Inhibition of HERG K⁺ current and prolongation of the guinea-pig ventricular action potential by 4-aminopyridine

Abstract

4-Aminopyridine (4-AP) has been used extensively to study transient outward K⁺ current (I_TO,1) in cardiac cells and tissues. We report here inhibition by 4-AP of HERG (the human ether-à-go-go-related gene) K⁺ channels expressed in a mammalian cell line, at concentrations relevant to those used to study I_TO,1. Under voltage clamp, whole cell HERG current (I_HERG) tails following commands to +30 mV were blocked with an IC₅₀ of 4.4 ± 0.5 mm. Development of block was contingent upon HERG channel gating, with a preference for activated over inactivated channels. Treatment with 5 mm 4-AP inhibited peak I_HERG during an applied action potential clamp waveform by ∼59 %. It also significantly prolonged action potentials and inhibited resurgent I_K tails from guinea-pig isolated ventricular myocytes, which lack an I_TO,1. We conclude that by blocking the α-subunit of the I_Kr channel, millimolar concentrations of 4-AP can modulate ventricular repolarisation independently of any action on I_TO,1.

4-Aminopyridine (4-AP) has been used extensively in cardiac electrophysiology to study the physiological roles of transient outward potassium current (I_TO,1; Wang et al. 1991; Gillis et al. 1998; Elizalde et al. 1999; Lei et al. 2000; Kocic et al. 2002) and, in recent years, of the ‘ultra-rapid’ delayed rectifier current (I_Kur; Li et al. 1996; Yue et al. 1996). 4-AP is generally considered selective for these currents in the heart, with micromolar concentrations being used to block I_Kur (Li et al. 1996) and millimolar concentrations to block I_TO,1 (typically 0.5 to 10 mm; e.g. Wang et al. 1991; Gillis et al. 1998; Elizalde et al. 1999; Lei et al. 2000; Kocic et al. 2002). However, limited published data raise the possibility that 4-AP may not be entirely selective for I_TO,1 over the rapid delayed rectifier K⁺ current (I_Kr) at concentrations used to study I_TO,1 (Mitcheson & Hancox, 1999; Liu et al. 1999; Zhang et al. 2000). Given the importance of I_Kr in regulating action potential repolarization and duration (Mitcheson & Sanguinetti, 1999), a non-selective effect of 4-AP on I_Kr would significantly influence the interpretation of data from experiments in which 4-AP is applied to cardiac muscle cells or tissues. The aim of this study was to determine whether or not 4-AP inhibits ionic current (I_HERG) carried by K⁺ channels encoded by HERG (the human ether-à-go-go-related gene, which encodes the α-subunit of the I_Kr channel; Warmke & Ganetzky, 1994). 4-AP was found both to inhibit I_HERG recorded from human embryonic kidney cells and to prolong action potentials from guinea-pig ventricular myocytes, which lack I_TO,1. These findings have widespread implications for the use of 4-AP as an investigative tool in cardiac muscle electrophysiology.

METHODS

Maintenance of a mammalian cell line stably expressing HERG

Measurements of I_HERG were made from human embryonic kidney (HEK 293) cells stably expressing HERG (cell line generously donated by Professor Craig January, University of Wisconsin; Zhou et al. 1998). Cells were passaged using a non-enzymatic agent (Splittix, AutogenBioclear) and plated out onto fragments of sterilized glass coverslips in 30 mm petri dishes containing a modification of Dulbecco's modified Eagle's medium with Glutamax-1 (DMEM; Gibco), supplemented with 10 % fetal bovine serum, 400 μg ml⁻¹ gentamicin (Gibco) and 400 μg ml⁻¹ geneticin (G418; Gibco). Cells were incubated at 37 °C (5 % CO₂) for a minimum of 2 days prior to any electrophysiological study.

Isolation of guinea-pig ventricular myocytes

Male adult guinea-pigs were killed by cervical dislocation (a Schedule 1 procedure according to the Home Office UK Animals (Scientific Procedures) Act 1986) and ventricular myocytes were enzymatically dispersed from both ventricles using a method described previously (Levi & Issberner, 1996). Prior to use, the isolated cells were kept at 4 °C in Kraft-Brühe medium (Isenberg & Klockner, 1982).

