Isoprenaline and fenoterol relax the arterial smooth muscle through activation of β-adrenoceptors (1). Increased levels of cytosolic cyclic adenosine monophosphate (cAMP) are thought to mediate the isoprenaline response (2). More recently, additional sites have been reported for the action of isoprenaline in blood vessels. For example, isoprenaline might activate adenosine triphosphate (ATP)-sensitive K+ channels (KATP channels) through a β1-adrenoceptor in the rat isolated perfused mesenteric arterial bed (3), in dog coronary artery (4), and in rat isolated aortic rings (5), whereas electro-physiologic study has shown that isoprenaline could hyperpolarize vascular smooth muscle of the canine saphenous vein, probably through opening of the KATP channels (6). These reports indicate that KATP channels play a significant role in the vasorelaxant effect of isoprenaline, which seems independent of the elevation of cytosolic cAMP (3). In view of recent findings that cAMP and cAMP-dependent protein kinase could activate Ca2+-activated K+ channels (KCa channels) in cultured smooth-muscle cells from porcine coronary artery (7), and KCa channel blockers such as iberiotoxin and tetraethylammonium ions inhibited coronary vasodilation induced by adenosine and cAMP (8), it is possible that some other K+ channels might be also involved in β-adrenoceptor-mediated vascular response.
Our study was intended to examine whether KATP channels are involved in β-adrenoceptor-mediated vasorelaxant responses in the arterial rings isolated from the rat superior mesenteric artery and whether KCa channel or other K+ channels also contribute to the vasodilator effects of isoprenaline and fenoterol. Experimentally, the vasorelaxant effects of isoprenaline, fenoterol, forskolin, and cromakalim have been compared in the absence and presence of various blockers of K+ channels.
Male Sprague-Dawley rats of ∼300 g were killed by cervical dislocation and bled. The main branch of the superior mesenteric artery was dissected, and the surrounding connective tissues were cleaned. Arterial rings of ∼3 mm in length were prepared and placed in 10-ml organ baths containing the Krebs-Henseleit solution of the following composition (mM): NaCl, 119; KCl, 4.7; CaCl2, 2.5; MgCl2, 1; NaHCO3, 25; KH2PO4, 1.2; and D-glucose, 11.1. The bath solution was constantly gassed with 95% O2 + 5% CO2 and maintained at 37°C. Tissues were allowed to equilibrate for 90 min under 0.5 g resting tension. The isometric tension was measured with force-displacement transducers (Grass Instrument Co.). In some experiments, the endothelium was mechanically disrupted by rubbing the lumen of the artery with plastic tubing. Successful removal of the endothelial layer was verified by the lack of the relaxant response of the preparation to acetylcholine (1 μM) at the beginning of each experiment.
Twenty minutes after setting up in the organ baths, preparations were first contracted with a single concentration of phenylephrine (1 μM) to test for their contractile responses, after which they were washed 3 times in Krebs-Henseleit solution to restore tension to the basal level. Tissues were then treated with the cyclooxygenase inhibitor indomethacin (3 μM) for the remainder of the experiment. Rings of artery were incubated with each putative K+ channel blocker for 20 min before they were contracted with 1 μM phenylephrine to establish the sustained tone, and then vasorelaxants were added cumulatively to induce concentration-dependent relaxation. In experiments using Ba2+, the phenylephrine-induced tone was significantly increased; therefore the concentration of phenylephrine was reduced to between 0.1 and 0.3 μM to match the initial tension in the absence of extracellular Ba2+.
The following drugs were used: phenylephrine hydrochloride, forskolin, glibenclamide, charybdotoxin (CTX), cromakalim, fenoterol, and acetylcholine chloride (Research Biochemicals International, Natick, MA, U.S.A.). Forskolin, glibenclamide, and cromakalim were dissolved in dimethyl sulfoxide (DMSO). DMSO at a final concentration of 0.2% (vol/vol) did not affect the isoprenaline response. Other chemicals were dissolved in Krebs-Henseleit solution.
