Block of cardiac delayed-rectifier and inward-rectifier K⁺ currents by nisoldipine

Abstract

1. The objective of this study was to determine the concentration-dependent effects of nisoldipine, a dihydropyridine Ca²⁺ channel blocker, on K⁺ currents in guinea-pig ventricular myocytes.

2. Myocytes in the conventional whole-cell configuration were bathed in normal Tyrode's solution or K⁺-free Tyrode's solution for the measurement of the effects of 0.01–100 μM nisoldipine on rapidly activating delayed-rectifier K⁺ current (I_Kr), slowly activating delayed-rectifier K⁺ current (I_Ks), inwardly rectifying K⁺ current (I_K1), and reference L-type Ca²⁺ current (I_Ca,L).

3. Nisoldipine inhibited I_Kr with an IC₅₀ of 23 μM, and I_Ks with an IC₅₀ of 40 μM. The drug also had weak inhibitory effects on inward- and outward-directed I_K1; the IC₅₀ determined for outward-directed current was 80 μM.

4. Investigation of nisoldipine action on I_Ks showed that inhibition occurred in the absence of previous pulsing, and with little change in the time courses of activation and deactivation. However, the drug-induced inhibition was significantly weaker at ⩾+30 mV than at +10 mV.

5. We estimate that nisoldipine is about 30 times less selective for delayed-rectifier K⁺ channels than for L-type Ca²⁺ channels in fully polarised guinea-pig ventricular myocytes, and several orders less selective in partially depolarised myocytes.

Introduction

Members of the 1,4-dihydropyridine class of compounds have important clinical application in the management of cardiovascular disorders, and the basis of this use is the binding of drug molecules to high-affinity inhibitory sites on L-type Ca²⁺ channels (Kohlhardt & Fleckenstein, 1977; Singh et al., 1983; Sanguinetti & Kass, 1984; Schwartz & Triggle, 1984; McDonald et al., 1994; Roden, 1996; Kamp et al., 1999). Dihydropyridines can also block other ion channels, including voltage-dependent Na⁺ channels (Yatani & Brown, 1985) and delayed-rectifier K⁺ channels, in a variety of excitable and inexcitable cells (Nishi et al., 1983; Hume, 1985; Terada et al., 1987; Richard et al., 1988; Jacobs & DeCoursey, 1990; Tatsuta et al., 1994).

Delayed-rectifier K⁺ channels in cardiac cells are also affected by dihydropyridines. In particular, ca. 10 μM concentrations of nifedipine and/or nisoldipine have been shown to have marked inhibitory effects on the fast-activating K⁺ channels that carry transient outward current (I_to) (Kass, 1982; Gotoh et al., 1991; Jahnel et al., 1994) and ultrarapid K⁺ current (I_Kur) (Zhang et al., 1997). However, the susceptibility of these channels to nifedipine/nisoldipine action is not representative of all cardiac delayed-rectifier K⁺ channels. For example, we have recently reported that 10 μM nifedipine has no significant effect on the two components of delayed-rectifier K⁺ current (I_K) in guinea-pig ventricular myocytes, rapidly activating I_Kr and slowly activating I_Ks; however, higher concentrations produced a concentration-dependent block with IC₅₀ values near 300 μM (Zhabyeyev et al., 2000). Whether nisoldipine (which is a much more potent inhibitor of L-type Ca²⁺ channel blocker than nifedipine) is an equally weak inhibitor of these K⁺ currents has not been established.

The primary objective of the present study was to determine the concentration-dependent effects of nisoldipine on delayed-rectifier I_Kr and I_Ks in guinea-pig ventricular myocytes. For comparison with these data, we also measured the effects of the drug on inwardly rectifying K⁺ current (I_K1) and on I_Ca,L. Nisoldipine had concentration-dependent inhibitory actions on all the three K⁺ currents, and these are discussed and compared with pertinent findings in earlier studies.

Methods

All the procedures were carried out in accordance with the national and university regulations on the care and treatment of laboratory animals.

