Quick Change Artist: EDHF in resistance arteries

Vascular endothelial cells importantly contribute to the regulation of vascular smooth muscle tone through the production of a variety of vasoactive substances and by direct electrical communication through myoendothelial gap junctions^1-3. Moncada and Vane discovered the first endothelium-derived vasoactive substance, prostacyclin (PGI₂), in the mid 1970's⁴ and rapidly identified its structure and chemical composition⁵. This was followed by the landmark observations of Furchgott, and discovery of endothelium-derived relaxing factor (EDRF) in 1980⁶. Seven years later, Furchgott's EDRF was positively identified as nitric oxide (NO)⁷. The existence of a non-prostaglandin, non-NO endothelium-derived hyperpolarizing factor (EDHF) was postulated by Komori et al.,⁸ in 1988. Twenty-three years later, we know that there is no one EDHF, but rather multiple pathways by which endothelial cells can produce hyperpolarization-induced relaxation of overlying vascular smooth muscle cells independent from PGI₂ and NO (Figure 1)^1-3. The study by Boettcher and de Wit⁹ in this issue of Hypertension supports this hypothesis by demonstrating that the phenotype of non-prostaglandin-, non-NO-mediated, acetylcholine (ACh)-induced vasodilatation can be modulated by the preparation used to study the phenomenon. These authors show that while ACh-induced relaxation of vessels studied in a wire-myograph under isometric conditions is highly dependent on endothelial cell expression of connexin40, ACh-induced vasodilatation of the same vessels studied under isobaric conditions in a pressure-myograph, or studied in vivo, appear independent from connexin40 expression. The implications of this study are at least two-fold. First, as the authors point out, one must take care in extrapolating results from in vitro experiments to regulation of vascular tone, in vivo. Second, and more interestingly, the authors’ findings suggest that the mechanisms underlying non-prostaglandin, non-NO-mediated endothelium-dependent (EDHF-like) vasodilatation can rapidly change depending on the circumstances, in the same blood vessel. This implies that the pathways involved are mutable not only by the experimental conditions, but possibly by the physiological or pathophysiological status of the system. The authors’ findings add to our understanding of endothelial cell function in health and disease^1-3.

Figure 1.

Pathways for non-prostacyclin-, non-nitric oxide-mediated, endothelium-dependent hyperpolarization of vascular smooth muscle in resistance arteries. Schematic diagram of a longitudinal section through a resistance artery showing smooth muscle and endothelial cell cross sections, and depicting several potential pathways by which acetylcholine (ACh) can produce endothelium-dependent smooth cell hyperpolarization and vasodilatation independent from prostaglandins and nitric oxide. On the left, ACh-induced increases in intracellular Ca²⁺ activate endothelial small conductance (sK_Ca) and intermediate conductance (IK_Ca) Ca²⁺-activated K⁺ channels. The released K⁺ ions can then act as an EDHF through activation of smooth muscle inward rectifier K⁺ channels (K_IR) or Na⁺/K⁺ ATPase. Smooth muscle hyperpolarization then closes voltage-gated Ca²⁺ channels (VGCC) leading to a fall in smooth muscle Ca²⁺ and vasodilatation. Alternatively, as shown in the center of the diagram, endothelial cell hyperpolarization produced by activation of sK_Ca or IK_Ca channels, as described above, can be directly transmitted to smooth muscle cells via myoendothelial gap junctions to produce vasodilatation. As shown in the right side of the diagram, the increase in endothelial cell Ca²⁺ produced by ACh can also lead to the synthesis of factors such as epoxides (epoxyeicoatrienoic acids, EETs) of arachidonic acid (AA) produced by cytochromes P450 (CYP), or lipoxygenase (LOX) products such as trihydroxyeicosatrienoic acid (THETA). EETs may activate transient receptor potential V4 (TRPV) channels on smooth muscle cells leading to Ca²⁺ influx which then triggers Ca²⁺ release from smooth muscle stores through ryanodine receptors (RyR) in the form of Ca²⁺ sparks which, in turn, activate smooth muscle large conductance Ca²⁺-activated K⁺ channels (BK_Ca). This leads to smooth muscle hyperpolarization and vasodilatation as outlined above. THETA may directly activate BK_Ca channels to produce smooth muscle hyperpolarization and vasodilatation. Additional pathways also are possible; see references^1-3 for additional information. PLA₂, Phospholipase A₂; PLCβ, Phospholipase Cβ; IP₃, inositol 1,4,5-trisphosphate; SERCA, smooth endoplasmic reticulum Ca²⁺ ATPase.

