Differential Modulation of the Cardiac L- and T-type Calcium Channel Currents by Isoflurane

The isolation procedure for cardiac myocytes from guinea pigs was approved by the Medical College of Wisconsin Animal Care and Use Committee. Single atrial and ventricular myocytes were isolated from the hearts of adult guinea pigs (SASCO, Wilmington, MA) weighing 180–250 g. The procedure of the cell isolation is a modification of that of Mitra and Morad 13and has been described previously. 14Briefly, the guinea pigs were injected intraperitoneally with 10,000 USP units of heparin and anesthetized with pentobarbital sodium (250 mg/kg intraperitoneally). Hearts were rapidly extracted after thoracotomy and mounted on a Langendorff apparatus for perfusion via  the ascending aorta. The hearts were perfused at a rate of 7 ml/min at 37°C with Joklik medium (Gibco minimum essential medium) containing heparin at pH 7.23 (95% O²–5% CO²oxygenation). After 3–4 min, the Joklik solution was removed, and the hearts were perfused with recirculating enzyme solutions containing a heparin-free Joklik medium with 0.6 mg/ml collagenase (GIBCO type II; Life Technologies, Grand Island, NY), 0.1 mg/ml protease Type XIV (Sigma Chemical Co., St. Louis MO), and 1 mg/ml bovine serum albumin solution (Pentex, Bayer Corp., Kankakee IL), pH 7.23 (95% O²-5% CO², oxygenation).

Atrial cells were obtained during the initial stage of the cell isolation procedure. After 3–4 min of perfusion with enzymes on the Langendorff apparatus, the auricles were cut and minced in 10 ml of Tyrode solution at 37°C. The minced auricles were then transferred into a small flask, placed in a water bath (37°C), and shaken for 3 min. After shaking, the dispersed tissue and cells were filtered through a 210-μm spectral mesh in a test tube and left to settle before experimentation.

From the same heart after the auricles were cut, the ventricles were removed after 14 min of enzyme digestion and minced in the enzyme solution. They were then transferred into small flasks containing enzyme solution and incubated for an additional 3–8 min in a shaker water bath (37°C) to disperse the cells. The dispersed cells were then filtered and centrifuged. The resulting pellet was first washed in the recovery solution (Joklik medium supplemented with 0.1 mm CaCl²and 1 mg/1 ml bovine albumin fraction V; Pentex) and then with Tyrode solution. The isolated ventricular myocytes were stored in Tyrode solution at room temperature for studies.

Modified Tyrode solution contained 132 mm NaCl, 4.8 mm KCl, 1.2 mm MgCl², 10 mm HEPES, 5 mm dextrose, and 1 mm CaCl²and was adjusted to pH 7.32 with NaOH. For recording of calcium currents, Ca²⁺was used as the charge carrier in all experiments. To isolate the calcium current, the following reagents were used for the external solution: 2 mm CaCl², 132 mm NMDG (for sodium replacement), 4.8 mm CsCl to block potassium currents, 2 mm MgCl², 10 mm HEPES, and 5 mm glucose, adjusted to pH 7.4 with HCl. The pipette solution contained 110 mm CsCl to block potassium currents, 10 mm HEPES, 1 mm MgCl², 1 mm CaCl², 11 mm EGTA, and 5 mm potassium-sensitive adenosine triphosphate adjusted to pH 7.3 with CsOH. All chemicals were obtained from Sigma Chemical Co. (St. Louis, MO).

Isoflurane (Ohmeda Caribe Inc., Liberty Corner, NJ) was delivered in the superfusate after a vigorous mixture of an aliquot of the anesthetic with a fixed volume of external Ca²⁺solution. To determine the effect of isoflurane on I^Ca, the inflow perfusate was changed to one in which a given concentration of the anesthetic agent had been equilibrated. At the end of each experiment, solution samples of the chamber were taken and analyzed by gas chromatography (GC-8AIF, Shimadzu Corp., Kyoto, Japan) to verify the anesthetic concentration surrounding the cell.

