- Email: [email protected]
NO in the isolated rat basilar artery
Neuroscience Letters 313 (2001) 21–24 www.elsevier.com/locate/neulet
Role of potassium channels in the relaxation induced by the nitric oxide (NO) donor DEA/NO in the isolated rat basilar artery Ralf G. Hempelmann a,*, Jo¨rg Seebeck b, Marie-Luise Kruse c, Albrecht Ziegler b, H. Maximilian Mehdorn a a
Department of Neurosurgery, University of Kiel, 24106, Kiel, Germany Department of Pharmacology, University of Kiel, 24106, Kiel, Germany c Department of Internal Medicine, University of Kiel, 24106, Kiel, Germany b
Received 16 July 2001; received in revised form 13 August 2001; accepted 21 August 2001
Abstract This study investigates whether potassium ion (K 1) channels are involved in the nitric oxide (NO)-induced relaxation in segments of the isolated rat basilar artery, mounted onto a wire myograph. A high extracellular K 1 concentration partly inhibited the relaxant effects of the NO donors DEA/NO and SIN-1 (3-morpholino-sydnonimine). Whereas single applications of the K 1 channel inhibitors tetraethyl-ammonium (10 23 M), glibenclamide (10 26 M), 4-aminopyridine (10 23 M), or BaCl2 (5 £ 10 25 M) did not affect the responses to DEA/NO, a combination of these inhibitors reduced the effects of DEA/ NO. These data suggest, that the relaxant effects of NO donors are partly mediated via activation of K 1 channels. Different K 1 channel types seem to be involved that function in a redundant manner and compensate for each other. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Basilar artery; Potassium channels; Nitric oxide; Rats
Potassium ion (K 1) channels play a substantial role in the regulation of vascular tone including that of cerebral arteries [3,8]. Activation of different K 1 channel types by nitric oxide (NO) donors or cyclic guanosine monophosphate (cGMP) analogs could be demonstrated in cerebral arteries from various species [1,12–14,16]. In the basilar artery of anaesthetised rats, voltage-dependent K 1 channels (KV), seem to be involved in the relaxation induced by the NO donor sodium nitroprusside [16], whereas a role of adenosine triphosphate (ATP)-sensitive (KATP) or calcium-sensitive K 1 channels (KCa) could not be observed under normal control conditions [15]. The present study addresses the role of K 1 channels in the NO-induced relaxation of the isolated rat basilar artery. For this purpose, we systematically investigated the effects of different K 1 channel inhibitors on the relaxation induced by the NO donor DEA/NO. Male Sprague–Dawley rats (300–450 g) were anaesthetised with diethylether and killed. Segments of the basilar
* Corresponding author. Klinik fu¨r Neurochirurgie, Weimarer Strasse 8, D-24106 Kiel, Germany. Tel.: 149-431-5974810; fax: 149-431-5974918. E-mail address: [email protected] (R.G. Hempelmann).
artery (2 ^ 0.5 mm long) were mounted onto a wire myograph [10]. A passive diameter-tension curve was established after the bathing solution had reached 37 ^ 0.58C [9,10]. As in previous experiments [6], the vessels were pre-loaded to a tension occurring at 90% of the diameter that resulted in a transmural pressure of 100 mmHg by means of the Laplace equation, using a computer model [9,10]. After the equilibration period, during which two successive exposures to K 1 (80 mM) were performed, the endothelial integrity was tested by the relaxant response to application of carbachol (10 25 M). Two cumulative control concentration-response curves to the NO donor DEA/NO were constructed after a contraction by 5-hydroxytryptamine (5-HT, 10 26 M) had reached its maximum, followed by a third concentration-response curve in the presence of an inhibitor. The inhibitors were administered 15 min before the next cumulative concentration response curve started. The concentration-response curves were separated by a period of at least 30 min. The experiments were finished by application of papaverine (10 24 M) to achieve a maximum relaxation of the vessel. In some experiments, the mounted vessels were de-endothelialised by rubbing off the intima with a hair. The functional de-endothelialisation
