Apamin‐sensitive potassium current and the Mechanisms of ventricular arrhythmia: Perspectives in heart failure treatment

In patients with advanced heart failure (HF), recurrent spontaneous ventricular fibrillation (SVF)/electrical storm (ES) is a frequent complication requiring multiple defibrillation shocks within a short space of time. The mechanism underlying initiation of SVF/ES, spontaneously or immediately after defibrillation shocks, remains unclear. In 2009, Ogawa et al. developed a pacing-induced HF model, in which SVF occurred frequently in failing rabbit ventricles [1]. In that model, acute but reversible post-shock action potential duration (APD) shortening could induce recurrent SVF (Fig. 1, panel A) [1]. The induction of SVF was associated with persistent intracellular calcium (Cai) elevation during late phase 3 and/or phase 4 of the action potential (AP), indicating that APD shortening, in conjunction with persistent post-shock Cai elevation, is a novel mechanism of post-shock SVF/ES [1]. However, the mechanisms underlying acute APD shortening after fibrillation/defibrillation episodes in HF ventricles remains to be determined. Pretreatment with apamin prevents late-phase 3 early after-depolarizations (EAD) and recurrent ventricular fibrillation (VF). Black and red lines indicate optical tracings for Vm and Cai, respectively (obtained from the endocardial surface in a failing ventricle). (A-a) Before apamin administration, a defibrillation shock (arrow) was followed by 2 post-shock beats (1 and 2). After a short pause, there was a large Cai transient (red dot), a short action potential duration (APD) (beat 3), and a first beat (beat 4) of spontaneous VF (SVF) arising from late phase 3. (A-b) Snapshots of Cai and Vm ratio maps taken from 10ms before to 20ms after the onset of beat 4. Note: Cai remained elevated throughout the mapped field, whereas Vm had already repolarized. Beat 4 arose during persistently high Cai and initiated the SVF. (B) After apamin (1μmol/L) infusion, the post-shock beats (5 and 6) had longer APDs than beats 1 to 4 in (A). The APD and Cai transient durations were approximately the same, and SVF episodes were completely prevented. PM indicates papillary muscle; S, interventricular septum. From Chua et al. [2] with permission. HF is associated with the down-regulation of multiple potassium (K) currents (i.e., Ito, IKs, IKr, IK1, and IKATP). The only K channel that is upregulated in HF is the small conductance Ca-activated K (SK) channel. The SK channel was originally found in most neurons. The increase in Cai evoked by the AP allows the SK channel to become activated, and generates a long-lasting after-hyperpolarization (AHP). Physiologically, the AHP can inhibit repetitive firing, and prevent deleterious tetanic activity in the nervous system. Previous studies show that the SK current is abundantly present in cardiac atrial cells, but not in normal ventricular cells. Because ventricular fibrillation (VF) is associated with Cai accumulation, especially in HF ventricles, Chua et al. hypothesized that SK channels might exist in HF ventricles and contribute to the post-shock APD shortening observed in Ogawa's study [1], [2]. To test this hypothesis, they used apamin, a selective SK channel blocker that specifically inhibits apamin-sensitive K current (IKAS), to explore the roles of the SK channel in mediating ventricular arrhythmia of failing ventricles. Chua et al. found that apamin administration effectively prevented post-shock APD shortening, late phase 3 early after-depolarizations (EAD), and triggered activity and recurrent SVF in failing rabbit ventricles (see Fig. 1, panel B) [2]. Using a voltage-clamp technique, they reported that the IKAS current density was significantly larger in failing ventricular cardiomyocytes than in normal ones [2]. The IKAS sensitivity to Cai was increased in cardiomyocytes isolated from failing ventricles compared to those from normal ventricles [2]. Similar findings were reproduced in failing human ventricles by Chang et al. [3]. They also showed that the IKAS current density was lower in the mid-myocardial cells than in the epicardial and endocardial cells [3]. The subtype 2 of SK (SK2) protein expression was 3-fold higher in HF than in non-HF human ventricles [3]. These findings indicated for the first time that HF heterogeneously increases the sensitivity of IKAS to Cai, leading to the up-regulation of IKAS, post-shock APD shortening, late phase 3 EAD, triggered activity, and recurrent SVF in both animal models and failing human ventricles. In HF patients and animal models, the APD shortens more rapidly than in normal ventricles during rapid pacing, leading to an increased slope of the APD restitution (APDR) curve. A steep APDR curve promotes dynamic instability, wave breaks, and VF. Because rapid pacing causes Cai accumulation, IKAS activation in failing ventricles might lead to increased APD shortening during a rapid pacing rate, and