Dissecting the transcriptional network governing differentiation of periderm

点击翻译按钮获取中文摘要

Our understanding of the pathogenic mechanisms for orofacial clefting (OFC) is limited by the fact that less than half of the heritable risk for this disorder has been assigned to specific genes. Towards identifying pathological sequence variants among the many irrelevant ones detected in exomes and whole genomes of patients with this disorder, an understanding of the gene regulatory networks (GRNs) that govern the development of relevant tissues, including the oral periderm, is essential. We propose a systems biology approach to analyzing the periderm GRN. Using this approach in the past enabled us to identify three novel OFC risk genes. We will utilize two model organisms, zebrafish and mouse, because the periderm differentiation GRN appears to be highly conserved. In zebrafish, the periderm differentiates very early in embryogenesis, greatly facilitating the execution and interpretation of genetic perturbation analyses. Mouse, on the other hand, has the advantage that its craniofacial anatomy is more similar to that of humans. In Aim 1, we will determine the zebrafish periderm differentiation GRN using a state-of-the-art network inference algorithm, NetProphet 2. This tool carries out both a coexpression analysis and a differential expression analysis. Input data sets will include RNA-seq expression profiles we will generate from loss-of-function (LOF) embryos for 4 key transcription factors (TF) known to participate in this GRN. We will also identify the direct gene linkages of these key TFs in the periderm GRN. Finally, we will test a novel candidate member of the periderm GRN, Tead, by carrying out LOF tests in zebrafish, thereby exploiting the strength of this model system. In Aim 2 we will deduce the murine oral periderm differentiation GRN, also using the NetProphet algorithm. Input datasets will include expression profiles of periderm isolated from the palate shelves of wild-type mouse embryos, and from heterozygous mutants of three key TFs: Irf6, Grhl3 and Tfap2a. For each of the mutant genotypes there is evidence of abnormal periderm differentiation. We will also identify murine periderm enhancer candidates by sorting GFP-positive and -negative cells from Krt17-gfp transgenic embryos, performing ATAC-seq on both populations, and H3K27Ac ChIP-seq on cells from palate shelves and the nasal cavity. As in Aim 1, we will also identify the direct gene linkages of the key TFs. We will train a machine learning algorithm on palate periderm enhancers, and use the resulting scoring function to prioritize OFC-associated SNPs near genes that are expressed in periderm for those that are likely to directly affect risk for OFC. Finally, we will perform allele- specific reporter assays on the top candidate SNPs from each of three loci. The expected outcome is a deeper understanding of the specific TFs and cis-regulatory elements that control differentiation of the periderm. This will have a broad impact because it will enable human geneticists to prioritize candidate risk variants that emerge from whole-exome and -genome sequencing analyses of OFC.