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 22 5- X
22
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
was demonstrated by the abolition of the carbachol-induced relaxation. The relaxant responses are presented in percent of the preceding contractions, regarding the pre-contraction as 100%, and the maximum relaxation by papaverine as 0%. The bath solution contained 20 ml of a Tyrode buffer with the following composition (mM): Na 1 147; K 1 2.7; Ca 21 1.8; Mg 21 1.1; Cl 2 145.5; HCO32 12; H2PO42 0.21; glucose 5.5, EGTA 0.01 and ascorbic acid 5.7. K 1 (80 mM) was obtained by replacing 77.3 mM Na 1 by K 1. The buffer was bubbled with a mixture of 95% O2 and 5% CO2 to maintain a pH of 7.4, the temperature was kept at 37 ^ 0.58C. Charybdotoxine was solved in saline solution, apamin in 0.05 M acetic acid, DEA/NO (DEA-NONOate), ((Z)-1-[2(2-Aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium1,2-diolate]) in TRIS buffer (pH 8.8). Further dilutions were performed with water. Stock solutions of glibenclamide, 8bromo-cGMP, and indomethacin were prepared with ethanol, stock solution of 4,4 0 Diisothiocyanatostilbene2,2 0 Disulfonic Acid (DIDS) with dimethyl sulfoxide. Stock solutions of the other substances were prepared with water. The data were compared using ANOVA, followed by the Bonferroni’s post-test. The differences were regarded statistically significant when Bonferroni’s P-value was , 0:05 for comparisons between data in the presence of an inhibitor and both controls. 5-HT (10 26 M) elicited reproducible and submaximal contractions. The relaxant effects of DEA/NO were not
influenced by indomethacin (10 25 M, n ¼ 6, Fig. 1a), the Na 1-K 1-ATPase inhibitor ouabain (10 25 M, n ¼ 6, Fig. 1b), or the chloride channel inhibitor DIDS (10 24 M, n ¼ 4, Fig. 1c). A high extracellular K 1 concentration (120 mM) partly inhibited the effect of DEA/NO when compared with the control concentration-response curves in the presence of 5HT (Fig. 1d). A comparable inhibition of the effect of DEA/ NO by K 1 (120 mM) was seen in de-endothelialised rat basilar arteries (data not shown). Also the relaxation in response to the NO donor SIN-1 (Fig. 1e) was inhibited by increasing K 1 concentrations, as was the relaxation induced by 8-bromo-cGMP (data not shown), whereas the concentration-response curve to papaverine was not considerably affected by K 1 (120 mM) (Fig. 1f). The relaxant effect of DEA/NO was not reduced by tetraethyl-ammonium (TEA) (10 23 M, Fig. 2a), charybdotoxin (CTX) (10 27 M, Fig. 2b, inhibitors of large conductance KCa channels [BKCa]), glibenclamide (10 26 M, Fig. 2c, inhibitor of KATP channels), 4-aminopyridine (4-AP) (10 23 M, Fig. 2d, inhibitor of KV channels), BaCl2 (5 £ 10 25 M, Fig. 2e, inhibitor of inward-rectifier K 1 channels, KIR), or apamine (10 26 M, Fig. 2f, inhibitor of small-conductance KCa channels). However, a combined application of TEA (10 23 M), glibenclamide (10 26 M), 4-AP (10 23 M), and BaCl2 (5 £ 10 25 M) significantly reduced the relaxant effect evoked by the NO donor (Fig. 3). While glibenclamide (10 26 M), BaCl2 (5 £ 10 25 M), and
Fig. 1. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curves) in the presence of: (a) indomethacin (10 25 M, n ¼ 7); (b) ouabain (10 25 M, n ¼ 6); (c) DIDS (10 24 M, n ¼ 4); (d) K 1 (120 mM, n ¼ 8). (e) Concentration-response-curves to SIN-1 (significantly different from each other at concentrations .10 25 M) in the presence of K 1 (40 mM, circles), (80 mM, triangles), and (120 mM, squares) (n ¼ 6). (f) Concentration-response-curves to papaverine in the presence of K 1 (40 mM, circles) and K 1 (120 mM, squares) (n ¼ 5). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test, *P , 0:05, **P , 0:01.
