Abstract 20546: Dynamic Transcriptomics Analyses Uncover Gene Regulatory Networks for Atrial Septation and Cardiac Chamber Development
Background: Cardiovascular development is regulated by a complex gene regulatory network that includes a number of cardiac transcription factors (TFs) and their downstream target genes. The genes interact with each other in non-linear, spatial and temporal manners, which present a huge challenge for decoding the gene network. Here we aimed to gain a comprehensive understanding of the gene regulation machinery of early heart development using a systems biology approach.
Methods and Results: We used transgenic mouse models to generate mutant embryos haploinsufficient or null for cardiac TFs Gata4, Tbx5 and Osr1. Posterior second heart field (pSHF), supplier of progenitor cells for the atrial septum, of Gata4+/-, Tbx5+/-, Tbx5+/-;Osr1+/-, Osr1+/-, and Osr1-/- embryos were collected at embryonic day (E) 9.5 and 10.5 and total mRNA was subjected to RNA sequencing. A 38-gene regulatory network was constructed from the transcriptomes using distance correlation, which considered both linear and non-linear relationships between genes. One of the network genes, Pcsk6, showed decreased expression in Tbx5+/- pSHF compared to wild-type (p < 0.05) upon real-time PCR validation and TBX5 activation of Pcsk6 was further confirmed by ChIP-qPCR, luciferase assay and immunohistochemical staining. In addition, we identified one SNP (rs4965381) located in the ninth intron of PCSK6 that was associated with the atrial septal defect phenotype in a Spanish family (p = 0.004). The pSHF gene network module was compared with gene network models derived from a large scale time course microarray study of the mouse atria and ventricles from E10.5 to 18.5 (NCBI GEO Accession: GSE1479). We developed a statistical method, conditional Maximal Information Coefficient (cMIC), to identify genes showing non-linear expression patterns over time. Cardiac TF genes, as well as genes involved in cell cycle regulation, Notch signaling pathway and bone morphogenetic protein pathway were found in both atrial and ventricular datasets.
Conclusion: Heart development is orchestrated by a hierarchical spatio-temporal gene network. We generated three gene network modules that play important roles in regulating the development of pSHF, atria and ventricles respectively.
Author Disclosures: M. Xiang: None. H. Lai: None. J. Liu: None. D. Heine Suñer: None. L. Xie: None. K. Zhang: None.
- © 2016 by American Heart Association, Inc.