Differential DNA-Methylation Pattern in Cardiac Tissue Samples from Patients with Congenital Heart Disease
02 February 2017 (online)
Similar to other types of diseases, congenital heart disease (CHD) has been associated with aberrant epigenetic modifications. In particular DNA-methylation, essential during early mammalian development, plays a key role in tissue differentiation. Although it is possible to distinguish different types of cardiac cells by their protein and/or RNA expression profiles, patterns of differentially methylated DNA have not been systematically assessed among heart tissue samples to identify consistent changes in patients with CHD. Therefore, we analyzed tissue samples from different heart regions of affected individuals to further investigate the role of DNA-methylation in CHD. We investigated 68 genomic DNA samples from patients with different types of CHD, including samples from left and right atrium (LA, RA), left and right ventricle (LV, RV), as well as the interatrial septum (IAS). Furthermore, we accessed nine non-failing heart samples to recapitulate our methods and findings. The analysis was done using Illumina's Infinnium Human Methylation 450 Bead Chip Array. Following quality control and normalization, statistical analysis was performed using the Qlucore Omics Explorer 2.3 software. For downstream verification of identified differentially methylated loci, we applied mass spectrometric Epityper technique and bisulfite pyrosequencing. Principal component analysis revealed a distinct aggregation of atrial (LA, RA, IAS) and ventricular (LV, RV) samples. Overall, 168 CpGs could be shown to be significantly differentially methylated between atrium and ventricle samples (q ≤ 1 × 10–6), showing methylation differences of greater than 30%. Atrial samples displayed hypomethylation of these loci compared with ventricular samples. These results have been verified by techniques mentioned above and independently validated in the non-failing heart samples. With the current sample size and lacking access to healthy age-matched controls we were not able to find disease-specific loci. Nevertheless, the results of this study suggest that different heart regions, in particular atrium and ventricle, could be distinguished by specific DNA-methylation profiles. To test whether this observation holds true, we are currently testing additional samples in an independent cohort of adult specimens. In summary, our study shows the importance of methylation patterns in different parts of the heart and could provide novel insight for in vivo differentiation of cardiomyocytes.