Eleclazine Suppresses Ventricular Fibrillation in Failing Rabbit Hearts with Ischemia-Reperfusion Injury Undergoing Therapeutic Hypothermia

Cardiac ischemia-reperfusion (IR) increases intracellular Na⁺ concentration via the Na⁺/H⁺ exchanger, which results in intracellular Ca²⁺ (Ca_i) overload, leading to myocardial injury [1]. In addition, IR activates Ca²⁺/calmodulin-dependent protein kinase II to enhance late sodium current (I_Na,L) to increase action potential duration (APD) dispersion (APD_dispersion) and susceptibility to ventricular tachyarrhythmia [2]. Eleclazine, a novel Na⁺-channel inhibitor with high selectivity for I_Na,L, was reported experimentally to reduce cardiac I_Na,L with minimal effects on other cardiac ion channels, including peak sodium current (I_Na), L-type Ca²⁺ channel and rapid delayed rectifier current, to exert potent antiarrhythmic effects in rabbit ventricular, canine, and pig atrial myocytes [3, 4]. While selective I_Na,L inhibition may achieve effective antiarrhythmic activity and circumvent proarrhythmic influences, a number of early clinical trials investigating the safety and antiarrhythmic effect of eleclazine in patients with implantable cardioverter defibrillators (NCT02104583), in the setting of hypertrophic cardiomyopathy (Liberty-HCM; NCT02291237), and in patients with gain-of-function mutations in the Na⁺-channel leading to long-QT syndrome type-3 (NCT02300558), have been discontinued due to limited efficacy or safety concerns. The main issue with most of the selective I_Na,L inhibitors is they function in a voltage-dependent manner and tend to block peak I_Na more at the physiologically relevant membrane voltage (V_m) range, thus losing their selectivity [5]. How specific was the inhibition of I_Na,L by eleclazine under the conditions studied? [6] The therapeutic value of eleclazine remains uncertain and requires further investigation in a disease-specific context.

Therapeutic hypothermia (TH) ameliorates oxidative injury and interrupts the early stages of the apoptotic pathway during acute IR injury [7, 8]. However, perturbations in electrophysiological properties during TH might predispose hearts with acute IR injury to reentrant arrhythmias. We previously reported that TH increased APD_dispersion, aggravated conduction slowing, and prolonged Ca_i decay to facilitate spatially discordant alternans (SDA) and ventricular fibrillation (VF) induction in failing rabbit hearts with acute IR injury [9]. It is possible that eleclazine suppresses TH-enhanced VF inducibility by ameliorating Ca_i overload to reduce ventricular repolarization alternans [10]. In this study, we investigated the effects of eleclazine administration before and after TH on cardiac electrophysiological parameters and vulnerability to VF by performing electrophysiological examinations and optical mapping studies. We hypothesize that eleclazine attenuates arrhythmogenic alternans and reduces VF inducibility in isolated IR-injured hearts undergoing TH.

The research protocol was approved by the Institutional Animal Care and Use Committee of Chang Gung Memorial Hospital, Taiwan (Approval No. 2021062201) and conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. New Zealand white rabbits (2.8–3.9 kg) were randomly divided into the following three groups: Group 1: heart failure (HF) with acute IR injury treated with eleclazine, followed by TH (n = 9); Group 2: HF with acute IR injury treated with TH, followed by eleclazine (n = 8); Group 3: control with acute IR injury treated with eleclazine, followed by TH (n = 8).

Rapid right ventricular pacing was used to induce HF as described previously [11]. Briefly, the rabbits were premedicated with intramuscular injections of zoletil (15 mg/kg) and xylazine (5 mg/kg). The surgical procedure was performed under general anesthesia with isoflurane (2%). An epicardial pacing lead was fixed on the right ventricle epicardium through a right lateral thoracotomy. The lead was then connected to a modified pacemaker (Adapta, Medtronic, Minneapolis, MN, USA). After 1-week recovery period, the heart was paced at a fixed rate of 320 bpm for 4 weeks to induce HF. Left ventricular (LV) function was assessed using echocardiography before and after 4 weeks of pacing.