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
23
Fig. 2. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curves) in the presence of: (a) TEA (10 23 M, n ¼ 6); (b) CTX (10 27 M, n ¼ 6); (c) glibenclamide (10 26 M, n ¼ 6); (d) 4-AP (10 23 M, n ¼ 6); (e) BaCl2 (5 £ 10 25 M, n ¼ 6); and (f) apamin (10 26 M, n ¼ 6). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test.
apamine (10 26 M) did not change the basal tone, TEA (10 23 M), CTX (10 27 M) as well as 4-AP (10 23 M) elicited contractile effects, that were unstable in most of the preparations, suggesting that BKCa and KV channels influence the basal tone, in contrast to KATP, KIR and small-conductance KCa channels. The combined application of TEA, 4-AP, glibenclamide, and BaCl2 markedly increased the basal
Fig. 3. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curve) in the presence of a combination of TEA (10 23 M), glibenclamide (10 26 M), 4-AP (10 23 M) and BaCl2 (5 £ 10 25 M) (n ¼ 7). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test, **P , 0:001.
tone, but did not significantly augment the pre-contraction induced by 5-HT. In all experiments, the pre-contractions induced by 5-HT (10 26 M) did not significantly differ in the absence and presence of the respective K 1 channel blockers. K 1 (120 mM) induced a significantly stronger pre-contraction than 5-HT, but only in intact, not in de-endothelialized preparations. Since the inhibitory effect of potassium was also seen in denuded arteries, the amount of precontraction is not a sufficient explanation for the inhibitory K 1 effect. The background of the present study is our previous observation of an inhibitory effect of a high extracellular K 1 concentration on the relaxant responses of isolated rat basilar arterial segments to substances that elicit endothelium-dependent vasorelaxations in the isolated rat basilar artery such as noradrenaline and carbachol [4,6]. This inhibitory action of K 1 suggests that the responses to NO are influenced by mechanisms that affect the membrane potential such as the electrogenic Na 1-K 1-ATPase, Cl 2 channels, or, most probably, K 1 channels [1,12–14,16]. Therefore, this study systematically investigates the effects of K 1 channel inhibitors on the responses to an NO donor in the isolated rat basilar artery. A number of K 1 channel inhibitors have been revealed to block the different channel types more or less selectively; BKCa channels are inhibited by TEA in millimolar concentrations or by scorpion toxins such as iberiotoxine or CTX. KATP channels are blocked by micromolar concentrations of glibenclamide. Barium ions also block KATP channels. KV channels can be inhibited by millimolar concentrations of 4-
24
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
AP, and KIR channels by low barium concentrations (for reviews see Refs. [8,11]). Applications of those K 1 channel blockers on cerebral arteries in vivo or in vitro as well as patch-clamp studies gave evidence to the existence of the four different potassium channels in rat cerebral arterial smooth muscle cells [2,7,16–18] (for reviews see Refs. [3,8]). A high extracellular K 1 concentration attenuated the responses to DEA/NO, SIN-1, and 8-bromo-cGMP, but did not substantially affect the relaxation generated by papaverine. The findings suggest that the vasorelaxant effects of NO and its ‘second messenger’ cGMP are partly mediated via membrane potential-dependent mechanisms. A substantial involvement of the Na 1-K 1-ATPase and of Cl 2 channels can be excluded by the lack of inhibitory effects of ouabain (10 25 M) and of DIDS (10 24 M) on the response to DEA/NO, an involvement of the endothelium is excluded by the experiments using de-endothelialised arteries. The single application of glibenclamide (10 26 M), TEA (10 23 M), CTX (10 27 M), 4-AP (10 23 M), or BaCl2 (5 £ 10 25 M) did not change the effects of DEA/NO, arguing against a predominant role of KATP, BKCa, KV, and KIR channels. The differences between our findings and those of Sobey et al. [16], who observed an inhibition of the NO-induced relaxation after single application of 4-AP in the rat basilar artery, may be explained by different conditions in in vivo and in vitro experiments. However, our data are in line with those of Sobey et al. [16] regarding the involvement of K 1 channels in the NO-induced relaxation of the rat basilar artery: a regimen of combined application of glibenclamide, TEA, 4-AP, and BaCl2 in the aforementioned concentrations significantly reduced the relaxant response to DEA/NO. These observations are in accordance with our previous findings in isolated human cerebral arteries [5]. Thus, as in human cerebral arteries, a portion of the relaxant effect of NO can be explained by activation of different K 1 channel types in a redundant manner. These K 1 channel types may compensate for each other. In our opinion, the lack of an inhibitory effect after application of a single K 1 channel inhibitor should not lead to the conclusion, that the respective K 1 channel type is not involved. We thank Mrs Gabriele Drexel for her sophisticated technical help. The study was supported by the Deutsche Forschungsgemeinschaft (He 2500/3-1). [1] Armstead, W.M., Role of ATP-sensitive K 1 channels in cGMP-mediated pial artery vasodilation, Am. J. Physiol., 270 (1996) H423–H426.