Ventricular myocytes

Male guinea-pigs (250–300 g) were killed by cervical dislocation, and single ventricular myocytes were enzymatically isolated as described previously (Ogura et al., 1995). The excised hearts were mounted on a Langendorff column, and retrogradely perfused through the aorta with Ca²⁺-free Tyrode's solution (37°C) containing collagenase (0.08–0.12 mg ml⁻¹; Yakult Pharmaceutical Co., Tokyo, Japan) for 10–15 min. The cells were dispersed and stored at 22°C in a high-K⁺, low-Na⁺ solution supplemented with 50 mM glutamic acid and 20 mM taurine. A few drops of the cell suspension were placed in a 0.3-ml perfusion chamber mounted on an inverted microscope stage. After the cells had settled to the bottom, the chamber was perfused (2 ml min⁻¹) with Tyrode's solution at 36°C. The Tyrode's solution contained (in mM) NaCl 140, KCl 5.4, CaCl₂ 1.8, MgCl₂ 1, glucose 10, and N-2-hydroxyethylpiperazine-N′-2-ethanesulphonic acid (HEPES) 5 (pH 7.4 with NaOH). In some experiments, the Tyrode's solution was replaced by a Na⁺-free solution (equimolar substitution with trimethylammonium ion), or by a K⁺-, Ca²⁺-free Tyrode's solution (KCl and CaCl₂ omitted) that also contained 0.2 mM Cd²⁺ to suppress Ca²⁺ channel current.

Whole-cell membrane currents were recorded using an EPC-7 amplifier (List Electronic, Darmstadt, Germany). Recording pipettes were fabricated from thick-walled borosilicate glass capillaries (H15/10/137, Jencons Scientific Ltd, Bedfordshire, U.K.) and filled with a solution that contained (in mM) KCl 40, potassium aspartate 106, MgCl₂ 1, K₂-ATP 4, ethylene glycol-bis(b-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA) 5, and HEPES 5 (pH 7.2 with KOH). The pipettes had resistances of 1.5–2.5 MΩ when filled with pipette solution, and liquid junction potentials between external and pipette-filling solution were nulled prior to patch formation. Series resistance ranged between 3 and 7 MΩ, and was compensated by 60–80%. Membrane current signals were filtered at 3 kHz, and digitised with an A/D converter (Digidata 1200A, Axon Instruments) and pCLAMP software (Axon Instruments) at a sampling rate of 8 kHz prior to analysis.

Drugs

Nisoldipine was kindly provided by Bayer Inc. (Etobicoke, ON, Canada). The drug was dissolved in dimethyl sulphoxide (DMSO) (Sigma Chemical Co., St Louis, MO, U.S.A.) (0.1 M stock solution) and stored in the dark at −20°C. Appropriate amounts of stock solution were freshly added to bathing solutions, and these were protected from the light during all experiments. The highest final concentration of DMSO was 0.1%, a concentration that has no significant effect on membrane currents in guinea-pig ventricular cells (Ogura et al., 1995). Nevertheless, the appropriate concentrations of DMSO were included in the control solutions used in experiments with high concentrations of nisoldipine. E4031 was obtained from Eisai (Tokyo, Japan) and dissolved in the bathing solution.

Statistics

Results are expressed as means±s.e.m., and Student's t-test or one-way ANOVA followed by Dunnett's multiple comparison test was used to determine the significance of drug effects. Differences were considered to be significant when P<0.05.

Results

Inhibition of total I_K

Figure 1a shows records of membrane currents obtained from a representative myocyte bathed in normal Tyrode's solution and depolarised for 500 ms from −40 mV to potentials up to +70 mV. The depolarisations elicited inward I_Ca,L that reached a maximal amplitude near 0 mV, and activated time-dependent outward K⁺ current that increased with positive potential and deactivated when the membrane was repolarised to −40 mV (I_K,tail). Exposure to 1 μM nisoldipine for 10 min resulted in near-complete inhibition of I_Ca,L, but had little effect on the amplitudes of I_K elicited by depolarisations and repolarisations. However, concentrations of nisoldipine higher than 1 μM had marked concentration-dependent inhibitory effects on I_K. In the examples shown in Figure 1b and c, 10 μM drug depressed the time-dependent current by about 20% and 100 μM depressed it by about 80%. These inhibitions were registered in myocytes pretreated with 1 μM nisoldipine, that is, they were essentially independent of changes in I_Ca,L.