Rabbits were premedicated with intramuscular injections of zoletil (15 mg/kg) and xylazine (5 mg/kg), intubated, and anesthetized with isoflurane (2%). After they were completely anesthetized and unresponsive to physical stimuli, the chests were opened through left thoracotomy. An obtuse marginal branch of the left circumflex artery was ligated midway between the atrioventricular groove and the cardiac apex for 60 min, followed by reperfusion. If spontaneous VF was induced, external defibrillation of 10–25 J was delivered.

After a 30-min reperfusion period, the hearts were excised and Langendorff-perfused with 37°C Tyrode’s solution (composition in mmol/L: NaCl 125, KCl 4.5, MgCl₂ 0.25, NaHCO₃ 24, NaH₂PO₄ 1.8, CaCl₂ 1.8, glucose 5.5, and albumin 50 mg/L in deionized water) and equilibrated with 95% O₂ and 5% CO₂ to maintain a pH of 7.4. Rhod-2AM (Ca_i indicator, 5 μm, Molecular Probes, OR, USA; in 20% Pluronic F-127 dissolved in dimethyl sulfoxide) and RH237 (V_m indicator, 1 μm, Molecular Probes; dissolved in dimethyl sulfoxide) were administered. The coronary perfusion pressure was regulated and maintained at 80–90 cm H₂O. The hearts were illuminated using a solid-state, frequency-doubled laser light source (Millennia, Spectra-Physics Inc., Newport Corporation, Irvine, CA, USA) with a wavelength of 532 nm. Epifluorescence was acquired and filtered (715 mm for V_m and 580 nm for Ca_i) with two MiCAM Ultima cameras (BrainVision, Tokyo, Japan) at 2 ms/frame temporal resolution and 100 × 100 pixels with spatial resolution of 0.30 × 0.30 mm² per pixel. Motion artifacts were suppressed with blebbistatin (10 μmol/L; Tocris Bioscience, Minneapolis, MN, USA).

A bipolar catheter was inserted into the right ventricle for pacing at twice the threshold. The ventricular effective refractory period (ERP) was measured by giving a premature stimulus after 8 beats at a pacing cycle length (PCL) = 400 ms. APD and Ca_i alternans were induced by a dynamic pacing protocol, as previously reported [11]. VF inducibility was defined as the ability to provoke sustained VF (>30 s) by a dynamic pacing protocol or an extrastimulus pacing protocol (up to S₅). Quantification of VF severity was based on the VF duration, which was classified as score 0: <10 s, 1: 10–30 s, and 2: ≥30 s, with larger values indicating greater severity. Defibrillation using epicardial patch electrodes was performed if VF persisted >2 min. Eleclazine (1 μm) [3] was then infused for 20 min, and electrophysiological studies were repeated.

Two thermostatic systems were connected in parallel to the Langendorff system. Induction of TH was performed by switching the thermostatic system and replacing the superfusate to 33°C. During cooling, LV temperature was monitored continuously with a thermometer implanted into the LV chamber via a left atriotomy. Electrophysiological studies were repeated after hypothermia for 30 min.

APD was measured at the level of 80% of repolarization. APD_dispersion was defined as the difference between maximal (APD_max) and minimal APD (APD_min) from the entire mapped areas at PCL = 300 ms [12]. We used monoexponential fittings to compute the time constant tau (τ) value of the decay portion of the Ca_i transient between 70% of the transient peak and the diastolic baseline. The thresholds of spatially concordant alternans (SCA) of APD and Ca_i were defined as the longest PCL required to produce a 10-ms difference in APD and a 10% difference in Ca_i amplitude between consecutive beats, respectively. The phase was considered positive for a short-long APD and a small-large Ca_i amplitude sequence (color coded in red) and negative for a long-short APD and a large-small Ca_i amplitude sequence (color coded in green). SDA was evidenced by the presence of both red and green regions separated by a nodal line. The SDA threshold was defined as the longest PCL required to reach the alternans threshold on both sides of a nodal line. To estimate conduction velocity (CV), we measured the distance and conduction time between the earliest activation point and two epicardial points in the non-IR (CV_non-IR) and IR (CV_IR) zones [11].

