Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Potassium channels are critical for regulating excitability of cardiac myocytes, where they maintain the resting membrane potential, modulate action potential duration, and determine pacemaking activity. Potassium channels can be gated by voltage (Coraboeuf and Nargeot, 1993; Deal et al., 1996; Barry and Nerbonne, 1996), neurotransmitters such as acetylcholine, or intracellular ligands such as ATP, Ca²⁺, or Na⁺ (Kurachi, 1995; Isomoto et al., 1997). Voltage-gated channels that activate and then inactivate very rapidly in response to membrane depolarization are called transient outward K⁺ channels. In the heart, these channels activate during the upstroke of the action potential and initiate the first phase of membrane repolarization. Rat ventricular myocytes are commonly used to study transient outward K⁺ current (I_to) because it is the major determinant of repolarization in these cells (Josephson et al., 1984).

Class IA antiarrhythmic drugs block sodium channels and prolong action potential duration by blocking one or more K⁺ channels (Campbell, 1983). Examples of this class of antiarrhythmic agent are disopyramide and quinidine (Zipes and Troup, 1978). Microelectrode studies have shown that disopyramide depresses the maximum rate of repolarization, increases conduction time, and prolongs the terminal phase of cardiac repolarization (Kus and Sasyniuk, 1975). In isolated cardiac myocytes, disopyramide blocks sodium current in a use-dependent manner, probably by binding to the activated state of the channel (Sunami et al., 1991; Koumi et al., 1992; Zilberter et al., 1994). Disopyramide also blocks cardiac potassium currents, including the inward rectifier (Coraboeuf et al., 1988; Martin et al., 1994), the ATP-sensitive K⁺ current (De Lorenzi et al., 1995), the muscarinic receptor-operated K⁺ current (Watanabe et al., 1997), and I_to (Coraboeuf et al., 1988; Virag et al., 1997).

In this study, we determined the mechanism of block of I_to channels by disopyramide in isolated rat ventricular myocytes. We conclude that disopyramide blocks the open state, and unblocks from the inactivated state of I_to channels.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Preparation. Single ventricular myocytes were obtained from the right ventricular free wall of adult rats as described previously (Sanchez-Chapula, 1992). The hearts were mounted on a Langendorff apparatus and perfused for 5 min with normal Tyrode's solution, then switched to a nominally calcium-free solution for an additional 5 min. Afterwards, the hearts were perfused for 20 min with a zero-calcium solution containing 1 mg/ml type I collagenase (Sigma, St. Louis, MO). The enzymes were washed out by perfusion with a high-potassium, low-chloride saline for 5 min. The free wall of the right ventricle was dissected away from the rest of the heart and cut into small pieces. Single cells were obtained by mechanical agitation with a pipette. The cells were maintained in a high-potassium, low-chloride solution at 4°C for up to 10 h before use in electrophysiological experiments.

Solutions. Tyrode's solution had the following composition (mM): 125 NaCl, 24 NaHCO₃, 0.42 NaH₂PO₄, 5.4 KCl, 1.8 CaCl₂, 1.0 MgCl₂, 11 glucose, and 10 taurine. The solution was equilibrated with 95% O₂/5% CO₂, at pH 7.4. Nominally Ca- free solution was prepared by omitting CaCl₂ from the Tyrode's solution. The high-potassium, low-chloride solution had the following composition (mM): 80 K-glutamate, 50 KCl, 20 taurine, 3 KH₂PO₄, 10 glucose, 10 HEPES, and 0.2 EGTA. The pH was adjusted to 7.4 with KOH.

