Antagonistic Actions of Halothane and Sevoflurane on Spontaneous Ca²⁺Release in Rat Ventricular Myocytes

These studies conformed to the Guide for the Care and Use of Laboratory Animals ,21and all animal procedures conformed to schedule 1 regulations described in the Animals (Scientific Procedures) Act, 1986, of the United Kingdom government Home Office. Rat ventricular myocytes were prepared as described previously.6 

Freshly dissociated cells were loaded with fura-2 by gentle agitation of a 2-ml aliquot of cell suspension with 6.25 μl fura-2 AM, 1 mm, in dimethyl sulfoxide for 12 min followed by centrifugation and resuspension in the physiologic salt solution (see Solutions, first sentence). Cells were transferred to a tissue chamber (volume, 0.1 ml) attached to the stage of an inverted microscope (Nikon Diaphot; Nikon, Surrey, United Kingdom) and allowed to settle for several minutes onto the glass bottom of the chamber before being superfused at a rate of approximately 3 ml/min with the physiologic salt solution. Solutions were delivered to the experimental chamber by magnetic drive gear metering pumps (Micropump, Concord, CA), and solution level and temperature were maintained by feedback circuits.22All experiments were conducted at 30°C to maximize the retention of cytosolic fura-2. To record Ca²⁺transients, the fura-2–loaded cells were excited alternately with light at 340 and 380 nm using a monochromator based system (Cairn; Faversham, Kent, United Kingdom) and fluorescence detected at 510 nm. The ratio of fluorescence at 510 nm in response to excitation at 340 and 380 nm was used as a measure of the intracellular Ca²⁺concentration. Ca²⁺transients were digitized at 1 kHz and analyzed using Ionoptix software (Milton, MA).

During experimentation, cells were perfused with the following physiologic salt solution: 140 mm NaCl, 5.4 mm KCl, 1.2 mm MgCl², 0.4 mm NaH²PO⁴, 5 mm HEPES, 10 mm glucose, and 1 or 2 mm CaCl², pH 7.4 (NaOH) at 30°C. Halothane, isoflurane, and sevoflurane were delivered from stock solutions made up in dimethyl sulfoxide; the final concentration of dimethyl sulfoxide never exceeded 0.2%, which had no significant effect on Ca²⁺transients (not shown). The maximum final bath concentration of anesthetic used in the experiments, 0.6 mm, approximates to twice the minimum alveolar concentration of halothane, isoflurane, and sevoflurane, and therefore, these concentrations are both clinically relevant and broadly equianesthetic in rat ventricle.23Concentrations of anesthetics in the superfusate were verified with gas–liquid chromatography4and found to be stable over the course of an experiment.

When cells were perfused with solution containing increased (2 mm) Ca²⁺, spontaneous Ca²⁺transients were evoked, which propagated down the cells as Ca²⁺waves. The frequency and amplitude of these transients were recorded in the absence of anesthetic to establish baseline conditions. If frequency or amplitude was not regular over a period of at least 2 min, the cell was discarded. When steady baseline conditions were achieved, the cells were exposed to 0.6 mm of halothane, isoflurane, or sevoflurane (or a mixture of anesthetics with a total anesthetic concentration of 0.6 mm) for a period of 4 min before control perfusate was restored. The frequency and amplitude of spontaneous Ca²⁺transients were recorded during periods of 30 s. For purposes of analysis of Ca²⁺transient amplitude, if a cell did not display any spontaneous transients during a 30-s period (e.g. , see fig. 1A), that cell was not included in the calculation of mean amplitude for that time period because this would artificially reduce amplitude of transients in cells where they did occur. However, to give a more complete picture of total SR Ca²⁺release, data are also presented as the product of frequency and amplitude. In these experiments, each cell acted as its own control and was only exposed to one anesthetic (or one anesthetic mixture) except for the protocols in electrically stimulated cells (see Discussion). In all groups of cells, there were no significant differences in either the frequency or amplitude of spontaneous Ca²⁺release under baseline conditions (analysis of variance).

