Identification of Rare Copy Number Variants Associated With Pulmonary Atresia With Ventricular Septal Defect

Abstract

Copy number variants (CNVs) are major variations contributing to the gene heterogeneity of congenital heart diseases (CHD). pulmonary atresia with ventricular septal defect (PA-VSD) is a rare form of cyanotic CHD characterized by complex manifestations and the genetic determinants underlying PA-VSD are still largely unknown. We investigated rare CNVs in a recruited cohort of 100 unrelated patients with PA-VSD, PA-IVS, or TOF and a population-matched control cohort of 100 healthy children using whole-exome sequencing. Comparing rare CNVs in PA-VSD cases and that in PA-IVS or TOF positive controls, we observed twenty-two rare CNVs only in PA-VSD, five rare CNVs only in PA-VSD and TOF as well as thirteen rare CNVs only in PA-VSD and PA-IVS. Six of these CNVs were considered pathogenic or potentially pathogenic to PA-VSD: 16p11.2 del (PPP4C and TBX6), 5q35.3 del (FLT4), 5p13.1 del (RICTOR), 6p21.33 dup (TNXB), 7p15.2 del (HNRNPA2B1), and 19p13.3 dup (FGF22). The gene networks showed that four putative candidate genes for PA-VSD, PPP4C, FLT4, RICTOR, and FGF22 had strong interaction with well-known cardiac genes relevant to heart or blood vessel development. Meanwhile, the analysis of transcriptome array revealed that PPP4C and RICTOR were also significantly expressed in human embryonic heart. In conclusion, three rare novel CNVs were identified only in PA-VSD: 16p11.2 del (PPP4C), 5q35.3 del (FLT4) and 5p13.1 del (RICTOR), implicating novel candidate genes of interest for PA-VSD. Our study provided new insights into understanding for the pathogenesis of PA-VSD and helped elucidate critical genes for PA-VSD.

Keywords: FLT4; PPP4C; RICTOR; congenital heart defects; copy number variants; network; pulmonary atresia with ventricular septal defect; whole exome sequencing.

FIGURE 1

Venn diagram outlining overlap between rare CNVs in PA-VSD (gray), PA-IVS (yellow), and TOF (blue). We compared rare CNVs in PA-VSD cases with that in PA-IVS or TOF positive controls. There were twenty-two rare CNVs only in PA-VSD, five rare CNVs only in PA-VSD and TOF as well as thirteen rare CNVs only in PA-VSD and PA-IVS.

FIGURE 2

Rare CNVs overlapping novel candidate gene for PA-VSD: RICTOR, PPP4C, and FLT4. The dotted rectangles represent the part of candidate genes which are not within the CNVs. Genomic parameters from Ensembl (GRCh37.p13).

FIGURE 3

Expression pattern of candidate genes in human embryonic heart. Human embryonic heart in different Carnegie stages from S10 to S16 were performed the gene expression analysis using microarray.

FIGURE 4

Network analysis between candidate genes and genes associated with CHD, outflow tract development, the secondary heart field (SHF) or cardiac neural crest (CNC). We used the Cytoscape, a bioinformatic software with STRING database, to perform network interaction of proteins. The red bold fonts represent candidate genes, the blue nodes represent rare CNVs loci genes in this study and the yellow nodes represent the genes in list 1. The different width of line connecting proteins represents different intensity of the protein interaction, and the wider the connecting line is, the closer the interaction is.

FIGURE 5

Network analysis between candidate genes and genes involved in blood vessel development. We used the Cytoscape, a bioinformatic software with STRING database, to perform network interaction of proteins. The red bold fonts represent candidate genes, the blue nodes represent rare CNVs loci genes in this study and the yellow nodes represent the genes in list 2. The different width of line connecting proteins represents different intensity of the protein interaction and the wider the connecting line is, the closer the interaction is.

FIGURE 6

Network analysis between candidate genes and genes related to well-known syndromes with heart defects. We used the Cytoscape, a bioinformatic software with STRING database, to perform network interaction of proteins. The red bold fonts represent candidate genes, the blue nodes represent rare CNVs loci genes in this study and the yellow nodes represent the genes in list 3. The different width of line connecting proteins represents different intensity of the protein interaction, and the wider the connecting line is, the closer the interaction is.