Figure 1.

Effects of nisoldipine on membrane currents in guinea-pig ventricular myocytes. The myocytes were bathed in normal Tyrode's solution, held at −80 mV, and depolarised after 200-ms prepulses (−40 mV) to more positive potentials for 500 ms at 0.1 Hz; tail currents were recorded on repolarisations to −40 mV. (a) Selective inhibition of inward I_Ca,L by 1 μM nisoldipine (Nis). (b,c) Inhibition of outward K⁺ currents at positive potentials, and K⁺ tail currents at −40 mV, by 8-min exposures to 10 and 100 μM nisoldipine. The myocytes were pretreated with 1 μM nisoldipine for 10 min. The dashed lines on the records here and in other figures indicate zero-current levels. (d) I_K,tail–V relationships determined before, during, and 10 min after removal of 100 μM nisoldipine. The calibration bars on the I_K,tail records (top) indicate 100 ms and 200 pA.

The pronounced inhibition of I_K by high concentrations of nisoldipine was not a manifestation of toxicity because ca. 10-min washouts generally restored the current to at least 50% of predrug amplitude (e.g., Figure 1d) (also see below). By comparison, washout of 1 μM nisoldipine for 13±2 min only restored I_Ca,L to 29±5% of control (n=4), and washout of 50 μM nisoldipine for 14±2 min only restored the current to 9±4% of control (n=5).

Myocytes were repolarised to −40 mV for measurement of I_K,tail elicited after 500-ms depolarisations to test potentials (V) between −30 and +70 mV. Average I_K,tail–V relationships recorded before and 7–10 min after additions of 3, 30, and 100 μM nisoldipine are shown in Figure 2a. The low concentration had little effect on the I_K,tail–V relationship, whereas the 30 and 100 μM concentrations depressed its amplitude by about 50 and 80%, respectively. For reference in the interpretation of these data, Figure 2a also shows the effects of 3 μM E4031 on the I_K,tail–V relationship. The selective I_Kr inhibitor (Sanguinetti & Jurkiewicz, 1990) almost completely abolished I_K,tail elicited after depolarisations from −40 mV up to test potential 0 mV, but had little effect on the increment in tail amplitude induced by depolarisations to more positive potentials.

Figure 2.

Inhibition of I_Kr by nisoldipine. Myocytes were bathed in normal Tyrode's solution and depolarised from prepulse −40 mV to more positive potentials for 500 ms at 0.1 Hz for the measurement of I_K,tail on repolarisations to −40 mV. (a) Average I_K,tail–V relationships determined from myocytes treated with single concentrations of nisoldipine (3 μM (n=4), 30 μM (n=5), 100 μM (n=6)). Also shown (lower right) are the effects of the I_Kr-selective inhibitor E4031 (3 μM) (n=7); the results of the E4031 series (near complete inhibition of I_K,tail at ⩽0 mV) validate the method (I_K,tail amplitude after pulses to 0 mV) used for measurement of I_Kr in (b, c) below. ‡P<0.001 (paired t-test at +70 mV). (b) Time course of inhibition of I_Kr,tail (0 mV) by 30 μM nisoldipine. (c) Dependence of I_Kr inhibition on nisoldipine concentration. Myocytes were treated for 7–10 min with one or two concentrations of the drug. The Hill equation fitting the data has an IC₅₀ of 23±2 μM and a coefficient of 1.05. Numbers of observations are shown in parentheses.

Inhibition of I_Kr

The result with E4031 in Figure 2a indicated that the amplitudes of tail currents after 500-ms depolarisations to 0 mV could be used to evaluate the inhibition of I_Kr by nisoldipine. In practice, the amplitude of I_Kr,tail was measured following 500-ms depolarisations to 0 mV at 0.1 Hz. The data in Figure 2b indicate that the inhibition by nisoldipine reached a steady state within 5 min. As illustrated by the significant recovery of current at 0 mV following 10-min washout of 100 μM drug in Figure 1d, the inhibition of I_Kr by high concentrations of nisoldipine was slowly reversible. For example, in four myocytes treated with 30 μM nisoldipine for 5–8 min, I_Kr recovered to 89±6% of predrug amplitude after a 10–15 min washout period.

