Modulation of vasorelaxant responses to potassium channel openers by basal nitric oxide in the rat isolated superior mesenteric arterial bed

Reduced EDHF responses and connexin activity in mesenteric arteries from the insulin-resistant obese Zucker rat

AIMS/HYPOTHESIS

The objective of this study was to examine the effect of insulin resistance on endothelium-derived hyperpolarising factor (EDHF) and small mesenteric artery endothelial function using 25-week-old insulin-resistant obese Zucker rats (OZRs) and lean littermate control rats (LZRs). The involvement of gap junctions and their connexin subunits in the EDHF relaxation response was also assessed.

METHODS

Mesenteric arteries were evaluated using the following assays: (1) endothelial function by pressure myography, with internal diameter recorded using video microscopy; (2) connexin protein levels by western blotting; and (3) Cx mRNA expression by real-time PCR.

RESULTS

Relaxations in response to acetylcholine were significantly smaller in mesenteric arteries from the OZRs than the LZRs, whereas there was no difference in relaxations in response to levcromakalim. Responses to acetylcholine were not altered by nitric oxide inhibitors, but were abolished by charybdotoxin in combination with apamin, which blocked the EDHF component of the response. 40Gap27 significantly attenuated the response to acetylcholine in the LZRs, but had no effect in the OZRs. Connexin 40 protein and Cx40 mRNA levels in mesenteric vascular homogenates were significantly smaller in the OZRs than in the LZRs, with no difference in connexin 43 or Cx43 mRNA levels.

CONCLUSIONS/INTERPRETATION

These findings demonstrate that endothelial dysfunction in mesenteric arteries from the insulin-resistant OZRs can be attributed to a defect in EDHF. The results also suggest that the defective EDHF is at least partly related to an impairment of connexin 40-associated gap junctions, through a decrease in connexin 40 protein and Cx40 mRNA expression in the OZRs.

Effect of phosphodiesterase 5 inhibitor on alteration in vascular smooth muscle sensitivity and renal function in rats with liver cirrhosis

Previous studies suggested that increased activity of phosphodiesterase (PDE)5 in the kidneys of cirrhotic rats contributes to sodium retention. This study examined the role of PDE5 in the changes in vascular reactivity, hemodynamics, and sodium excretion in rats with liver cirrhosis. Four weeks after bile duct ligation (BDL) or sham operation (SO), in vitro reactivity of aortic rings to various agents and in vivo effects of a PDE5-selective inhibitor [1,3-dimethyl-6-(2-propoxy-5-methanesulfonylamidophenyl)pyrazolo[3,4d]-pyrimidin-4-(5H)-one, DMPPO] were studied. The vasodilator responses to nitroglycerin and S-nitroso-N-acetyl-penicillamine (SNAP) in phenylephrine-precontracted rings without endothelium were attenuated in BDL compared with SO rats. Pretreatment with DMPPO (0.1 microM) enhanced these responses and eliminated the differences between the two groups. Vasodilation to DMPPO itself was also less in BDL rats. The responses to phenylephrine were attenuated in endothelium-rich aorta from BDL relative to SO rats, but they were similar in endothelium-denuded aorta and remained similar despite preincubation with SNAP (0.1 microM) alone or with SNAP and DMPPO. In vivo, BDL rats were vasodilated relative to SO rats; DMPPO (5 mg/kg i.v.) decreased arterial pressure and vascular resistance in both groups equally and caused significant increase in sodium excretion in BDL rats only. In conclusion, the results are in accordance with a possible increase in PDE5 activity in aorta and kidney of cirrhotic rats that results in reduced responses to NO donors and contributes to the increase in sodium retention. PDE5 inhibitors may ameliorate sodium retention in cirrhosis but may worsen vasodilation. Examining the effect of PDE5 inhibitors after chronic administration will be more revealing.

Lysophosphatidylcholine potentiates phenylephrine responses in rat mesenteric arterial bed through modulation of thromboxane A2

Lysophosphatidylcholine (LPC) plays important physiological and pathophysiological roles in the cardiovascular system. Despite this, there is little information about its effects on vasore-activity of resistance vessels. The present study was designed to characterize the effects of LPC in the isolated perfused rat mesenteric arterial bed (MAB) and to investigate the underlying mechanisms of the changes it produced. Perfusion with 10 microM LPC for 40 min did not significantly affect basal perfusion pressure or reactivity of MAB to the alpha(1)-adrenoceptor agonist phenylephrine (PE) but almost completely abolished the maximal endothelium-dependent relaxation to acetylcholine (Ach), reducing it from 93 +/- 5 to 7 +/- 4% (p < 0.001). After washout of LPC for 60 min, the vasodilator response to Ach partially recovered, whereas the vasoconstrictor response to PE was markedly enhanced, the pD(2) value increasing from 7.50 +/- 0.04 to 8.13 +/- 0.15 and maximum response to 199 +/- 24% of control (p < 0.001). Pretreatment with either indomethacin, a nonselective inhibitor of cyclooxygenase, or SQ-29548 [[1S-[1a,2a(Z),3a,4a]]-7-[3-[[2-[(phenylamino)carbonyl]hydrazino] methyl]-7-oxabicyclo [2.2.1]hept-2-yl]-5-heptanoic acid], a selective thromboxane receptor antagonist, completely prevented the potentiation of the PE response after washout of LPC. In untreated MABs, only the highest concentration of PE produced a significant increase in thromboxane A(2) (TxA(2)) production (assessed by enzyme-immunoassay of thromboxane B(2) levels). This was prevented by perfusion with LPC but was significantly increased after LPC washout. The basal release of TxA(2) was not modified by LPC. These results demonstrate that LPC exerts both immediate and residual effects on the reactivity of the rat MAB and that these effects are at least partially due to modification of PE-induced TxA(2) production.