A drop of cell suspension (atrial or ventricular myocytes) was placed on a Plexiglas chamber mounted on the stage of an inverted microscope (Olympus IMT2, Tokyo, Japan) and superfused with Tyrode solution. Experiments involving atrial cells were conducted within 6 h after isolation, and the selected cells were spindle shaped and showed distinct borders and striations. For ventricular myocytes, the cells were used within 10 h after isolation, and rod shaped cells with clear borders and striations were selected. Pipettes from borosilicate glass were pulled with a multistage puller (Sachs-Flaming, PC-84; Sutter Instrument Company, Novato, CA) and heat polished (Narishige microforge, MF-83; Setagaya-KU, Tokyo, Japan). The polished pipettes had tip resistances of 2.5–4.5 mΩ. High-giga-ohm seals and whole cell voltage clamp were attained in Tyrode solution, followed by superfusion of the chamber with the external Ca²⁺solution.

Whole cell calcium currents using the patch clamp technique were recorded at room temperature. Currents were recorded using a List EPC-7 patch clamp amplifier (Adams and List Assoc., Great Neck, NY), and the output was filtered at 3 kHz to reduce high frequency noise. Data acquisition and analysis were performed using the pCLAMP software package (Version 6.0; Axon Instruments, Inc., Foster City, CA). Additional analysis was performed using ORIGIN version 4.0 software (Microcal Software Inc., Northampton, MA).

In ventricular and atrial myocytes, I^Ca,Lwas elicited during 50-ms depolarizing test pulses (200 ms duration) to +60 mV (10-mV increments) from a holding potential of −50 mV. To elicit I^Ca,Tin atrial cells, depolarizing test pulses were initiated from holding potentials of −90 and −50 mV. I^Ca,Twas obtained by digital subtraction of currents obtained at holding potential of −50 mV (where the L-type is elicited) from currents elicited from a holding potential of −90 mV (where both L- and T-type are elicited) in the same cell.

After obtaining whole cell configuration, calcium current was initially recorded for 15 min to monitor time-dependent changes in peak calcium current amplitude. Cells that showed a significant change in current amplitude because of rundown (i.e. , > 5%) during the stabilization period of 15–20 min, or did not show at least 80% recovery during washout, were not included in analysis of the data. Current–voltage (I-V) relations were obtained during control (anesthetic-free external Ca²⁺solution), in the presence of anesthetic, and during washout (recovery) for each cell. The effects of anesthetic on peak I^Caare reported as percentage block of the current amplitude in the presence of isoflurane compared with control. Concentration–response curves for block of the peak I^Caby isoflurane were obtained by nonlinear regression to a sigmoidal equation:

where B is a block by anesthetic, M is the maximum block by the anesthetic, [A] is concentration of the anesthetic, K^dis the anesthetic concentration for half maximal effect, and n^His the Hill coefficient.

Steady-state activation was obtained by calculating conductance from:

where I^Cais the current amplitude, V is the test potential, and E^revis the reversal potential for calcium, which was approximately 50–60 mV during control conditions. The conductance was normalized to maximum conductance, g^max. To monitor steady-state inactivation of atrial and ventricular I^Ca,L, cells were subjected to 1-s preconditioning test pulses ranging from −50 mV (holding potential) to +40 mV (in 10-mV increments) before stepping to the test pulse of +20 or +30 mV, the potential at which peak I^Ca,Lcurrent occurred. Steady state T-type current inactivation was determined by imposing 1-s preconditioning voltage pulses ranging from −90 (holding potential) to −30 mV before stepping to the test pulse of −30 mV, the potential at which peak I^Ca,Toccurs. At −30 mV test pulse, I^Ca,Lwas not activated. Peak currents obtained at the test potential were normalized to the maximum current (I^max). Both steady- state activation and inactivation were fitted to a Boltzmann equation of the form:

where V is the membrane potential (for activation) or preconditioning potential (for inactivation), V^1/2is the voltage where activation or inactivation is half maximal, and k is the slope factor.

