The factor in EDHF: Cytochrome P450 derived lipid mediators and vascular signaling

Abstract

Cytochrome P450 (CYP) epoxygenases metabolize arachidonic acid to generate epoxyeicosatrienoic acids (EETs). The latter are biologically active and reported to act as an endothelium-derived hyperpolarizing factor (EDHF) as well as to affect angiogenic and inflammatory signaling pathways. In addition to arachidonic acid, the CYP epoxygenases also metabolize the Ω-3 polyunsaturated fatty acids (PUFAs), eicosapentaenoic acid and docosahexaenoic acid, to generate bioactive lipid epoxide mediators. The latter can be more potent than the EETs but their actions are under investigated. The Ω3-epoxides, like the EETs, are metabolized by the soluble epoxide hydrolase to corresponding diols and epoxide hydrolase inhibition increases epoxide levels and demonstrates anti-hypertensive as well as anti-inflammatory effects. It seems that the overall consequences of CYP epoxygenase activation largely depend on enzyme substrate preference and the endogenous Ω-3/Ω-6 PUFA ratio. This review outlines the evidence for a physiological role for EETs in the regulation of vascular homeostasis.

Introduction

Local vascular tone is determined by a variety of factors such as neuro-transmitters released from autonomic nerves, circulating vasoactive compounds, tissue metabolites and endothelium-derived autacoids. The best characterized vasodilator autacoids are nitric oxide (NO) and prostacyclin (PGI₂), but a substantial component of the vasodilator response observed in response to receptor-dependent agonists or increases in flow is insensitive to inhibitors of NO synthases or cyclooxygenases (COXs). The existence of a NO/PGI₂-independent component of endothelium-dependent relaxation is particularly prominent in the microcirculation as well as in coronary, mesenteric and renal arteries. Since the NO/PGI₂-independent vasodilatation originally described was co-incident with vascular smooth muscle hyperpolarization, and was abolished by depolarizing concentrations of potassium, it was proposed to be mediated by an endothelium-derived hyperpolarizing factor or “EDHF” [1], [2].

As the name implies an endothelium-derived hyperpolarizing factor would be expected to be a substance generated in and released from endothelial cells that is able to stimulate the hyperpolarization of underlying vascular smooth muscle cells. Such a factor would not be expected to be generated in large amounts under resting conditions but its production should be stimulated by hemodynamic stimuli such as cyclic stretch, fluid shear stress and increases in pressure (myogenic response), as well as in response to classical vasodilating agonists. Characterizing the nature of the proposed factor was partially a case for “fridge pharmacology” i.e. the testing of a number of enzymatic inhibitors and ion channel blockers on agonist induced hyperpolarization responses. Relatively early on the field split into two camps: those studying arteries in which the EDHF response was blocked by iberiotoxin and implicating large conductance calcium activated potassium (BK_Ca) channels [3], [4], [5], and those studying arteries in which the response was abrogated by the combination of charybdotoxin and apamin implying small and intermediate conductance K_Ca (SK_Ca and IK_Ca) channels [6], [7], [8], [9]. While the former responses were largely also sensitive to substances such as the cytochrome P450 (CYP) inhibitors, clotrimazole, miconazole and 17-octadecynoic acid, arteries in which EDHF responses were linked to SK_Ca and IK_Ca channel activation were generally not. For quite a while the field failed to advance significantly with an “it is” versus “it is not” type of discussion dominating the literature. The realization that myo-endothelial gap junctional communication may also represent a means of transferring hyperpolarization between endothelial cells and smooth muscle cells without the need for a diffusible factor per se [10], [11], [12] prompted a review of the available evidence and the realization that at least three different mechanisms may underlie the agonist induced NO- and PGI₂-independent hyperpolarization of smooth muscle cells from different vascular beds [13]. Thus, the original EDHF type responses can now be attributed to (i) the activation of endothelial cell SK_Ca and IK_Ca channel activation and the induction of K⁺ ion-induced vascular smooth muscle cell hyperpolarization, (ii) myo-endothelial gap junctional transfer, and (iii) the generation of CYP-derived products of arachidonic acid [14], [15]. There may also be an additional role for endothelial cell derived hydrogen peroxide [16], [17]. Which of these mechanisms dominates in which vascular bed is determined by a number of factors, including the architecture of the vasculature as well as by tissue specific differences in gene expression.

