Geneticsand epigenetics of congenital heart disease

The genetic architecture behind the development of congenital heart disease (CHD) still remains unclear, although CHD represents the most common type of birth defect affecting nearly 1% of live births. CHD occurs through heterogeneous and complex processes where common variants and/or (low/intermediate penetrance) mutations in genes critical for cardiac morphogenesis as well as dysregulation of post-transcriptional control play a crucial role. The purpose of this thesis was to expand the knowledge about the etiology of CHDs through 3 main objectives: 1. to investigate the presence of gene mutations in key cardiac transcription factors in both germline and somatic cells; 2. to analyze the association between SNPs and haplotypes in the 3’UTR of candidate genes and CHD risk as well as their influence on the post-transcriptional control through the binding with specific miRNAs; 3. to investigate the presence of novel rare mutations in 17 genes related to CHD by using a next generation sequencing (NGS) approach. We performed a Sanger sequencing of GATA4 and NKX2.5 genes in cardiac tissues of 15 sporadic (7 males; 12.4±7.5 years) and in blood samples of 15 familial (6 males; 14.1±9.0 years) patients with non-syndromic CHD. For each SNP in the 3’UTR, computational analysis was used to detect putative miRNAs and to assess the sum of all the Differences of Minimum Free Energy of hybridization (|ΔMFEtot| = Σ|ΔMFE|) in order to predict the biological importance of a SNP binding more miRNAs. Subsequently, we performed a validation reporter gene assay based on the luminescence generated by the luciferase protein, to validate the SNP whit the highest |ΔMFEtot| as a direct target of selected miRNAs. A case-control study and haplotype analysis were completed in order to define the association between the four common variants with the highest |ΔMFEtot| in the GATA4 3’UTR and the CHD risk. A total of 12 DNA samples of CHD patients was sequenced in a NGS study by using the MiSeq® sequencing platform. The Design Studio included 17 genes related to CHD: NODAL, ZIC3, ELN, NKX2.5, JAG1, CFC1, LEFTY2, ACVR2B, GATA4, GATA6, GDF1, MYH6, TFAP2A, NOTCH1, TBX1, TBX20, and ZFPM2. No evidence of novel GATA4 and NKX2.5 mutations was found in both sporadic and familial patients. However, we found several SNPs in the 3’UTR of GATA4 gene. Computational analysis revealed 27 putative miRNAs binding to GATA4 3'UTR variants. The most relevant |ΔMFEtot| (21.66 kcal/mol) was found for the miR-583, specifically targeting the +1521C>G SNP. A validation reporter gene assay based on the luminescence generated by the luciferase protein indicated that miR-583 was dose-dependently effective in regulating +1521 C allele compared with +1521 G allele. The case-control study revealed that the +1158 C>T and +1521 C>G SNPs were significantly associated with the CHD risk (both, p=0.01). The haplotype analysis showed that the C-A-A-C haplotype (more common in CHD than controls) 4-fold increased the risk of CHD (p=0.04). Conversely, the T-T-G-C haplotype (more uncommon in CHDs) was associated with a significantly decreased CHD risk (p=0.035). As regarded NGS, each single variant resulted from the targeted resequencing analysis has been annotated using the web application wANNOVAR (http://wannovar.usc.edu/) and the Ingenuity Variant Analysis software tool (Ingenuity Systems, QIAGEN), following specific parameters, as quality, frequency and pathogenicity. After filtering, the NGS analysis revealed 8 novel pathogenetic mutations, likely involved in the pathogenesis of CHD. Specifically, the mutations were identified mainly in genes (NOTCH1, MYH6, CFC1, GDF1, LEFTY2) that are related to left and/or right isomerism, involving transposition of the great arteries, tetralogy of Fallot and common atrioventricular canal anomalies. These results provided important insights into the genetics of isolated, non-syndromic CHDs, underlining the need to investigate both the impact of SNPs on posttranscriptional mechanisms and the effects of novel, rare mutations with a low penetrance.