Data are presented as mean ± SEM. Dose-related inhibition of I^Cawas analyzed using one-way analysis of variance. Multiple-comparison analysis of variance with repeated measures was used to analyze the between-group differences. The Student-Newman-Keuls procedure (least significant range) was used a posteriori  to determine differences between mean values. All statistical analyses were conducted using commercially available software from SigmaStat (Jandel Scientific, Corte Madera, CA). A test result was considered significant when P ≤ 0.05.

Figure 1depicts a representation of the L- and T-type calcium channel currents recorded from atrial cells. Whole cell inward I^Caand corresponding I-V relations were recorded from atrial and ventricular cells bathed in Na⁺-free external solution containing 2 mm Ca²⁺. Current traces in figures 1A and Bshow I^Ca,Lelicited by depolarizing test pulses to −30 mV and +10 mV from holding potentials of −50 and −90 mV, respectively. With the cell held at −50 mV, I^Ca,Lwas present during a test step to +10 mV but not at −30 mV. From a holding potential of −90 mV, both I^Ca,Tand I^Ca,Lwere elicited by depolarizing step pulses to −30 and at +10 mV, respectively. Subsequent subtraction of traces recorded from a holding potential of −50 mV from those obtained at −90 mV yielded the T-type calcium current (fig. 1C) in atrial myocytes. During our recording conditions, ventricular T-type calcium channel currents were not studied because they were not observed.

Figure 2depicts representative whole cell ventricular L- and atrial L- and T-type calcium current traces and their corresponding I-V relations recorded in the absence (control) and presence of isoflurane. The average peak current amplitude during control conditions (2 mm external Ca²⁺) for atrial T- and L-type and ventricular L-type calcium currents were 101 ± 4 (n =  51), 211 ± 6 (n =  56), and 511 ± 13 pA (n =  61), respectively. In the examples shown in figure 2, isoflurane at 0.9 mm depressed peak I^Cain both L- (atrial and ventricular) and T-type (atrial) calcium channels. The corresponding I-V values show that the depression of I^Caby isoflurane was reversed after washout of the anesthetic.

To compare the relative sensitivity of atrial L- and T-type calcium channels and ventricular L-type calcium channels to isoflurane, concentration–response curves were determined. Isoflurane block was calculated as a percent reduction in the peak current amplitude measured during depolarizing test pulses in the presence of various concentrations of the anesthetic compared with control. The respective K^dvalues for atrial T- and L-type and ventricular L-type calcium channels were 0.23 ± 0.03, 0.34 ± 0.03, and 0.71 ± 0.06 mm (fig. 3). These significant differences in K^dvalues show that the atrial T-type calcium channel is most sensitive and the ventricular L-type calcium channel is the least sensitive to the blocking action of isoflurane. The Hill coefficients were 1.5 ± 0.3 and 1.7 ± 0.2 for atrial T- and L-type calcium channels, respectively, and 2.8 ± 0.5 for the ventricular L-type calcium channel. The differences were not statistically significant.

Figure 4demonstrates the effects of isoflurane on steady-state inactivation in atrial T- and L-type and ventricular L-type calcium channels. In the example shown, isoflurane produced a shift in the hyperpolarizing direction for all three cases. During control conditions, the average V^1/2and k values for atrial T-type calcium channel current were −57.4 ± 0.6 and 6.9 ± 0.3 mV, respectively. The average values for V^1/2and k for atrial and ventricular L-type calcium channel currents during control conditions were −15.1 ± 0.6 and 5.3 ± 0.3 mV and −14.9 ± 0.1 and 5.2 ± 0.1 mV, respectively. After the administration of isoflurane, there was a dose-dependent hyperpolarizing shift in V^1/2for both atrial and ventricular calcium channels. Table 1summarizes the dose-dependent shifts in V^1/2of steady-state inactivation by isoflurane. At 0.25 mm isoflurane, a small but significant shift in V^1/2occurred in atrial T-type channel but not in the atrial and ventricular L-type channels. At 0.9 mm isoflurane, the shift in steady state inactivation was significantly greater (P < 0.05) in the atrial L-type than the atrial T- and ventricular L-type calcium channels. However, at 1.20 mm isoflurane, the shift in steady state inactivation was significantly (P < 0.05) greater in both atrial and ventricular L-type calcium channels than the atrial T-type calcium channels. The k values did not change significantly in the presence of isoflurane.