The effects of vasodilators on the sustained tone were expressed as a percentage of the control value. Cumulative concentration-relaxation relations were analyzed with a nonlinear curve fitting by a logistic equation (Grafit; Erithacus Software Limited), and IC50 values were calculated as the drug concentration causing a half-maximal relaxation. The data were presented as mean ± SEM from n experiments. A level of probability of <0.05 obtained from unpaired Students' t test was considered as significant.
Effect of charybdotoxin on the vasorelaxant response to isoprenaline
Phenylephrine contracted the rat mesenteric artery containing intact endothelium with an EC50 of 0.136 ± 0.013 μM and maximal tension of 0.9 ± 0.04 g (n = 16). Sustained tension (0.53 ± 0.03 g; n = 8) was raised by phenylephrine at a submaximal concentration of 1 μM. Nonselective β-adrenoceptor agonist isoprenaline produced complete relaxation of phenylephrine-induced tension. The concentration-response curve for the vasorelaxant effect of isoprenaline in the absence of CTX is shown in Fig. 1A, and an IC50 of 0.0407 ± 0.0076 μM (n = 8) was calculated. CTX significantly shifted the concentration-relaxation curve for the vasorelaxant effect of isoprenaline to the right (p < 0.05). The IC50 values were 0.541 ± 0.082 (n = 6) and 7.99 ± 1.57 μM (n = 6) in the presence of CTX at 30 and 100 nM, respectively. In addition, CTX (100 nM) did not affect the basal tension and the phenylephrine-induced contractile response. Removal of endothelium shifted the isoprenaline concentration-relaxation curve to the right [IC50 of 2.56 ± 0.44 μM (n = 6), compared with an IC50 of 0.036 ± 0.011 μM (n = 7) in the presence of endothelium], and CTX (100 nM) further reduced the isoprenaline response (Fig. 1B).
Effect of CTX on the vasorelaxant responses to fenoterol and forskolin
Figure 2A shows that fenoterol, a β2-adrenoceptor agonist, concentration-dependently relaxed the phenylephrine-precontracted artery in the presence and absence of endothelium with respective IC50 values of 0.40 ± 0.09 (n = 11) and 21.09 ± 2.26 μM (n = 7). The maximal relaxation to fenoterol was reduced by 25.6 ± 5.0% (n = 7) on removal of endothelium. CTX (100 nM) significantly inhibited the fenoterol-induced vasorelaxation [IC50 of 7.78 ± 2.60 μM (n = 7), in the presence of endothelium; IC50 of 58.48 ± 7.63 μM (n = 6), in the absence of endothelium]. To investigate whether the cAMP-dependent mechanism is involved in the CTX-sensitive component of the vasorelaxant response of isoprenaline or fenoterol, the effect of forskolin in the absence and presence of CTX was compared. Figure 2B shows that CTX did not alter the vasorelaxant effect of forskolin (IC50 of 13.35 ± 2.48 nM (n = 6) in control, and IC50 of 11.17 ± 1.93 nM (n = 6) in 100 nM CTX]. Forskolin at 100 nM caused a complete relaxation and reduced the basal tone (Fig. 2B).
Effect of glibenclamide on the vasorelaxant responses to isoprenaline, fenoterol, forskolin, and cromakalim
Figure 3A shows that the concentration-dependent relaxation induced by isoprenaline was unaffected by the presence of glibenclamide (10 μM). IC50 values for the effect of isoprenaline were 0.0341 ± 0.0119 (n = 6) and 0.0297 ± 0.0086 μM (n = 6) for the absence and the presence of glibenclamide, respectively. Similarly, glibenclamide (10 μM) failed to influence the relaxant effect of fenoterol [IC50 of 0.41 ± 0.13 μM (n = 7) in control; IC50 of 0.36 ± 0.07 μM (n = 6) in glibenclamide; Fig. 3A]. Glibenclamide (10 μM) did not change the forskolin-induced relaxant effect [IC50 of 13.23 ± 1.14 nM (n = 5) in control and of 14.29 ± 1.95 nM (n = 6) in glibenclamide; Fig. 3B]. In contrast, glibenclamide at 3 μM caused a significant shift (p < 0.05) of the concentration-relaxation curve for cromakalim to the right, shown in Fig. 3C [IC50 of 0.0177 ± 0.0019 μM (n = 5) in control and of 0.8023 ± 0.3519 μM (n = 5) in glibenclamide]. Glibenclamide (10 μM) did not affect the induced tone.
