ACh-induced endothelium-dependent relaxation in rabbit small mesenteric arteries is resistant to N-nitro-l-arginine (l-NA) and indomethacin but sensitive to high K+, indicating the relaxations are mediated by endothelium-derived hyperpolarizing factors (EDHFs). The identity of the EDHFs in this vascular bed remains undefined. Small mesenteric arteries pretreated with l-NA and indomethacin were contracted with phenylephrine. ACh (10−10 to 10−6 M) caused concentration-dependent relaxations that were shifted to the right by lipoxygenase inhibition and the Ca2+-activated K+ channel inhibitors apamin (100 nM) or charybdotoxin (100 nM) and eliminated by the combination of apamin plus charybdotoxin. Relaxations to ACh were also blocked by a combination of barium (200 μM) and apamin but not barium plus charybdotoxin. Addition of K+ (10.9 mM final concentration) to the preconstricted arteries elicited small relaxations. K+ addition before ACh restored the charybdotoxin-sensitive component of relaxations to ACh. K+ (10.9 mM) also relaxed endothelium-denuded arteries, and the relaxations were inhibited by barium but not by charybdotoxin and apamin. With the use of whole cell patch-clamp analysis, ACh (10−7 M) stimulated voltage-dependent outward K+ current from endothelial cells, which was inhibited by charybdotoxin, indicating K+ efflux. Arachidonic acid (10−7 to 10−4 M) induced concentration-related relaxations that were inhibited by apamin but not by charybdotoxin and barium. Addition of arachidonic acid after K+ (10.9 mM) resulted in more potent relaxations to arachidonic acid compared with control without K+ (5.9 mM). These findings suggest that, in rabbit mesenteric arteries, ACh-induced, l-NA- and indomethacin-resistant relaxation is mediated by endothelial cell K+ efflux and arachidonic acid metabolites, and a synergism exists between these two separate mechanisms.
- potassium channels
- endothelium-derived hyperpolarizing factor
stimulation of the vascular endothelium with ACh, bradykinin, and increases in flow induces vasodilation by releasing soluble factors such as nitric oxide (NO), prostacyclin, and endothelium-derived hyperpolarizing factors (EDHFs; see Refs. 1, 5, 10, 20). The action of EDHFs has been demonstrated in a number of vascular beds, where a significant portion of endothelium-dependent relaxation persists in the presence of NO synthase (NOS) and cyclooxygenase inhibitors such as N-nitro-l-arginine (l-NNA) and indomethacin. This EDHF-mediated relaxation is associated with smooth muscle K+ channel activation and hyperpolarization and is sensitive to inhibition to high extracellular K+ or K+ channel blockers (1). The mediators of EDHF vary with species, vascular bed, and vascular size. In bovine, porcine, and canine coronary arteries, cytochrome P-450 metabolites of arachidonic acid, epoxyeicosatrienoic acids (EETs), act as EDHFs (3, 8, 11, 23), whereas K+ may represent EDHF in rat hepatic arteries (7). In addition, H2O2 and myoendothelial gap junctions may mediate smooth muscle hyperpolarization and EDHF responses in some murine vascular beds (16, 26).
The EDHF-mediated relaxations involve different subtypes of K+ channels. Blockade of the large-conductance Ca2+-activated K+ (KCa) channel (BKCa) inhibits EDHF-mediated relaxations to ACh in bovine coronary arteries (3), whereas the inhibition of intermediate-conductance KCa (IKCa) channel is sufficient to block EDHF-mediated relaxation in rat cerebral arteries (14). In rat hepatic arteries, stimulation of endothelial cells leads to K+ efflux into the myoendothelial space, which causes relaxation by activating inward-rectifying K+ (Kir) channels and Na+-K+-ATPase on the smooth muscle (7). In many other arteries, simultaneous application of inhibitors for both small-conductance KCa (SKCa) and IKCa channels is required to block the EDHF response (1).
