Sex differences in mitochondrial gene expression during viral myocarditis

Abstract

Background: Myocarditis is an inflammation of the heart muscle most often caused by an immune response to viral infections. Sex differences in the immune response during myocarditis have been well described but upstream mechanisms in the heart that might influence sex differences in disease are not completely understood.

Methods: Male and female BALB/c wild type mice received an intraperitoneal injection of heart-passaged coxsackievirus B3 (CVB3) or vehicle control. Bulk-tissue RNA-sequencing was conducted to better understand sex differences in CVB3 myocarditis. We performed enrichment analysis to understand sex differences in the transcriptional landscape of myocarditis and identify candidate transcription factors that might drive sex differences in myocarditis.

Results: The hearts of male and female mice with myocarditis were significantly enriched for pathways related to an innate and adaptive immune response compared to uninfected controls. When comparing females to males with myocarditis, males were enriched for inflammatory pathways and gene changes that suggested worse mitochondrial transcriptional support (e.g., mitochondrial electron transport genes). In contrast, females were enriched for pathways related to mitochondrial respiration and bioenergetics, which were confirmed by higher transcript levels of master regulators of mitochondrial function including peroxisome proliferator-activated receptor gamma coactivator 1 (PGC1α), nuclear respiratory factor 1 (NRF1) and estrogen-related receptor alpha (ERRα). TRANSFAC analysis identified ERRa as a transcription factor that may mediate sex differences in mitochondrial function during myocarditis.

Conclusions: Master regulators of mitochondrial function were elevated in females with myocarditis compared to males and may promote sex differences in mitochondrial respiratory transcript expression during viral myocarditis resulting in less severe myocarditis in females following viral infection.

Keywords: autoimmune disease; coxsackievirus B3; estrogen-related receptor alpha estrogen/ERRa; inflammation; innate immunity; interleukin-1 beta; mitochondria; nuclear respiratory factor 1/NRF1; peroxisome proliferator-activated receptor gamma coactivator 1/PGC1a.

Figure 1. Myocardial inflammation is increased in males compared to females

a, Myocarditis severity (% inflammation) between female controls (F-CON, n= 27), females with myocarditis (F-MYO, n = 41), male controls (M-CON, n= 30), and males with myocarditis (M-MYO, n = 40); b, representative heart sections (scale bars = 80μm); c-h, relative gene expression (RGE) for controls (CON, n = 31–34) versus mice with myocarditis (MYO, n = 32–41) and for F-CON (n = 15–16), F-MYO (n = 15–21), M-CON (n = 18), and M-MYO (n = 15–20).

Figure 2. RNA sequencing reveals sex differences in immune and mitochondrial genes

a, RNA-sequencing experimental pipeline; b, principal component analysis plot showing female controls (F-CON), females with myocarditis (F-MYO), male controls (M-CON), and males with myocarditis (M-MYO); c, results from GSEA pre-ranked plotted on Cytoscape with Enrichment Map and AutoAnnotate, pink = F-MYO and blue = M-MYO, nodes circled in black are mitochondrial-related pathways; heat map for d, the top 273 most differentially expressed genes and e, mitochondrial genes between all four groups.

Figure 3. Sex-specific gene expression pathways during myocarditis

The top ten most enriched pathways from GSEA ranked by normalized enrichment score comparing a, female controls (F-CON) to females with myocarditis (F-MYO), b, male controls (M-CON) to males with myocarditis (M-MYO) (pathway text marked with astricts indicate abbreviated pathway names, see supplement for abbreviations and full pathway names) c, Row normalized heatmaps for the most enriched pathways from F-CON vs F-MYO and M-CON vs M-MYO, respective to colors highlighting pathways in (a) and (b). NES = normalized enrichment score, Extracellular Matrix Structural (Struct.) Constituent (Const.). *FDRq<0.05,**FDRq<0.01, *** FDRq<0.001, **** FDRq<0.00001

Figure 4. Sex differences in differential gene expression pathways during myocarditis

The top ten most enriched pathways from GSEA ranked by normalized enrichment score comparing a, females with myocarditis (F-MYO) and males with myocarditis (M-MYO) and heatmap of auto-annotated cluster of pathways/nodes describing the mitochondrial respiratory complex, b, Row normalized heatmaps for pathways highlighted in pink for F-MYO (from F-MYO vs M-MYO comparison; other 2 pathways share common genes with those in (b) and are available in Supplemental Figure 7). *FDRq<0.05, ** FDRq<0.01, *** FDRq<0.001, **** FDRq<0.00001

