Comprehensive Analysis of Differences in N6-Methyladenosine RNA Methylation Groups in CVB3-Induced Viral Myocarditis and Identification of the Anti-Apoptotic Role of RBM15B

Introduction

Viral myocarditis (VMC) is a myocardial inflammatory disorder triggered by infection with cardiotropic viruses, often resulting in acute or chronic cardiac damage.^1–3 This condition arises primarily from adverse immune responses to viral infection, accompanied by myocardial injury, reparative fibrosis, and progression to heart failure during viral clearance. Among the various viral pathogens implicated, Coxsackievirus B3 (CVB3), a positive-sense single-stranded RNA virus, is the most frequently identified causative agent.^4,5 Despite extensive research efforts, effective antiviral therapies for VMC are still unavailable. This underscores the critical need for deeper investigation into the molecular mechanisms underlying CVB3-induced myocardial damage, which could pave the way for novel therapeutic approaches.

Recent studies emphasize the role of epigenetic regulation in inflammation, particularly through RNA methylation modifications. N⁶-Methyladenosine (m⁶A), the most common post-transcriptional modification in eukaryotic RNA, is regulated by “writers”, “erasers”, and “readers”.^6–9 Methyltransferases like METTL3, METTL14, and WTAP act as “writers” to add methyl groups, while demethylases such as ALKBH5 and FTO serve as “erasers” to remove them. “Readers”, including YTH domain proteins (eg, YTHDC1, YTHDF2), recognize m⁶A sites to influence RNA processing.¹⁰ m⁶A regulates key mRNA processes such as splicing, stability, and translation, playing a vital role in mRNA fate.^11–15 m⁶A modifications have been linked to inflammatory diseases like pulmonary inflammation, colitis, and coronary artery disease,^16–18 as well as to viral RNA regulation in HIV, Zika virus, and SARS-CoV-2.^19–22 However, the function and regulatory landscape of m6A modifications in CVB3-induced VMC remain largely unexplored, representing a major knowledge gap in the field.

In this research, high-throughput MeRIP-seq was utilized to map m⁶A modifications in HL-1 cardiomyocytes infected with CVB3. Analyses of differentially methylated genes (DMGs), differentially expressed genes (DEGs), and differentially methylated and expressed genes (DMEGs) were conducted, with gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment highlighting their functional significance. Furthermore, mouse mRNA Sequencing (RNA-Seq) data was employed to confirm the expression patterns of m⁶A-regulated genes and critical targets through qRT-PCR, revealing consistency across in vitro and in vivo models. Building upon these analytical findings, we hypothesize that m⁶A modifications contribute to CVB3-induced myocardial damage, with RBM15B playing a regulatory role in apoptosis, and subsequently conducted systematic experimental validations. Altogether, our findings provide the first comprehensive characterization of m⁶A epitranscriptomic alterations during CVB3 infection, highlighting the central role of m⁶A methylation in mediating viral myocarditis. This work offers new molecular insights into VMC progression and suggests potential therapeutic avenues targeting RNA methylation machinery.

Materials and Methods

Cell Culture

HeLa cells (obtained from the Institute of Cancer Research, Central South University, China) were grown in DMEM (Gibco, CA, USA) supplemented with 10% FBS (Biological Industries, Israel) and antibiotics (100 U/mL penicillin and 100 µg/mL streptomycin; Life Technologies, NY, USA). HL-1 cells (Merck, Darmstadt, Germany) were cultured in Claycomb medium (Sigma-Aldrich, MO, USA), enriched with 10% FBS, 1% antibiotics, 0.1 mm norepinephrine, and 2 mm glutamine (all from Sigma-Aldrich). All cells were cultured in a humidified incubator set at 37°C with 5% CO₂.