Electrophysiological recordings

Cells were superfused with a normal Tyrode solution, which contained (mm): 140 NaCl, 4 KCl, 2.5 CaCl₂, 1 MgCl₂, 10 glucose, 5 Hepes, (titrated to pH 7.45 with NaOH). Experimental solutions were applied using a home-built, warmed solution delivery system capable of changing the bathing solution surrounding a cell in <1 s. Patch-pipettes were fire-polished to a resistance of 3.5–5 MΩ. The pipette solution for I_HERG measurement contained (mm): 130 KCl, 1 MgCl₂, 5 EGTA, 5 MgATP, 10 Hepes (titrated to pH of 7.2 with KOH). A K⁺-based, EGTA-free pipette solution containing 10 mm NaCl was used for action potential recordings. The ‘pipette-to-bath’ liquid junction potential was small (−3.2 mV) and was uncorrected. Series resistance (R_s) values lay typically between 4–7 MΩ and were compensated by ≥ 70 %. The expected voltage drop across residual uncompensated R_s during I_HERG measurements was therefore only small (∼3.5 mV or less) and no correction was made for this. Measurements were made at 37 ± 1 °C.

Drugs

4-AP powder (Sigma) was dissolved directly in normal Tyrode solution to make test solutions of different concentrations referred to in the Results; pH was corrected to 7.45 with HCl.

Voltage protocols and analysis

Measurements were made using an Axopatch 200 or 200A amplifier (Axon Instruments) and a CV-201/2A headstage. Voltage-clamp commands were generated using ‘WinWCP’ (John Dempster, Strathclyde University, UK) or pCLAMP (v 6.0, Axon Instruments). Data were recorded via a Digidata interface (Axon Instruments) and stored on the hard disk of a personal computer.

Concentration–response relationships

I_HERG tails were measured on repolarization to −40 mV after 1 s depolarizing voltage commands to +30 mV. 4-AP concentrations between 100 μm and 100 mm were each applied to a minimum of five different cells. The mean fractional blockade was obtained as:

(1)

where I_HERG(4AP) and I_{HERG(CONTROL)} represent tail current amplitudes in the presence and absence of 4-AP, respectively. The agent was rapidly acting (maximal effects of a given concentration typically being observed within ∼2–4 pulses), and its effects could be observed without any marked overlying current run-down. Concentration-response data were fitted by the equation:

(2)

where IC₅₀ is [4-AP] producing half-maximal inhibition of the I_HERG tail and h is the Hill coefficient for the fit.

I_HERG inhibition at different voltages

From −80 mV, 2 s depolarizations were applied to potentials between −40 and +40 mV. I_HERG tails were measured on repolarization to −40 mV. Current-voltage (I–V) relations (not shown) were constructed for each of eight cells in the absence and presence of 5 mm 4-AP. Half-maximal activation voltages were obtained by fitting each I–V relation with a modified Boltzmann equation:

(3)

where I =I_HERG tail amplitude following test potential V_m, I_max is the maximal I_HERG tail current observed, V_1/2 is the potential at which I_HERG was half-maximally activated and k is the slope factor describing I_HERG activation.

Voltage-dependent activation curves for I_HERG were obtained by calculating activation variables at 2 mV intervals between −80 and +40 mV. Values for V_1/2 and k derived from fits to experimental I–V data with eqn (3) were inserted into the following equation:

(4)

where the ‘activation parameter’ at any test potential, V_m, occurs within the range 0 to 1 and V_1/2 and k have similar meanings to those in eqn (3).

Envelope of tails

From −80 mV, membrane potential was stepped to +40 mV for varying time periods between 12.5 and 600 ms and was repolarized to − 40 mV in order to monitor I_HERG tail amplitude. After control measurements, each cell was equilibrated in 4-AP-containing solution while at the holding potential and the protocol was then reapplied. For each cell, the fractional block of I_HERG tails at each test pulse duration was determined using eqn (1). The plot of mean fractional block versus test pulse duration was then fitted with an equation of the form:

(5)

where A represents maximal fractional block, x represents pulse duration and τ is the time constant of development of blockade (ms).

Action potential voltage clamp and measurements from guinea-pig ventricular myocytes

For action potential clamp measurements of I_HERG from HEK cells, the command waveform used was the same as that employed in previous studies of I_HERG from our laboratory (Hancox et al. 1998a, b).

Action potentials were elicited from guinea-pig ventricular myocytes by brief suprathreshold current pulses at 15 s intervals. For measurements of guinea-pig I_K tails, membrane potential was held at −80 mV, depolarized to −40 mV for 20 ms, then depolarized to +20 mV for 500 ms. I_K tails were measured during subsequent repolarization to −40 mV.

Data analysis and presentation

Data are presented as means ± s.e.m. Statistical comparisons were made using Student's paired t test or one-way analysis of variance (ANOVA) with a Bonferroni post hoc test. P values of less than 0.05 were taken as significant.

