Can inhibition of I_Kur promote atrial fibrillation?

Abstract

BACKGROUND

Block of ultra-rapid delayed rectified potassium current (I_Kur), present in atria, but not in ventricles, is thought to be a promising approach for atrial-specific therapy of atrial fibrillation (AF). However, it has been shown that I_Kur block may abbreviate atrial repolarization and that loss-of-function mutations in KCNA5, which encodes K_v 1.5 channels responsible for I_Kur, is associated with familial AF.

OBJECTIVE

Our objective in this study was to use low concentrations of 4-aminopyridine (4-AP, 10–50 µM), known to selectively block I_Kur, to assess the pro-and antiarrhythmic effects of I_Kur block in “healthy” and “remodeled” atria.

METHODS

Isolated canine coronary-perfused right atrial preparations were used. Acetylcholine or ischemia/reperfusion was utilized to acutely “remodel” the atria. Transmembrane action potentials and a pseudo-ECG were simultaneously recorded.

RESULTS

Normal (“healthy”) atria typically displayed action potentials (AP) with a prominent plateau, whereas remodeled atria displayed triangular-shaped APs (“remodeled”). In “healthy” atria, in which AF could not be induced with programmed stimulation, 4-AP abbreviated action potential measured at 90% repolarization (APD₉₀) and effective refractory period (ERP), permitting the induction of AF in 4/12 preparations (33%). In “remodeled” atria, 4-AP produced little (50 µM) to no (10–25 µM) prolongation of APD₉₀ or ERP and was either ineffective or poorly effective in terminating AF or preventing its induction.

CONCLUSIONS

Our findings suggest that block of I_Kur can provide the substrate for development of AF in “healthy” canine atria, presumably via abbreviation of APD and ERP.

INTRODUCTION

Ultra-rapid delayed rectified potassium current (I_Kur), carried by K_v 1.5 channels encoded by the KCNA5 gene, is present in atria, but not in the ventricles.^1–3 Block of I_Kur is thought to be a promising approach for atrial-specific therapy of atrial fibrillation (AF). However, a recent report has associated KCNA5 loss-of-function mutations with familial AF.⁴ It has been shown also that reduction of I_Kur abbreviates action potential duration (APD₉₀) in non-remodeled (“healthy”) atrial cells, but prolongs APD₉₀ in remodeled atrial cells (displaying a triangular-shaped action potential).^{5, 6} Knowing that abbreviation of atrial repolarization promotes AF, we hypothesized that I_Kur block could promote AF in “healthy” atria. We used low concentrations of 4-AP (10–50 µM), shown to be specific for I_Kur block,^{2, 3} to assess the pro-and antiarrhythmic effects of I_Kur block in “healthy” vs. “remodeled” canine isolated coronary-perfused atria.

METHODS

Experiments were performed using isolated arterially-perfused canine right atrial preparations. The methods used for isolation and perfusion of these preparations have been described in previous publications.^{5, 7} Briefly, right atria were dissected from hearts removed from anesthetized (sodium pentobarbital) adult mongrel dogs (20–25 kg). Unfolded atria with a rim of the right ventricle was cannulated and perfused through the ostium of the right coronary artery. Unperfused tissue was removed with a razor blade or scissors. The cut ventricular and atrial branches were ligated using silk thread. After these procedures (performed in cold cardioplegic solution, 4–8°C), the preparations were transferred to a temperature-controlled bath and arterially-perfused with Tyrode’s solution by use of a roller pump. The composition of the Tyrode’s solution was (in mM): NaCl 129, KCl 4, NaH₂PO₄ 0.9, NaHCO₃ 20, CaCl₂ 1.8, MgSO₄ 0.5, and D-glucose 5.5, buffered with 95% O₂ and 5% CO₂ (37±0.5 °C, pH=7.35)

Transmembrane action potential (AP) recordings were obtained using floating glass microelectrodes (2.7 M KCl, 10–25 MΩ DC resistance). A pseudo-electrocardiogram (ECG) was recorded using two electrodes consisting of Ag/AgCl half cells placed in the Tyrode’s solution bathing the preparation, 1.0 to 1.2 cm from opposite ends of the atrial preparations. Effective refractory period (ERP) was measured by delivering premature stimuli of twice diastolic threshold intensity after every 10^th basic beat at a pacing CL of 500 ms.

