions can be attributed to the release of intracellular endothelial calcium stores, whereas the entry of extracellular calcium into the endothelial cell prolongs the response to acetylcholine or bradykinin and accounts entirely for the response to A23187. Like the release of nitric oxide, endothelium-dependent hyperpolarization apparently depends on calcium calmodulin, because inhibitors of this enzyme, calmidazolium and fendiline, decrease the endothelium-dependent hyperpolarization and relaxations of the porcine coronary artery to bradykinin, which are resistant to inhibitors of nitric oxide.^{82 83}

Electrical Conduction
An important issue that remains unresolved is a potential electrical conduction of hyperpolarization between endothelial and smooth muscle cells,^{84 85} which might explain the inability to assay a soluble EDHF in arteries such as the porcine coronary artery.^{86 87} Although simultaneous hyperpolarization of endothelial and smooth muscle cells occurs in the porcine coronary artery in response to bradykinin, it could not be demonstrated by current injections that the two cell types are electrically connected or by lucifer yellow injections that gap junctions exist between the two cell types.^{85 88 89} These studies likely exclude, at least in this artery, the possibility that hyperpolarization is conducted between the endothelial and smooth muscle cells. The possibility of electrical coupling between endothelial and smooth muscle cells may be more likely in the microcirculation, where contacts between the two cell types are much more intimate.^{90 91}

Vasoconstrictors such as phenylephrine, U46619, and endothelin have been reported to depolarize the overlying endothelial cells.⁹² Hyperpolarization by acetylcholine reduces the endothelial depolarization induced by these vasoconstrictors. The depolarization of the endothelial cell by the contractile agonist does not occur if the endothelium is separated from the smooth muscle, even if the smooth muscle and endothelium are reapposed, suggesting electrical transmission or an extremely short-lived mediator.⁹² Direct electrical signaling between endothelial and smooth muscle cells, if it is shown to occur in some arteries, could explain the difficulty in identifying a distinct diffusable EDHF.

Endothelial Cell Membrane Potential
Membrane potential is as important a determinant of endothelial cell function as it is for smooth muscle cells. Endothelial cells also are hyperpolarized by agents that cause endothelium-dependent hyperpolarization,^{85 92 93 94} an event that has been attributed to the opening of calcium-dependent potassium channels on the endothelial cell membrane in response to the increase in cell calcium.^{95 96 97 98 99 100 101 102 103 104} Because the endothelial cell lacks typical voltage-dependent calcium channels,^{97 105 106} the influx of calcium into the hyperpolarized cell is dependent on and enhanced by the greater electrochemical gradient for the cation¹⁰⁷ (Fig 1). Although it is not known if shear stress regulates EDHF release, shear stress–induced release of nitric oxide from endothelial cells is inhibited by calcium-dependent potassium channel blockers, emphasizing the importance of these potassium channels¹⁰² to the release mechanism. Thus, endothelial cell hyperpolarization may enhance release of all endothelial cell autacoids whose liberation depends on calcium, including EDHF.^{93 108 109} Because endothelial cells release bradykinin,¹¹⁰ adenine nucleotides,^{111 112} and endothelin,¹¹³ these autocoids could be involved in autocrine regulation of nitric oxide and EDHF release. It is even possible that EDHF acts in an autocrine fashion, hyperpolarizing the endothelial cell, and that a significant physiological role of EDHF is to modulate the release of nitric oxide and other autacoids, such as prostaglandins. This might occur by potassium channel or calcium channel activation. For example, an endothelial cell cytochrome P-450 product that is a potential EDHF, 5,6-epoxyeicosatrienoic acid, increases endothelial cell calcium influx.¹¹⁴ Cytochrome P-450 inhibitors diminish the sustained endothelial cell calcium rise caused by histamine, suggesting that endogenous epoxyeicosatrienoic acids synthesized by endothelial cells promote calcium influx and therefore stimulate autacoid release.¹¹⁴

Response of Smooth Muscle Cells to EDHF
How Does Hyperpolarization of the Smooth Muscle Occur?
The hyperpolarization of the smooth muscle membrane mediated by the endothelium has been attributed to an increase in conductance to potassium ions. The evidence for this rests on the facts that (1) the cell membrane conductance of smooth muscle measured electrically during hyperpolarization is increased^{24 26 71} ; (2) the magnitude of endothelium-dependent hyperpolarization is decreased in a concentration-dependent manner by extracellular potassium⁷¹ ; (3) radioactive rubidium efflux is increased during hyperpolarization^{37 115} ; (4) the endothelium-dependent hyperpolarizations in certain cases have been prevented by potassium channel blockers^{27 80 116} ; and (5) many parallels can be drawn between the action of EDHF and pharmacological potassium channel openers, which cause relaxation primarily by causing hyperpolarization.¹¹⁷

The role of potassium channels has been proposed also because these channels, particularly large conductance calcium-dependent potassium channels, account for the passage of large amounts of ion current across the membrane, far greater than, for example, the Na⁺,K⁺-ATPase, and because the membrane potassium gradient accounts to the greatest extent for the resting membrane potential.¹¹⁸ It has been estimated that only a small increase in the opening rate of large conductance potassium channels is required to nearly maximally hyperpolarize the cell membrane.¹¹⁸ Nevertheless, other potential explanations exist that could contribute to endothelium-dependent hyperpolarization, including activation of the Na⁺,K⁺-ATPase⁷⁹ or inactivation of chloride channels.^{119 120} For example, taking the latter as a hypothetical example, vasoconstrictors increase smooth muscle cell intracellular calcium, which can lead to depolarization, in part due to the activation of calcium-activated chloride channels. Nitric oxide released from the endothelium decreases intracellular calcium in the smooth muscle.¹²¹ Thus, the decrease in calcium caused by nitric oxide could deactivate chloride channels and account for the hyperpolarizing effects of nitric oxide. This example points out that reduction in intracellular calcium induced by nitric oxide, or by another EDHF, could cause secondary changes in membrane potential that may contribute to the regulation of contractile tone. Thus, although experiments with nitric oxide synthase inhibitors suggest that endothelial factors other than nitric oxide can cause hyperpolarization, it is difficult to exclude membrane potential changes due to endothelium-derived nitric oxide when its synthesis is not blocked.