The dependence of I_Kr,tail inhibition on the concentration of nisoldipine is shown in Figure 2c. The lowest concentration that produced a significant inhibition was 3 μM (reduction to 90±2% of predrug control (n=10), P<0.05, paired t-test). The overall data are well described by the Hill equation with an IC₅₀ of 23±2 μM and a coefficient of 1.05.

Inhibition of I_Ks

Dependence on drug concentration

Inhibition of I_Ks by nisoldipine was evaluated in myocytes that were superfused with a solution (K⁺-, Ca²⁺-free plus 0.2 mM Cd²⁺) that enhances I_Ks and suppresses I_Kr and Ca²⁺-dependent currents (Sanguinetti & Jurkiewicz, 1990; Jones et al., 1998). Under these conditions, both the time-dependent currents elicited by 500-ms depolarisations and the subsequent tail currents at −30 mV were insensitive to 5 μM E4031 (Figure 3a). However, I_Ks isolated in this manner was rapidly inhibited by high concentrations of nisoldipine, and the inhibition was only slowly and partially reversible upon removal of the drug (Figure 3b). In four myocytes treated with 100 μM nisoldipine for 5 min, a 10-min washout period restored I_Ks to 73±5% of predrug control amplitude.

Figure 3.

Inhibition of I_Ks by nisoldipine. Myocytes were bathed in K⁺-, Ca²⁺-free Cd²⁺ solution to suppress I_Kr and Ca²⁺-dependent currents, and I_Ks was activated by 500-ms or 2-s depolarisations from −30 mV to more positive potentials at 0.1 Hz. (a) Results from two myocytes demonstrating the lack of effect of I_Kr inhibitor E4031 on the I_K activated under the foregoing experimental conditions. (b) Time course of inhibition of I_Ks,tail by 30 and 100 μM nisoldipine. (c) Dependence of inhibition on nisoldipine concentration. Left: examples of records obtained from myocytes depolarised with 2-s pulses. Right: data (open symbols) obtained from myocytes exposed to one or two concentrations of the drug for 6–10 min each. Inhibition was measured from changes in the amplitudes of tail currents after 2-s depolarisations to +70 mV, and the Hill equation fitting the data has an IC₅₀ of 40±1 μM and a coefficient of 1.07. Numbers of observations are shown in parentheses. For comparison, the plot also shows data (filled symbols) obtained when the inhibition of I_Ks was evaluated as the drug-induced change in the amplitude of time-dependent current elicited by 500-ms pulses to +70 mV in myocytes bathed in normal Tyrode's solution (n=4–10).

The dependence of I_Ks inhibition on nisoldipine concentration was evaluated from measurements of the amplitudes of tail currents after 2-s depolarisations to +70 mV before and during steady-state drug action (Figure 3c). The data from these determinations are well described by the Hill equation with an IC₅₀ of 40±1 μM and a coefficient of 1.06 (Figure 3c, open circles).

The dependence of I_Ks inhibition on ⩾10 μM nisoldipine was also assessed from changes in the amplitude of time-dependent outward currents elicited by 500-ms depolarisations to +70 mV in myocytes superfused with normal Tyrode's solution and pretreated with 1 μM nisoldipine (see Figure 1 for representative traces). The filled circles on the plot in Figure 3c indicate that the nisoldipine–I_Ks relationship determined in this manner was similar to that determined from measurements of tail I_Ks under K⁺-free conditions.

Time- and voltage-dependent features

Inhibition of fast-activating types of delayed-rectifier K⁺ current by dihydropyridines generally increases with time during a depolarising pulse to a positive potential (Gotoh et al., 1991; Jahnel et al., 1994; Zhang et al., 1997). In the present study, we examined the effect of 100 μM nisoldipine on the time course of development of I_Ks during long (5 s) depolarisations to +50 mV. With the exception of the first 100–200 ms, the current records were satisfactorily fitted with single exponential functions (Figure 4a, left). Although the time constants of these functions were longer in the presence of the drug (587±37 ms) than in its absence (predrug, 424±31 ms; after washout, 518±23 ms) (n=4 myocytes), the differences were not significant at the 0.05 level (paired t-tests). When records obtained before and after addition of nisoldipine were scaled and superimposed, it was evident that drug treatment did not modify the ratio of the amplitudes of activating I_Ks and I_Ks,tail (Figure 4a, right). The same figure indicates that the time course of I_Ks,tail was also unchanged.