Effect of sildenafil on the isolated rat aortic rings

Sildenafil, a highly selective inhibitor of PDE 5, is effective in the treatment of erectile dysfunction. Penile erection involves relaxation of smooth muscle of corpus cavernosum and its associated arterioles. The objective of this study was to investigate the effect of sildenafil on nitric oxide/cyclic guanosine monophosphate (NO/cGMP)-dependent relaxation of rat aortic rings. The contribution of sildenafil to the vasorelaxation of rat aortic rings was also investigated. Sildenafil produced significant potentiation of acetylcholine (ACh, 2 x 10(-6) m)-induced relaxation at concentration > or =1 x 10(-8) m. Addition of sildenafil (1 x 10(-7) m) to aortic rings failed to alter the effect of N(G)-nitro-L-arginine (l-NNA, 3 x 10(-5) m) or methylene blue (MB, 3 x 10(-5) m) on ACh response. Similarly, sildenafil (1 x 10(-7) m) augmented significantly the vasorelaxation induced by sodium nitroprusside over the range of 1 x 10(-9)-1 x 10(-8) m. When added to phenylephrine (3 x 10(-6) m)-precontracted rat aortic rings, sildenafil (1 x 10(-9)-1 x 10(-4) m) induced concentration-dependent relaxation reaching a maximum of 96.48 +/- 1.44%. These relaxations were not significantly attenuated by previous incubation with L-NNA (3 x 10(-5) m) or MB (3 x 10(-5) m). Denudation did not significantly affect the vasorelaxant effect of sildenafil. Sildenafil may act in the rat aortic rings through the amplification of NO/cGMP pathway. It may augment both basal endothelial NO function and exogenous NO-dependent vasodilatation. However, sildenafil may act by a mechanism independent of NO/cGMP pathway and this mechanism contributes to its smooth muscle relaxant effect.

KATP-channel-induced vasodilation is modulated by the Na,K-pump activity in rabbit coronary small arteries

The purpose of the study was to evaluate the importance of the Na,K-pump in relaxations induced by K(ATP)-channel openers in rabbit coronary small arteries. Arterial segments were mounted in myographs for recording of isometric tension. Whole-cell patch clamp was used to assess K(ATP)-channel currents in isolated smooth muscle cells from the arteries. In arteries preconstricted with the thromboxane A(2) analogue U46619 pinacidil and cromakalim induced concentration-dependent relaxations. In arteries preconstricted with potassium (124 mM) only high concentrations of pinacidil had a small relaxant effect. In arteries preconstricted with U46619 pinacidil-induced relaxations were unaffected by pretreatment with N(omega)-nitro-L-arginine (L-NNA) and only slightly reduced after mechanical removal of the endothelium. Pinacidil induced relaxations were not significantly affected by 1 microM glibenclamide. However, the relaxations were partly inhibited in potassium-free media and by 1 microM ouabain. In contrast, the concentration-dependent relaxation to cromakalim was partly blocked by 1 microM glibenclamide and partly by 1 microM ouabain and when both drugs were present the inhibition increased. Ouabain (1 microM) and glibenclamide (1 microM) each partly inhibited an ATP-sensitive current induced by pinacidil and cromakalim. In the presence of both inhibitors a greater inhibition was seen. When the solution in the patch pipette was sodium-free the current was reduced and ouabain had no effect. The study suggests that the relaxation to cromakalim and most likely pinacidil is mediated through opening of K(ATP) channels. Inhibition of the Na,K-pump, however, may change the local environment for the K(ATP) channels (i.e. increases the ATP/ADPratio and/or decreases the transmembrane potassium gradient), which partly prevents the activation of the K(ATP)-channel current.

Inhibition of human and pig ureter motility in vitro and in vivo by the K(+) channel openers PKF 217-744b and nicorandil

The relaxing property of the K(+) channel opener and nitric oxide donor nicorandil and the new K(+) channel opener PKF 217-744b was investigated on isolated human ureteral tissue in vitro and in intact ureters of anesthetized pigs in vivo. In addition, nicorandil and its antagonists, glibenclamide and methylene blue, were tested on isolated pig ureter tissue in vitro. Nicorandil decreased the frequency of spontaneous contractions in isolated pig ureter rings. This effect was antagonized by glibenclamide and methylene blue suggesting that the nicorandil induced relaxation of the ureter is mediated by activation of ATP-sensitive K(+) channels and involvement of soluble guanylate cyclase. Moreover, nicorandil and PKF 217-744b reduced the amplitude of electrically induced contractions in isolated human ureter rings. Calculations of EC(50) values showed that PKF 217-744b [EC(50) = 4.83 x 10(-8) M] was more potent than nicorandil [EC(50) = 4.38 x 10(-5) M]. Both drugs reduced the contraction frequency of the pig ureter after intravenous and topical administration in vivo. Intravenous, but not topical, administration of nicorandil and PKF 217-744b significantly decreased arterial blood pressure but did not affect the heart rate. The in vitro findings suggest that K(+) channel opening and nitric oxide release mediate the effect of nicorandil. Our in vivo results indicate that PKF 217-744b and nicorandil are promising drugs for clinical application in patients with acute stone colic to relieve obstruction and facilitate stone passage or to relax the ureter before ureteroscopy.

