Halothane directly inhibits contraction of airway smooth muscle, mainly by decreasing the intracellular concentration of free Ca2+ ([Ca2+]i). The role of intracellular Ca2+ stores, sarcoplasmic reticulum, is still unclear. We investigated the role of sarcoplasmic reticulum in the inhibitory effect of halothane on contraction of airway smooth muscle by measuring [Ca2+]i and intracellular concentration of inositol 1,4,5-triphosphate ([IP3]i), a second messenger for release of Ca2+ from sarcoplasmic reticulum.
[Ca2+]i was monitored by measuring the 500-nm light emission ratio (F340/F380) of a Ca2+ indicator fura-2 with isometric tension of canine tracheal smooth muscle strip. During Ca2+-free conditions, carbachol (10(-5) M) was introduced with pretreatment of halothane (0-3%). During Ca2+-free conditions, 20 mM caffeine, a Ca2+-induced Ca2+ release channel opener, was introduced with or without halothane. We measured [IP3]i during exposure to carbachol and halothane by radioimmunoassay technique.
Pretreatment with halothane significantly diminished carbachol-induced increases in [Ca2+]i by 77% and muscle tension by 83% in a dose-dependent manner. Simultaneous administration of halothane significantly enhanced caffeine-induced transient increases in [Ca2+]i and muscle tension in a dose-dependent manner, by 97% and 69%, respectively. Pretreatment with halothane abolished these responses. Rapid increase in [IP3]i produced by carbachol was significantly inhibited by 32% by halothane in a dose-dependent manner.
Halothane, during Ca2+-free conditions, inhibits transient contraction of airway smooth muscle induced by muscarinic receptor stimulation, mainly by attenuating the increase in [Ca2+]i. Depletion of Ca2+ from sarcoplasmic reticulum via Ca2+-induced Ca2+ release channels also may contribute to the attenuation of the increase in [Ca2+]i by halothane.
HALOTHANE has a potent and direct relaxing effect on airway smooth muscle. [1,2]Because the intracellular concentration of free Ca2+([Ca2+]i) plays a central role in the regulation of airway smooth muscle tone, [3,4]a possible mechanism for relaxation by this anesthetic agent is a decrease in [Ca2+]i. Yamakage and Jones et al., using the Ca2+indicator fura-2, demonstrated that relaxation of contracted canine tracheal smooth muscle by halothane at clinically relevant concentrations was associated with a decrease in [Ca2+]i. [Ca2+]iis regulated by influx of Ca2+through membrane-associated Ca (2+) channels (voltage-dependent and Ca2+depletion-activated Ca2+channels) and by release of Ca2+from intracellular Ca2+stores, especially from sarcoplasmic reticulum (SR)(Figure 1). Entry of extracellular Ca2+through voltage-dependent channels is necessary for maintenance of the contraction of airway smooth muscle. [2,6]Yamakage et al., using patch clamp techniques, demonstrated that halothane had an inhibitory effect on the voltage-dependent channels of porcine tracheal smooth muscle cells at clinically relevant concentrations. The role, however, of intracellular Ca2+stores, called SR, in the inhibitory effect of halothane on airway smooth muscle contraction is still unclear.
Release of Ca2+from SR in airway smooth muscle is regulated by two mechanisms: inositol 1,4,5-triphosphate (IP3)-induced Ca2+release (IICR)and Ca2+-inducedCa2+release (CICR) channels (Figure 1). The current study therefore was designed to clarify the role of SR in the inhibitory effect of halothane on contraction of airway smooth muscle (1) by measuring [Ca2+]isimultaneously with muscle tension during exposure to a muscarinic receptor agonist carbachol or a CICR channel opener caffeine, with or without halothane, during Ca2+-freeconditions; and (2) by measuring intracellular concentration of IP3([IP (3)]i) during exposure to carbachol, with or without halothane.