There are several important additional caveats of which to be aware when considering a potential regulation of membrane potential in mediating endothelium-dependent relaxation. First, there is only a certain range of membrane potentials over which an EDHF that operates by opening potassium channels might be expected to contribute significantly to the regulation of membrane potential. On the one hand, very negative membrane potentials are close to the equilibrium potential for potassium, and therefore opening of potassium channels would have little further effect. On the other hand, at depolarized potentials the conductance for potassium is already high, so the effect of further increasing the opening of channels may be negligible. Thus, just as changes in intracellular calcium could cause secondary changes in membrane potential, differences in resting membrane potential could explain differences in experimental results that have otherwise been attributed to the altered release or the absence of EDHF.

Second, caution must be exercised in interpreting experiments in which inhibitors of hyperpolarizing mechanisms are used, such as ouabain to block the Na⁺,K⁺-ATPase, glibenclamide to inhibit ATP-dependent potassium channels, or tetraethylammonium and charybdotoxin to block calcium-dependent potassium channels. Although these inhibitors may very well inhibit their intended target, they also may, simply by depolarizing the membrane, nonspecifically inhibit a response initiated by the mechanism that is the actual target of EDHF in the smooth muscle. In addition, many of the inhibitors have been used in concentrations that greatly exceed the range over which they are considered specific. For example, studies have been performed using glibenclamide to inhibit ATP-dependent potassium channels at concentrations exceeding the micromolar range that is more than 10 000 times higher than the concentration that inhibits the ATP-dependent channels in pancreatic ß cells.¹²² At micromolar concentrations, glibenclamide can inhibit responses mediated by calcium-activated potassium channels.^{123 124} Another factor that complicates interpretation of experiments is that both endothelial cells and smooth muscle cells possess a large complement of ion channels. None of the inhibitors are specific for one cell type or the other, making it difficult to determine if a pharmacological agent is affecting EDHF release from endothelial cells, its action on endothelial cells, or its action on the smooth muscle.

Third, contractile agents can have various independent effects on membrane potential or intracellular calcium, both of which can influence the response to an EDHF. To avoid movement of the preparation during contraction and the associated difficulty in maintaining cell impalements, many measurements of membrane potential in response to endothelium-dependent agonists have been made in the absence of contractile agents. Different membrane potential responses to endothelium-dependent agents have been recorded in the presence and absence of contractile agents,^{125 126} so caution is necessary in extrapolating data obtained in resting vascular smooth muscle as to how it relates to the mechanisms by which contracted smooth muscle relaxes. These factors emphasize the importance of future studies to determine the role of EDHF in vivo independent of exogenous contractile agents and at physiological membrane potentials.

How Does Hyperpolarization Cause Relaxation?
The mechanism by which hyperpolarization causes relaxation is controversial. The most direct and obvious explanation is that hyperpolarization of the smooth muscle cell membrane inhibits the opening of voltage-dependent calcium channels, allows calcium sequestration and removal mechanisms to lower intracellular calcium, and leads to relaxation.¹²⁷ Although this mechanism may operate in some blood vessels under some conditions, it does not fully explain the mechanism of relaxation. As has been pointed out previously,¹¹⁷ the isolated rabbit thoracic aorta is not depolarized during contractions induced by norepinephrine, and those contractions are poorly inhibited by blockers of voltage-dependent calcium channels such as nifedipine. Nevertheless, potassium channel activators, like chromakalim, whose relaxations are attributed solely to opening of potassium channels and hyperpolarization, induce potent relaxations of that preparation. Taken together, these observations suggest that potassium channel openers may relax by mechanisms other than inhibition of voltage-dependent calcium channels.¹¹⁷

This point was made more evident in an investigation of the bradykinin-induced relaxations of the porcine coronary artery that persist in the presence of nitric oxide synthesis inhibitors. The artery was treated with nifedipine before contracting the artery with the thromboxane mimetic, U46619.¹²⁸ Despite the fact that potassium could no longer contract the artery in the presence of nifedipine (indicating that voltage-dependent calcium channels were blocked), the potassium-induced depolarization inhibited the relaxations to bradykinin. This is consistent with the relaxations being mediated by a hyperpolarizing mechanism. However, in the absence of potassium, the bradykinin-induced relaxations of arteries treated with nifedipine were similar to those observed by others in the absence of nifedipine.^{30 68 129} This suggests that inhibition of voltage-dependent channels, at least of the L-type blocked by nifedipine, is not important in mediating relaxations associated with hyperpolarization that occur in the presence of nitric oxide synthase inhibitors and makes it likely that EDHF can induce relaxations by other mechanisms.

Potassium channel openers may relax by inhibiting vasoconstrictor-induced, phospholipase C–mediated inositol triphosphate production that would decrease intracellular calcium release.^{130 131 132} Also, possibly as a result of a reduction in diacylglycerol production by phospholipase C, there is a reduction in calcium sensitivity mediated by protein kinase C.^{133 134 135 136} These actions of potassium channel openers can all be attributed to hyperpolarization because they are prevented by elevated extracellular potassium and are blocked by glibenclamide.¹¹⁷ There are several lines of additional evidence that the potassium channel openers also may relax by mechanisms independent of the changes in membrane potential that they produce,¹¹⁷ possibly by enhancing the refilling of intracellular calcium stores.^{137 138} These observations regarding potassium channel openers have important implications for mechanisms attributed to an EDHF. First, relaxations caused by EDHF may differ depending on the mechanism by which the contractile agonist causes contraction. Contractions mediated by agonists that depolarize and activate voltage-dependent calcium channels may be affected differently by a hyperpolarizing stimulus than are those that do not. Thus, rather than indicating that different endothelial factors relax an artery contracted by two different agents, it may be that the difference is explained by the differences in the mechanisms of contraction.¹³⁹ Second, one should be aware that an EDHF could potentially relax by mechanisms other than those dependent on membrane hyperpolarization, and its effect on membrane potential may be an additional mechanism of relaxation or even an epiphenomenon.