Virus and Infection

The CVB3 Nancy strain (obtained from Shanghai Jiaotong University School of Medicine, China) was amplified in HeLa cells and preserved at −80°C. Viral titers were determined using plaque assays. For infection experiments, HL-1 cells were seeded in 6-well plates one day prior and serum-starved overnight in medium containing 2% FBS. Then infect the cells with CVB3 (MOI=10, based on the optimal infection conditions established through prior studies). Following a 1-hour incubation, the cells were washed with PBS and maintained in fresh 2% FBS medium for an additional 48 hours. Control cells (Sham group) were exposed to 2% FBS medium without viral infection. Supernatants were collected post-culture and preserved at −80°C for later plaque assay analysis.

Animal Models

Male BALB/c mice (4 weeks old, 18–20 g, SPF-grade) were purchased from the Laboratory Animal Center of Central South University and housed under SPF conditions. Following a 1-week acclimation period, all mice were divided into different groups based on complete randomization. The CVB3 group received intraperitoneal injection of CVB3 as reported in literature²³ (10³ TCID₅₀, diluted in PBS), while Sham group received an equal volume of saline. Mice were observed daily for changes in appearance, behavior, and survival status. On day 7 post-infection, they were euthanized via cervical dislocation, and collected blood samples and cardiac tissues. Mice that either succumbed during the experiment or showed no evidence of myocardial inflammatory cell infiltration in the CVB3 group were excluded from the study. All procedures involving animals were approved by the Central South University Animal Experimentation Management Committee (Approval No. 2021sydw0104) and adhered to ethical standards to minimize animal suffering.

Hematoxylin and Eosin (HE) Staining

The heart tissues were preserved in 4% paraformaldehyde, embedded in paraffin, and cut into 5 µm sections. These sections underwent deparaffinization in xylene, rehydration through a graded series of ethanol, and staining with HE. Finally, the stained sections were mounted with neutral balsam and examined using an optical microscope.

ELISA Assay

The levels of CK-MB and cTnT were conducted using ELISA kits provided by JINGMEI Biotechnology (Jiangsu, China). The assays were conducted following the manufacturer’s protocol, and absorbance measurements were obtained using a Multilabel Plate Reader (PerkinElmer, MA, USA).

Immunofluorescence (IF) Assay

Paraffin-embedded tissue sections underwent dewaxing, rehydration, and antigen retrieval, followed by overnight incubation at 4°C with primary antibody targeting CVB3 (1:100). Cells were fixed in 4% paraformaldehyde, then treated with 0.5% Triton X-100 for permeabilization and incubated with 5% goat serum for blocking (2 hours). Primary antibodies against FOXO1 (1:100) were applied and incubated overnight at 4°C. Subsequently, Secondary antibodies conjugated to Alexa Fluor 594 or 488 (1:200) were applied for 1 hour, followed by DAPI staining (125 µg/mL) for 15 minutes. Visualization was carried out using a Zeiss fluorescence microscope (Jena, Germany). The antibodies in this study are provided in Supplementary Table S2.

Viral Plaque Assay

Culture media underwent serial dilutions and were applied to HeLa cells plated in six-well dishes, each well containing 1 × 10⁶ cells. After an hour of incubation, cells were washed by PBS and overlaid with 2% soft agar. The cultures were then incubated for 72 hours. Viral plaques were visualized using the VisionWorks system (Analytik Jena AG, Jena, Germany), and the titers were quantified in PFU/mL.

Cell Proliferation and Cytotoxicity Assay

Cell proliferation and cytotoxicity were assessed with the Calcein AM/PI Double Stain Kit (MKBio, Shanghai, China). HL-1 cells were incubated with Calcein-AM (2 µM) and PI (4.5 µM) at 37°C for 10 minutes. Fluorescence imaging was performed and recorded using a Zeiss fluorescence microscope.

RNA Extraction and Quantitative Real-Time PCR (RT-PCR)

Total RNA was extracted with TRIzol reagent (Invitrogen) and quantified using the NanoDrop 2000 (Thermo Scientific, MA, USA). cDNA was synthesized with HiScript III RT SuperMix (Vazyme, Nanjing, China). qRT-PCR was conducted on a LightCycler 480II system (Roche, Basel, Switzerland), with β-actin as the normalization control. Primer sequences are provided in Supplementary Table S1.