Interaction of cyclic AMP modulating agents with levcromakalim in the relaxation of rat isolated mesenteric artery

The effect of cyclic AMP modulating agents on levcromakalim-induced relaxation was investigated in myograph-mounted rat mesenteric arteries. Forskolin (adenylyl cyclase activator), dibutyryl cyclic AMP (protein kinase A activator) and 5'-N-ethylcarboxamidoadenosine (NECA; adenosine receptor agonist) all potentiated the vasorelaxant effects of levcromakalim. The modulatory and relaxant effects of dibutyryl cyclic AMP, NECA and forskolin were sensitive to the protein kinase A inhibitor, Rp-cAMPS. However, relaxation to these three agents was unaffected by the K(ATP) inhibitor, glibenclamide. Dibutyryl cyclic AMP and NECA also caused levcromakalim to induce relaxation in the sub-nanomolar concentration range, however, this effect was Rp-cAMPS- and glibenclamide-insensitive. These results suggest that cyclic AMP modulating agents modulate K(ATP), even though this channel does not contribute to their relaxant effects.

Characterization of adenosine action in isolated rat renal artery. Possible role of adenosine A(2A) receptors

Adenosine (0.1-300 microM) induced concentration- and endothelium-dependent relaxation of rat renal artery (RRA). N(G)-Nitro-L-arginine (L-NOARG, 10 microM) significantly reduced adenosine-elicited dilatation, but not the application of indomethacin (10 microM), ouabain (100 microM) or tetraethylammonium (TEA, 500 microM). In the presence of high concentration of K(+) (100 mM) or glibenclamide (1 microM), adenosine-evoked relaxation was almost abolished. 8-(3-Chlorostyril)caffeine (CSC, 0.3-3 microM), a selective A(2A)-antagonist, significantly reduced adenosine-evoked dilatation in a concentration-dependent manner (pA(2)=7.29). Conversely, 1,3-dipropyl-8-cyclopentylxanthine (DPCPX, 10 nM), an A(1)-antagonist, did not alter adenosine-induced relaxation. These results indicate that adenosine produces endothelium-dependent relaxation of isolated RRA. Dilatation evoked by adenosine is mediated by predominant releasing of endothelium-derived hiperpolarizing factor (EDHF) and also in one part of nitric oxide (NO) from endothelial cells. The obtained results also suggest that RRA response to adenosine is most likely initiated by activation of endothelial adenosine A(2A) receptors.

Levcromakalim causes indirect endothelial hyperpolarization via a myo-endothelial pathway

1. Effects of K+ channel opener, levcromakalim, on vascular endothelial cells were examined. Under voltage- and current-clamp conditions, application of acetylcholine to dispersed endothelial cells isolated from rabbit superior mesenteric artery (dispersed RMAECs) produced hyperpolarization and outward currents. On the other hand, dispersed RMAECs did not respond to levcromakalim. 2. When membrane potential was recorded from endothelium in a mesenteric arterial segment, exposure to levcromakalim in a concentration range of 0.1 to 3 microM caused concentration-dependent hyperpolarization. The hyperpolarization was observed in the absence of external Ca2+ and was inhibited by 10 microM glibenclamide. 3. The presence of 1 mM heptanol did not affect the levcromakalin-induced hyperpolarization, whereas treatment of the mesenteric arterial segment with 20 microM 18 beta-glycyrrhetinic acid significantly reduced the hyperpolarization. The response to acetylcholine of RMAECs in an arterial segment with 18 beta-glycyrrhetinic acid was, however, similar to that without 18 beta-glycyrrhetinic acid. 4. These suggest that although RMAECs themselves are functionally insensitive to levcromakalim, those in an arterial segment are hyperpolarized by levcromakalim via myo-endothelial electrical communication.

Vasodilator effects of sodium nitroprusside, levcromakalim and their combination in isolated rat aorta

1. The endothelial modulation of the relaxant responses to the nitric oxide (NO) donor sodium nitroprusside (SNP) and the KATP channel opener levcromakalim (LEM) and the interactions between these agents were analysed in isolated rat aorta. 2. LEM-induced relaxation was unchanged by endothelium removal or by the presence of L-NAME (10-4 M) or ODQ (10-6 M). In contrast, in KCl- (25 mM), but not in noradrenaline- (NA, 10-6 M) contracted arteries, SNP-induced relaxation was augmented by endothelium removal but not by L-NAME, indomethacin, glibenclamide nor charybdotoxin plus apamin. 3. The isobolographic analysis of the interactions between exogenously activated KATP channels and cyclic GMP using mixtures of SNP and LEM revealed that there were no interactions between both drugs at the proportions at which both drugs were active. However, the points for the SNP : LEM mixtures in proportions 10:1 and 1:10,000 (i.e. at concentrations at which LEM and SNP were inactive, respectively) fell significantly above the line of additivity indicating that there were negative interactions between both drugs at these selected proportions (about 5- and 2 fold inhibition, respectively). The former interaction was sensitive to glibenclamide, whereas the latter was insensitive ODQ. The magnitude of the 10:1 SNP:LEM interaction was smaller in endothelium-intact arteries and was absent in arteries stimulated by NA. 4. In conclusion, the relaxations induced by LEM and SNP were additive. However, the presence of endothelium and low concentrations of LEM inhibited SNP-induced relaxation. Both inhibitory effects were not additive and were only observed in KCl- and not in NA-contracted aortae.