[2] Faraci, F.M. and Heistad, D.D., Role of ATP-sensitive potassium channels in the basilar artery, Am. J. Physiol., 264 (1993) H8–H13. [3] Faraci, F.M. and Heistad, D.D., Regulation of the cerebral circulation: Role of endothelium and potassium channels, Pharmacol. Rev., 78 (1998) 53–97. [4] Hempelmann, R., Mehdorn, H.M. and Ziegler, A., Catecholamine-induced relaxation of rat cerebral arteries is dependent on intact endothelium, Naunyn-Schmiedeberg’s Arch. Pharmacol., 347 (1994) R63. [5] Hempelmann, R.G., Seebeck, J., Ziegler, A. and Mehdorn, H.M., Effects of potassium channel inhibitors on the relaxation induced by the NO donor DEA/NO in isolated human cerebral arteries, J. Neurosurg., 93 (2000) 1048–1054. [6] Hempelmann, R.G. and Ziegler, A., Endothelium-dependent noradrenaline-induced relaxation of rat isolated cerebral arteries: pharmacological characterization of receptor subtypes involved, Br. J.. Pharmacol., 110 (1993) 1321– 1328. [7] Johnson, T.D., Marrelli, S.P., Steenberg, M.L., Childres, W.F. and Bryan, R.M., Inward rectifier potassium channels in the rat middle cerebral artery, Am. J. Physiol., 274 (1998) R541–R547. [8] Kitazono, T., Faraci, F.M., Taguchi, H. and Heistad, D.D., Role of potassium channels in cerebral blood vessels, Stroke, 26 (1995) 1713–1723. [9] McPherson, G.A., Assessing vascular reactivity of arteries in the small vessel myograph, Clin. Exp. Pharmacol. Physiol., 19 (1992) 815–825. [10] Mulvany, M.J. and Halpern, W., Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats, Circ. Res., 41 (1977) 19–26. [11] Nelson, M.T. and Quayle, J.M., Physiological roles and properties of potassium channels in arterial smooth muscle, Am. J. Physiol., 268 (1995) C799–C822. [12] Onoue, H. and Katusic, Z.S., Role of potassium channels in relaxations of canine middle cerebral arteries induced by nitric oxide donors, Stroke, 28 (1997) 1264–1271. [13] Paterno`, R., Faraci, F.M. and Heistad, D.D., Role of Ca 21dependent K 1 channels in cerebral vasodilatation induced by increases in cyclic GMP and cyclic AMP in the rat, Stroke, 27 (1996) 1603–1608. [14] Robertson, B.E., Shubert, R., Heschelar, J. and Nelson, M.T., cGMP-dependent protein kinase activates Ca 21 activated K 1 channels in cerebral artery smooth muscle cells, Am. J. Physiol., 265 (1993) C299–C303. [15] Sobey, C.G. and Faraci, F.M., Effect of nitric oxide and potassium channel agonists and inhibitors on basilar artery, Am. J. Physiol., 272 (1997) H256–H262. [16] Sobey, C.G. and Faraci, F.M., Inhibitory effects of 4-aminopyridine on responses of the basilar artery to nitric oxide, Br. J. Pharmacol., 126 (1999) 1437–1443. [17] Stockbridge, N., Zhang, H. and Weir, B., Potassium currents of rat basilar artery smooth muscle cells, Pflug. Arch., 421 (1992) 37–42. [18] Wang, Y. and Maters, D.A., Ca 21 -dependent K 1 channels of high conductance in smooth muscle cells isolated from rat cerebral arteries, J. Physiol., 462 (1993) 529–545.