Western Blotting

Myocardial tissues and cells were lysed in RIPA buffer (Beyotime, Shanghai, China) supplemented with protease and phosphatase inhibitors (Roche). The homogenates were separated by SDS-PAGE (8–12%) and transferred onto PVDF membranes (Millipore, CA, USA). Membranes were first blocked with 5% non-fat milk, then incubated overnight at 4°C with primary antibodies, and subsequently exposed to secondary antibodies at room temperature for 1 hour. Protein bands were visualized utilizing an ECL detection kit (Biosharp, Anhui, China), and the bands were subsequently analyzed with the VisionWorks imaging system. Antibodies are detailed in Supplementary Table S2.

m⁶A Dot Blot Analysis

A total of 1 µg of RNA was denatured at 65°C for 10 minutes, then cooled on ice before being transferred onto nylon membranes. After UV cross-linking for 10 minutes and blocking with 5% non-fat milk, the membranes underwent overnight incubation at 4°C with the m⁶A antibody (1:600, HA601049). Afterward, a secondary antibody conjugated with HRP (1:2000) was applied for 1 hour. Protein bands were then detected using a chemiluminescence imaging system. Equal loading was confirmed by staining membranes with 0.02% methylene blue for 30 minutes, followed by rinsing and photographing.

RNA-Seq and m⁶A-RNA Immunoprecipitation Sequencing (m⁶A-Seq)

Total RNA from mouse heart tissue and HL-1 cells was extracted using TRIzol. Poly(A) RNA was extracted from 50 µg of total RNA through two rounds of purification using Dynabeads Oligo(dT) (Thermo Fisher). The quality of the RNA was assessed using a Bioanalyzer 2100 (Agilent), and ribosomal RNA was eliminated through the use of the Epicentre Ribo-Zero Gold Kit (Illumina). Fragmented RNA underwent m⁶A immunoprecipitation with Dynabeads and an m⁶A antibody (Synaptic Systems, Germany), followed by cDNA synthesis for sequencing.

Bioinformatics Analysis

Raw data quality was assessed using Fastp (https://github.com/OpenGene/fastp) and aligned to the Mus musculus genome with HISAT2 (http://daehwankimlab.github.io/hisat2). Differential peak analysis was performed using exomePeak (https://bioconductor.org/packages/release/bioc/html/exomePeak.html), and annotations were completed with ANNOVAR (http://www.openbioinformatics.org/annovar/). Motif analysis utilized MEME2 (http://meme-suite.org) and HOMER (http://homer.ucsd.edu/homer/motif). Gene expression was quantified with StringTie (https://ccb.jhu.edu/software/stringtie), and differential expression was analyzed using edgeR (https://bioconductor.org/packages/edgeR) with thresholds of fold change (FC) ≥ 2 or ≤ 0.5 and p-value < 0.05.

Statistical Analysis

All experiments were conducted in triplicate to ensure consistency, and the data are presented as mean ± standard deviation (S.D). Statistical analyses were performed using SPSS software (IBM SPSS 26.0). Two-tailed independent t-tests were employed for comparisons between two groups with normally distributed data. For comparisons between multiple groups, one-way ANOVA with Tukey’s post hoc test was applied. For non-normally distributed data, the Mann–Whitney U-test was used. Statistical significance was defined as p < 0.05.

Results

Increased m⁶A Levels in CVB3-Induced Viral Myocarditis

To explore the mechanisms underlying CVB3-induced viral myocarditis, a mouse model was established via intraperitoneal injection of 10³ TCID₅₀ CVB3. CVB3-infected mice exhibited significant weight loss, smaller heart sizes, and irregular myocardial surfaces (Figure 1A and B). Serum CK-MB and cTnT levels were markedly higher compared to Sham group (Figure 1C). HE staining revealed intense inflammatory infiltration in myocardial fibers and perivascular regions (Figure 1D), while immunofluorescence confirmed substantial CVB3 presence in myocardial tissues (Figure 1E), validating the model. Notably, myocardial m⁶A methylation levels were significantly elevated in CVB3-infected mice (Figure 1G). A similar increase was observed in HL-1 cells, with peak m⁶A levels at 48 hours post-infection (Figure 1F). HL-1 cells infected with CVB3 (MOI = 10) for 48 hours were subsequently analyzed by sequencing.