Figure 5. Females with myocarditis are enriched for pathways related to mitochondrial respiration compared to males with myocarditis

Metascape enrichment results for females with myocarditis (comparing females and males with myocarditis) a, the top enriched pathways colored by cluster/pathway and b, by p-value. Top enriched pathways c, Protein-protein interaction analysis clustered by interaction outside of pathways (i) and interaction inside pathways (ii). Log10(P) vals are derived by averaging the Log10(P) vals for the 3 MCODE annotations, rounded to whole number with colors indicating respective pathways. Images in (i) are cropped to show the bulk of pathways and interactors and the top 3 pathways only are shown in (ii).

Figure 6. Males with myocarditis are enriched for pathways related immune activation compared to females with myocarditis

Metascape enrichment results for males with myocarditis (comparing females and males with myocarditis) a, the top enriched pathways colored by cluster/pathway and b, by p-value. Top enriched pathways c, Protein-protein interaction analysis clustered by interaction outside of pathways (i) and interaction inside pathways (ii). Log10(P) vals are derived by averaging the Log10(P) vals for the 3 MCODE annotations, rounded to whole number with colors indicating respective pathways. Images in (i) are cropped to show the bulk of pathways and interactors and the top 3 pathways only are shown in (ii).

Figure 7. Metascape Enrichment QC shows cell specific signals and identifies potential transcriptional regulators of sex differences in myocarditis

a, Enrichment quality control (QC) for males with myocarditis (M-MYO) shows cell specific enrichment signals from Pattern Gene Database (PaGenBase) and suggested transcription factors from Transcriptional Regulatory Relationships Unraveled by Sentence-based Text mining (TRRUST) and for b, females with myocarditis (F-MYO)

Figure 8. Sex differences in mitochondrial electron transport genes during myocarditis

Row normalized RPKM comparing female controls (FC), females with myocarditis (FM), male controls (MC) and males with myocarditis (MM) for nuclear encoded genes for a, complex I, b, color-coded illustration of the mitochondrial electron transport chain; c, complex II, d, complex III, e, complex IV, and f, ATP synthase. *p<0.05, ** p<0.01, *** p<0.001,****p<0.0001

Figure 9. TRANSFAC analysis identifies interferon regulatory factors and estrogen-related receptors as potential mediators of sex difference during myocarditis

a, TRANSFAC results comparing females with myocarditis (F-MYO) and males with myocarditis (M-MYO); RPKM (reads per kilobase per million) using false discovery rate to compare F-MYO and M-MYO for b, interferon regulatory factors (IRFs) and c, estrogen-related receptors (ERRs); d, predicted binding capacity of ERRs for electron transport chain transcripts; e, significantly different transcripts by sex in electron transport genes are indicated by bold blue lettering, green boxes indicate genes that ERRa predicted to bind to.

Figure 10. Females with myocarditis express higher levels of mitochondrial master regulators PGC1a and NRF1

Relative gene expression (RGE) for controls (CON, n = 33–35) versus mice with myocarditis (MYO, n = 35–39) and for F-CON (n = 15), F-MYO (n = 20–21), M-CON (n = 17–18), and M-MYO n = 19–20) for a-b, PGC1a; and c,d, NRF1.

Figure 11. Females with myocarditis express higher levels of ERRαcompared to males

Relative gene expression (RGE) for controls (CON, n = 33–35) versus mice with myocarditis (MYO, n = 35–39) and for F-CON (n = 15), F-MYO (n = 20–21), M-CON (n = 17–18), and M-MYO n = 19–20) for a,b, ERRa ELISA of ERRa protein from whole heart homogenate supernatant comparing c, CON (n = 14) to MYO (n= 42); d, all females (F, n = 27) to all males (M, n = 29) regardless of disease state; e, two-way ANOVA F-CON (n = 7), F-MYO (n = 20), M-CON (n = 7), and M-MYO (n = 22). Two-tailed assessment of correlations for, f ERRα gene expression and global longitudinal strain (GLS (%)), and g for ERRα gene expression and myocarditis severity (Inflammation %) scored by H+E stain.

Figure 12. In situ expression of ERRα demonstrates sex differences

Representative heart sections (based on H&E scores) for F-CON, M-CON, F-MYO, and M-MYO stained for ERRα. Images of the myocardium were taken at the base, middle (mid) and apex of the heart for each sample. Sale bars = 70 μm.