Effects of Volatile Anesthetics on Store-operated Ca²⁺Influx in Airway Smooth Muscle

Details of the ASM cell preparation technique have been previously described.10,26Porcine tracheae were obtained from a local abattoir. The smooth muscle layer was excised, minced in Hanks’ balanced salt solution (HBSS) with 10 mm HEPES (pH 7.4; Invitrogen, Carlsbad, CA), and then incubated first for 2 h with 20 U/ml papain and 2000 U/ml DNase (Worthington Biochemical Corp., Lakewood, NJ) and subsequently for 1 h at 37°C with 1 mg/ml type intravenous collagenase (Worthington). The sample was triturated, centrifuged, and resuspended in minimum essential medium with 10% fetal calf serum. Isolated cells were plated on collagen-coated glass coverslips, and incubated for 1 h in 5% CO²at 37°C.

Details of the confocal imaging technique for ASM cells have also been previously published.10,26Briefly, cells attached to coverslips were loaded with 5 μm fluo-3 am (Molecular Probes, Eugene, OR) for 30–45 min at 37°C in HBSS, washed, mounted on an open slide chamber (RC-25F, Warner Instruments, Hamden, CT), and perfused with HBSS at room temperature. An Odyssey XL real-time confocal system (Noran Instruments, Middleton, WI) with an Ar-Kr laser (488 nm line) and a Nikon 40×/1.3 oil-immersion objective lens was used to visualize ASM cells. Using manufacturer-provided software, regions of interest encompassing single cells were outlined for measurements of [Ca²⁺]ⁱ. Empirical calibration of fluo-3 fluorescence concentrations for [Ca²⁺]ⁱbased on exposure to known Ca²⁺concentrations was performed as described previously.26 

As previously described,2calibrated online vaporizers were used to add halothane (Wyeth-Ayerst Laboratories, Philadelphia, PA), isoflurane, and sevoflurane (Abbott Laboratories, Deerfield, IL) to the aerating gas mixture. Aqueous anesthetic concentrations equivalent to 1 and 2 adult porcine minimum alveolar concentration (MAC) at room temperature (25°C) were determined for halothane and isoflurane by gas chromatography and electron capture detector (Hewlett-Packard 5880A; Hewlett-Packard, Sunnyvale, CA) and for sevoflurane by a flame ionization detector. Concentrations were halothane 1 MAC 0.35 ± 0.09 mm, 2 MAC 0.53 ± 0.10 mm, isoflurane 1 MAC 0.4 ± 0.09 mm, 2 MAC 0.6 ± 0.11 mm, and sevoflurane 1 MAC 0.50 ± 0.10 mm, 2 MAC 0.68 ± 0.10 mm.

Baseline Ca²⁺concentrations were measured while ASM cells were perfused with HBSS. Extracellular Ca²⁺([Ca²⁺]^o) was then removed by exposure to zero-Ca²⁺HBSS (5 mm EGTA). In the continued absence of [Ca²⁺]^o, cells were also exposed to 1 μm nifedipine and 10 mm KCl to ensure that L-type Ca²⁺channels were not activated during the protocol. Cells were then rapidly exposed to 1 μm cyclopiazonic acid (CPA) in zero-Ca²⁺HBSS. This technique has been frequently used to induce passive SR depletion by inhibiting SR Ca²⁺adenosine triphosphatase with continued SR Ca²⁺leak, presumably from both inositol 1,4,5-trisphosphate- and ryanodine-sensitive SR stores.13,16Continued SR Ca²⁺leak resulted in increased [Ca²⁺]ⁱconcentrations that typically reached a plateau concentration or started to trend down (reflecting increasing Ca²⁺efflux via  the plasma membrane). At this point, 2.5 mm [Ca²⁺]^owas rapidly reintroduced (in the continued presence of CPA). The observed [Ca²⁺]ⁱresponse was then characterized using various pharmacological manipulations.

In separate sets of ASM cells, we performed additional studies using caffeine for selective depletion of ryanodine-sensitive Ca²⁺stores and the assessment of any interactions between Ca²⁺release via  RyR channels and SOCC. The protocol involved replacing CPA in the above experiments with 5 mm caffeine. As with the CPA protocols, pharmacological manipulations were used to characterized SOCC.