Figure 1 Establishment of CVB3 infection model and determination of total m6A. (A) Body weights of Sham and CVB3 groups; n = 6. (B) General morphology of the hearts of Sham and CVB3 groups. (C) Comparison of serum myocarditis inflammatory markers CK-MB and cTnT between Sham and CVB3 groups; n = 6. (D) Representative images of HE staining of mouse myocardium from Sham and CVB3 groups (scale bar = 50 μm); n = 6. (E) IF staining of CVB3 (red) in mouse myocardial tissue in the Sham and CVB3 groups, blue staining indicates DAPI nuclear staining (scale bar = 50 μm); n = 6. (F) m6A dot blot analysis of myocardial tissues from Sham and CVB3 groups of mice; n=3. (G) m6A dot blot analysis of Sham and CVB3 groups of HL-1 cell; n=3. *p < 0.05, **p < 0.01, ****p < 0.0001.

Overview of m⁶A RNA Immunoprecipitation Sequencing in CVB3-Infected HL-1 Cells

In the MeRIP-seq analysis, the CVB3 and Sham groups produced an average of 90,779,293 and 90,038,791 valid reads, respectively, while RNA-seq yielded averages of 89,876,025 and 91,116,208 reads (Supplementary Table S3). Valid read mapping rates were 95.04% (Sham) and 94.47% (CVB3) for immunoprecipitated (IP) samples, and 91.87% (Sham) and 87.82% (CVB3) for input samples (Supplementary Table S4). Clean reads predominantly mapped to exonic regions, accounting for 92.68% (Sham) and 93.05% (CVB3) of total reads (Figure S1). m⁶A modifications were enriched near the transcription start site (TSS) and transcription end site (TES) (Figure 2A). Differential m⁶A peaks were primarily located in the 3′UTR (Sham: 32.5%; CVB3: 33.5%) and coding sequence (CDS) regions (Sham: 32%; CVB3: 32.9%) (Figure 2B and C), with similar peak distributions across groups (Figure 2D). A total of 327 significant differential peaks were identified, with 116 showing upregulation and 211 showing downregulation, based on criteria of |log2(FC)| > 1 and p-value < 0.05 (Figure 2E). Both groups exhibited the characteristic m⁶A motif 5′-RRACH-3′ (where R = A or G; H = A, C, or U) (Figure 2F).

Figure 2 Significant dysregulation of m6A peaks during CVB3 infection of HL-1 cells. (A) Enrichment of peaks near gene transcription start sites. (B, C and D) Distribution of differentially methylated m6A peaks in Sham and CVB3 groups. (E) Volcano plot of genes with differential m6A peaks (|log2(FC)|>1 and p-value < 0.05). (F) Motif map of m6A modifications in the Sham and CVB3 groups.

Differential m⁶A Modifications are Associated with Key Biological Pathways

GO and KEGG enrichment analyses highlighted the functional roles of m⁶A-modified genes in CVB3-infected HL-1 cells. GO analysis showed that hypermethylated genes were primarily associated with the “membrane” in the cellular component category, while hypomethylated genes were enriched in the “nucleus” (Figure 3A and B). Both groups were linked to biological processes such as DNA transcription regulation, RNA polymerase II activity regulation, and cellular differentiation. For molecular functions, “protein binding”, “RNA binding”, and “metal ion binding” were common associations for both hyper- and hypomethylated genes (Figure 3A and B). KEGG pathway analysis revealed distinct patterns: upregulated m⁶A-modified genes were significantly enriched in pathways such as “virus infection”, “tryptophan metabolism”, “neuroactive ligand-receptor interaction”, and “phenylalanine metabolism” (Figure 3C). Conversely, downregulated genes were predominantly associated with “Regulation of actin cytoskeleton”, “cAMP signaling pathway”, and “Notch signaling pathway” (Figure 3D).

