Isoflurane and Sevoflurane Modulate Inactivation Kinetics of Calcium sup 2+ Currents in Single Bullfrog Atrial Myocytes

All solutions were made with glass-distilled water and were kept saturated with 100% oxygen. The following solutions were used:

1. Standard Ringer's solution (mmol *symbol* l sup -1): NaCl 110.0, KCl 2.5, CaCl²2.5, MgCl²5.0, HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid) 5.0, glucose 10.0 (pH was adjusted to 7.4 with 1N NaOH).

2. Potassium sup + -free Caesium sup + solution: NaCl 110.0, CsCl 20.0, CaCl²2.5, MgCl²5.0, HEPES 5.0, glucose 10.0 (pH was adjusted to 7.4 with 1N NaOH).

3. Caesium sup + internal solution: Caesium aspartate 120.0, EGTA (ethyleneglycol-bis-[beta-aminoethyl ether]-N,N,N',N'-tetraacetic acid) 0.5, HEPES 5.0 (pH was adjusted to 7.2 with 1N CsOH).

The calcium-free Ringer's solution was identical to the standard Ringer's solution except that CaCl²was omitted.

The current technique for isolation of atrial myocytes from bullfrog hearts is a modification of the method described previously by Hirota et al. [8]After obtaining approval from the Animal Care Committee of Toyama Medical and Pharmaceutical University, adult bullfrogs (Rana Catesbeiana) were killed by decapitation under ether anesthesia and their hearts removed. The atrium was dissected and then transferred into a 20-ml solution of calcium-free Ringer's solution containing 0.15% crude bacterial collagenase (Type I; Sigma, St Louis, MO) and 0.1% bovine pancreatic trypsin (Type III; Sigma). The enzyme solution was passed through 0.22-micro meter micropore filter before use. The atrial myocytes were dispersed by gently triturating the tissue segment with a Pasteur pipette. The isolated atrial myocytes were stored in a calcium-free Ringer's solution containing 0.1% bovine albumin (Sigma).

Small aliquots of the cell suspension were placed in the recording chamber (volume 0.5 ml) mounted on the stage of an inverted microscope. After cells had settled at the bottom of the chamber, a Potassium sup + -free Caesium sup + solution was superfused at a rate of 1 ml *symbol* min sup -1.

A single suction micropipette technique [9]was used for the whole-cell voltage clamp recordings. The voltage clamp experiments controls the membrane potential by holding it at chosen values and keeping it uniform within the cell so that ionic currents can be recorded. [10]Electrodes were made from 1.0-mm ID, 1.5-mm OD borosilicate capillary tubing (World Precision Instruments, Sarasota, FL) and the electrode resistance was 1-5 M omega when filled with the Caesium sup + internal solution. A List-Medical Electronics (Darmstadt, Germany) EPC7 amplifier was used for voltage-clamp recordings and corrections for liquid junction potentials between the external and internal solutions. [11].

Membrane currents were monitored on a storage oscilloscope and were digitized simultaneously (1024 samples per record; 12-bit resolution) using a type DT2801A (Data Translation, Marlborough, MA) A/D converter board, and then stored on a microcomputer (IBM AT). [12]Scientific graphing software (SigmaPlot; Jandel Scientific, San Rafael, CA) was used for analysis, curve fits and plots of the digitized data and computer simulations of I^Ca.

The methods for recording of the Calcium²+ currents and the analysis of the inactivation kinetics in the Calcium²+ currents basically are the same as those described by Hume and Giles [13]and by Isenberg and Klockner. [14]Calcium²+ currents were recorded in response to 200-ms depolarizing pulses from a holding potential of -40 mV with a Potassium sup + free Caesium sup + external solution and a Caesium sup + internal solution, a condition that induced a complete blockage of the Sodium sup + and Potassium sup + currents. The steady-state inactivation curve for I^Cawas obtained from double-pulse protocols. Every 10 s a 200-ms prepulse from -40 mV to different potentials was followed by a 200-ms test pulse from -40 mV to +20 mV. The amplitude of I sub Ca elicited by the test-pulse was plotted as percentage of the I^Caevoked in the absence of a prepulse. Reactivation (recovery from inactivation) was analyzed by depolarizing the cell every 10 s from -40 mV to +20 mV, returning to -40 mV for 11 intervals from 0 ms to 200 ms.

All experiments were performed at room temperature (20-22 degrees C).