Electrical Recording. A few drops of the cell suspension were placed in a chamber (0.5-ml volume) mounted on a modified stage of an inverted microscope (Nikon Diaphot, Tokyo, Japan). The chamber was superfused at a rate of 0.5 ml/min with normal external solution at room temperature (21-23°C). Currents were recorded using the whole-cell patch-clamp method (Hamill et al., 1981) and an Axopatch IC patch-clamp amplifier (Axon Instruments, Inc., Burlingame, CA). A Labmaster-TL/1 interface (Axon Instruments) controlled by pClamp 6.0 software (Axon Instruments) was used to generate voltage-clamp command protocols and acquire data. Currents were filtered at 2 kHz with a four-pole Bessel filter, digitally sampled at 4 kHz and stored on the hard disk of an Epson 486Dx/33 computer. Micropipettes were pulled from borosilicate glass capillary tubes (TW 150-6, World Precision Instruments, Sarasota, FL) on a programmable horizontal puller (Sutter Instruments, Novato, CA). When filled with the intracellular solution, the pipette tip resistance was 1 to 2 M. Whole-cell capacitance and series resistance (Rs) compensations were optimized to minimize capacitive currents and reduce voltage errors.

Protocols and Analysis. Cells with resting potentials of 75 mV or more negative were used for voltage clamp experiments. After membrane patch rupture, the cells were superfused with the Ca-Co or the TEA-Ca-Co external solutions. In rat ventricular myocytes, at least three different calcium-independent outward potassium currents activated by depolarization have been identified (Apkon and Nerbonne, 1991; Slawsky and Castle, 1994; Scamps, 1996). These include an I_to sensitive to 4-aminopyridine (4-AP), a delayed rectifying outward K⁺ current (I_K) that activates and inactivates slowly and is sensitive to external TEA, and a sustained current (I_sus) that is 4-AP- and TEA-insensitive.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Tonic Block of I_to by Disopyramide. Tonic effects of disopyramide on I_to were obtained in the presence of TEA-Ca-Co external solution. Most of the experiments were performed in the TEA-containing solution unless otherwise indicated. Disopyramide decreased the peak outward current in a concentration-dependent manner, and accelerated the rate of apparent inactivation. Figure 1, A and B, shows I_to elicited by a voltage step from 70 to +50 mV in control, and in the presence of 100 and 300 µM. The concentration-response relationship for reduction of the integral of I_to by disopyramide is shown in Fig. 1C. The data was fit with a Hill equation to obtain a K_d of 259 µM and a Hill coefficient (n_H) of 1.07. 

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Effects of disopyramide on Ito. A, superimposed currents traces of Ito obtained under control conditions and the presence of 100 µM disopyramide. B, effect of 300 µM disopyramide. C, concentration-response curve of the effect of disopyramide on the integral of Ito during a pulse of 200-ms duration.

(1)

1

1

1

View larger version (10K): [in this window] [in a new window]   Fig. 2.   A, inactivation time course of Ito under control conditions. Time course was fitted by one exponential with  = 49.3 ms. B, inactivation time course Ito in the presence of 100 µM disopyramide. Time course was fitted by two exponentials with 1 = 16.9 ms and 2 = 174 ms. C, rate of block (1/block) as a function of the concentration.

Use-Dependent Effects of Disopyramide on I_to. Fig. 3, A and B, illustrates results obtained in an experiment designed to test for use-dependent block of I_to by disopyramide. A train of 16 pulses (30-ms duration, to +50 mV) was applied from an HP of 70 mV at a frequency of 1 Hz. Under control conditions, the current induced by each pulse remained constant during the pulse train. In the presence of 100 µM disopyramide, the amplitude of peak outward current during the first pulse was 15% less than control, and declined with successive pulses to reach a steady-state level equivalent to 60% of the control amplitude after four to six pulses. The average data from 12 cells is plotted in Fig. 3C.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Use-dependent effects of disopyramide on Ito. A, superimposed records of pulses 1st, 2nd, 3rd, and 16th under control conditions. B, in the presence of 100 µM disopyramide. C, use-dependent development of block during the train of pulses. Peak Ito for each pulse in the train is normalized by dividing it by peak amplitude of Ito of the first pulse. Mean ± S.D. of 12 cells is shown. Disopyramide produced a block during the first pulse of 21 ± 5% and an additional use-dependent block of 26 ± 6% with a  = 0.68 pulse 1.

