Salidroside Alleviates Mobocertinib-Induced Cardiotoxicity by Improving Cardiac Electrophysiology

Abstract

Mobocertinib, a tyrosine kinase inhibitor approved for EGFR exon 20 insertion-mutated non-small cell lung cancer, induces cardiotoxicity manifesting as QT prolongation. Salidroside, a bioactive compound from Rhodiola rosea, exhibits cardioprotective properties. We investigated electrophysiological mechanisms underlying mobocertinib cardiotoxicity and salidroside's protective efficacy using Langendorff-perfused rat hearts and neonatal rat cardiomyocytes (NRCMs). Hearts were treated with escalating mobocertinib concentrations (214, 428, 856 nM) with or without salidroside (20 μM) pretreatment; cardiac parameters including heart rate, PR, JT, QTc intervals and conduction were mapped via high-resolution multichannel electrophysiology. NRCMs underwent chronic mobocertinib exposure with/without salidroside, followed by transcriptomic sequencing, qRT-PCR, and western blot validation. Mobocertinib concentration-dependently prolonged PR, JT, QTc intervals, increased conduction time/heterogeneity, reduced heart rate/velocity, and JT/QTc prolongation persisted after washout. Salidroside pretreatment markedly attenuated these abnormalities. In NRCMs, mobocertinib prolonged field potential duration and impaired conduction; salidroside normalized parameters. Transcriptomic analysis identified enrichment in calcium signaling, aldosterone synthesis, cGMP-PKG, and adrenergic pathways. Notably, Atp2b2 was upregulated in mobocertinib group but downregulated in salidroside combination group, whereas Agtr1b showed opposite pattern. Thus, salidroside mitigates mobocertinib-induced electrophysiological toxicity by modulating Atp2b2/Agtr1b-related signaling networks, providing a novel cardioprotective strategy against mobocertinib-associated cardiotoxicity.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s12012-026-10169-w.

Introduction

Lung cancer is a leading cause of malignancy-related morbidity and mortality worldwide [1]; among which non-small cell lung cancer (NSCLC) accounts for approximately 85% of these cases [2]. Within this category, non-classical mutations, such as epidermal growth factor receptor (EGFR) exon 20 insertions, present significant therapeutic challenges and are associated with poor prognosis [3]. With the rapid advancement of molecular targeted therapies, EGFR tyrosine kinase inhibitors (TKI) have emerged as the standard of care for patients with advanced NSCLC harboring EGFR mutations. Mobocertinib, an EGFR-TKI designed to target EGFR exon 20 insertion mutations, was approved for treating locally advanced or metastatic NSCLC [4]. Although mobocertinib demonstrates promising antitumor activity, cardiac safety concerns have emerged in clinical practice. Clinical trial data revealed that 11% of patients experienced corrected QT interval (QTc) prolongation exceeding 60 ms from baseline, with 1.2% exhibiting a QTc interval surpassing 500 ms and the threshold associated with a 2–threefold increased risk of Torsades de Pointes [5, 6]. Moreover, cardiac failure occurred in 2.3% of patients, with fatal cases reported [5]. These adverse events have limited its clinical application.

Sudden cardiac death resulting from non-cardiac drug-induced electrophysiological disturbances represents a major public health concern [7]. QT prolongation, reflecting delayed ventricular repolarization, is a critical indicator of cardiotoxicity caused by non-cardiac medications and may predispose individuals to life-threatening cardiac events [8]. Evidence suggests that mobocertinib may disrupt cardiac electrophysiological stability by interfering with cardiomyocyte ion channel function or signal transduction pathways [9, 10]. However, systematic research regarding the molecular mechanisms underlying mobocertinib-induced cardiotoxicity is lacking, and the electrophysiological remodeling processes associated with acute and long-term exposure remain unelucidated.

Rhodiola rosea is a medicinal herb widely used in traditional Chinese medicine. Salidroside, its primary bioactive component, exhibits a broad spectrum of pharmacological activities, including antioxidant, anti-inflammatory, anti-apoptotic, and anti-hypoxic effects, as well as the ability to improve energy metabolism [11, 12]. Notably, salidroside maintains cardiac structure and function by regulating myocardial energy metabolism, inhibiting oxidative stress, and modulating cardiomyocyte calcium homeostasis. It exhibits protective efficacy in various cardiovascular disease models, including myocardial ischemia–reperfusion injury and heart failure [13, 14]. Given that cardiotoxicity limits the safe clinical application of mobocertinib, the cardioprotective properties of salidroside suggest it may offer therapeutic value in this context.

