Introduction

Top Abstract Introduction Methods Results Discussion References

Class III antiarrhythmic agents exert their actions by prolonging the cardiac action potential and thereby increasing the refractory period, without altering phase 0 sodium current or conduction velocity. A variety of class III drugs are in current clinical use or in development, including sotalol (Singh and Nademanee, 1987), dofetilide (Rasmussen et al., 1992), E-4031 (Fujiki et al., 1994) and ambasilide (Takanaka et al., 1992). Although data have been presented that suggest that sotalol is a highly selective antagonist of the delayed rectifier K⁺ current (Carmeliet, 1985), and particularly of the rapid component I_Kr (Sanguinetti and Jurkiewicz, 1990), other work has suggested that sotalol potently inhibits the transient outward current (Berger et al., 1989). Dofetilide and E-4031 have been characterized as specific blockers of I_Kr (Rasmussen et al., 1992; Fujiki et al., 1994; Colatsky et al., 1990), largely on the basis of experiments with cells isolated from experimental animals. Ambasilide has been found to inhibit both components of the delayed rectifier: I_Kr and the slower component I_Ks (Zhang et al., 1992). In an experimental canine model of AF, ambasilide was found to be a more potent antiarrhythmic agent than d-sotalol and to prolong refractoriness with much less reverse use-dependence than d-sotalol (Wang et al., 1994b).

Recent work has helped to clarify the ionic currents governing human atrial repolarization. Whereas delayed rectifier K⁺ currents in human atrium resemble corresponding currents in a variety of animal cells (Wang et al., 1993a; Wang et al., 1994a), some K⁺ currents in human atrial cells show important differences from other species. For example, human atrial I_to has quite different kinetic properties from I_to in rabbit atrium (Fermini et al., 1992), and a novel type of delayed rectifier with kinetic and pharmacologic properties resembling those of the cloned human K⁺ channel Kv1.5 appears to be important in human atrial repolarization (Wang et al., 1993b). The latter current has been designated I_Kur, or the ultra-rapid delayed rectifier, because its activation kinetics are two orders of magnitude faster than those of I_Kr (Wang et al., 1993b).

Relatively little is known about the effects of antiarrhythmic drugs on ionic currents in human heart cells. We have shown that quinidine produces open-channel block of I_to in human atrial cells, whereas flecainide inhibits I_to in a fashion that suggests the highest affinity for the inactivated state (Wang et al., 1995). Quinidine was found to inhibit I_Kur significantly at concentrations in the clinically relevant range, which suggests that I_Kur block may contribute to the drug's antiarrhythmic properties in the human, whereas flecainide had no detectable effect on I_Kur at concentrations as large as 10 µM (Wang et al., 1995).

I_to and I_Kur appear to play important roles in human atrial repolarization (Shibata et al., 1989; Escande et al., 1987; Wang et al., 1993b) and may therefore be important targets for antiarrhythmic drug action. The effects of class III drugs on I_Kur have not, to our knowledge, been studied. There is limited information about class III drug effects on I_to, and the results that have been obtained, largely relating to sotalol, are somewhat contradictory (Carmeliet, 1985; Berger et al., 1989). We therefore designed the present study to evaluate the effects of three class III antiarrhythmic drugs, d-sotalol, E-4031 and ambasilide, on I_to and I_Kur in human atrial myocytes.

	Methods

Top Abstract Introduction Methods Results Discussion References

Isolation of human atrial cells. Specimens of human right atrial appendage were obtained from the hearts of 36 patients (58 ± 3 years old) undergoing aortocoronary bypass surgery. All patients were free of supraventricular tachyarrhythmias, and the atria were grossly normal at the time of surgery. The procedure for obtaining the tissue was approved by the Ethics Committee of the Montreal Heart Institute. Samples were quickly immersed in nominally Ca⁺⁺-free Tyrode's solution (100% O₂, 37°C) of the following composition (mM): NaCl, 126; KCl, 5.4; MgCl₂, 1.0; NaH₂PO₄, 0.33; dextrose, 10; HEPES (Sigma Chemicals, St. Louis, MO), 10; pH adjusted to 7.4 with NaOH. The myocardial specimens were chopped with scissors into cubic chunks and placed in a 25-ml flask containing 10 ml of the Ca²⁺-free Tyrode's solution. The tissue was gently agitated by continuous bubbling with 100% O₂ and stirring with a magnetic bar. After an initial 5 min in this solution, the chunks were reincubated in a similar solution containing 390 U/ml collagenase (CLS II, Worthington Biochemical, Freehold, NJ) and 4 U/ml protease (Type XXIV, Sigma Chemicals). The first supernatant was removed after 45 min, and the chunks were reincubated in a fresh enzyme-containing solution. Microscopic examination was performed every 15 min to determine the number and quality of the isolated cells. When the yield appeared to be maximal, the chunks were suspended in a solution containing (mM): KCl, 20; KH₂PO₄, 10; glucose, 10; glutamic acid, 70; -hydroxybutyric acid, 10; taurine, 10; EGTA, 10; albumin, 1%; pH adjusted to 7.4 with KOH, and gently pipetted.