Figure 3 Differential m6A modifications are involved in important biological pathways. (A) GO enrichment analysis of the hypermethylated peaks. (B) GO enrichment analysis of the hypomethylated peaks. (C) KEGG pathway analysis of the hypermethylated peaks. (D) KEGG pathway analysis of the hypomethylated peaks.

RNA-Seq Differential Gene Expression Analysis

To investigate the connection between m⁶A modifications and gene expression, RNA-seq was performed. Hierarchical clustering revealed distinct gene expression profiles between the Sham and CVB3 groups (Figure 4A). A total of 4539 DEGs were detected, comprising 1597 upregulated and 2942 downregulated genes (|log2(FC)| > 1 and p-value < 0.05) (Figure 4B). GO analysis revealed that these DEGs were primarily associated with terms such as “membrane”, “calcium ion binding”, and “cell adhesion” (Figure 4C). Moreover, KEGG pathway analysis demonstrated significant enrichment in pathways such as “PI3K-Akt signaling”, “MAPK signaling”, “cytokine-cytokine receptor interaction”, and “calcium signaling” (Figure 4D). These findings indicate that m⁶A modification is closely linked to pathways essential for cellular signaling, inflammation, and viral infection processes.

Figure 4 Differential expression gene analysis by RNA-Seq. (A) Heatmap showing mRNAs differentially expressed in three CVB3 group input samples and three Sham group input samples. (B) Volcano plot showing mRNAs differentially expressed between CVB3 and Sham groups with statistical significance (|log2(FC)| ≥1.0 and p < 0.5). (C) GO enrichment analysis of differential genes. (D) KEGG pathway analysis of differential genes.

Combined Analysis of m⁶A-Seq and RNA-Seq

To investigate the interplay between gene expression and m⁶A modifications, RNA-seq and m⁶A-seq datasets were integrated. Genes were categorized based on their m⁶A modification and expression changes: 38 genes exhibited hypermethylation with upregulation (hyper-up), 23 showed hypermethylation with downregulation (hyper-down), 65 displayed hypomethylation with downregulation (hypo-down), and 13 were hypomethylated with upregulation (hypo-up) (Figure 5A). KEGG pathway analysis revealed the following: Hypo-up genes were significantly enriched in pathways such as “glycerophospholipid metabolism”, “virus infection”, “TNF signaling pathway”, and “Ras signaling pathway” (Figure 5B). Hyper-down genes were associated with “complement and coagulation cascades” and “fluid shear stress and atherosclerosis” (Figure 5C). In addition, hypo-up genes were mainly enriched in “Cytokine-cytokine receptor interaction”, “PI3K-Akt signaling pathway”, “MAPK signaling pathway” and “FoxO signaling pathway” (Figure 5D). Hypo-down genes were linked to pathways like “axon guidance”, “Wnt signaling pathway”, and “calcium signaling pathway” (Figure 5E). Additionally, Table 1 lists the top 20 transcripts exhibiting both differential m⁶A modifications and altered mRNA expression.

Table 1 Top 20 Transcripts of Differential m6A Modification and mRNA Expression Between Sham Group and CVB3 Group

Figure 5 Combined analysis between m6A-Seq and RNA-Seq. (A) Four-quadrant diagram of hyper-up, hyper-down, hypo-up, and hypo-down DMEG. (B) KEGG pathway analysis of hyper-up DMEG (C) KEGG pathway analysis of hyper-down DMEG. (D) KEGG pathway analysis of hypo-up DMEG (E) KEGG pathway analysis of hypo-down DMEG.

Protein-Protein Interaction (PPI) Network Analysis

A PPI network was generated for the 139 DEGs linked to m⁶A modifications using the STRING database (Figure S2A). The top 5 hub genes in the network, identified using the CytoHubba plugin in Cytoscape, were Sqstm1, Ar, Notch3, Hspa1b, and Col4a1 (Figure S2B).