	Nature of EDHF

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
Two Known EDHFs: Nitric Oxide and Prostacyclin
Nitric oxide and prostacyclin are two known endothelial products that can be released from the endothelium in sufficient quantities to cause relaxation and hyperpolarization at least in some blood vessels and thus fulfill the criteria to be called EDHFs. Nitric oxide and prostacylin are distinguishable from a putative distinct EDHF by having their hyperpolarizing and relaxing actions annulled by competitive inhibitors of nitric oxide synthase or inhibitors of cyclooxygenase, respectively.

Prostacyclin
Prostacyclin contributes to endothelium-dependent relaxation of several isolated blood vessels and vasodilation of perfused organs.^{140 141 142 143 144 145 146 147} The vasodilation of the rabbit coronary circulation by prostacyclin is blocked by glibenclamide,¹⁴⁸ suggesting an underlying opening of potassium channels as the mechanism. Prostacyclin causes relaxation by increasing cAMP in smooth muscle cells. Prostacylin has also been shown to hyperpolarize cultured canine carotid artery smooth muscle cells,¹⁴⁹ and a prostacylin analogue hyperpolarizes the isolated guinea pig coronary artery.¹⁵⁰ In the latter study, cAMP also hyperpolarized the smooth muscle, suggesting that hyperpolarization may occur via cyclic nucleotide-dependent protein kinases, which are known to modulate potassium channels. Likewise, in the canine saphenous vein denuded of endothelium, other activators of adenylate cyclase induce hyperpolarization by opening ATP-dependent potassium channels.¹⁵¹

Nitric Oxide
Nitric oxide in solution or nitrovasodilators can hyperpolarize smooth muscle of several different isolated blood vessels, including the small mesenteric artery¹²⁵ and aorta¹⁵² of the rat, the uterine¹²⁶ and coronary artery of the guinea pig,¹⁵⁰ and the basilar artery of the rabbit.⁷⁸ Nitric oxide in concentrations exceeding 5x10^-7 mol/L hyperpolarized all vessels studied, including various arteries of rat, rabbit, dog, and guinea pig.¹²⁶ In some arteries, nitric oxide caused relaxation at lower concentrations without hyperpolarization, but hyperpolarization was always observed at higher concentrations. In the present study, it was found to be necessary to depolarize and contract the arteries with phenylephrine to observe hyperpolarization to nitric oxide. The failure to observe a hyperpolarizing response to nitric oxide in resting arteries was attributed to a highly negative resting membrane potential. Stretching the guinea pig coronary artery was found to be necessary to observe nitric oxide-induced hyperpolarizations.¹⁵⁰ In contrast, nitric oxide hyperpolarized the small mesenteric artery of the rat only in the absence of contractile force.¹²⁵ These differences may, as discussed previously, depend on differences in resting membrane potential in the various preparations or on other experimental conditions. Although the concentrations of nitric oxide in vivo are not known, the concentrations that caused hyperpolarization in several of the above studies^{125 126 150} were in the micromolar range, which is equal to the concentration of nitric oxide measured in rabbit aortic medial smooth muscle after stimulation of the endothelium by acetylcholine.¹⁵³

It was also demonstrated that the hyperpolarization to acetylcholine in the guinea pig artery was reduced by 41% by a relatively low concentration of N^G-monomethyl-L-arginine.¹²⁶ Another nitric oxide synthase inhibitor, N^G-L-arginine methyl ester, as well as N^G-monomethyl-L-arginine, had no effect on peak acetylcholine-induced hyperpolarization in the rat aorta but made the response more transient, reducing the more sustained hyperpolarization by about 50%.¹⁵² These studies suggest that at least under some circumstances, the release of nitric oxide accounts for part of the hyperpolarizing response to endothelium-dependent vasodilators.

Even in the absence of an effect of inhibitors of nitric oxide on the hyperpolarization stimulated by endothelium-dependent vasodilators, hemoglobin,^{29 37} methylene blue,^{71 126} or nitric oxide synthase inhibitors^{152 154} can depolarize resting arteries. This suggests that nitric oxide tonically released from the endothelium can under certain circumstances hyperpolarize the smooth muscle cell. Despite these findings, removal of the endothelium does not usually depolarize the underlying smooth muscle cells, as one would expect by removing a source of hyperpolarizing factor.^{27 28 29 43 71 77 80 81 151} This is perhaps due to the physical trauma to the blood vessel associated with removal of the endothelium or to the simultaneous removal of depolarizing factors released by the endothelium.

One mechanism by which nitric oxide could hyperpolarize smooth muscle is by increasing cGMP, which in turn causes protein kinase–dependent activation of calcium-dependent potassium channels.^{67 155 156} In addition, ATP-dependent potassium channels appear to be activated by nitric oxide, through a cGMP-dependent mechanism.¹⁵⁷ All endothelium-dependent relaxations and hyperpolarizations have been attributed to cGMP-dependent effects on potassium channels.¹⁵⁸ This is clearly not the case, as demonstrated by endothelium-dependent relaxations that persist after blocking the rise in cGMP.^{30 32 37} Some investigators have relied on the ability of methylene blue to block cGMP,^{27 28 71 74 126 158 159} without accounting for its ability to scavenge nitric oxide,⁵⁷ as well as to inhibit nitric oxide synthesis by the endothelium.¹⁶⁰