In rabbit small mesenteric arteries, ACh-induced relaxation is resistant to inhibition by l-NNA and indomethacin but sensitive to high K+, indicating the activity of EDHF (9). The identity of EDHFs in this vasculature remains elusive. Recently, we reported that the lipoxygenase metabolites of arachidonic acid, trihydroxyeicosatrienoic acids (THETAs), activate apamin-sensitive or SKCa currents on vascular smooth muscle. These lipoxygenase metabolites serve as EDHFs and mediate non-NO and nonprostacyclin relaxations to arachidonic acid and ACh in rabbit aorta and rabbit mesenteric arteries (4, 13, 31). The present study evaluated the effects of various K+ channel blockers on ACh-induced relaxations of rabbit small mesenteric arteries. Based on the results with K+ channel blockers, the potential roles of arachidonic acid metabolites and K+ efflux as EDHFs were explored.
MATERIALS AND METHODS
Experiments were performed on isolated mesenteric arteries of 4- to 6-wk-old, male New Zealand White rabbits (New Franken Research Rabbits, New Franken, WI). The experimental protocol was approved by the Animal Care Committee of the Medical College of Wisconsin. Second- or third-order branches from the superior mesentery arteries (200–300 μm) were isolated and placed in HEPES solution consisting of (in mM): 150 NaCl, 5.0 KCl, 1.8 CaCl2, 1.0 MgCl2, 10 HEPES, and 5.5 glucose, pH 7.4. Arterial segments (1.5 mm long) were threaded on two stainless steel wires (40 μm diameter) and mounted on a four-chamber wire myograph (model 610M; Danish Myo Technology). Arteries were equilibrated at 37°C for 30 min in physiological saline solution (PSS) containing (in mM): 119 NaCl, 4.7 KCl, 2.5 CaCl2, 1.17 MgSO4, 24 NaHCO3, 1.18 KH2PO4, 0.026 EDTA, and 5.5 glucose, bubbled with 95% O2-5% CO2 (29). The resting tension was set at 1 mN. Arteries were stimulated two times with KCl (60 mM) plus phenylephrine (10 μM) for 3–5 min at 10-min intervals before the initiation of experimental protocols.
Vascular relaxation responses.
A submaximal concentration of phenylephrine (0.5–2 μM) was added to the bath to precontract the arteries to 50–75% of the maximal phenylephrine response. After the contraction reached steady state, cumulative concentration responses to ACh (10−10 to 10−6 M) were determined. Relaxations were examined as paired rings or before and after the application of inhibitors. To examine the role of NO and cyclooxygenase metabolites, arteries were pretreated for 15–30 min with the endothelial NOS inhibitor l-NNA (30 μM) or the cyclooxygenase inhibitor indomethacin (10 μM). To examine the contribution of smooth muscle hyperpolarization to the relaxation responses, arteries were preconstricted with KCl (60 mM), and the response to ACh was determined. Where indicated, the endothelium was removed by gently rubbing the intimal surface of the artery with a human hair. The endothelium was considered intact if ACh (1 μM) caused >80% relaxation of the phenylephrine-precontracted arteries and effectively denuded if ACh induced <10% relaxation.
All subsequent experiments on EDHF-mediated relaxations were performed in the presence of l-NNA (30 μM) and indomethacin (10 μM). To determine the contribution of lipoxygenase metabolites, arteries were pretreated for 15–30 min with cinnamyl-3,4-dihydroxy-α-cyanocinnamate (CDC, 1 μM), nordihydroguaiaretic acid (NDGA, 1 μM), or ebselen (1 μM). Stepwise addition of KCl (2.5 mM, 8.4–18.4 mM final concentration) was performed to determine the concentration-dependent relaxations to K+. To characterize the K+ channels that mediate the relaxations to ACh, arteries were pretreated for 15–30 min with the following agents alone or in combination: charybdotoxin (100 nM), an IKCa channel blocker; apamin (100 nM), an SKCa channel blocker; barium (200 μM), a Kir channel blocker; iberiotoxin (100 nM), a BKCa channel blocker; or ouabain (10 or 100 μM), an Na+-K+-ATPase inhibitor. To examine the role of K+ efflux and arachidonic acid metabolites as EDHFs, vascular responses to a low concentration of K+ (5 mM, 10.9 mM final), arachidonic acid (10−7 to 10−4 M), and THETAs (0.1–1× concentration) were examined in the absence and presence of the various K+ channel blockers. To examine the interaction of K+ efflux and arachidonic acid metabolites, vascular responses to arachidonic acid or THETAs were determined in the presence of KCl (5 mM, final concentration 10.9 mM).