“Healthy” and “Remodeled” atria

Arterially-perfused atrial preparations not subjected to any interventions were referred to as “Healthy” atria. Acute “remodeling” of atrial coronary perfused preparations was achieved with two experimental approaches known to abbreviate atrial APD and promote AF. In the first, acetylcholine (ACh, 0.5 µM) was used to abbreviate APD and ERP and mimic a “remodeled” atrium.^{7, 8} The second approach involved a recently developed AF model in which isoproterenol (ISO, 0.1–0.2 µM) was added to the coronary perfusate after the atrial preparations are exposed to ischemia and reperfusion.^{9, 10} Ischemia and reperfusion were produced by stopping (for 30–40 min) and reinitiating coronary flow with continuous superfusion flow. On the 15–30th minutes after the start of reperfusion, ISO was added to the coronary perfusate, which reproducibly generated non-sustained AF in the ischemia/reperfusion-damaged atria.^{9, 10} Note that β-adrenergic stimulation only mildly promotes AF in “healthy” atria.^{9, 11}

Experimental Protocols

The equilibration period for the preparations was 30–40 min. The concentration of 4-aminopyridine (4-AP, 10–50 µM in “healthy” atria and 10–200 µM in “remodeled” atria) was increased in a step-wise manner (in ≥ 20 min after the start of each concentration). APs were recorded from the endocardial crista terminalis (CT) and pectinate muscle (PM). Arrhythmogenicity was tested using programmed electrical stimulation (PES) protocols at a CL of 500 ms (a single premature beat with a twice-threshold stimulus intensity) at three locations (two in PM and one in CT) under baseline conditions and in the presence of each concentration of 4-AP. In some experiments involving ACh, persistent AF was induced by a rapid pacing protocol (CL = 100–50 ms for 3–5 seconds).

Drugs

4-AP, ACh, and ISO (all SIGMA, MO) were dissolved in distilled water and prepared fresh as a stock of 1–100 mM before each experiment.

Statistics

Statistical analysis was performed using paired or unpaired t test and one way repeated measures or multiple comparison analysis of variance (ANOVA) followed by Bonferroni’s test, as appropriate. All data are expressed as mean ± SD.

RESULTS

In “healthy” coronary-perfused right atrial preparations, block of I_Kur with low concentrations of 4-AP significantly elevated the plateau level and prolonged the early repolarization phase (APD₂₀ from 5±3 to 42±19 ms, p<0.001; in CT; 25 µM; n=10), but abbreviated APD₉₀ and ERP (Fig. 1). The magnitude of phase 1 of the action potential was reduced from 30±6 mV in control to 22±5, 17±4, and 8±3 mV in the presence of 10, 25, 50 µM of 4-AP, respectively (all p<0.05 vs. control; CT, n=12 for each; CL = 500 ms; Fig. 1). Plateau voltage (dome) was significantly elevated from −4±4 mV in control to +3±5, +10±3, and +15±6 mV after 10, 25, and 50 µM of 4-AP, respectively (all p<0.05 vs. control; CT, n=12 for each; CL = 500 ms).

Figure 1.

Inhibition of I_Kur using low concentrations of 4-aminopyridne (4-AP) abbreviates APD₉₀ and ERP in “healthy” atrial preparations. Shown are superimposed action potentials recorded from the crista terminalis (CT) under control conditions (C) and after the addition of 4-AP (A). Graphs plot summary data of APD₉₀ (B) and ERP (C) recorded from CT and pectinate muscle (PM) sites as a function of 4-AP concentration. CL = 500 ms. n=7–12. * − p<0.05 vs. respective control. Note that atrial ERP corresponds to the level of APD₇₅.

Under baseline conditions, 2 out of 14 “healthy” atria developed short AF (<1 sec) in response to PES. These two preparations were removed from further study. In the presence of low concentrations of 4-AP (10–50 µM), AF could be induced in up to 33% of “healthy” atrial preparations, which did not develop AF in the absence of 4-AP (Fig. 2 and Table I). All episodes of AF were non-sustained (≤3 sec). Spontaneous arrhythmias were not observed.

Figure 2.

Non-sustained AF induced by a single premature beat (S₁–S₂ = 115 ms) in the presence of 25 µM 4-AP in a “healthy” atrial preparation.

Table I.