Effect of Ba2+ on the vasorelaxant effects of isoprenaline and forskolin
Because Ba2+ augmented the initial tension caused by 1 μM phenylephrine [0.53 ± 0.05 g (n = 5) in control and 0.68 ± 0.04 g (n = 6) in 100 μM Ba2+, and 0.8 ± 0.05 g (n = 6) in 300 μM Ba2+], the induced initial tension was matched by reducing the concentration of phenylephrine to 0.1-0.3 μM. Figure 4A shows that Ba2+ shifted the concentration-response curve for the vasorelaxant effect of isoprenaline to the right in a concentration-dependent manner. In the presence of 300 μM Ba2+, ∼52% of the induced tone cannot be further reduced by isoprenaline. In addition, Ba2+ also inhibited the vasorelaxant response to forskolin [Fig. 4B; IC50 values: 15.02 ± 2.15 nM (n = 6) in control; 41.16 ± 8.08 nM (n = 5) in 100 μM Ba2+; 66.17 ± 10.42 nM (n = 5) in 300 μM Ba2+]. Unlike isoprenaline, forskolin was still able to cause complete relaxation in the presence of 300 μM Ba2+.
The results of our study demonstrate that in the rat isolated mesenteric artery, β-adrenoceptor-mediated activation of KCa channels might contribute in part to the vasorelaxant effects of nonselective β-adrenoceptor agonist isoprenaline and β2-adrenoceptor agonist fenoterol (1). KATP channels, however, did not appear to be involved in the effects of the β-adrenoceptor agonists.
In our experiments, CTX, an selective blocker of large-conductance KCa channels (9), decreased the vasodilator potency of isoprenaline and fenoterol in arteries with intact endothelium. Relaxation to isoprenaline and fenoterol was also inhibited by mechanical removal of the endothelium, suggesting that in this preparation, β-adrenoceptor agonists may act partially through endothelium-derived relaxing or hyperpolarizing factors. This result supports a role of endothelium in the β-adrenoceptor-mediated vasorelaxation previously reported in rat mesenteric arteries (10). On removal of the endothelium, CTX was still able to reduce significantly the relaxant effects of β-adrenoceptor agonists. These results point to the possible direct involvement of KCa channels in β-adrenoceptor-mediated vasorelaxation. However, factors from endothelium may also play a minor part in the CTX-sensitive portion of relaxation because nitric oxide has recently been shown to activate CTX-sensitive KCa currents in single smooth-muscle cells, which is likely responsible for nitric oxide-induced relaxation of rat pulmonary artery (11). More recently, isoprenaline was clearly demonstrated to increase tetraethylammonium-sensitive whole-cell KCa current (12) and to promote open-state probability of large-conductance KCa channels recorded from single smooth-muscle cells of guinea-pig basilar artery and rat aorta (12,13). Isoprenaline was previously shown to elevate intracellular levels of cAMP in vascular smooth muscle (2). cAMP or cAMP-dependent protein kinase activates KCa channels in coronary and basilar arteries (8,12) and in cultured smooth muscle from porcine coronary artery and rat aorta (7,13). In addition, the relaxant effect of isoprenaline or the β2-adrenoceptor agonist, salbutamol, in human or guinea pig airway smooth muscle was inhibited by CTX or iberiotoxin (14-16). Isoproterenol stimulates KCa channels through cAMP-dependent phosphorylation in cultured aortic smooth-muscle fibers (13). Therefore it is possible that cAMP mediates the coupling between β-adrenoceptor activation and KCa channels. However, the KCa channel blocker did not influence the vasorelaxant effect of forskolin, a direct activator of adenylyl cyclase. In coronary artery and tracheal smooth muscle cells, a direct G protein-mediated mechanism that is independent of phosphorylation by protein kinase A has been proposed for effects of β-adrenoceptor stimulation of KCa channels (17,18). These results indicate that there may be species or tissue differences in the way in which β-adrenoceptors are linked to K+ channels.