Continuous variables are expressed as mean ± standard deviation and categorical variables are represented by numbers and percentages. Repeated-measure analysis of variance with post hoc LSD analysis was performed to analyze differences in ERP, APD, CV, and Ca_i decay. Student’s t test was performed to compare CV and Ca_i decay between the non-IR and IR zones. Categorical variables were evaluated using Fisher’s exact test. Statistical analyses were performed using IBM SPSS V25.0 (Armonk, NY, USA). Differences were considered significant at p < 0.05.

After a 4-week pacing period, the LV ejection fraction decreased from 73 ± 3% to 41 ± 3% (p < 0.001) in Group 1 and from 73 ± 4% to 43 ± 5% (p < 0.001) in Group 2. The baseline LV ejection fraction was 71 ± 2% in Group 3.

The electrophysiological effects of eleclazine and TH are summarized in Table 1. Eleclazine had no significant effects on ERP in failing hearts at normothermia (p = 0.35, Group 1) and TH (p = 0.67, Group 2) and also in control hearts (p = 0.09, Group 3). TH prolonged ERP in failing hearts with eleclazine (p < 0.001, Group 1) and without eleclazine (p < 0.001, Group 3) and also in control hearts with eleclazine (p < 0.001, Group 3).

As summarized in Figure 1 and Table 1, eleclazine had no significant effects on APD_max but prolonged APD_min and thus decreased APD_dispersion; TH prolonged APD_max and APD_min and increased APD_dispersion. In failing hearts, eleclazine decreased APD_dispersion at normothermia (p = 0.002, Fig. 1a) and TH (p = 0.046, Fig. 1b). TH prolonged APD_max and APD_min (p < 0.001, Fig. 1a, b) and increased APD_dispersion with eleclazine (p = 0.002, Fig. 1a) and without eleclazine (p < 0.001, Fig. 1b). In control hearts, eleclazine also decreased APD_dispersion at normothermia (p < 0.001, Fig. 1c) and subsequent TH increased APD_dispersion (p < 0.001, Fig. 1c). Representative examples of APD maps (right subpanels) show that eleclazine prolonged APD_min and thus reduced APD_dispersion, and TH prolonged APD_max and APD_min and increased APD_dispersion in all three groups.

As summarized in Table 1 and Figure 2, Ca_i decay time was significantly longer in the IR zone than in the non-IR zone at all stages. In failing hearts, eleclazine accelerated Ca_i decay during normothermia, but decelerated Ca_i decay after TH; and TH prolonged Ca_i decay time with and without eleclazine (Fig. 2a, b, respectively). In control hearts, Ca_i decay time was shortened by eleclazine, and prolonged by subsequent TH (Fig. 2c). As shown in the representative examples of Ca_i tracings (lower subpanels), eleclazine at normothermia shortened Ca_i decay time in the failing and control hearts, and subsequent TH significantly increased τ values (Fig. 2a, c). In Figure 2b, TH prolonged Ca_i decay time, but subsequent eleclazine administration further increased τ values from 89.5 ms to 128.3 ms in the non-IR zone and from 97.4 ms to 133.6 ms in the IR zone. Even if the effect of eleclazine on Ca_i decay was different under normothermia and TH, the combined effects of eleclazine and TH on Ca_i decay were not significantly different between Group 1 and Group 2: the τ value increased 2.02 ± 0.27 times the baseline value after “eleclazine + TH” in Group 1 and 2.22 ± 0.32 times the baseline value after “TH + eleclazine” in Group 2 (p = 0.261).

