Cardiac α4β2 nicotinic receptors as a therapeutic target for fatal ventricular arrhythmias

Abstract

Background and Aims

Fatal ventricular tachyarrhythmias (FVTs) are a major cause of sudden cardiac death globally. Despite their clinical importance, current antiarrhythmic therapies remain constrained by limited efficacy and proarrhythmic risks. Although endogenous cardiac cholinergic signalling contributes to electrophysiological regulation, the therapeutic potential of targeting ventricular α4β2 nicotinic acetylcholine receptors (nAChRs), a pivotal component of this system, for FVT treatment remains to be elucidated.

Methods

A pharmacological screen identified positive allosteric modulators (PAMs) of α4β2 nAChRs as possessing antiarrhythmic properties. Through structure-based design, salvage-1 was developed and its efficacy and safety profile were evaluated in rodent, porcine, and human ex vivo heart models of FVTs. The underlying mechanism was investigated using patch-clamp electrophysiology, high-resolution optical mapping, and molecular dynamics simulations.

Results

Pharmacological screening validated the potentiation of α4β2 nAChRs as a promising antiarrhythmic strategy. Structure-guided development yielded salvage-1, a PAM that selectively engages Phe312 and Phe316 on the α4 subunit to stabilize the receptor in an open-channel state. Across rodent, porcine, and human ex vivo heart models of FVTs, salvage-1 consistently prevented arrhythmogenesis and rapidly restored sinus rhythm. Mechanistically, salvage-1 enhanced acetylcholine-gated currents, selectively improved conduction velocity in injured myocardium, and suppressed re-entry, without compromising electrophysiology in healthy tissue.

Conclusions

This study identifies ventricular α4β2 nAChRs as a druggable target for FVTs and introduces salvage-1 as a first-in-class therapeutic candidate, thereby establishing a new direction for the pharmacological therapy of cardiac arrhythmias.

Structured Graphical Abstract

Structured Graphical Abstract.

See the editorial comment for this article ‘Non-neural cholinergic modulation of cardiac electrophysiology: fact or hypothesis?’, by A. Zaza and A. Becchetti, https://doi.org/10.1093/eurheartj/ehag425.

Translational Perspective.

This study identifies ventricular α4β2 nicotinic acetylcholine receptors as a previously unrecognized druggable target for fatal ventricular arrhythmias (FVTs). The first-in-class, positive allosteric modulator salvage-1 demonstrates potent antiarrhythmic efficacy across rodent, porcine, and human ex vivo heart models, outperforming the standard-of-care lidocaine in both effectiveness and safety. These properties support its potential application in both FVT prevention, such as in patients with a history of sudden cardiac arrest or those with underlying structural heart disease, and FVT treatment, including acute episodes of ventricular tachycardia and ventricular fibrillation, as well as an adjunctive therapy during cardiac surgery or catheter ablation procedures. These findings establish a new therapeutic paradigm that shifts from non-selective ion channel blockade to targeted receptor modulation.

Introduction

Despite significant advancements in arrhythmia management, the safe and effective pharmacological treatment of cardiac arrhythmias remains a critical challenge in modern clinical practice.^1–4 Currently approved antiarrhythmic agents, such as ion channel blockers, are limited by their narrow therapeutic windows and substantial proarrhythmic risks. These drugs not only have the potential to worsen existing arrhythmias but also to trigger new, life-threatening arrhythmic events, thereby increasing the risk of all-cause and cardiovascular mortality.^2,3,5–10 These safety concerns underscore the urgent need for more selective, mechanism-based antiarrhythmic strategies.

Clinically, fatal ventricular tachyarrhythmias (FVTs), such as ventricular tachycardia (VT) and ventricular fibrillation (VF), are among the deadliest arrhythmias, often leading to sudden cardiac death within minutes if untreated.^5,11–14 Our recent research has revealed that the endogenous cholinergic system (ECS) in ventricular cardiomyocytes modulates local electrophysiology independently of autonomic input, in which nicotinic acetylcholine receptors (nAChRs) act as key signalling mediators. Importantly, our experimental evidence establishes that dysregulated α4β2 nAChR signalling contributes to FVT pathogenesis and demonstrates that targeted receptor potentiation effectively suppresses arrhythmia susceptibility.¹⁵ These findings collectively point to α4β2 nAChR as a potential therapeutic target for antiarrhythmic intervention. To translate this promise into clinical reality, a dual approach is essential: systematic validation of its target credibility, coupled with the development of therapeutically viable agents featuring innovative structural architectures.

In this work, we therefore aimed to explore the antiarrhythmic potential of targeting α4β2 nAChRs in the setting of FVTs. Using a combination of structural analysis and pharmacological screening, we developed modulators of α4β2 nAChRs and evaluated their ability to influence receptor function and suppress arrhythmogenesis. Our findings identify modulation of α4β2 nAChRs as an effective antiarrhythmic strategy and highlight a non-canonical target beyond the traditional ion-channel–centric framework of antiarrhythmic drug discovery.

Methods

Full details of the methodology are available in the Supplementary Materials. Below is a summarized version of the key methods used in this study.