Data acquisition. Only quiescent rod-shaped cells showing clear cross-striations were used. A small aliquot of the solution containing the isolated cells was placed in a 1-ml chamber mounted on the stage of an inverted microscope. Five minutes was allowed for cell adhesion to the bottom of the chamber, and then the cells were superfused at 3 ml/min with a solution containing (millimolar concentrations): NaCl, 126; KCl, 5.4; MgCl₂, 0.8; CaCl₂, 1.0; NaH₂PO₄, 0.33; HEPES, 10; glucose, 5.5; pH adjusted to 7.4 with NaOH. In order to minimize possible contamination from I_K, I_K1, I_K,Ach and choline-activated K⁺ current (Fermini and Nattel, 1994), the following chemicals were present during current recording: TEA (Sigma Chemicals, 10 mM, to inhibit I_K), BaCl₂ (Sigma Chemicals, 1 mM, to inhibit I_K1) and atropine (Sigma Chemicals, 1 µM, to inhibit I_K,Ach and choline-activated K⁺ current). In preliminary experiments and previously published studies (Wang et al., 1993a; Wang et al., 1993b), we found these interventions to be without effect on I_to and I_Kur. Sodium current was inhibited with the use of a holding potential of 50 mV and/or equimolar choline replacement of Na⁺ in the superfusate. CdCl₂ (200 µM) was added to the superfusate to inhibit Ca⁺⁺ current. 4AP was obtained from Sigma Chemicals, prepared as a 1 M stock solution with pH adjusted to 7.4 with the use of 1 N HCl and added at selected concentrations as specified below. Experiments were conducted at room temperature in order to resolve the rapid activation and deactivation kinetics of I_Kur; previous studies have shown that the amplitude of I_Kur at room temperature is similar to that at 37°C (Wang et al., 1993b).

Data analysis. The amplitude of I_to was measured as the difference between the peak of the transient outward current and the sustained current at the end of the pulse, as previously described (Wang et al., 1993b; Wang et al., 1995). To record I_Kur in the absence of contamination by I_to, we used a 1-sec prepulse to +40 mV to inactivate I_to 10 msec before a depolarizing test pulse, a procedure that we have previously developed and validated (Wang et al., 1993b). I_Kur was measured in two ways: 1) as described previously (Wang et al., 1993b; Wang et al., 1995), based on the maximum current upon depolarization in the presence of a depolarizing prepulse to inactivate I_to, and 2) in terms of the tail current upon repolarization from a depolarizing test potential to 20 mV.

	Results

Top Abstract Introduction Methods Results Discussion References

Effects of d-sotalol and E-4031 on I_to and I_Kur. Concentrations of d-sotalol up to 500 µM, which fully inhibit I_Kr (Sanguinetti and Jurkiewicz, 1990) and are substantially higher than the maximum therapeutic concentration in the human (Wang et al., 1986), failed to affect I_to recorded on 300-msec depolarizing pulses delivered at 0.1 Hz from 50 mV. Similarly, I_Kur elicited by 160-msec pulses from 50 mV (after a 1-sec prepulse to +40 mV to inactivate I_to) was not altered by the drug. Overall, concentrations of 5, 10, 50 and 500 µM d-sotalol (in five cells at each concentration), produced 1.2 ± 0.3, 0.9 ± 0.4, 1.5 ± 0.7 and 1.4 ± 0.9% changes in I_to and 0.8 ± 0.3, 2.1 ± 0.7, 1.3 ± 0.4 and 1.5 ± 0.9% changes in I_Kur, respectively, at +40 mV. When a train of 15 conditioning 90-msec pulses to +50 mV at a frequency of 1, 2 and 3.3 Hz was introduced before the test pulse to evaluate possible use-dependent actions, no effect of 100 µM d-sotalol on I_to or I_Kur was noted. I_to averaged 850 ± 65, 845 ± 59 and 831 ± 62 pA at +50 mV at 1, 2 and 3.3 Hz, respectively, before and 849 ± 62, 846 ± 61 and 829 ± 58 pA after 100 µM d-sotalol (P = NS for d-sotalol vs. control for each). Similarly, I_Kur averaged 499 ± 52, 488 ± 49 and 475 ± 46 pA at +50 mV at 1, 2 and 3.3 Hz, respectively, before and 490 ± 49, 490 ± 52 and 480 ± 50 pA after 100 µM d-sotalol (P = NS for d-sotalol vs. control).