Eleclazine slowed CV in a rate-dependent manner, especially at TH. In failing hearts, eleclazine did not change CV at PCL = 300 ms but slowed CV at PCL = 120 ms at normothermia. As shown in Figure 3a (right subpanel), eleclazine slowed CV_non-IR from 78 cm/s to 58 cm/s and CV_IR from 54 cm/s to 51 cm/s at PCL = 120 ms. At TH, eleclazine slowed CV_non-IR and CV_IR even at PCL = 300 ms. As shown in Figure 3b (left subpanel), eleclazine administration after TH further slowed CV_non-IR (from 56 cm/s to 21 cm/s) and CV_IR (from 34 cm/s to 15 cm/s) at PCL = 300 ms. TH slowed CV, especially in the presence of eleclazine. In the absence of eleclazine, TH slowed CV_non-IR from 84 cm/s to 56 cm/s and CV_IR from 60 cm/s to 34 cm/s (left subpanels, Fig. 3b); in the presence of eleclazine, TH slowed CV to a greater degree: CV_non-IR from 94 cm/s to 36 cm/s and CV_IR from 60 cm/s to 25 cm/s at PCL = 300 ms (left subpanels, Fig. 3a). In control hearts, eleclazine also did not change CV at PCL = 300 ms but slowed CV at PCL = 120 ms at normothermia. The representative isochronal maps reveal that eleclazine slowed CV_non-IR from 99 cm/s to 74 cm/s and CV_IR from 70 cm/s to 48 cm/s at PCL = 120 ms (right subpanels Fig. 3c).

To demonstrate a synergistic effect on CV, we calculated the extent to which TH altered CV in the presence or absence of eleclazine. In Group 1, TH slowed CV_non-IR to 0.39 ± 0.04 times and CV_IR to 0.36 ± 0.07 times baseline values with eleclazine; in Group 2, TH slowed CV_non-IR to 0.59 ± 0.12 times and CV_IR to 0.55 ± 0.11 times baseline values without eleclazine in failing hearts (p < 0.001 for both comparisons). In Group 3, TH also slowed CV_non-IR to 0.31 ± 0.16 times and CV_IR to 0.33 ± 0.08 times baseline values with eleclazine in control hearts. These results imply that TH and eleclazine acted synergistically to slow conduction.

The effects of eleclazine and TH on SCA and SDA in acute IR-injured hearts are summarized in Figure 4a and Table 1. SCA was provoked in all hearts. In failing hearts, eleclazine elevated the threshold of SCA at normothermia (p = 0.016, Group 1), but lowered the threshold of SCA at TH (p = 0.036, Group 2). In control hearts, eleclazine elevated the threshold of SCA at normothermia (p = 0.016, Group 3). TH lowered the threshold of SCA in failing hearts with and without eleclazine (p < 0.001 for both comparisons) and in control hearts with eleclazine (p < 0.001, Group 3).

Eleclazine suppressed SDA induction in failing hearts. At normothermia, SDA was induced in 9 of 9 hearts at baseline and 4 of 9 hearts with eleclazine (p = 0.029, Group 1). At TH, eleclazine did not suppress SDA induction but elevated the SDA threshold (p = 0.006, Group 2). TH lowered the SDA threshold with eleclazine (p = 0.004, Group 1) and without eleclazine (p < 0.001, Group 2). In control hearts, SDA was induced in 8 of 8 hearts at baseline and 3 of 8 hearts with eleclazine at normothermia (p = 0.026), and the SDA threshold was also lowered by TH (p < 0.001, Group 3). Figure 4b–d show representative examples of SDA induction under the effects of eleclazine and TH. In Group 1 (panel B), SDA was induced at PCL = 100 ms at baseline, non-inducible after eleclazine administration, and induced at PCL = 240 ms after TH; in Group 2 (panel C), SDA was induced at PCL = 100 ms at baseline, at PCL = 280 ms at TH, and not induced at PCL = 260 ms, but PCL = 260 ms induced VF after eleclazine administration; in Group 3 (panel D), SDA was induced at PCL = 100 ms at baseline, non-inducible after eleclazine administration, and not induced at PCL = 240 ms, but PCL = 240 ms induced VF after TH.

