Introduction
Acute myocarditis (AM), including fulminant myocarditis (FM) and non-fulminant myocarditis (NFM), is an inflammatory disease of the heart muscle with variable clinical presentations. AM can be caused by a broad range of infectious agents, including viruses, bacteria, fungi, and protozoa, as well as non-infectious triggers, such as toxins and hypersensitive reactions (Sagar et al., 2012; Fung et al., 2016). Among these triggers, viral infection (especially coxsackie B3 virus, CVB3) has been recognized as the most common cause of myocarditis (Fairweather et al., 2012; Sagar et al., 2012). Pathological changes in the virus-infected myocardium have been well recognized, ranging from viral replication, activation of innate and acquired immune responses, to virus-induced cardiac damage and dysfunction (Shirani et al., 1993; Caforio et al., 2013). Pathogens and immunological responses can cause acute structural and functional abnormalities in the heart, which in turn leads to regional or global cardiac contractile impairment. Advances in cardiac magnetic resonance imaging (cMRI) and endo-myocardial biopsy have improved the ability to characterize changes in the heart and facilitate the diagnosis of myocarditis (Friedrich et al., 2009). However, the molecular mechanisms underlying AM are not fully understood.
FM is the most severe form of AM, which is characterized by a rapidly progressing clinical course that results in serious hemodynamic compromise and even death (Gupta et al., 2008; Wang et al., 2019). Initially, FM might present heterogeneously with symptoms ranging from fever, lethargy, and myalgia to severe heart failure. Given the variability of symptoms, it may be difficult to differentiate FM from NFM at the disease onset. FM progresses to severely impaired LVEF more rapidly than NFM. Late gadolinium enhancement (LGE) cMRI shows that patients with FM exhibit a greater inflammatory involvement of the myocardium and a greater extent of edema and fibrosis than those with NFM (Ammirati et al., 2019; Sharma et al., 2019). However, the various pathophysiological mechanisms involved in the progression of FM and NFM remain elusive. Molecular investigations in the heart tissue, such as transcriptomics, could identify common and specific molecules or pathways in FM and NFM (Heidecker et al., 2011; Jin et al., 2017).
Non-coding RNAs (ncRNAs) are a class of non-coding ribonucleic acids that are critical contributors to various biological processes, among which long non-coding RNAs (lncRNAs) spanning more than 200 nucleotides in length are the major members (Batista and Chang, 2013). Our group and others have revealed that lncRNAs could regulate the occurrence and development of viral infection and that circulating non-coding RNAs might have a great value in the diagnosis of myocarditis (Chen et al., 2018; Nie et al., 2020; Zhang C et al., 2020). Meanwhile, the roles of lncRNAs in myocarditis have attracted increasing attention (Zhang et al., 2019; Xue et al., 2020; Zhang Y et al., 2020). For example, lncRNA MEG3 inhibits M2 macrophage polarization by activating TRAF6 via miRNA-223 downregulation in viral myocarditis (Xue et al., 2020). Altered lncRNA expression profiles were identified in the plasma by microarray analysis of children with FM (Liu et al., 2019). Nevertheless, the role of myocarditis-associated lncRNAs in the heart is still not fully understood.
In the current study, the expression profiles of lncRNAs in the heart of various CVB3-induced AM mice were assessed by RNA-seq. Several lncRNAs exhibited significantly altered expression levels, which were related to inflammatory responses in both FM and NFM. Furthermore, several lncRNAs showed distinct expression patterns in FM than in NFM, which indicated that lncRNAs might play important roles in AM, especially in FM.
Materials And Methods
Animals
This study was approved by the Institutional Animal Research Committee of Tongji Medical College. All animal experimental protocols complied with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. Male A/J (~6-week-old) mice were obtained from GemPharmatech (Nanjing, China). Male C57BL/6 (~6-week-old) mice, male BALB/c (~6-week-old) mice, and male C3H mice were purchased from Beijing Vital River Laboratory Animal Technology (Beijing, China). All four mice strains were randomly assigned to two groups: control (n = 10) and CVB3-treated (n = 15) groups. A/J mice and BALB/c mice were intraperitoneally injected with 104 PFU CVB3, while C57BL/6 and C3H mice were intraperitoneally injected with 105 PFU CVB3, as described previously (Huber et al., 2001; Althof et al., 2018). Phosphate-buffered saline (PBS) was used as the control. All animals were anesthetized with intraperitoneal injections of a mixture of xylazine (5 mg/kg) and ketamine (80 mg/kg). Echocardiography was performed on day 7 after virus injection, and the mice were sacrificed. Subsequently, the organs were collected and frozen in liquid nitrogen, followed by storage at −80°C.