Figure 4.

Characteristics of nisoldipine block of I_Ks in myocytes bathed in K⁺-, Ca²⁺-free Cd²⁺ solution. (a) Lack of significant effect of 100 μM nisoldipine on the time courses of I_Ks activation and deactivation. Left: original records whose activation time courses have been fitted with single exponentials (small dots), having time constants of 488 ms (control) and 576 ms (nisoldipine). Right: superimposition after scaling of the nisoldipine record to the control record. (b) Effect of rest on the development of block. Left: measurements of I_Ks amplitudes. Right: records obtained on the last pulse before rest (Ctl) and the first postrest pulse (PR 1).

To determine whether block of I_Ks by nisoldipine has classical use-dependent features, myocytes were stimulated at 0.1 Hz (500-ms pulses from −30 to +50 mV), rested at −80 mV, exposed to 100 μM drug, and restimulated 4 min later. The records and time plot in Figure 4b indicate that near-maximal inhibition was reached on the first postrest depolarisation. In four experiments of this type, tail I_Ks on the first postrest pulse was 32±6% of predrug control, and tail I_Ks on the 30th postrest pulse was a similar 33±4%.

The dependence of block on I_Ks-activating voltage was determined by measuring the amplitudes of tail currents elicited after 2-s depolarisations to different test potentials. Data from a representative myocyte illustrate that, in contrast to low concentrations of the drug, 30 μM nisoldipine decreased the amplitude of tail currents after depolarisations to all test voltages between +10 and +90 mV (Figure 5a). When tail I_Ks–V data from six experiments with 30 μM nisoldipine were normalised, it was evident that tails after small depolarisations were inhibited to a greater degree than tails after large depolarisations. As depicted in Figure 5b, the relative block declined from 0.58±0.04 at +10 mV to 0.34±0.03 at +90 mV, with most of the voltage dependence arising between +10 and +50 mV. Analysis of the data (one-way ANOVA followed by Dunnett's multiple comparison test) indicates that block at +10 mV was significantly larger than block at +30, +50, +70, and +90 mV (P<0.05–0.01).

Figure 5.

Dependence of nisoldipine block of I_Ks on the voltage of the activating pulse. Myocytes were bathed in K⁺-, Ca²⁺-free Cd²⁺ solution and depolarised for 2 s to potentials V for evaluation of I_Ks,tail amplitude on repolarisation. (a) Records and I_Ks,tail analysis from a myocyte exposed to 3 μM and then 30 μM nisoldipine. (b) Dependence of the degree of block on voltage. The data are from six myocytes that were exposed to 30 μM nisoldipine. ^*P<0.05, ‡P<0.01 (versus +10 mV data).

Inhibition of I_K1

When myocytes were bathed in Tyrode's solution that contained 0.2 mM Cd²⁺ (to block I_Ca,L), 500-ms pulses from −40 mV to potentials between −120 and −10 mV elicited currents that were large and inward up to −80 mV, and smaller, outward, and quasi-time-independent at more positive potentials (Figure 6a and b); these features, and the negative slope in the I–V relationship at <−50 mV, are typical for I_K1 over this voltage region (Jones et al., 1999). High concentrations of nisoldipine lowered the end-of-pulse amplitudes of both inward and outward I_K1; in four myocytes, 30 μM nisoldipine inhibited the inward current at −120 mV by 34±4%, and the outward current at −40 mV by 35±5% (both P<0.01, multiple comparison test). Inhibition caused by ⩽30 μM nisoldipine was only partially reversed by washout of the drug for ca. 10 min (not shown).

Figure 6.