Nitric oxide in concentrations from 5x10^-7 to 10^-5 mol/L increases the activity of single calcium-dependent potassium channels in isolated membrane patches of rabbit aortic smooth muscle cells.⁷⁰ Because the membrane patches in these experiments were devoid of all nucleotides and cofactors necessary for cGMP generation, this observation provides a mechanism whereby nitric oxide could hyperpolarize smooth muscle cells independent of cGMP. Relaxations to nitric oxide of the rabbit aorta were incompletely blocked by methylene blue, which effectively prevented the increase in cGMP, and the residual relaxations were blocked by charybdotoxin.⁷⁰ This suggests that cGMP-independent potassium channel activation by nitric oxide can cause relaxation. These effects of exogenous nitric oxide were reproduced by native EDRF. In this case, the effluent of a perfused rabbit aorta whose intact endothelium was stimulated with acetylcholine also was shown to activate single potassium channels in a membrane patch.⁷⁰ The cGMP-independent activation of potassium channels was apparently dependent on a chemical reaction of nitric oxide with protein sulfhydryl groups, because blockade of these constituents prevented the channel activation. This mechanism provides a molecular mechanism independent of cGMP for nitric oxide to act as an EDHF.⁷⁰

Tetraethylammonium or charybdotoxin depolarizes and constricts cerebral microvessels of the rabbit.¹⁶¹ Because the constrictions are blocked by calcium channel blockers or by lowering intravascular pressure, hyperpolarization mediated by calcium-dependent potassium channels may provide an important physiological brake to pressure-induced, calcium-dependent myogenic tone. It is possible that nitric oxide released tonically from the endothelium can regulate membrane potential via a potassium channel–dependent mechanism and account for these observations.

Another potential hyperpolarizing action of nitric oxide is its ability to stimulate the Na⁺,K⁺-ATPase pump. In the rabbit aorta, removal of the endothelium or inhibition of nitric oxide synthesis decreased ouabain-sensitive radioactive rubidium influx used as a marker of pump activity,¹⁶² and sodium nitroprusside stimulates pump activity.¹⁶³ The stimulation of pump activity by sodium nitroprusside appears to be independent of cGMP and attributable to Na⁺/H⁺ exchange.¹⁶³ An endothelium-dependent increase in Na⁺,K⁺-ATPase activity might help to explain the depolarizing effects of inhibitors of nitric oxide under resting conditions. In addition, the sodium pump inhibitor ouabain decreases endothelium-dependent hyperpolarization caused by acetylcholine in the coronary artery of the dog⁷⁹ and the middle cerebral artery of the cat.⁷⁴

Nonprostanoid, Non–Nitric Oxide Endothelium-Derived Hyperpolarizing and Relaxing Factor(s)
Despite the above discussion of the potential role of prostacyclin and nitric oxide as EDHFs, the evidence that another nonprostanoid, non–nitric oxide factor is an EDHF is overwhelming. In many arteries, blockade of prostaglandin synthesis with, for example, indomethacin has little effect on endothelium-dependent relaxation or hyperpolarization.^{37 74 79 80} In addition, inhibition of the L-arginine–nitric oxide pathway with arginine analogues or with hemoglobin or methylene blue has no or only minimal inhibitory effects on endothelium-dependent relaxations in many blood vessels, including the pig coronary artery^{30 42 68 128 129} ; rat small mesenteric artery^{125 164} ; guinea pig uterine¹²⁶ and coronary artery¹⁵⁰ ; dog mesenteric²⁸ and coronary artery³⁸ ; cat middle cerebral artery⁷⁴ ; rabbit femoral artery,²⁹ carotid artery, and abdominal aorta³² ; and dog femoral vein.¹⁵⁹ Also, in several of the studies,^{30 32 38} relaxations were resistant to combined inhibition of prostaglandins and nitric oxide, excluding participation by one factor in the absence of the other. Endothelium-dependent relaxations under these conditions are associated with hyperpolarization as indicated by changes in membrane potential^{28 29 68 71 74 78 80 125 126 152 164} and rubidium efflux.^{37 115} In the porcine coronary artery^{30 128} and rabbit carotid artery and abdominal aorta,³² relaxations resistant to prostanoid and nitric oxide inhibitors are prevented by elevated extracellular potassium, and in the rabbit arteries,³² but not in the pig coronary artery,^{42 128} blockade by charybdotoxin implicates the role of calcium-dependent potassium channels. Tetraethylammonium, a less specific inhibitor of calcium-dependent potassium channels, but not the ATP-dependent channel antagonist glibenclamide, inhibits endothelium-dependent hyperpolarization to acetylcholine of the rat aorta¹⁶⁵ and guinea pig coronary artery.⁸⁰ It should be noted that tetraethylammonium has been reported to be a muscarinic receptor antagonist and could block the action of acetylcholine by that mechanism.¹⁶⁶ Glibenclamide, at a concentration of 3x10^-6 mol/L, inhibits the endothelium-dependent hyperpolarization to acetylcholine in the rabbit cerebral artery, and the antagonism is accentuated by blockade of nitric oxide with methylene blue.²⁷ In the isolated, perfused rat mesentery, apamin, an inhibitor of small conductance potassium channels, blocks vasodilation after treatment with a nitric oxide synthase inhibitor.⁴¹ These studies all indicate that a nonprostaglandin, non–nitric oxide mediator of relaxation and hyperpolarization exists in some isolated arteries whose effects are inhibited by either elevated extracellular potassium or potassium channel blockers. Regardless of the identity of the mediator in these studies of isolated arteries or vascular beds, its importance is suggested by vasodilator responses in in vivo human and animal studies that are resistant to the inhibitors of both nitric oxide and prostacyclin.^{9 33 34 41}

Possible Identities of Nonprostanoid, Non–Nitric Oxide EDHFs
Hydrogen peroxide relaxes arterial smooth muscle by a cGMP-dependent mechanism,^{167 168 169 170} activates potassium channels,^{171 172} and causes hyperpolarization.¹⁷³ Beny and vonderWeid¹⁷³ showed that catalase that effectively blocked the hyperpolarization to hydrogen peroxide does not inhibit the endothelium-dependent hyperpolarization of the porcine coronary artery to bradykinin, excluding a role for hydrogen peroxide in the response. However, hydrogen peroxide has been shown to be a product of nitric oxide synthase when arginine utilization by the enzyme is prevented¹⁷⁴ and therefore remains a candidate EDHF, perhaps in pathological states, where arginine availability is limited or blocked.¹⁷⁵ Superoxide anion is also generated by endothelial cells, may be another product of nitric oxide synthase,¹⁷⁶ and, in distinction to its role as an inactivator of nitric oxide, has been implicated in causing endothelium-dependent relaxation.¹⁷⁷ Although this action is unlikely to occur directly, superoxide anion can react with nitric oxide to form peroxynitrite,¹⁷⁸ which can relax smooth muscle, or, alternatively, superoxide anion chemically degenerates to hydrogen peroxide, which could be the active agent.

