Analysis of miRNAs Involved in Mouse Heart Injury Upon Coxsackievirus A2 Infection

Abstract

Coxsackievirus A2 (CVA2) has recently been constantly detected, and is associated with viral myocarditis in children. Our previous study demonstrated that CVA2 led to heart damage in a neonatal murine model. However, the molecular mechanism of heart injury caused by CVA2 remains largely unknown. Emerging evidence suggests the significant functions of miRNAs in Coxsackievirus infection. To investigate potential miRNAs involved in heart injury caused by CVA2, our study, for the first time, conducted a RNA-seq in vivo employing infected mice hearts. In total, 87, 101 and 76 differentially expressed miRNAs were identified at 3 days post infection (dpi), 7 dpi and 7 dpi vs 3 dpi. Importantly, above 3 comparison strategies shared 34 differentially expressed miRNAs. These results were confirmed by quantitative PCR (qPCR). Next, we did GO, KEGG, and miRNA-mRNA integrated analysis of differential miRNAs. The dual-luciferase reporter assay confirmed the miRNA-mRNA pairs. To further confirm the above enriched pathways and processes, we did Western blotting and immunofluorescence staining. Our results suggest that inflammatory responses, T cell activation, apoptosis, autophagy, antiviral immunity, NK cell infiltration, and the disruption of tight junctions are involved in the pathogenesis of heart injury caused by CVA2. The dysregulated miRNAs and pathways recognized in the current study can improve the understanding of the intricate interactions between CVA2 and the heart injury, opening a novel avenue for the future study of CVA2 pathogenesis.

Keywords: Coxsackievirus A2; RNA-seq; and mouth disease; foot; hand; miRNAs; molecular mechanism.

Figure 1

CVA2 infection leads to heart injury in a neonatal mouse model. (A) Symptoms of CVA2-infected mice at 5 dpi. The body weight (B), survival rate (C), and clinical scores (D) of Control (n=10) and CVA2-infected mice (n=10) were recorded from 1 dpi to 10 dpi. Viral titers (E) in heart tissues of mice represented by the fold change of VP1 mRNA were detected by qRT-PCR (n=4). Histopathological changes (F) in heart slices of mice were evaluated by H&E staining and TUNEL staining. Red arrows indicate limb paralysis of infected mice. Black arrows indicate inflammatory cell infiltration, myocardial fiber breakage, myocardial interstitial edema, and white arrows indicate apoptotic cells in heart slices of infected mice. *p < 0.05; ***p < 0.001.

Figure 2

The miRNA expression patterns in heart tissues of CVA2-infected mice. (A) Study design of RNA-seq analysis for CVA2-infected heart tissues. A total of 224 miRNAs with TPM > 1 were detected in heart tissues of CVA2-infected mice at 3 and 7 dpi. (B) Principal component analysis of the miRNA profile. Biological replicates were generated for each sample, represented by different color points in the figure. (C) The expression patterns of miRNAs at 3 dpi, and 7 dpi.

Figure 3

The differentially expressed miRNAs in heart tissues of CVA2-infected mice. (A) Statistics of differentially expressed miRNAs among samples. (B) Venn diagram representation of differentially expressed miRNAs sharing and exclusive constitutively presented among samples. (C–E) Volcano plots of the −log₁₀ p value vs the log₂ miRNAs abundance comparisons between control and CVA2 infected hearts. MiRNAs outside the significance threshold lines with p-value < 0.05 were colored in blue, and miRNAs with p value < 0.05 and |log2 (fold change)| > 1.0 were colored in red. (F–H) Heatmaps of differentially expressed miRNAs.

Figure 4

Visualization of differential co-expressed miRNAs in heart tissues of CVA2-infected mice. A total of 34 differential co-expressed miRNAs were identified in this study. The heatmap of these miRNAs was presented in the figure.

Figure 5

Verification of RNA-seq using qPCR. The expression of the 7 differentially expressed miRNAs identified by RNA-seq was analyzed using qPCR. (A) The 7 miRNAs changed at 3 dpi. (B) The 7 miRNAs changed at 7 dpi. The detection for each miRNA was repeated at least 3 times and the standard deviation was denoted as error bar.

Figure 6

GO and KEGG enrichment analysis of significantly dysregulated miRNAs in heart tissues. (A–C) GO enrichment analysis of significantly dysregulated miRNAs in heart tissues. (D–F) KEGG enrichment analysis of significantly dysregulated miRNAs in heart tissues.

Figure 7

Target validations of differential miRNAs. (A) The series of miR-130a-3p combined with TRIM37; nucleotides in the seed region of pmirGLOTRIM37 and pmirGLOTRIM37-mut constructs are highlighted in red. (B) Luciferase reporter assays were carried out in HEK293 cells, using renilla luciferase as the endogenous control. Data were presented as mean ± SD and were generated from three independent experiments. (C) The series of miR-29c-3p combined with Eif4e2; nucleotides in the seed region of pmirGLOEif4e2 and pmirGLOEif4e2-mut constructs are highlighted in red. (E) The series of miR-511-3p combined with TLR4; nucleotides in the seed region of pmirGLOTLR4 and pmirGLOTLR4-mut constructs are highlighted in red. (B, D, F) Luciferase reporter assays were carried out in HEK293 cells, using renilla luciferase as the endogenous control. Data were presented as mean ± SD and were generated from three independent experiments (**p < 0.01, ***p < 0.001).

Figure 8

Validation of functional enrichments. At 7 dpi, CVA2-infected mice (n = 3) and controls (n = 3) were euthanized, and heart tissues were taken out for Western blotting (n = 3) and immunofluorescence staining (n = 3). (A) Western blotting analyses of pJNK, JNK2, pERK1/2, pp38, p38, pAKT, AKT, pSTAT1, pmTOR, LC3B, CTNI, VE-Cadherin, Caspase-3, Cl-Caspase-3. (B) Relative expression of pJNK (a), pERK1/2 (b), pp38 (c), pAKT (e), pSTAT1 (f), pmTOR (g), LC3B (h), CTNI (i), VE-Cadherin (j), Cl-Caspase-3 (k) were normalized by non-phosphorylated form or actin. (C) CD11b immunofluorescence staining of heart slices of control (a) and CVA2-infected mice (b). CD11b positive cells (C-c) were calculated by Image J. Data were presented as mean ± SD and were generated from three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 9

MiRNA-mRNA interaction network. It illustrates the predicted interactions of differential miRNAs with their targets. Red and green represent differential miRNAs and target genes, respectively.

Figure 10

A hypothetical schematic of the mechanisms of heart injury caused by CVA2.