Cell Culture
HL-1 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) in a humidified atmosphere of 95% air and 5% CO2 at 37°C, as described previously (Fan et al., 2021). Cells were treated with CVB3 (1 × 104 plaque forming unit) and collected 24 h later. Each experiment was repeated independently at least three times.
LncRNA Sequencing Analysis
Heart tissue samples were collected from A/J, BALB/c, C57BL/6, and C3H mice. RNA isolation, quality control, library construction, and sequencing were performed by Personal Biotechnology Co. (Shanghai, China). TRIzol reagent (Invitrogen Life Technologies, CA, USA) was used to extract total RNA, and the concentration, quality, and integrity of RNAs were detected using a NanoDrop spectrophotometer (Thermo Scientific, CA, USA). Approximately 3 mg of RNA was used to generate RNA-Seq cDNA libraries. Sequencing libraries were then sequenced on a Hiseq X ten platform/NovaSeq 6000 (Illumina, San Diego, CA, USA). Cutadapt (v2.7) was used to filter the sequencing data and obtain clean reads for further analysis. The reference genome index was built using Bowtie2 (v2.4.1), and high-quality sequences were mapped to the reference genome using HISAT2 (v2.1.0). The resulting P-values were adjusted using Benjamini and Hochberg’s approach for controlling the false discovery rate. Genes with an adjusted P <0.05, and absolute LogFC >1 measured by DESeq2 were assigned as differentially expressed.
Validation by Quantitative Real-Time PCR
Total RNA extracted from the heart tissue was transcribed into cDNA, and lncRNAs were quantified by qRT-PCR using Hieff qPCR SYBR Green Master Mix (Yeasen Biotech, Shanghai, China) on a 7900HT Fast Real-Time PCR system (Life Technologies, Carlsbad, CA, USA). qRT-PCR was performed using standard settings: 95°C for 2 min, 40 cycles of 95°C for 10 s, 60°C for 30 s, and 72°C for 15 s. LncRNA primers were designed by us and synthesized by AuGCT (Wuhan, China) (
Supplemental Table S1
). PCR reactions were performed in a total volume of 10 μl containing 1 μl of 100 ng/ml sample cDNA, 5 Ml of SYBR Green Master Mix, 0.5 μl of 5 mM forward primer, 0.5 μl of 5 mM reverse primer, and 3 μl of RNase/DNase-free water. Relative expression levels were calculated using the 2-ΔΔCt relative quantification method, as previously described (Nie et al., 2020).
Cardiac Function Detection in Mice
Echocardiography analysis was performed on day 7 after virus injection using a high-resolution imaging system with a 30-MHz high-frequency scanhead (VisualSonics Vevo770, VisualSonics, Toronto, Canada), as described previously (Nie et al., 2018).
Histological Analysis
After echocardiography, the heart samples were collected, fixed with 4% paraformaldehyde, embedded in paraffin, and sectioned into 4 μm slices. Heart morphology was evaluated by hematoxylin and eosin (H&E) staining and measured using Image-Pro Plus Version 6.0 software (Media Cybernetics, Washington, USA). Six views of one H&E-stained section under a microscope were randomly collected and averaged as one data point. For each group, the results were analyzed using at least six groups of data.
Functional Analysis of lncRNA Target Genes
Cis-target gene prediction was performed using the UCSC database (Haeussler et al., 2019). DAVID Bioinformatics Resources was employed for KEGG pathway analysis of the target genes. Statistical significance was set at P <0.05. MiRNAs targets of lncRNAs were predicted by miRnada and psRobot and analyzed by Cytoscape 3.7.
Statistical Analysis
Data are presented as the mean ± standard error of the mean (SEM). Student’s t-test was used to compare the differences in normally distributed continuous values; Mann–Whitney test was used to evaluate the differences in non-normally distributed continuous values. Pearson or Spearman rank correlation was conducted to evaluate the relationships between candidate lncRNAs and clinical parameters according to the normal distribution. Statistical significance was set at P <0.05.
Discussion
In the current study, the expression profiles of lncRNAs in the heart were detected in CVB3-induced AM mouse models. Several lncRNAs were confirmed to have common or specific expression patterns in FM and NFM. Functional analyses of these dysregulated lncRNAs revealed the varied pathogenesis of FM and NFM.