SOCC in ASM cells was first established by performing a control protocol using CPA or caffeine in the presence of nifedipine and KCl (as above). Once SOCC-mediated Ca²⁺influx was observed, the cells were washed for 15–20 min with HBSS to remove CPA (or caffeine) and to replenish SR Ca²⁺stores. [Ca²⁺]^owas then removed and the cells exposed to either CPA or caffeine. Once a [Ca²⁺]ⁱresponse was observed (slow development of plateau with CPA or a transient with caffeine), the cells were exposed for 1 min to 1 or 2 MAC halothane, isoflurane, or sevoflurane in the continued presence of CPA or caffeine. This technique ensured that the anesthetics did not influence the SR Ca²⁺release process itself, but were present in sufficient concentration before activation of SOCC. In the continued presence of CPA (or caffeine), and anesthetic, [Ca²⁺]^owas rapidly reintroduced and the [Ca²⁺]ⁱresponse was recorded. In control experiments, the SOCC protocol was performed twice without anesthetic, with an intervening wash in HBSS.

As above, SOCC in ASM cells was first established by performing a control protocol using CPA or caffeine in the presence of nifedipine and KCl. Cells were then washed in HBSS. [Ca²⁺]^owas removed, nifedipine and KCl were added, and SR Ca²⁺release was induced by CPA or caffeine. In the continued presence of CPA or caffeine, the cells were exposed to 10 nm, 100 nm, or 1 μm acetylcholine (Ach) for 2 min. This ensured activation of muscarinic receptors occurred after SR Ca²⁺release had already occurred. Activation of muscarinic receptors occurred with both protocols; however, in the case of CPA, SR Ca²⁺depletion is passive, whereas for caffeine, complete release through RyR channels occurs, although inositol 1,4,5-trisphosphate-induced release is dependent on ACh concentration. In the continued presence of CPA (or caffeine) and ACh, [Ca²⁺]^owas rapidly reintroduced to activated SOCC. Cells were then washed for a second time, and [Ca²⁺]^oremoved (with addition of nifedipine and KCl). After CPA or caffeine exposure, ACh was added as above and after 1 min, cells were exposed to 1 or 2 MAC halothane, isoflurane, or sevoflurane for an additional 1 min. This ensured that muscarinic activation occurred before anesthetic exposure and that confounding effects of anesthetic interactions with the receptor itself were avoided. [Ca²⁺]^owas then rapidly reintroduced. In control protocols, after a control SOCC activation, the ACh protocol was performed twice, without anesthetic.

For each ASM cell, comparisons before and after exposure to a drug were made using paired Student t  test. Repeated measures analysis of variance was used for multiple comparisons with post hoc  Bonferroni and Sheffé F tests. All experimental protocols were not performed in the same cells. Results were replicated in at least 3–5 cells obtained from each of five animals (paired comparisons within cells, independent testing across cells). Statistical significance was tested at P < 0.05. Values are reported as mean ± SD.

At the beginning of experiments, baseline [Ca²⁺]ⁱconcentrations of ASM cells ranged from 70 to 100 nm (78 ± 10 nm; n = 128). Neither removal of [Ca²⁺]^onor addition of nifedipine and KCl significantly affected [Ca²⁺]ⁱconcentrations. In the absence of [Ca²⁺]^o, 1 μm CPA resulted in increased [Ca²⁺]ⁱconcentrations that reached a plateau (228 to 518 nm; n = 65). Subsequent rapid reintroduction of [Ca²⁺]^oresulted in a further, sustained increase of [Ca²⁺]ⁱ(fig. 1A; 340 to 660 nm; P < 0.05 when compared to the first plateau). In contrast to CPA, exposure to 5 mm caffeine in the absence of [Ca²⁺]^oresulted in a transient [Ca²⁺]ⁱresponse (fig. 1B; n = 63). Reintroduction of [Ca²⁺]^oresulted in a relatively sustained increase in [Ca²⁺]ⁱranging from 110 to 386 nm (compared to a peak caffeine response ranging from 455 to 890 nm, the influx response was ∼25% of the caffeine peak). The influx observed with reintroduction of [Ca²⁺]^oafter caffeine exposure was significantly smaller than that after CPA exposure (P < 0.05).

