This Article Full Text (PDF) 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. Search for Related Content PubMed PubMed Citation Articles by Cohen, R. A. Related Collections Cardiovascular Pharmacology Animal models of human disease Endothelium/vascular type/nitric oxide Other Vascular biology Related Article

(Circulation. 2005;111:724-727.)
© 2005 American Heart Association, Inc.

Editorial

The Endothelium-Derived Hyperpolarizing Factor Puzzle

A Mechanism Without a Mediator?

Richard A. Cohen, MD

From the Vascular Biology Unit, Department of Medicine, Boston University School of Medicine, Boston, Mass.

Correspondence to Richard A. Cohen, MD, Director, Vascular Biology Unit, X708, Dept of Medicine, Boston University School of Medicine, 650 Albany St, Boston, MA 02118. E-mail racohen{at}bu.edu

Key Words: Editorials • endothelium • vasodilation

	Introduction

Top Introduction What Is EDHF and... How Does EDHF Relax... Where Do We Go... References

 
The article by Scotland et al¹ published in this issue of Circulation reports on a unique mouse model that might help to identify a longstanding, intensely investigated, and controversial substance called endothelium-derived hyperpolarizing factor (EDHF). EDHF is a "holy grail," a "third factor" that provides local control of vascular homeostasis in addition to nitric oxide (NO) and prostacyclin. The discovery of the latter substances, which are known to cause both hyperpolarization and relaxation, led to Nobel prizes. But beware acronyms! The term EDHF is >15 years old, and as yet no one substance and no one mechanism has been proposed that explains all of its features. The story of the investigation has been a meandering one, with many stops along the way. Each stop has taught us how blood vessels function, but there is more to be learned if we are to have a complete understanding.

See p 796

Twenty-five years ago, Robert Furchgott identified that acetylcholine and bradykinin generated a diffusible, short-lived mediator from endothelial cells that relaxed underlying smooth muscle cells by a mechanism similar to that of nitrovasodilators.² This seminal observation led to the identification of the substance as NO, contributed in a major way to the growth of vascular biology,³ and increased our understanding of why cardiovascular risk factors and disease diminish vasodilatation. Notwithstanding the importance of NO and cyclic guanosine monophosphate (cGMP), the major intracellular mediator of NO-induced smooth muscle relaxation, many of us in the field have had the nagging feeling that something was missing.⁴ That this feeling is warranted is no better exemplified than by the findings of Scotland et al.¹ These authors show that transgenic mice devoid of endothelial NO synthase (eNOS) and cyclooxygenase (COX)-1 maintain a mechanism by which acetylcholine can hyperpolarize, relax, and vasodilate resistance blood vessels. Because their experimental strategy seemingly excludes the enzymatic sources of NO and prostacyclin, the authors conclude that a non-NO, nonprostacyclin substance must be involved. Furthermore, they make the fascinating observation that the mechanism shows a marked gender preference for female mice. To help the reader put the authors’ observations in perspective, I would like to briefly discuss their findings in relationship to our prior knowledge and current understanding of EDHF and to suggest potential fruitful areas of investigation. Scotland et al¹ review the pertinent literature, and the subject of EDHF has been reviewed in depth in Circulation by this author and a colleague,⁴ and more recently by leaders in the field.⁵

	What Is EDHF and What Is Its Endothelial Source?

Top Introduction What Is EDHF and... How Does EDHF Relax... Where Do We Go... References

