Acute myocarditis (AM) has been defined as a clinical manifestation of cardiac inflammation leading to myocardial damage, fibrosis and heart failure (Gupta et al., 2008). There are a variety of causes for AM, including viruses, bacteria, toxins and autoimmune diseases, among which enterovirus (especially coxsackie B3 virus, CVB3) was the most common pathogen (Maisch et al., 2014). Patients with myocarditis may present a wide range of symptoms, ranging from mild dyspnea or chest pain to cardiogenic shock and even death (Cooper, 2009). Infection by cardiotropic viruses triggers different symptoms, varying from severe myocarditis to mild or asymptomatic presentation. Fulminant myocarditis (FM), a severe form of myocarditis, has higher rates of cardiac death and heart transplantation compared with patients with non-FM (NFM) (Ammirati et al., 2019). Varying degrees of myocardial inflammation and cardiomyocyte necrosis are pathological features of AM (Ammirati et al., 2018). The molecular mechanisms underlying the various clinical presentation and severity of AM are largely elusive. It was well-recognized that A/J mice showed a higher susceptibility towards viral infection than C57BL/6 mice. CVB3 infected A/J mice exhibit severe myocarditis at day 3 and the viremia which persist up to 5–7 days after infection (Muller et al., 2015). Even after 21 days, severe heart lesions were still found in A/J mice (Wolfgram et al., 1986). Whereas the CVB3-resistant strains such as C57BL/6 mice could eliminate the virus after the early acute phase. Although histological examination showed regional myocardial inflammatory infiltrations in the initial stage, almost all signs of inflammation in the hearts of CVB3-treated C57BL/6 mice disappeared 21 days after infection (Turk, 1978; Gauntt and Huber, 2003; Corsten et al., 2012; Garmaroudi et al., 2015).

Recent studies have revealed that multiple non-coding RNAs participated in AM progression. Expression profiles of lncRNAs and miRNAs were identified in both children and adults with FM (Liu et al., 2019; Nie et al., 2020; Wang et al., 2021). A novel plasma miRNA hsa-miRChr8:96 could be used to distinguish patients with myocarditis from those with myocardial infarction, and the expression level of hsa-miRChr8:96 was positively correlated with the severity of myocarditis (Blanco-Domínguez et al., 2021). However, the involvement of circRNAs in AM pathogenesis remains largely unknown.

CircRNAs are a novel class of non-coding RNAs (ncRNAs), which originate from pre-mRNAs (Kristensen et al., 2019). CircRNAs are widely expressed in eukaryotic cells with high abundance, stable structure and tissue-specific pattern (Jeck and Sharpless, 2014). CircRNAs are emerging powerful regulators in cardiovascular diseases (Viereck and Thum, 2017; Huang et al., 2019), as well as in autoimmune diseases (Chen et al., 2019; Zhang et al., 2020a; Saaoud et al., 2020). For example, overexpression of circPPM1F could promote pancreatic islet injury by enhancing M1 macrophage activation through the circPPM1F-HuR-PPM1F-NF-κB axis (Zhang et al., 2020b). CircHIPK3 expression was significantly upregulated in myocardial tissue when exposed to LPS, while knockdown of circHIPK3 efficiently alleviated LPS-induced myocarditis in C57BL/6 mice (Fan et al., 2020). Nevertheless, expression patterns and functions of circRNAs in myocardial immune response and cardiac function remained elusive.

In the present study, circRNAs expression profiles were identified in A/J and C57BL/6 myocarditis mice models. CircRNAs showed diverging expression patterns, and might play important roles in the pathogenesis of viral myocarditis. Importantly, silencing of circArhgap32 that was downregulated in CVB3-induced A/J mice promoted cardiomyocytes apoptosis in vitro.

The current study compared the circRNA expression profile related to immune response and cardiac function between CVB3-induced FM in A/J mice and NFM in C57BL/6 mice, respectively. The results indicated that circRNA dysregulation is an important characteristic of AM.

The initial pathogenesis of both FM and NFM is considered to be similar, regardless of the triggers including viral infections, toxic substances and autoimmune conditions. FM is the most severe type of myocarditis and is characterized by acute onset and rapid progress. FM might present with abnormity in hemodynamics, severe heart failure, or even death (Sharma et al., 2019). In contrast to patients with NFM who present with New York Heart Association (NYHA) class II-III symptoms of heart failure, patients with the FM exhibit symptoms of heart failure that meet NYHA class IV criteria (Cooper, 2009). Patients with FM exhibit a greater inflammatory involvement of the myocardium and a greater extent of edema and fibrosis as well as more severely impaired LVEF than those with NFM (Ammirati et al., 2019). Cardiomyocyte death and extensive inflammation are the predominant histopathological changes during FM (Ryu et al., 2013). Biomarkers of cardiac injury are elevated in patients with acute myocarditis. A number of biomarkers can assist in diagnosing FM and estimating extent of disease. FM have higher plasma concentrations of C-reactive protein and Creatinine kinase MB than NFM (Smith et al., 1997). FM patients should receive active symptomatic management and life-support treatment (Wang et al., 2019). FM is associated with overall worse outcomes that include lower LVEF at last follow-up, higher in-hospital mortality, and increased rates of cardiac transplantation (Sagar et al., 2012).