Preparation of Muscle Strips
This study was approved by the Sapporo Medical University Ethical Committee on Animal Research. Adult mongrel dogs (weight, 9–12 kg) were anesthetized with intravenous thiamylal (20 mg/kg). After a surgical level of anesthesia was attained, the trachea was quickly excised and placed in physiologic salt solution (PSS) at room temperature. The PSS contained (in mM) NaCl 136.9, KCl 5.4, CaCl21.5, MgCl21.0, NaHCO323.9, glucose 5.5, and EDTA 0.01. The solution was aerated continuously with a 95% O2/5% CO2gas mixture (pH 7.4). The smooth muscle was dissected free of epithelium, cartilage, and connective tissue and cut into small strips [almost equal to]1 mm wide and [almost equal to]8 mm long.
Measurement of [Ca2+]iand Muscle Tension
Fura-2 loading was performed according to the previously described method. The muscle strips were pretreated with a 5-[micro sign]M acetoxymethyl ester of fura-2 (fura-2/AM), an indicator of Ca2+, in PSS for [almost equal to]7 h at room temperature (22–24 [degree sign]C). Cremophor EL (0.02% vol/vol), a noncytotoxic detergent, was added to increase the solubility of fura-2/AM. After fura-2 loading, the muscle strip was held horizontally in a temperature-controlled (37 [degree sign]C) 5-ml organ bath. One end of the muscle strip was connected to a strain gauge transducer (120T-20B; Kyowa Co., Tokyo, Japan). The strip was then washed with PSS for 30 min to remove uncleaved fura-2/AM. Experiments were conducted within 60 min after washing.
Experiments used a fluorescence spectrometer (CAF-110; Japan Spectroscopic Co., Tokyo, Japan) specially designed to measure the surface fluorescence of living tissue. Excitation light obtained from a xenon high-pressure lamp (75 W) was passed through a rotating filter wheel (128 Hz) that contained 340 and 380 nm filters. The emitted light from the muscle strip at 500 nm was measured with a photomultiplier. The time constant of the optical measurements was 0.25 s. The ratio of the fluorescence from excitation at 340 nm to that at 380 nm (F340/F380) was calculated from successive illumination periods and used as an indicator of [Ca2+](i) as has been reported. [2,10]
The first contraction evoked by a 72.7 mM high K+solution served as a control (100%). The high K+solution was made by substituting NaCl in the PSS with equimolar KCl. After washing the muscle strip with PSS and determining the resting tension, the organ bath solution was substituted transiently (for [almost equal to]5 s) with Ca2+-freePSS (with 5 mM EGTA) to remove Ca2+from the cell surface and then substituted with Ca2+-freePSS (with 50 [micro sign]M EGTA) to maintain [Ca2+]iat the normal resting level. During this condition, the muscle strip was stimulated with 10-5M carbachol, a potent muscarinic receptor agonist. After this protocol, the muscle strip was reincubated with PSS, including 1.5 mM Ca2+. A second contraction was evoked by 72.7 mM high K+solution to restore Ca2+in SR. The PSS was again substituted with Ca2+-freePSS as described earlier. During this condition, halothane (1.0, 2.0, or 3.0% in the gas phase) was introduced into a bath solution for 3 min, and the muscle strip was stimulated with 10-5M carbachol. The concentration of carbachol (10-5M) used in this study could induce maximum contraction and maximum increase in [IP3]i. The order of these two protocols was randomized.
In another experiment, the first contraction was similarly evoked by a 72.7 mM high K+solution, which served as a control (100%). After washing the muscle strip with PSS, including 1.5 mM Ca2+, the organ bath solution was substituted with Ca2+-freePSS as described earlier. During this condition, the muscle strip was exposed to 20 mM caffeine, a CICR opener. After this protocol, the same strip was similarly reincubated with PSS, including 1.5 mM Ca2+, and the PSS was again substituted with Ca2+-freePSS. Halothane (1.0, 2.0, or 3.0% in the gas phase) and caffeine (20 mM) were introduced simultaneously into the bath solution or halothane was preintroduced into the bath solution for 3 min. The muscle strip was then exposed to 20 mM caffeine during this condition. The order of these three protocols was randomized.
To further investigate the effect of other anesthetic agents on the increase of [Ca2+]iattributable to release of Ca2+from SR, we performed additional experiments using isoflurane (range, 0.0–4.5%) in the absence of external Ca2+.