Inhibition of I_K1 by nisoldipine. Myocytes were bathed in Tyrode's solution that contained 0.2 mM Cd²⁺, and held at −80 or −90 mV. (a) Effects of 30 μM nisoldipine on I_K1 measured as the end-of-pulse current amplitude on 500-ms pulses from prepulse −40 mV to potentials between −120 and −10 mV. Left: records from a representative experiment; right: summary of data from four experiments. †P<0.01 (multiple comparison test at −120 mV). (b) Time courses of changes in two representative experiments. Left: lack of effect of 1 μM nisoldipine on outward-directed I_K1 at −40 mV; right: inhibition of inward (−90 mV) and outward (−50 mV) I_K1 by 30 μM nisoldipine. (c) Dependence of inhibition of outward I_K1 on the concentration of nisoldipine. The Hill equation fitting the data has an IC₅₀ of 80±3 μM and a coefficient of 0.83. Noncumulative drug exposures lasted for 8–10 min. Numbers of myocytes are shown in parentheses.

The time course and concentration dependence of nisoldipine-induced inhibitions of outward-directed I_K1 were evaluated by measuring the outward current amplitude at −40 mV (Jones et al., 1999). The time plots in Figure 6b indicate that outward I_K1 at −40 mV was generally stable over tens of minutes, and unaffected by 1 μM nisoldipine, but quickly inhibited by 30 μM nisoldipine. The steady-state effects of the drug are well described by the Hill equation with an IC₅₀ of 80±3 μM and a coefficient of 0.83 (Figure 6c).

Inhibition of I_Ca,L

For comparison with the foregoing data on the inhibition of K⁺ currents by nisoldipine, we determined concentration–response relations for inhibition of I_Ca,L by the drug. To obtain data that might be applicable to inhibition during an action potential in fully polarised ventricular myocytes, a group of myocytes was bathed in Na⁺-free Tyrode's solution to eliminate Na⁺ current, held at –90 mV, and depolarised to 0 mV for 200 ms at 0.1 Hz. The myocytes were pretreated with 3 μM E4031 to suppress I_Kr, and exposed to 0.4 mM Cd²⁺ after 5–7 min nisoldipine to establish the background current level. A second group of myocytes was bathed in normal Tyrode's solution, and subjected to a voltage protocol similar to that used in many of the experiments on K⁺ currents. These myocytes were depolarised by a prepulse to –40 mV for 200 ms to inactivate Na⁺ current, and then pulsed to 0 mV at 0.1 Hz; they were also treated with 3 μM E4031 and exposed to Cs²⁺ after the effect of nisoldipine had reached a steady state.

The results obtained from the two groups of myocytes are summarised in Figure 7. The data from the group held at –90 mV are described by the Hill equation with IC₅₀ of 0.73±0.13 μM and a coefficient of 1.03, whereas the data from the other group have an IC₅₀ of 0.08±0.01 μM and a coefficient of 1.07. The difference in IC₅₀ is an indication of the voltage dependence of nisoldipine action on L-type Ca²⁺ channels (Kass, 1982; McDonald et al., 1994).

Figure 7.

Inhibition of I_Ca,L by nisoldipine. Myocytes were either bathed in Na⁺-free Tyrode's held at –90 mV, and depolarised to 0 mV for 200 ms (0.1 Hz), or bathed in normal Tyrode's solution, held at –80 mV and depolarised by a prepulse to −40 to 0 mV for 200 ms (0.1 Hz). At 5 min postpatch, the myocytes were pretreated with 3 μM E4031 for 3 min and then exposed to single concentrations of nisoldipine for 5–7 min prior to application of 0.4 mM Cd²⁺ to establish the background current level at 0 mV. Peak I_Ca,L was measured as the background current level minus inward peak level. The Hill equation fitting the data obtained from myocytes held at –90 mV has an IC₅₀ of 0.73±0.13 μM and a coefficient of 1.03. The fit of the data from the other group of myocytes has an IC₅₀ of 0.08±0.01 μM and a coefficient of 1.07.

Discussion

The present study provides new information on the sensitivity of cardiac inward-rectifier and delayed-rectifier K⁺ channels to inhibition by nisoldipine. The results are discussed and compared with findings from previous studies on interactions between nisoldipine and K⁺ channels.