EDHF-mediated relaxation is impaired in fructose-fed rats

Insulin resistance (IR) is associated with endothelial dysfunction. A defect in endothelium-dependent relaxation via outward potassium conductance has been observed in mesenteric arteries from IR rats. The purpose of this study was to assess whether this defect in endothelium-dependent relaxation was due to impaired endothelium-derived hyperpolarizing factor (EDHF) and to determine which specific potassium channel(s) are involved in relaxation. This was accomplished by using specific potassium channel inhibitors in the presence of nitric oxide synthase and cyclooxygenase inhibition. In addition, we sought to assess the function of smooth muscle cell adenosine triphosphate (ATP)-dependent potassium (K(ATP)) channels. Sprague-Dawley rats were randomized to control or IR. To determine EDHF-mediated relaxation, acetylcholine (ACh)-induced (10(-9)-10(-5) M) relaxation was measured (in vitro) in mesenteric arteries in the presence of indomethacin (10(-5) M) and N-nitro-L-arginine (L-NNA) (10(-4) M). Subsequently the combination of charybdotoxin (CTX) (0.1 microM) and apamin (0.5 microM) or glibenclamide (Glib) (10 microM) was added to the bath to inhibit KCa or K(ATP), respectively. In separate experiments, relaxation to pinacidil (10(-13)-10(-5) M), a K(ATP) activator, was assessed in vessels with intact endothelium, endothelium denuded, or with L-NNA. Maximal relaxation to ACh in the presence of L-NNA and indomethacin was 68+/-6% for control and 12+/-3% for IR (p<0.01). The addition of CTX + apamin almost abolished EDHF-mediated relaxation in control (Emax, 8+/-5% vs. 68+/-6%; p<0.01), whereas Glib had little affect. Neither CTX + apamin nor Glib had any affect on IR. Additionally, IR arteries were less sensitive to pinacidil than were controls (EC50, 1.5+/-0.9 microM vs. 5x10(-4)+/-3x10(-4) microM, respectively; p<0.01). Endothelial removal or L-NNA pretreatment of control arteries decreased the response to pinacidil similar to IR, whereas IR vessels were unaffected. EDHF-mediated relaxation is impaired in IR arteries. In addition, the K(Ca) channel appears to be imperative for activity of EDHF in rat small mesenteric arteries. Moreover, activation of K(ATP) channels by pinacidil is impaired in IR, and this appears to be a result of endothelial dysfunction.

Involvement of ATP-sensitive potassium channels in a model of a delayed vascular hyporeactivity induced by lipopolysaccharide in rats

We have investigated the role of ATP-sensitive potassium (K(ATP)) channels in an experimental model of a delayed phase of vascular hyporeactivity induced by lipopolysaccharide (LPS) in rats. After 24 h, from LPS treatment, in anaesthetized rats the bolus injection of phenylephrine (PE) produced an increase in mean arterial pressure (MAP) significantly (P<0.05) reduced in LPS-treated rats compared to the vehicle-treated rats. This reduction was prevented by pre-treatment of rats with glibenclamide (GLB), a selective inhibitor of K(ATP) channels. GLB administration did not affect the MAP in vehicle-treated rats but produced an increase of MAP in rats treated with LPS. Cromakalim (CRK), a selective K(ATP) channel opener, produced a reduction of MAP that was significantly (P<0.05) higher in LPS- than in vehicle-treated rats. In contrast, the hypotension induced by glyceryl trinitrate (GTN) in LPS-treated rats was not distinguishable from that produced in vehicle-treated rats. Experiments in vitro were conducted on aorta rings collected from rats treated with vehicle or LPS 24 h before sacrifice. The concentration-dependent curve to PE was statistically (P<0.005) reduced in aorta rings collected from LPS- compared to vehicle-treated rats. This difference was totally abolished by tetraethylammonium (TEA), a non-selective inhibitor of K+ channels. CRK produced a relaxation of PE precontracted aorta rings higher in rings from LPS- than in vehicle-treated rats. GLB inhibited CRK-induced relaxation in both tissues, abolishing the observed differences. In conclusion, our results indicate an involvement of K(ATP) channels to the hyporesponsiveness of vascular tissue after 24 h from a single injection of LPS in rats. We can presume an increase in the activity of K(ATP) channels on vascular smooth muscle cells but we cannot exclude an increase of K(ATP) channel number probably due to the gene expression activation.

Cellular mechanisms by which adenosine induces vasodilatation in rat skeletal muscle: significance for systemic hypoxia