Therefore, this study employed a Langendorff-perfused isolated heart model and neonatal rat cardiomyocytes (NRCMs), combined with high-resolution multichannel electrophysiological mapping, to systematically evaluate the effects of mobocertinib on cardiac electrophysiology and investigate the protective efficacy of salidroside. Furthermore, using transcriptomic sequencing and molecular validation, we aimed to elucidate the underlying protective mechanisms, thereby providing a theoretical foundation and novel therapeutic strategy for preventing mobocertinib-associated cardiotoxicity.

Materials and Methods

Experimental Animals

Healthy adult male Sprague–Dawley rats (8 weeks old, 200–300 g) and neonatal Sprague–Dawley rats (postnatal days 1–3) were used in this study. Animals were obtained from Henan Province SCBS Biotechnology Co., Ltd. Rats were housed in a standard animal facility with ad libitum access to food and water and underwent a one-week acclimatization period under a 12-h light/dark cycle. Following isoflurane anesthesia, rats were euthanized by cervical dislocation; rapidly excised the hearts and mounted them on a Langendorff perfusion system. Retrograde aortic perfusion was performed at a constant flow rate of 8–10 mL/min with Krebs–Henseleit (K-H) solution oxygenated with 95% O₂ and 5% CO₂ at 37 °C. Isolated hearts were randomly assigned to the following groups: mobocertinib (Mob, n = 8), salidroside (Sal, n = 5), and salidroside pretreatment + mobocertinib (Sal_Mob, n = 8). In the Mob group, hearts were perfused for 30 min with K-H solution containing low (214 nM), medium (428 nM), or high (856 nM) concentrations of mobocertinib (HY-135815, MedChemExpress, China), corresponding to one, two, and four times the maximum clinical plasma concentration, respectively [15]. The Sal group was perfused with K-H solution containing 20 μM salidroside (SS8080, Solarbio, China). In the Sal_Mob group, hearts were pretreated with 20 μM salidroside for 30 min, followed by perfusion with mobocertinib at the specified concentrations. Approval for all animal experiments was obtained from the Ethics Committee of the First Affiliated Hospital of Henan Medical College.

Isolated Hearts Electrophysiological Mapping

Two-lead electrocardiogram (ECG) electrodes were placed on the right atrium and left ventricle, respectively, while a 64-channel electrode array was positioned on the left ventricular surface. A bipolar stimulation electrode was placed on the left ventricle. Field potential duration (FPD) and ECG signals were recorded using a 64-channel electrical mapping system (EMS64-USB-1003, MappingLab Ltd., UK). Electrical stimulation was delivered via a stimulator controller (VCS-3002, MappingLab Ltd., UK) (Fig. 1A), and the pacing protocol consisted of 1 mA stimulation intensity, 6 Hz frequency, and 2 ms pulse width. Data recording time points are illustrated in Fig. 1B. Data were acquired using MappingLab EMapRecord v5.8.3 software and analyzed using MappingLab EMapScope 5 v5.9.41 software. Conduction time was defined as the interval between the earliest and latest activation across all 64 channels. Conduction velocity was calculated as the distance between the corresponding electrodes divided by this interval. Conduction heterogeneity was quantified as the standard deviation of activation times across the 64‑channel array. All electrophysiological parameters were manually measured from three distinct cardiac cycles per heart, and the average value was used for analysis (Figs. 2, 3).

Fig. 1.

Effects of Salidroside on blank isolated hearts. A Schematic diagram of isolated heart electrophysiological mapping; B Flowchart of isolated heart experiment; C Representative ECG; D Representative conduction and conductive heterogeneity heatmap; E–K Effects of Salidroside on heart rate, PR interval, JT interval, QTc interval, conduction time, conduction velocity, and conduction heterogeneity in blank isolated hearts. Data are presented as mean ± SD. A one-way repeated measures ANOVA was used for within-group comparisons over time, followed by Tukey’s HSD post-hoc test to compare each time point with baseline. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 vs. the corresponding value at baseline

Fig. 2.