Effects on inwardly rectifying K⁺ current

Nisoldipine inhibited guinea-pig ventricular I_K1 with an IC₅₀ of 80 μM. The inhibition appeared to be independent of voltage and direction of the current, and was at least partially reversible with prolonged washout of the drug. In an earlier study on nisoldipine, Kass (1982) observed that concentrations up to 100 μM had little effect on outward I_K1 in cardiac Purkinje fibres. A possible reason for the discrepancy with the present findings is that the Purkinje fibres had been injected with tetrabutylammonium bromide to block I_to prior to the trials with nisoldipine. The pretreatment may have interfered with the measurements because the quaternary compound is itself a blocker of inwardly rectifying K⁺ current (Barros et al., 1992).

Inhibition of delayed-rectifier K⁺ currents

Fast-activating currents

Dihydropyridine Ca²⁺-channel blockers have previously been shown to inhibit fast-activating types of delayed-rectifier K⁺ currents in a variety of noncardiac cells. In most cases, nifedipine was the drug under investigation (e.g., Nishi et al., 1983; Gola & Ducreux, 1985; Jacobs & DeCoursey, 1990; Tatsuta et al., 1994). However, Nerbonne & Gurney (1987) included nisoldipine in their study on fast-activating inactivating K⁺ current in Aplysia bag cell neurons, and reported that the IC₅₀ of the drug was ca. 4 μM.

The fast-activating inactivating I_to in rabbit atrial myocytes has also been found to be quite sensitive to nisoldipine (IC₅₀ 4.7 μM) (Gotoh et al., 1991).

Slower-activating cardiac K⁺ currents

There have been two previous studies on the effects of nisoldipine on slower-activating global I_K in cardiac cells. In the first of these, Kass (1982) found that 10 μM nisoldipine had no effect on I_K (I_x) elicited by 1-s depolarisations of cardiac Purkinje fibres. Subsequently, Hume (1985) reported that 10 μM nisoldipine reduced I_K in frog atrial myocytes by about 40%. In relation to the present findings on I_Kr and I_Ks, it is important to note that frog atrial I_K has been characterised as a single K⁺ conductance (Hume & Giles, 1983) that resembles I_Ks in some respects (slowly activating, noninactivating), but not in others (half-activation near 0 mV, saturation at ca. +40 mV) (see Hume, 1985). Thus, direct comparison of the findings on guinea-pig ventricular I_Kr (IC₅₀ 23 μM) and I_Ks (IC₅₀ 40 μM) with those on frog atrial I_K (IC₅₀ 16 μM) may not be justified. Nevertheless, these determinations taken together indicate that nisoldipine has a weaker action on slower-activating types of cardiac delayed-rectifier K⁺ channels than has been reported for fast-activating I_to-carrying channels by Gotoh et al. (1991).

Characteristics of I_Ks block

Block of I_Ks by nisoldipine had the following features: (i) the time course of nonblocked current in the presence of the drug was similar to that of current in the absence of the drug, (ii) block was fully established on the first pulse after a rest in the presence of the drug, and (iii) the degree of block was dependent on the voltage of the activating depolarisation. The first two features suggest (but do not prove) that, in contrast to dihydropyridine block of cardiac I_to (Gotoh et al., 1991; Jahnel et al., 1994) and hKv1.5 current (Zhang et al., 1997), block of I_Ks by nisoldipine occurs via binding of drug molecules to closed rather than open channels.

