PROJECT SUMMARY Congenital heart disease is the result of abnormal development of discrete sub-types of cardiogenic cells during early embryogenesis, and is the most common of all human birth defects. The recently developed ability to analyze the transcriptional and epigenetic state of thousands of individual cells at a time makes it tractable to discern the consequences of genetic alterations on small subsets of cells that could lead to congenital heart defects (CHDs). Nearly 30% of all CHDs involve the developing valvuloseptal region, but the mechanisms underlying valvuloseptal development are incompletely understood. Trisomy 21, the cause of Down Syndrome, offers an opportunity to investigate the genetic basis of valvuloseptal defects, as 50% of Down Syndrome children have CHD, and it is the most common known genetic cause of atrioventricular (AV) septal defects and other defects of septation. However, which of the 241 genes on Ch21 are responsible for causing cardiac defects when present in three copies remains unclear. A minimal critical region containing 148 duplicated syntenic genes sufficient for recapitulating cardiac defects in mice has been established (Dp1Tyb). Using our single cell RNAseq atlas of mouse cardiogenesis, we found only 66 of the 148 genes present in the Dp1Tyb region were expressed during cardiac development, and we defined cardiac cell types with greatest transcriptional and epigenetic changes in the Dp1Tyb mouse model. Furthermore, we differentiated a human trisomy 21 iPS cell line and its isogenic disomic iPS control toward the cardiac lineage, and observed substantial single cell transcriptional and epigenetic dysregulation in trisomic AV myocardial cells. We engineered the disomic human iPS cell line with a CRISPR-based gene activation system, introduced guide RNAs to moderately activate the 66 candidate genes on Ch 21, and used a machine learning approach to identify thirteen Ch21 genes that shifted the transcriptional signature in the AV myocardium to be more similar to that observed in trisomy 21 cells. Among these were 5 transcription factors, including BACH1 and HMGN1. In parallel, scATACseq analysis of trisomy 21 AV myocardial cells revealed a dramatic increase in motifs for the transcriptional repressor BACH1 among more closed chromatin regions. Based on these data, we hypothesize that the cardiac defects of Trisomy 21 arise due to the presence of three copies of one or more of the 13 genes identified in our pooled CRISPR screen, and that increased dosage of one or more transcription factors, such as BACH1 or HMGN1, alters AV myocardium to contribute to such defects. To test this, we propose to: 1) identify trisomy 21 gene(s) duplicated in Down Syndrome that are necessary and sufficient for modulation of iPS-derived AV myocardium; 2) determine the mechanism by which transcription factors from our screen affect transcriptional and epigenetic signatures in AV myocardium; and 3) determine the in vivo dose dependency of candidate genes in the mouse Down Syndrome region on cardiogenesis. Execution of this proposal will reveal the critical gene(s) that cause cardiac defects in Trisomy 21, and provide insight into the mechanisms that drive CHD more generally.

PROJECT NARRATIVE Valvuloseptal defects account for the majority of congenital heart defects, which are the most common human birth defect. Down Syndrome, involving trisomy of chromosome 21 affects 1 in 700 live births and causes valvuloseptal heart defects. We will investigate the molecular mechanisms that cause valvuloseptal defects in mouse and human models, including models of Down Syndrome, to understand the sets of genes that when improperly regulated cause syndromic and non-syndromic heart defects.