RESULTS

Figure 1A shows that 4-AP produced a concentration-dependent inhibition of I_HERG and Fig. 1B shows the mean concentration–response relationship for this effect. The IC₅₀ derived from the data was 4.4 ± 0.5 mm (Hill coefficient 0.7 ± 0.1). The voltage dependence of the effect was assessed by applying commands to a range of test potentials (Fig. 1Ci and Cii). No I_HERG tail inhibition was evident after pre-pulses to −30 mV and −20 mV, whilst at more positive potentials marked inhibition of I_HERG was observed (Fig. 1Ci and Cii). Mean data are shown in Fig. 1D, which also contains mean activation curves in control (continuous line) and 4-AP (dashed line). In the presence of 4-AP a leftward shift in the half-maximal activation voltage (V_1/2; by −7.4 ± 0.9 mV, n = 8; P < 0.0001, t test) was observed. The activation shift may account for the apparent augmentation/ lack of blockade of by 4-AP observed at −30 and −20 mV. The voltage range over which inhibition was voltage dependent corresponded with that over which voltage-dependent activation of I_HERG developed (Mergenthaler et al. 2001; Paul et al. 2001). Treatment with 5 mm 4-AP did not significantly alter the mean deactivation time course of tails measured at − 40 mV, following a voltage command to +30 mV.

Figure 1. Concentration and voltage-dependent inhibition of I_HERG by 4-AP.

A, upper traces show representative current traces illustrating the effect of different concentrations of 4-AP. Lower trace represents voltage protocol used. B, concentration response relation for I_HERG tail blockade by 4-AP (n = 5–8 cells at each concentration). C, representative records in the absence (Ci) and presence (Cii) of 5 mm 4-AP. D, the mean (± s.e.m.) fractional block of I_HERG by 5 mm 4-AP over a range of potentials (n = 8 cells). I_HERG activation curves are shown in the absence (continuous line) and presence (dashed line) of 5 mm 4-AP.

The dependence of I_HERG inhibition by 4-AP upon duration of depolarization was studied using an ‘envelope of tails’ protocol (Trudeau et al. 1995) (see Fig. 2A and B). Figure 2A shows representative current traces obtained in control (Fig. 2Ai) and in 4-AP (Fig. 2Aii); Fig. 2B shows a plot of the resulting mean levels of fractional block against pulse duration. The time course of development of blockade was described by a mono-exponential curve that intersected the origin, with a time constant of 32.3 ± 6.4 ms (n = 5). Considered collectively with Fig. 1D these data demonstrate that inhibition of I_HERG by 4-AP requires channel gating to occur.

Figure 2. Time dependence of I_HERG block by 4-AP.

A, traces from an ‘envelope of tails’ protocol, in the absence (Ai) and presence (Aii) of 5 mm 4-AP. Lower traces represent the voltage protocol used. B, the mean (± s.e.m.) fractional block of I_HERG with varying pulse duration (n = 5 cells). The τ for development of inhibition was 32.3 ± 6.4 ms (n = 5).

An additional protocol (see Fig. 3A, lower trace) was employed to ascertain whether or not 4-AP showed a preference for activated or inactivated HERG channels. In control solution, I_HERG amplitude decreased markedly during a depolarisation from +20 to +80 mV (Fig. 3A, upper traces, Control) as a consequence of increased I_HERG inactivation. In the presence of 5 mm 4-AP, blockade of I_HERG at +20 mV was greater than that at +80 mV (Fig. 3A, upper traces, 4-AP). Mean data shown in Fig. 3B indicate that blockade during the +80 mV step (‘4000 ms’) was significantly lower than that at +20 mV (‘1000 ms’ and 9000 ms’). Thus, when I_HERG was reduced by extensive I_HERG inactivation the blocking effect of 4-AP was diminished. This observation suggests that 4-AP binds preferentially to activated over inactivated HERG channels. In further experiments we observed an acceleration of the rate of development of I_HERG inactivation by 4-AP (data not shown). This effect is in marked contrast with that reported for externally applied tetraethylammonium ions (TEA), which slow the distinctive, rapidly developing C-type inactivation of I_HERG (Smith et al. 1996). This action of 4-AP could have resulted either from a genuine modification of inactivation gating, or from the rapid development of open-channel block during the protocol used to study inactivation.

Figure 3. Effect of strong depolarisation on the action of 4-AP.