The effects of isoflurane on steady-state activation for atrial L- and T-type and ventricular L-type calcium channel currents were less pronounced than those observed for steady-state inactivation (fig. 5). During control conditions, the average V^1/2for steady-state activation for atrial T- and L-type and ventricular L-type calcium channels were −30.6 ± 0.8, 7.4 ± 0.8, and 2.0 ± 0.4 mV, respectively. The k values for activation for the atrial L- and T-type channels were 10.1 ± 0.2 and 9.1 ± 0.4, respectively, and 8.3 ± 0.2 mV for the ventricular L-type calcium channel. Table 2summarizes the effects of isoflurane on steady-state activation in atrial and ventricular L- and T-type calcium channels. For the ventricular L-type calcium channel, only at higher isoflurane concentrations of 1.2 and 1.6 mm was a depolarizing shift in V^1/2significant. Isoflurane did not significantly affect the steady state activation of the atrial L- and T-type calcium currents. The slope factor, k, did not change significantly in all cases after the addition of isoflurane.

In addition to the inhibitory effect on I^Ca, isoflurane also caused a significant hyperpolarizing shift in the apparent reversal potential (E^rev) for the atrial and ventricular L- but not for the T-type calcium channels (fig. 2). E^revwas determined graphically from the I-V relation. During control conditions (2 mm extracellular Ca²⁺), the average E^revfor the atrial and ventricular L-type calcium channels were 59.7 ± 2.3 and 59.1 ± 0.7 mV, respectively. Isoflurane induced dose-dependent negative changes in E^revin atrial and ventricular L-type calcium channel currents (figure 6A). These hyperpolarizing shifts in E^revwere reversed toward control values after washout of the anesthetic. To test whether these hyperpolarizing shifts in E^revwere dependent on extracellular Ca²⁺, experiments were conducted using ventricular myocytes to determine the effects of 0.9 mm isoflurane on E^revin the presence of increased extracellular Ca²⁺concentration (fig. 6B). At higher concentrations of extracellular Ca²⁺, the shifts in E^revinduced by isoflurane were diminished.

This study compared the effects of isoflurane on atrial L- and T-type and ventricular L-type calcium channel currents from the guinea pig heart. At clinically relevant concentrations, isoflurane differentially depressed atrial and ventricular calcium channel currents. Atrial T-type calcium channels showed the greatest sensitivity to isoflurane, followed by atrial L-type and the ventricular L-type calcium channels. This dose-dependent block of I^Caby isoflurane within the clinical range shows saturation at higher concentrations for atrial L- and T-type but not for the ventricular L-type calcium channel. Isoflurane also induced dose-dependent hyperpolarizing shifts in V^1/2of steady-state inactivation in atrial L- and T-type and ventricular L-type calcium channel currents. At higher concentrations, isoflurane induced a significant depolarizing shift in steady-state activation in ventricular L-type but not in atrial L- or T-type calcium channel current. Isoflurane also induced a hyperpolarizing shift in E^revfor the atrial and ventricular L-type calcium channel currents, and this effect was diminished in the presence of elevated extracellular Ca²⁺concentrations.