Effects of ambasilide on I_to. The response of I_to to ambasilide is illustrated in figure 1. Figure 1A shows representative currents recorded in one cell under control conditions. Ambasilide produced a slight decrease in I_to at a concentration of 10 µM (fig. 1B). At a higher concentration (50 µM, fig. 1C), ambasilide decreased I_to substantially and caused apparent acceleration in the initial decay of I_to after peak values were attained. I_to inhibition by ambasilide was almost fully reversed after 20 min of drug washout (fig. 1D).

View larger version (40K): [in this window] [in a new window]   Fig. 1.   Concentration-dependent effects of ambasilide on Ito in a representative cell. Ito was elicited by 300-msec depolarizing pulses from 50 mV to +50 mV at 0.1 Hz. Recordings are shown under control conditions (panel A) and then in the presence of 10 µM (panel B) and 50 µM (panel C) ambasilide, followed by washout (panel D).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   A) Current-voltage relations under control, 10, 50 and 100 µM ambasilide and washout conditions (mean ± S.E.M., n = 8). B) Percent reduction in Ito (relative to control at the same voltage). *P < .05; **P < .01 vs. control (n = 8). C) Concentration-response relation for ambasilide effect on Ito at +50 mV. Symbols represent experimental data (mean ± S.E.M., n = 6 cells exposed to all concentrations); solid line is best-fit equation of the form E = Emax (1 + [K/C]n), where E is the effect at concentration C, Emax is the maximum effect, K is the concentration for half-maximal action, and n is the Hill coefficient. D) Mean data for voltage dependence of Ito inactivation and activation in the absence (circles) and presence (triangles) of 50 µM ambasilide (mean ± S.E., n = 6). Inactivation was assessed with a two-pulse protocol, with a 1-sec prepulse to voltages between 120 and +20 mV followed by a 1-sec test pulse to +60 mV. Activation was assessed by the tail current at 30 mV after 5-msec depolarizations to a variety of test potentials. The curves shown are fits of mean data by Boltzmann distribution equations (see text).

Effects of ambasilide on I_Kur. The effects of ambasilide on I_Kur in a representative myocyte are illustrated in figure 3A. I_Kur was recorded with the use of the protocol shown in the inset, including a 1-sec prepulse to +40 mV to inactivate I_to, followed by a 160-msec test pulse delivered at 0.1 Hz to a variety of potentials between 40 and +50 mV (results at +50 mV are shown in the figure). Current under control conditions (fig. 3A) shows the rapid activation with little or no inactivation typical of I_Kur. Ambasilide (100 µM) caused a substantial reduction in both step current elicited by depolarization and tail current at 20 mV. The effect of ambasilide was qualitatively similar to, although quantitatively somewhat less than, that of 4AP at a concentration (5 mM) we have previously shown to block I_Kur fully (Wang et al., 1993b). The drug-sensitive difference currents shown in figure 3B indicate the similar morphology of the current inhibited by ambasilide and 4AP. Figure 3C shows the mean ± S.E.M. current-voltage relationships of step current sensitive to 100 µM ambasilide () and 5 mM 4AP (), which were quite similar in form.