Toxic effects of Mobocertinib on isolated hearts. A Representative ECG; B–H Effects of different concentrations of Mobocertinib on heart rate, PR interval, JT interval, QTc interval, conduction time, conduction velocity, and conduction heterogeneity in isolated hearts. Data are presented as mean ± SD. Comparisons among different concentrations were performed by one-way ANOVA followed by Tukey’s HSD post-hoc test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 vs. the control group

Fig. 3.

Protective effects of Salidroside against Mobocertinib induced cardiotoxicity in isolated hearts. A Representative ECG; B Representative conduction heatmap of Mob group and Sal_Mob group; C Representative conductive heterogeneity heatmap of Mob group and Sal_Mob group; D–J Effects of different concentrations of Mobocertinib on heart rate, PR interval, JT interval, QTc interval, conduction time, conduction velocity, and conduction heterogeneity in isolated hearts following Salidroside pretreatment. Data are presented as mean ± SD. Within-group comparisons over time were assessed by one-way repeated measures ANOVA. Between-group comparisons were performed using Student’s t-test with Holm-Bonferroni correction for multiple tests. ^*P < 0.05/3, ^**P < 0.01/3, ^***P < 0.001/3 vs. Mob group at same dose; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs. self-control

Isolation and Culture of NRCMs

Neonatal Sprague–Dawley rats (postnatal days 1–3) were obtained from Henan Province SCBS Biotechnology Co., Ltd. Neonates were anesthetized with isoflurane and euthanized; hearts were excised, minced, and washed with phosphate-buffered saline (D1040, Solarbio, China) to remove residual blood. Tissue fragments were pretreated with 0.25% trypsin (CSP045, Zqxzbio, China) overnight at 4 °C, followed by digestion with type II collagenase (C8150, Solarbio, China) at 37 °C. The resulting cell suspension was collected and subjected to differential adhesion for 2 h to enrich for cardiomyocytes. Isolated cells were seeded and cultured in DMEM/F12 medium (10–092-CVRC, Corning, USA) containing 10% fetal bovine serum (10099141c, Gibco, USA) and maintained at 37 °C with 5% CO₂ atmosphere. For the first 48 h, 0.1 μM 5-Bromo-2’-deoxyuridine (B110731, Aladdin, China) was added to inhibit fibroblast proliferation. NRCM experiments included four groups: control (Con), mobocertinib (Mob), salidroside (Sal), and salidroside pretreatment + mobocertinib (Sal_Mob). In the Sal and Sal_Mob groups, 20 μM salidroside was added on day 3 and maintained for 5 days. In the Mob and Sal_Mob groups, 428 nM mobocertinib was added on day 5 and maintained for 72 h. Experiments were conducted after a total culture period of seven days.

Assessment of NRCM Purity

NRCMs were seeded onto confocal culture dishes and cultured in drug-free medium for seven days; purity was subsequently assessed via cardiac Troponin T (cTnT)/4’,6-diamidino-2-phenylindole (DAPI) immunofluorescence staining. After fixation with 4% paraformaldehyde (BL539A, Biosharp, China), cells were permeabilized using 0.1% Triton X-100 (T8200, Solarbio, China). Samples were incubated overnight at 4 °C with an anti-cTnT primary antibody (1:300; 26592–1-AP, Proteintech, China). Cells were then incubated with an Alexa Fluor 488-conjugated secondary antibody (1:1000; 4412S, Cell Signaling Technology, USA) and DAPI (1:1000; C1002, Beyotime, China) for 1 h at room temperature in the dark to label cardiomyocytes and nuclei, respectively. Images were captured using a confocal laser scanning microscope to quantify cell purity.