1. In anaesthetized rats, we recorded arterial blood pressure (ABP), heart rate (HR), femoral blood flow (FBF) and femoral vascular conductance (FVC). We tested the effects of the nitric oxide (NO) synthesis inhibitor L-NAME (nitro-L-arginine methyl ester), or the ATP-sensitive K+ (KATP) channel inhibitor glibenclamide, on responses evoked by systemic hypoxia (breathing 8% O2 for 5 min) or i.a. infusion for 5 min of adenosine, the NO donor sodium nitroprusside (SNP), the adenosine A1 receptor agonist CCPA (2-chloro-N6-cyclopentyladenosine) or the adenosine A2A receptor agonist CGS 21680 (2-p-(2-carboxyethyl)-phenethylamino-5'-N-ethylcarboxamidoadeno sin e hydrochloride). 2. L-NAME (10 mg kg-1 i.v.) greatly reduced the increase in FVC induced by hypoxia or adenosine, as we have shown before, but had no effect on the increase in FVC evoked by SNP. In addition, L-NAME abolished the increase in FVC evoked by CCPA and greatly reduced that evoked by CGS 21680. These results substantiate the view that muscle vasodilatation induced by systemic hypoxia and infused adenosine are largely NO dependent. They also indicate that muscle dilatation induced by A1 receptor stimulation is entirely NO dependent while that induced by A2A receptors is largely NO dependent; dilatation may also be induced by direct stimulation of A2A receptors on the vascular smooth muscle. 3. Glibenclamide (10 or 20 mg kg-1 i.v.) reduced the increase in FVC induced by hypoxia, preferentially affecting the early part (< 1 min). In addition, glibenclamide greatly reduced the increase in FVC induced by adenosine, but it had no effect on that evoked by SNP. Further, glibenclamide abolished the increase in FVC evoked by CCPA and greatly reduced that evoked by CGS 21680. These results substantiate the view that hypoxia-induced muscle vasodilatation is initiated by KATP channel opening. They also indicate that NO does not induce muscle vasodilatation by opening KATP channels on the vascular smooth muscle, but indicate that the dilatation induced by adenosine and by A2A receptor stimulation is largely dependent on KATP channel opening, while that induced by A1 receptor stimulation is wholly dependent on KATP channel opening. 4. These results, together with previous evidence that hypoxia-induced vasodilatation in skeletal muscle is largely mediated by adenosine acting on A1 receptors, lead us to propose that adenosine is released from endothelium during systemic hypoxia and acts on endothelial A1 receptors to open KATP channels on the endothelial cells and cause synthesis of NO, which then acts on the vascular smooth muscle to cause dilatation. During severe systemic hypoxia we propose that adenosine may also act on A2A receptors on the endothelium to cause dilatation by a similar process and may act on A2A receptors on the vascular smooth muscle to cause dilatation by opening KATP channels.

Characterization of endothelium-dependent relaxations in mesenteries from transgenic hypertensive rats

Endothelial dysfunction has been reported to be a feature of hypertension. We have investigated the relative contributions of nitric oxide (NO) and the endothelium-derived hyperpolarizing factor (EDHF) to endothelium-dependent relaxations in isolated mesenteries from (mREN-2)-27 transgenic hypertensive (TGH) rats and their normotensive controls (Hannover Sprague-Dawley). Relaxation to the endothelium-dependent relaxant, carbachol, was unimpaired in mesenteries from TGH rats compared to the Hannover Sprague-Dawley controls. Inhibition of NO synthase (with 100 microM Nomega-nitro-L-arginine methyl ester) had greater inhibitory effects against these relaxations in the mesenteries from Hannover Sprague-Dawley compared to TGH. Inhibition of EDHF activity with high K+ also had greater inhibitory effects against endothelium-dependent relaxations in the mesenteries from the Hannover Sprague-Dawley compared to TGH. The present results show that, although endothelium-dependent relaxation is unimpaired in mesenteries from TGH rats, there are differences in the relative contributions of NO and EDHF, such that inhibition of either NO or EDHF alone in TGH mesenteries has less impact compared to Hannover Sprague-Dawley. It is suggested that the recently identified reciprocal relationship between NO and EDHF is upregulated in the mesenteries from the TGH rats.

Modulation of relaxation to levcromakalim by S-nitroso-N-acetylpenicillamine (SNAP) and 8-bromo cyclic GMP in the rat isolated mesenteric artery

1. Levcromakalim caused concentration-dependent relaxations of methoxamine-induced tone in both endothelium-denuded and intact vessels. Its potency was reduced by the nitric oxide donor, S-nitroso-N-acetylpenicillamine (SNAP; 0.1 microM or 1 microM) in both denuded and intact vessels. The maximal relaxation (Rmax) was reduced only in denuded vessels. 2. SNAP was more potent in endothelium-denuded than intact vessels but there were no differences in Rmax. Glibenclamide (10 microM) did not affect relaxation to SNAP in endothelium-denuded or intact vessels. 3. The soluble guanylyl cyclase inhibitor, 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ, 10 microM) increased the potency and Rmax of levcromakalim in endothelium-intact vessels. ODQ had no effect in denuded vessels. 4. ODQ (10 microM) reduced the vasorelaxant potency of SNAP in both intact and endothelium-denuded vessels by 190-fold and 620-fold, respectively. 5. 8-bromo cyclic GMP (10 or 30 microM) reduced both the potency and Rmax of levcromakalim in de-endothelialized vessels, but had no effect in intact vessels although it reduced both the potency and Rmax of levcromakalim in intact vessels incubated with ODQ (10 microM). 6. In the presence of ODQ (10 microM), SNAP (0.1 microM or 1 microM) reduced the potency of levcromakalim in intact vessels, without altering Rmax, but had no effect in denuded vessels. SNAP (50 microM) reduced both the potency and Rmax of levcromakalim in intact and endothelium-denuded vessels. 7. Therefore, although SNAP causes relaxation principally through generation of cyclic GMP, it can modulate the actions of levcromakalim through mechanisms both dependent on, and independent of, cyclic GMP; the former predominate in endothelium-denuded vessels and the latter in intact vessels.

