Dynamic DNA methylation orchestrates cardiomyocyte development, maturation and disease

The heart is a highly specialized organ with essential function for the organism throughout life. The significance of DNA methylation in shaping the phenotype of the heart remains only partially known. Here we generate and analyse DNA methylomes from highly purified cardiomyocytes of neonatal, adult healthy and adult failing hearts. We identify large genomic regions that are differentially methylated during cardiomyocyte development and maturation. Demethylation of cardiomyocyte gene bodies correlates strongly with increased gene expression. Silencing of demethylated genes is characterized by the polycomb mark H3K27me3 or by DNA methylation. De novo methylation by DNA methyltransferases 3A/B causes repression of fetal cardiac genes, including essential components of the cardiac sarcomere. Failing cardiomyocytes partially resemble neonatal methylation patterns. This study establishes DNA methylation as a highly dynamic process during postnatal growth of cardiomyocytes and their adaptation to pathological stress in a process tightly linked to gene regulation and activity.

cardiomyocyte nuclei from newborn (i), adult healthy (j) and failing (k) murine hearts (n = 6 per group). Data are displayed as mean values ± s.e.m., n = 6.  Fig. 4: a, Schematic representation of the murine Atp2a2 gene and position of the primers used for pyrosequencing to assess CpG methylation (assays I-IV, for primers see Supplementary Tab. 6). b, Methylation of three CpGs (assay II) in Atp2a2 in healthy and failing adult cardiomyocytes (PCM1-positive nuclei) and non-myocytes (PCM1-negative nuclei) as compared with heart tissue (n = 3-4 samples per group). c, Average methylation of 4 CpGs (assay IV) upstream of Atp2a2 in healthy and failing adult cardiomyocytes (PCM1positive nuclei) and non-myocytes (PCM1-negative nuclei) as compared with heart tissue (n = 3-4 samples per group, **P < 0.01, ANOVA, Bonferroni post hoc test). d, Fraction of cardiomyocyte nuclei in adult and failing hearts determined either by flow cytometry using PCM1 staining or by calculation of CpG methylation levels in cardiomyocytes, non-myocytes and cardiac tissue (average methylation of 11 CpGs using assays I-IV, n = 3-6). Data are shown as mean values ± s.e.m.  Fig. 6: a-d, Genomic annotation of the average genome (a) and DNA regions with differential CpG methylation (DMRs) between newborn, adult and failing cardiomyocytes vs. ES cells (b-d). e-h, CpG island annotation of the average genome (e) and DNA regions with differential CpG methylation (DMRs) between newborn, adult and failing cardiomyocytes vs.

Supplementary Figure 7: DNA methylation at cardiac transcription factor binding sites in differentially methylated regions (DMRs)
Suppl. Fig. 7: a, CpG methylation of 5 kbp flanking regions of cardiac transcription factor (TF) binding motifs identified in DMRs which were hypomethylated in adult cardiomyocytes vs. ES cells (TF factor motifs highlighted in bold in Fig. 1f). b-c, Transcription factor motif enrichment in hypomethylated DMRs of cardiomyocytes (b) and hearts (c) as compared with ES cells. Graphs display the 15 most significantly enriched motifs identified by HOMER (hypergeometric test). d-e, GO biological analysis of genes with proximal (-5 -+1 kbp) or distal (up to 50 kbp) DMRs by GREAT. Functional significance of hyper-(d) or hypomethylated (e) DMRs in adult cardiomyocytes as compared with ES cells. Numbers of genes per group are indicated next to the columns (binomial test).

Supplementary Figure 8: Gene ontology analysis of hypermethylated gene bodies in adult cardiomyocytes vs. ES cells
Suppl. Fig. 8: Gene ontology analysis of genes which were hypermethylated (from Fig. 2, group I) in adult cardiomyocytes vs. ES cells. P value per GO term < 10 -3 for hypermethylated genes, hypergeometric test, Bonferroni step down correction.

Supplementary Figure 9: Gene ontology analysis of hypomethylated gene bodies in adult cardiomyocytes vs. ES cells
Suppl. Fig. 9: Gene ontology analysis of genes which were hypomethylated (from Fig. 2, group II) in adult cardiomyocytes vs. ES cells. P value per GO term < 10 -13 for hypomethylated genes, hypergeometric test, Bonferroni step down correction.

Supplementary Figure 10: Gene ontology terms of genes with hypomethylated gene bodies in adult vs. newborn cardiomyocytes
Suppl. Fig. 10: Gene ontology analysis of hypomethylated genes in adult vs. neonatal cardiomyocytes (P value per GO term < 0.001, hypergeometric test, Bonferroni step down correction).

Supplementary Figure 11: Gene ontology terms of genes with hypermethylated gene bodies in adult vs. newborn cardiomyocytes
Suppl. Fig. 11: Gene ontology analysis of hypermethylated genes in adult vs. neonatal cardiomyocytes (P value per GO term < 0.05, hypergeometric test, Bonferroni step down correction).

Supplementary Figure 12: EZH2 binding and H3K27me3 of genes with low genic CpG methylation in cardiomyocytes
Suppl. Fig. 12: a, Heat maps of EZH2 binding and H3K27me3 marks for different developmental stages of genes which are demethylated in cardiomyocytes as compared with ES cells. Shown are results of E12.5 hearts 1 , newborn and adult cardiomyocytes. Displayed genes have either very low (< 1 FPKM, from  Fig. 16: Chronic cardiac pressure overload in mice after 3 weeks of transverse aortic constriction was assessed by echocardiography (a, representative M-mode tracings, vertical bar 2 mm, horizontal bar 200 msec) and led to reduced left ventricular ejection fraction, increased diastolic wall dimensions and chamber dilatation (b, n = 10-11), cardiac hypertrophy and pulmonary edema (c, n = 10-11), increased expression of brain natriuretic peptide (d, Nppb, n = 4-5) and increased cardiac fibrosis (Sirius red staining, bar 50 µm) and connective tissue growth factor (Ctgf) mRNA expression (e, n = 4-5). Data are shown as mean values ± s.e.m.; ***P < 0.001, **P < 0.01, Student's t-test.