Figure 5a summarizes the effects of eleclazine and TH on VF inducibility in acute IR-injured hearts. Eleclazine could not suppress VF inducibility at normothermia in failing (Fig. 5a(A)) and control hearts (Fig. 5a(C)). TH alone had no significant effect on VF inducibility (Fig. 5a(B)), but eleclazine combined with TH significantly reduced VF inducibility by extrastimulus pacing in failing hearts, no matter which one was administered first. As shown in Figure 5a(A, B), sustained VF was induced by extrastimulus pacing in 5, 6, and 0 of 9 hearts (p = 0.009) at baseline, eleclazine, and “eleclazine + TH” in Group 1, and in 5, 4, and 0 of 8 hearts (p = 0.027) at baseline, TH, and “TH + eleclazine” in Group 2. In control hearts, eleclazine combined with TH had a lower VF inducibility by dynamic pacing than eleclazine alone, but had no significant difference compared to baseline. As shown in Figure 5a(C), sustained VF was induced in 5, 5, and 3 of 8 hearts by extrastimulus pacing (p = 0.670) and in 3, 6 and 1 of 8 hearts by dynamic pacing (p = 0.056) at baseline, eleclazine, and “eleclazine + TH,” respectively.

Figure 5b summarizes the effects of eleclazine and TH on VF severity. In failing hearts, eleclazine had no significant effects on VF severity at normothermia (Fig. 5b(A)), but eleclazine administration after TH counteracted TH-enhanced VF persistence by dynamic pacing and significantly reduced VF severity by extrastimulus pacing (Fig. 5b(B)). TH alone increased VF severity by dynamic pacing (p = 0.026, Fig. 5b(B)). TH after eleclazine administration significantly reduced VF severity by extrastimulus pacing (p = 0.029) but increased VF severity by dynamic pacing (p = 0.001, Fig. 5b(A)). In control hearts, eleclazine increased VF severity by dynamic pacing at normothermia (p = 0.026), which was counteracted by TH (p = 0.013, Fig. 5b(C)).

Figure 6 shows a representative example of eleclazine-enhanced arrhythmogenicity by dynamic pacing in a control IR-injured heart. At baseline, SDA was induced at PCL = 120 ms, but VF was not induced even at PCL = 100 ms. Eleclazine elevated the SDA threshold to 110 ms and decreased CV_IR from 46 cm/s to 33 cm/s at PCL = 110 ms (Fig. 6a). VF was induced at PCL = 110 ms (Fig. 6b). Figure 6c shows the V_m and Ca_i tracings at sites A, B, and C (labeled in Fig. 6d) during VF induction. As shown in Figure 6d, rapid pacing induced a functional conduction block at site B (beats 2 and 3), which facilitated reentry formation (beats 4 and 5) and was followed by multiple wavelets (beats 6 and 7). Note that no significant APD alternans and Ca_i amplitude alternans were induced at sites A, B, and C before the onset of the conduction block (Fig. 6c), suggesting that eleclazine-induced functional conduction block was due to an exacerbation of rate-dependent conduction disturbance at the peri-infarct zone rather than SDA induction.

Figure 7 shows a representative example of the combined TH and eleclazine effects on VF inducibility and severity by extrastimulus pacing in a failing IR-injured heart. Panel A shows the mapping field. Sustained VF was induced by extrastimulus pacing at baseline and TH, but VF was less inducible and non-sustained (25 s) after eleclazine administration (Fig. 7b). Figure 7c shows the V_m tracings at sites A and B (labeled in Fig. 7a) during VF induction by extrastimulus pacing (400/240/160 ms) in TH. Figure 7d shows the isochronal and phase maps. Extrastimulus pacing induced multiple lines of functional conduction block (red bars), which facilitated multiple wavelets formation (surrounding phase singularities, beats 6–8) and initiated VF. After eleclazine administration, repeat extrastimulus pacing (400/300/220/190 ms) could only induce non-sustained VF. Figure 7e, f show the V_m tracings and the isochronal and phase maps during VF induction, respectively. There were fewer phase singularities during VF initiation in the presence of eleclazine than with TH alone. Beats 4–6 showed a reentrant wavefront surrounding a phase singularity (black arrow) at site B, whereas V_m tracing showed double potentials during reentry. Note that eleclazine reduced CV_IR from 52 cm/s to 25 cm/s even at PCL = 400 ms, implying that a very slow CV prevented multiple wavebreaks to simplify wavefront propagation.