Basal nitric oxide release differentially modulates vasodilations by pinacidil and levcromakalim in goat coronary artery

In the current investigation, the role of basal nitric oxide (NO) in modulating the vasorelaxant responses to pinacidil and levcromakalim was examined in goat isolated coronary artery. Pinacidil (10(-8) 10(-4) M) elicited concentration-dependent relaxations of the coronary artery ring segments (with intact endothelium) constricted with 30 mM K+ saline solution. The EC50 of the vasodilator was 2.57 x 10(-6) M (95% CL, 1.9-3.46 x 10(-6) M). The removal of endothelium by mechanical rubbing caused a rightward shift in the concentration-response curve of pinacidil with a corresponding increase in EC50 value (1.90 x 10(-5) M; 95% CL, 1.12-3.23 x 10(-5) M). Similar to endothelium removal, treatment of endothelium-intact rings either with the NO synthesis inhibitor L-NAME (NG-nitro-L-arginine methyl ester; 3 x 10(-5) M) or the guanylate cyclase inhibitor, methylene blue (3 x 10(-6) M) resulted in a marked inhibition in the relaxant responses to pinacidil. Hence, the EC50 values of the potassium channel opener were significantly higher in tissues treated either with L-NAME (7.41 x 10(-6) M; 95% CL, 6.02-9.12 x 10(-6) M) or methylene blue (2.29 x 10(-5) M; 95% CL, 1.58-3.31 x 109-5) M) as compared to untreated controls. The ATP-sensitive potassium (KATP) channel blocker glibenclamide, which caused a significant rightward shift in the concentration-relaxation curve of pinacidil in control tissues, was found to be less potent in antagonising the relaxant responses of the KATP channel opener in endothelium-denuded rings and in rings with intact endothelium but treated with either L-NAME or methylene blue. In contrast to the observations made with pinacidil, the vasodilator responses to another KATP channel opener, levcromakalim, were potentiated in the absence of basal NO. Thus, the EC50 of levcromakalim was 1.33 x 10(-8) M (95% CL, 0.8-2.21 x 10(-8) M) in control tissues with intact endothelium, which was significantly higher than those obtained in endothelium-deprived rings (4.81 x 10(-9) M; 95% CL, 4.04-5.73 x 10(-9) M) or endothelium intact rings treated either with L-NAME (2.63 x 10(-9) M; 95% CL, 1.58-4.36 x 10(-9) M) or methylene blue (2.82 x 10(-9) M; 95% CL, 1.7-4.68 x 10(-9) M). The selective modulation by basal NO of the arterial relaxations elicited with the KATP channel openers was evident from the findings that papaverine-induced relaxations were not affected in the absence of basal NO. Taken together, the results of the present study suggest that basal NO differentially modulates the interaction of pinacidil and levcromakalim with the KATP channels in goat coronary artery through a cGMP-dependent pathway.

Role of K+ channel opening and stimulation of cyclic GMP in the vasorelaxant effects of nicorandil in isolated piglet pulmonary and mesenteric arteries: relative efficacy and interactions between both pathways

1. The effects of the K+ channel opener levcromakalim, the guanylate cyclase stimulant nitroprusside and the dual drug nicorandil (K+ channel opener and guanylate cyclase stimulant) were analysed in piglet isolated endothelium-denuded pulmonary (PA) and mesenteric (MA) arteries stimulated by noradrenaline (NA) or by the thromboxane A2 mimetic U46619. 2. Nicorandil, levcromakalim and verapamil were less potent in PA than in MA, the efficacy of levcromakalim was also reduced in PA. The effects of nicorandil and levcromakalim were similar in arteries pre-contracted by NA and U46619, whereas verapamil was more potent in arteries pre-contracted by NA. Nitroprusside was equipotent in MA pre-contracted by either NA or U46619 and in PA pre-contracted by NA whereas in PA pre-contracted by U46619, nitroprusside showed lower potency and efficacy. 3. The relaxant effects of levcromakalim and nitroprusside were inhibited by 10(-5) M glibenclamide and 10(-6) M ODQ, respectively. Nicorandil-induced relaxation was inhibited by ODQ in all experimental conditions, whereas glibenclamide had inhibitory effects in PA and MA pre-contracted by U46619, had no effect in PA pre-contracted by NA and in MA pre-contracted by NA it was only inhibitory in the presence of ODQ. 4. No apparent interactions were found between nitroprusside and levcromakalim as indicated by the lack of effects of pretreatment with one of them (producing 20-35% relaxation) on the potency of the relaxant response to the other. However, in PA pre-contracted by U46619, where nitroprusside or levcromakalim induced only partial relaxation, the combination of both mechanisms (either by combining nitroprusside plus levcromakalim or by nicorandil) was able to induce full vasodilatation. 5. In conclusion, K+ channel opening and guanylate cyclase stimulation are independent pathways that induce additive vasorelaxation in piglet PA and MA. The mechanism of action of nicorandil is dependent on the artery and on the nature of the agonist employed to precontract the artery. The relative efficacy of K+ channel opening vs guanylate cyclase stimulation may partially explain the preferential contribution of each mechanism to the relaxant effects of nicorandil.

Endothelium and cannabinoid receptor involvement in levcromakalim vasorelaxation

Levcromakalim was more potent at relaxing rat small mesenteric arteries with endothelium (EC50, 84+/-10 nM) than denuded vessels (EC50, 779+/-101 nM). The cannabinoid receptor antagonist SR 141716A (N-(piperidin-1-yl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-me thyl-1H-pyrazole-3-carboxamide hydrochloride; 1 microM) shifted the levcromakalim concentration/response curve 7.6-fold rightwards in intact vessels but had no effect in de-endothelialised vessels. Similar effects occurred with pinacidil. Combination of the K+ channel blockers apamin (1 microM) and charybdotoxin (100 nM) shifted the levcromakalim concentration/response curve 3-fold rightwards only in intact vessels. It is concluded that levcromakalim and pinacidil relax mesenteric arteries partly by releasing a relaxing factor from endothelium, possibly an endogenous cannabinoid.