Validation of Differential Genes

Comparative analysis of the in vivo and in vitro RNA-seq results revealed that 15 genes were upregulated and 1 gene downregulated in both models (Figure 6A). Genes with significant changes (|log2(FC)| > 2.5 and p-value < 0.05) included Igtp, ApoI9b, Ddit3, and Irgm3 (Figure 6B). qRT-PCR validation demonstrated that ApoI9b expression in the myocardium of the mouse model differed from the RNA-seq results, while the other genes were consistent (Figure 6C). In the HL-1 cardiomyocyte model, expression aligned with the sequencing results (Figure 6D). These findings confirm the accuracy of our sequencing data. Additionally, IGV visualization of these genes showed significant differences in m⁶A levels (Figure 6E).

Figure 6 Validation of differentially expressed genes. (A) Wayne plots of in vivo and in vitro RNA-seq sequencing results. (B) Volcano plot showing mRNAs that were differentially expressed simultaneously after comparison of in vivo and in vitro RNA-seq sequencing results with statistical significance (|log2(FC)| ≥ 1.0 and p-value < 0.5). (C and D) In vivo and in vitro validation of differentially expressed genes. (E) IGV visualization showing three key m6A-modified genes. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Abbreviation: ns, no statistical significance.

Validation of m⁶A-Related Differential Genes

The expression levels of 36 m⁶A regulators were evaluated using mouse RNA-seq data (Figure 7A). qRT-PCR validation of 10 common regulators in both in vivo and in vitro models revealed significant changes in the expression of IGF2BP2, EIF3H, RBM15B, and YTHDC2 (Figure 7B and C). The trends observed in qRT-PCR were consistent with the RNA-seq data. Notably, RBM15B, which exhibited the most pronounced change in expression, was further validated by protein analysis. Consistent with qRT-PCR results, CVB3 infection led to increased RBM15B expression both in vivo (Figure 7D and E) and in vitro (Figure 7F and G).

Figure 7 Validation of m6A-associated differentially expressed genes (A) Heatmap of m6A regulators in sequenced mouse samples. (B and C) qRT- PCR results of 10 common m6A regulators in vivo and in vitro. (D and E) Analysis of RBM15B protein levels in mouse samples from Sham and CVB3 groups by Western blotting. (F and G) Analysis of RBM15B protein levels in HL-1 cell samples from Sham and CVB3 groups by Western blotting. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Abbreviation: ns, no statistical significance.

Knockdown of RBM15B Ameliorates CVB3 Infection in HL-1 Cardiomyocytes

To explore the function of RBM15B during CVB3 infection, HL-1 cells were transfected with si-RBM15B or a non-targeting control siRNA (si-NC) prior to infection with CVB3. Plaque assays revealed a significant reduction in viral replication in the si-RBM15B group compared to the si-NC group (Figure 8D). Calcein AM/PI double staining showed decreased cytotoxicity in the si-RBM15B group (Figure 8E). Analysis of apoptotic proteins showed that RBM15B knockdown led to a significantly reduced Bax/Bcl-2 ratio and lower Cleaved-Caspase3 levels compared to the control (Figure 8A–C). These findings indicate that silencing RBM15B mitigates CVB3-induced apoptosis in HL-1 cardiomyocytes.

Figure 8 Knockdown of RBM15B ameliorates HL-1 cell apoptosis caused by CVB3 infection. (A–C) After transfection of HL-1 cells with si-RBM15B or si-NC, the cells were harvested under the same infection conditions and analyzed by Western blotting to determine the expression of RBM15B, cleaved-PARP, Bax/Bcl2, and cleaved-caspase3; n = 3. (D) Cultures from the si-NC group and si- RBM15B group were collected 48 h after infection with CVB3, respectively, and viral plaque assay was performed using Hela cells; n = 3. (E) Calcein AM/PI double staining assay was used to assess the cytotoxicity of HL-1 cells after transfection with si-NC and si-RBM15B at CVB3 infection (48 h) (scale bar = 50 μm). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Abbreviation: ns, no statistical significance.