 
Since the original description that the endothelium-dependent vasodilator acetylcholine causes smooth muscle hyperpolarization⁶ (and, hence, coining of the term EDHF⁷), considerable research has been done to identify a factor that, like NO, diffuses from the endothelium and hyperpolarizes the smooth muscle cell. Hyperpolarization is thought to promote relaxation by inhibiting voltage-dependent calcium channels, although there are other possibilities.⁴ Despite the research, no consensus has been reached as to the mediator of endothelium-dependent hyperpolarization or its mechanism of action. It has always been assumed that the mediator and mechanism of EDHF supplements the actions of NO and prostacyclin, both of which are confirmed endothelium-derived hyperpolarizing and relaxing factors.^4,8 Indeed, vascular hyperpolarization and relaxation in the presence of inhibitors of NOS and COX-1 have been taken to be synonymous with EDHF. Scotland et al¹ show that these inhibitors leave a substantial portion of the mesenteric artery relaxation to acetylcholine or bradykinin intact. By showing that the aortic relaxation to acetylcholine is blocked by the same inhibitors, they also demonstrate another characteristic of EDHF. Although some large arteries, in particular, pig, bovine, and human coronary arteries, have potent EDHF mechanisms, this mechanism appears to be most important in small arteries that account for vascular resistance and therefore blood pressure regulation.

NO was in part identified as a short-lived endothlium derived relaxing factor (EDRF) by its ability to diffuse from an artery with intact endothelium and to relax a neighboring artery denuded of endothelium. Unfortunately, except under special circumstances, inhibitors of NOS and COX-1 block the substantial relaxing factor present in the perfusate of arteries with intact endothelium.^4,5,9 This leaves open the possibility that EDHF is even more short-lived than NO or that it is present in too low a concentration to function in such an experiment. Nevertheless, the hyperpolarization and relaxation of arteries, which persist in the presence of NOS and COX-1 inhibitors, bear some resemblance to EDRF. The response requires an intact endothelium and an increase in endothelial intracellular Ca²⁺ levels, which activates a calmodulin-dependent mechanism similar to that required for eNOS.¹⁰ Derivatives of endothelial cell arachidonic acid have been suspected to be EDRF even since the first report,² and this possibility is compatible with increased activity of Ca²⁺-dependent phospholipase A₂, which releases large amounts of arachidonic acid from endothelial cells stimulated by acetylcholine or bradykinin. Inhibitors of cytochrome P450, a major pathway of nonprostanoid arachidonic acid metabolism, were recognized early on to inhibit endothelium-dependent relaxation, and products of this enzyme, eicosatetraenoic acids (EETs), were shown to directly activate potassium channels and to cause smooth muscle cell hyperpolarization and relaxation. The difficulty of bioassaying a diffusable hyperpolarizing or relaxing factor in the presence of NOS and COX-1 inhibitors has led to the conclusion that although EETs may be involved, they may instead act primarily on endothelial cell potassium channels to hyperpolarize that cell. Unlike smooth muscle cells, endothelial cells are devoid of typical voltage-sensitive Ca²⁺ channels, and hyperpolarization promotes Ca²⁺ entry into the cell through nonselective ion channels. This in turn may further stimulate as-yet unidentified Ca²⁺-dependent processes in the endothelium that promote vascular relaxation. One interesting proposal is that opening of endothelial K⁺ channels, perhaps mediated by EETs, allows for the efflux of potassium from endothelial cells in sufficient amounts that it hyperpolarizes the smooth muscle cells.⁵ This suggestion depends on endothelium-dependent smooth muscle cell hyperpolarization occurring via inward rectifier potassium channels and Na⁺/K⁺ ATPase, both of which are stimulated by low concentrations of K⁺. K⁺ as an EDHF remains a possibility, and its dilution in solution may explain why it is difficult to bioassay. Nevertheless, identification of a non-NO, nonprostanoid factor that is released from endothelial cells in sufficient amounts to cause smooth muscle hyperpolarization is lacking. What explains the fact that after 20 years of investigation, no factor has been found?