Nonprostanoid products of the metabolism of arachidonic acid remain particularly appealing candidates as EDHFs because as products of phospholipase A₂ they would be expected to be released in response to elevated endothelial cell calcium levels. Phospholipase A₂ inhibitors such as mepacrine and lipoxygenase inhibitors such as nordihydroguaiaretic acid and BW755c were some of the earliest recognized inhibitors of endothelium-dependent relaxations.^{1 142 179 180 181 182 183 184 185} These actions have been attributed to nonspecific actions of the drugs in some cases¹⁸⁶ ; however, lipoxygenase products are made in abundance by endothelial cells,^{187 188 189} and some can activate potassium channels,^{190 191} maintaining their candidacy as EDHFs.

Products of cytochrome P-450 are also produced by endothelial cells^{192 193} and can activate potassium channels,¹⁹⁴ and inhibitors of cytochrome P-450 have been reported to inhibit endothelium-dependent relaxations,^{195 196} making this third class of products of arachidonic acid candidate EDHFs as well.^{45 47} Relaxations of isolated bovine and porcine coronary arteries¹⁹⁷ and vasodilatation of the coronary circulation of the rat⁸⁷ in the presence of N^G-nitro-L-arginine are prevented by SKF 525a or clortrimazole, inhibitors of cytochrome P-450, and by tetrabutylammonium or apamin. 11,12-Epoxyeicosatetraenoic acid, a cytochrome P-450 metabolite, relaxes the isolated coronary arteries, and these were blocked by tetrabutylammonium, further suggesting that cytochrome P-450 products could serve as EDHFs. However, in these and previous studies,⁸⁶ a diffusable, bioassayable vasodilator released from endothelial cells could not be demonstrated after blockade of nitric oxide synthesis. This leaves open the question of whether products of cytochrome P-450 are released in sufficient quantity by endothelial cells to account for relaxation and hyperpolarization, which is resistant to nitric oxide synthase inhibitors. Furthermore, inhibitors of cytochrome P-450 are notoriously nonspecific. SKF 525a and its analogues are potent direct inhibitors of calcium-dependent potassium channels, and therefore they could block the actions of another EDHF on endothelial and smooth muscle cells as well as the synthesis of the products of cytochrome P-450_.¹⁹⁸

	How Important Is Endothelium-Dependent Hyperpolarization to Relaxation Caused by Endothelium-Dependent Vasodilators?

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
In situations in which endothelium-dependent relaxations resistant to prostanoid and nitric oxide synthesis inhibitors are blocked by potassium channel antagonists or elevated potassium, hyperpolarization may be implicated in causing the relaxation. However, some studies suggest that the degree to which hyperpolarization contributes to relaxation may be small or nonexistent. Although acetylcholine and another muscarinic agonist, oxotremorine, cause endothelium-dependent relaxation, only acetylcholine causes hyperpolarization.¹⁹⁹ Repeated administration of acetylcholine results in the desensitization and disappearance of hyperpolarization but has no effect on relaxation of the rabbit basilar artery, suggesting that the hyperpolarization is superfluous to relaxation.⁷⁸ Hemoglobin does not significantly affect acetylcholine-induced hyperpolarization of the rabbit femoral artery but effectively inhibits relaxation, and hyperpolarization cannot be implicated to account for more than 23% of acetylcholine-induced relaxation.²⁹ An estimate of a 20% to 30% contribution of hyperpolarization to relaxation also was made by comparing the hyperpolarization and relaxation induced in the rat aorta by acetylcholine and a potassium channel opener, BRL38227.¹⁶⁵ Although nitric oxide inhibitors reduce the sustained hyperpolarization of the rat aorta in response to acetylcholine by only 50%,¹⁵² the endothelium-dependent relaxation of this blood vessel is nearly completely blocked by these inhibitors.³⁵ In the coronary artery of the pig,³⁰ the carotid artery and abdominal aorta of the rabbit,³² or the perfused rat mesentery,⁴¹ 25 to 30 mmol/L extracellular potassium has little effect on relaxation or vasodilation to bradykinin or acetylcholine. Only after inhibiting nitric oxide synthesis does the same concentration of potassium produce an effective blockade. Also, potassium channel antagonists have little effect on endothelium-dependent relaxation unless nitric oxide synthesis is first inhibited.^{27 32} These results suggest that endothelium-dependent hyperpolarization may back up or enhance the relaxing action of nitric oxide but that it is not an important primary mediator of endothelium-dependent relaxation in most normal blood vessels. Although studies on perfused rat mesentery⁴¹ appear to include resistance arteries in this conclusion, it should be noted that relatively few studies of EDHF have been made in resistance arteries, leaving this question open. It is possible that nitric oxide inhibits the formation or action of EDHF. For example, if EDHF is a cytochrome P-450 metabolite, nitric oxide might inhibit its formation by binding to the heme group of the P-450 reductase. This could explain why the hyperpolarizing mechanism appears to take over when nitric oxide is inhibited.

	Is the Role of EDHF Altered in Vascular Disease?

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
It is important to note that the supremacy of the nitric oxide–cGMP mode of endothelium-dependent relaxation is based on studies of normal blood vessels in which endothelium-derived nitric oxide normally increases cGMP, which can relax by multiple calcium regulatory mechanisms not involving membrane potential. Some disease states may interfere with nitric oxide by either limiting its production or, in cases where oxygen-derived free radicals are increased, limiting nitric oxide access to or activation of guanylate cyclase. Decreased endothelium-dependent, presumed nitric oxide–induced cGMP production has been noted in hypercholesterolemia,^{200 201 202 203} hypertension,^{204 205} and diabetes mellitus.^{206 207} If under these circumstances a distinct EDHF mechanism were spared, it may take on greater importance.