Our results show the lack of effect of glibenclamide on the vasodilator responses to the β-adrenoceptor agonists and forskolin in the isolated arterial rings of the rat mesentery, whereas glibenclamide inhibited the vasorelaxation induced by cromakalim, the KATP channel opener (19). In contrast, glibenclamide caused a significant shift of the concentration-response curve for the vasodilator response of isoprenaline to the right in the rat isolated perfused mesenteric arterial bed (3). In addition, glibenclamide prevents the metabolic vasodilation induced by β1-adrenoceptor stimulation in perfused dog coronary artery (4). It is not clear what might cause this difference in the way in which glibenclamide exerts its effect. It is hard to compare the induced tension in our experiments with the established perfusion pressure in the study of Randall and McCulloch (3). In their experiments, the whole mesenteric arterial bed was constantly perfused at 5 ml/min. It is therefore unknown whether the level of established tone or perfusion rate influences the effect of isoprenaline. Glibenclamide was previously shown to inhibit the flow-evoked endothelial release of ATP that caused vasodilation in the pulmonary vascular bed of rat (20). However, Nakashima and Vanhoutte (6) recently demonstrated that isoproterenol-induced hyperpolarization was mediated through cAMP-dependent opening of KATP channels in smooth muscle of the canine saphenous vein, whereas CTX did not affect the effects of isoproterenol. Glibenclamide did not alter the vasodilator effect of dibutyl cAMP in the rat perfused mesenteric arterial bed (3). Thus the K+ channels linked to the activation of adenylyl cyclase might differ among tissues and species. The mechanism underlying activation of KATP channels by the intracellular cAMP-dependent pathway remains to be elucidated.
Our results also show that Ba2+ attenuated the vasodilator responses to isoprenaline and forskolin. The effect of Ba2+ did not seem to be caused by its potentiation of the phenylephrine-induced tension because the concentration of α1-adrenoceptor agonist was reduced to match the initial tone induced by 1 μM phenylephrine in the absence of Ba2+. It appears that Ba2+-sensitive K+ channels may be involved in the effect of isoprenaline in isolated mesenteric artery of rat; this effect, however, might be nonspecific because Ba2+ also inhibited the forskolin-induced relaxation.
In summary, the findings of our investigation show a component of β-adrenoceptor-mediated vasorelaxant response is caused by the activation of KCa channels and probably Ba2+-sensitive K+ channels in the rat isolated mesenteric artery. The cAMP-dependent mechanism may not be involved in regulation of KCa channels. These results indicate an additional site for actions of β-adrenoceptor agonists in the artery. However, our work does not support the involvement of KATP channels in β-adrenoceptor-mediated vasodilation, as found in the whole perfused mesenteric arterial bed of rat (3) and in perfused dog coronary artery (4), suggesting that coupling activation of β-adrenoceptor to K+ channels may differ among tissues or species or both.
Acknowledgment: This work was supported by Hong Kong RGC and Direct Grants awarded to Y.H. (A/C 221401280 and A/C 220403710).