Optical mapping

Optical mapping was performed as described previously.^16–18 Briefly, rat or mouse hearts were excised and perfused on a Langendorff apparatus with oxygenated Tyrode's solution. After recovery, hearts were loaded with Di-4-ANEPPS (10 µM). For ischaemia-reperfusion (I/R) studies, rat hearts underwent left anterior descending (LAD) artery ligation for 60 min, followed by 60 min of reperfusion before optical recording. Transverse ventricular slices (300 µm) were superfused with oxygenated Tyrode's solution at 37 °C. Fluorescence was excited at 530 ± 25 nm and collected >610 nm at 1000 frames per second using a high-speed CMOS camera. Data analysis was performed using BV Workbench software.

Rats transverse aortic constriction model

Pressure overload heart failure (HF) was induced by transverse aortic constriction surgery with minor modifications.¹⁹ Briefly, a midline sternotomy was performed, exposing the aortic arch, which was ligated using a 6–0 silk suture tied around a 27-G needle. Sham controls underwent identical procedures, excluding aortic ligation. The chest was closed, and animals were maintained for subsequent assessments.

Model of ischaemia-reperfusion-induced fatal ventricular tachyarrhythmias

For the ex vivo rat model, hearts were rapidly excised and mounted on a Langendorff perfusion system. After equilibration, LAD ligation was applied for 60 min, followed by 60 min of reperfusion. PES commenced immediately after the I/R injury protocol. ECGs were recorded continuously, and electrical stimulation was delivered via platinum electrodes positioned at the cardiac apex. Ten stimulations were applied at 2-min intervals, with pauses for rhythm recovery if needed. The entire experiment was completed within 60 min after reperfusion.

For the porcine model of myocardial I/R-induced FVT, male domestic pigs (30–40 kg) were anesthetized, and a balloon catheter was inserted into the LAD to induce complete occlusion for 60 min, followed by 72 h of reperfusion. Following reperfusion, pigs were anesthetized with Zoletil 50 (2 mg/kg) and xylazine (1.5 mg/kg), intubated, and anticoagulated with heparin (500 IU/kg). Hearts were excised and mounted on a Langendorff apparatus, perfused with oxygenated Tyrode’s solution. Programmed stimulation was applied using the S1S2S3 protocol. Stimulation continued until VT, ventricular flutter (VFL), or VF lasting >1 s was induced. The stimulation protocol was conducted within 30 min, and the entire experimental period was confined to within 60 min.

For the human model of HF-induced FVT, discarded human hearts from transplant recipients were maintained in HTK cardioplegic solution at 4°C for preservation. Langendorff perfusion was established via cannulation of the coronary ostia and maintained at 36.5 ± 0.5°C using oxygenated Tyrode's solution. Programmed stimulation was applied using the S1-S2-S3 protocol. Stimulation was repeated with decrements in S3 until VT, VFL, or VF was induced. The entire experiment was conducted within 60 min.

Artificial intelligence-based drug design and synthesis

A generative deep learning model (FragGPT) was used to design novel chemical entities. Candidate molecules were identified using molecular docking and structural evaluation. The compounds were synthesized using standard organic synthetic techniques, purified to >98% purity, and characterized through ¹H/¹³C-NMR spectroscopy and high-resolution mass spectrometry.

Chemical synthesis of salvage-1

Salvage-1 was synthesized by reacting 5-methoxytryptamine with 2-nitrobenzyl bromide in tetrahydrofuran, followed by purification using flash chromatography. The product was confirmed by NMR and mass spectrometry.

Statistical analysis

All statistical analyses were conducted using GraphPad Prism software and IBM SPSS. The normality of data distribution was evaluated by the Shapiro-Wilk test and Q-Q plots. Data are reported as mean ± standard deviation for continuous variables that are approximately normally distributed, and as median (interquartile range) for skewed data. Statistical significance was defined as P < .05.

RESULTS

Functional screening identifies α4β2 nAChRs as a novel therapeutic target for FVTs

To identify novel druggable targets for FVTs, we conducted a systematic pharmacological screening of compounds targeting the ECS—a recently identified local regulator of ventricular electrophysiology.¹⁵ Using an ex vivo rat model of ischaemia-reperfusion (I/R)-induced FVTs that reliably produced sustained arrhythmic episodes (Supplementary data online, Figure S1A), we screened 56 compounds sourced from DrugBank and PubChem databases,^20,21 including 19 clinically approved drugs and 37 tool compounds targeting ACh-related pathways (Figure 1A and Supplementary data online, Table S1). Notably, varenicline, a partial agonist of α4β2 nAChRs, and desformylflustrabromine (dFBr), a positive allosteric modulator (PAM) of α4β2 nAChRs, emerged as the most effective candidates (Supplementary data online, Table S2 and S3). At 10 µM, varenicline reduced FVT incidence from 100% to 25% (P < .0001) and shortened arrhythmia duration from 9.5 min (95% CI 5.59–13.36) to 0.76 min (95% CI 0.07–1.45) (P < .0001) (Figure 1B and C). Strikingly, dFBr exhibited even stronger efficacy, significantly lowering FVT incidence and duration (Figure 1D and E). These antiarrhythmic effects were further confirmed in vivo, where both compounds (10 mg/kg) significantly suppressed FVT inducibility and duration in I/R-injured rats, with dFBr consistently outperforming varenicline across all endpoints (Figure 1F–I; Supplementary data online, Figure S2). Taken together, these results not only validate α4β2 nAChRs as a promising target for FVT intervention but also suggest that allosteric modulation may provide superior antiarrhythmic efficacy compared to direct receptor activation.