The study by Boettcher and de Wit⁹ also leaves a number of questions unanswered. First, the identity of the EDHF, or the EDHF-like pathway that is responsible for the ACh-induced smooth muscle relaxation observed in the pressure-myograph, and also in vivo, remains to be established. The lack of effect of connexin40 deletion on ACh-induced vasodilatation under isobaric conditions indicates that direct electrical communication from endothelial cells to smooth muscle cells via connexin40-based gap junctions is not involved under isobaric conditions, in vitro or in vivo. Previously¹⁰ it was shown that ACh-induced dilatation of pressurized mouse gracilis arteries is mediated by release of K⁺ ions through endothelial Ca²⁺-activated K⁺ channels and subsequent activation of smooth muscle inward rectifier K⁺ channels and the Na⁺/K⁺ ATPase (see left side of Figure 1). This mechanism would not require myoendothelial gap junction communication to produce ACh-induced vasodilatation of these murine blood vessels under isobaric conditions. However, in pressurized rat gracilis arteries studied in vitro, cell-cell communication via gap junctions appears requisite for ACh-induced vasodilatation^{11, 12}. This either means that there are substantial species-dependent differences in mechanisms (which is clearly possible²), or that the EDHF-like mechanisms in gracilis arteries are more complicated than previously assumed. It is worthy to note that neither the study by Boettcher and de Wit⁹, nor that by Krummen, et al.,¹⁰, exclude gap junction-based mechanisms involving connexins other than connexin40 in the murine gracilis artery.

Second, the mechanisms responsible for the switch of the phenotype of the EDHF-like responses have yet to be identified. Boettcher and de Wit⁹ suggest that it may be related to differences in wall tension between the isometric and isobaric conditions. However, this hypothesis was not tested, and the signaling pathways involved were not provided. Previous studies in rat mesenteric resistance arteries suggest that the mechanism of ACH-induced, EDHF-like dilation depends on the level of vasoconstrictor-induced tone: at high levels of agonist-induced tone, connexin40-based mechanisms dominate, whereas at more moderate levels of tone, EDHF-like dilation appears independent from connexin40-based mechanisms¹³. Boettcher and de Wit⁹ used similar concentrations of norepinephrine to constrict mouse gracilis arteries in both the wire-myograph (isometric) and pressure-myograph (isobaric) preparations in an attempt to produce similar levels of smooth muscle activation. However, the authors did not verify that the concentration of vasoconstrictor agonist used was equi-effective in the two preparations. Direct comparison of norepinephrine concentration-response relationships between vessels studied in a wire-myograph (isometric) and a pressure-myograph (isobaric) have shown that norepinephrine is actually more potent in the isobaric system¹⁴. Thus, a given, sub-maximal concentration of this vasoconstrictor will produce a greater response (i.e., more tone) in the isobaric preparation. This is the opposite of what would be required to produce a “switch” in EDHF-like mechanisms as proposed by Mather, et al.¹³.

Obviously additional experiments will be required to determine what is responsible for the differences reported by Boettcher and de Wit⁹, and particularly to define their role in the etiology and progress of diseases like hypertension. Ideally these experiments should include simultaneous measurement of smooth muscle and endothelial cell membrane potential, more direct assessment of gap junction-based communication (using current injection, for example) and precise quantitation of the level of agonist-induced smooth muscle activation in the two experimental preparations. More attention to the study of EDHF mechanisms in vivo would also appear warranted. It also remains to be established if changes in wall tension, as could be produced by changes in luminal pressure, alone, in the absence of added vasoconstrictors, can modulate the mechanisms underlying EDHF-like responses in resistance arteries and arterioles that develop substantial myogenic tone. Better understanding of these pathways may help to clarify changes in vascular function that produce, or are affected by diseases like hypertension. These mechanisms also may provide new targets for development of therapies to treat and potentially prevent hypertension or other pathologies that affect vascular function.