Whole cell recordings of K+ currents were obtained in freshly isolated mesenteric endothelial cells using a modification of published methods (12, 24). In brief, rabbit small mesenteric arteries were dissected and incubated with low calcium (0.1 μM) PSS containing 1 mg/ml albumin followed by sequential incubation at 37°C with papain (1 mg/ml) and dithiothreitol (0.5 mg/ml) for 15 min, then collagenase (2.5 mg/ml), trypsin inhibitor (1 mg/ml), and elastase (0.5 mg/ml) for 15–20 min. All enzymes were purchased from Sigma Chemical. Endothelial cells were dislodged first with gentle trituration. The remaining tissue was removed, and the cells were placed on ice. Endothelial cells were placed in a 1-ml patch-clamp chamber on an inverted microscope (Olympus IX70). Currents were recorded with an Axopatch 200B amplifier (Axon Instruments) and pClamp 8 software (Axon instruments). Cells were dialyzed with a pipette solution composed of (in mM) 145 potassium glutamate, 1 MgCl2, 10 HEPES, 1 EGTA, 1 Na2ATP, 0.05 Na2GTP, and 100 nM Ca2+ (pH 7.2) and perfused with a bath solution composed of (in mM) 145 NaCl, 4 KCl, 1 MgCl2, 10 glucose, 10 HEPES, and 2 CaCl2 (pH 7.4) at room temperature. Pipette tip resistance averaged 4–8 MΩ. Macroscopic K+ currents were generated by progressive, stepwise, 10-mV depolarizing steps (500-ms duration and 5-s intervals) from a constant holding potential of −60 mV. Currents were sampled at 3 kHz and filtered at 1 kHz. After control currents were recorded, ACh (10−7 M) was applied in the absence and presence of charybdotoxin (100 nM) or apamin (100 nM). The effect of charybdotoxin on control currents was also determined. Indomethacin (10 μM) was present in all perfusate solutions. Recordings were performed in triplicate and averaged to estimate K+ current density. The membrane capacitance of each cell was estimated by integrating the capacitative current generated by a 10-mV hyperpolarizing pulse after electronic cancellation of pipette-patch capacitance. Current density is expressed as picoamperes per picofarad (pA/pF) to account for differences in cell membrane area.
Phenylephrine, ACh, arachidonic acid (sodium salt), l-NNA, indomethacin, NDGA, ebselen, apamin, charybdotoxin, iberiotoxin, barium, and ouabain were purchased from Sigma Chemical. CDC was obtained from Bio-Mol Research laboratories. 11,12,15-THETA was synthesized and isolated as previously described (31).
Relaxation responses are expressed as a percentage relaxation relative to phenylephrine or KCl precontraction, with 100% relaxation representing basal tension. Current density is expressed as normalized current density, with maximum control current at +60 mV representing 100%. Data are presented as means ± SE. Significance of differences between mean values was evaluated by Student's t-test or ANOVA followed by the Student-Newman-Keuls multiple-comparison test. P < 0.05 was considered statistically significant.
We first determined the contribution of three major relaxing factors, i.e., NO, prostacyclin, and EDHFs in ACh relaxation responses. In phenylephrine-contracted arteries, ACh (10−10 to 10−6 M) elicited concentration-dependent relaxations, with a maximal relaxation of 99 ± 0.3% (Fig. 1A). Pretreatment of arteries with l-NNA (30 μM) caused a rightward shift of the concentration-response curve to ACh (maximal relaxation of 99 ± 0.2%; Fig. 1A). In contrast, indomethacin (10 μM) was without effect (Fig. 1B). l-NNA plus indomethacin had no significant effect on basal tension. High K+ (60 mM) also shifted ACh responses to the right (maximal relaxation of 89 ± 0.3%; Fig. 1C). Removal of the endothelium eliminated ACh-induced relaxations (Fig. 1D).