View larger version (26K): [in this window] [in a new window]   Fig. 3.   A) IKur recorded from one cell with the protocol shown in the inset of panel B, under control conditions and then after exposure to 100 µM ambasilide and 5 mM 4AP. B) Currents sensitive to 5 mM 4AP (a) and to 100 µM ambasilide (b), based on digital subtraction of currents in the presence of drug from those in the absence of drug in panel A. C) Current-voltage relation of 4AP- and ambasilide-sensitive current obtained as illustrated in panels A and B. D) Current-voltage relations for step current under control, 10 (), 50 () and 100 () µM ambasilide and washout conditions (mean ± S.E.M., n = 8). E) Percent reduction in IKur (relative to control at the same voltage). *P < .05; **P < .01 vs. control (n = 8). F) Concentration-response relation for ambasilide inhibition of IKur at +50 mV. Open symbols represent experimental data (mean ± S.E.M., n = 6 cells exposed to all concentrations); solid lines are best-fit concentration-response equations. Filled symbols show data corrected for nonspecific component by calculating ambasilide's effects as (b/a) × 100%, where b is current inhibited by the drug at a given concentration and a is current inhibited by 5 mM 4AP, obtained as shown in figure 3B.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Effects of ambasilide on IKur tail currents. A) Recordings of IKur obtained in one cell with the voltage protocol shown in inset, under control, ambasilide (10, 50 and 100 µM) and washout conditions. B) Current-voltage relations (mean ± S.E.M. for five cells exposed to each condition) for IKur tail currents recorded as shown in the inset. C) Percent change in IKur tail current produced by various concentrations of ambasilide. **P < .01 vs. control at same voltage. D) Concentration-response relation for ambasilide inhibition of tail currents. Symbols are mean data at +50 mV; solid line is best-fit concentration-response curve. Dashed line is concentration-response curve for drug effects on step current, corrected for 4AP-sensitive component as shown by filled symbols and corresponding curve fit in figure 3F.

State-dependence of ambasilide actions. The above results show that ambasilide inhibits I_to and I_Kur in a concentration-dependent and voltage-independent fashion. To evaluate further the possibility of state-dependent blocking actions, we assessed the time-dependence of block. If ambasilide interacted preferentially with open or inactivated I_to channels with recovery from the rested state slower than spontaneous recovery from inactivation, then slowed recovery after a depolarizing pulse would result. Figure 5A presents an analysis of the time-dependent recovery of I_to as determined with the two-pulse protocol shown in the inset. Mean values for I_to of the test pulse (P₂) normalized to current during a basic pulse (P₁, at 0.1 Hz) are shown as a function of the P₁ to P₂ interval in six cells. The best-fit exponential curves to each set of data are shown; they are very similar. Exponential curve fitting to recovery data in each cell provided recovery time constants of 31.8 ± 2.9 and 33.2 ± 3.0 msec under control and 100 µM ambasilide conditions, respectively, which indicated no change in I_to recovery kinetics.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   A) Time course of recovery from inactivation of Ito in absence (open circles) and presence (closed circles) of 100 µM ambasilide obtained with protocol in inset. Results are mean ± S.E.M. of five cells studied under both conditions. Best-fit exponential recovery curves are provided. B) Time to peak of Ito as a function of test potential under control conditions and in the presence of 100 µM ambasilide (mean ± S.E.M., n = 6, reduction by drug significant at P < .01 for all voltages).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effects of ambasilide on the time course of Ito inactivation. Representative recordings from one cell upon depolarization to +50 mV under control conditions (panel A) and in the presence of 50 µM ambasilide (panels B and C). Best-fit monoexponential relations to raw data are shown in panels A and B, whereas panel C shows a biexponential fit to results in the presence of ambasilide. D) Time constants of Ito inactivation under control and drug conditions. C = control; AM10, AM50 and AM100 = 10, 50 and 100 µM ambasilide, respectively. **P < .01 vs. control.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   The time course of development of inhibition of Ito (panel A) and IKur (panel B) by 10 and 50 µM ambasilide upon depolarization to +50 mV from a holding potential of 50 mV. Reductions in current over time are expressed as changes from control values (IC) to values in the presence of ambasilide (ID). The curves are least-squares fits to the equations f = a + b × [1  exp(t/1)] × exp(t/2) in panel A and f = a  b × exp(t/) in panel B. The average time constants are as follows: for Ito, 1 6.4 ± 0.5 msec and 2.8 ± 0.3 msec, 2 525 ± 60 msec and 588 ± 62 msec at 10 and 50 µM ambasilide, respectively; for IKur, 6.5 ± 0.8 msec and 2.7 ± 0.3 msec at 10 and 50 µM ambasilide.

1

1

1

1

1

1

	Discussion

Top Abstract Introduction Methods Results Discussion References

We have shown that the class III antiarrhythmic agents E-4031 and d-sotalol do not alter I_to or I_Kur in human atrial myocytes. On the other hand, the experimental class III drug ambasilide inhibited both currents. These results point to the possibility of developing class III antiarrhythmic agents with a different profile of channel blocking actions from previous class III drugs, whose use has been limited by proarrhythmic properties (Hondeghem and Snyders, 1990).