Isoflurane and sevoflurane were vaporized with isoflurane (Muraco, Tokyo, Japan) and sevoflurane vaporizers (Ohmeda, West Yorkshire, UK), respectively, in 100% oxygen (1 liter *symbol* min sup -1) and then bubbled into the Potassium sup + -free Caesium sup + solution in a reservoir for at least 5 min before application to cells. Anesthetic concentrations are presented as volume percentage (vol%) in 100% oxygen. The concentration of isoflurane and sevoflurane in the gas phase were verified by an anesthetic gas analyzer (Capnomac; Datex, Finland) and the concentration in solution was analyzed by a gas chromatograph (Shimazu, Kyoto, Japan). The concentrations of isoflurane and sevoflurane in solution were found to be linear (0.64 and 0.55 mmol *symbol* 1 sup -1 per 1% respectively) with the percentage in gas phase, up to 5.0%.

When breaking the patch after a gigaseal was made with an electrode, there was a gradual decline or increase in I^Cathat could be related to "run down" or "run up" of current during the stabilizing period. After I^Cahad reached its steady state, control data were measured. Subsequently, cells were exposed to the anesthetic-equilibrated solution for at least 3 min, which was sufficient to produce stable effects, and the effects of anesthetics were examined. Recovery responses were recorded at least 15 min after washout of the anesthetic-containing solution from the chamber. Data were analyzed only if complete recovery (95-105% of control) occurred on washout of anesthetics.

Forty-six atrial myocytes from 18 bullfrogs were used for the current experiments. Values were expressed as mean+/-SEM, and comparisons were made by paired t test. A probability of chance occurrence of less than 5% (P < 0.05) was considered significant.

(Figure 1) shows the representative recordings of I^Caand the current-voltage curve in the presence and the absence of isoflurane (5.0 vol%) and sevoflurane (5.0 vol%), respectively. The currents elicited at different potentials for each cell under both conditions were normalized to the peak I^Caat the maximum response in the absence of anesthetic (control). Calcium²+ currents were strongly depressed by isoflurane and sevoflurane. However, they had no significant effect on either the activation threshold or the apparent reversal potential for I^Ca. Effects of isoflurane (1.25-5.0 vol%) and sevoflurane (2.5-5.0 vol%) on characteristics of I^Caare summarized in Table 1. Isoflurane and sevoflurane decreased the amplitude of Calcium²+ currents in a dose-dependent manner.

(Figure 2) shows kinetics of the inactivation of I^Cain the presence and absence of isoflurane (2.5 vol%) and sevoflurane (2.5 vol%). Figure 2(A) shows the representative recordings of the effects of isoflurane on I^Ca. In Figure 2(B and C), time courses of inactivation of I^Ca(indicated by dashed squares in Figure 2(A)) were plotted in semilogarithmic charts. The time decays of currents were satisfactorily fitted by a single exponential equation of the form: Equation 1where t is the time after a peak of I^Ca, I^Ca,max is the maximum I^Caamplitude, and tau^fis the time constant for inactivation time course of I^Ca. The calculated tau^fvalues from I^Carecordings in Figure 2(B) were 52.3 ms in the absence of isoflurane. Isoflurane (2.5 vol%) markedly reduced the tau^fof I^Cato 25.2 ms. Sevoflurane (2.5 vol%) decreased the tau^fof I^Cafrom 41.8 ms to 34.2 ms (Figure 2(C)).

Effects of isoflurane (1.25-2.5 vol%) and sevoflurane (2.5-5.0 vol%) on tau^fin I^Caare summarized in Table 1. Every concentration of isoflurane and sevoflurane significantly reduced tau^f.

(Figure 3)(A) shows the effects of isoflurane (2.5 vol%) on the steady-state inactivation of I^Ca(n = 5). After a 200-ms conditioning prepulse between -40 and +20 mV (to inactivate I^Ca), remaining I^Cawere measured by a 200 ms test-pulse at +20 mV. The test pulse to +20 mV elicits I^Cawhose peak depends on the prepulse potential. I/I^maxis (peak I^Caelicited by test-pulse)/(maximum I^Ca). When the prepulse was +20 mV (inactivation was maximal), peak I^Caof the test pulse was minimal. When the prepulse went to -40 mV, inactivation was least pronounced and I^Cawas largest. The observed curves were fitted to a Boltzmann function of the form: Equation 2where V^mis prepulse potential, V¹/2 is the mid-point potential, and k is the slope factor. The value of V¹/2 in I^Cawas -0.3+/-0.9 mV in control state. Isoflurane (2.5 vol%) significantly shifted the V¹/2 toward the negative to -2.5+/-0.1 mV without a change in the slope factor (k). Thus, the data indicate that isoflurane increases the probability of closing the inactivation gate of the Calcium²+ channel.

(Figure 3)(B) shows a double-pulse experiment measuring the time course of reactivation (recovery from inactivation) of I^Ca. The prepulse to +20 mV was used to inactivate I^Ca. The test pulse separated by several intervals (0, 20, 40, 60, 80, 100, 120, 140, 160, 180, and 200 ms) determines what fraction has recovered in that time.

