Cardiomyocyte cytosolic nuclear self-DNA contributes to the pathogenesis of desmoplakin cardiomyopathy

Figure 1. Detection of cytosolic nDNA stained with phosphorylated H2AFX or TP53BP1 in cardiomyocytes. (A) Immunofluorescence (IF) staining of isolated cardiomyocytes from wild-type (WT), Myh6-McmTam, and Myh6-McmTam Dspfl/fl mice stained for DNA with SYBR Green and phosphorylated (p-) H2A histone family member X (pH2AFX). Individual IF panels are shown along with the panels superimposed on the differential interference contrast (DIC) panel. The arrow shows cytosolic (extranuclear) nDNA stained positive for pH2AFX. (B) IF panels showing cardiomyocytes stained for DNA (SYBR Green) and the DNA damage marker tumor protein 53 binding protein 1 (TP53BP1). (C) Quantitative data representing cytosolic nDNA stained with pH2AFX depicted along with the corrected pairwise P values obtained by Dunn’s test. Only P values < 0.05 are shown. (N = 6 mice per genotype.) (D) Quantitative data representing cytosolic nDNA stained with TP53BP1. Data were analyzed by Welch’s ANOVA tests followed by Dunnett’s T3 test for multiple comparisons (N = 6 mice per genotype).

Figure 2. Detection of cytosolic nDNA stained with H3K9ac in cardiomyocytes. (A) Myocytes stained for acetylated histone 3 lysine 9 (H3K9ac) and DNA (SYBR Green). Arrows point to cytosolic nDNA. (B) Quantitative data representing data for cytosolic nDNA stained with H3K9ac (N = 6 mice per genotype). (C) Dot blots showing the individual data points, the median value, and the 95% CI of PCR amplification of selected nuclear genes in the cytosolic DNA extracts from cardiomyocytes in the WT and Myh6-McmTam Dspfl/fl groups. The cytosolic nDNA is presented as the percentage of the nuclear DNA and compared between the 2 genotypes by the Mann-Whitney test. Genotypes are denoted by different colors, and the same color is used throughout all panels in each figure.

Figure 3. Detection of cytosolic mtDNA in cardiomyocytes. (A) Confocal IF panels showing individual cardiomyocytes stained for DNA with SYBR Green and mitochondrial protein ATP5F1A along with the overlay in the WT, Myh6-McmTam, and Myh6-McmTam Dspfl/fl genotypes. (B) Spectral display of the chromophore in merged image in the 3 genotypes. The extra-mitochondrial cytoplasmic mtDNA was identified by staining of DNA with SYBR Green but not with ATP5F1A after excluding the nucleus. (C) The spectral overlaps between the SYBR Green and ATP5F1A were calculated in at least 6 to 9 myocytes and 4 hearts per genotype and were displayed as mean and SD. The differences were compared by ANOVA followed by pairwise comparison by Tukey’s method. Only the P values less than 0.05 are depicted in the graph. (D) Dot blots showing the individual data points, the median value, and the 95% CI of PCR amplification of selected mitochondrial genes in the cytosolic DNA extracts from cardiomyocytes in the WT and Myh6-McmTam Dspfl/fl mice. The cytosolic mtDNA is presented as the percentage of total cellular DNA and compared between the 2 genotypes by the Mann-Whitney test.

Figure 4. Activation of the CDSP pathway in the Myh6-McmTam Dspfl/fl cardiomyocytes. (A) Immunoblots showing expression of the selected proteins in the CDSP pathway in the WT, Myh6-McmTam, and Myh6-McmTam Dspfl/fl cardiomyocytes. (B) Dot plots showing the individual data points, median, and 95% CI representing the expression levels of the selected proteins shown in A. Data were compared among the 3 genotypes using ANOVA followed by pairwise comparisons per Bonferroni method (N = 4 to 8 per group). Only P values < 0.05 are depicted. (C) GSEA plot showing enrichment of interferon regulatory factor 3 (IRF3) target genes in the Myh6-McmTam Dspfl/fl myocytes compared with the WT. The normalized enrichment score (NES) and the q value are depicted. (D) GSEA plot showing enrichment of NF-κB target genes in the Myh6-McmTam Dspfl/fl myocytes compared with the WT. NES and the q value are depicted.

Figure 5. Activation of the ATM/TP53 pathway in the Myh6-McmTam Dspfl/fl cardiomyocytes. (A) Circos map of the differentially expressed genes showing activation of the DDR ATM/TP53 and the CDSP CGAS/STING1 pathways in the Myh6-McmTam Dspfl/fl (compared with the WT) myocytes. Fold change in gene expression is depicted as log2 units. (B) GSEA plot showing enrichment of TP53 target genes in the Myh6-McmTam Dspfl/fl myocytes compared with the WT. NES and the q value are depicted. (C) Immunoblot of selected proteins in the nuclear cell cycle checkpoint DDR pathway. (D) Dot plots showing individual data points, median, and 95% CI of the data on the expression levels of the selected proteins in the nuclear DDR pathway. ANOVA P values are shown along with Bonferroni-corrected pairwise P values (N = 4 per group).