It has been previously reported that SOCC-mediated Ca²⁺influx to be inhibited by Ni²⁺and La³⁺, at least in tissues other than ASM. After control activation of SOCC with CPA (n = 10) or caffeine (n = 10), and HBSS wash, ASM cells were exposed to nifedipine, KCl and 1 μm NiCl²or LaCl³in HBSS. [Ca²⁺]^owas then removed and, in the continued presence of nifedipine, KCl and Ni²⁺(or La³⁺), cells were reexposed to CPA or caffeine, with subsequent reintroduction of [Ca²⁺]^o. In the presence of either 1 μm Ni²⁺or La³⁺, reintroduction of [Ca²⁺]^odid not significantly increase [Ca²⁺]ⁱ(Table 1; P < 0.05 compared to control) indicating that SOCC-mediated Ca²⁺influx in ASM cells is sensitive to both ions.

To determine whether the observed Ca²⁺influx was mediated by mechanisms other than SOCC (because L-type Ca²⁺channels were already inhibited by nifedipine and clamped membrane potential), inhibitors of store-operated Ca²⁺influx (SKF-96365; 10 μm) and Na⁺/Ca²⁺exchange (KBR-7943; 10 μm) were tested. Although the bidirectional plasma membrane Na⁺/Ca²⁺exchanger may not play a major role in most smooth muscle types, expression of this protein has been found in ASM.27When operating in the “reverse” or influx mode, the exchanger can bring Ca²⁺intracellularly, in exchange for Na⁺. After control SOCC activation with CPA (n = 11) or caffeine (n = 11) and HBSS wash, ASM cells were exposed to nifedipine, KCl, and 10 μm SKF-96365 or KBR7943 in HBSS. The SOCC protocol was then repeated in the continued presence of these agents. In 90% of cells, SKF-96365 inhibited SOCC-mediated Ca²⁺influx with both CPA and caffeine protocols (Table 1; P < 0.05 compared with controls). In contrast, the presence of KBR-7943 had no effect on the [Ca²⁺]ⁱresponse upon reintroduction of [Ca²⁺]^o(Table 1), demonstrating that Na⁺/Ca²⁺exchange did not substantially contribute to the observed Ca²⁺influx.

Compared with controls where no ACh was present, preexposure of ASM cells to either 100 nm or 1 μm ACh significant increased Ca²⁺influx with reintroduction of [Ca²⁺]^ofor both CPA and caffeine protocols (P < 0.05; Table 1). In contrast, in other controls where the ACh protocol was repeated, there was ∼5–7% decrease in the Ca²⁺influx upon repetition (not shown).

In ASM cells where SOCC-mediated Ca²⁺influx was first established, repetition of the protocol (CPA or caffeine) resulted in a 5–8% decrease in the observed Ca²⁺influx (rundown control). In contrast, exposure to halothane, isoflurane, or sevoflurane all resulted in significant decreases in the extent of Ca²⁺influx after reintroduction of [Ca²⁺]^o(P < 0.05 compared to control for each anesthetic at 1 or 2 MAC, both same cell and rundown control; fig. 2). For halothane and isoflurane, the effects at 1 MAC were significantly less than at 2 MAC, whereas for sevoflurane, the effects were comparable between 1 and 2 MAC. Among the three anesthetics, sevoflurane caused the least inhibition. At 2 MAC, halothane and isoflurane produced comparable inhibition (fig. 2). Between the CPA and caffeine protocols, for each volatile anesthetic, the effect on Ca²⁺influx was significantly greater for CPA compared to caffeine (P < 0.05).

In ASM cells in which both SOCC-mediated Ca²⁺influx and an ACh-induced augmentation of influx were first established using the CPA or caffeine protocols, exposure to halothane and isoflurane, but not sevoflurane, resulted in significant decreases in the Ca²⁺influx after reintroduction of [Ca²⁺]^o(P < 0.05 compared to both same cell and ACh rundown control for 1 or 2 MAC; fig. 3). Compared to anesthetic effects on Ca²⁺influx in the absence of ACh, anesthetic inhibition of ACh-induced augmentation of influx was significantly greater (P < 0.05). For halothane and isoflurane, the effects at 1 MAC were significantly less than at 2 MAC, with halothane causing greater inhibition. For both halothane and isoflurane, the effect on ACh augmentation of Ca²⁺influx was significantly greater for the CPA protocol (P < 0.05).