Figure 1.

Functional screening identifies ventricular α4β2 nAChRs as a novel therapeutic target for FVTs. (A) Three-stage screening pipeline: > 4000 FDA-approved drugs and >10 000 compounds were triaged to 56 ventricular endogenous cholinergic system (ECS) modulators, functionally assessed in an ex vivo rat ischaemia-reperfusion (I/R) model, and verified by optical mapping, patch clamp, and protein assays. (B–E), Representative electrocardiograms (ECGs) and bar graphs show that varenicline (Var) (B) and desformylflustrabromine (dFBr) (D) uniquely suppressed fatal ventricular tachyarrhythmia (FVT) induction and shortened episode duration. n = 6–11 per group. (F – I) Representative ECG and bar graphs from rats treated with 10 mg/kg Var (F–G) or dFBr (H–I), n = 5–10 per group. Data are reported as median (IQR) or mean (SD). One-way ANOVA with Dunnett multiple comparison test for (I, left); Kruskal–Wallis test followed by Dunn’s multiple comparison test for (C), (E), (G), and (I, right)

Given that impaired ventricular electrical conduction is a key electrophysiological factor in arrhythmogenesis,^22,23 we next assessed whether the two compounds modulate conduction velocity (CV) following ex vivo rat I/R injury. Optical mapping with Di-4-ANEPPS revealed a dramatic reduction in CV in the I/R-injured ventricular myocardium (0.51 ± 0.048 mm/ms) compared to healthy myocardium (0.98 ± 0.095 mm/ms, P < .0001). Application of varenicline and dFBr partially reversed this conduction deficit, increasing CV from 0.51 ± 0.048 mm/ms to 0.69 ± 0.11 mm/ms (P = .0095) and 0.77 ± 0.055 mm/ms (P < .0001), respectively (Figure 2A). This functional readout further supports the antiarrhythmic efficacy of α4β2 nAChRs modulation.

Figure 2.

α4β2 nAChRs modulation improves conduction and restores receptor function after I/R injury. (A) Isochronal maps reveal I/R-induced CV impairment, which is partially reversed by Var or dFBr. n = 7–8 per group. (B) Whole-cell recordings reveal a significant loss of acetylcholine (ACh)-evoked current after I/R, partially restored by both agents (n = 9–10 cells from 3–4 rats per group). (C and D) Capillary western blot (C) and immunofluorescence (D) show reduced CHRNA4 and CHRNB2 expression in ischaemia-reperfusion (I/R)-injured rat hearts compared to sham controls. Scale bars, 50 μm. n = 4 independent biological samples per group. (E and F) Capillary Western (E) and immunofluorescence (F) demonstrate a reduction in CHRNA4/CHRNB2 protein in failing human ventricles (n = 6 per group). Scale bar, 75 μm. Data are reported as median (IQR) or mean (SD). Brown–Forsythe and Welch ANOVA followed by Dunnett’s T3 multiple comparisons test for (A); Kruskal–Wallis test followed by Dunn’s multiple comparison test for (B); Unpaired two-tailed t-test for (C) and (E, right); Mann-Whitney test for (E, left)

To further assess the functional relevance of α4β2 nAChRs as therapeutic targets, we performed patch clamp recordings on rat ventricular cardiomyocytes isolated from hearts subjected to I/R injury. Following I/R injury, we observed a dramatic reduction in ACh-gated currents (2.8 ± 2.4 vs 0.39 ± 0.02 pA/pF, P = .00012), suggesting substantial receptor dysfunction. Importantly, treatment with varenicline and dFBr partially restored these currents (varenicline: P = .036, and dFBr: P = .0030), respectively (Figure 2B). These findings provide direct evidence supporting the therapeutic potential of α4β2 nAChRs in stabilizing electrical activity in the injured myocardium.

Additionally, we observed significant reductions in the expression of α4β2 nAChRs subunits in both I/R-injured rat hearts and failing human hearts. In I/R-injured rat myocardium, protein levels of the α4 subunit (encoded by Chrna4) and β2 subunit (encoded by Chrnb2) were significantly reduced (α4: P = .0049; β2: P = .033) (Figure 2C). This was corroborated by immunofluorescence imaging, which showed a near-complete loss of α4β2 nAChRs staining compared to sham control (Figure 2D and Supplementary data online, Figure S3). A corresponding reduction in α4 and β2 subunit protein expression was also confirmed at the tissue level in failing human myocardium (Figure 2E and F). These findings indicate that pathological remodelling of α4β2 nAChRs contributes to the molecular substrate underlying arrhythmogenesis.

In summary, these multifaceted results—encompassing arrhythmia suppression, conduction restoration, receptor engagement, and disease-related receptor expression loss—collectively establish α4β2 nAChRs as an unrecognized but highly promising target for antiarrhythmic therapy. Our data highlight allosteric activation of this receptor class, exemplified by dFBr, as a novel and mechanistically distinct strategy for treating FVTs.