Electrophysiological Mapping of NRCMs

The microelectrode array was coated with 0.1% polyethyleneimine (P3143, Sigma, USA) and incubated at 37 °C for 1 h. After rinsing with sterile water, the array was coated with a substrate working solution (CA3003100, Cellapybio, China) and incubated at 37 °C for 1 h again. After removing the coating solution, NRCMs were seeded onto the array (Fig. 4A). NRCMs were assigned to four groups: Con (n = 7), Sal (n = 7), Mob (n = 6), and Sal_Mob (n = 7). The culture protocol is illustrated in Fig. 4B. A temperature control system (PCTC1001, MappingLab Ltd., UK) maintained the culture at 37 °C during recordings. FPD signals were recorded using a 64-channel mapping system (EMS64-USB-1003CS, MappingLab Ltd., UK). Electrical stimulation was delivered by configuring two channels as bipolar stimulation electrodes and the pacing protocol for NRCMs consisted of stimulation intensity 0.01 mA, pulse width 2 ms, and inter-stimulus interval 0.1 s. Data were acquired using MappingLab EMapRecord v5.8.3 and analyzed with MappingLab EMapScope 5 v5.9.41.

Fig. 4.

The electrophysiological effects of Salidroside and Mobocertinib on NRCMs. A Schematic diagram of NRCMs electrophysiological mapping; B Flowchart of NRCMs experiment; C Representative FPD; D Representative conduction and conductive heterogeneity heatmap; E–H Changes in FPD, conduction time, conduction velocity, and conduction heterogeneity in NRCMs across different treatment groups. Data are presented as mean ± SD. Comparisons among groups were performed using one‑way ANOVA with Tukey’s HSD post‑hoc test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 vs. Con

Transcriptomics Analysis

NRCMs were seeded in culture dishes and assigned to three groups (n = 3 per group): Con, Mob, and Sal_Mob. After seven days, cells were harvested, and TRIzol reagent was employed for total RNA extraction. Its concentration and purity (A260/A280) were assessed via a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA); RNA integrity (RIN > 8.0) was confirmed using an Agilent 2100 Bioanalyzer with the RNA 6000 Nano Kit (Agilent Technologies Inc., Santa Clara, CA, USA). For the construction of strand-specific libraries, the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (New England Biolabs Inc., Ipswich, MA, USA) was employed; cDNA fragments (400–500 bp) were selected and amplified. Sequencing was conducted on an Illumina NovaSeq X Plus platform with paired-end 150-bp reads.

Quantitative Real-Time PCR (qRT-PCR) Analysis

Total RNA was extracted from NRCMs in the Con, Mob, and Sal_Mob groups (n = 4 per group) using TRIzol reagent. Reverse transcription was performed using the 5 × Evo M-MLV RT Reaction Mix Ver.2 Kit (AG11728, Accurate Biology, Changsha, China). All primers were designed to amplify products ranging from 80 to 200 bp in length. qRT-PCR was conducted using 2 × SYBR Green Pro Taq HS Premix (Rox Plus) (AG11718, Accurate Biology) on a QuantStudio 6 Flex Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA). The thermal cycling conditions consisted of initial denaturation at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 30 s. Primer sequences are listed in Table 1. The expression of target genes was normalized against glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and relative expression was calculated using the 2 − ΔΔCt method.

Table 1.

Primer sequences for qRT-PCR

Gene name	Forward sequences (5’-3’)	Reverse sequences (5’-3’)
Agtr1b	TTCGTGTTCCCTTTCCTTAT	GTCATCATTTCTTGGCGTGT
Atp2b2	GTGACATGACCAACAGCGACT	GTCTCCTTGATCTTGACCACAG
GAPDH	ACCCATCACCATCTTCCAGG	GACTGTGGTCATGAGCCCTT

Western Blot Analysis

The proteins of the Con, Mob, and Sal_Mob groups (n = 4 per group) were lysed from NRCMs using RIPA buffer, and protein concentrations were then determined by the bicinchoninic acid assay. SDS-PAGE was used to separate the collected proteins, which were then transferred onto PVDF membranes. Following a 1 h blocking step in 5% skim milk, the membranes were probed overnight at 4 °C with primary antibodies against AGTR1B (1:1000; R22571, Zen-bio, China), ATP2B2 (1:500; 370,028, Zen-bio, China), and GAPDH (1:10,000; 10,494–1-AP, Proteintech, China). Subsequently, the membranes were incubated with a rabbit secondary antibody (1:10,000; RGAR001, Proteintech, China) for 1 h. Protein bands were visualized using an ECL chemiluminescent solution. ImageJ software was utilized to quantify the relative protein expression levels, which were first normalized to the GAPDH internal reference.