Studies by other investigators reported similar hyperpolarizing shifts in steady-state inactivation by volatile anesthetics in ventricular I^Ca,L. 3,15The hyperpolarizing shift in V^1/2of the inactivation curve suggests that isoflurane depresses I^Cain guinea pig atrial and ventricular myocytes, in part by reducing the availability of calcium channels necessary for the influx of Ca²⁺ions on depolarization. Furthermore, in ventricular cells, the depolarizing shift in activation curve at 1.2 and 1.6 mm isoflurane indicates that, at higher concentrations (above clinical levels), the depression of peak I^Ca,Lmay also involve a decrease in ion conductance. The change in steady-state activation may be caused by isoflurane-induced alteration in the rate constants from closed to open states. The magnitude of shifts in steady-state inactivation curve was less in the atrial T-type than the L-type calcium channels (table 1), although the depression of peak I^Cawas greater in the T-type channel (fig. 3). These differences in shifts in inactivation curves between the atrial and ventricular L- and T-type calcium channels suggest that the depression of I^Ca,Tin atrial cells may depend less on the effects of isoflurane on calcium channel availability when compared with the L-type calcium channel. Furthermore, the shift in steady-state activation in the ventricular but not in the atrial L-type calcium channels by high concentrations of isoflurane suggest that the L-type calcium channels in the two chambers may be modulated differently by volatile anesthetics. No significant changes in k by isoflurane were observed, indicating that the channels voltage sensor was not affected.

The pharmacologic results of this study may suggest heterogeneity in the L- and T-type calcium channels in the guinea pig heart. Studies by other investigators have compared the effects of volatile anesthetics on calcium channels between and also within cell types. For example, McDowell et al.  10showed that, in a thyroid C-cell line, isoflurane and enflurane blocked T-type calcium channels more potently than either L- or N-type calcium channels. In a recent preliminary study, Yamakage et al.  16reported that, in porcine bronchial smooth muscle cells, the T-type calcium channel was more responsive than the L-type calcium channel to inhibition by volatile anesthetics. However, anesthetic depression of the bronchial L-type calcium channel was similar to its depression of the tracheal L-type calcium channel. In smooth muscle cells obtained from guinea pig portal vein, the T-type calcium channel was less sensitive than the L-type calcium channel to isoflurane. 17Previous studies from this laboratory showed that, in the canine Purkinje fiber, the L- and T-type calcium channels showed similar sensitivities to isoflurane, enflurane, and halothane. 12However, in the current study using guinea pig hearts, the atrial T-type calcium channels were more sensitive to isoflurane than the atrial and ventricular L-type calcium channels. In addition, the atrial L-type calcium channels were more responsive to isoflurane than the ventricular L-type calcium channel. The discrepancies in these studies may be attributed to tissue and functional differences between atrial and Purkinje fibers. In the canine heart, the ventricular L-type calcium channel current has been shown to have properties that are different from the conducting Purkinje fiber L-type calcium channel. For example, the decay of I^Ca,Lis slower in ventricular cells than in Purkinje cells. 18The differences in responses may also be ascribed to species differences.

Heterogeneity in the L- and T-type calcium channel functions have been attributed to the expression of distinct pore forming α¹subunits, the selective association with accessory subunits and other modulatory proteins associated with the channels. 19The underlying molecular structure of the L-type calcium α^1Csubunit present in heart, smooth muscle, and neurons 20–22is different from the recently cloned human heart α^1Hsubunit of the T-type calcium channel. 23–26The α^1Csubunit is also modulated by accessory subunits that may be different from the accessory subunits associated with the α^1Hsubunit of the T-type calcium channel. 19,24,25,27,28The L- and T-type calcium channels are also modulated differentially by neuromediators and intracellular mechanisms involving protein kinases and phosphorylation-dependent pathway. 8,18Takenoshita and Steinbach 29reported that, in rat DRG cells, the greater sensitivity of the low-voltage (T-type) calcium channel to halothane when compared with the high-voltage (L-type) calcium channel might be attributed to the differences in signal transduction pathways that modulate the two channels. 29These differences in α¹isoforms and differences in intracellular modulators could contribute to the differences in responses of the L- and T-type calcium channels to isoflurane.