Volatile anesthetic effects on ASM have been previously shown to involve inhibition of Ca²⁺influx via  L-type Ca²⁺channels5–7as well as increased SR Ca²⁺leak via  both inositol 1,4,5-trisphosphate and RyR channels.1,2,8Such a leak may be expected to trigger SOCC-mediated Ca²⁺influx in ASM cells, thus providing a mechanism for replenishment of depleted SR Ca²⁺stores. However, in support of our hypothesis, the results of this study demonstrate that volatile anesthetics also inhibit such store-operated Ca²⁺influx, thus accentuating the state of SR depletion. The results further demonstrate that volatile anesthetics, especially halothane and isoflurane, interfere with ACh-induced enhancement of SOCC-mediated Ca²⁺influx, potentially maintaining bronchodilation even during agonist stimulation.

Regulation of [Ca²⁺]ⁱin ASM is mediated by both extracellular Ca²⁺influx and SR Ca²⁺release via  both inositol 1,4,5-trisphosphate receptor9and RyR10channels. Previous studies have demonstrated that Ca²⁺influx in ASM occurs through both voltage-gated11and receptor-gated channels.12,28–30Agonists such as ACh are known to produce SR Ca²⁺release via  both inositol 1,4,5-trisphosphate and RyR channels.10Indeed, it is thought that the initial [Ca²⁺]ⁱresponse to ACh stimulation is SR Ca²⁺release, whereas maintenance of ASM [Ca²⁺]ⁱand thus muscle tone is thought to involve sustained Ca²⁺influx. However, we have previously shown that the [Ca²⁺]ⁱresponse to ASM is oscillatory, mediated by repetitive SR Ca²⁺release and reuptake, where initiation of oscillations is dependent on Ca²⁺release through inositol 1,4,5-trisphosphate receptor channels, but sustenance of oscillations occurs through Ca²⁺-induced Ca²⁺release mechanisms via  RyR channels.10,26The amplitude of [Ca²⁺]ⁱoscillations represent SR Ca²⁺content, and frequency depends on sensitivity for Ca²⁺release via  RyR channels. These studies also showed that Ca²⁺influx serves to maintain and replenish SR Ca²⁺because oscillations were not sustained in the absence of extracellular Ca²⁺or in the presence of inhibitors of Ca²⁺influx.10,26Our previous study on SOCC31as well as the current report emphasize the importance of this SR- and [Ca²⁺]^o-dependent mechanism in maintaining the repletion state of the SR in ASM.

Volatile anesthetics have been shown to target several [Ca²⁺]ⁱregulatory mechanisms in ASM at both SR and the plasma membrane. In terms of SR Ca²⁺release, we and others have shown that volatile anesthetics deplete SR Ca²⁺by decreasing SR Ca²⁺content (store) via  increased “leak” through both inositol 1,4,5-trisphosphate and RyR channels.2,32–35We have previously demonstrated that clinically relevant concentrations of halothane decrease both amplitude and frequency of ACh-induced [Ca²⁺]ⁱoscillations. We interpreted these findings as a decrease in SR Ca²⁺content and reduced sensitivity for Ca²⁺induced Ca²⁺release by volatile anesthetics.35These data collectively demonstrate that volatile anesthetics deplete SR Ca²⁺stores, thus preventing an increase of [Ca²⁺]ⁱin response to agonist stimulation. Anesthetic-induced depletion of SR Ca²⁺then raises the question of whether Ca²⁺influx via  SOCC (if present) is then activated.

Given the existence of voltage-gated and receptor-operated Ca²⁺influx mechanisms in ASM, there is already considerable evidence in the literature that the decrease in [Ca²⁺]ⁱby volatile anesthetics involves inhibition of Ca²⁺influx3,5through voltage-gated L-type Ca²⁺channels.5,6These studies underline the inhibitory effect of volatile anesthetics on plasma membrane Ca²⁺regulatory proteins. Accordingly, the results of the current study demonstrating inhibition of SOCC in the plasma membrane are consistent. Functionally, such inhibition only emphasizes the accentuation of anesthetic effects on [Ca²⁺]ⁱin ASM because prevention of influx would only maintain the SR in a state of depletion, thus keeping [Ca²⁺]ⁱlow even with agonist stimulation.