A dependence of dihydropyridine block on voltage has been ruled out in some studies on noncardiac delayed-rectifier K⁺ channels (e.g., Nerbonne & Gurney, 1987; Barros et al., 1992; Pappone & Ortiz-Miranda, 1993), but has been detected in two previous studies on cardiac delayed-rectifier channels. Hume (1985) observed that inhibition of frog atrial I_K by 10 μM nisoldipine was weaker on depolarisations to positive potentials than to negative ones, and Zhang et al. (1997) reported that block of hKv1.5 channels by 30 μM nifedipine was weaker at potentials above +20 mV. In the present case, the block of I_Ks by 30 μM nisoldipine was approximately 40% weaker at ⩾+50 mV than at +10 mV. Since nisoldipine has a very low pK_a and is almost completely in the neutral form at pH 7.4, it is difficult to view the voltage dependency in terms of charged drug molecules under the influence of the electrical field. However, Hume (1985) noted that this type of voltage-dependent block could be due to a drug-induced positive shift of the voltage dependence of current activation. It seems unlikely that a shift is the major cause of the voltage dependence of the I_Ks block studied here, because the time course of I_Ks at +50 mV was little affected by 100 μM nisoldipine. Rather, the voltage dependence may arise from a coupling of the binding site to a voltage-sensitive process such that unbinding is enhanced at more positive potentials. As proposed by Zhang et al. (1997) for nifedipine block of cardiac hKv1.5 channels, the dihydropyridine site on Ks channels may be near the external mouth of the pore, and relief of block may be linked to increased K⁺ permeation at more positive voltages.

Selectivity of nisoldipine action on cardiac K⁺ and Ca²⁺ channels

The IC₅₀ values for nisoldipine block of I_Kr, I_Ks, and I_K1 were 23, 40, and 80 μM, respectively. Thus, nisoldipine is far from being a potent blocker of these K⁺ pathways. This is particularly the case with regard to I_Kr because there are numerous clinical and experimental drugs that inhibit this current with an IC₅₀<1 μM (e.g., Sanguinetti & Jurkiewicz, 1990; Carmeliet, 1992; Salata et al., 1995; Roden, 1996; Yang et al., 1997; Jones et al., 1998; Zhang et al., 1999). Nisoldipine is a potent blocker of L-type Ca²⁺ channels in cardiac cells, and is commonly used to abolish I_Ca,L in electrophysiological studies on these cells. The present results indicate that any spillover effects on K⁺ currents are likely to be negligible when the drug is used to abolish I_Ca,L in studies on guinea-pig ventricular myocytes. The placing of a numerical value on the relative selectivity of nisoldipine for its binding site in L-type Ca²⁺ channels over those in delayed-rectifier K⁺ channels is complicated by the fact that binding to the Ca²⁺ channels is strongly dependent on the holding potential (Sanguinetti & Kass, 1984; McDonald et al., 1994). For this reason, the IC₅₀ for I_Ca,L in cardiac Purkinje fibres declined from 1.34 μM at a holding potential −80 mV, to ca. 0.2 μM at −50 mV, and a calculated 0.001 μM at ca. −20 mV (Sanguinetti & Kass, 1984).

In the present study, the IC₅₀ for block of I_Ca,L at 0 mV was 0.73±0.13 μM when myocytes were held at –90 mV, and a much lower 0.08±0.1 μM when myocytes were held at –80 mV and depolarised by a prepulse to –40 mV for 200 ms prior to the test pulses to 0 mV. Since the block of K⁺ currents was not influenced by holding potential, the relative selectivity of nisoldipine for block of Ca²⁺ channels over delayed-rectifier channels is estimated to be about 30 when well-polarised myocytes undergo rapid depolarisation to a plateau potential (for example, during an action potential). The selectivity increases to near 400 when depolarisation to a plateau potential occurs after a prepulse from –80 to –40 mV (a protocol frequently used in electrophysiological studies). The latter estimate of relative selectivity is close to the 1000-fold value estimated by Hume (1985), whose reference IC₅₀ for frog atrial I_Ca,L was 0.016 μM (holding potential –60 mV).

Abbreviations

DMSO

dimethyl sulphoxide

EGTA

ethylene glycol-bis(b-aminoethyl)-N,N,N,N-tetraacetic acid

HEPES

N-2-hydroxyethylpiperazine-N′-2-ethanesulphonic acid

I–V

current–voltage

IC₅₀

concentration that produces 50% of maximal inhibition

I_Ca,L

L-type Ca²⁺ current

I_K

delayed-rectifier K⁺ current

I_Kr

rapidly activating component of I_K

I_Ks

slowly activating component of I_K

I_Kur

ultrarapid K⁺ current

I_K1

inward-rectifying K⁺ current

I_to

transient outward current