Hypercholesterolemia
Many studies, including those in patients, have demonstrated impaired endothelium-dependent vasodilation due to hypercholesterolemia.^{200 201 202 203 208 209} However, in hypercholesterolemic rabbit carotid arteries, acetylcholine-induced relaxations are normal.²⁰⁰ This is despite the fact that cGMP accumulation in response to acetylcholine is absent.²⁰⁰ Possibly as a result of the interruption in the nitric oxide–cGMP signaling pathway, the acetylcholine-induced relaxations, unlike those in normal arteries, are inhibited by increased extracellular potassium or charybdotoxin.²⁰⁰ These observations suggest that acetylcholine-induced relaxations of hypercholesterolemic rabbit carotid arteries are mediated differently than those in normal arteries and are subserved to a greater degree by a potassium channel–dependent hyperpolarizing mechanism. This does not necessarily indicate that a distinct hyperpolarizing factor mediates the acetylcholine-induced relaxation because unlike in normal arteries, nitric oxide–induced relaxations are also inhibited by charybdotoxin in the hypercholesterolemic arteries.²¹⁰ Also, inhibitors of nitric oxide synthesis decrease the acetylcholine-induced relaxations in the hypercholesterolemic rabbit arteries to an even greater extent than in normal arteries,²⁰⁰ suggesting a role of nitric oxide in the cGMP-independent, potassium channel–dependent, acetylcholine-induced relaxation.⁷⁰ These results emphasize the existence of two primary mechanisms of endothelium-dependent relaxation: one mediated primarily by cGMP, and the other by a mediator(s) of smooth muscle hyperpolarization, which may be of major importance only when the nitric oxide–cGMP–dependent pathway of relaxation is impaired, as it appears to be by hypercholesterolemia. The alternate pathway may maintain vasodilation despite inhibition of what is normally the primary mechanism.

Hypertension
Endothelium-dependent relaxation and hyperpolarization to acetylcholine in the aorta of two-kidney, one clip renal hypertensive rats is decreased compared with that of normal rats.¹⁶⁵ The resting smooth muscle cell membrane is also more depolarized in the hypertensive rat aorta. Tetraethylammonium inhibits the relaxation and hyperpolarization to acetylcholine in normal rats but does not inhibit further the relaxation of the hypertensive artery. Glibenclamide has no effect, suggesting that tetraethylammonium acts on calcium-dependent potassium channels. This suggests that a calcium-dependent potassium channel on endothelial or smooth muscle cells is important in mediating endothelium-dependent smooth muscle cell hyperpolarization and that this mechanism is impaired in hypertension. A similar decrease in endothelium-dependent hyperpolarization has been described in the mesenteric artery of the spontaneously hypertensive rat.³⁹

Diabetes Mellitus
Unlike in normal arteries where the presence of endothelium causes a twofold increase in arterial Na⁺,K⁺-ATPase activity, the pump activity in diabetic rabbit aorta and carotid arteries is the same in the presence and absence of the endothelium.^{162 211} The sodium pump activity measured in diabetic arteries with endothelium is similar to that of normal arteries treated with nitric oxide inhibitors or denuded of endothelium. This suggests that a hyperpolarizing mechanism present in normal arteries, which is accounted for by an enhancement of sodium pump activity by endothelium-derived nitric oxide, is impaired in diabetes.

	Future Directions for Research

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 
Many questions remain open to further investigation of the mechanisms of endothelium-dependent vasodilatation. Perhaps it has been no better demonstrated than it has for nitric oxide that the study of any physiological mechanism is greatly enhanced once the mediator is identified and specific pharmacological inhibitors are produced. Thus, a primary goal of future work on EDHF will be to identify the mediator(s) that are released from endothelial cells in sufficient quantities to cause hyperpolarization of the smooth muscle, as well as to find corresponding inhibitors. An important question relates to the relevance of the endothelial cell hyperpolarization initiated either by an increase in endothelial intracellular calcium, increasing activity of calcium-dependent potassium channels, or by an autocrine mediator, to subsequent endothelium-dependent relaxation. Similarly, the existence of electrical signaling between endothelial cells and smooth muscle cells would be very important to delineate in further studies. Another area that requires study is the intracellular signaling, which leads to hyperpolarization of both smooth muscle and endothelial cells. These studies would necessarily address interrelationships that might exist between cGMP-dependent and -independent mechanisms of ion channel regulation and relaxation. This would also require a greater understanding of the regulation of membrane potential by ion channels and transporters and, in turn, the regulation of contractile mechanisms by membrane potential. Finally, not only may it have important therapeutic implications to delineate the role of endothelium-dependent hyperpolarization in vascular diseases, but also disease remains an experiment of nature, the study of which can lead to greater understanding of normal physiology.

Received April 10, 1995; revision received June 21, 1995; accepted July 24, 1995.

	References

Top Introduction Mechanism of Endothelium... Nature of EDHF How Important Is Endothelium... Is the Role of... Future Directions for Research... References

 

1. Furchgott RF, Zawadzki JV. The obligatory role of the endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature. 1980;288:373-376. [Medline] [Order article via Infotrieve]
   
2. Furchgott RF, Vanhoutte PM. Endothelium-derived relaxing and contracting factors. FASEB J. 1989;3:2007-2018. [Abstract]
   
3. Luscher TF, Vanhoutte PM. The Endothelium: Modulator of Cardiovascular Function. Boca Raton, Fla: CRC Press; 1990.
   
4. Busse R, Mulsch A. Calcium-dependent nitric oxide synthesis in endothelial cytosol is mediated by calmodulin. FEBS Lett. 1990;265:133-136. [Medline] [Order article via Infotrieve]
   
5. Rubanyi GM, Schwartz A, Vanhoutte PM. Calcium transport mechanisms in endothelial cells regulating the synthesis and release of endothelium-derived relaxing factor. In: Vanhoutte PM, ed. Relaxing and Contracting Factors. Clifton, NJ: The Humana Press; 1988:179-188.
   