Role of potassium channels in the relaxation induced by the nitric oxide (NO) donor DEA/NO in the isolated rat basilar artery Ralf G. Hempelmann a,*, Jo¨rg Seebeck b, Marie-Luise Kruse c, Albrecht Ziegler b, H. Maximilian Mehdorn a a
Department of Neurosurgery, University of Kiel, 24106, Kiel, Germany Department of Pharmacology, University of Kiel, 24106, Kiel, Germany c Department of Internal Medicine, University of Kiel, 24106, Kiel, Germany b
Received 16 July 2001; received in revised form 13 August 2001; accepted 21 August 2001
Abstract This study investigates whether potassium ion (K 1) channels are involved in the nitric oxide (NO)-induced relaxation in segments of the isolated rat basilar artery, mounted onto a wire myograph. A high extracellular K 1 concentration partly inhibited the relaxant effects of the NO donors DEA/NO and SIN-1 (3-morpholino-sydnonimine). Whereas single applications of the K 1 channel inhibitors tetraethyl-ammonium (10 23 M), glibenclamide (10 26 M), 4-aminopyridine (10 23 M), or BaCl2 (5 £ 10 25 M) did not affect the responses to DEA/NO, a combination of these inhibitors reduced the effects of DEA/ NO. These data suggest, that the relaxant effects of NO donors are partly mediated via activation of K 1 channels. Different K 1 channel types seem to be involved that function in a redundant manner and compensate for each other. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Basilar artery; Potassium channels; Nitric oxide; Rats
Potassium ion (K 1) channels play a substantial role in the regulation of vascular tone including that of cerebral arteries [3,8]. Activation of different K 1 channel types by nitric oxide (NO) donors or cyclic guanosine monophosphate (cGMP) analogs could be demonstrated in cerebral arteries from various species [1,12–14,16]. In the basilar artery of anaesthetised rats, voltage-dependent K 1 channels (KV), seem to be involved in the relaxation induced by the NO donor sodium nitroprusside [16], whereas a role of adenosine triphosphate (ATP)-sensitive (KATP) or calcium-sensitive K 1 channels (KCa) could not be observed under normal control conditions [15]. The present study addresses the role of K 1 channels in the NO-induced relaxation of the isolated rat basilar artery. For this purpose, we systematically investigated the effects of different K 1 channel inhibitors on the relaxation induced by the NO donor DEA/NO. Male Sprague–Dawley rats (300–450 g) were anaesthetised with diethylether and killed. Segments of the basilar
* Corresponding author. Klinik fu¨r Neurochirurgie, Weimarer Strasse 8, D-24106 Kiel, Germany. Tel.: 149-431-5974810; fax: 149-431-5974918. E-mail address: [email protected] (R.G. Hempelmann).
artery (2 ^ 0.5 mm long) were mounted onto a wire myograph [10]. A passive diameter-tension curve was established after the bathing solution had reached 37 ^ 0.58C [9,10]. As in previous experiments [6], the vessels were pre-loaded to a tension occurring at 90% of the diameter that resulted in a transmural pressure of 100 mmHg by means of the Laplace equation, using a computer model [9,10]. After the equilibration period, during which two successive exposures to K 1 (80 mM) were performed, the endothelial integrity was tested by the relaxant response to application of carbachol (10 25 M). Two cumulative control concentration-response curves to the NO donor DEA/NO were constructed after a contraction by 5-hydroxytryptamine (5-HT, 10 26 M) had reached its maximum, followed by a third concentration-response curve in the presence of an inhibitor. The inhibitors were administered 15 min before the next cumulative concentration response curve started. The concentration-response curves were separated by a period of at least 30 min. The experiments were finished by application of papaverine (10 24 M) to achieve a maximum relaxation of the vessel. In some experiments, the mounted vessels were de-endothelialised by rubbing off the intima with a hair. The functional de-endothelialisation