Data are presented as mean ± SEM, and statistical comparisons were conducted with SigmaStat, (SPSS, Chicago, IL) using one-way repeated-measures analysis of variance followed by Tukey tests for multiple comparisons or Friedman repeated-measures analysis of variance on ranks followed by Tukey tests for multiple comparison if data failed normality or equal variance tests. Independent sample t  tests were used for comparison of the effects of halothane alone versus  halothane–sevoflurane mixtures on the frequency of spontaneous Ca²⁺release. All figures were prepared using SigmaPlot (SPSS).

Figure 1Aillustrates spontaneous Ca²⁺transients before, during, and after a 4-min exposure to 0.6 mm halothane. The frequency (fig. 1B) and amplitude (fig. 1C) of spontaneous Ca²⁺transients assessed over consecutive 30-s periods were stable over the initial control period (0–120 s). Addition of 0.6 mm halothane led to an immediate increase in frequency (from 2.6 ± 0.21 to 7.5 ± 0.5 transients per 30 s; P < 0.05 vs.  control), which then declined to a level not significantly different from control (fig. 1B). At the end of the 4-min exposure, Ca²⁺transient amplitude was significantly reduced to 54 ± 10% of control (P < 0.05). On removal of halothane, the frequency of spontaneous Ca²⁺release declined (P < 0.05) for a period of 60 s before returning to control. The amplitudes of events (when they occurred, see Materials and Methods) remained below control but after 2 min of control perfusate were restored. The product of frequency and amplitude (fig. 1D), which gives a measure of total Ca²⁺release, reflected changes in both variables.

In contrast, sevoflurane initially reduced the frequency of spontaneous Ca²⁺release (P < 0.05) over a period of 60 s (figs. 2A and B), which returned toward control levels during the 4-min exposure, with no significant change in the amplitude of spontaneous Ca²⁺transients (fig. 2C). On removal of sevoflurane, frequency was enhanced transiently (P < 0.05) before returning to control, but again, this was not accompanied by any change in Ca²⁺transient amplitude. Figure 2Dillustrates that total Ca²⁺release reflected the effect of sevoflurane on frequency, not amplitude, of release.

Isoflurane (0.6 mm) had no significant effect on the frequency or amplitude of spontaneous Ca²⁺release when first applied, during the exposure, or on removal (not shown), suggesting that isoflurane had no net influence on the sensitivity of SR Ca²⁺release.7 

Because spontaneous release of Ca²⁺from the SR is strongly dependent on SR Ca²⁺content,24this was assessed by rapid application of 20 mm caffeine at the time point when the next spontaneous Ca²⁺release was expected. Figure 3Aillustrates this protocol; under control conditions, 20 mm caffeine was applied to the cell under study to discharge SR Ca²⁺content. After removal of caffeine, spontaneous Ca²⁺transients were restored. Anesthetic, 0.6 mm, was then applied, and the caffeine protocol was repeated at the beginning and at the end of the 4-min anesthetic exposure. Figures 3B and Cillustrate that sevoflurane significantly increased the magnitude of caffeine-induced release of Ca²⁺from the SR (a measure of SR Ca²⁺content) during the exposure (to 110 ± 12%, n = 10; P < 0.05), as would be expected if SR Ca²⁺release were inhibited, whereas halothane reduced content to 77 ± 24% (n = 9; P < 0.05), consistent with a halothane-induced enhancement of SR Ca²⁺release (i.e. , spontaneous Ca²⁺release evoked at a lower SR Ca²⁺content).

These results illustrate that halothane and sevoflurane have opposing effects on the SR Ca²⁺release process, in which case the effects of halothane might be ameliorated or reversed by coapplication of sevoflurane. The experiments in figure 4illustrate the results of experiments in which mixtures of halothane and sevoflurane were applied to spontaneously active cells; application of a mixture of 0.3 mm halothane and 0.3 mm sevoflurane (fig. 4A) significantly increased frequency of release (P < 0.05) from an average value of 1.8 (over the four 30-s control periods) to 3.1 over the first 30-s period of exposure, before returning to control (fig. 4B). This compares to 0.6 mm halothane alone (fig. 1B), where frequency increased from an average value of 2.6 (over four 30-s periods) to 7.5 (over the first 30-s period). This could result from the lower actual concentration of halothane, but this was investigated in a separate series of experiments (see next paragraph, second sentence). The mixture of 0.1 mm halothane and 0.5 mm sevoflurane (fig. 4C) effectively abolished the initial transient increase in frequency that we observed with 0.1 mm halothane alone and figure 4Dshows that during the first 30-s period, frequency of release was actually reduced. After removal of this anesthetic mixture, there was no effect on the frequency of release.