Recovery from Inactivation of I_to. Recovery of I_to from disopyramide-induced block was assessed with a paired voltage step protocol. A 100-ms conditioning pulse to +50 mV was used to inactivate I_to. The membrane potential was then clamped to 70 mV to allow recovery of channels from inactivation or from drug block. Recovery was assessed with a second pulse to +50 mV after a variable time at 70 mV. Recovery from inactivation of I_to was complete when the conditioning interval was >200 ms (Fig. 4A), and was best fit by a single exponential function with a time constant (_r) of 45.9 ms (n = 16; Fig. 4C). In the presence of disopyramide, recovery was prolonged with a rapid and slow phase (Fig. 4B). The fast time constant was 51 ± 11 and 49 ± 013 ms in the presence of 30 and 100 µM disopyramide, respectively, similar to the single time constant of recovery (_r) found under control conditions. The slow time constant (_s) was two orders of magnitude greater than _r; 5211 ± 822 ms at 30 µM, and 5048 ± 755 ms at 100 µM disopyramide. The relative magnitude of the slow component of recovery was 0.22 ± 0.08 at 30 µM and 0.35 ± 0.11 at 100 µM disopyramide.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Disopyramide induced a component of slow recovery from block. We have used a standard two-pulse protocol to +50 mV (100-ms duration) from an HP of 70 mV. A, superimposed current traces under control conditions. B, current traces in the presence of 100 µM disopyramide. C, time course of the recovery from block under control conditions was complete in about 200 ms, fitted by a single exponential with a  = 45 ± 9 ms (mean ± S.D.; n = 6). In the presence of 30 and 100 µM disopyramide the process was biexponential, at 30 µM disopyramide the fast component time constant f was 51 ± 11 ms and slow component time constant s was 5211 ± 822 ms. In the presence of 100 µM drug f was 49 ± 13 ms and s was 5048 ± 755 ms. The magnitude of the slow component was 0.22 ± 0.08 at 30 µM disopyramide and 0.35 ± 0.11 at 100 µM.

Voltage-Dependent Onset and Recovery from Block. To determine if the use-dependent effects of disopyramide was voltage-dependent, the HP was varied between 40 and 90 mV during pulse trains to +50 mV. The duration of each pulse was 30 ms, and they were applied at a frequency of 1 Hz (Fig. 5A). The use-dependent block induced by 100 µM disopyramide was accentuated at more negative HPs. For example, the steady-state level of block was 7% at 40 mV and 47% at 90 mV. In Fig. 5B, the recovery from block in the presence of disopyramide 100 µM is shown at an HP of 50 and 90 mV. The onset of block was biexponential, with a _f of 85 ms at 50 mV and 26 ms at 90 mV. However, _s was 454 ms at 50 mV and 7900 ms at 90 mV. The values of _r for control, and _f and _s in the presence of disopyramide using an HP of 40 mV to 90 mV are shown in Fig. 5C. The value of _r and _f at all voltages was similar, suggesting that the fast component of recovery observed in the presence of disopyramide corresponds to the recovery from inactivation of drug-free I_to channels, and that _s reflects the recovery of channels from block. In contrast to _r and _f the values of _s increased at more negative membrane potentials.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   A, modulation by HP of the use-dependent effects of disopyramide on Ito. Normalized peak Ito during trains of pulse in the presence of 100 µM disopyramide. The trains of 30-ms pulse to +50 mV at a frequency of 1 Hz were applied from different HPs. The use-dependent block was 7 ± 3% at 40 mV, 31 ± 9% at 60 mV, 39 ± 8% at 70 mV, and 47 ± 11% at 90 mV. B, time course of recovery from block of Ito in the presence of disopyramide, at different HPs. f was 85 ms at 50 mV and 26 ms at 90 mV. s was 454 at 50 mV and 7900 at 90 mV. C, voltage dependence of the recovery from inactivation (r) under control conditions, and disopyramide f and s recovery from block.