The main findings in this study are as follows: In failing hearts with IR injury, (1) eleclazine exerted antiarrhythmic effects by reducing APD_dispersion and accelerating Ca_i uptake to suppress induction of arrhythmogenic alternans at normothermia. However, eleclazine also exacerbated rate-dependent conduction slowing, resulting in a neutral effect on VF inducibility and severity. (2) TH increased VF severity by increasing APD_dispersion, impairing Ca_i uptake, exacerbating conduction disturbance, and enhancing arrhythmogenic alternans induction. (3) Eleclazine administration after TH ameliorated TH-induced APD_dispersion and further slowed CV to reduce VF inducibility and severity by extrastimulus pacing. (4) TH following eleclazine administration also further slowed CV to reduce VF inducibility and severity by extrastimulus pacing, although TH increased APD_dispersion and impaired Ca_i uptake in the presense of eleclazine. Thus, combined treatment with eleclazine and TH slowed CV synergistically to reduce VF inducibility and severity by extrastimulus pacing, regardless of which one was administered first. (5) In control hearts with IR injury, eleclazine had no significant effects on VF inducibility and severity by extrastimulus pacing, but increased VF severity by dynamic pacing, which was counteracted by subsequent TH.

I_Na,L is normally small, but its amplitude is greater in certain acquired or heritable conditions, including HF and IR [13, 14]. Inhibition of I_Na,L can reduce intracellular Na⁺ overload to enhance Ca²⁺ exclusion from the cytosol via the Na⁺-Ca²⁺ exchanger, thereby reducing Ca_i overload [15]. This improves myocardial relaxation and reduces LV diastolic stiffness, in turn enhancing myocardial contractility and perfusion [16], which may reduce APD_dispersion caused by IR injury. Improvement of Ca_i cycling prevents beat-to-beat fluctuation of sarcoplasmic reticulum (SR) Ca²⁺ content to reduce susceptibility to arrhythmogenic alternans [17]. Therefore, I_Na,L inhibition would be an effective antiarrhythmic treatment for failing hearts with IR injury. Eleclazine is a recently described I_Na blocker demonstrated to have a high affinity for I_Na,L, with IC₅₀ values of 0.62 and 51 µm for I_Na,L and peak I_Na blockade, respectively [18]. Zablocki et al. [18] showed that eleclazine suppressed spontaneous ventricular tachyarrhythmia from 66% (vehicle) to 0% (eleclazine 2.5 μm) in anesthetized rabbits with acute myocardial ischemia. However, our study revealed that eleclazine (1 μm) exerted a neutral effect on VF inducibility in Langendorff-perfused rabbit hearts with acute IR injury. The different effects of eleclazine on VF suppression between the two studies may be related to the animal models and study protocols. In that study, the infarct size was very large (ligation at 2–3 mm from the origin of left circumflex artery), resulting in a very high incidence of spontaneous ventricular tachyarrhythmia (66%). To perform optical mapping studies, we did not generate such a large infarct model, and only 2 of 25 (8%) rabbits had spontaneous VF during IR creation (1 during ischemia, 1 during early reperfusion). Furthermore, our model using isolated Langendorff-perfused hearts post 60-min ischemia and 30-min reperfusion was different from that study using in vivo hearts with 30-min ischemia. Our study showed that eleclazine exhibited its antiarrhythmic effects by reducing APD_dispersion, shortening Ca_i decay, and suppressing SDA induction. However, eleclazine also aggravated rate-dependent conduction slowing to facilitate VF induction. Potet et al. [19] reported that eleclazine exerts strong effects on slow inactivation and recovery from inactivation at high activation rates, resulting in substantial use-dependent block of peak I_Na. It is conceivable that eleclazine would affect peak I_Na, particularly in the case of partially depolarized membrane observed in diseased hearts or during tachycardia [20]. The concentration used in this study was sufficient for use-dependent block of peak I_Na (IC₅₀ = 0.6 μm at 10 Hz). As shown in Figure 7, eleclazine-induced rate-dependent conduction block occurred at the border of the IR zone, whereas SDA was not induced by rapid pacing, implying that the underlying mechanism of eleclazine-induced functional conduction block was due to rate-dependent peak I_Na inhibition at sites with anatomic heterogeneities rather than due to dynamically induced discordant repolarization alternans.