Figure 5Aillustrates that as the proportion of sevoflurane was increased in the mixture, the frequency of spontaneous events on application declined with the line crossing baseline conditions (dotted line) at approximately 0.1 mm halothane–0.5 mm sevoflurane. The effect of 0.1 and 0.3 mm halothane alone (filled symbols) on the initial increase in frequency was significantly (P < 0.05) greater than that induced by the anesthetic mixtures containing the same level of halothane. Data for frequency of events on removal (fig. 5B) also illustrates that sevoflurane ameliorates the effect of halothane on suppression of SR Ca²⁺release, and the line again intercepts baseline conditions at approximately 0.1 mm halothane–0.5 mm sevoflurane.

Halothane (0.6 mm) significantly (P < 0.05) increased diastolic Ca²⁺throughout the exposure (figs. 1A and 6A), which returned to control levels on removal of halothane. In contrast, 0.6 mm sevoflurane significantly reduced diastolic Ca²⁺for a period of 120 s, after which diastolic Ca²⁺returned to control levels (figs. 2A and 6B). In cells pretreated with 1 μm ryanodine to inhibit SR function, no significant change in diastolic Ca²⁺occurred during exposure to halothane (n = 14) or sevoflurane (n = 12). Figure 6Cshows that 0.1 mm halothane also led to a sustained increase in diastolic Ca²⁺similar to that observed with 0.6 mm halothane. Application of a mixture of 0.1 mm halothane and 0.5 mm sevoflurane also produced a sustained increase in diastolic Ca²⁺; however, there was a 90-s delay before the increase in diastolic Ca²⁺reached significance.

These data describe the effects of halothane, isoflurane, and sevoflurane on the properties of spontaneous Ca²⁺release in isolated ventricular myocytes and the associated changes in diastolic Ca²⁺that accompany the application and removal of anesthetic, which help to further our understanding of the inotropic actions of these agents in ventricular muscle. One intriguing aspect of these data is that the effects of halothane on SR Ca²⁺regulation can be reversed by coapplication of sevoflurane.

During normal excitation–contraction coupling, Ca²⁺entry via  I^Cainduces the release of Ca²⁺from the SR via  the process of calcium-induced calcium release.25,26However, in certain pathologic conditions (e.g. , glycoside toxicity), cellular Ca²⁺overload occurs, which can result in arrhythmogenic activity. This results from spontaneous release of Ca²⁺from the SR, which induces delayed after depolarizations (via  inward Na⁺–Ca²⁺exchange current27as Ca²⁺is extruded from the cell), potentially leading to the generation of spontaneous action potentials. The extent of spontaneous release of Ca²⁺from the SR is thought to be heavily influenced by the level of Ca²⁺within the lumen of the SR,28and it has been proposed that luminal Ca²⁺modulates the sensitivity of the ryanodine receptor (RyR) to cytoplasmic Ca²⁺. In these studies, we have induced mild Ca²⁺overload to assess the impact of anesthetics on the sensitivity of the SR Ca²⁺release process.

Our data show that halothane induced an immediate release of SR Ca²⁺, which was ryanodine sensitive and consistent with sensitization of the Ca²⁺release process.3,4,7In contrast, application of sevoflurane (although less potent than halothane in molar terms) initially inhibited SR Ca²⁺release. This result is similar to the decreased efficacy of SR Ca²⁺release in response to rapid cooling induced by sevoflurane.29Coapplication of halothane and sevoflurane significantly ameliorated the sensitizing effect of halothane on the frequency of spontaneous SR Ca²⁺release, with the mixture of 0.1 mm halothane and 0.5 mm sevoflurane having little impact on the frequency or amplitude of SR Ca²⁺release, and effectively mimicked the results produced by 0.6 mm isoflurane.