The fact that arachidonic acid was able to induce vascular smooth muscle cell hyperpolarization independent of an increase in cyclic nucleotides together with studies using pharmacological inhibitors of phospholipase A₂ (PLA₂) and CYP enzymes certainly suggested that CYP-derived metabolites of arachidonic acid may act as an EDHF [18], [19], [20]. Moreover, many CYP enzymes are sensitive to inhibition by high concentrations of NO, which fit with observations that EDHF-mediated hyperpolarization and vasodilatation was most prominent when NO production was inhibited [21]. Also, in bioassay systems it was possible to demonstrate the release of a physically transferable endothelium-derived CYP metabolite that could elicit smooth muscle cell hyperpolarization [3]. However, definitive demonstration of the importance of these enzymes came with the report that two epoxides of arachidonic acid i.e. 11,12- and 14,15-epoxyeicosatrienoic acid (EET) were able to elicit the activation of K_Ca channels in bovine coronary arteries to induce hyperpolarization and relaxation [5]. A short time later, it was possible to identify a CYP2C enzyme in porcine coronary artery endothelial cells and demonstrate that the downregulation of this enzyme attenuated agonist-induced hyperpolarization and relaxation [4], [22].

There is at least circumstantial evidence for a similar CYP-dependent vasodilator pathway in humans as inhibitors such as sulfaphenazole have been found to be both ineffective [23] as well as effective [24], [25], [26], [27] in modulating vasodilatation in healthy subjects. The reason for the discrepancy seems to be the vascular bed studied. For example, while CYP inhibitor-sensitive responses in the forearm vasculature in its entirety are difficult to demonstrate, more clear-cut data were obtained when skeletal muscle arterioles [24] and the radial artery [25], [26], [27] were studied; tissues in which it was possible to confirm the expression of CYP2C protein [24], [27].

How are EETs generated? Ca^2 + elevating receptor dependent agonists have frequently been used to increase CYP activity and EET production with bradykinin has been particularly effective in the endothelium [4], [5]. Adenosine 2A receptors (A_2AR) seemingly play an important role in the kidney, in particular during the development of salt sensitive hypertension [28]. Following an increase in cell Ca^2 +, PLA₂ is activated to liberate arachidonic acid from membrane phospholipids. The latter serves as a substrate for CYP enzymes and results in the generation of epoxides that can affect vascular tone. As metabolism of arachidonic acid seems to take place more or less immediately, it seems that the regulation of CYP activity is largely determined by substrate availability combined with the expression of the enzymes involved. Some of the CYP enzymes are known to be phosphorylated [29], [30] but to date there has been no indication that this plays a major role in the acute generation of vasodilator EETs.

As far as endothelial cells are concerned the vast majority of reports indicate that CYP-derived lipid mediators elicit vasodilation, however CYP enzymes in vascular smooth muscle cells also catalyze the generation of hydroxy metabolites such as 20-hydroxyeicosatetraenoic acid (20-HETE) that elicit vasoconstriction — at least in the systemic circulation. The situation is reversed in the pulmonary circulation in which 20-HETE can act as a vasodilator [31] and 11,12-EET as a vasoconstrictor [32], [33]. This review focuses on CYP-derived epoxides and the systemic endothelium, but for more detailed information on the role of CYP metabolites in vascular smooth muscle cells see two excellent recent reviews [34], [35].

Intracellular levels of the EET epoxides are tightly regulated and metabolism occurs relatively rapidly by hydrolysis, β-oxidation and chain elongation [36]. The soluble epoxide hydrolase (sEH) is the most important epoxide-metabolizing enzyme that generates dihydroxyeicosatrienoic acids (DHETs) that are generally less active than the parent epoxides (at least as far as eliciting vasodilatation is concerned). The development of sEH inhibitors and transgenic sEH mice — both of which increase tissue and circulating EET levels, has contributed immensely to the identification of the biological actions of the CYP epoxides and diols [37], [38].

While CYP activity is determined by expression and substrate availability, the activity of the sEH can be regulated by post-translational modification. For example, two tyrosine residues (Tyr383 and Tyr466) in the active site of the hydrolase that are reportedly essential for enzyme activity [39], where found to be nitrated by peroxynitrite in vitro and in vivo in mouse models of type 1 and type 2 diabetes, leading to a decrease in sEH activity [40]. The sEH was also recently reported to be nitrosated in leptin-stimulated wild-type but not endothelial NO synthase knockout mice, suggesting that the effects of NO on arachidonic acid metabolism may be partly related to the modulation of sEH activity [41]. The sEH can also be targeted by high concentrations of electrophilic oxidants such as 15-deoxy-Δ-prostaglandin J₂, e.g. during inflammation, to attenuate sEH activity and increase EET levels [42].