further steepen the APDR curve. This hypothesis was confirmed by Hsieh et al., who showed that, in failing rabbit ventricles, apamin flattens the APDR curve at fast pacing rates [4]. Apamin also decreases wave breaks, reduces the dominant frequency, and shortens the duration of VF. On the contrary, apamin lengthens the APD, steepens the APDR curve, and abolishes repolarization heterogeneity during slow pacing rate (cycle length, CL 300–350 ms). This finding also suggests that IKAS activation is a major factor that underlies the repolarization heterogeneity and preservation of repolarization reserve in HF ventricles. In Hsieh's study, a secondary rise in Cai during the late AP plateau was observed in 9–19% of the epicardial area during slow heart rate in failing ventricles. The prolonged availability of Ca2+ during secondary Cai rise might activate IKAS to shorten the APD, thereby maintaining repolarization reserve and preventing ventricular arrhythmias in HF. Because slower heart rates (longer diastolic intervals) are associated with higher availability of L-type Ca2+ current (ICa,L) and longer Cai transient duration that might maximize IKAS activation, an experimental model with extremely slow heart rate (CL>500 ms) might better elucidate the role of IKAS in maintaining repolarization reserve. Chang et al. therefore performed cryoablation of the atrioventricular node in failing rabbit hearts to reduce the ventricular rate [5]. In that extremely slow heart rate (CL>500 ms) model, apamin induced EAD, premature ventricular beats (PVBs) and torsades de pointes (TdPs) in failing ventricles. Interestingly, the earliest activation site of the EAD and PVB always occurred at the area with long APD and large amplitude of the secondary rise of Cai. This study further confirmed the importance of IKAS in maintaining repolarization reserve and preventing TdP in HF ventricles. Similar to HF, myocardial infarction (MI) is followed by significant down-regulation of multiple K currents in the peri-infarct zone. These changes may contribute to after-depolarization and ventricular arrhythmia in MI ventricles. However, none of the previous studies reported MI in an increased IKAS. By delicate control of infarction size, Lee et al. studied IKAS in a rabbit model of chronic MI (5 weeks) with normal left ventricular function [6]. In that model, they found that apamin prolonged the APD in the peri-infarct zone by 9.8%, which is greater than that (2.8%) of normal ventricles. Post-pacing APD shortening is also prevented by apamin. A patch clump study showed that the sensitivity of IKAS to Cai was significantly increased in myocytes isolated from the peri-infarct zone. This study firstly proved that MI is associated with IKAS up-regulation, even without HF. From the serial studies of IKAS in diseased hearts, it was clear that up-regulation of IKAS during slow heart rate might increase the repolarization reserve and prevent after-depolarizations, whereas IKAS up-regulation at tachycardia (short CL) might shorten APD and steepen APDR, promoting ventricular arrhythmia. Therefore, similar to other K channel blockers, IKAS blockers can be both pro-arrhythmic and anti-arrhythmic, depending on the clinical situations associated with arrhythmogenesis. IKAS blockade may prevent transition from ventricular tachycardia (VT) to VF and prevent the spontaneous re-initiation of VF. On the other hand, if the arrhythmia is bradycardia dependent, such as that induced by EAD, IKAS blockers may promote triggered activity and ventricular arrhythmia. However, the optimal heart rate for preventing VT to VF transition while avoiding bradycardia-induced EAD in terms of IKAS blockade remains to be determined. Our knowledge of IKAS is derived from experimental animal models (i.e., HF, MI); whether the IKAS hypothesis could be applied to human anti-arrhythmic therapy warrants further investigation. Apamin is a neurotoxin that may induce significant neurological side effects when used in live animals and humans. We need to develop non-toxic IKAS blockers before we can test the effects of IKAS blockade on VF storm in human patients. Amiodarone is effective in the treatment of recurrent VT/VF and is commonly used as the first line therapy for ES. Recently, Turker et al. used a patch-clamp technique to study the effects of amiodarone on IKAS in human embryonic kidney 293 (HEK-293) cells transiently expressing human SK2 channel [7]. They found that amiodarone inhibited IKAS in a dose-dependent manner. The degree of IKAS inhibition by amiodarone is dependent on the Cai concentration. This study indicated that SK2 current inhibition might in part underlie the effects of amiodarone in preventing ES in failing ventricles. A better understanding of the effects of anti-arrhythmic drugs on IKAS may be important in the prevention of sudden death in patients with HF in the future. There is no conflict of interest in this study. We thank Dr. Peng-Sheng Chen for his critical comments.