Measurement of [IP3]i
The muscle strips also were used for measuring [IP3]i. After preincubating three or four muscle strips for 30 min in PSS at 37 [degree sign]C, the muscle strips were first incubated with halothane-containing (0.0, 1.0, 2.0, or 3.0% in the gas phase) PSS for 2 min and then stimulated with 10-5M carbachol. The reactions were terminated after 0, 5, 10, 15, 30, 60, or 120 s of stimulation with carbachol by freezing the tissue samples in liquid nitrogen. 
The technique of Uemura et al. was used to measure the [IP (3)]i. The frozen tissue sample was homogenized with 2 ml of 10%(vol/vol) ice-cold HClO4for 20 min. A 200-[micro sign]l aliquot of the homogenized solution was used to measure concentrations of protein. The remaining aliquots were centrifuged at 2,000g for 15 min to remove insoluble materials. The pH of the supernatant was adjusted precisely to 7.5 with 10 N KOH/HEPES. Insoluble precipitates (primarily KClO4) were removed by centrifugation at 2,000g for 10 min. The resultant supernatant was lyophilized and stored at -20 [degree sign]C. The lyophilized samples were dissolved in 100 [micro sign]l distilled water, and the amount of IP3was measured using the Amersham IP3assay system (code TRK 1,000; Amersham Japan Co., Tokyo, Japan). This assay is based on competition between unlabeled IP3in the samples and a fixed quantity of tritium-labeled IP (3) for a limited number of high-affinity binding sites on a specific IP (3) binding protein. The determinations were made in duplicate, and the results were expressed as pmoles per milligram of protein.
Determination of Concentrations of the Anesthetic Agent in the Bath Solution
The tissue samples in all experiments were quickly (within 10 s) exposed to a bath solution equilibrated with halothane (1.0, 2.0, or 3.0% in the gas phase) or isoflurane (1.5, 3.0, or 4.5% in the gas phase). The bath solution was continuously bubbled with the same concentration of the anesthetic agent. Concentrations of the anesthetic agents in bath solution samples were analyzed with a gas chromatograph (GC-12A; Shimadzu Co., Kyoto, Japan) equipped with a flame ionization detector (FTD-8; Shimadzu) and an integrator (Chromatopac C-R 3A; Shimadzu). The mean concentrations of halothane in the solution (1.0, 2.0, and 3.0% in the gas phase) were 0.33, 0.75, and 1.15 mM, respectively, whereas the mean concentrations of isoflurane in the solution (1.5, 3.0, and 4.5% in the gas phase) were 0.35, 0.79, and 1.21 mM, respectively. The concentration of the anesthetic agents in the bath solution had close linear correlation with the concentration in the gas phase, and the anesthetic potencies in dogs between these agents were comparable. [18,19]
With the exceptions noted later, reagents were obtained from Sigma Chemical Co. (St. Louis, MO) and Dojindo Co. (Kumamoto, Japan). Halothane, caffeine, and the IP3assay system were obtained from ICI Co. (Dighton, MA), Wako Pure Chemical Co. (Osaka, Japan), and Amersham Japan Co. (Tokyo, Japan), respectively.
All data are expressed as mean +/- SD. For the measurement of [Ca (2+)]iand muscle tension, high K+-inducedsustained changes in [Ca2+]i(indicated by F340/F380ratio) and muscle tension were used as references (100%). [2,10]All data were analyzed using paired/unpaired two-tailed t test or one-factor analysis of variance with Fisher's a posteriori test. In all comparisons, a probability value <0.05 was considered significant.