Comparison with previous studies of ionic mechanisms of the drugs studied. Several studies have indicated that E-4031 is a highly selective blocker of I_Kr in animal cells (Sanguinetti and Jurkiewicz, 1990; Colatsky et al., 1990). The present study extends the specificity of E-4031 action by excluding a blocking effect on I_Kur and I_to in human atrium. The specificity of sotalol's blocking action has been less clear. Carmeliet (1985) showed a high degree of selectivity for I_K but significant effects on other currents at higher concentrations. Sanguinetti and Jurkiewicz (1990) showed a high degree of sotalol selectivity for the I_Kr component, but other studies have suggested relatively strong effects on I_to (Berger et al., 1989). The present study indicates that E-4031 and d-sotalol, even at very high concentrations, have no effect on human atrial I_to or I_Kur.

Potential significance of our findings. Several newer class III agents appear to block the rapid component of I_K with relatively high selectivity (Colatsky et al., 1990). These agents tend to have strong reverse use-dependent effects on repolarization (Wang et al., 1994b; Hondeghem and Snyders, 1990; Jurkiewicz and Sanguinetti, 1993; Colatsky and Argentieri, 1994), which are associated with important risks of proarrhythmic reactions because of excessive delays in repolarization at slow HR (Hondeghem and Snyders, 1990; Nattel and Zeng, 1984). Balser et al. (1991) and Zhang et al. (1992) have suggested that class III agents without selectivity for I_Kr may have a more favorable profile of rate-dependent actions. In vivo studies suggest that amiodarone and ambasilide do, indeed, have less reverse use-dependent actions on repolarization than highly selective I_Kr-blocking compounds (Wang et al., 1994b; Sager et al., 1993). The present work points to another potentially interesting action of class III drugs: blockade of currents particularly important in repolarizing human atrial cells. Both I_to and I_Kur have been shown to play important roles in human atrial repolarization (Shibata et al., 1989; Escande et al., 1987; Wang et al., 1993b). Furthermore, I_Kur appears to be absent in human ventricle (Li et al., 1996). Therefore, blockade of I_Kur and/or I_to may be an advantageous property for class III compounds.

Mechanisms of channel blocking action. We found that ambasilide produced time-dependent block of I_to and I_Kur, a result that suggests an open-channel blocking mechanism. These properties are similar to blocking mechanisms we noted previously for quinidine on I_to and I_Kur (Wang et al., 1995). Koidl et al. (1996) recently reported effects of ambasilide on rapidly and slowly inactivating components of I_to in human atrial cells. They observed inhibiting effects of the drug on each component, which they interpreted as representing blocking actions on I_to and I_Kur, respectively. They noted that ambasilide accelerated I_to inactivation, decreasing the time constants of both phases. The acceleration of I_to inactivation that they noted is compatible with an open-channel blocking action, as demonstrated in the present study. Although Koidl et al. (1996) suggested that the effect of ambasilide on the slowly inactivating component of I_to may be due to an effect on I_Kur, they did not study the latter directly. Our findings indicate that the effects of ambasilide they hypothesized on I_Kur do, in fact, occur and that the acceleration of inactivation they noted may be due to open-channel blockade. Unlike Koidl et al. (1996), we did not observe a slowly inactivating component of I_to in human atrial myocytes. The difference is probably due to differences in bath temperature. We studied currents at room temperature in order to observe accurately the rapid activation of I_Kur (Wang et al., 1993b), whereas Koidl et al. (1996) worked at 37°C, at which temperature I_Kur inactivation might accelerate enough to be measurable during a 300-msec depolarizing pulse.

Potential limitations. Studies in native myocytes always present difficulties in terms of isolating the currents of interest. I_to is relatively distinct in terms of its rapid inactivation. I_Kur is more difficult to isolate, and we have used two previously described approaches (Wang et al., 1993b): a depolarizing pulse to inactivate I_to and sensitivity to 4AP to identify the highly 4AP-sensitive I_Kur component. In addition, we analyzed effects on I_Kur tail currents and obtained results that were qualitatively consistent with the different methods used. The use of cloned channels in expression systems allows for clearer study of single currents but is limited by uncertainties regarding the relationship between cloned channels and their native counterparts, as well as by potential distortions due to differences between model expression systems and native tissues in the properties of membranes and in important intracellular regulatory processes.