Design of salvage-1 as a novel structural PAM of α4β2 nAChRs

Building on the potent antiarrhythmic activity of the α4β2 nAChR-PAM dFBr, we implemented an artificial intelligence (AI)-assisted drug discovery pipeline to identify structurally novel and functionally optimized PAMs (Figure 3A). Using dFBr as a structural template, we pursued a two-stage design strategy: (i) fragment-based molecular generation was conducted with FragGPT, a generative deep learning model that constructs diverse chemical scaffolds while preserving key pharmacophoric elements from the template.^24,25 This approach broadly explored chemical space and yielded an initial set of candidate molecules. Molecular docking and structural evaluation of this AI-generated library identified promising initial leads, salvage-2 to salvage-4, each containing an aromatic core and hydrogen-bond-capable side chains suitable for further optimization. (ⅱ) These AI-derived leads then underwent systematic structural optimization via rational design strategies such as scaffold hopping and fragment replacement. Aromatic cores were systematically varied to identify optimal molecular frameworks, while side chains with hydrogen-bonding motifs were introduced to synergistically enhance binding affinity for α4β2 nAChRs, thereby prioritizing compounds with enhanced therapeutic potential.

Figure 3.

Salvage-1 demonstrates marked antiarrhythmic efficacy in rodent models of I/R and HF. (A) Left: An AI-assisted workflow integrating lead-compound and target-based molecular generative modelling to generate 23 novel candidate compounds based on the structure of desformylflustrabromine (dFBr). Right: Functional assessment of the compounds across rodent, porcine, and human models. (B and C) Preventive effect of 10 µM salvage-1 abolishes inducible fatal ventricular tachyarrhythmias (FVT) ex vivo (n = 5–10 per group). (D and E) Preventive effect of 1 mg/kg salvage-1 abolishes inducible FVT in vivo (n = 7–8 per group). (F–I) A therapeutic bolus of salvage-1 (200 µM) rapidly terminates established FVT in both I/R (n = 9 per group, F and G) and HF (n = 6–8 per group, H and I) rat models. Data are reported as median (IQR). Kruskal–Wallis test followed by Dunn’s multiple comparison test for (C) and (E); Fisher’s exact test for (G, left) and (I, left); Mann-Whitney test for (G, right) and (I, right)

Employing this dual-pronged design approach, we developed multiple ligand series and established efficient gram-scale synthetic routes. The target compounds were readily accessible through standard medicinal chemistry methodologies, including substitution reactions, Suzuki–Miyaura cross-couplings, and amidation reactions, achieving overall yields of 35%–80%. In total, 23 structurally diverse novel small molecules were synthesized, representing chemotypes such as indoles, benzofurans, benzothiazoles, and quinoxalines. This final collection included the initial leads (salvage-2 to salvage-4) along with their rationally optimized derivatives (Supplementary data online, Figure S4). All compounds were synthesized via convergent routes suitable for parallel synthesis, purified to >98% purity by flash chromatography, and fully characterized through comprehensive ¹H/¹³C-NMR spectroscopy and high-resolution mass spectrometry. Collectively, this carefully curated chemical library provides a robust foundation for subsequent pharmacodynamic studies, safety assessments, and efficacy evaluations.

Salvage-1 demonstrates marked antiarrhythmic efficacy in rodent models of FVTs

To identify the most promising α4β2 nAChRs-targeting compounds for FVT treatment, we screened 23 candidates (salvage-1 to salvage-23) in ex vivo perfused hearts from healthy rats. Salvage-22 induced dose-dependent bradycardia (at 10 μM), suggesting proarrhythmic potential and was excluded from further analysis (Supplementary data online, Table S4). The remaining 22 compounds exhibited stable ECG parameters across 0.1–10 μM, showing no significant alterations in heart rate, PR interval, QRS duration, or QTc interval (Supplementary data online, Table S4) and were thus deemed suitable for efficacy testing. Antiarrhythmic activity was then assessed in a rat ex vivo model of I/R-induced FVT. Notably, salvage-1 demonstrated the most potent effect, reducing arrhythmia inducibility from 100% (vehicle) to 5% at 10 μM (P < .0001) and suppressing arrhythmia duration from a median of 10 (IQR: 6.3–11) to a median of 0 (IQR: 0–0.04) min (P < .0001) (Figure 3B and C; Supplementary data online, Table S5). Optical mapping of infarcted ventricular slices revealed that salvage-1 significantly improved CV within infarct zones compared to vehicle (Supplementary data online, Table S6).

Further in vivo studies in a rat model of I/R-induced FVTs corroborated the ex vivo findings, demonstrating the robust antiarrhythmic efficacy of salvage-1. At 1 mg/kg, salvage-1 reduced FVT incidence from a median of 95% (IQR: 90–100) to 5% (IQR: 0–10) (P < .0001) and shortened arrhythmia duration from a median of 15 s (IQR: 10–18) to 0.08 s (IQR: 0–0.54) (P < .0001) (Figure 3D and E). Importantly, salvage-1 also effectively terminated ongoing arrhythmias: a single bolus of 200 μM salvage-1 resolved 77.8% of FVT episodes within 2 min (Figure 3F and G), highlighting its suitability for both prophylactic and acute intervention.

To assess the cross-model robustness of salvage-1, we evaluated its efficacy in a rat model of pressure overload-induced HF. In this model, salvage-1 successfully terminated 87.5% of FVT episodes within 2 min, whereas no termination occurred in vehicle-treated controls (Figure 3H and I).