1. O'Donnell SR, Wanstall JC. Demonstration of both β1
- and β2
-adrenoceptors mediating relaxation of isolated ring preparations of rat pulmonary artery. Br J Pharmacol
2. Linz W, Wiemer G, Scholkens BA. Effects of colforsin, trequinsin and isoprenaline on norepinephrine-induced contractions and cyclic nucleotide levels of isolated vascular tissue. Arzneimittelforschung
3. Randall MD, McCulloch, AI. The involvement of ATP-sensitive potassium channels in β-adrenoceptor-mediated vasorelaxation in the rat isolated mesenteric arterial bed. Br J Pharmacol
4. Narishige T, Egashira K, Akatsuka Y, et al. Glibenclamide prevents coronary vasodilation induced by β1
-adrenoceptor stimulation in dogs. Am J Physiol
5. Hüsken BCP, Pfaffendorf M, van Zweiten PA. The influence of glucose and glibenclamide on vascular β-adrenoceptor-mediated responses. Br J Pharmacol
6. Nakashima M, Vanhoutte PM. Isoprenaline causes hyperpolarization through opening of ATP-sensitive potassium channels in vascular smooth muscle of the canine saphenous vein. J Pharmacol Exp Ther
7. Minami K, Fukuzawa K, Nakaya Y, Zeng XR, Inoue I. Mechanism of activation of the Ca2+
channel by cyclic AMP in cultured porcine coronary artery smooth muscle cells. Life Sci
8. Cabell F, Weiss DS, Price JM. Inhibition of adenosine-induced coronary vasodilation by block of large-conductance Ca2+
channels. Am J Physiol
9. Miller C, Moczydlowski E, Latorre R, Phillis M. Charybdotoxin, a protein inhibitor of single Ca2+
channels from skeletal muscle. Nature
10. Graves J, Poston L. β-Adrenoceptor agonist mediated relaxation of rat isolated resistance arteries: a role for the endothelium and nitric oxide. Br J Pharmacol
11. Archer SL, Huang JMC, Hampl V, Nelson DP, Shultz PJ, Weir EK. Nitric oxide and cGMP cause vasorelaxation by activation of a charybdotoxin-sensitive K+
channel by cGMP-dependent protein kinase. Proc Natl Acad Sci U S A
12. Song YM, Simard JM. β-Adrenoceptor stimulation activates large-conductance Ca2+
channels in smooth muscle cells from basilar artery of guinea pig. Pflugers Arch
13. Sadoshima JI, Akaike N, Kanaide H, Nakamura M. Cyclic AMP modulates Ca-activated K channel in cultured smooth muscle cells of rat aortas. Am J Physiol
14. Miura M, Belvisi MG, Stretton CD, Yacoub MH, Barnes PJ. Role of potassium channels in bronchodilator responses in human airways. Am Rev Resp Dis
15. Huang JC, Garcia ML, Reuben JP, Kacsorowski GJ. Inhibition of β-adrenoceptor agonist relaxation of airway smooth muscle by Ca2+
channel blockers. Eur J Pharmacol
16. Jones TR, Charette L, Garcia ML, Kaczorowski GJ. Interaction of iberiotoxin with β-adrenoceptor agonists and sodium nitroprusside on guinea pig trachea. J Appl Physiol
17. Scornik FS, Codina J, Bimbaumer L, Toro L. Modulation of coronary smooth muscle KCa
channels by Gs alpha independent of phosphorylation by protein kinase A. Am J Physiol
18. Kume H, Hall IP, Washabau RJ, Takagi K, Kotlikoff MI. β-Adrenergic agonists regulate KCa
channels in airway smooth muscle by cAMP-dependent and -independent mechanisms. J Clin Invest
19. Standen N, Quayle J, Davies N, Brayden J, Huang Y, Nelson M. Hyperpolarizing vasodilators activate ATP-sensitive K+
channels in arterial smooth muscle. Science
20. Hassessian H, Bodin P, Burnstock G. Blockade by glibenclamide of the flow-evoked endothelial release of ATP that contributes to vasodilatation in the pulmonary vascular bed of the rat. Br J Pharmacol