Effects of Halothane and Isoflurane on [Ca2+]iand Muscle Tension
(Figure 2) shows the effect of high K+(72.7 mM) with 1.5 mM Ca2+and the effects of carbachol (10-5M) with or without halothane 2.0%/isoflurane 4.5% during Ca2+-freeconditions on [Ca2+]iand the tension of canine tracheal smooth muscle. The ratio F340/F380, an indicator of [Ca2+]i, was increased rapidly by high K+with a concomitant muscle contraction (Figure 2A). After washout with the Ca2+-freePSS, the resting levels of [Ca2+]iand muscle tension remained unchanged (Figure 2B). During this condition, carbachol (10-5M) significantly increased muscle tension. This increased tension was followed by a slight decrease before the muscle tension reached a steady state. The peak and plateau levels of the muscle contraction were 71.2 +/- 9.7 and 65.7 +/- 8.6% of the contraction compared with the muscle tension induced by 72.7 mM high K+with 1.5 mM Ca2+. In contrast, carbachol during Ca2+-freeconditions induced a transient increase of [Ca2+]i, followed by a substantial reduction. The percent peak of [Ca2+]iwas 68.4 +/- 7.6% compared with the [Ca2+]iinduced by 72.7 mM high K+with 1.5 mM Ca2+. [Ca2+](i) and muscle tension reached their respective peaks in 10–30 s. Pretreatment of halothane (2.0%) during Ca2+-freeconditions induced a slight and transient increase of [Ca2+]i, followed by a substantial reduction without change in the muscle tension (Figure 2C). During 2.0% halothane, carbachol (10-5M) induced slight and transient increases of [Ca2+]iand muscle tension, followed by substantial reductions (Figure 2C). These changes in [Ca2+]iand muscle tension were smaller than those induced by carbachol without halothane. The order of the two protocols shown in Figure 2B and Figure 2C were randomized. There were no significant differences in the peak of [Ca2+]ior muscle tension obtained by consecutive carbachol stimulation (data not shown). To determine whether these effects were induced by other anesthetic agents as well, we performed additional experiments using isoflurane in the absence of external Ca2+(Figure 2D). Isoflurane at concentrations of up to 4.5% in the gas phase had no apparent effect on either muscle contraction (inhibited by [almost equal to]9 +/- 4% at 4.5% isoflurane) or the increase in [Ca2+]i(inhibited by [almost equal to]7 +/- 2% at 4.5% isoflurane) induced by carbachol (n = 8 at each point).
(Figure 3) shows the relation between concentrations of halothane and the percent response of [Ca2+]ior muscle tension. Halothane significantly decreased the peaks of [Ca2+]iby [almost equal to]77% and muscle tension by [almost equal to]83% in a dose-dependent manner.
(Figure 4) shows the effects of caffeine (20 mM) and halothane (2.0%) or isoflurane (4.5%) during Ca2+-freeconditions on [Ca2+](i) and the tension of canine tracheal smooth muscle. Caffeine (20 mM) induced increased in [Ca2+]iand muscle tension during the Ca2+-freecondition. These increases were transient (Figure 4A) and similar to those obtained by carbachol stimulation during a Ca2+-freecondition (Figure 2B). [Ca2+]iand muscle tension induced by caffeine reached their respective peaks in 10–30 s. Simultaneous administration of 2.0% halothane augmented the transient increases in [Ca2+]iand muscle tension (Figure 4B). Figure 5shows the relation between concentrations of halothane and the percent response of [Ca2+]ior muscle tension. Halothane significantly increased the peaks of [Ca2+](i) by [almost equal to]97% and muscle tension by [almost equal to]69% in a dose-dependent manner. As shown in Figure 2C, the pretreatment with halothane (2.0%) during the Ca2+-freecondition induced a slight and transient increase in [Ca2+]iwithout changing the muscle tension (Figure 4C). During this conditions, caffeine (20 mM) exerted almost no effect on either [Ca2+]ior muscle tension during any concentration of halothane up to 3.0%. The order of these three protocols was randomized, and there were no significant differences in the peaks of either [Ca2+](i) or muscle tension obtained by consecutive carbachol stimulation (data not shown).
We performed additional experiments using isoflurane in the absence of external Ca2+as well (Figure 4D). Isoflurane at concentrations up to 4.5% in the gas phase had no apparent effect on either muscle contraction (increased by [almost equal to]6 +/- 2% at 4.5% isoflurane) or the increase in [Ca2+]i(increased by [almost equal to]3 +/- 2% at 4.5% isoflurane) induced by caffeine (n = 8 at each point).