One issue is that many of the experimental observations have necessarily required the use of pharmacological inhibitors that may not block 100% of the intended target or that lack specificity. For instance, cytochrome P450 inhibitors used in early experiments were later shown to block potassium channels. Also, NOS inhibitors, which have been widely used to exclude the role of NO, may not completely inhibit NO release, and their efficacy has rarely been verified. As an example, 8 years ago my colleagues and I were in the midst of investigating the mechanisms of endothelium-dependent relaxation of the rabbit carotid artery. This blood vessel has all of the typical characteristics of EDHF, including endothelium-dependent hyperpolarization and relaxation that persist in the presence of NOS and COX-1 inhibitors,¹¹ which are blocked by K⁺ channel and cytochrome P450 inhibitors. Using 2 separate methods of measurement, we made the striking finding that despite using high concentrations of NOS inhibitors, almost 30% of the release of nitric oxide persisted, and the hyperpolarization and relaxation of the artery in the presence and absence of the NOS inhibitors correlated with NO release.⁸ This may simply be because the NOS inhibitors that are generally used are competitive antagonists that by their nature do not provide insurmountable blockade of arginine use by NOS. Also, smaller arteries with EDHF-like properties such as the rabbit carotid artery release higher amounts of NO that are more difficult to block with inhibitors.¹²

Scotland et al¹ appear to have circumvented any possibility that they did not prevent the release of NO or prostacyclin because they used genetic deletion of both the enzymes that normally produce them. They demonstrated that prostacyclin release was diminished in the blood of the mice they studied. They did not, however, evaluate whether NO was eliminated. This possibility would seem to be unlikely were it not for a series of elegant studies by Meng et al,¹³ who showed that in small cerebral pial blood vessels of the eNOS-deficient mouse, endothelium-dependent vasodilation persists. These authors showed that through a striking demonstration of genetic flexibility, the persistent responses are mediated by the other calcium-dependent NOS isoform, neuronal NOS (nNOS), that replaces the deleted eNOS in the endothelium.¹³ Although neither the authors nor the reviewers of the article by Scotland et al¹ appear to have been aware of this work, Scotland et al did show that addition of NOS inhibitor to the arteries deficient in eNOS had no further effect. Although this should obviate any possibility that release of NO persists, nNOS is particularly difficult to completely inhibit, and it is possible that nNOS is producing something other than NO that is unimpeded by NOS inhibitors.¹⁴ Finally, the female gender preference for the EDHF response is compatible with an increased expression of nNOS caused by estrogen.¹⁵

Until recently, the possibility that NOS could produce a mediator other than NO that could function as an EDHF would have seemed unlikely. Like Scotland and associates, however, Matoba et al¹⁶ showed that hyperpolarizing responses to acetylcholine of mouse mesenteric arteries resisted NOS and COX-1 inhibitors and persisted in eNOS-deficient mice. They showed that catalase inhibited acetylcholine-induced hyperpolarization and relaxations that persisted in the presence of the NOS and COX-1 inhibitors in wild-type mice, and proposed that hydrogen peroxide produced by eNOS is EDHF. Morikawa et al¹⁷ provided evidence that strengthened the prospects that hydrogen peroxide may be a NOS-derived vasodilator by showing that mice deficient in Cu/Zn superoxide dismutase, which catalyzes the formation of hydrogen peroxide from superoxide anion, display diminished EDHF vascular responses. They concluded that superoxide dismutase may be an "EDHF synthase." In our own studies, we found that the rabbit carotid artery still relaxed to acetylcholine, despite blocking superoxide dismutase activity.¹² As expected, however, this relaxation greatly increased superoxide levels and decreased the ability of NO to stimulate cGMP. This suggests the possibility that NO released from NOS reacts with superoxide anion to form the product peroxynitrite, which might constitute another mediator of relaxation.¹² Further research is required to determine the precise role of hydrogen peroxide, peroxynitrite, and perhaps other reactive oxygen and nitrogen species as EDHF.

	How Does EDHF Relax Vascular Smooth Muscle?

Top Introduction What Is EDHF and... How Does EDHF Relax... Where Do We Go... References

 
For NO, the simple view is that relaxation of smooth muscle occurs through the stimulation of guanylyl cyclase to produce its product, cGMP, which causes multiple protein kinase G substrates to decrease intracellular Ca²⁺ and inhibit contractile proteins to relax smooth muscle. For EDHF, without knowing precisely what the mediator is, it is difficult to define precisely how hyperpolarization of the smooth muscle occurs. Nevertheless, inward rectifier K⁺ channels and Na⁺/K⁺ ATPase were identified and shown by the use of barium and ouabain, respectively, to play a role in endothelium-dependent hyperpolarization. Scotland et al¹ show that these inhibitors, and, in addition, removal of the endothelium, prevent the residual relaxations in the mesenteric arteries of mice deficient in NOS and COX-1.