Statistical Analysis

The results are expressed as mean ± standard deviation. A one-way repeated measures analysis of variance (ANOVA) was conducted for within-group comparisons. One-way ANOVA was employed for comparisons among multiple groups, while Student’s t-test was used for comparisons between two groups. Statistical significance was defined as a P-value < 0.05. To ensure control of the false discovery rate when conducting multiple t-test comparisons, P-values were corrected using the Holm-Bonferroni method (P < α/N), where N represents the number of comparisons. Furthermore, all experimental procedures and electrophysiological analyses were performed in a blinded manner, with investigators unaware of group assignments. All data are presented as the mean ± standard deviation (SD).

Results

Effect of Salidroside on Baseline Cardiac Electrophysiology

Perfusion of isolated rat hearts with salidroside did not substantially alter heart rate, JT interval, or QTc interval at any examined time point. However, the PR interval shortened in a time-dependent manner (Fig. 1C, E–H). Salidroside treatment did not substantially alter cardiac conduction parameters. Specifically, conduction velocity and heterogeneity remained stable. Although conduction time tended to shorten with increased perfusion duration, this trend was not statistically significant (Fig. 1D, I–K).

Mobocertinib-Induced Cardiotoxicity in Isolated Hearts

Compared with controls, mobocertinib caused concentration-dependent reductions in heart rate and prolongation of PR, JT, and QTc intervals. Following washout with drug-free K-H solution, heart rate and PR interval partially recovered; however, JT and QTc prolongation persisted and exhibited progressive deterioration (Fig. 2A–E). Concurrently, mobocertinib dose-dependently prolonged conduction time, reduced conduction velocity, and increased conduction heterogeneity (Fig. 2F–H).

Protective Efficacy of Salidroside Against Mobocertinib-Induced Cardiotoxicity

Following 30 min of salidroside pretreatment, mobocertinib exposure elicited no significant changes in heart rate, PR, JT, or QTc intervals. Although JT and QTc intervals exhibited a slight, concentration-dependent trend toward prolongation, these changes were not statistically significant compared with controls (Fig. 3A, D–G). Furthermore, salidroside pretreatment preserved conduction time and velocity. Although conduction heterogeneity increased with escalating mobocertinib concentrations, this elevation was not statistically significant compared with controls and was significantly attenuated compared with the Mob group (Fig. 3B, C, H–J).

Electrophysiological Effects of Salidroside and Mobocertinib on NRCMs

Immunofluorescence staining confirmed NRCM purity > 95% (Figure S1). Compared with the Con group, the Sal group exhibited no significant changes in FPD, conduction time, velocity, or heterogeneity. Mobocertinib treatment prolonged FPD and conduction time, reduced conduction velocity, and increased conduction heterogeneity. However, salidroside pretreatment normalized these electrophysiological parameters (Fig. 4C–H).

Transcriptomics Analysis

Pearson correlation analysis was performed to assess sample similarity. Intragroup correlation coefficients exceeded 0.9 for all samples, indicating high reproducibility of expression patterns (Fig. 5A). Using a threshold of |log2FoldChange|> 1 and P < 0.05, we identified 298 differentially expressed genes (DEGs) in the Mob group compared with controls (172 upregulated, 126 downregulated). Comparison of the Sal_Mob and Mob groups revealed 50 DEGs (27 upregulated, 23 downregulated). Notably, ATPase plasma membrane Ca₂⁺ transporting 2 (Atp2b2) and angiotensin II type 1 receptor (Agtr1b) were identified as key overlapping DEGs in both comparisons (Fig. 5B, D, E). Hierarchical clustering analysis revealed distinct gene expression signatures across groups (Fig. 5C). Gene Ontology enrichment analysis of Mob vs. Con DEGs highlighted processes including extracellular matrix organization, external encapsulating structure, and multicellular organismal processes (Fig. 5F). Conversely, DEGs from the Sal_Mob vs. Mob comparison were primarily associated with viral defense responses (Fig. 5G). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis identified common enrichment in calcium signaling, aldosterone synthesis and secretion, cGMP-PKG signaling, and adrenergic signaling in cardiomyocytes across both comparisons (Fig. 5H, I). Furthermore, Atp2b2 and Agtr1b are key components of these four signaling pathways.

Fig. 5.