0304-3940/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 22 5- X
22
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
was demonstrated by the abolition of the carbachol-induced relaxation. The relaxant responses are presented in percent of the preceding contractions, regarding the pre-contraction as 100%, and the maximum relaxation by papaverine as 0%. The bath solution contained 20 ml of a Tyrode buffer with the following composition (mM): Na 1 147; K 1 2.7; Ca 21 1.8; Mg 21 1.1; Cl 2 145.5; HCO32 12; H2PO42 0.21; glucose 5.5, EGTA 0.01 and ascorbic acid 5.7. K 1 (80 mM) was obtained by replacing 77.3 mM Na 1 by K 1. The buffer was bubbled with a mixture of 95% O2 and 5% CO2 to maintain a pH of 7.4, the temperature was kept at 37 ^ 0.58C. Charybdotoxine was solved in saline solution, apamin in 0.05 M acetic acid, DEA/NO (DEA-NONOate), ((Z)-1-[2(2-Aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium1,2-diolate]) in TRIS buffer (pH 8.8). Further dilutions were performed with water. Stock solutions of glibenclamide, 8bromo-cGMP, and indomethacin were prepared with ethanol, stock solution of 4,4 0 Diisothiocyanatostilbene2,2 0 Disulfonic Acid (DIDS) with dimethyl sulfoxide. Stock solutions of the other substances were prepared with water. The data were compared using ANOVA, followed by the Bonferroni’s post-test. The differences were regarded statistically significant when Bonferroni’s P-value was , 0:05 for comparisons between data in the presence of an inhibitor and both controls. 5-HT (10 26 M) elicited reproducible and submaximal contractions. The relaxant effects of DEA/NO were not
influenced by indomethacin (10 25 M, n ¼ 6, Fig. 1a), the Na 1-K 1-ATPase inhibitor ouabain (10 25 M, n ¼ 6, Fig. 1b), or the chloride channel inhibitor DIDS (10 24 M, n ¼ 4, Fig. 1c). A high extracellular K 1 concentration (120 mM) partly inhibited the effect of DEA/NO when compared with the control concentration-response curves in the presence of 5HT (Fig. 1d). A comparable inhibition of the effect of DEA/ NO by K 1 (120 mM) was seen in de-endothelialised rat basilar arteries (data not shown). Also the relaxation in response to the NO donor SIN-1 (Fig. 1e) was inhibited by increasing K 1 concentrations, as was the relaxation induced by 8-bromo-cGMP (data not shown), whereas the concentration-response curve to papaverine was not considerably affected by K 1 (120 mM) (Fig. 1f). The relaxant effect of DEA/NO was not reduced by tetraethyl-ammonium (TEA) (10 23 M, Fig. 2a), charybdotoxin (CTX) (10 27 M, Fig. 2b, inhibitors of large conductance KCa channels [BKCa]), glibenclamide (10 26 M, Fig. 2c, inhibitor of KATP channels), 4-aminopyridine (4-AP) (10 23 M, Fig. 2d, inhibitor of KV channels), BaCl2 (5 £ 10 25 M, Fig. 2e, inhibitor of inward-rectifier K 1 channels, KIR), or apamine (10 26 M, Fig. 2f, inhibitor of small-conductance KCa channels). However, a combined application of TEA (10 23 M), glibenclamide (10 26 M), 4-AP (10 23 M), and BaCl2 (5 £ 10 25 M) significantly reduced the relaxant effect evoked by the NO donor (Fig. 3). While glibenclamide (10 26 M), BaCl2 (5 £ 10 25 M), and
Fig. 1. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curves) in the presence of: (a) indomethacin (10 25 M, n ¼ 7); (b) ouabain (10 25 M, n ¼ 6); (c) DIDS (10 24 M, n ¼ 4); (d) K 1 (120 mM, n ¼ 8). (e) Concentration-response-curves to SIN-1 (significantly different from each other at concentrations .10 25 M) in the presence of K 1 (40 mM, circles), (80 mM, triangles), and (120 mM, squares) (n ¼ 6). (f) Concentration-response-curves to papaverine in the presence of K 1 (40 mM, circles) and K 1 (120 mM, squares) (n ¼ 5). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test, *P , 0:05, **P , 0:01.