In addition to previously established modes of Ca²⁺influx, i.e. , voltage-gated11and receptor-gated channels,12,28–30in a recent study, we31found that Ca²⁺influx in ASM occurs in response to SR Ca²⁺depletion. Such a mechanism for influx in response to the state of SR Ca²⁺stores was initially reported in parotid acinar cells and vascular smooth muscle in which influx was found to be dependent on [Ca²⁺]^obut not L-type or receptor operated channels.13,36Other studies have demonstrated that SR Ca²⁺adenosine triphosphatease inhibitors (such as thapsigargin, similar to CPA) increase Ca²⁺influx or smooth muscle tone.20,37–40In these studies, whereas [Ca²⁺]^ois necessary for maintenance of muscle tone or [Ca²⁺]ⁱ, Ca²⁺-channel blockers do not consistently prevent influx, suggesting that both voltage-gated Ca²⁺channels and SOCC contribute to smooth muscle Ca²⁺regulation and tone. Studies to date suggest that SOCC may play a significant role in tonic muscles such as blood vessels.13,17,18Only a few studies, including our own, have reported the existence of SOCC in ASM cells.31,41 

Although the channels that represent SOCC have not been definitively identified, members of the tryptophan family have been implicated in several studies, and mammalian tryptophan genes have been identified.15,42Some of these isoforms have been shown to be involved in SOCC.15,43We have previously verified the expression of several tryptophan C isoforms in porcine ASM,31although other studies have reported tryptophan C expression in guinea pig ASM.44Therefore, it is possible that SOCC in porcine ASM is mediated via  tryptophan C proteins.

There are scarce data on anesthetic effects on SOCC (or capacitative Ca²⁺influx per some authors). Horibe et al.  45found that propofol inhibits capacitative Ca²⁺entry in pulmonary artery smooth muscle. Tas et al.  46found that in primary human endothelial cells, isoflurane inhibits histamine induced Ca²⁺influx via  SOCC. However, this group also found that enflurane enhances Ca²⁺influx in rat glioma cells.47Pochet et al.  48reported that trichloroethanol, the active metabolite of chloral hydrate, inhibits SOCC-mediated Ca²⁺influx in submandibular acinar cells. The current study is the first to report the effects of anesthetics on SOCC in airway smooth muscle.

Studies from several cell types have now shown that SOCC does not appear to be activated by changes in membrane potential, thus making this mechanism different from L-type Ca²⁺channels.13In accordance, in our previous study, we found that changes in membrane potential (induced by altering extracellular K⁺concentration) did not affect SOCC-mediated Ca²⁺influx in ASM.31Furthermore, Ca²⁺influx after SR Ca²⁺depletion by CPA was found to be insensitive to nifedipine, confirming that L-type Ca²⁺channels do not mediate the observed influx. However, the existence of L-type Ca²⁺channels in ASM has already been established,11,49as has the inhibitory effect of volatile anesthetics.5,6Certainly, volatile anesthetic inhibition of L-type channels would also prevent SR repletion, and blunt the [Ca²⁺]ⁱresponse to ACh. Therefore, in the current study, we used nifedipine to block the confounding effect of this mechanism on the observed influx and volatile anesthetic effects. Addition of KCl to the extracellular medium also helped maintain membrane potential at a value where L-type Ca²⁺channels were unlikely to be activated.

Consistent with our previous study, and with studies in other smooth muscle types, we found that SOCC is blocked by Ni²⁺17,19,21,31,41and La³⁺.17,21,31Although both ions are capable of nonspecifically inhibiting influx via  several mechanisms, inhibition of influx by μM concentrations of either ion strongly indicate that the observed influx is mediated via  SOCC.

