Halothane Attenuates Endothelium-dependent Pulmonary Vasorelaxant Response to Lemakalim, an Adenosine Triphosphate (ATP)-sensitive Potassium Channel Agonist

All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee.

Thirty-two healthy dogs weighing 20-30 kg were anesthetized with pentobarbital sodium (30 mg/kg given intravenously) and fentanyl citrate (15 micro gram/kg given intravenously). After tracheal intubation, the lungs were mechanically ventilated (Harvard respirator, South Natick, MA). A catheter was placed in the femoral artery, and the dogs were exsanguinated by controlled hemorrhage and killed with a bolus of saturated KCl injected intravenously. A left lateral thoracotomy was performed, and the heart and lungs were removed en bloc. Right and left intralobar pulmonary arteries (2-4 mm inner diameter) were dissected free and immersed in cold modified Krebs-Ringer bicarbonate solution composed of 118.3 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO⁴, 1.2 mM KH²PO⁴, 2.5 mM CaCl², 25 mM NaHCO³, 0.016 mM Ca-EDTA, and 11.1 mM glucose. The arteries were cut into 0.5-cm rings with care taken not to damage the endothelium. In some arterial rings, the endothelium was intentionally removed by gently rubbing the intimal surface with a cotton swab. Removal of the endothelium was later verified in each denuded ring by the absence of a vasorelaxant response to acetylcholine (10 sup -6 M). The rings were suspended between two stainless steel stirrups in organ chambers filled with 25 ml modified Krebs-Ringer bicarbonate (37 [degree sign] Celsius) and a 95% oxygen and 5% carbon dioxide mixture. One of the stirrups was anchored, and the other was connected to a strain gauge (model FT03 force displacement transducer; Grass Instruments, West Warwick, RI) to measure isometric tension (3000S; Gould, Valley View, OH). Halothane was delivered to one half of the organ chambers through a calibrated vaporizer (Fluotec MKIII; Ohmeda Inc., Austell, GA) to achieve a bath concentration of 0.5, 1.0, or 1.5 minimum alveolar concentration (MAC). Actual concentrations of halothane in the solution measured by gas chromatography (n = 7) were 0.17 +/- 0.01 mM, 0.35 +/- 0.01 mM, and 0.51 +/- 0.03 mM, respectively. These concentrations are equivalent to 0.46 +/- 0.01, 0.98 +/- 0.01, and 1.43 +/- 0.03 MAC in dogs, respectively. Rings from the same relative anatomic locations in the right and left lungs were used as paired rings (e.g., no anesthetic versus halothane).

Pulmonary arterial rings were stretched at 10-min intervals in increments of 0.5 g to achieve optimal resting tension. Optimal resting tension was defined as the minimum amount of stretch required to achieve the largest contractile response to 20 mM KCl and was determined in preliminary experiments to be 5 g for the size of arteries used in these studies. After the rings had been stretched to their optimal resting tension, the contractile response to 60 mM KCl was measured. After washout of KCl from the organ chamber and return of isometric tension to prestimulation values, halothane was delivered for 30 min to one half of the organ chambers. Phenylephrine dose-response relations were obtained for each ring to calculate 50% of the maximal contractile response to phenylephrine (ED⁵⁰). This was achieved by increasing the concentration of phenylephrine in half-log increments (from 10 sup -8 to 3 x 10 sup -5 M) after the response to each preceding concentration had reached a steady state. Initial experiments showed that phenylephrine caused beta-adrenergic relaxation in addition to alpha-adrenergic contraction in these arteries. Thus all rings were pretreated with the beta-adrenergic antagonist, propranolol (5 x 10 sup -6 M; incubated for 30 min), before phenylephrine administration in all protocols. After washout and the return of isometric tension to baseline values, concentration-effect curves to the K sup +^ATPchannel agonist, lemakalim, were obtained under the following conditions.

Protocol 1. In this experimental series, pulmonary vascular rings with or without endothelium were used (n = 10 dogs). Halothane (1 MAC) was delivered to one half of the organ chambers. After a 30-min equilibration period, rings with or without endothelium were contracted to the ED⁵⁰level of tension with phenylephrine, followed by the cumulative administration of lemakalim (10 sup -9 to 10 sup -5 M). To determine the halothane dose-effect relation, we also investigated the effects of halothane at 0.5 MAC (n = 3) and 1.5 MAC (n = 5) on lemakalim-induced pulmonary vasorelaxation in rings with or without endothelium.