Figure 6. Genetic blockade of the CDSP pathway by deleting the Mb21d1 (CGAS) gene in the Myh6-McmTam Dspfl/fl mice. (A) Immunoblots showing the expression levels of selected protein members of the CDSP pathway in the WT, Myh6-McmTam, Mb21d1–/–, Myh6-McmTam Dspfl/fl, and Myh6-McmTam Dspfl/fl Mb21d1–/– mice. (B) Dot plots showing the individual data points, median, and 95% CI of the quantitative data representing the expression levels of the selected proteins shown in A (N = 6 to 9 per group). Statistical analysis was performed by ANOVA followed by pairwise comparisons per Bonferroni’s method. Genotypes are denoted by color and the same color is used throughout all panels in each figure.

Figure 7. Effects of deletion of Mb21d1 gene on survival, cardiac function, and expression of heart failure biomarkers. (A) Kaplan-Meier survival plots in mice in 5 control and experimental groups. Each data point identifies 1 mouse alive at that time point, and each vertical drop indicates a death (N = 21 to 40 per genotype). The survival rates were compared by the log-rank test. (B) Dot plots showing heart weight/body weight ratio in the experimental groups (N = 17 to 21 mice per group). The P values were calculated by ANOVA and pairwise comparisons by the Bonferroni method. Only P values < 0.05 are depicted. (C) Selected echocardiographic indices of cardiac size and function are shown as dot plots in the 5 groups. Data on left ventricular end-diastolic diameter (LVEDD), LV end-systolic diameter (LVESD), LV fractional shortening (LVFS), and LV ejection fraction (LVEF) are depicted. The differences were compared by ANOVA followed by pairwise comparison by the Bonferroni method. Only P values < 0.05 are shown (N = 10 to 15 per genotype). (D) Dot plots showing cardiomyocyte transcript levels of a dozen selected genes that are the targets of IRF3 or NF-κB1 and are considered biomarkers for…

Figure 8. Effects of deletion of Mb21d1 gene on the expression levels of selected heart failure biomarkers in cardiomyocytes and nonmyocytes. (A) Immunoblots showing expression levels of matrix metalloproteinase-9 (MMP9); secreted phosphoprotein 1, also known as osteopontin (SPP1); secreted frizzled-related protein 3 (SFRP3); and vimentin (VIM) in cardiomyocytes in the control and experimental groups. (B) Quantitative data representing the blots shown in A. Individual data points, the mean, and 95% CIs are shown. The P values are calculated by ANOVA followed by pairwise comparisons. Only P values < 0.05 are shown. (C) Immunoblots showing expression levels of MMP9, SPP1, SFRP3, and VIM in nonmyocyte cells in the heart in the control and experimental groups. (D) Quantitative data representing the blots in C. Data are shown as described in B.

Figure 9. Electrocardiographic phenotype. Two-channel surface rhythm recordings in mice in the experimental groups. Normal sinus rhythm was recorded in the WT, Myh6-McmTam, and Mb21d1 mice. The Myh6-McmTam Dspfl/fl and Myh6-McmTam Dspfl/fl Mb21d1 mice showed cardiac arrhythmias, including ventricular arrhythmias and atrioventricular blocks. There were no differences in the prevalence of arrhythmias between Myh6-McmTam Dspfl/fl and Myh6-McmTam Dspfl/fl Mb21d1 mice. Selected ventricular arrhythmias and atrioventricular blocks (AVB) are depicted on the electrocardiograms. uv, microvolt.

Figure 10. Effects of deletion of Mb21d1 gene on myocardial apoptosis. (A) TUNEL-stained low-magnification myocardial sections along with the DAPI-stained sections and the overlay in the control and experimental groups. The TUNEL-stained nuclei are identified as green. TUNEL and DAPI costained nuclei are considered apoptotic cells. (B) Dot blots show the percentage of the apoptotic nuclei in the control and experimental groups. Each dot represents the mean value of the TUNEL-positive cells in each mouse (N = 4 to 5 mice per genotype). The differences were compared by ANOVA followed by pairwise comparison by the Bonferroni method. Only P values < 0.05 are shown.

Figure 11. Effects of deletion of Mb21d1 gene on protein markers of cell death programs. (A) Immunoblots showing protein levels of selected cell death markers composed of representatives of the PANoptosis pathways (apoptosis, necroptosis, and pyroptosis). (B) Dot blots showing fold change in the expression levels of the selected proteins compared with the corresponding protein in the WT mice. Data are shown as dot plots and 95% CI and compared using ANOVA followed by Bonferroni’s pairwise comparison (N = 5 to 6 mice per genotype).

Figure 12. Effects of deletion of Mb21d1 gene on selected cell death proteins in myocytes and nonmyocyte cells in the heart. (A) Immunoblots showing protein levels of selected cell death protein markers in cardiomyocytes isolated from mice in the control and experimental groups. (B) Dot blots showing fold change (FC) in the expression levels of the selected proteins in cardiomyocytes representing the blots shown in A. The level of the protein of interest in each group is compared with that in the WT mice and presented as FC. The differences among the groups were compared by ANOVA, followed by pairwise comparison (N = 6 mice per genotype). (C) Immunoblots showing protein levels of selected cell death protein markers in nonmyocyte cells in the heart in the control and experimental groups. (D) Corresponding quantitative data of protein levels shown in C. Data are shown as dot plots and 95% CI and compared using ANOVA followed by Bonferroni’s pairwise comparison (N = 6 per group). Only P values < 0.05 are depicted on the graphs.