Comparative pharmacology of endothelium-derived hyperpolarizing factor and anandamide in rat isolated mesentery

We have recently proposed that anandamide, or a related cannabinoid, is the endothelium-derived hyperpolarizing factor (EDHF) and have now compared EDHF-mediated responses (induced by carbachol in the presence of both nitric oxide and prostanoid synthesis inhibitors) with those induced by anandamide in the rat isolated superior mesenteric arterial bed. Both EDHF-mediated and anandamide-induced relaxations were inhibited in the presence of high K+ (60 mM) and opposed by blockade of K+ channels with 10 mM tetraethylammonium. The cytochrome P450 inhibitors, and putative EDHF inhibitors, clotrimazole (10 microM) and proadifen (SKF 525A) (10 microM), opposed both anandamide-induced and EDHF-mediated relaxations and also relaxant responses to the K+ channel activator levcromakalim. Therefore, EDHF-mediated and anandamide-induced vasorelaxations show very similar pharmacological characteristics, with both responses being mediated via K+ channel activation. Further, the actions of EDHF and anandamide are both sensitive to proadifen and clotrimazole, EDHF antagonists which appear to act through K+ channel inhibition. Accordingly, these results support our proposal that an endocannabinoid is an EDHF.

Synergistic interaction between endothelium-derived NO and prostacyclin in pulmonary artery: potential role for K+ATP channels

1. The aim of the present study was to assess interactions between nitric oxide (NO) and prostacyclin (PGI2) during endothelium-dependent relaxations evoked by bradykinin, calcium ionophore (A23187) and acetylcholine in canine isolated pulmonary artery. 2. Relaxations to low concentrations of bradykinin and A23187 were abolished by combined inhibition of NO-synthase (by N omega-nitro-L-arginine methyl ester L-NAME, 30 microM) and cyclo-oxygenase (indomethacin, 10 microM), suggesting mediation by NO and PGI2. The individual contributions of NO and PGI2 to the dilator responses were quantified by use of areas above the separate indomethacin-insensitive and L-NAME-insensitive components of the concentration-effect curves, respectively. Individually, NO and PGI2 accounted for only 53 +/- 5% and 16 +/- 9% of total bradykinin-induced relaxation, and 46 +/- 10% and 20 +/- 9% of total A23187-induced relaxation, suggesting that NO and PGI2 acted synergistically to cause endothelium-dependent relaxation. 3. Relaxation to low concentrations of acetylcholine was abolished by L-NAME but not affected by indomethacin, suggesting the response was mediated solely by NO with no interaction from PGI2. 4. Glibenclamide (1 microM), an inhibitor of ATP-sensitive potassium (K+ATP) channels, inhibited responses to bradykinin or A23187 but did not affect relaxations evoked by acetylcholine. Glibenclamide did not affect endothelium-independent relaxations to PGI2 or the NO-donor, 3-morpholinosydnonimine (SIN-1). 5. With bradykinin, glibenclamide attenuated total relaxation by 49 +/- 8%, but did not alter the individual NO and PGI2-mediated components of the response. Glibenclamide abolished the synergistic interaction between endothelium-derived NO and PGI2. 6. At high concentrations, bradykinin, A23187 or acetylcholine caused endothelium-dependent relaxation that was insensitive to L-NAME + indomethacin. With bradykinin or A23187, this component of relaxation was inhibited by glibenclamide, whereas with acetylcholine, glibenclamide had no effect. 7. The synergistic interaction between endothelium-derived NO and PGI2 in canine pulmonary artery is mediated by activation of K+ATP channels, presumably by an endothelium-derived hyperopolarizing factor (EDHF). The pattern of endothelial dilator mediators and the presence of this synergistic interaction is dependent on the nature of the endothelial stimulus.

Characterization and modulation of EDHF-mediated relaxations in the rat isolated superior mesenteric arterial bed