Protocol 2. In the second experimental series, only endothelium-intact pulmonary vascular rings were used (n = 9). Rings were pretreated with one of the following drugs: N^w-nitro-L-arginine methyl ester (L-NAME; 3 x 10 sup -5 M), a nitric oxide synthase inhibitor, indomethacin (10 sup -5 M), an inhibitor of cyclooxygenase, a combination of both inhibitors, or no pretreatment. [16]Rings were contracted to the ED⁵⁰level of tension with phenylephrine 30 min after pretreatment with the inhibitors, followed by the cumulative administration of lemakalim (10 sup -9 to 10 sup -5 mM) with or without exposure to halothane (1 MAC). The inhibitors remained in the bath solution for the duration of the exposure to lemakalim.

Protocol 3. In the third experimental series, pulmonary vascular rings with or without endothelium were used (n = 5). Rings were pretreated with the selective K sup +^ATPchannel antagonist, glybenclamide (10 sup -6 M). Rings were contracted to the ED⁵⁰level of tension with phenylephrine 30 min after pretreatment with glybenclamide, followed by the cumulative administration of lemakalim (10 sup -9 to 3 x 10 sup -6 M). Glybenclamide remained in the bath solution for the duration of the exposure to lemakalim.

The following drugs and chemicals were used: lemakalim (BRL38227, SmithKline Beecham, Herts, UK), indomethacin, L-NAME HCl, glybenclamide, phenylephrine HCl, acetylcholine chloride, and propranolol HCl (Sigma Chemical, St. Louis, MO). All concentrations are expressed as the final molar concentration in the organ chamber. Lemakalim was dissolved in 95% ethanol and diluted in distilled water (final organ chamber ethanol concentration, 0.054%). Indomethacin was dissolved in NaHCO³and diluted in distilled water (final organ chamber NaHCO³concentration, 2 x 10 sup -4 M). Glybenclamide was dissolved in methanol and diluted in distilled water (final organ chamber methanol concentration, 0.16% v/v). The vehicles have no effect on relaxation responses at the concentrations used for drug preparation. [16]All other drugs were dissolved in distilled water.

Values are expressed as the mean +/- SEM. Responses to lemakalim are expressed as a percentage of the contraction to phenylephrine. The effects of endothelial denudation, halothane, and the inhibitors on the concentration-response curves to lemakalim were assessed by calculating the inhibitory concentration causing 50% relaxation of the contraction to phenylephrine (IC⁵⁰). This value was interpolated from the linear portion of the concentration-effect curve by regression analysis and is presented as log IC⁵⁰. Student's t test for paired samples was used to compare the IC⁵⁰values. Values were considered to be significantly different when P <0.05.

Compared with endothelium-intact rings, denudation of the endothelium decreased the ED⁵⁰for phenylephrine without altering the maximum response (Table 1). Lemakalim caused concentration-dependent relaxation in endothelium-intact and endothelium-denuded rings (Figure 1). However, as summarized in Figure 1, lemakalim-induced relaxation was attenuated (P < 0.05) in endothelium-denuded rings (log IC⁵⁰= -6.82 +/- 0.06) compared with endothelium-intact rings (log IC⁵⁰= -7.24 +/- 0.09). These results indicate that lemakalim-induced pulmonary vasorelaxation involves a vascular smooth muscle and an endothelium-dependent component.

