EDHF and NO: different pathways for production--similar actions.

In this issue of Circulation Research, Véquaud and Thorin1 present observations that attest to differential G protein signaling in dilation to nitric oxide (NO) and endothelium-derived hyperpolarization factor (EDHF). The investigators also report that EDHF of the mesenteric artery is not a metabolite of phospholipase C, which strongly suggests that this vasodilator is not a lipid metabolite of cytochrome P450. Véquaud and Thorin also reported that EDHFmediated dilation is calcium independent, which contrasts to the calcium dependency of agonist-induced NO-mediated dilation. And finally the authors report that heat shock protein 90 (Hsp90) participates in the production of NO but not EDHF. This study not only provides important contributions to our understanding of NOand EDHF-mediated vasodilation but also presents a methodological tour de force in the study of microvascular signaling using the intracellular application of antibodies as specific inhibitors of the signaling pathways. With regard to the methodological advance, the authors’ method for the study of signaling pathways in intact microvessels is novel and is based on a technique described to study transduction pathways in cultured cells.2 Specifically, antibodies to specific G proteins or Hsp90 were incorporated into the endothelium of isolated pressurized microvessels using osmotic shock—hyperosmotic conditions followed by hypoosmotic shock. After this treatment, the nonspecific effects appeared nonexistent as evidenced by restoration of spontaneous tone and myogenic responses. The advantage of this approach is that the effects of an antibody should be specific. This contrasts to the application of so-called “specific” inhibitors of enzymes—most of these inhibitors are not specific, and preferential is a better descriptive term. Although many inhibitors have reported Ki values, it is very difficult to estimate the intracellular concentrations in systems as simple as cultured monolayers of cells, and even more problematic in cells in isolated vessels in organ chambers. The use of antibodies to probe signaling pathways represents a truly novel approach to study signaling in intact blood vessels and microvessels and circumvents many of the problems related to specificity with pharmacological inhibitors. The observation that Hsp90 is required for endothelial NO synthase (eNOS) function is well documented by several laboratories,3,4 and the present observations confirm these reports.1 The novel aspect of these particular findings is that Hsp90 is not required for the synthesis and/or actions of EDHF. Véquaud and Thorin1 also reported that EDHF is not a metabolite of phospholipase C and does not require increases in intracellular calcium for production. Although these data do not reveal the identity of EDHF in the mesenteric artery, they suggest what it is not. Along this line, it is worth mentioning that the identity of EDHF is controversial, and our opinion for the basis of this controversy relates to the likelihood that there are many EDHFs. Some groups report that EDHF in coronary arteries and arterioles is a lipid metabolite of cytochrome P450,5–7 and the general presumption is that the lipid substrate for cytochrome P450 is produced by the actions of phospholipase C on membrane phospholipids. However, several other groups have reported that EDHF is not a metabolite of cytochrome P450, and the identity has been suggested as enandimides, K, or other fatty acid metabolites.8–11 The authors also observed that Ca signaling is not involved in the production of EDHF, which contrasts greatly to the signal cascade activated during agonist-induced production of NO. Despite their compelling results, a caveat should be mentioned: with the probable existence of many EDHFs, one should not necessarily conclude that Ca is not involved, because in another EDHF, this cation may hold a seminal role. An important aspect of the results presented by Véquaud and Thorin1 relates to the differential G protein signaling involved in the production of NO and EDHF. In a similar vein, one should also exercise caution regarding a general conclusion about the involvement of G proteins in all EDHFs. Because of the many organ system and species differences that seem to underscore the many faces of EDHF, a universal involvement of a certain G protein in the production of this vasodilator seems unlikely. Despite this caveat, the authors’ results that G protein a-subunits and b-subunits are involved in NOand EDHF-mediated vasodilation, respectively, are important. This conclusion, and advance, was revealed by administration of the specific antibody against the particular G-protein subunit to the endothelium of the intact vessel. Because the antibody was administered intraluminally, the effect is largely confined to endothelial cells. This is important to highlight because it engendered the authors to discriminate between the production of the vasodilator versus the actions of the substance. The results also imply a level of discrete regulation of the production of EDHF and NO. Moreover, the existence of discrete signaling pathways to The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Department of Physiology and Cardiovascular Center, Medical College of Wisconsin, Milwaukee, Wis. Correspondence to William M. Chilian, PhD, Department of Physiology, Cardiovascular Center, 8701 Watertown Plank Rd, Milwaukee, WI 53226-0509. E-mail [email protected] (Circ Res. 2001;89:648-649.) © 2001 American Heart Association, Inc.