Myocarditis is an inflammatory heart disease with variable clinical outcomes. There is firm evidence that enteroviruses account for many infections in cardiomyocytes. It is also thought that acute and chronic myocarditis might develop because of infection with adenoviruses, parvovirus B19 (PVB19), human herpesvirus 6, and Epstein–Barr virus (Cooper, 2009; Breinholt et al., 2010; Pollack et al., 2015). Pathogen-recognition receptors such as TLR-3 and TLR-4 recognize positive-sense single-stranded RNA viruses such as CVB3 (Garmaroudi et al., 2015). Mice deficient in TLR4 or TLR9 show attenuated myocarditis phenotypes accompanied by reduced production of inflammatory cytokines (Hollingsworth et al., 2004; Pagni et al., 2010). Animals lacking the damage-associated molecular patterns S100A8 and S100A9 showed reduced cardiac inflammation (Marinkovic et al., 2020; Muller et al., 2020). Upon infection, cardiac-resident cells, such as cardiomyocytes, endothelial cells, and fibroblasts, may contribute to acute inflammation by secreting cytokines such as IL-1β, IL-6, TNF-α, and IL-18. The inflammatory infiltrates in the EMB specimens are primarily composed of T lymphocytes and macrophages (Rivadeneyra et al., 2018). Various innate (NK cells and macrophages) and adaptive (T and B cells) immune cells infiltrate into tissues and contribute to tissue damage by secreting inflammatory cytokines (Epelman et al., 2015). Viral myocarditis presents with decreased ejection fraction and increased left ventricular end-diastolic diameter and is usually associated with cardiomyocyte loss (Sagar et al., 2012). Canonical activation of the inflammasome is critical to promote caspase-1-dependent maturation of the proinflammatory cytokines IL-1β and IL-18, as well as to induce cell death in response to pathogens (Rathinam and Fitzgerald, 2016; Toldo et al., 2018). The classical inflammasome complex consists of a cytosolic sensor, an adaptor protein ASC, and an effector caspase, pro-caspase-1. Activated pro-caspase-1 promotes the cleavage of pro-IL-1β and pro-IL-18 and the generation of biologically mature active cytokines. In addition to the canonical mode of inflammasome activation, additional caspases, such as caspase-8 or caspase-11 (caspase 4/5 in humans), also contribute to inflammasome-dependent control of IL-1β/IL18 processing and pyroptotic cell death (Rathinam and Fitzgerald, 2016).
Viral infections are the most common cause of myocarditis, which is usually caused by viruses such as enteroviruses and adeno viruses (Cooper, 2009). Modern advances in PCR technology have enabled the detection of human herpesvirus 6 and parvovirus B19 in patients with AM (Breinholt et al., 2010; Pollack et al., 2015). Coxsackieviruses belong to the Enterovirus genus within the Picornaviridae family, comprising of six Coxsackieviruses group B (CVB). Among the six serotypes of CVB, only CVB1, CVB3, and CVB5 are notably cardiotropic (Gauntt et al., 1983; Garmaroudi et al., 2015). Group B Coxsackieviruses frequently cause cardiac inflammation in humans. In murine models of CVB3-induced myocarditis, a wide spectrum of histopathology was observed, which could mimic the diverse processes in diseased patients (Garmaroudi et al., 2015). CVB3 induced myocarditis in susceptible strains of mice, such as A/J and BALB/c mice, has provided insights into the pathogenesis of viral myocarditis, as it shares many biological parameters of CVB3-induced disease in humans (Gauntt and Huber, 2003; Fairweather et al., 2012). In addition to viral infections, Trypanosoma cruzi and bacterial infections still contribute to the global burden of acute myocarditis in the developing world (Sagar et al., 2012). The lncRNA expression pattern and their potential differentiation function in these factors mediated by FM and NFM require further study.
FM is the most severe form of myocarditis. Patients with FM present with acute, severe heart failure, cardiogenic shock, and even death. Moreover, patients with FM also exhibit a greater inflammatory involvement of the myocardium and more severely impaired LVEF than those with NFM (Ammirati et al., 2019). Cardiomyocyte damage and inflammation are the predominant histopathological changes in the progression of FM (Shirani et al., 1993; Maisch et al., 2014). Here, RNA-seq was performed in the heart tissue from CVB3-induced mouse models to differentiate FM from NFM. Several dysregulated lncRNAs were detected with vital roles in the pathogenesis of FM. Longevity regulating and Hippo signaling pathways are vital contributors to cell growth and death (Zheng and Pan, 2019). FM-specific dysregulated lncRNAs might be core mediators of the longevity regulating pathway and cell growth, which play more important roles in cardiac function than the commonly dysregulated lncRNAs during AM.