In summary, these results identify salvage-1 as a potent antiarrhythmic agent with dual therapeutic action—preventing FVT onset and promoting rapid termination. Its consistent efficacy across I/R and HF models and rapid onset of action underscore its potential as a next-generation antiarrhythmic therapy.

Salvage-1 also rapidly terminates FVTs in porcine and human hearts

To assess the translational potential of salvage-1 beyond rodent models, we investigated its antiarrhythmic efficacy in large-animal and human ex vivo models of FVTs. In the porcine I/R injury model, programmed electrical stimulation consistently induced FVTs that lasted over 10 min in all hearts tested (Supplementary data online, Figure S1B). Treatment with salvage-1 (200 μM bolus followed by 10 μM continuous perfusion) achieved sinus rhythm restoration in all specimens (100% vs 0% in vehicle-treated controls, P = .0079) and significantly shortened arrhythmia duration from 38 ± 6.2 to 3.1 ± 0.76 min (P = .0060) (Figure 4A; Supplementary data online, Video S1).

Figure 4.

Salvage-1 demonstrates superior efficacy compared to lidocaine in both porcine and human models of FVTs. (A) Top, experimental timeline for porcine ischaemia-reperfusion (I/R) hearts; Electrocardiogram (ECG) panels illustrate model fidelity. Middle and bottom, salvage-1 demonstrates superior efficacy to lidocaine in restoring sinus rhythm in porcine hearts, with significantly higher success rate and faster termination time (n = 5 per group). (B) Top, workflow of ex vivo working-heart perfusion using the recipient's hearts. Middle and bottom, in human recipient hearts, salvage-1 effectively restores atrial pacing rhythm and suppresses FVT, demonstrating superior efficacy to lidocaine (n = 6 per group). (C and D) Multi-electrode array maps demonstrate salvage-1's superior ability over lidocaine to enhance conduction velocity and reduce electrical heterogeneity in both porcine and recipient hearts. (pig, n = 5 per group; recipient, n = 6 per group). Data are reported as median (IQR) or mean (SD). Kruskal–Wallis test followed by Dunn’s multiple comparison test for (A, right), (C, left) and (D, right); Brown–Forsythe and Welch ANOVA, followed by Dunnett’s T3 multiple comparisons test for (D, left); One-way ANOVA with Dunnett multiple comparison test for (B, right) and (C, right); Fisher’s exact test for (A, left) and (B, left)

To further evaluate the clinical translatability of salvage-1, we extended our assessment to explanted human hearts from patients with end-stage HF. These specimens required atrial pacing due to surgical preservation of the sinoatrial node. Under this clinically relevant configuration, programmed ventricular stimulation successfully induced sustained FVTs in all hearts (Figure 4B; Supplementary data online, Figure S1C and Supplementary data online, Table S7). Notably, in this human ex vivo model, salvage-1 (200 μM bolus followed by 10 μM continuous perfusion) achieved termination of 100% of arrhythmic episodes within 4 min (Figure 4B; Supplementary data online, Video S2).

The consistent efficacy across porcine and human heart models, with approximately 90% reduction in arrhythmia duration, demonstrates cross-species conservation of salvage-1’s therapeutic effect. These results establish salvage-1 as a potent antiarrhythmic agent with rapid onset of action, supporting its clinical relevance in emergency settings such as cardiac arrest or as an adjunct during catheter ablation procedures.

In summary, these findings position salvage-1 as a mechanistically guided, clinically viable candidate for terminating life-threatening ventricular arrhythmias. Its efficacy in both human and large-animal models strongly supports its translational promise and progression towards clinical development.

Superior efficacy of salvage-1 over lidocaine with preserved baseline electrophysiology

Given that lidocaine remains a first-line clinical agent for managing FVTs,^26–28 we directly compared its efficacy with salvage-1 in ex vivo porcine and human FVT models. In the porcine I/R injury model, salvage-1 terminated all FVT episodes within 4 min, whereas lidocaine achieved termination in only 20% of cases. Moreover, salvage-1 markedly reduced arrhythmia duration compared with lidocaine (3.1 ± 0.76 vs 25 ± 16 min, P = .039) (Figure 4A, bottom). Similarly, in explanted human hearts, salvage-1 consistently terminated all FVT episodes within 2 min, while lidocaine succeeded in only 17% of samples (Figure 4B, bottom). These findings establish salvage-1 as a potent antiarrhythmic agent with superior efficacy and a pronounced capacity to rapidly restore sinus rhythm.

Importantly, beyond its efficacy, salvage-1 maintained a stable electrophysiological profile. No significant alterations in baseline electrophysiological parameters—including heart rate, PR interval, QRS duration, and QTc interval—were observed following administration of salvage-1 in ex vivo porcine hearts compared with vehicle controls (P > .05 for all parameters; Supplementary data online, Figure S5A).

Multi-electrode mapping demonstrated that salvage-1 selectively enhanced conduction in injured myocardium. In porcine I/R hearts, salvage-1 increased CV in injured ventricular regions from 0.44 mm/ms (95% CI 0.32–0.55) to 0.83 mm/ms (95% CI 0.72–0.94; P = .018; Figure 4C), while CV in non-injured regions remained unchanged. Salvage-1 also significantly reduced conduction dispersion from 15% (95% CI 8.9–21) to 2.9% (95% CI 0.94–4.8; P = .0014). In contrast, lidocaine affected conduction in both diseased and healthy myocardium (Supplementary data online, Figure S5B). Similar findings were observed in explanted human hearts, where CV increased from 0.41 mm/ms (95% CI 0.38–0.44) to 0.68 mm/ms (95% CI 0.59–0.77; P = .00072), accompanied by a reduction in electrical dispersion from 16% (95% CI 10–21) to 5.2% (95% CI: 2.3–8; P = .016) (Figure 4D).