Incidence of PES-induced non-sustained AF in “healthy” atrial preparations in the presence of low concentration of 4-AP.

Control	4-AP 10 µM	4-AP 25 µM	4-AP 50 µM
0/12 (0%)*	1/12 (8%)	3/12 (25%)	4/12 (33%)

^*PES induced non-sustained arrhythmias (≤ 1 sec) in 2 out of 14 atrial preparations under control conditions. These two preparations were excluded from further study.

In contrast to “healthy” atria, block of I_Kur using 10 and 25 µM 4-AP produced little to no change in APD₉₀ or ERP in “remodeled” atria (Fig. 3). Higher concentrations of 4-AP (50 and 100 µM) prolonged APD₉₀ and ERP in the “remodeled” atrial preparations (Fig. 4).

Figure 3.

Effects of 4-AP on APD₉₀ and ERP in “acutely remodeled” atrial preparations. Shown are superimposed action potentials obtained from atria pre-treated with acetylcholine (ACh, A) and from atria exposed to ischemia and reperfusion + isoproterenol (Isch/Rep/Iso, B) before and after addition of 4-AP. C–D: Summary APD₉₀ and ERP data from pectinate muscle. n=7–12. CL = 500 ms. * − p<0.05 vs. respective control.

Figure 4.

Concentrations of 4-AP that selectively block I_Kur fail to prevent or terminate atrial fibrillation in “remodeled” atria. Shown are simultaneously recorded ECG and AP traces. A: 25 µM 4-AP fails to terminate persistent AF induced in the presence of acetylcholine (ACh, 0.5 µM). B: 25 µM 4-AP fails to prevent the induction of non-sustained AF by a single premature beat (S₁–S₂ = 85 ms, which is ERP) in ischemia/reperfusion+isoproterenol AF model (Isch/Rep/Iso). C: An increase of 4-AP concentration to 100 µM prolongs ERP and prevents the initiation of the arrhythmia by premature stimulation (S₁–S₂ = 110 ms, which is ERP).

4-AP at concentrations of 10–25 µM was not able to prevent or terminate AF and at a concentration of 50 µM exerted only mild antiarrhythmic actions in both experimental AF models used in our study (Table II and III). Higher concentrations of 4-AP (100–200 µM) were more effective in preventing AF initiation and termination of persistent AF, but still showed limited antiarrhythmic efficacy in our experimental AF models (Table II and III).

Table II.

Incidence of acetylcholine (ACh) -mediated AF and the ability of 4-AP to prevent the initiation of AF or terminate persistent AF in atrial preparations.

	Ach (0.5 µM)	ACh + 4-AP (10–25 µM)	ACh + 4-AP (50 µM)	ACh + 4-AP (100–200 µM)
Incidence of AF initiation	100% (10/10)	100% (4/4)	75% (3/4)	50% (3/6)
Termination of persistent AF	0% (0/10)	0% (0/4)	0% (0/5)	33% (2/6)

Experiments to prevent AF and terminate AF were performed in different preparations. Incidence of AF initiation: 4-AP was added to ACh-containing Tyrode’s solution and AF induction was attempted using PES. Termination of persistent AF: Persistent AF was induced in the exposure to Ach and 4-AP was added 5–7 min after the initiation of AF.

Table III.

Incidence of PES-induced non-sustained AF with and without 4-AP in ischemia/reperfusion + isoproterenol model of AF in atrial preparations.

Ischemia/Reperfusion + isoproterenol	+ 4-AP (10–25 µM)	+ 4-AP (50 µM)	+ 4-AP (100–200 µM)
100% (5/5)	100% (5/5)	80% (4/5)	40% (2/5)

DISCUSSION

The main findings of our study are that block of I_Kur with low concentrations of 4-AP abbreviates APD₉₀ and ERP and promotes AF in “healthy” canine coronary-perfused atria.