On wash-off of halothane, spontaneous SR Ca²⁺release was inhibited, whereas removal of sevoflurane induced a burst of Ca²⁺release. As above, these effects of halothane were ameliorated by sevoflurane (figs. 4 and 5). Although these data do not permit a specific subcellular site to be identified, a plausible explanation of these results is that halothane and sevoflurane bind to the RyR, inducing either an increase7or a decrease in its open probability, respectively. Their binding seems antagonistic because 0.5 mm sevoflurane blocked the effects of 0.1 mm halothane almost completely, but again, this may not reflect antagonism at the same site of action as halothane. These data also suggest that isoflurane seems to have no net effect on the RyR under these conditions.

The transient effects of halothane on the frequency of spontaneous Ca²⁺release are similar to those induced by a low dose of caffeine,24,30which are thought to involve luminal feedback of RyR sensitivity. To explain data with halothane (and also caffeine),24,30the initial increase in RyR opening would decrease SR Ca²⁺content (fig. 3). The reduction in SR luminal Ca²⁺would reduce RyR activation and therefore Ca²⁺efflux from the SR and a new steady state would be reached when Ca²⁺efflux was balanced by SR Ca²⁺uptake. On wash-off of halothane, the decrease in Ca²⁺release would result from a combination of lower SR Ca²⁺content and reduced RyR activity.24 

The effects of sevoflurane on the frequency of spontaneous Ca²⁺release are similar to those of tetracaine,24,31which has been reported to decrease RyR open probability.31–33To explain data with sevoflurane (and tetracaine),23,31inhibition of SR Ca²⁺release would increase SR Ca²⁺content (fig. 3). The consequent increase of luminal Ca²⁺would overcome the sevoflurane-induced decrease in open probability of the RyR, and Ca²⁺efflux would increase. This would reach a new steady state when the enhanced efflux at the higher SR Ca²⁺was balanced by SR Ca²⁺uptake. On removal of sevoflurane, the burst of spontaneous Ca²⁺release would result from the restoration of RyR sensitivity coupled with the increased SR Ca²⁺content.

Although the above-proposed mechanism is capable of explaining the effects of halothane and sevoflurane on the frequency of SR Ca²⁺release and SR Ca²⁺content, it can not explain fully the effects on diastolic Ca²⁺, because a mechanism centered around changes in the sensitivity of the RyR would be expected to induce only transient changes in diastolic Ca²⁺, as were observed with sevoflurane (and caffeine30). Halothane induced a ryanodine-sensitive maintained increase in diastolic Ca²⁺at doses between 0.1 and 0.6 mm, which suggests that additional mechanisms associated with the SR are affected. This could result from inhibition of SR Ca²⁺uptake by halothane,34because other agents that inhibit SR Ca²⁺uptake35,36also increase diastolic Ca²⁺. Figure 6Dillustrates that sevoflurane was able to delay the increase in diastolic Ca²⁺, suggesting that the effects of sevoflurane and halothane on diastolic Ca²⁺were additive rather than antagonistic.

One limitation is that these data might overstate the importance of SR-dependent mechanisms in the inotropic profiles of the anesthetics studied as excitation-contraction coupling is heavily SR dependent in the rat. Second, the experiments were conducted at 30°C rather than 37°C—a strategy used to maximize fura-2 retention by cells to allow accurate and consistent measurement of fluorescence signals during long protocols, and the balance of Ca²⁺fluxes along the various pathways may differ slightly at 30°C compared with 37°C.

To summarize, these data suggest that halothane sensitizes, sevoflurane desensitizes, and isoflurane has no net effect on the SR Ca²⁺release process. Figure 7illustrates representative traces of cell shortening and the cytosolic fura-2 fluorescence ratio in an electrically stimulated cell exposed sequentially to halothane, then sevoflurane, and finally to a mixture of halothane and sevoflurane. Figures 7A and Bshow the characteristic inotropic profiles induced by halothane and sevoflurane as described in the introduction. Figure 7Cillustrates that the initial positive inotropic effect associated with halothane and the positive inotropic effect on removal of sevoflurane are both abolished when halothane and sevoflurane are applied simultaneously, corroborating the results of experiments conducted in spontaneously contracting cells. This illustrates the important contribution of anesthetic-induced changes in the sensitivity of the SR Ca²⁺release process to the inotropic profiles of these agents and further suggests that halothane and sevoflurane affect SR Ca²⁺release in an antagonistic manner such that the effects of halothane are ameliorated by sevoflurane.