Collectively, these data indicate that salvage-1 not only achieves rapid and robust arrhythmia termination but also selectively restores conduction in diseased myocardium without perturbing physiological electrical stability, thereby supporting a favourable therapeutic window for the management of life-threatening ventricular arrhythmias.

Salvage-1 suppresses re-entry by allosterically potentiating α4β2 nAChRs

FVTs, such as VT and VF, are primarily maintained by re-entry, a mechanism closely linked to impaired electrical conduction in damaged myocardial tissue.^29,30 We have previously shown that acetylcholine activates nAChRs on ventricular cardiomyocytes, generating currents that modulate cellular excitability and conduction¹⁵. We therefore hypothesized that salvage-1 suppresses re-entrant arrhythmias by enhancing ACh-gated currents, thereby improving excitability and conduction to disrupt self-sustaining arrhythmic circuits.

To test this, we employed high-resolution optical mapping in I/R-injured rat ventricles. As Figure 5A shown, salvage-1 markedly suppressed re-entrant activity, as evidenced by significant reductions in both phase singularities (Supplementary data online, Video S3) and dominant frequency—key electrophysiological markers of re-entry.^31–33 We next evaluated the CV in the infarct border zone. Salvage-1 increased CV from 0.18 ± 0.098 mm/ms to 0.54 ± 0.16 mm/ms (P < .0001), an enhancement that likely contributes to re-entry disruption.

Figure 5.

Salvage-1 suppresses re-entry through α4β2 nAChR-mediated enhancement of myocardial excitability and conductivity. (A) Optical mapping reveals that salvage-1 stabilizes cardiac electrical activity in ischaemia-reperfusion (I/R) rat hearts by eliminating spiral wave re-entry (A, left), reducing dominant frequency (DF) (A, middle), and improving conduction velocity (A, right), compared with vehicle. n = 9–10 per group. (B and C) Patch clamp recording show that salvage-1 reduces action potential threshold (B) and increases acetylcholine (ACh)-gated currents (C) (n = 8–9 cells from three rats per group). Data are reported as median (IQR) or mean (SD). Mann-Whitney test for (A, left); Unpaired two-tailed t-test with Welch’s correction for (A, middle); Unpaired two-tailed t-test (A, right and B).

At the cellular level, salvage-1 significantly reduced the action potential initiation threshold (Figure 5B), indicating improved excitability in compromised myocardium. Accordingly, whole-cell patch clamp recordings showed that 10 μM salvage-1 potentiated ACh-gated currents by approximately 2-fold vs vehicle-treated cells (Figure 5C), consistent with its role as an α4β2 nAChRs PAM.

In summary, these results demonstrate that salvage-1 selectively enhances α4β2 nAChRs function, leading to increased ACh-gated currents, improved excitability and conduction, and ultimately suppression of re-entrant arrhythmias. This receptor-targeted mechanism offers a novel electrophysiological strategy for rhythm control in diseased ventricles.

Salvage-1 selectively potentiates α4β2 nAChRs by interacting with residues Phe312 and Phe316 of α4 subunit, thereby stabilizing the receptor in its open-channel state

To elucidate the molecular mechanism underlying the antiarrhythmic effect of salvage-1, we conducted thermal shift assays (TSA). The results indicated that treatment with 10 μM salvage-1 induced a rightward shift in the temperature-dependent denaturation curve. Furthermore, analysis of the area under the curve revealed a significant increase in the curve area (4.16 ± 0.16 in the vehicle group vs 5.36 ± 0.32 in the salvage-1 group, P = .0017). These findings demonstrate that salvage-1 selectively stabilizes the α4 subunit (CHRNA4), with minimal effect on the β2 subunit (CHRNB2) (Figure 6A). Notably, at therapeutically relevant concentrations (10 μM), salvage-1 showed weak inhibition of key cardiac ion channels, Cav1.2 and Nav1.5 (Supplementary data online, Figure S6A and B), and a low affinity for the dopamine transporter (Supplementary data online, Figure S6C). These results collectively support the high target specificity of salvage-1.

Figure 6.

Salvage-1 acts as an allosteric stabilizer that conformationally locks the α4β2 nAChR. (A) Thermal shift assays (TSA) demonstrate selective stabilization of CHRNA4. n = 4 independent experiments. (B) CL-MS analysis of α4β2 nAChRs in the absence and presence of salvage-1. Inset shows the details of over-length cross-link products. (C) Single-channel analysis confirms that salvage-1 increased open probability without altering the unitary conductance. Data are reported as mean (SD). Unpaired two-tailed t-test with Welch’s correction for (C, left); Two-way ANOVA for (A)

To validate the in vivo relevance of α4-containing nAChRs, we generated cardiomyocyte-specific Chrna4 knockout (Chrna4-CKO) mice (Supplementary data online, Figure S7). Following optical mapping and ECG recording, we observed that salvage-1 failed to induce statistically significant changes in FVT incidence, FVT duration, phase singularities, or dominant frequency or CV in Chrna4-CKO mice (Supplementary data online, Figure S8), thereby confirming that α4β2 nAChRs are essential for its antiarrhythmic effects.