Transcriptomic analysis and validation of key genes. A Correlation analysis of samples in the Con, Mob, and Sal_Mob groups; B DEGs identified in the comparisons of Con vs Mob and Mob vs Sal_Mob; C Heatmap clustering of DEGs across the three groups; D–E Volcano plots of DEGs for Con vs Mob and Mob vs Sal_Mob; F–G GO enrichment analysis of DEGs for Con vs Mob and Mob vs Sal_Mob; H–I KEGG enrichment analysis of DEGs for Con vs Mob and Mob vs Sal_Mob. DEGs identified by DESeq2 (|log2FC|> 1, P < 0.05). Pearson correlation for sample similarity

Experimental Verification of Key DEGs

We validated Agtr1b and Atp2b2 expression via qRT-PCR. Atp2b2 mRNA levels were upregulated in the Mob group but downregulated in the Sal_Mob group. Conversely, Agtr1b expression was downregulated by mobocertinib and restored by salidroside pretreatment (Fig. 6A, B). Consistently, western blot analysis confirmed that the protein expression levels of AGTR1B and ATP2B2 paralleled their mRNA expression patterns across the three groups (Fig. 6C-E).

Fig. 6.

Validation of key genes by qRT-PCR and Western Blot. A, B The expression levels of Agtr1b and Atp2b2 in the Con, Mob, and Sal_Mob groups. C–E The protein expression levels of AGTR1B and ATP2B2 in the Con, Mob, and Sal_Mob groups. Data are presented as mean ± SD. Comparisons among groups were performed using one‑way ANOVA with Tukey’s HSD post‑hoc test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 vs. Con

Discussion

Mobocertinib is a therapeutic agent approved for locally advanced or metastatic non-small cell lung cancer harboring EGFR exon 20 insertion mutations. However, reports of cardiotoxic events have increased, primarily manifesting as QT prolongation. In severe cases, this condition can progress to Torsades de pointes, substantially limiting the clinical application of the drug. Salidroside, the principal bioactive component of Rhodiola rosea, exhibits pleiotropic cardioprotective properties, including antioxidant, anti-inflammatory, and anti-apoptotic effects, and mitigates myocardial ischemia–reperfusion injury. However, its protective efficacy against mobocertinib-induced cardiotoxicity and the underlying molecular mechanisms remain unclear. We systematically investigated the electrophysiological characteristics of mobocertinib-induced cardiotoxicity, alongside the protective efficacy and mechanisms of salidroside, utilizing electrophysiological mapping of isolated hearts and NRCMs, transcriptomic analysis, and molecular validation.

We utilized an isolated heart perfusion model to assess acute drug effects. We evaluated the impact of mobocertinib and salidroside on isolated hearts using four primary parameters: heart rate, PR interval, QTc interval, and JT interval. The QTc interval was calculated using the Fridericia formula (QT/∛(RR)), which corrects for heart rate‑dependent fluctuations in the QT interval and provides more accurate rate correction than the Bazett formula in rodents with high heart rates. Ventricular repolarization abnormalities are typically more closely associated with cardiotoxicity than depolarization changes. Therefore, we utilized the JT interval (QT interval minus QRS duration) to specifically isolate the ventricular repolarization process, providing a more accurate reflection of cardiotoxicity. Additionally, cultured NRCMs were employed to evaluate the effects of prolonged drug exposure. In the isolated heart experiments, statistically significant changes in electrophysiological parameters were observed only at the 428 nM concentration of mobocertinib. Therefore, to balance the observation of pharmacological effects with cell viability, we selected this medium concentration for the NRCM experiments, thereby avoiding insufficient toxicity at lower concentrations and non‑specific cellular damage at higher concentrations that could interfere with the assessment of protective mechanisms. Findings from both models demonstrated that mobocertinib dose-dependently impaired electrophysiological function: it prolonged PR, JT, and QTc intervals in isolated hearts, prolonged FPD in NRCMs, and increased conduction time and heterogeneity in both systems. Furthermore, several of these electrophysiological alterations persisted after drug washout, indicating the induction of irreversible myocardial electrical remodeling. In contrast, salidroside pretreatment significantly attenuated mobocertinib-induced bradycardia and the prolongation of FPD, PR, JT, and QTc intervals, while also preserving conduction function. Compared with the Mob group, the Sal_Mob group exhibited significantly shorter conduction times, faster conduction velocities, and reduced conduction heterogeneity, confirming that salidroside protects against mobocertinib-induced myocardial electrical disturbances.