There have been no studies investigating the electrophysiological effect of eleclazine under hypothermic conditions. Our study revealed that eleclazine decreased APD_dispersion, further slowed Ca_i decay and CV, and lowered the threshold of SCA induction, but also elevated the threshold of SDA induction in failing hearts with acute IR injury undergoing TH. The VF inducibility studies revealed that eleclazine counteracted TH-enhanced VF inducibility. Compared with TH alone, eleclazine administration at TH resulted in relatively smooth propagation wavefronts with less wavebreaks during VF. One of the possible antifibrillatory mechanisms of eleclazine at TH could be a synergistic effect on CV slowing, which allowed for more time for the downstream sites to repolarize, making them no longer refractory during impulse propagation and thereby preventing conduction block for VF initiation. Additionally, further depressing conduction might convert unidirectional block to bidirectional block, thereby interrupting reentry for VF maintenance. Myocardial cooling was associated with slowed conduction, which is most likely explained by a decrease in I_Na availability via its temperature-dependent slowing of activation/inactivation kinetics [21]. It is possible that peak I_Na is more susceptible to I_Na inhibitors at TH, accounting for the synergistic effect of eleclazine and TH on slowing conduction.

An augmentation of I_Na,L is expected to increase Ca²⁺ loading of SR and thereby spontaneous Ca²⁺ release and susceptibility to develop Ca_i and APD alternans [22]. Inhibition of I_Na,L would antagonize these changes in Ca_i overload and rate-dependent calcium cycling. The reduction of intracellular Na⁺ by I_Na,L inhibition also leads to increased forward activity of the Na⁺/Ca²⁺-exchanger [23], thus increasing the rate of Ca²⁺ efflux through the sarcolemma. Consistent with this, our data showed that eleclazine fastened Ca_i decay and suppressed arrhythmogenic alternans at normothermia. Hypothermia slowed SR Ca²⁺ release and uptake through ryanodine receptors and SR Ca²⁺-ATPase dysfunction [24]. Unexpectedly, eleclazine further prolonged Ca_i decay and decreased the threshold of SCA induction at TH. Note that the combined effects of eleclazine and TH on Ca_i decay were not significantly different, regardless of which one was administered first. The mechanism by which eleclazine enhanced the effect of TH on slowing Ca_i decay is unclear and warrants further investigation.

Given that electrophysiological and optical mapping studies were performed in isolated Langendorff-perfused rabbit hearts, electrophysiological responses to TH and eleclazine may be different when the hearts are studied in vivo.

In acute IR-injured hearts, eleclazine had neutral effects on VF inducibility in failing and control hearts and on VF severity in failing hearts, but increased VF severity in control hearts at normothermia. At TH, eleclazine ameliorated TH-induced APD_dispersion and further slowed CV to reduce VF inducibility and severity by extrastimulus pacing and counteract TH-increased VF severity by dynamic pacing in failing hearts. TH increased VF severity, but TH in the presence of eleclazine also reduced VF inducibility and severity by extrastimulus pacing in failing hearts and counteracted eleclazine-increased VF severity by dynamic pacing in control hearts.

We thank Laboratory Animal Center, Chang Gung Memorial Hospital, Linkou, Taiwan, for the animal husbandry and care.

This study protocol was approved by the Institutional Animal Care and Use Committee of Chang Gung Memorial Hospital (Approval No. 2021062201) in line with the current NIH guidelines for the care and use of laboratory animals.

The authors have no conflicts of interest to declare.

This study was supported by Chang Gung Memorial Hospital Medical Research Program, Taiwan (CMRPG3L1141), to C.-C. Chou.

Chung-Chuan Chou designed and performed the experiments and revised the manuscript. Hui-Ling Lee analyzed the data and wrote the first draft of the manuscript. Po-Cheng Chang, Hong-Ta Wo, and Hao-Tien Liu analyzed the data. Ming-Shien Wen and Po-Cheng Chang contributed to the concept of the study. All authors read and approved the final manuscript.

The data that support the findings of this study are included in the article. The data of optical mapping are not publicly available due to large size of dataset (more than 400 GB) but are available from the corresponding author (C.-C.C.) upon reasonable request.