Non-coding RNAs (ncRNAs) are a class of transcriptional RNAs with no protein-coding potential, mainly comprising lncRNAs, circRNAs, and miRNAs. ncRNAs are known to play vital roles in viral infection and inflammatory responses (Chen et al., 2018). Altered expression of lncRNAs in response to viral infection in eukaryotes has been reported in recent studies (Peng et al., 2010; Zhang et al., 2016; Liu et al., 2019). LncRNA HIF1A-AS1 levels were significantly increased in CVB3-induced myocardium and cardiomyocytes, and silencing of HIF1A-AS1 reduced the release of pro-inflammatory cytokines and alleviated the rate of late apoptosis and ROS production by targeting miR-138 (Cao et al., 2020). Downregulation of lncRNA-MEG3 leads to the inhibition of inflammation and induces M2 macrophage polarization via the miR-223/TRAF6/NF-κB axis in CVB3-induced viral myocarditis (Xue et al., 2020). LncRNA AK085865 promotes macrophage M2 polarization by regulating the ILF2-ILF3 complex in CVB3-induced viral myocarditis (Zhang Y et al., 2020). In addition to the host cell generating diverse lncRNAs, the viruses themselves express ncRNAs to mediate cellular antiviral activity and immune response. Human adenovirus-encoded VA RNAs bind dicer and function as competitive substrate-suppressing dicer to inhibit RNAi. VA RNA also binds and consequently blocks PKR activity and inhibits the activation of eIF-2a and viral mRNA translation (Svensson and Akusjärvi, 1984; Smith et al., 1994). Although much effort has been made to understand the effect of lncRNAs in myocarditis, the features of lncRNAs in different subtypes of acute myocarditis remain unknown. In this study, we identified and confirmed the variably and commonly expressed lncRNAs and preliminarily disclosed their potential roles in FM and NFM. However, the exact function of lncRNAs in myocarditis requires further study.
LncRNA ENSMUST00000222401, termed DIO3OS, was commonly upregulated in four mouse myocarditis models. Previously, the function of DIO3OS has been mainly studied in cancers. DIO3OS levels were considerably increased in the thyroid cancer tissue samples. Knocking down DIO3OS within thyroid carcinoma cells suppressed cancer cell viability, cell invasion, and migration via regulation of the let-7d/NF-κB2 axis (Wang et al., 2021). Another study showed that DIO3OS exhibited oncogenic properties by stimulating pancreatic cancer cell proliferation and invasion and promoting cancer growth by binding to miRNA-122 (Cui et al., 2019). Conversely, DIO3OS levels were lower in hepatocellular carcinoma tissues, and upregulation of DIO3OS repressed malignant biological behavior by sponging miR-328 (Wang et al., 2020). Other studies have also shown that the expression levels of DIO3OS were significantly lower in patients with Crohn’s disease and ulcerative colitis than those in healthy controls. The area under the receiver operating characteristic (ROC) curve between DIO3OS expression in patients with Crohn’s disease or ulcerative colitis and the healthy controls was 0.794 and 0.653, respectively, which suggests that DIO3OS has potential diagnostic value for detecting inflammatory bowel disease (Wang et al., 2018). This study suggested that lncRNA DIO3OS is related to inflammation, but the role of DIO3OS in myocarditis requires further study.
Interestingly, the expression level of ENSMUST00000199139 was increased in heart tissues but decreased in cultured cardiomyocytes upon CVB3 treatment. A similar finding was also reported in our previous study. MiR-320 levels were significantly elevated in cardiomyocytes but were decreased in cardiac fibroblasts, thereby leading to a slight increase in miR-320 expression in the global heart tissue of TAC mice (Zhang et al., 2021). Moreover, another study showed a decreased expression trend in SARS-CoV-2-infected human induced pluripotent stem cell (iPSC)-derived cardiomyocytes, as detected by RNA-seq (Sharma et al., 2020). These results suggested that ENSMUST00000199139 might be expressed by multiple cell types with different changes upon CVB3 infection. In fact, we have explored ENSMUST00000199139 expression in other cell types of the heart in A/J mice by RNA-FISH. The results showed that ENSMUST00000199139 expression was mainly increased in macrophages upon CVB3 infection (data not shown), and its function requires further investigation.
In this study, we reported altered expression profiles of lncRNAs and their correlation with cardiac function and inflammatory response in myocarditis in detail; however, there are still limitations to this study. Further in-depth studies are needed to demonstrate the exact biological function of lncRNAs in myocarditis and to investigate the diagnostic value of differential lncRNAs in plasma. Noteworthily, results of lncRNA expression levels might vary across the source material, platforms, and sample sizes, requiring careful and critical evaluation when interpreting findings and comparing results from different groups.