The differences in sensitivities to isoflurane by the atrial and ventricular L-type calcium channels may also suggest the existence of variants of the cardiac L-type calcium α^1Csubunit. Splice variants of the human fibroblast α^1Csubunit for the L-type calcium channel have been shown to respond differently to the inhibitory action of isradipine. 30In the cardiovascular system, variants of the α^1Csubunit are responsible for greater sensitivity of smooth muscle L-type calcium channel to nitrendipine when compared with cardiac L-type calcium channel. 20These differential responses may be attributed to specific segments of the transmembrane protein and splice variants of the accessory subunits that modulate the α¹subunit. 19,21,30If similar variations in isoforms of the L-type exist in the heart, this may explain why the atrial L- and ventricular L-type calcium channels showed different sensitivities to isoflurane. However, whether isoforms of the L-type calcium channel α^1Csubunit and variations in their accessory subunits exist for atrial and ventricular myocytes of the two chambers remains to be tested.

Another plausible explanation for the differences in the sensitivity of the atrial and ventricular L-type calcium channels could be diversity in the regulation of the L-type calcium channels in the two chambers. Thus, in the presence of isoflurane, the modulation of the L-type calcium channels in the atria and ventricle could involve other modulatory proteins in the local environment around the channel or in subsequent steps in signal transduction pathways. In support of this notion, we have recently shown that isoflurane might depress atrial and ventricular L-type calcium channel currents by different regulatory pathways. Preliminary results show that, at clinically relevant concentrations, isoflurane might inhibit ventricular but not atrial I^Ca,Lvia  a cholinergic-dependent signal transduction cascade. 31,32Thus, atrial and ventricular cells may contain receptors or intracellular mediators that may contribute to differential anesthetic sensitivity.

An interesting observation of this study is the reversible dose-dependent hyperpolarizing shift in the apparent E^revof the atrial and ventricular I^Ca,Linduced by isoflurane. This negative shift in reversal potential by isoflurane was attenuated by increases in extracellular Ca²⁺concentration (fig. 6). This reduction in the shift in E^revby isoflurane in the presence of elevated external Ca²⁺suggests that the shift may partly be attributed to selectivity changes of the calcium channel by the anesthetic, although further experiments are needed to confirm this. The notion that the calcium channel may be permeable to other monovalent ions has been suggested. 33,34Study 35reported a similar hyperpolarizing shift in the calcium E^revin the presence of isoflurane in hippocampal pyramidal neurons. According to Study, the shift in the reversal potential may be attributed to an increase in outward rectification caused, in part, by movement of Cl⁻, 36which is reversed in the presence of Cd²⁺. 35Studies by Mitra and Morad 37have also shown that Cs⁺ion fluxes through the inward rectifier potassium channel may contribute to shift in the reversal potential. In the current study, however, it is unlikely that Cs⁺flux through the inward rectifier potassium channel contributed to the shift in the reversal potential. At the two holding potentials used in this study, −50 and −90 mV, the holding current amplitude was zero in control and in the presence of isoflurane. Furthermore, the concentrations of extracellular Cs⁺, 30–120 mm, was significantly higher in the study by Mitra and Morad. 37Consequently, the contribution of the inward rectifier potassium channel to changes in the observed reversal potential is likely negligible.

In summary, the current study shows that the calcium channels in the atria and ventricles respond differently to isoflurane. The differences in sensitivity and actions of the anesthetic on the channels may represent phenotypic differences in the calcium channels or differences in the regulation of the channels. Nonetheless, the differential sensitivities of the channels to isoflurane, particularly at the lower concentrations, could cause a disparity in the refractory periods in the two chambers, which could cause conduction abnormalities during anesthesia.