Another important mechanism by which endothelial cells cause smooth muscle hyperpolarization arose when it was realized that electrical conductance can occur directly from endothelium to smooth muscle. An anatomic feature called "gap junctions," formed between the cells by connexin proteins, underlie this property. They are present particularly in small arteries where the apposition of the 2 cells is closest, potentially explaining the larger role for EDHF in small arteries. Inhibitors of gap junctions block the hyperpolarization caused by acetylcholine in smooth muscle cells of guinea pig mesenteric arteries without blocking the hyperpolarization of the endothelium, suggesting the importance of this mechanism.¹⁸ Thus, EDHF may not be an enzymatically derived "factor" that, like NO and prostacyclin, diffuses from endothelial cells to smooth muscle; rather it may be a phenomenon that begins as a calcium-activated event that hyperpolarizes the endothelium and a signal that is transferred to the smooth muscle by K⁺ or via gap junctions. The endothelial hyperpolarization could be initiated by EETs or NOS-derived NO, hydrogen peroxide, peroxynitrite, or other reactive species.

As mentioned above, the hyperpolarization and relaxation of the rabbit carotid artery persist in the presence of NOS and COX-1 inhibitors but can be explained by the NO release that persists. More striking is that despite the residual NO, the NOS inhibitors largely prevent the rise in smooth muscle cGMP, suggesting that NO may mediate hyperpolarization and relaxation of the smooth muscle by other means. One possibility is that NO can activate K⁺ channels by a thiol redox-dependent mechanism.¹⁹ Another is that NO may regulate other mechanisms of membrane potential regulation such as Na⁺/K⁺ ATPase. Still another is that NO, without the aid of cGMP, increases the activity of the sarco(endo)plasmic reticulum Ca²⁺ ATPase (SERCA). This in turn increases intracellular stores of Ca²⁺, resulting in the closing of store-operated, nonselective ion channels.²⁰ Because these channels conduct Na²⁺ in addition to Ca²⁺, their closing hyperpolarizes the smooth muscle cell membrane. We recently demonstrated the molecular mechanism by which NO activates SERCA via formation of peroxynitrite within smooth muscle cells, which in turn S-glutathiolates reactive cysteine thiols on SERCA.²¹ Indeed, SERCA was S-glutathiolated during endothelium-dependent relaxation of the rabbit carotid artery, consistent with this mechanism.

	Where Do We Go From Here?

Top Introduction What Is EDHF and... How Does EDHF Relax... Where Do We Go... References

 
This story, like any search for a holy grail, is long and complex, but I think it is fair to conclude that much productive research has been performed in pursuit of the goal of more fully understanding the mechanisms of vasodilatation. Clearly, the story goes well beyond NO and cGMP.⁴ Other mediators, including NOS-derived reactive oxygen and nitrogen species, as well as arachidonic acid derivatives, are likely to be involved. We have moved well beyond the simple concept of diffusable endothelial factors to a need for deeper understanding of the mechanisms by which endothelial cell Ca²⁺ is regulated and by which endothelial membrane potential is regulated and transmitted to the smooth muscle. The mechanisms by which smooth muscle relaxes involve multiple pathways, and an integrated approach is required to fully understand vasodilation. Such an experimental approach requires new methodologies to monitor not only Ca²⁺ but also K⁺, Na⁺, and Cl^– ions, along with membrane potential and vessel diameter. New concepts of redox regulation by reactive oxygen and nitrogen species need to be integrated with our knowledge of the function of ion channels and transporters, as well as contractile proteins. The physiological responses must be correlated with measurements of the suspected mediators, whether they be intra- or extracellular, so that their role is validated.