Halothane at 1 MAC decreased the ED⁵⁰for phenylephrine without altering the maximum response in endothelium-intact rings, but not in endothelium-denuded rings (Table 1). The effect of 1 MAC halothane on lemakalim-induced pulmonary vasorelaxation is summarized in Figure 2. With the endothelium intact (Figure 2(A)), 1 MAC halothane attenuated (P < 0.05) the relaxation induced by lemakalim (log IC⁵⁰= -6.80 +/- 0.08) compared with rings without halothane (log IC⁵⁰= -7.24 +/- 0.09). In contrast, in endothelium-denuded rings (Figure 2(B)), relaxation induced by lemakalim was unchanged by 1 MAC halothane (log IC⁵⁰= -6.67 +/- 0.08) compared with rings without halothane (log IC⁵⁰= -6.82 +/- 0.06). These results indicate that 1 MAC halothane attenuates the endothelium-dependent component of lemakalim-induced pulmonary vasorelaxation. Halothane at 0.5 MAC did not alter lemakalim-induced relaxation in either endothelium-intact (Figure 3(A)) rings (log IC⁵⁰= -7.05 +/- 0.03 with halothane vs. -7.08 +/- 0.07 without halothane), or endothelium-denuded (Figure 3(B)) rings (log IC⁵⁰= -6.78 +/- 0.03 with halothane vs. -6.89 +/- 0.06 without halothane). In endothelium-intact rings (Figure 4(A)), halothane at 1.5 MAC attenuated (P < 0.05) the relaxation induced by lemakalim (log IC⁵⁰= -6.97 +/- 0.11) compared with rings without halothane (log IC⁵⁰= -7.24 +/- 0.11). The magnitude of the attenuation was similar to that observed with 1 MAC halothane. In endothelium-denuded rings (Figure 4(B)), 1.5 MAC halothane also attenuated (P < 0.05) the relaxation induced by lemakalim (log IC⁵⁰= -6.68 +/- 0.06) compared with rings without halothane (log IC⁵⁰= -7.03 +/- 0.10).

Pretreatment with L-NAME decreased the ED⁵⁰for phenylephrine and increased the maximum response, whereas indomethacin had no effect (Table 2). The effects of L-NAME and indomethacin, alone and in combination, on lemakalim-induced relaxation are summarized in Figure 5. N^w-nitro-L-arginine methyl ester alone (Figure 5(A)) did not alter lemakalim-induced relaxation (log IC⁵⁰= -7.52 +/- 0.14) compared with the no-drug condition (log IC⁵⁰= -7.54 +/- 0.12). In contrast, indomethacin alone (Figure 5(B)) attenuated (P <0.05) lemakalim-induced relaxation (log IC⁵⁰= -7.15 +/- 0.09). Combined inhibition with L-NAME and indomethacin (Figure 5(C)) had no additional attenuating effect on lemakalim-induced relaxation (log IC sub 50 = -7.05 +/- 0.09) compared with indomethacin pretreatment alone. These results indicate that in endothelium-intact rings, a component of lemakalim-induced pulmonary vasorelaxation is mediated by a vasodilator metabolite of the cyclooxygenase pathway but not by nitric oxide.

Halothane at 1 MAC decreased the ED⁵⁰for phenylephrine in rings pretreated with indomethacin, but not in rings pretreated with L-NAME (Table 2). The effects of 1 MAC halothane on lemakalim-induced pulmonary vasorelaxation in endothelium-intact rings pretreated with either L-NAME or indomethacin, alone and in combination, are summarized in Figure 6. In L-NAME pretreated rings (Figure 6(A)), 1 MAC halothane still attenuated (P < 0.05) lemakalim-induced relaxation (log IC⁵⁰= -7.02 +/- 0.19) compared with rings without halothane (log IC⁵⁰= -7.52 +/- 0.14). In contrast, in indomethacin pretreated rings (Figure 6(B)), 1 MAC halothane had no significant effect on lemakalim-induced relaxation (log IC⁵⁰= -6.93 +/- 0.09) compared with rings without halothane (log IC⁵⁰= -7.15 +/- 0.09), although the maximum relaxation response to lemakalim was slightly decreased (P <0.05) during halothane at 1 MAC (79 +/- 3%) compared with rings without halothane (87 +/- 2%). In rings pretreated with combined L-NAME and indomethacin (Figure 6(C)), 1 MAC halothane had no effect on either the log IC⁵⁰(halothane: -6.93 +/- 0.12 vs. no anesthetic: -7.05 +/- 0.09) for lemakalim or the maximum response (halothane: 80 +/- 3% vs. no anesthetic: 84 +/- 3%). These results support the concept that halothane inhibits the cyclooxygenase-dependent component of lemakalim-induced pulmonary vasorelaxation.

The selective K sup +^ATPchannel antagonist, glybenclamide, attenuated (P < 0.05) lemakalim-induced pulmonary vasorelaxation in rings with endothelium (Figure 7(A)), and in rings without endothelium (Figure 7(B)). At lemakalim concentrations below 10 sup -6 M, glybenclamide reduced lemakalim-induced pulmonary vasorelaxation to less than 20% of phenylephrine precontraction.