Effect of Halothane on [IP3]i
(Figure 6A) shows the time course and effects of 3.0% halothane on [IP3]iin carbachol-stimulated canine tracheal smooth muscle. The [IP3]iat time 0 was 10.6 +/- 0.8 pmol/mg protein (n = 8) and failed to change with the addition of halothane (10.2 +/- 0.9, 10.6 +/- 0.8, and 9.9 +/- 0.7 pmol/mg protein at 1.0, 2.0, and 3.0% halothane, respectively). Carbachol (10-5M) produced a rapid increase in the [IP (3)]i, which reached maximum (23.8 +/- 2.1 pmol/mg protein) 10 s after the stimulation. The rapid increase in [IP3]iinduced by carbachol was followed by a rapid and substantial decrease to a concentration of [almost equal to]10 pmol/mg protein. Halothane (3.0%) significantly inhibited the increase in [IP3]iinduced by carbachol 5–15 s after stimulation with carbachol with no apparent change in the time course of [IP (3)]i. Figure 6B summarizes the effects of various concentrations of halothane (0.0, 1.0, 2.0, and 3.0%) on the peak [IP3]i10 s after stimulation with carbachol. Halothane significantly inhibited in a dose-dependent manner the increase in [IP3]iby [almost equal to]32% induced by carbachol.
The major findings of this study are that, in canine tracheal smooth muscle in vitro, clinically relevant concentrations of halothane altered the intracellular free Ca2+transient and attenuated the muscle contraction induced by muscarinic receptor stimulation even without external Ca2+. These effects were dose-dependent at the concentrations of halothane studied and are consistent with the previous study that used the luminescent Ca2+indicator aequorin. 
In airway smooth muscle, muscarinic receptor stimulation activities the plasma membrane-bound phospholipase C via G protein (Figure 1). Phospholipase C subsequently catalyzes the hydrolysis of membrane-associated phosphatidylinositol 4,5-bisphosphate to IP3and diacylglycerol. A rapid increase in [IP3]iinduces release of Ca (2+) from the SR via IICR channels. [8,22]Diacylglycerol activates Ca (2+/phospholipid-dependent) protein kinase at its membrane site, resulting in sensitization of the contractile elements to intracellular Ca2+. [2,5,12]Stimulation of muscarinic receptors also increases the slow influx of extracellular Ca2+across the plasma membrane. [3,23]An additive increase in [Ca2+]iactivates the Ca2+-and calmodulin-dependent myosin light chain kinase, resulting in the contraction of the muscle cells. Release of Ca2+from the SR is therefore important to initiate the muscle contraction. [2–4,12,20]
This study shows that the airway smooth muscle-sustained contraction can be obtained during Ca2+-freeconditions although influx of Ca2+is important to maintain the muscle contraction. [2–4,12]The muscle contraction seen during the Ca2+-freecondition should be divided into two parts:(1) initial transient contraction with a concomitant increase in [Ca2+]i; and (2) decreased but sustained contraction with a substantial decrease in [Ca2+]i. Because in airway smooth muscle IP (3) is the primary regulator for release of Ca2+from SR, and because the time course of the increase in [IP3]iinduced by carbachol was very similar to that of the change in [Ca2+]i(Figure 2and Figure 6A), we suggest that IP3is an important determinant of [Ca2+]iduring agonist stimulation during Ca2+-freeconditions, whereas IICR is also regulated by [Ca2+]i. [9,24]Accordingly, our results that halothane significantly attenuated the increase in [IP3]iinduced by muscarinic receptor stimulation (Figure 6) supported the results wherein the initial transient increase in muscle tension was significantly inhibited by halothane with a concomitant reduction of increase in [Ca2+]i(Figure 2and Figure 3). Because the release of Ca2+from SR depends on the cube of [IP (3)]i, and the resting level of [IP3]iis [almost equal to]10 pmol/mg protein (Figure 6), it seems reasonable that the changes in carbachol-induced [IP3]iproduced by halothane do not linearly parallel the halothane-induced changes in carbachol-induced Ca2+/tensionchanges (Figure 3).