Effects of Duration of Conditioning Pulse on Recovery from Block. To determine if disopyramide blocks I_to channels in the inactivated state, recovery from block in the presence of drug was studied using conditioning pulses of 50 and 500 ms. Recovery from block was very slow after a 50-ms conditioning pulse (Fig. 6A), but when the duration of the conditioning pulse was increased to 500 ms, 95% of the peak amplitude was recovered after an interval of 150 ms (Fig. 6B). In Fig. 6C, the whole time course of the recovery from block is plotted for the two different conditioning pulses. The slow component of recovery had a similar time constant, 4870 ± 675 and 5042 ± 895 ms (n = 7) with 50- and 500-ms conditioning pulses, respectively. However, the relative magnitude of the slow component of recovery was 0.46 ± 0.06 after a 50-ms duration conditioning pulse, as opposed to 0.07 ± 0.04 after a 500-ms conditioning pulse (n = 7). These data show that block of I_to was reduced by prolonged depolarizations, suggesting unblock of drug from the inactivated state of the channel.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Recovery from block after 50 (A)- and 500 (B)-ms conditioning pulses, in the presence of 100 µM disopyramide. C, time course of recovery from block using both conditioning pulse durations. Time constants were similar using both protocols. However, the magnitude of the slow component was 46 ± 6% using 50-ms duration conditioning pulse and 7 ± 4% using 500-ms duration conditioning pulse (n = 7 cells).

Time Course of Block Onset by Disopyramide. The time course of disopyramide-induced block of I_to during a depolarizing pulse to +50 mV was determined using a paired pulse protocol from an HP of 70 mV. The duration of the first conditioning pulse was varied between 1 ms and 2 s. The duration of the second pulse was fixed at 100 ms. The conditioning and test pulses were separated by a gap of 300 ms at 70 mV to allow sufficient time for the channels that were not blocked by disopyramide during the conditioning pulse to recover from inactivation. Hence, comparisons of the test current amplitudes in the presence or absence of the conditioning pulse gave a measure of the disopyramide-induced block.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Time course of disopyramide block during depolarization. Block was determined from the current during the test pulse to +50 mV applied after a conditioning pulse to +50 mV with variable duration (1-1000 ms) and a gap of 300 ms at 70 mV. A and B, selected tracings for conditioning pulses of 10-, 30-, 50-, 100-, 300-, and 500-ms duration. C, normalized peak current amplitude as a function of the duration of the conditioning depolarization. Note that block increased up to conditioning pulses of 50 ms, but declined with further prolongation of the conditioning depolarization. Inset shows the first 50 ms of the plot. The decaying phase was fitted by an exponential function with  = 10.5 ± 2.4 ms. The rising phase was fitted by another exponential function with  = 373 ± 26 ms (mean ± S.D.; n = 6 cells).

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Time course of disopyramide block in the presence of a Ca-Co external solution. The protocol was the same as that used in Fig. 7. A, records obtained under control conditions. B, records obtained in the presence of 100 µM disopyramide. C, normalized peak current amplitude as a function of the duration of the conditioning pulse. Under control conditions, the peak current amplitude of the test pulse decreased as a function of the increase in duration of the conditioning pulse, reaching a 15 ± 5% decrease after a 2-s duration conditioning pulse. In the presence of disopyramide block increased up to a conditioning pulse duration of 50 ms, but declined with further prolongation of the conditioning depolarization.