Given that the isolated heart perfusion model represents acute drug exposure, alterations in mRNA levels are typically limited, with electrophysiological changes primarily attributed to direct ion channel modulation. Therefore, to comprehensively elucidate the molecular mechanisms underlying mobocertinib-induced cardiotoxicity, particularly the long-term transcriptional regulatory mechanisms beyond ion channels, we performed transcriptomic sequencing on neonatal rat cardiomyocytes subjected to prolonged drug treatment. Transcriptome analysis indicates pathways related to calcium signaling, aldosterone synthesis and secretion, cGMP-PKG signaling, and adrenergic signaling in cardiomyocytes were significantly enriched in both DEG comparisons. Notably, Atp2b2 and Agtr1b, key components of these enriched pathways, exhibited inverse expression patterns in the Mob and Sal_Mob groups. These findings suggest that these pathways do not function in isolation; rather, they collectively regulate cardiomyocyte electrophysiological homeostasis through a complex interactive network. Consequently, they may serve as key regulatory targets in the context of mobocertinib-induced cardiotoxicity and the cardioprotection observed with salidroside co-treatment. Our qRT-PCR and western blot results further validated the transcriptomic findings, providing robust evidence for the molecular networks mediating mobocertinib toxicity and salidroside rescue.

Mobocertinib-induced cardiotoxicity is closely associated with its direct effects on cardiomyocyte electrophysiology. Under acute conditions, mobocertinib dose-dependently inhibits the hERG potassium channel (IC₅₀ 4.96 μM) and the Cav1.2 calcium channel (IC₅₀ 12.60 μM), while suppressing the Nav1.5 sodium channel and the Kv4.3 transient outward potassium channel at higher concentrations[9]. This multi-channel blockade not only prolongs the QT interval but also accounts for PR prolongation and reduced conduction velocity by impairing atrioventricular and ventricular conduction. Furthermore, phosphoproteomic analysis indicates that mobocertinib suppresses lysine deficient protein kinase 1 kinase activity, leading to the downregulation of hERG and KCNQ1 potassium channel expression and the consequent prolongation of action potential duration [10]. This mechanism, distinct from direct channel blockade, likely contributes to the long-term, irreversible electrophysiological toxicity of the drug. Thus, mobocertinib induces cardiotoxicity not only through acute channel blockade but also by disrupting ion channel protein homeostasis, resulting in persistent electrical remodeling.

In the present study, salidroside significantly alleviated mobocertinib-induced electrophysiological abnormalities. The cardioprotective effects of salidroside have been demonstrated across various pathological models. Regarding oxidative stress and apoptosis, salidroside attenuates doxorubicin-induced oxidative damage by inhibiting NADPH oxidase 1 expression and enhancing catalase and manganese superoxide dismutase activities; concurrently, it upregulates the anti-apoptotic protein B-cell lymphoma 2 (Bcl-2), downregulates Bcl-2-associated X protein, and inhibits caspase-3 activity, thereby preventing cardiomyocyte apoptosis [16]. Furthermore, salidroside alleviates doxorubicin-induced cardiac dysfunction by activating the QKI/FoxO1 pathway [17]. Regarding calcium homeostasis, proteomic analyses reveal that salidroside inhibits mitochondrial calcium overload in hypoxic cardiomyocytes and improves mitochondrial energy metabolism via the EGLN1/HIF-1α pathway, thereby attenuating hypoxia-induced apoptosis [18]. In the present study, salidroside reversed the mobocertinib-induced upregulation of Atp2b2, which encodes a key pump protein responsible for calcium extrusion; its upregulation likely reflects a compensatory response to intracellular calcium overload. Therefore, restoring calcium homeostasis may represent a pivotal mechanism associated with the preservation of electrophysiological function observed in the Sal_Mob group.