Structural analysis using chemical cross-linking mass spectrometry revealed multiple cross-link products in the absence of salvage-1, most of which exceeded the distance constraint of the cross-linker DCD (31.86 Å),³⁴ and were classified as over-length cross-links (Figure 6B and Supplementary data online, Table S8). Upon salvage-1 binding, we observed a marked reduction in total and over-length cross-links (Figure 6B and Supplementary data online, Table S8), indicating that salvage-1 enhances structural stability and locks the receptor into a defined conformation. Single-channel recordings confirmed that salvage-1 significantly increases open-channel probability without affecting unitary conductance (Figure 6C).

To identify the structural basis for salvage-1 binding, we performed molecular docking against the α4β2 nAChRs cryo-EM structure (PDB: 8ST0),³⁵ identifying five candidate sites on the α4 subunit (Figure 7A). Alanine-scanning mutagenesis revealed that mutations at Phe312 and Phe316 (Mut.5) markedly impaired salvage-1 binding (Supplementary data online, Figure S9A and B). Consistently, F312A and F316A abolished salvage-1–induced potentiation of ACh-gated currents (Figure 7B and Supplementary data online, Figure S9C and D), establishing these residues as critical for its activity.

Figure 7.

Molecular mechanism of α4β2 nAChR potentiation by salvage-1. (A) Molecular docking analysis predicts five groups of possible binding sites of salvage-1 to the α4 subunit. (B) Typical traces of acetylcholine (ACh)-gated currents in HEK293T cells expressing full CHRNA4 wild-type mutant 1–5 (Mut.1: K50A/W51A/W119A/R120A; Mut.2: W88A/D202A; Mut.3: F170A/I251A/W622A; Mut.4: E270A/E589A/M596A; Mut.5: F312A/F316A). Salvage-1 potentiation is lost in Mut.5: F312A/F316A (n = 8–11 per group). (C–F) Molecular dynamics simulations: ACh + salvage-1 yields the widest minimum pore radius and highest open probability (C–D). Dynamic cross-correlation matrix and root mean square fluctuation analyses reveal dampened anticorrelated motion and reduced flexibility in α4 residues 72–176 and 248–289 (E–F). Data are reported as mean (SD). One-way ANOVA with Dunnett multiple comparison test for (B)

Molecular dynamics simulations of four receptor states (apo, salvage-1–bound, ACh-bound, ACh&salvage-1–bound) showed the highest open probability (∼35%) in the ACh&salvage-1–bound state vs lowest (∼7%) in the apo state (Figure 7C). The ACh&salvage-1–bound state exhibited wider pore dimensions (Figure 7D), decreased correlated motions between regions 72–176 and 248–289 (Figure 7E), and reduced flexibility in residues 72–176 (Figure 7F), collectively indicating that salvage-1 binding stabilizes the receptor and promotes pore opening.

In summary, these results demonstrate that salvage-1 binds specifically to Phe312 and Phe316 on the α4 subunit of α4β2 nAChRs, stabilizing the open-channel state and enhancing antiarrhythmic efficacy through allosteric potentiation.

Discussion

This study establishes ventricular α4β2 nAChRs as a therapeutic target for life-threatening ventricular arrhythmias. Our work advances the field in three major ways. First, we identify ventricular α4β2 nAChRs as a previously unrecognized druggable target. Allosteric activation of this receptor confers greater antiarrhythmic efficacy than lidocaine and reveals an electrophysiological role for the cardiac cholinergic system beyond its established functions in cellular homeostasis.³⁶ Second, we develop salvage-1, a selective PAM that engages Phe312 and Phe316 within the α4 subunit, providing structural specificity at the receptor level. Third, we provide comprehensive functional, structural, and computational validation demonstrating that salvage-1 suppresses FVTs through selective enhancement of α4β2 nAChR-mediated cholinergic signalling (Structured Graphical Abstract).

The traditional paradigm for managing cardiac arrhythmias, established in the late 20th century, relied on broad-spectrum blockade of Na⁺, K⁺, or Ca²⁺ currents.^1,3,37–40 This approach was fundamentally challenged by the CAST, CAST II, and SWORD trials, which demonstrated that non-selective channel inhibition could paradoxically increase mortality,^7–9,41 largely halting drug development in this area for decades. In contrast, we identify α4β2 nAChRs as a previously unrecognized target for FVTs, demonstrating markedly improved safety and efficacy over the standard therapy, lidocaine. Our novel compound, salvage-1, acts as a PAM of α4β2 nAChRs, offering a targeted mechanism to suppress re-entrant activity—the primary electrophysiological basis of FVTs. Thus, our work represents a strategic shift from non-selective channel blockade to targeted receptor modulation.