1. We have used the isolated, buffer-perfused, mesenteric arterial bed of the rat (preconstricted with methoxamine or 60 mM K+) to characterize nitric oxide (NO)-independent vasorelaxation which is thought to be mediated by the endothelium-derived hyperpolarizing factor (EDHF). 2. The muscarinic agonists carbachol, acetylcholine (ACh) and methacholine caused dose-related relaxations in preconstricted preparations with ED50 values of 0.18 +/- 0.04 nmol (n = 8), 0.05 +/- 0.02 nmol (n = 6) and 0.26 +/- 0.16 nmol (n = 5), respectively. In the same preparations NG-nitro-L-arginine methyl ester (1-NAME, 100 microM) significantly (P < 0.05) decreased the potency of all the agents (ED50 values in the presence of L-NAME: carbachol, 0.66 +/- 0.11 nmol; ACh, 0.28 +/- 0.10 nmol; methacholine, 1.97 +/- 1.01 nmol). The maximal relaxation to ACh was also significantly (P < 0.05) reduced (from 85.3 +/- 0.9 to 73.2 +/- 3.7%) in the presence of L-NAME. The vasorelaxant effects of carbachol were not significantly altered by the cyclo-oxygenase inhibitor indomethacin (10 microM; n = 4). 3. The K+ channel blocker, tetraethylammonium (TEA, 10 mM) also significantly (P < 0.001) reduced both the potency of carbachol (ED50 = 1.97 +/- 0.14 nmol in presence of TEA) and the maximum relaxation (Rmax = 74.6 +/- 3.2% in presence of TEA, P < 0.05, n = 3). When TEA was added in the presence of L-NAME (n = 4), there was a further significant (P < 0.001) decrease in the potency of carbachol (ED50 = 22.4 +/- 13.5 nmol) relative to that in the presence of L-NAME alone, and Rmax was also significantly (P < 0.05) reduced (74.6 +/- 4.2%). The ATP-sensitive K+ channel inhibitor, glibenclamide (10 microM), had no effect on carbachol-induced relaxation (n = 9). 4. High extracellular K+ (60 mM) significantly (P < 0.01) reduced the potency of carbachol (n = 5) by 5 fold (ED50: control, 0.16 +/- 0.04 nmol; high K+, 0.88 +/- 0.25 nmol) and the Rmax was also significantly (P < 0.01) reduced (control, 83.4 +/- 2.7%; high K+, 40.3 +/- 9.2%). The residual vasorelaxation to carbachol in the presence of high K+ was abolished by L-NAME (100 microM; n = 5). In preparations preconstricted with high K+, the potency of sodium nitroprusside was not significantly different from that in preparations precontracted with methoxamine, though the maximal response was reduced (62.4 +/- 3.4% high K+, n = 7; 83.1 +/- 3.1% control, n = 7). 5. In the presence of the cytochrome P450 inhibitor, clotrimazole (1 microM, n = 5 and 10 microM, n = 4), the dose-response curve to carbachol was significantly shifted to the right 2 fold (P < 0.05) and 4 fold (P < 0.001) respectively, an effect which was further enhanced in the presence of L-NAME. Rmax was significantly (P < 0.01) reduced by the presence of 10 microM clotrimazole alone, being 86.9 +/- 2.5% in its absence and 61.8 +/- 7.8% in its presence (n = 6). 6. In the presence of the cell permeable analogue of cyclic GMP, 8-bromo cyclic GMP (6 microM), the inhibitory effects of L-NAME on carbachol-induced relaxation were substantially enhanced (ED50: L-NAME alone, 0.52 +/- 0.11 nmol, n = 5; L-NAME + 8-bromo cyclic GMP, 1.42 +/- 0.28 nmol, n = 7, Rmax: L-NAME alone, 82.2 +/- 2.4%; L-NAME + 8-bromo cyclic GMP, 59.1 +/- 1.8%. P < 0.001). These results suggest that the magnitude of the NO-independent component of vasorelaxation is reduced when functional cyclic GMP levels are maintained, suggesting that basal NO (via cyclic GMP) may modulate EDHF activity and, therefore, on loss of basal NO production the EDHF component of endothelium-dependent relaxations becomes functionally greater. 7. The present investigation demonstrates that muscaranic receptor-induced vasorelaxation in the rat mesenteric arterial bed is mediated by both NO-dependent and independent mechanisms. The L-NAME-insensitive mechanism, most probably occurs via activation of a K+ conductance and shows the characteristics of EDHF-mediated responses. Finally, the results demonstrate that EDHF activity may become upregulated on inhibition of NO production and this may compensate for the loss of NO.

Endothelium-mediated and N omega-nitro-L-arginine methyl ester-sensitive responses to cromakalim and diazoxide in the rat mesenteric bed

The effects of two 'K+ channel openers', (+/-)-6-cyano-3,4-dihydro-2,2-dimethyl-trans-4-(2-oxo-1-pyrrolidyl )-2 H-benzo[b]-pyran-3-ol (cromakalim) and 7-chloro-3-methyl-2 H-1,2,4-benzothiadiazine 1,1-dioxide (diazoxide), were studied on the rat isolated mesenteric bed. Differences in the perfusion pressure were measured as a parameter of vascular resistance. Cromakalim (0.1-700 microM) and diazoxide (1 microM-1 mM) reduced to 60% the contractions elicited by 10 microM noradrenaline and to 30% those evoked by 100 mM KCl. The relaxant effects of cromakalim and diazoxide on the noradrenaline-induced contractions were reduced by the K(+)-ATP channel blocker, 5-chloro-N-[2-[4-[[[(cyclohexylamino) carbonyl]amino]-sulfonyl]phenyl]ethyl]-2-methoxybenzamide (glibenclamide, 0.01-0.3 microM), endothelium removal with 0.1% saponin and pretreatment with the nitric oxide synthesis inhibitor, S(+/-)-N5-[imino(nitroamino)methyl]-L-ornithine methyl ester hydrochloride (L-NAME, 500 microM). Reductions in the relaxant responses after endothelium removal or L-NAME pretreatment were observed with 1-100 microM cromakalim and with 30 microM diazoxide but not with 100 and 300 microM diazoxide. Pretreatment with the inactive stereoisomer D-NAME as well as with the prostanoid synthesis inhibitor, 1-[p-chlorobenzoyl]-5-methoxy-2-methylindole-3-acetic acid (indomethacin, 10 microM), did not affect the reductions in contractile responses to noradrenaline caused by either cromakalim or diazoxide. It is concluded that the relaxant effects of cromakalim and diazoxide in the rat mesenteric bed are endothelium-mediated and L-NAME-sensitive and could at least partially involve the participation of nitric oxide.