Viral infection is the most common cause of myocarditis. CVB3, a cardiotropic virus, is the most common etiology for myocarditis (Garmaroudi et al., 2015); (). CVB3-induced acute myocarditis includes early myocyte damage. CVB3 infection begins by coupling of the virus to host-cell receptors such as CAR (coxsackievirus, and adenovirus receptor), and DAF (decay-accelerating factor). CVB3 variant PD can replicate in DAF and CAR negative cells suggesting that other receptors such as heparan sulfate are also involved in attachment of virus to the host cells (Rivadeneyra et al., 2018). Once entering into host cells, viral RNA was rapidly translated and the virus was packaged, followed by lysis of host cells, mainly through apoptosis, necrosis and autophagy. Besides the direct injury effect, aberrant immune response-mediated host cell damage was also involved in CVB3 infection. CVB3 might be sensed by well-known pattern recognition receptors (PPRs) including the Toll-like receptors (TLRs) and the RIG-I-like RNA helicases (RLHs), especially TLR3, TLR4 and MDA5 (Massilamany et al., 2014; Weithauser et al., 2016). Following ligand recognition, these receptors initiate an intracellular signaling cascade resulting in the production of type-1 IFN and pro-inflammatory cytokines such as TNF-α, IL-1, IL-18 and IL-4. Signal-transduction pathways such as GSK3β/β-catenin, MAPKs/ERK, NF-κB signal pathway and survival pathways such as protein kinase B or Akt were also significantly activated upon CVB3 infection in host cells (Daba et al., 2019).

Over-activated immune responses evoked by viral infection as well as subsequent cardiac myocyte destruction, reparative fibrosis, and heart failure are important for the pathogenesis of myocarditis, especially in FM (Fung et al., 2016). Acute myocarditis with multi-organ dysfunction is more common in young children, whereas adult patients usually present with asymptomatic, with only a few presenting with acute myocarditis (Sagar et al., 2012). The annual incidence of acute myocarditis is estimated at approximately 22 cases per 100,000 population, with heart failure just 4% of these cases (Vos et al., 2015). In studies of patients hospitalized with myocarditis, approximately 30% were considered as fulminant myocarditis (Ammirati et al., 2017). Much of our understanding of the pathophysiology and mechanisms of human viral myocarditis is based on studies of experimental murine models. Susceptible mouse strains (A/J; BALB/c) develop severe myocarditis as represented by the sustained presence of viral RNA within the myocardium, whereas more resistant strains such as C57BL/6 are able to eliminate the virus after the early acute phase (Gauntt and Huber, 2003; Corsten et al., 2012; Garmaroudi et al., 2015). The different gene backgrounds of mice showed variable susceptibility for viral infection. CVB3-treated A/J mice usually exhibit severer inflammatory infiltration and cardiac dysfunction even at 10⁴ PFU virus, which is considered as a model of severe myocarditis, while C57BL/6 mice was recognized with hereditary low susceptibility to virus-induced myocarditis at 10⁵ PFU (Althof et al., 2018). A/J mice with virus at 10⁴ PFU alone showed high mortality over time (Nakamura et al., 2013). Low dose of virus at 5.0 × 10¹ TCID₅₀ resulted in 20% mortality of the infected A/J mice, whereas doses ranging from 1 × 10³ to 1 × 10⁵ TCID₅₀ led to high mortalities (89–100%) as early as 5–14 days, and the inflammatory foci was similar at 1.0 × 10³ and 1.0 × 10⁵ TCID₅₀ (Gangaplara et al., 2012). CVB3 infection at 10⁴ PFU in C57BL/6 mice displayed mild inflammatory infiltration and less than 20% mortality (De Giusti et al., 2015; Huang et al., 2018). What determines the variable clinical presentation and severity of this disease are largely unknown.