Ion channel specificity of 4-AP

Within the range of concentrations used in the present study to block I_Kur (10–50 µM), 4-AP is thought to exert little or no inhibition of I_to and not to affect other currents.^{2, 3} When measured in isolated atrial myocytes, 4-AP blocks I_Kur and I_to with an IC₅₀ of 5 and 471 µM in dogs (potency ratio = 94),² and 8 and 1000 µM in human (potency ratio = 125).³ Apparently, all I_Kur blockers inhibit I_to. The I_to/I_Kur blocking potency ratios for AVE0118, AVE1231, DPO-1, AZD7009, RSD1235, and NIP-141 are 3.1, 1.6, 8.0, 0.9, 2.3, and 3.1, respectively (data in most cases obtained from heterologous expression systems such as CHO or HEK cells).^12–17 Apart from I_to, most of these I_Kur blockers also inhibit I_K(ACh), and/or I_Kr, and/or I_Na with affinities comparable to that of I_Kur block. Thus, 4-AP at low concentrations is a more selective I_Kur blocker than the above-listed agents.

In mathematical models of canine or human “healthy” atrial action potential, 80–90% percent block of I_Kur results in about 20–25% reduction of phase 1 magnitude.^{6, 18, 19} The large reduction of the phase 1magnitude induced by 50 µM of 4-AP (by 71%; see Fig. 1) suggests that in the coronary-perfused canine atria 50 µM 4-AP may also block a substantial amount of I_to. A relatively selective I_Kur inhibition appears to be achieved with ≤25 µM of 4-AP (which causes a 43% reduction in phase 1magnitude). It is noteworthy that the pro-arrhythmic potency of 25 and 50 µM 4-AP was comparable (Table I).

Atrial-selective approaches for management of AF

A number of antiarrhythmic agents have been shown to be effective in terminating and/or preventing AF in the clinic. Most of these agents have as a primary action the ability to reduce I_Na (e.g., propafenone and flecainide) or I_Kr (e.g., dofetilide) or to block multiple ion channels (e.g., amiodarone). An important limitation of these antiarrhythmic agents is their potential ventricular pro-arrhythmic actions and/or organ toxicity at therapeutically effective doses.^{20, 21}

As a consequence, much of the focus in recent years has been on development of atrial-selective drugs that take advantage of the presence of I_Kur channels in atrial but not in ventricular cells.^1–3 A number of studies have demonstrated that agents capable of blocking I_Kur (AVE0118, AVE1231, S9947, S20951, ISQ1, DPO-1, RSD1235; AZD7009; NIP141, NIP-142) selectively prolongs atrial ERP both in electrically-remodeled and non-remodeled atria in vivo and in vitro.^{13, 22–24 17, 25–27}

It is well appreciated that the effect of I_Kur inhibition on APD_70–90 is dependent on the baseline AP morphology.⁶ I_Kur block has been reported to prolong APD_70–90 in atrial cells displaying a triangular AP morphology (typical for remodeled atria) and abbreviate APD_70–90 in atrial cells having a plateau AP morphology (typical for non-remodeled atria)⁶ (as observed in the current study as well). Such APD₉₀ abbreviation is explained by elevation of AP plateau voltage (see fig. 1), which leads to augmentation of I_Kr and I_Ks.^{5, 6} I_Kur block-induced ERP prolongation in electrically-remodeled atria is consistent with APD_70–90 prolongation. On the other hand, I_Kur block-induced ERP prolongation in non-remodeled atria is not readily explained by APD_70–90 abbreviation,^{5, 6} and may be attributable to concomitant block of sodium channel current. I_Na blockers are known for their ability to prolong ERP without prolonging APD. This effect of I_Na is due to the development of post-repolarization refractoriness (PRR), secondary to the depression of excitability.^{10, 28} Atrial-selective ERP prolongation attributable to sodium channel blockade, has been reported secondary to development of PRR preferentially in atria.^{10, 29}

Interestingly, atrial selective agents AZD7009 and RSD1235 (or vernakalant) have been shown to potently block I_Na, along with I_Kur, I_to, and I_Kr.^{16, 25, 26, 30} The atrial selectivity of these agents to prolong ERP is largely attributed to their ability to block I_Kur. However, AZD7009 also depresses sodium channel-dependent parameters (conduction velocity and diastolic threshold of excitation) in an atrial selective manner.²⁶ AZD7009 prolongs atrial ERP much more than APD₉₀, leading to capture failure; due to development of PRR.³⁰ These observations suggest that atrial-selective or predominant sodium channel block may contribute prominently to the atrial-selective ERP prolongation of AZD7009. ISQ-1 may also block I_Na, since this agent slows down conduction velocity in atria in vivo.³¹ Note that with the exception of AZD7009,²⁶ comparison of the effects of I_Kur blockers on I_Na or sodium channel-dependent parameters in atria and ventricles has not been conducted.