Discussion

CVB3, belonging to the enterovirus genus and characterized by a positive-sense RNA genome, is the primary pathogen responsible for VMC.²⁴ VMC is characterized by viral-induced myocardial injury followed by a persistent inflammatory response.²⁴ Previous studies have suggested that acute myocarditis induced by CVB3 results from direct viral damage to myocardial cells, which is subsequently followed by inflammation.²⁵ m⁶A, a reversible RNA modification, plays a critical role in regulating gene expression and has been implicated in inflammatory responses associated with pulmonary inflammatory diseases, spontaneous colitis, and coronary artery disease.^16–18 Furthermore, m⁶A modifications have been identified in viral RNA transcripts, including those of Rous sarcoma virus, influenza virus, simian virus 40, avian sarcoma virus, adenovirus, and coronaviruses.^22,26–28 However, the distribution, functional roles, and regulatory mechanisms of m6A in CVB3-induced viral myocarditis remain unexplored. This study focuses on elucidating the impact of m⁶A modifications on the progression of CVB3-induced VMC.

We observed that CVB3 infection significantly increased m⁶A methylation in both in vitro and in vivo models. MeRIP-seq analysis revealed distinct m⁶A modification profiles between CVB3-infected and Sham controls, identifying 327 differentially methylated peaks—116 upregulated and 211 downregulated. These modifications were primarily enriched in the 3′ UTR and CDS regions, which aligns with previous studies linking m⁶A modifications to mRNA stability, localization, and translation.²⁹ Notably, m⁶A modifications were linked to key biological pathways.³⁰ Upregulated m⁶A peaks were enriched in pathways associated with “viral infection”, “tryptophan metabolism”, and “neuroactive ligand-receptor interactions”, while downregulated peaks were involved in pathways such as “actin cytoskeleton regulation”, “cAMP signaling”, and “Ras signaling.” These pathways are critical for viral replication, inflammation, and autophagy, all of which play pivotal roles in CVB3-induced myocarditis.^31–33 These results indicate that m⁶A modifications are not randomly distributed but functionally associated with critical signaling cascades relevant to CVB3 pathogenesis.

To investigate how m⁶A modifications influence CVB3 infection in HL-1 cells, we integrated m⁶A methylation patterns with transcription factor motifs to uncover key signaling pathways involved. As in previous studies, pathways like PI3K-Akt, MAPK, Rap1, and calcium signaling were significantly enriched.^34–36 m⁶A modifications have been shown to regulate gene expression in a context-dependent manner.^37–39 Therefore, we categorized the DMEGs into four groups: hyper-up, hyper-down, hypo-up, and hypo-down. KEGG analysis revealed that hypo-up genes were enriched in pathways related to “glycerophospholipid metabolism”, “viral infection”, and “TNF signaling”, whereas hyper-down genes were associated with “complement and coagulation cascades” and “Th17 cell differentiation.” These findings highlight the regulatory complexity of m⁶A in modulating transcriptional responses during CVB3 infection and its potential role in orchestrating both immune and metabolic pathways.