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
23
Fig. 2. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curves) in the presence of: (a) TEA (10 23 M, n ¼ 6); (b) CTX (10 27 M, n ¼ 6); (c) glibenclamide (10 26 M, n ¼ 6); (d) 4-AP (10 23 M, n ¼ 6); (e) BaCl2 (5 £ 10 25 M, n ¼ 6); and (f) apamin (10 26 M, n ¼ 6). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test.
apamine (10 26 M) did not change the basal tone, TEA (10 23 M), CTX (10 27 M) as well as 4-AP (10 23 M) elicited contractile effects, that were unstable in most of the preparations, suggesting that BKCa and KV channels influence the basal tone, in contrast to KATP, KIR and small-conductance KCa channels. The combined application of TEA, 4-AP, glibenclamide, and BaCl2 markedly increased the basal
Fig. 3. Two control concentration-response curves to DEA/NO after precontraction by 5-HT (10 26 M) (dashed curves), followed by a third curve (solid curve) in the presence of a combination of TEA (10 23 M), glibenclamide (10 26 M), 4-AP (10 23 M) and BaCl2 (5 £ 10 25 M) (n ¼ 7). 100%, maximum precontraction; 0%, maximum relaxation by papaverine. The data represent the means ^ SEM, ANOVA followed by Bonferroni’s post-test, **P , 0:001.
tone, but did not significantly augment the pre-contraction induced by 5-HT. In all experiments, the pre-contractions induced by 5-HT (10 26 M) did not significantly differ in the absence and presence of the respective K 1 channel blockers. K 1 (120 mM) induced a significantly stronger pre-contraction than 5-HT, but only in intact, not in de-endothelialized preparations. Since the inhibitory effect of potassium was also seen in denuded arteries, the amount of precontraction is not a sufficient explanation for the inhibitory K 1 effect. The background of the present study is our previous observation of an inhibitory effect of a high extracellular K 1 concentration on the relaxant responses of isolated rat basilar arterial segments to substances that elicit endothelium-dependent vasorelaxations in the isolated rat basilar artery such as noradrenaline and carbachol [4,6]. This inhibitory action of K 1 suggests that the responses to NO are influenced by mechanisms that affect the membrane potential such as the electrogenic Na 1-K 1-ATPase, Cl 2 channels, or, most probably, K 1 channels [1,12–14,16]. Therefore, this study systematically investigates the effects of K 1 channel inhibitors on the responses to an NO donor in the isolated rat basilar artery. A number of K 1 channel inhibitors have been revealed to block the different channel types more or less selectively; BKCa channels are inhibited by TEA in millimolar concentrations or by scorpion toxins such as iberiotoxine or CTX. KATP channels are blocked by micromolar concentrations of glibenclamide. Barium ions also block KATP channels. KV channels can be inhibited by millimolar concentrations of 4-
24
R.G. Hempelmann et al. / Neuroscience Letters 313 (2001) 21–24
AP, and KIR channels by low barium concentrations (for reviews see Refs. [8,11]). Applications of those K 1 channel blockers on cerebral arteries in vivo or in vitro as well as patch-clamp studies gave evidence to the existence of the four different potassium channels in rat cerebral arterial smooth muscle cells [2,7,16–18] (for reviews see Refs. [3,8]). A high extracellular K 1 concentration attenuated the responses to DEA/NO, SIN-1, and 8-bromo-cGMP, but did not substantially affect the relaxation generated by papaverine. The findings suggest that the vasorelaxant effects of NO and its ‘second messenger’ cGMP are partly mediated via membrane potential-dependent mechanisms. A substantial involvement of the Na 1-K 1-ATPase and of Cl 2 channels can be excluded by the lack of inhibitory effects of ouabain (10 25 M) and of DIDS (10 24 M) on the response to DEA/NO, an involvement of the endothelium is excluded by the experiments using de-endothelialised arteries. The single application of glibenclamide (10 26 M), TEA (10 23 M), CTX (10 27 M), 4-AP (10 23 M), or BaCl2 (5 £ 10 25 M) did not change the effects of DEA/NO, arguing against a predominant role of KATP, BKCa, KV, and KIR channels. The differences between our findings and those of Sobey et al. [16], who observed an inhibition of the NO-induced relaxation after single application of 4-AP in the rat basilar artery, may be explained by different conditions in in vivo and in vitro experiments. However, our data are in line with those of Sobey et al. [16] regarding the involvement of K 1 channels in the NO-induced relaxation of the rat basilar artery: a regimen of combined application of glibenclamide, TEA, 4-AP, and BaCl2 in the aforementioned concentrations significantly reduced the relaxant response to DEA/NO. These observations are in accordance with our previous findings in isolated human cerebral arteries [5]. Thus, as in human cerebral arteries, a portion of the relaxant effect of NO can be explained by activation of different K 1 channel types in a redundant manner. These K 1 channel types may compensate for each other. In our opinion, the lack of an inhibitory effect after application of a single K 1 channel inhibitor should not lead to the conclusion, that the respective K 1 channel type is not involved. We thank Mrs Gabriele Drexel for her sophisticated technical help. The study was supported by the Deutsche Forschungsgemeinschaft (He 2500/3-1). [1] Armstead, W.M., Role of ATP-sensitive K 1 channels in cGMP-mediated pial artery vasodilation, Am. J. Physiol., 270 (1996) H423–H426.