We recently reported that halothane anesthesia attenuates the pulmonary vasodilator response to the K sup +^ATPchannel agonist, lemakalim, in dogs fitted with instruments for long-term observation. [5]Using an in vitro model of isolated pulmonary arterial rings, our goal in this study was to elucidate the cellular mechanism by which halothane attenuates lemakalim-induced pulmonary vasodilation. Our results indicate that halothane attenuates the endothelium-dependent component of the pulmonary vasorelaxant response to lemakalim via an inhibitory effect on vasodilator metabolites of the cyclooxygenase pathway. This conclusion is based on the following results. First, the pulmonary vasorelaxant response to lemakalim was attenuated in endothelium-denuded rings compared with endothelium-intact rings. Second, halothane (1 MAC) reduced the pulmonary vasorelaxant response to lemakalim in endothelium-intact rings, but not in endothelium-denuded rings. Third, pretreatment with indomethacin, but not L-NAME, attenuated the pulmonary vasorelaxant response to lemakalim in endothelium-intact rings. And finally, pretreatment with indomethacin, but not L-NAME, abolished the halothane-induced attenuation of the pulmonary vasorelaxant response to lemakalim in endothelium-intact rings.

Increasing evidence suggests that K sup +^ATPchannels exist not only in vascular smooth muscle cells but also in vascular endothelial cells. Using the patch-clamp technique in rat brain microvascular endothelial cells, intracellular dialysis of ATP resulted in an increase in K sup + conductance that was inhibited by glybenclamide. [11]Adenosine triphosphate-sensitive, glybenclamide-inhibitable K sup + currents have also been observed in freshly dissociated endothelial cells from rabbit aorta and pulmonary artery. [12]These effects of intracellular ATP depletion on K sup + currents are mimicked by the K sup +^ATPagonists, pinacidil and lemakalim. [11,12]An increase in K sup + conductance results in membrane hyperpolarization, which increases Ca²+ influx in endothelial cells. [13]Pinacidil and cromakalim have been shown to increase intracellular calcium concentration in stimulated porcine aortic endothelial cells. [14]This effect could lead to activation of endothelial enzymes and the increased production of endothelium-derived vasorelaxants. [17]Thus K sup +^ATPchannel-induced vasodilation could be mediated by an endothelium-dependent component and by direct activation of vascular smooth muscle.

There have been differential reports in the literature concerning the functional role of the endothelium in the vasodilator response to K sup +^ATPchannel agonists. In rat aorta, vasorelaxation in response to pinacidil has been reported to be endothelium independent. [18,19]In contrast, cromakalim- and pinacidil-induced increases in canine coronary artery diameter were reduced after endothelial denudation. [20]In the present study, we observed that the pulmonary vasorelaxant response to lemakalim was attenuated in endothelium-denuded rings. The role of endothelium-derived mediators of K sup +^ATPchannel-induced vasodilation is also controversial. In rat aorta, cromakalim-induced vasorelaxation was not altered by either cyclooxygenase or lipoxygenase inhibition. [21]In contrast, we observed that the pulmonary vasorelaxant response to lemakalim was attenuated after pretreatment with the cyclooxygenase inhibitor, indomethacin. In fetal lambs, inhibition of nitric oxide synthase attenuated pulmonary vasodilation in response to pinacidil, [10]whereas we observed that pretreatment with the nitric oxide synthase inhibitor, L-NAME, had no effect on lemakalim-induced pulmonary vasorelaxation. Our results are consistent with the concept that pulmonary vasorelaxation in response to lemakalim involves an endothelium-dependent component that does not involve nitric oxide but is mediated by a vasodilator metabolite of the cyclooxygenase pathway (e.g., prostacyclin). Additional studies are needed to elucidate the cellular mechanism responsible for this selective action of lemakalim on prostacyclin formation, release, or vasodilator activity.