6. Nakane M, Mitchell J, Forstermann W, Murad F. Phosphorylation by calcium calmodulin-dependent protein kinase II and protein kinase C modulates the activity of nitric oxide synthase. Biochem Biophys Res Commun. 1991;180:1396-1402. [Medline] [Order article via Infotrieve]
   
7. Lopez-Jaramill P, Gonzalez MC, Palmer RMJ, Moncada S. The crucial role of physiological Ca²⁺ concentrations in the production of endothelial nitric oxide and the control of vascular tone. Br J Pharmacol. 1990;101:489-493. [Medline] [Order article via Infotrieve]
   
8. Amezcua JL, Palmer RMJ, de Souza BM, Moncada S. Nitric oxide synthesized from L-arginine regulates vascular tone in the coronary circulation of the rabbit. Br J Pharmacol. 1989;97:1119-1124. [Medline] [Order article via Infotrieve]
   
9. Vallance P, Collier JG, Moncada S. Effect of endothelium-derived nitric oxide on peripheral arteriolar tone in man. Lancet. 1989;2:997-1000. [Medline] [Order article via Infotrieve]
   
10. Persson MG, Gustafsson LE, Wiklund NP, Hedqvist P, Moncada S. Endogenous nitric oxide as a modulator of rabbit skeletal muscle microcirculation in vivo. Br J Pharmacol. 1990;100:436-466.
    
11. Wiklund NP, Persson MG, Gustafsson LE, Moncada S, Hedqvist P. Modulatory role of endogenous nitric oxide in pulmonary circulation in vivo. Eur J Pharmacol. 1990;185:123-124. [Medline] [Order article via Infotrieve]
    
12. van Gelderen EM, Den Boer MO, Saxena PR. N^G-Nitro-L-arginine methyl ester: systemic and pulmonary haemodynamics, tissue blood flow and arteriovenous shunting in the pig. Naunyn-Schmiedebergs Arch Pharmacol. 1993;348:417-423.
    
13. Beierwaltes WH, Sigmon DH, Carretero OA. Endothelium modulates renal blood flow but not autoregulation. Am J Physiol. 1992;262:F943-F949. [Abstract/Free Full Text]
    
14. Ribeiro MO, Antunes E, deNucci G, Lovisolo SM, Zatz R. Chronic inhibition of nitric oxide synthesis. Hypertension. 1992;20:298-303. [Abstract/Free Full Text]
    
15. Rees DD, Palmer RMJ, Moncada S. Role of endothelium-derived nitric oxide in the regulation of blood pressure. Proc Natl Acad Sci U S A. 1989;86:3375-3378. [Abstract/Free Full Text]
    
16. Lahera V, Salom MG, Miranda-Guardiola F, Moncada S, Romero JC. Effects of N^G-nitro-L-arginine methyl ester on renal function and blood pressure. Am J Physiol. 1991;261:F1033-F1037. [Abstract/Free Full Text]
    
17. Shepherd JT, Vanhoutte PM. The Human Cardiovascular System: Facts and Concepts. New York, NY: Raven Press; 1979.
    
18. Moncada S, Herman AG, Vanhoutte PM. Endothelium-derived relaxing factor is identified as nitric oxide. Trends Pharmacol Sci. 1987;8:365-368.
    
19. Palmer RMJ, Ashton DS, Moncada S. Vascular endothelial cells synthesize nitric oxide from L-arginine. Nature. 1988;333:664-666. [Medline] [Order article via Infotrieve]
    
20. Feelisch M, te Poel M, Zamora R, Deussen A, Moncada S. Understanding the controversy over the identity of EDRF. Nature. 1994;368:62-65. [Medline] [Order article via Infotrieve]
    
21. Funaki S, Bohr DV. Electrical and mechanical activity of isolated vascular smooth muscle of the rat. Nature. 1967;203:192-194.
    
22. Funaki S. Electrical and mechanical activity of isolated smooth muscle from the portal vein of the rat. Bibl Anat. 1966;8:5-10.
    
23. Nakajima A, Horn L. Electrical activity of single vascular smooth muscle fibers. Am J Physiol. 1967;213:25-30.
    
24. Kuriyama H, Suzuki H. The effects of acetylcholine on the membrane and contractile properties of smooth muscle cells of the rabbit superior mesenteric artery. Br J Pharmacol. 1978;64:493-501. [Medline] [Order article via Infotrieve]
    
25. Kitamura K, Kuriyama H. Effects of acetylcholine on the smooth muscle cells of isolated main coronary artery of the guinea-pig. J Physiol. 1979;293:119-133. [Abstract/Free Full Text]
    
26. Bolton TB, Lang RJ, Takewaki T. Mechanism of action of noradrenaline and carbachol on smooth muscle guinea-pig anterior mesenteric artery. J Physiol. 1984;351:549-572. [Abstract/Free Full Text]
    
27. Brayden JE. Membrane hyperpolarization is a mechanism of endothelium-dependent cerebral vasodilation. Am J Physiol. 1990;259:H668-H673. [Abstract/Free Full Text]
    
28. Komori K, Lorenz RR, Vanhoutte PM. Nitric oxide, ACh, and electrical and mechanical properties of canine arterial smooth muscle. Am J Physiol. 1988;255:H207-H212. [Abstract/Free Full Text]
    
29. Huang AH, Busse R, Bassenge E. Endothelium-dependent hyperpolarization of smooth muscle cells in rabbit femoral arteries is not mediated by EDRF (nitric oxide). Naunyn-Schmiedebergs Arch Pharmacol. 1988;338:438-442.
    