With good reason, Scotland et al¹ suggest that the mouse model that they have created provides an excellent model with which to identify and further understand EDHF. They provide evidence that without eNOS or COX-1 a major mechanism of blood pressure control remains that counters the hypertension that is expected from the lack of eNOS. Understanding the mechanism is a worthy goal with great potential for the development of therapeutics, particularly for systemic and pulmonary hypertension. Scotland et al provide the "teaser" that the vasodilator mechanism observed in their female double-knockout mice might help to explain the relative cardioprotection enjoyed by women. If this is true, then their findings will have epidemiological importance as well.

	Footnotes

 
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.

	References

Top Introduction What Is EDHF and... How Does EDHF Relax... Where Do We Go... References

 

1. Scotland RS, Madhani M, Cauhan S, Moncada S, Andresen J, Nilsson H, Hobbs AJ, Ahluwalia A. Investigation of vascular responses in endothelial nitric oxide synthase/cyclooxygenase-1 double-knockout mice: key role for endothelium-derived hyperpolarizing factor in the regulation of blood pressure in vivo. Circulation. 2005; 111: 796–803.[Abstract/Free Full Text]
   
2. 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.[CrossRef][Medline] [Order article via Infotrieve]
   
3. Cohen RA. The potential clinical impact of 20 years of nitric oxide research. Am J Physiol. 1999; 276: H1404–H1407.[Medline] [Order article via Infotrieve]
   
4. Cohen RA, Vanhoutte PM. Endothelium-dependent hyperpolarization: beyond nitric oxide and cyclic GMP. Circulation. 1995; 92: 3337–3349.[Free Full Text]
   
5. Busse R, Edwards G, Feletou M, Fleming I, Vanhoutte PM, Weston AH. EDHF: bringing the concepts together. Trends Pharmacol Sci. 2002; 23: 374–380.[CrossRef][Medline] [Order article via Infotrieve]
   
6. 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]
   
7. 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]
   
8. Cohen RA, Plane F, Najibi S, Huk I, Malinski T, Garland CJ. Nitric oxide is the mediator of both endothelium-dependent relaxation and hyperpolarization of the rabbit carotid artery. Proc Natl Acad Sci U S A. 1997; 94: 4193–4198.[Abstract/Free Full Text]
   
9. Kauser K, Rubanyi GM. Bradykinin-induced, N-nitro-L-arginine-insensitive endothelium-dependent relaxation of porcine coronary arteries is not mediated by bioassayable relaxing substances. J Cardiovasc Pharmacol. 1992; 20: S101–S104.[Medline] [Order article via Infotrieve]
   
10. Nagao T, Illiano S, Vanhoutte PM. Calmodulin antagonists inhibit endothelium-dependent hyperpolarization in the canine coronary artery. Br J Pharmacol. 1992; 107: 382–386.[Medline] [Order article via Infotrieve]
    
11. 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]
    
12. Pagano PJ, Griswold MC, Najibi S, Marklund SL, Cohen RA. Resistance of endothelium-dependent relaxation elevation of O₂^– levels in rabbit carotid artery. Am J Physiol Heart Circ Physiol. 1999; 46: H2109–H2114.
    
13. Meng W, Ayata C, Waeber C, Huang PL, Moskowitz MA. Neuronal NOS-cGMP-dependent ACh-induced relaxation in pial arterioles of endothelial NOS knockout mice. Am J Physiol. 1998; 274: H411–H415.[Medline] [Order article via Infotrieve]
    
14. Heinzel B, John M, Klatt P, Bohme E, Mayer B. Ca²⁺/calmodulin-dependent formation of hydrogen peroxide by brain nitric oxide synthase. Biochem J. 1992; 281: 627–630.[Medline] [Order article via Infotrieve]
    
15. Weiner CP, Lizasoain I, Baylis SA, Knowles RG, Charles IG, Moncada S. Induction of calcium-dependent nitric oxide synthases by sex hormones. Proc Natl Acad Sci U S A. 1994; 91: 5212–5216.[Abstract/Free Full Text]
    