A, current records (upper traces) elicited by the voltage protocol shown (lower trace), applied from a holding potential of −80 mV. Arrows indicate traces obtained in control and 4-AP. B, bar charts (n = 5 cells) compare the mean level of blockade at 1000 ms into protocol (at +20 mV), at 4000 ms (at +80 mV) and at 9000 ms (after membrane potential was returned to +20 mV).

The effect of 4-AP on I_HERG elicited by an applied ventricular action potential waveform was also determined (Fig. 4A). In control solution, I_HERG amplitude increased progressively through the action potential plateau and was maximal near −30 mV (−30.2 ± 3.4 mV; n = 5; cf. Hancox et al. 1998a). Treatment with 5 mm 4-AP attenuated current throughout the action potential command, inhibiting maximal I_HERG by 58.5 ± 4.1 % (n = 5). Also, the voltage at which I_HERG peaked was shifted by −10.6 ± 4.1 mV, consistent with the voltage shift in current activation described in Fig. 1. These data suggest that 4-AP might increase the action potential duration in native cardiac tissue through blockade of native I_Kr, independent of any inhibitory action on I_TO,1. Therefore, we measured action potentials from guinea-pig ventricular myocytes, which lack I_TO,1 (Coraboeuf et al. 1998). 4-AP exerted little effect immediately after the action potential peak, but the times to reach both 50 % and 90 % repolarization (APD₅₀ and APD₉₀, respectively) were significantly prolonged by 4-AP (Fig. 4B; in control: mean APD₅₀ = 162.2 ± 12.2 ms and APD₉₀ = 184.3 ± 12.2 ms; in 4-AP: APD₅₀ = 210.9 ± 14.6 ms and APD₉₀ = 243.1 ± 14.8 ms, n = 8; P < 0.01 and P < 0.001, respectively). In complementary voltage-clamp experiments, resurgent I_K tails (sensitive to the I_Kr blocker E-4031) from guinea-pig myocytes were also inhibited by 4-AP (Fig. 4C).

Figure 4. Effect of 4-AP under action potential conditions.

A, representative records of I_HERG, after P/4 leak subtraction, (upper traces) activated by ventricular action potential command (lower trace), in control and in the presence of 5 mm 4-AP. B, effect of 5 mm 4-AP on action potentials from guinea-pig ventricular myocytes. C, effect of 4-AP on resurgent I_K tails from a guinea-pig ventricular myocyte (upper traces; lower trace shows a section of the voltage protocol. A mean inhibition of 71.1 ± 4.3 % was observed (n = 6).

DISCUSSION

The dependence of HERG blockade by 4-AP upon channel gating observed in this study differs from the mechanism underlying its blockade of cardiac I_TO,1, which depends considerably upon the agent binding to the closed channel (Castle & Slawsky, 1993; Wang et al. 1995). I_TO,1 from rat, rabbit and human have been reported to be blocked by 4-AP with IC₅₀ values of 0.14, 0.33 and 0.67 mm, respectively, whilst Kv 4.3 current exhibits an IC₅₀ of 1.54 mm (Xu et al. 1998; Faivre et al. 1999; Lei et al. 2000). Therefore, whilst 4-AP is clearly a low affinity HERG blocker, its IC₅₀ for I_HERG inhibition falls within a 4-AP concentration range used to block I_TO,1 selectively in previous studies (typically 0.5–10 mm; Wang et al. 1991; Gillis et al. 1998; Elizalde et al. 1999; Lei et al. 2000; Kocic et al. 2002). The fact that some level of I_Kr blockade may also have occurred in such experiments is supported by the present data from guinea-pig myocytes, which lack I_TO,1, and by previous data regarding reduction of I_K tails from rabbit nodal cells (Mitcheson & Hancox, 1999). On the basis of our findings, we conclude that millimolar concentrations of 4-AP can block I_HERG/I_Kr and, thereby, prolong cardiac action potential repolarization independently of any action on I_TO,1. In light of these observations, it may be necessary to consider carefully the interpretation of experiments in which millimolar 4-AP has been used as a pharmacological tool to study cardiac I_TO,1 in species that also exhibit I_Kr. Of course, in vitro data must only be extrapolated to the in vivo situation with caution and there is evidence that 4-AP infusion regimens used to study the role of I_TO,1 in the dog (del Balzo & Rosen, 1992) may actually attain submillimolar plasma levels (Nattel et al. 2000). Our data are concordant with the notion that micromolar 4-AP may justifiably be used to study I_Kur in cells that exhibit I_Kr, as comparatively little I_Kr blockade would be expected at the low concentration levels (e.g. Yue et al. 1996) required to inhibit I_Kur.