Voltage Dependence of Depolarization-Induced Block by Disopyramide. The voltage dependence of activation of I_to and disopyramide-induced block was compared (Fig. 9). The voltage dependence of activation for I_to was measured from an HP of 70 by applying 15-ms pulses to potentials ranging from 30 and +100 mV. After this activating pulse, the cell was immediately repolarized to 40 mV and the tail current amplitude after repolarization was used as a measure of I_to activation. Figure 9C (open circles) shows the voltage dependence of tail current amplitude, normalized to the peak at +60 mV. The membrane potential dependence of activation was fitted by a Boltzmann function, with the following equation:

(2)

V_m is membrane potential, V_h is the voltage at which 50% of the channels are open, and s represents the slope factor for the relationship. V_h was 0.5 mV and s was 10.9 mV. The voltage dependence of the disopyramide-induced block of I_to was measured using a paired-pulse protocol (Fig. 9B). A conditioning pulse of 50-ms duration was applied to membrane potentials between 50 and +100 mV from an HP of 70 mV. This pulse was followed by 300 ms at 70 mV and a test pulse to +50 mV. Under control conditions, this protocol produced less that 5% depression of I_to after the most positive conditioning potential. In the presence of 100 µM disopyramide, conditioning depolarizations positive to about 40 mV produced measurable block of I_to and the amount of block increased at more positive potentials. The voltage dependence of disopyramide I_to block was plotted and fitted by the equation:

(3)

where the first part of the equation is similar to the single Boltzmann function used to fit the activation curve. D is the disopyramide concentration; K_d is the apparent binding constant; and z, F, R, and T have the usual meanings.  represents the fraction of the transmembrane electrical field sensed by a single charge at the receptor site. The calculated values were 239 µM for K_d, and 0.19 for  (n = 5). The K_d value obtained in these experiments is close to the K_d value obtained by different methods (Figs. 2 and 3B). It is clear that channel blockade has a steep voltage dependence coincident with channel opening and an additional weakly voltage-dependent component that reflects the effect of the transmembrane electrical field on the charged drug (Snyders et al., 1992).

View larger version (16K): [in this window] [in a new window]   Fig. 9.   A, current traces of an experiment studying voltage dependence of activation of Ito. Activation was measured by analysis of tail currents obtained after brief activating pulses (15 ms) to potentials between 30 and 100 mV. These activating pulses were followed by a test pulse to 40 mV. Tail currents were measured after pulses to 30, 10, +10, +20, +50, and +70 mV. B, current traces from the same experiment after superfusion with 100 µM disopyramide. From an HP of 70 mv, conditioning pulses of 50-ms duration to different voltages between 50 and +100 mV were applied every 30s. These conditioning pulses were followed by a rest interval of 300 ms at 70 mV to allow the recovery from inactivation of Ito channels; the rest period was followed by the test pulse to +50 mV. Test pulses were preceded by conditioning pulses to 50, 0, +50, and +90 mV. C, steady-state voltage dependence of activation data were fitted by a Boltzmann function. The best fit was obtained with a Vh value of 0.5 mV and a slope factor of 10.9. The voltage dependence of block was fitted by eq. 3. The values for drug block (100 µM disopyramide) were Kd = 239 µM and  = 0.192.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The main goal of the present work was to study the effect of disopyramide on I_to. Three different outward currents have been described in these cells (Apkon and Nerbonne, 1991; Slawsky and Castle, 1994; Scamps, 1996): 1) a rapidly activating and inactivating I_to, sensitive to 4-AP; 2) a more slowly activating and inactivating current I_K, sensitive to external TEA; and 3) an apparently time-independent component I_sus, insensitive to 4-AP and TEA. I_to was isolated from the two other types of outward currents by using external TEA (50 mM) to block I_K and by digitally subtracting the noninactivating, time-independent I_sus. Externally applied TEA has been shown to selectively block I_K (Apkon and Nerbonne, 1991; Slawsky and Castle, 1994). The effects of disopyramide on I_to, including the decrease in peak amplitude, the initial acceleration of apparent inactivation, and the slowing of the late phase of repolarization were similar in the presence and in the absence of TEA. These findings suggest that TEA does not modify the tonic block of I_to by disopyramide.