To further define the role of EDHFs in ACh-induced relaxations, arteries in all subsequent experiments were treated with l-NNA (30 μM) and indomethacin (10 μM) to block NO and prostacyclin pathways. In the presence of l-NNA and indomethacin, a significant portion of the relaxation to ACh remained (maximal relaxation of 99 ± 0.2%), which was blocked by high K+ (60 mM KCl), indicating the involvement of K+ channels and EDHF (Fig. 2A). The l-NNA- and indomethacin-resistant relaxation was eliminated by a combination of the SKCa channel blocker apamin (100 nM) and IKCa channel blocker charybdotoxin (100 nM). Apamin and charybdotoxin alone shifted the ACh responses to the right without affecting maximal relaxations (Fig. 2B). Barium (200 μM), a Kir channel blocker, caused a rightward shift of the concentration-response curve to ACh. The combination of barium and apamin blocked the ACh responses, whereas barium plus charybdotoxin had no further inhibition compared with the blockers alone (Fig. 2C). Relaxations to ACh were not affected by 10 μM ouabain, an Na+-K+-ATPase inhibitor, but were significantly inhibited by 100 μM ouabain (Fig. 2D). The indomethacin- and l-NNA-resistant relaxations to ACh were not altered by iberiotoxin (Fig. 2E), and the combination of apamin plus iberiotoxin inhibited the relaxations to similar extent as apamin alone. These studies suggest the involvement of the following two parallel pathways in the relaxations to ACh: 1) an SKCa apamin-sensitive component and 2) an IKCa and Kir component sensitive to charybdotoxin and barium. Iberiotoxin-sensitive BKCa channels are not involved in the responses.
Previous studies on the rabbit aorta indicate that THETAs, lipoxygenase metabolites of arachidonic acid, mediate a portion of the relaxations to ACh by activation of apamin-sensitive K+ channels (4, 13). To determine whether lipoxygenase metabolites contribute to the apamin-sensitive component of ACh-induced relaxations, arteries were treated with the lipoxygenase inhibitors CDC, NDGA, and ebselen (Fig. 3, A–C). CDC, NDGA, and ebselen inhibited the relaxations to ACh and shifted the concentration response to the right. This concentration of the inhibitors blocked the conversion of [14C]arachidonic acid to [14C]-hydroxyepoxyeicosatrienoic acids (HEETAs) and -THETAs by 50–70% (data not shown; see Ref. 31). However, higher concentrations of the lipoxygenase inhibitors inhibited phenylephrine contractions in these arteries and therefore could not be tested on ACh-induced relaxations. These findings indicate that THETAs contribute to the relaxations to ACh in rabbit mesenteric arteries.
In l-NNA- and indomethacin-treated preconstricted rings, stepwise increases in extracellular K+ caused concentration-related relaxations (Fig. 4A). Maximal relaxations were 64 ± 4% at 13.4 mM K+. Increases in K+ >15.9 mM resulted in constriction. In endothelium-denuded arteries, 10.9 mM K+ induced sustained relaxations that were inhibited by barium but not by charybdotoxin or apamin (Fig. 4B). Thus, in rabbit mesenteric arteries, K+ alone causes relaxation independent of NO, cyclooxygenase metabolites, and the endothelium. The relaxations to K+ are inhibited by barium, implicating Kir channels in the response.
The role of K+ efflux via IKCa channels in EDHF-mediated relaxations to ACh was examined. Arteries were treated with charybdotoxin to block K+ efflux. Addition of a low concentration of K+ (10.9 mM, final concentration) to mimic K+ efflux elicited small sustained relaxations. K+ addition before ACh restored the charybdotoxin-sensitive component of relaxations to ACh (Fig. 4C).
Patch-clamp studies were performed to confirm K+ efflux via IKCa channels in vascular endothelial cells (Fig. 5). Whole cell K+ currents of freshly isolated endothelial cells were elicited by 10-mV depolarizing steps from −60 to +60 mV. Capacitance of the cells averaged 15.7 ± 2.2 pF. Perfusion of the endothelial cells with ACh (10−7 M) resulted in a significant increase in outward K+ currents. This increase was subsequently inhibited by charybdotoxin (100 nM; Fig. 5A) but not apamin (100 nM; Fig. 5B). Under control conditions, charybdotoxin decreased the outward K+ currents by 30%.