There is an apparent contradiction between our data and those of others^{6, 32} showing that 4-AP abbreviates atrial APD₉₀ in healthy atrial preparations in vitro with data reported by Nattel et al³³ demonstrating that 4-AP at a plasma concentration range of 25–50 µM significantly increases ERP in healthy canine atria in vivo. This discrepancy may be due to the fact that Nattel et al³³ recorded ERP at relatively rapid pacing rates (300–400 ms CLs), at which AP shape tends to become triangular. Under these conditions, 4-AP may prolong late repolarization and thus ERP. The in vivo vs. in vitro differences may be due to the effect of 4-AP on autonomic tone in vivo.³⁴

Inhibition of I_Kur promotes non-sustained AF in healthy atria

Most experimental models of AF, as well as clinical cases of AF, are associated with an abbreviation of atrial repolarization.^{7, 8, 35} It is therefore no surprise that abbreviation of APD₉₀ and ERP by 4-AP in “healthy” atria promoted AF in our study. The extent of APD and ERP abbreviation induced by 4-AP is much smaller that that induced by ACh or electrical remodeling. This may account for the relatively low incidence of AF and its brief duration in our experiments. Our findings relative to the ability of I_Kur inhibition to promote AF are consistent with a recent report demonstrating an association of KCNA5 loss-of-function mutations with familial AF.⁴

Can I_Kur block alone suppress AF?

At concentrations that terminate AF, I_Kur blockers relatively potently inhibit I_to and/or I_KACh (e.g., AVE0118 or NIP-142)^{22, 27} and/or I_Na (e.g., AZD7009 and vernakalant).^{16, 26, 36} Our experimental findings suggest that “pure” I_Kur block may be incapable of effectively suppressing AF. Note the IC₅₀ of 4-AP block of atrial I_to is one third of that of ventricular I_to.^{3, 33} If it is the case with the other I_Kur/I_to blockers, then I_to block may importantly contribute to the atrial-selectivity of I_Kur blockers.

A possible reason for the failure of “pure” I_Kur block to terminate AF in our experimental models may be the fact that I_Kur density is reduced at rapid activation rates,^{37, 38} so that at rates encountered during AF I_Kur may already be greatly reduced without much left to block.

It is commonly believed that anti-AF actions of potassium channel blockers are mediated through the prolongation of APD₉₀/ERP. A recent mathematical modeling study suggests that I_Kur and/or I_to blockade may terminate AF by prolonging APD at the level of the plateau, but not the terminal phase of the action potential, leading to random tip meander and wavebreak, resulting in rotor termination.³⁹

Study limitations

Our experiments were performed in isolated atria coronary-perfused with Tyrode’s solution. The presence of autonomic influences and other factors present in vivo may modulate the effect of I_Kur inhibition, resulting in outcomes different from those observed in the present study. I_Kur is known to be modulated by β-and α-adrenergic sympathetic activity.⁴⁰ I_Kur blockers, including 4-AP, are known to increase atrial contractility (i.e., intracellular calcium activity), apparently secondary to elevation of plateau voltages and augmentation of I_Ca and reverse mode of sodium-calcium exchange.⁴¹ It is not known whether these influences secondary to I_Kur block, or possibly 4-AP directly, contribute to the arrhythmogenesis observed. If due to I_Kur block, this effect should be relevant to all I_Kur blockers.

Our results obtained in “acutely remodeled” atrial preparations, lacking a number of principal electrical (I_Ca,L, I_to, I_K1, etc) and structural changes observed in chronically remodeled atria,³⁵ should be interpreted with due caution. In that there are important inter-atrial electrophysiological differences, the data obtained from the right atrium in the present study may not be directly applicable to the left atrium.

CONCLUSIONS

Our findings suggest that inhibition of I_Kur can create the substrate for development of AF in “healthy” canine atria, via an abbreviation of ERP. These observations advance our understanding of why KCNA5 loss-of-function mutations are associated with the development of AF.⁴

ABBREVIATIONS LIST

AF

atrial fibrillation

AP

transmembrane action potential

APD

action potential duration

ERP

effective refractory period

I_Kur

ultra-rapid delayed rectified potassium current

I_to

transient outward potassium current

PES

programmed electrical stimulation

PRR

post-repolarization refractoriness

RMP

resting membrane potential