Beyond mechanistic novelty, targeting α4β2 nAChRs confers distinct therapeutic advantages over conventional sodium channel blockade. In direct head-to-head comparisons, salvage-1 achieved faster and more complete termination of FVTs than lidocaine in both porcine and explanted human hearts, underscoring its superior antiarrhythmic efficacy. Importantly, this enhanced efficacy was not accompanied by global electrophysiological perturbation. Salvage-1 did not significantly alter baseline heart rate, PR interval, QRS duration, or QTc interval, indicating preservation of fundamental cardiac electrical properties. Rather than exerting diffuse conduction slowing, salvage-1 selectively improved CV and reduced electrical dispersion within injured myocardium, while leaving non-injured regions unaffected. This spatially restricted enhancement of excitability contrasts with the non-selective sodium channel inhibition of lidocaine, which can affect conduction in both diseased and healthy tissue. By restoring conduction homogeneity specifically in the arrhythmogenic substrate while maintaining physiological stability elsewhere, α4β2 nAChR potentiation establishes a wider therapeutic window and represents a more refined strategy for rhythm control in structurally compromised ventricles.

This study delineates the antiarrhythmic mechanism of salvage-1, which specifically targets α4β2 nAChRs to suppress re-entrant activity, the primary electrophysiological substrate for FVTs. Re-entry is perpetuated by a pathological circuit of slow conduction or unidirectional block (zigzag conduction) within injured myocardium.^29,30,42 Our findings demonstrate that salvage-1, a PAM, binds selectively to residues Phe312 and Phe316 on the α4 subunit, stabilizing the receptor in an open-channel conformation. This molecular interaction potentiates acetylcholine-gated inward currents in cardiomyocytes. The resultant current depolarizes the membrane, thereby lowering the action potential excitation threshold and selectively enhancing CV specifically in the compromised infarct border zone. These electrophysiological actions collectively disrupt the conditions essential for re-entry maintenance. High-resolution optical mapping confirmed that salvage-1 eliminates spiral wave re-entry and phase singularities, restoring electrical synchrony without perturbing electrophysiology in healthy tissue. This targeted action underscores a receptor-specific mechanism to terminate re-entrant FVTs.

Several limitations warrant consideration. First, despite safeguards such as randomization, blinding, and cross-species validation, confounders including model heterogeneity, human sample variability, and the modest sample size may limit the generalizability of our findings, and larger studies would help confirm these results. Second, although key in vivo efficacy has been demonstrated in acute settings, further translational development of salvage-1 will require comprehensive chronic model evaluations to assess long-term safety and efficacy, along with full pharmacokinetic studies. Third, potential off-target effects in other organ systems expressing nAChRs (e.g. the central nervous system) remain to be fully elucidated. Fourth, the long-term safety and pharmacokinetics of salvage-1 warrant further investigation. Fifth, translational relevance to other arrhythmogenic contexts (e.g. non-ischaemic cardiomyopathy) requires additional validation. Sixth, across two complementary models, ACh and vagal stimulation reduced ischaemia-related ventricular arrhythmias in association with preservation of phosphorylated Cx43 and gap-junction integrity.^43,44 Aiba et al. showed that ACh attenuated border zone conduction slowing and suppressed VT/VF inducibility while preventing ischaemia-induced loss of phosphorylated Cx43.⁴³ Ando et al. demonstrated that efferent vagal stimulation markedly reduced ischaemia-induced VT in vivo in an atropine-sensitive manner and preserved phosphorylated Cx43 and its junctional localization; in vitro, ACh similarly maintained phosphorylated Cx43 and intercellular coupling during hypoxia.⁴⁴ However, Cx43 expression and phosphorylation were not directly examined in the present study, and further work is needed to determine whether α4β2 nAChR activation influences gap-junction remodelling during ischaemia. Finally, the full structural dynamics of α4β2 nAChRs modulation in vivo may involve conformational states not captured in current models. Addressing these limitations will be essential for advancing salvage-1 toward future clinical application.

Conclusions

This study establishes cardiac α4β2 nAChRs as a novel therapeutic target for FVTs. Using a structure-based approach targeting this receptor, we developed salvage-1, a first-in-class PAM that binds Phe312 and Phe316, locking the receptor into an open-channel state. This molecular action translates to tissue-specific therapy: salvage-1 enhances acetylcholine-gated currents selectively in injured myocardium, suppresses re-entry, and terminates FVTs, all while sparing healthy electrophysiology. Such targeted efficacy and safety distinguish it profoundly from conventional ion channel blockers. Together, these findings advance α4β2 nAChRs potentiation as a new antiarrhythmic strategy and position salvage-1 as a mechanistically distinct therapeutic, offering a promising path to address long-standing gaps in the management of life-threatening arrhythmias.

Supplementary data

Supplementary data are available at European Heart Journal online.

Declarations

Disclosure of Interest

All authors declare no disclosure of interest for this contribution.

Data Availability

All data needed to evaluate the conclusions in the paper are present in the main text or the Supplementary Materials.

Funding

This work was supported by the National Natural Science Foundation of China (82088101 to Y.-H.C., 82321005 to H.H., and 82222008 and 82470330 to D.X.); Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China (JYB2025XDXM601); Key Research Center Construction Project of Shanghai (2022ZZ01008 to Y.-H.C.); and the National Key Clinical Specialty.

Ethical Approval

All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (U.S. National Institutes of Health) and were approved by the Institutional Animal Care and Use Committee of Tongji University School of Medicine. The experimental protocol involving human samples complied with the principles of the Declaration of Helsinki and was approved by the ethics committee of Wuhan Union Hospital, Wuhan, China