In addition to calcium regulation, neurohumoral regulation plays a crucial role in cardiac electrical remodeling. Agtr1b encodes the angiotensin II type 1b (AT1b) receptor, a core effector molecule of the renin–angiotensin–aldosterone system [19]. Upon activation, the AT1b receptor couples to Gq/11 proteins, initiating the phospholipase C/inositol trisphosphate cascade; this triggers cytosolic calcium release, mediates vasoconstriction and enhanced myocardial contractility, and increases aldosterone synthesis and secretion [20]. In the present study, Agtr1b expression was downregulated in the mobocertinib-treated group, a response potentially reflecting negative feedback following prolonged receptor activation. This downregulation may impair myocardial responsiveness to neurohumoral regulatory signals, exacerbating electrophysiological instability in the context of calcium overload. Therefore, future mechanistic investigations must account for the complex dynamics of neurohumoral regulation. The cGMP-PKG signaling cascade is a well-established cardioprotective pathway. Activation of this pathway attenuates myocardial reperfusion injury by regulating intracellular pH recovery, inhibiting calcium oscillations, and preventing the opening of the mitochondrial permeability transition pore opening [21]. Furthermore, cGMP-PKG-mediated cardioprotection involves the modulation of sarcoplasmic/endoplasmic reticulum calcium ATPase 2a function and ryanodine receptor 2 stability; conversely, inhibiting this pathway impairs calcium cycling and promotes proarrhythmic calcium leak [22].

Although transcriptomic analysis identified Atp2b2 and Agtr1b as key DEGs, their functional roles have not been directly validated via gene knockout or overexpression models. Future functional experiments such as siRNA/CRISPR-mediated knockdown or adeno-associated virus (AAV)-mediated overexpression will need to be conducted to clarify their specific roles in cardiac electrophysiology. Additionally, the present study did not directly measure salidroside’s effects on specific ion channel currents using patch-clamp techniques. Whether salidroside directly reverses mobocertinib-induced inhibition of hERG, Cav1.2, Nav1.5, or other channels remains to be determined. Future patch-clamp studies in isolated cardiomyocytes or heterologous expression systems are warranted to elucidate the direct ion channel targets of salidroside in the context of mobocertinib-induced cardiotoxicity. Furthermore, mobocertinib was dissolved in dimethyl sulfoxide (DMSO), whereas salidroside was dissolved in ultrapure water. A separate vehicle control group was not included. However, the final concentration of DMSO in the perfusate was less than 0.01% for all mobocertinib concentrations used, and such low concentrations of DMSO have been widely reported to have no significant toxicity to cardiac muscle cells [23].

In summary, mobocertinib induces myocardial electrophysiological abnormalities by disrupting calcium homeostasis and neurohumoral regulation. Salidroside pretreatment effectively reverses these alterations, providing robust cardioprotection. This protective effect observed with salidroside co-treatment involves the interactive regulation of multiple signaling pathways, with Atp2b2 and Agtr1b serving as key molecular targets. These findings provide a theoretical foundation and novel therapeutic strategy for preventing mobocertinib-associated cardiotoxicity, establishing a basis for future mechanistic studies.

Conclusion

This study demonstrates that mobocertinib induces myocardial electrophysiological abnormalities by disrupting calcium homeostasis and neurohumoral regulation, leading to irreversible QTc and JT prolongation alongside conduction disturbances. Salidroside pretreatment is associated with reversal of these cardiotoxic effects, and this protective effect involves the modulation of calcium signaling, aldosterone synthesis, cGMP-PKG, and adrenergic pathways. Notably, Atp2b2 and Agtr1b act as key regulatory genes within these networks. These findings highlight a novel therapeutic strategy for preventing mobocertinib-associated cardiotoxicity and establish a foundation for future targeted research.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

CY wrote the main manuscript text and prepared all figures. XHL, XYZ, PW, XFL, DXL, GHZ, and ML performed the experiments, data curation, and formal analysis. XYZ, GJL, and FL supervised the project, acquired funding, and revised the manuscript. All authors reviewed the manuscript.

Funding

This work was supported by Key Research and Development and Promotion Project of Henan Province (No. 252102311285), Research Projects of Higher Education Institutions in Henan Province (No. 25B360015), Natural Science Foundation of Henan Province (No. 252300420157) and Key Research and Development and Promotion Project of Henan Province (No. 232102310425).

Data Availability

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request and the transcriptomic data have been deposited in the NCBI database.

Declarations

Competing Interests

The authors declare that they have no conflict of interest.