Activation of inflammatory cells and immune response is important to eliminate viral infection, but over-activation of inflammatory reaction may also lead to adverse inflammatory cytokines secretion and result in severe myocardial damage. Viral entry into the target cells is facilitated by host receptors such as decay-accelerating factor and coxsackievirus-adenovirus receptor, which cause myocardial injury via apoptosis and necrosis of cardiomyocytes within 3–4 days post-infection (Garmaroudi et al., 2015; Rivadeneyra et al., 2018). Upon infection, various cardiac-resident cells, such as cardiomyocytes, endothelial cells, mast cells and fibroblasts, may contribute to acute inflammation by secreting cytokines such as IL-1, IL-6, TNF-α and IL-18 (Tschöpe et al., 2021). Innate leukocytes may produce type I IFNs to prevent viral replication after infection. Over-activation of NK cells may exacerbate myocardial damage owing to excessive release of cytotoxic molecules within the myocardium (Ong et al., 2017). The involvement of T cell subsets differs among various strains of mice and appears to be an important determinant of host susceptibility to viral myocarditis. Researchers showed that host susceptibility was correlated with the increased T cell responses, particularly those expressing the γδ T cell receptor (Huber et al., 1999; Esfandiarei and McManus, 2008). Th1 response is considered to be protective in acute myocarditis because it prevents viral replication (Uitendaal et al., 1979; Zhang et al., 2016). Th2 responses can reduce acute myocarditis by promoting T regulatory (Treg) cells and secreting IL-4 and IL-10 (Cunningham, 2001). Antibodies produced by B cells help to neutralize and clear the infectious virus within 2 weeks post-infection (Krebs et al., 2007). IL-10-producing B cells downregulate the proportion of Th1 and Th17 cells to alleviate inflammatory damage in the myocardium during AM (Wei et al., 2019). Hereditary susceptibility for viral infection, immune cells over-activation, and excessive release of cytotoxic molecules may be significant contributors in diverse pathogenesis of myocarditis.

Non-coding RNA (ncRNA) refers to RNA that is not translated into protein and plays a vital role in many biological processes. There is expanding evidence revealing that ncRNAs regulate the occurrence and development of AM (Liu and Ding, 2017; Zhang et al., 2020a). Dysregulated miRNA expressions were detected in the plasma from FM patients, and miR4763-3p could serve as a potential biomarker for FM diagnosis (Nie et al., 2020). MiR-21 and miR-146b were upregulated in a mouse model of AM, inhibition of miR-21 and miR-146b decreased the expression levels of IL-17 and RORγt (Liu et al., 2013). Expression profile of lncRNAs in the heart of mice with coxsackie B3 virus-induced myocarditis was also explored, and lncRNA expression levels displayed a strong correlation with immune response and cardiac function. LncRNA AK085865 promoted macrophage M2 polarization in CVB3-induced viral myocarditis by regulating ILF2-ILF3 signaling pathway (Zhang et al., 2020c). CircRNAs are considered as critical modulators in immune and inflammatory reactions (Zhang et al., 2020a). CircKcnt2 recruits the nucleosome remodeling deacetylase (NuRD) complex onto Batf (basic leucine zipper transcription factor) promoter to suppress its expression, which inhibits ILC3 activation to promote innate colitis resolution (Liu et al., 2020). Overexpression of circPPM1F could promote pancreatic islet injury by enhancing M1 macrophage activation through the HuR-PPM1F-NF-κB axis (Zhang et al., 2020b). CircSirt1 inhibits vascular inflammation by sequestering NF-κB p65 in the cytoplasm (Yang et al., 2007). Circ_001253, termed as circ_SIPA1L1, was aberrantly elevated in osteosarcoma tissues and cell lines, and over-expression of circ_SIPA1L1 promotes osteosarcoma progression via miR-379-5p/MAP3K9 axis (Jun et al., 2020). Circ-RHOT1 was increased in non-small cell lung cancer. Circ-RHOT1 overexpression abolished proliferation, migration, and invasion induced by propofol in non-small cell lung cancer by regulating miR-326 (Huang et al., 2020; Zhang et al., 2021).

Here, we reported the expression patterns of circRNAs in heart tissues during AM. A set of circRNAs were dysregulated in mild and severe myocarditis respectively. In A/J mice, a severe form of myocarditis, specifically dysregulated circRNAs were strongly correlated with the immune response and cardiac function during AM. Functional analyses showed that immune pathways, such as viral infection and T cell or B cell receptor signaling pathways, were notably activated in A/J-specific dysregulated circRNAs. Silencing of circArhgap32 impaired cardiomyocyte viability and promoted cell apoptosis. These results indicated that circRNAs dysregulation was an important characteristic of AM and may determine the severity of myocarditis. However, there are still limitations in this study. It should be mentioned that results might vary across the genetic backgrounds, environmental factors, platforms and sample sizes. Further in-depth studies are ongoing to demonstrate the exact biological functions of circRNAs in different animal models of myocarditis.

Overall, this study provided a comprehensive expression profile of differentially expressed circRNAs in AM, whose expression levels were related to the severity of myocarditis, and functional analyses indicated potential mechanisms of dysregulated circRNAs in AM.