[2] Faraci, F.M. and Heistad, D.D., Role of ATP-sensitive potassium channels in the basilar artery, Am. J. Physiol., 264 (1993) H8–H13. [3] Faraci, F.M. and Heistad, D.D., Regulation of the cerebral circulation: Role of endothelium and potassium channels, Pharmacol. Rev., 78 (1998) 53–97. [4] Hempelmann, R., Mehdorn, H.M. and Ziegler, A., Catecholamine-induced relaxation of rat cerebral arteries is dependent on intact endothelium, Naunyn-Schmiedeberg’s Arch. Pharmacol., 347 (1994) R63. [5] Hempelmann, R.G., Seebeck, J., Ziegler, A. and Mehdorn, H.M., Effects of potassium channel inhibitors on the relaxation induced by the NO donor DEA/NO in isolated human cerebral arteries, J. Neurosurg., 93 (2000) 1048–1054. [6] Hempelmann, R.G. and Ziegler, A., Endothelium-dependent noradrenaline-induced relaxation of rat isolated cerebral arteries: pharmacological characterization of receptor subtypes involved, Br. J.. Pharmacol., 110 (1993) 1321– 1328. [7] Johnson, T.D., Marrelli, S.P., Steenberg, M.L., Childres, W.F. and Bryan, R.M., Inward rectifier potassium channels in the rat middle cerebral artery, Am. J. Physiol., 274 (1998) R541–R547. [8] Kitazono, T., Faraci, F.M., Taguchi, H. and Heistad, D.D., Role of potassium channels in cerebral blood vessels, Stroke, 26 (1995) 1713–1723. [9] McPherson, G.A., Assessing vascular reactivity of arteries in the small vessel myograph, Clin. Exp. Pharmacol. Physiol., 19 (1992) 815–825. [10] Mulvany, M.J. and Halpern, W., Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats, Circ. Res., 41 (1977) 19–26. [11] Nelson, M.T. and Quayle, J.M., Physiological roles and properties of potassium channels in arterial smooth muscle, Am. J. Physiol., 268 (1995) C799–C822. [12] Onoue, H. and Katusic, Z.S., Role of potassium channels in relaxations of canine middle cerebral arteries induced by nitric oxide donors, Stroke, 28 (1997) 1264–1271. [13] Paterno`, R., Faraci, F.M. and Heistad, D.D., Role of Ca 21dependent K 1 channels in cerebral vasodilatation induced by increases in cyclic GMP and cyclic AMP in the rat, Stroke, 27 (1996) 1603–1608. [14] Robertson, B.E., Shubert, R., Heschelar, J. and Nelson, M.T., cGMP-dependent protein kinase activates Ca 21 activated K 1 channels in cerebral artery smooth muscle cells, Am. J. Physiol., 265 (1993) C299–C303. [15] Sobey, C.G. and Faraci, F.M., Effect of nitric oxide and potassium channel agonists and inhibitors on basilar artery, Am. J. Physiol., 272 (1997) H256–H262. [16] Sobey, C.G. and Faraci, F.M., Inhibitory effects of 4-aminopyridine on responses of the basilar artery to nitric oxide, Br. J. Pharmacol., 126 (1999) 1437–1443. [17] Stockbridge, N., Zhang, H. and Weir, B., Potassium currents of rat basilar artery smooth muscle cells, Pflug. Arch., 421 (1992) 37–42. [18] Wang, Y. and Maters, D.A., Ca 21 -dependent K 1 channels of high conductance in smooth muscle cells isolated from rat cerebral arteries, J. Physiol., 462 (1993) 529–545.