Disopyramide Is An Open Channel Blocker of I_to. The inhibition of I_to by disopyramide is characterized by a concentration-dependent reduction in peak I_to and an acceleration of the apparent rate of current inactivation. These results are similar to those found with different I_to open channel blockers like tedisamil (Dukes et al., 1990; Wettwer et al., 1998), bupivacaine (Castle, 1990), clofilium (Castle, 1991), quinidine (Slawsky and Castle, 1994; Clark et al., 1995), propafenone, and flecainide (Slawsky and Castle, 1994). The characteristics of the disopyramide-induced block of I_to strongly suggest that disopyramide blocks the open state of the channel. Evidence for this mechanism includes: 1) disopyramide accelerated the initial apparent inactivation rate of I_to, 2) at the onset of the depolarizing pulses there was no inhibition of I_to, indicating that disopyramide does not bind block channels in the rested state, 3) a close correlation between the voltage dependence of current activation and disopyramide-induced block, and 4) the unblock of I_to during prolonged depolarizations, which decrease the open state of the channel.

Competition between Drug Binding and Inactivation of I_to. We propose that disopyramide competes with the inactivation gate of the I_to channel. A key experiment that supports this proposal was performed in the presence of Ca-CoTEA external solution (Fig. 7). However, qualitatively similar results were found in a solution without TEA (Fig. 8). These results show that the presence of TEA in the external solution also did not modify the phasic effects of disopyramide on I_to. Therefore, we conclude that TEA did not modify the blocking effects of disopyramide on I_to.

Kinetic Scheme. To explain the disopyramide block of I_to, we propose a simple kinetic scheme:
where C, O, and I are closed, open, and inactivated states of the channel, respectively; n indicates that there is a series of several closed states leading to an open state (e.g., Zagotta and Aldrich, 1990); B is a nonconducting, disopyramide-blocked channel. At positive membrane potentials, channel closure in the absence of drug occurs preferentially by inactivation. Hence, a prolonged depolarization results in rapid activation of the channel, followed by a slower decay. In the presence of disopyramide, drug block cannot occur until the channels open, and the decay of current will be initially accelerated by disopyramide because the open channel can close by two pathways, namely, inactivation (O  I) and disopyramide block (O  B). The model also includes a mutually exclusive interaction between drug block and channel inactivation, that is, blocked channels cannot inactivate and inactivated channels cannot be blocked. This constraint can explain the slowing of the final phase of apparent inactivation observed in the presence of disopyramide that produced a crossover of the current traces (Fig. 1; Wettwer et al., 1998). It can also explain the time-dependent development of block of I_to during the first 50 ms of the depolarizing pulses, whereas longer depolarizations induce unblocking of the channel (Fig. 8B).

Comparison with Previous Studies. In rabbit ventricular myocytes, it was recently reported (Virag et al., 1998) that disopyramide decreased the amplitude of I_to by an open channel-blocking mechanism. However, this effect was not voltage- or use-dependent. In addition, disopyramide did not affect I_to recovery from inactivation. These results are in apparent contradiction to the results of the present work. Similar contradictory results have been obtained with quinidine. In rabbit atrial myocytes, Liu et al. (1996) found that quinidine induced a tonic decrease of I_to without additional use-dependent effects, but did not affect recovery from inactivation. However, in rat ventricular myocytes, quinidine produced significant tonic and use-dependent effects and a slowing of the I_to recovery from inactivation (Slawsky and Castle, 1994; Clark et al., 1995). Possible explanations for these apparent discrepancies are that the kinetics of recovery from inactivation of I_to in rabbit atrial and ventricular myocytes is a very slow process, with time constants in the range of seconds (Giles and Imaizumi, 1988), similar to the unblocking kinetics of disopyramide (this study). Another possibility is that the molecular bases of I_to and/or the I_to mechanism of block by disopyramide in rat are different than in rabbit. In rat ventricular muscle, it has been suggested that Kv 4.2 and Kv 4.3 isoforms contribute to I_to (see Fiset et al., 1997). The slow recovery from inactivation of I_to in rabbit cardiac myocytes suggests that Kv 1.4 could be the molecular basis of this current (see Yeola and Snyders, 1997).