To examine the contribution of arachidonic acid metabolites to the EDHF response in rabbit small mesenteric arteries, the effect of K+ channel inhibitors was examined on arachidonic acid-induced relaxations. As shown in Fig. 6, arachidonic acid (10−7 to 10−4 M) caused concentration-dependent relaxations in small mesenteric arteries contracted with phenylephrine in the presence of l-NNA and indomethacin. The maximal relaxation to arachidonic acid was 94 ± 2%. The relaxations to arachidonic acid were markedly inhibited by apamin (maximal relaxation of 45 ± 10%). In contrast, charybdotoxin and barium did not have a significant effect on relaxation responses. The combination of apamin and charybdotoxin or barium had similar inhibition on arachidonic acid-induced relaxations as apamin alone. Because both arachidonic acid and 10.9 mM K+ relax mesenteric arteries, we tested the interaction of these two relaxing factors. Addition of 10.9 mM K+ potentiated the relaxations to arachidonic acid when compared with control relaxations with arachidonic acid (Fig. 6C).
We also examined the effects of 11,12,15-THETA on the vascular tone of rabbit small mesenteric arteries (31). 11,12,15-THETA (0.1–1× concentration) relaxed preconstricted, denuded arteries in a concentration-dependent manner (Fig. 7A). These relaxations were blocked by apamin. Similar to arachidonic acid and K+, K+ enhanced the relaxations to 11,12,15-THETA (Fig. 7B).
In the present study, we characterized relaxation responses to ACh in isolated rabbit small mesenteric arteries. We found that ACh induced a potent endothelium-dependent relaxation that is resistant to l-NNA and indomethacin. The ACh-induced NO- and prostacyclin-independent relaxation is eliminated in arteries depolarized by high K+, thus confirming the previous findings that ACh-induced relaxation is mediated by membrane potential-sensitive mechanisms or EDHFs (9). With regard to the mediators of this EDHF response, we provided the first evidence that the EDHF-dependent relaxation response to ACh is mediated by two parallel mechanisms (i.e., K+ efflux and arachidonic acid metabolites) and a synergism exists between these two mechanisms.
EDHF-mediated relaxations may involve different subtypes of K+ channels in various vascular beds (1, 3, 7, 14). The present study shows that, in rabbit mesenteric arteries, apamin and charybdotoxin partially reduced ACh-induced relaxation, and a combination of both toxins abolished the response. In contrast, apamin plus iberiotoxin, a selective BKCa channel blocker, did not further inhibit ACh-induced relaxation compared with apamin alone. These results indicate that the EDHF response involves the activation of SKCa and IKCa channels in rabbit mesenteric arteries. Additionally, ACh-induced relaxation was abolished by a combination of barium and apamin, whereas the relaxation was only partially inhibited by barium alone. This suggests that Kir channels are also involved in the EDHF response. Because barium plus charybdotoxin elicited a similar inhibition on ACh-induced relaxation compared with either toxin alone, it is likely that ACh stimulates a sequential activation of IKCa and Kir channels. Alternatively, ouabain inhibited the ACh relaxations only at the concentration of 100 μM. Because of nonspecific actions of ouabain at high concentrations, the significance of this result remains inconclusive (15). Taken together, our results suggest that the EDHF-dependent relaxation to ACh involves two parallel pathways, i.e., the activation of IKCa and Kir channels in one pathway and the activation of SKCa channels in another.
