DNA methylome and transcriptome profiling reveal key electrophysiology and immune dysregulation in hypertrophic cardiomyopathy

Abstract

Hypertrophic cardiomyopathy (HCM) is the most common inherited heart disease. However, a detailed DNA methylation (DNAme) landscape has not yet been elucidated. Our study combined DNAme and transcriptome profiles for HCM myocardium and identify aberrant DNAme associated with altered myocardial function in HCM. The transcription of methylation-related genes did not significantly differ between HCM and normal myocardium. Nevertheless, the former had an altered DNAme profile compared with the latter. The hypermethylated and hypomethylated sites in HCM tissues had chromosomal distributions and functional enrichment of correlated genes differing from those of their normal tissue counterparts. The GO analysis of network underlying the genes correlated with DNAme alteration and differentially expressed genes (DEGs) shows functional clusters centred on immune cell function and muscle system processes. In KEGG analysis, only the calcium signalling pathway was enriched either by the genes correlated with changes in DNAme or DEGs. The protein-protein interactions (PPI) underlying the genes altered at both the DNAme and transcriptional highlighted two important functional clusters. One of these was related to the immune response and had the estrogen receptor-encoding ESR1 gene as its node. The other cluster comprised cardiac electrophysiology-related genes. Intelliectin-1 (ITLN1), a component of the innate immune system, was transcriptionally downregulated in HCM and had a hypermethylated site within 1500 bp upstream of the ITLN1 transcription start site. Estimates of immune infiltration demonstrated a relative decline in immune cell population diversity in HCM. A combination of DNAme and transcriptome profiles may help identify and develop new therapeutic targets for HCM.

Keywords: Cardiac electrophysiology; DNA methylation; hypertrophic cardiomyopathy; immune responses; myocardial remodelling; transcriptome.

Figure 1.

Study design and HCM methylation landscape. (a) HCM samples were obtained during septal myectomy. Twenty-four HCM and eight normal control samples were used for DNAme data collection. Sixteen HCM and seven normal control samples were used for RNA-seq data collection. Differential methylation, differentially expressed gene, and combinatorial analyses were performed to identify key genes and pathways in HCM pathogenesis. (b) Representative images of histological results. HE staining and Masson’s staining results of LV from HCM patients and normal control. Scale bar, 50 μm. (c) Principal component analysis (PCA) of DNAme in 32 samples revealed overall differences between HCM and normal samples. (d) β-value density distribution. HCM myocardium harboured hypomethylated and hypermethylated DNA. (e) △β-value distribution and range of DMPs; −0.29 to−0.1 for HCM_hypo DMPs and 0.1 to 0.34 for HCM_hyper DMPs. (f) DMP distributions in gene models. (g) DMP distributions in CpG islands (CGIs) (h) DMP distributions across autosomes. Red and blue lines on each chromosome indicate hypermethylation and hypomethylation sites, respectively. Heatmap at the bottom shows the significance of HCM_hyper DMP and HCM_hypo DMP enrichment for each chromosome. Scaled colour bar = p-value. (i) DMP methylation levels in 32 samples. Scaled colour bar = β-value.

Figure 2.

Differentially expressed genes (DEGs) in HCM and DEGs and DMP gene enrichment analyses. (a) PCA of gene expression profiles for all 32 samples. (b) DEGs with high/low expression. Compared with healthy samples, HCM samples had 554 downregulated and 351 upregulated genes. (c) DMP transcription levels in HCM and healthy groups. Scaled colour bar = FPKM values. (d) GSEA of DEGs related to the ‘innate immune system’ and ‘adaptive immune system.’ DMMT transcriptional levels. No significant intergroup differences were observed in our data (e) or GSE130036 public dataset (f). No transcriptional changes were observed for TETs in our data (g) or GSE130036 public dataset (h). (i) Functionally grouped networks of DEG and DMP gene-enriched GO items in enrichment analysis of both gene sets. (j) Fifteen most significant KEGG-enriched pathways of DEGs and DMP genes. Circle size is proportional to the number of genes in the pathway. Color represents P-value.

Figure 3.

Relationship between gene methylation and transcription and results of xCell analysis. (a) PPI networks of HCM DEGs harbouring DMPs. PPIs network of 79 DEGs affected by DNAme alterations consisted of two major clusters. A larger cluster harboured genes related to tissue and organ growth and development and was associated with genes regulating immune functions such as ESR1, STC2, TGFA, TGFB2, and VEGFC. (b) Twenty-five DEGs with regulatory DNAme alterations. CpG hypermethylation in gene promoters downregulates gene transcription (orange dots). CpG hypermethylation/hypomethylation in the gene body indicates/downregulates active gene transcription (green dots). Estimation of immune cell composition. Immune cells with distinct enhancement scores in HCM myocardium compared with normal tissue in xCell analysis are shown with their significance levels (*P value<0.05, **P value<0.01, ***P value<0.001) for our data (c) and GSE130036 public dataset (d). Immune cells arranged in order ‘Myeloids,’ ‘Stromal cells,’ ‘Stem cells,’ ‘Lymphoids,’ and ‘Others’.