16. Matoba T, Shimokawa H, Nakashima M, Hirakawa Y, Mukai Y, Hirano K, Kanaide H, Takeshita A. Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in mice. J Clin Invest. 2000; 106: 1521–1530.[Medline] [Order article via Infotrieve]
    
17. Morikawa K, Shimokawa H, Matoba T, Kubota H, Akaike T, Talukder MA, Hatanaka M, Fujiki T, Maeda H, Takahashi S, Takeshita A. Pivital role of Cu,Zn-superoxide dismutase in endothelium-dependent hyperpolarization. J Clin Invest. 2003; 112: 1871–1879.[CrossRef][Medline] [Order article via Infotrieve]
    
18. Yamamoto Y, Imaeda K, Suzuki H. Endothelium-dependent hyperpolarization and intercellular electrical coupling in guinea-pig mesenteric arterioles. J Physiol. 1999; 514: 505–513.[Abstract/Free Full Text]
    
19. Bolotina VM, Najibi S, Palacino JJ, Pagano PJ, Cohen RA. Nitric oxide directly activates calcium-dependent potassium channels in vascular smooth muscle cells. Nature. 1994; 368: 850–853.[CrossRef][Medline] [Order article via Infotrieve]
    
20. Cohen RA, Weisbrod RM, Gericke M, Yaghoubi M, Bierl C, Bolotina VM. Mechanism of nitric oxide-induced vasodilatation. Refilling of intracellular stores by sarcoplasmic reticulum Ca²⁺ ATPase and inhibition of store-operated Ca²⁺ influx. Circ Res. 1999; 84: 210–219.[Abstract/Free Full Text]
    
21. Adachi T, Weisbrod RM, Pimentel D, Ying J, Sharov VS, Schoneich C, Cohen RA. S-Glutathiolation by peroxynitrite activates SERCA during arterial relaxation by nitric oxide. Nat Med. 2004; 10: 1200–1207.[CrossRef][Medline] [Order article via Infotrieve]
    

Related Article:

Investigation of Vascular Responses in Endothelial Nitric Oxide Synthase/Cyclooxygenase-1 Double-Knockout Mice: Key Role for Endothelium-Derived Hyperpolarizing Factor in the Regulation of Blood Pressure in Vivo

Ramona S. Scotland, Melanie Madhani, Sharmila Chauhan, Salvador Moncada, Jørgen Andresen, Holger Nilsson, Adrian J. Hobbs, and Amrita Ahluwalia
Circulation 2005 111: 796-803. [Abstract] [Full Text]

This article has been cited by other articles:

				D. S. Celermajer Reliable Endothelial Function Testing: At Our Fingertips? Circulation, May 13, 2008; 117(19): 2428 - 2430. [Full Text] [PDF]

				J D Raffetto and R A Khalil Mechanisms of varicose vein formation: valve dysfunction and wall dilation Phlebology, April 1, 2008; 23(2): 85 - 98. [Abstract] [Full Text] [PDF]

				R. A. Rose and W. R. Giles Natriuretic peptide C receptor signalling in the heart and vasculature J. Physiol., January 15, 2008; 586(2): 353 - 366. [Abstract] [Full Text] [PDF]

				H. Si, W.-T. Heyken, S. E. Wolfle, M. Tysiac, R. Schubert, I. Grgic, L. Vilianovich, G. Giebing, T. Maier, V. Gross, et al. Impaired Endothelium-Derived Hyperpolarizing Factor-Mediated Dilations and Increased Blood Pressure in Mice Deficient of the Intermediate-Conductance Ca2+-Activated K+ Channel Circ. Res., September 1, 2006; 99(5): 537 - 544. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) 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. Search for Related Content PubMed PubMed Citation Articles by Cohen, R. A. Related Collections Cardiovascular Pharmacology Animal models of human disease Endothelium/vascular type/nitric oxide Other Vascular biology Related Article