Several recent studies have investigated the effects of volatile anesthetics on vascular smooth muscle K sup + channel activity. In canine cerebral arteries, halothane and isoflurane suppressed the Ca sup 2+ -dependent and -independent K sup + channel current. [22,23]In contrast, these inhalational anesthetics appear to induce coronary vasodilation by activating K sup +^ATPchannels, because the K sup + sub ATP antagonist, glybenclamide, attenuated the vasodilator responses to halothane and isoflurane. [24,25]In the present study, we observed that halothane at 1 MAC attenuated the pulmonary vasorelaxant response to lemakalim. However, this attenuating effect of 1 MAC halothane was apparent only when the endothelium was intact. This indicates that 1 MAC halothane had no effect on the vascular smooth component of lemakalim-induced pulmonary vasorelaxation but rather exerted its effect on the endothelium-dependent component of vasorelaxation. In contrast, 1.5 MAC halothane attenuated lemakalim-induced relaxation in endothelium-intact and endothelium-denuded rings. This result suggests that higher concentrations of halothane can have a direct effect on the vascular smooth muscle component of lemakalim-induced relaxation. It is interesting that the magnitude of the halothane-induced attenuation of lemakalim relaxation was similar at 1 MAC and 1.5 MAC in endothelium-intact rings. Furthermore, halothane exerted a similar effect on endothelium-intact and -denuded rings at 1.5 MAC. Together these findings suggest that the effect of halothane on lemakalim-induced vasodilation in vivo may be mediated primarily by an inhibitory effect on the endothelium-dependent component of the response. Loeb et al. [26]recently showed that halothane inhibits bradykinin-stimulated prostacyclin production in cultured bovine aortic endothelial cells. They also reported that halothane did not alter prostacyclin production stimulated by melittin (a nonreceptor-mediated phospholipase A²activator). [26]These results suggest that halothane did not alter endothelial phospholipase A²activity or cyclooxygenase activity directly. Furthermore, these investigators concluded that the potential site for halothane-mediated inhibition of bradykinin-stimulated prostacyclin production may involve inhibition of protein kinase C, because activation of protein kinase C completely reversed the effects of halothane. [26] 

Although indomethacin pretreatment abolished the halothane-induced alteration in the lemakalim IC⁵⁰value, halothane-induced attenuation of the maximum response to lemakalim was still observed (Figure 6(B)). This effect was eliminated with combined pretreatment with indomethacin and L-NAME (Figure 6(C)), whereas L-NAME alone had no effect (Figure 6(A)). Although speculative, one possible explanation for this finding is that high-dose lemakalim may stimulate a subthreshold amount of nitric oxide, which could act synergistically with prostacyclin to cause pulmonary vasorelaxation.

Removing the endothelium as well as pretreatment with L-NAME potentiated phenylephrine-induced pulmonary artery contraction, whereas pretreatment with indomethacin had no effect. These results suggest that endothelium-derived nitric oxide, but not prostacyclin, modulates phenylephrine-induced pulmonary artery contraction. This effect was shown previously in both pulmonary [27]and systemic vascular rings. [28]Halothane also potentiated phenylephrine-induced pulmonary artery contraction in an endothelium-dependent manner, and this effect was abolished by L-NAME but not by indomethacin. We have preliminary results that suggest that phenylephrine stimulates the a²receptor-mediated release of nitric oxide from pulmonary endothelial cells, and that this effect is attenuated by halothane. [29] 

The selective K sup +^ATPchannel antagonist, glybenclamide, markedly attenuated lemakalim-induced pulmonary vasorelaxation at concentrations of lemakalim less than 10 sup -6 M. This inhibitory effect of glybenclamide was apparent in both endothelium-intact and endothelium-denuded rings. It is important to note that the inhibitory effects of endothelial denudation, halothane, and indomethacin on the magnitude of lemakalim-induced vasorelaxation were apparent at lemakalim concentrations of 10 sup -6 M and less. Thus these various interventions appear to have effects that involve inhibition of K sup +^ATPchannel activation by lemakalim.

In summary, halothane (1 MAC) attenuated the pulmonary vasorelaxant response to lemakalim in an endothelium-dependent manner. Indomethacin, but not L-NAME, abolished the endothelium-dependent halothane-induced attenuation of the vasorelaxation response to lemakalim. These results suggest that lemakalim-induced pulmonary vasorelaxation involves both an endothelium-dependent and vascular smooth muscle component. Further, halothane attenuates the endothelium-dependent pulmonary vasorelaxant response to lemakalim via an inhibitory effect on vasodilator metabolites of the cyclooxygenase pathway.

The authors thank Steve Schomisch, Pantelis Konstantinopoulos, Mike Trentanelli, and George Markakis for technical assistance and Ronnie Sanders for secretarial support in preparing the manuscript. The authors also thank Dr. Zelko Bosnjak (Medical College of Wisconsin) for his assistance in measuring the halothane concentration in the organ bath. The authors thank SmithKline Beecham Pharmaceuticals for their generous gift of lemakalim.