We analyzed both in vivo and in vitro sequencing data and identified four genes—Igtp, ApoI9b, Ddit3, and Irgm2—that exhibited significant and consistent expression changes. qRT-PCR validation confirmed the RNA-seq findings for all genes, except for ApoI9b, in myocardial tissues from CVB3-infected mice. CVB3 infection notably upregulated the mRNA levels of Igtp, Ddit3, and Irgm2 in both models. Igtp and Irgm2, key regulators of interferon (IFN) signaling, are part of the immune-related GTPase family, which plays a crucial role in resistance to intracellular pathogens.^40,41 Specifically, Igtp is involved in vesicular trafficking for pathogen clearance, while Irgm2 may regulate autophagy during innate immune responses.^42,43 Previous research has shown that Igtp can inhibit CVB3-induced apoptosis in cardiomyocytes by activating the PI3K/Akt signaling pathway.⁴⁴ However, the role of Irgm2 in CVB3-induced diseases remains unexplored. DDIT3 (also known as GADD153 or CHOP) is a transcription factor activated by stress, including ER stress, and regulates cell cycle arrest and apoptosis—processes central to many diseases.^45–47 Previous studies have established that CHOP upregulation due to ER stress contributes to CVB3-induced VMC by promoting cardiomyocyte apoptosis.⁴⁸ Additionally, IGV analysis demonstrated alterations in m⁶A peak distributions in Igtp, Ddit3, and Irgm2, suggesting that m⁶A modifications may participate in regulating their expression and functional responses to viral infection.

Additionally, we analyzed the expression of 36 m⁶A regulators using RNA-seq data, followed by validation through qRT-PCR. Significant changes were observed in the expression of key regulators, including IGF2BP2, EIF3H, RBM15B, and YTHDC2. Among these regulators, IGF2BP2, EIF3H, and YTHDC2 function as m⁶A readers, while RBM15B serves as an m⁶A writer. IGF2BP2 is notably associated with signaling pathways such as PI3K/Akt,⁴⁹ TNF/NF-κB,^50,51 and FOXO.⁵² EIF3H is linked to the Wnt⁵³ and Hippo signaling pathways,⁵⁴ whereas YTHDC2 participates in pathways including TNF,⁵⁵ Akt,⁵⁶ NF-κB,⁵⁷ and Hippo.⁵⁸ Notably, previous studies have highlighted the close connection between CVB3 infection and these pathways. The PI3K/Akt signaling pathway, known for its roles in cell proliferation, apoptosis, and protein synthesis, is implicated in CVB3 infection.⁵⁹ Similarly, the MAPK signaling pathway is activated by CVB3, promoting inflammation, apoptosis, and viral replication in host cells.^35,60 The TNF signaling pathway contributes to inflammation, immune responses, and apoptosis, playing a pivotal role in creating an inflammatory environment and inducing cellular damage during CVB3 infection.^61,62 Furthermore, CVB3 has been shown to suppress the activation of NLRP3 inflammasomes by inhibiting the NF-κB signaling pathway, thereby enhancing host susceptibility to infection.⁶³ These findings suggest that m⁶A regulators may play a critical role in modulating these signaling pathways during VMC.

Intriguingly, our analysis of in vivo and in vitro RNA-seq data on m⁶A-related differential genes revealed that RBM15B exhibited the most significant upregulation, suggesting its potential as a key target in CVB3-induced VMC. RBM15B, a core component of the m6A methyltransferase complex, binds METTL3 and WTAP proteins to recruit them to specific RNA loci for site-specific methylation.⁶⁴ To explore its functional role, we demonstrated that RBM15B knockdown in HL-1 cells significantly attenuated both viral replication and apoptosis upon CVB3 infection. These findings highlight the pivotal involvement of m⁶A regulators in modulating critical signaling pathways driving VMC progression. Notably, studies indicate that the m⁶A methyltransferase complex deposits m⁶A marks on nascent transcripts of promoter-associated RNAs and enhancer RNAs, thereby protecting their 5′ ends from endonuclease cleavage.⁶⁵ This implies that CVB3 infection may exploit RBM15B-mediated m⁶A methylation to manipulate mRNA stability and translational efficiency, ultimately exacerbating cardiomyocyte apoptosis and viral replication. However, this mechanistic link requires further validation.

In summary, this study provides the first transcriptome-wide map of m⁶A modifications in CVB3-induced VMC, revealing that m⁶A serves as a dynamic regulator of gene expression and signaling networks in response to viral infection. These findings not only advance our understanding of CVB3 pathogenesis but also offer potential m⁶A-related targets for therapeutic intervention. Nevertheless, further studies incorporating clinical samples are essential to validate these experimental observations and evaluate their translational relevance.