Our results are in general agreement with studies in a variety of cell types, in which treatment with halothane has been associated with inhibition of the increase in [Ca2+]imediated by IP3. [27–30]These studies have demonstrated that halothane alters Ca2+homeostasis, an action that underlies the in vivo effect of the anesthetic agent. Smart et al. and Rooney et al., however, showed that halothane induced formation of IP3in SH-SY5Y neuroblastoma cells and turkey erythrocytes, respectively. These discrepancies may result from the differences in cell types and species or in the selective effects of halothane on certain receptors, G proteins, or phospholipase C isozymes. 
The latter part of the sustained contraction obtained by stimulation of muscarinic receptors during a Ca2+-freecondition could be in part attributable to the protein kinase C-induced sensitization of contractile elements to Ca2+. [2,12]Accordingly, the inhibition of the latter portion of the sustained contraction by halothane might partly be explained by the previously reported evidence that activity of protein kinase C is attenuated by halothane and by the finding that the increase in [IP3]iinduced by carbachol was inhibited by halothane in this study (Figure 6).
Another kind of Ca2+release channel, CICR channels, also exist in the SR membrane. The SR has been functionally separated into two components: SR-[Greek small letter alpha] and SR-[Greek small letter beta]. SR-[Greek small letter alpha] involves two types of channels, IICR and CICR channels, whereas SR-[Greek small letter beta] involves only IICR channels (Figure 1). [9,24,34]Because evidence shows that an increase in [Ca2+]iper se induces release of Ca2+from the SR via CICR channels, there is a possibility that release of Ca2+via the CICR channels partly involves the increase in [Ca2+]iinduced by stimulation of muscarinic receptors during the Ca2+-freecondition. We conducted another experiment on the effect of halothane on the CICR channels using the CICR opener caffeine. As shown in Figure 4and Figure 5, simultaneous administration of halothane in a dose-dependent manner significantly enhanced the release of Ca2+by stimulation with caffeine. Conversely, the pretreatment with halothane apparently abolished the effect of caffeine on [Ca2+]i. Because the sole administration of halothane as shown in Figure 2and Figure 4transiently increased [Ca (2+)]i, we conclude that halothane settles the CICR channels into an open state. This results in depletion of Ca2+from the SR-[Greek small letter alpha] and attenuation of the increase in [Ca2+]iinduced by caffeine. The preintroduction of halothane therefore could have a partial role in the inhibition of the initial increase in [Ca2+]iinduced by stimulation of muscarinic receptors. These results are consistent with some other investigations. In unstimulated cardiac, [35–37]skeletal, and vascular smooth [27,39]muscles, halothane causes depletion of Ca2+from the SR either by attenuating uptake of Ca2+from the cytosol or by release of Ca2+from the SR via CICR channels. Recently, Warner et al. showed that halothane decreased [Ca2+]iand muscle force in canine tracheal smooth muscle, only when they used submaximum stimulation and not maximum stimulation. This discrepancy may result from the differences in types and concentrations of agonists and from differences in experimental techniques we used. Further, the role of attenuation of uptake of Ca2+into SR by halothane is unknown in airway smooth muscle.
It is noteworthy that isoflurane had little effect on the muscle contraction and increase in [Ca2+]iinduced either by carbachol or by caffeine during Ca2+-freeconditions (Figure 2and Figure 4). This observation parallels the clinical observation that halothane is more effective than other anesthetic agents at inhibiting airway smooth muscle contraction at clinical concentrations. In addition, isoflurane does not activate release and depletion of Ca2+from the SR via CICR as does halothane.
Halothane, during Ca2+-freeconditions, inhibits carbachol-induced transient contraction of canine tracheal smooth muscle, mainly by attenuating the transient increase in [Ca2+]i. This attenuation of increase in [Ca2+]iinduced by stimulation of muscarinic receptors could be attributable to the inhibition of the increase in [IP3]i, which releases Ca2+from SR via IICR channels. Depletion of Ca2+from SR via CICR channels also may partly contribute to attenuation of the increase in [Ca2+]iinduced by stimulation of muscarinic receptors, especially with preintroduction of halothane.