30. Cowan CL, Cohen RA. Two mechanisms mediate relaxation by bradykinin of pig coronary artery: NO-dependent and independent responses. Am J Physiol. 1991;261:H830-H835. [Abstract/Free Full Text]
    
31. Cowan CL, Cohen RA. Different mechanisms of relaxation of pig coronary artery to bradykinin and chromakalim are distinguished by potassium channel blockers. J Pharmacol Exp Ther. 1992;260:248-253. [Abstract/Free Full Text]
    
32. Cowan CL, Palacino JJ, Najibi S, Cohen RA. Potassium channel-mediated relaxation to acetylcholine in rabbit arteries. J Pharmacol Exp Ther. 1993;266:1482-1489. [Abstract/Free Full Text]
    
33. Chowienczyk PJ, Cockcroft JR, Ritter JM. Differential inhibition by N^G-monomethyl-L-arginine of vasodilator effect of acetylcholine and methacholine in human forearm vasculature. Br J Pharmacol. 1003;110:736-738. [Medline] [Order article via Infotrieve]
    
34. Mugge A, Lopez JAG, Piegors DJ, Breese KR, Heistad DD. Acetylcholine-induced vasodilatation in rabbit hindlimb in vivo is not inhibited by analogues of L-arginine. Am J Physiol. 1991;260:H242-H247. [Abstract/Free Full Text]
    
35. Rees DD, Palmer RMJ, Hodson HF, Moncada S. A specific inhibitor of nitric oxide formation from L-arginine attenuates endothelium-dependent relaxation. Br J Pharmacol. 1989;96:418-424. [Medline] [Order article via Infotrieve]
    
36. Nagao T, Illiano S, Vanhoutte PM. Heterogeneous distribution of endothelium-dependent relaxations resistant to N^G-nitro-L-arginine in rats. Am J Physiol. 1992;263:H1090-H10994. [Abstract/Free Full Text]
    
37. Chen G, Suzuki H, Weston AH. Acetylcholine releases endothelium-derived hyperpolarizing factor and EDRF from rat blood vessels. Br J Pharmacol. 1988;95:1165-1174. [Medline] [Order article via Infotrieve]
    
38. Mombouli JV, Illiano S, Nagao T, Scott-Burden T, Vanhoutte PM. Potentiation of endothelium-dependent relaxations to bradykinin by angiotensin I converting enzyme inhibitors in canine coronary artery involves both endothelium-derived relaxing and hyperpolarizing factors. Circ Res. 1992;71:137-144. [Abstract/Free Full Text]
    
39. Fujii K, Tominaga M, Ohmori S, Kobayashi K, Koga T, Takata Y, Fumjishima M. Decreased endothelium-dependent hyperpolarization to acetylcholine in smooth muscle of the mesenteric artery of spontaneously hypertensive rats. Circ Res. 1992;70:660-669. [Abstract/Free Full Text]
    
40. Nagao T, Vanhoutte PM. Hyperpolarization as a mechanism for endothelium-dependent relaxations in the porcine coronary artery. J Physiol. 1992;445:355-367. [Abstract/Free Full Text]
    
41. Adeagbo ASO, Triggle CR. Varying extracellular [K⁺]: a functional approach to separating EDHF- and EDNO-related mechanisms in perfused rat mesenteric arterial bed. J Cardiovasc Pharmacol. 1993;21:423-429. [Medline] [Order article via Infotrieve]
    
42. Nagao T, Vanhoutte PM. Characterization of endothelium-dependent relaxations resistant to nitro-L-arginine in the porcine coronary artery. Br J Pharmacol. 1992;107:1102-1107. [Medline] [Order article via Infotrieve]
    
43. Nakashima M, Mombouli JV, Taylor AA, Vanhoutte PM. Endothelium-dependent hyperpolarization caused by bradykinin in human coronary arteries. J Clin Invest. 1993;92:2867-2871.
    
44. Holzmann S, Kikovetz WR, Windischhofer W, Paschke E, Graier WF. Pharmacologic differentiation between endothelium-dependent relaxations sensitive and resistant to nitro-L-arginine in coronary arteries. J Cardiovasc Pharmacol. 1994;23:747-756. [Medline] [Order article via Infotrieve]
    
45. Vanhoutte PM. The end of the quest? Nature. 1987;327:459-460. [Medline] [Order article via Infotrieve]
    
46. Vanhoutte PM. Other endothelium-derived vasoactive factors. Circulation. 1993;87(suppl V):V-9-V-17.
    
47. Komori K, Vanhoutte PM. Endothelium-derived hyperpolarizing factor. Blood Vessels. 1990;27:238-245. [Medline] [Order article via Infotrieve]
    
48. Taylor SG, Weston AH. Endothelium-derived hyperpolarizing factor: a new endogenous inhibitor from the vascular endothelium. Trends Pharmacol Sci. 1988;9:272-274. [Medline] [Order article via Infotrieve]
    
49. Feletou M, Vanhoutte PM. Endothelium-derived factor(s) and membrane potential of vascular smooth muscle. Blood Vessels. 1986;23:68.
    
50. Nagao T, Vanhoutte PM. Endothelium-derived hyperpolarizing factor and endothelium-dependent relaxation. Am J Respir Cell Mol Biol. 1993;8:1-6.
    
51. Ignarro LJ. Heme-dependent activation of soluble guanylate cyclase by nitric oxide: regulation of enzyme activity by porphyrins and metalloporphyrins. Semin Hematol. 1989;26:63-76. [Medline] [Order article via Infotrieve]
    
52. Martin W, Villani GM, Jothianandan D, Furchgott RF. Blockade of endothelium-dependent and glyceryl trinitrate-induced relaxation of rabbit aorta by certain ferrous hemoproteins. J Pharmacol Exp Ther. 1985;233:679-685. [Abstract/Free Full Text]
    
53. Gruetter CA, Gruetter DY, Lyon JE, Kadowitz PJ, Ignarro J. Relationship between cyclic guanosine 3':5'-monophosphate formation and relaxation of coronary arterial smooth muscle by glyceryl trinitrate, nitroprusside, nitrite and nitric oxide: effects of methylene blue and methemoglobin. J Pharmacol Exp Ther. 1981;219:181-186. [Abstract/Free Full Text]
    
54. Ignarro LJ, Byrns RE, Buga GM, Wood KS. Endothelium-derived relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical properties identical to those of nitric oxide radical. Circ Res. 1987;61:866-879.