Previous studies have suggested that IKCa channels are restricted to the endothelium, and K+ efflux via these channels may serve as an EDHF in some arteries (2, 7, 17, 27). In contrast, K+ does not appear to significantly contribute to the EDHF response in arteries such as porcine coronary, guinea pig carotid, and mouse mesenteric artery (6, 25). K+ release in the myoendothelial space relaxes the adjacent smooth muscle by activating smooth muscle Kir channels and possibly Na+-K+-ATPase (7, 22, 29). In the present study, ACh-induced EDHF-dependent relaxations were inhibited by charybdotoxin. Raising bath K+ concentration to 10.9 mM significantly enhanced the EDHF-dependent relaxations to ACh. Additionally, K+ (10.9 mM) induced relaxation in denuded mesenteric arteries, and these relaxations were inhibited by barium but not charybdotoxin. K+ efflux via IKCa channels, but not SKCa channels, was confirmed by whole cell patch-clamp of isolated mesenteric endothelial cells. In those cells, ACh elicited a significant increase in outward currents, and this increase was subsequently inhibited by charybdotoxin but not apamin. Based on these results, we suggest that K+ efflux via endothelial IKCa channels contributes to EDHF-dependent relaxation responses to ACh in rabbit small mesenteric arteries. K+ induces relaxation in these arteries that is mediated by activation of smooth muscle Kir channels.
Maximum relaxation to exogenous K+ (13.4 mM) averaged 64%, whereas maximum indomethacin-, l-NNA-, and lipoxygenase-resistant relaxation to ACh was 90–100%. Similarly, maximal indomethacin-, l-NNA-, and apamin-resistant relaxation to ACh was 90–100%. The reason for the discrepancy is unclear. It is possible that exogenous K+ may not be as efficacious as local release of endogenous K+. Additionally, because higher concentrations of the lipoxygenase inhibitors could not be used in these arteries, residual relaxations by lipoxygenase metabolites may occur.
As discussed previously, our data indicate that the activation of apamin-sensitive SKCa channels represents another pathway in ACh-induced EDHF-dependent relaxations. This is consistent with a previous report by Murphy and Brayden (21) that apamin inhibits the non-NO and non-prostanoid-dependent hyperpolarization to ACh in rabbit mesenteric arteries. With regard to the possible mechanisms responsible for SKCa channel activation, we reported that lipoxygenase metabolites of arachidonic acid (11,12,15-THETA) activate apamin-sensitive or SKCa currents on vascular smooth muscle, thereby acting as an EDHF in ACh-induced relaxations in rabbit aorta (13). In the current study, arachidonic acid-induced relaxations were markedly inhibited by apamin. Similarly, 11,12,15-THETA also relaxed the rabbit mesenteric arteries, which were inhibited by apamin. Therefore, arachidonic acid metabolites could serve as mediators of the SKCa channel activation and relaxations to ACh in rabbit small mesenteric arteries. The SKCa channels responsible for ACh-induced relaxations are probably present on vascular smooth muscle cells, since apamin did not affect the outward K+ currents activated by ACh in mesenteric endothelial cells.
It is important to note that K+ and arachidonic acid acted synergistically in inducing relaxations. Addition of arachidonic acid after 10.9 mM K+ elicited more potent relaxations compared with control relaxation with arachidonic acid. There is also greater relaxation to 11,12,15-THETA plus 10.9 mM K+ compared with either agent alone. These results provide further support for two parallel pathways in relaxation responses of rabbit small mesenteric arteries. A similar scheme of EDHF-dependent relaxations involving parallel pathways has been recently reported in porcine coronary arteries (28) and rat cremaster arteries (18). In these arteries, bradykinin hyperpolarizes smooth muscle cells by endothelial cell hyperpolarization and gap junctions as one mechanism and endothelial production of EETs and BKCa channel action as another mechanism.
In summary, the data from the present study indicate that ACh-induced EDHF-dependent relaxation in rabbit small mesenteric arteries involves two separate parallel mechanisms, i.e., K+ efflux and arachidonic acid metabolites (Fig. 8). The opening of endothelial IKCa channels in response to ACh induces K+ efflux, which activates smooth muscle Kir channels. In another pathway, ACh stimulates the metabolism of arachidonic acid in endothelial cells by the lipoxygenase pathway, and these metabolites diffuse into the myoendothelial space and activate smooth muscle SKCa channels. The two mediators have a synergistic effect to cause relaxation. The corelease of K+ and arachidonic acid lipoxygenase metabolites mediates the EDHF response in rabbit mesenteric arteries.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-37981. D. X. Zhang is a postdoctoral fellow of the American Heart Association, Greater Midwest Affiliate, and a recipient of the Jenkins Cardiovascular Research Fellowship.
We thank Gretchen Barg for secretarial assistance.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 by the American Physiological Society