Peak amplitude of Calcium²+ current recovers with an approximately exponential time course of the form: Equation 3where t is the interval between prepulse and test-pulse, and tau^ris the time constant for the reactivation of I^Ca.

The tau^rof I^Ca(57.8+/-2.7 ms [n = 5]) was significantly prolonged by isoflurane (2.5 vol%) to 68.9+/-3.7 ms.

According to the modification of the Hodgkin and Huxley model [7]by Bassingthwaighte and Reuter, [15]I^Cais represented by: Equation 4where G^Cais the membrane conductance to Calcium²+, and d(t) and f(t), the parameters of activation and inactivation of I^Ca, respectively, are defined by: Equation 5, Equation 6where tau^dand tau^fare the time constant of activation and inactivation of I sub Ca, respectively.

(Equation 4) attributed I^Cainactivation to a purely voltage-dependent process, analogous to inactivation of the Sodium sup + current. Although the model is widely used, [16]the idea of intracellular Calcium²+ -dependent inactivation of I^Cahas been suggested. [17]Because the inactivation of I^Cadepends on the intracellular Calcium²+ concentration as well as the membrane potential, [18]the Calcium²+ -dependent inactivation might be involved in the observed I^Cainactivation. In the current study, however, we used EGTA as the Calcium²+ chelator to prevent the Calcium²+ -dependent inactivation. [19]Additionally, the Calcium sup 2+ transient due to the release of Calcium²+ from intracellular Calcium²+ stores could be excluded because of the lack of sarcoplasmic reticulum in bullfrog myocytes. [20]This would be helpful to minimize the interference due to the Calcium²+ -dependent inactivation.

The results showed that isoflurane and sevoflurane reduced not only the amplitude of I^Cabut also tau^f. However, the apparent reversal potential remained unchanged in the presence of isoflurane and sevoflurane. The changes of tau^dby both anesthetics presumably are negligible because isoflurane (1.25-2.5 vol%) and sevoflurane (2.5-5.0 vol%) had no significant effect on the time to peak.

Here, we hypothesized that isoflurane and sevoflurane (up to 2.5 vol%) have no effect on G^Caand that the decrease of tau^fis mainly responsible for the I^Cadepression, and we then simulated I sub Ca using Equation 4. Figure 4shows the computer-simulated I^Caconstructed with tau^fvalues from Figure 2(B). The configurations of I^Cain Figure 2were well simulated in Figure 4. The peak amplitudes of the simulated I^Caafter isoflurane (2.5 vol%) and sevoflurane (2.5 vol%) calculated with tau^fvalues were 75.5 +/-8.6% (n = 5) and 91.7+/-4.3% (n = 8) of control, respectively. The results were not significantly different from the observed reduction by isoflurane (2.5 vol%) and sevoflurane (2.5 vol%) in amplitudes (71.1+/-5.2% and 89.9+/-3.9% of control; Table 1). These are consistent with our hypothesis that the observed decreases in I^Caby isoflurane and sevoflurane are mainly caused by the depression of tau^f. The model that the local anesthetic receptor of the Sodium sup + channel has additional states differing in their binding affinities and rate constants is called the modulated receptor model for local anesthetic action. [21]There is the possibility of an additional inactivated state of the Calcium²+ channel with the application of isoflurane or sevoflurane.

The reduction in tau^f, however, was not found to be linear with the reduction in amplitude at 5.0 vol% of isoflurane and sevoflurane (Figure 5). The observation suggests that a higher concentration of isoflurane and sevoflurane may affect the other components of the membrane currents, because a high concentration of isoflurane (5.0 vol%) significantly prolonged the time to peak (Table 1). It has been reported that a relatively high concentration of volatile anesthetics (4.0 vol% of sevoflurane [6]and 2.0 vol% of halothane [22]) reduces the intracellular concentration of myocardial cyclic adenosine 3',5'-monophosphate. The myocardial cyclic adenosine 3',5'-monophosphate depression [23,24]may explain the reduction of I^Cacaused by the decrease of G^Caafter the high concentration of volatile anesthetics application.

In summary, isoflurane and sevoflurane modified the inactivation process of I^Cain frog atrial myocytes. These results suggest that the volatile anesthetics enhance the probability of closing the inactivation gate of the Calcium²+ channel, and the volatile anesthetic depression of myocardial Calcium²+ current is mainly due to the acceleration of the inactivation in I^Ca.

The authors thank Dr. K. Toriizuka, for performing assays of isoflurane and sevoflurane bath concentration; Dinabot Co. Ltd., Osaka, Japan, for supplying isoflurane and a isoflurane vaporizer; and Maruishi Pharmaceutical Co., Osaka, Japan, for supplying sevoflurane and a sevoflurane vaporizer.