Previous studies in tissues other than ASM have found that SOCC-mediated Ca²⁺influx occurs after depletion of inositol 1,4,5-trisphosphate-sensitive SR Ca²⁺stores.13,22In our previous study, we found that SOCC in ASM is triggered after SR Ca²⁺release via  inositol 1,4,5-trisphosphate receptor channels.31It has been previously demonstrated that volatile anesthetics markedly increase inositol 1,4,5-trisphosphate concentrations in ASM cells, even in the absence of agonist stimulation2(but see Yamakage et al.  8). Accordingly, anesthetic exposure should enhance SOCC-mediated Ca²⁺influx. However, we31and others14have found that inositol 1,4,5-trisphosphate by itself does not trigger SOCC-mediated Ca²⁺influx even in the presence of agonist stimulation, and it is SR Ca²⁺release via  inositol 1,4,5-trisphosphate receptor channels that is important. In this regard, previous studies have demonstrated that volatile anesthetics enhance SR Ca²⁺leak via  inositol 1,4,5-trisphosphate receptor channels.2,32–35This should lead to enhanced SOCC-mediated Ca²⁺influx in the presence of anesthetics. However, the current study demonstrates that despite this potential enhancement, volatile anesthetics can inhibit Ca²⁺influx.

In addition to triggering of SOCC after release via  inositol 1,4,5-trisphosphate receptor channels, there are now data in skeletal muscle50,51and in ASM31demonstrating SOCC activation in response to depletion of caffeine-sensitive SR Ca²⁺stores and subsequent introduction of [Ca²⁺]^o. Again, it has been previously demonstrated that volatile anesthetics increase SR Ca²⁺leak via  RyR channels,2,52,53suggesting that in the presence of anesthetics, SOCC-mediated influx may be enhanced. However, as with inositol 1,4,5-trisphosphate receptor channels, the data from the current study (i.e. , the caffeine protocol), demonstrates that volatile anesthetics overwhelm any potential enhancement of influx resulting from depletion of caffeine-sensitive Ca²⁺stores.

In ASM, exposure to ACh appears to enhance SOCC-mediated influx. However, it is not entirely clear whether agonist stimulation modulates SOCC because influx is confounded by the presence of mechanisms for Ca²⁺influx that are neither voltage-gated nor mediated by SOCC, i.e. , G-protein coupled receptor-operated mechanisms.41,54,55Whether such a mechanism is involved in ASM is also not clear.41In our previous study,31we found that selective activation of the non-SOCC component by preventing SR Ca²⁺depletion resulted in significantly less enhancement of Ca²⁺influx by ACh compared with when SR depletion was present, suggesting the existence of both SOCC and noncapacitative Ca²⁺influx in ASM. Regardless, the data at least suggests that ACh does modulate SOCC in ASM.

Muscarinic modulation of SOCC has the potential for at least partial inhibition of anesthetic blockade of SOCC; however, it is possible that anesthetic effects on SOCC more likely help maintain bronchodilation, even during agonist stimulation. Furthermore, volatile anesthetics also inhibit muscarinic receptor activation.56,57Accordingly, even in the presence of ACh, volatile anesthetics likely maintain their inhibition of SOCC.

Increased [Ca²⁺]ⁱis key to force generation in ASM. SOCC-mediated Ca²⁺influx has been shown to produce ASM contraction.25Accordingly, anesthetic inhibition of SOCC has the potential to prevent SR Ca²⁺repletion and agonist-induced bronchoconstriction. Volatile anesthetics are known to affect other mechanisms that regulate both [Ca²⁺]ⁱ(e.g. , SR and voltage-gated Ca²⁺channels) and force (e.g. , Ca²⁺sensitivity of force generation58), albeit with different relative potencies (e.g. , sevoflurane has a much smaller effect on influx, but is more potent than isoflurane in decreasing Ca²⁺sensitivity). Nonetheless, anesthetic inhibition of SOCC may significantly contribute to the overall bronchodilatory effect of these agents by decreasing overall [Ca²⁺]ⁱavailability and thus help to maintain bronchodilation.

The authors thank Larry Hunter, M.S. (Associate in Research, Department of Anesthesiology, Mayo Clinic, Rochester, Minnesota) and Thomas Keller (Technician, Department of Anesthesiology, Mayo Clinic, Rochester, Minnesota) for their superb technical assistance.