Comprehensive identification and functional study of Esrp-regulated isoforms during epithelial-mesenchymal transition.

ABSTRACT Although over 90% of human genes undergo RNA alternative splicing, most studies fail to delineate the various isoforms that are functionally important in the biological or pathological process under study. Past efforts to identify isoforms were limited by incomplete sequencing read length, depth, or both. Meanwhile, alterations of exon usage are critical regulatory mechanisms in biology with dramatic clinical consequences in embryonic anomalies and cancer transformation. This failure to delineate the phenotype-causing RNA isoform derangement is a particular barrier when investigating epithelial-mesenchymal transition (EMT), as we and others have shown that Epithelial-splicing regulatory proteins ESRP1/2 are key regulators of RNA alternative splicing and EMT. Loss of Esrp1/2 function abrogates alternative splicing in the embryonic epithelium termed periderm, resulting in global alteration of mRNA isoforms in key EMT genes causing dramatic phenotype of orofacial cleft malformation in humans, mouse and zebrafish. Since orofacial cleft is one of the most common human birth structural anomalies, there is an urgent scientific and clinical unmet need to understand the role of RNA alternative splicing and EMT with isoform level resolution and specificity. This proposal leverages next generation long-read single-cell technologies and computation approaches based on rigorous prior research to comprehensively identify RNA isoforms transcriptome-wide. This proposal tests the central hypothesis that Esrp1/2 and Esrp-regulated isoforms modulate cell-cell interactions necessary for EMT, translating to tissue morphogenesis changes in craniofacial morphogenesis. To address this hypothesis, we applied long-read bulk RNAseq in zebrafish wildtype and esrp1/2 mutants and identified tp63, gsk3bb and other candidate genes important in EMT. We will functionally investigate these genes and their isoforms in complementary in vitro and in vivo models to mechanistically elucidate their function in EMT. We will further identify lineage specific isoforms at a whole-transcriptome scale using long-read single-cell RNAseq. We also developed a CleftTeq isoform panel using TEQUILA-seq, a rapid and cost-efficient isoform identification and quantification technology we pioneered, to analyze 103 isoforms of 464 cleft genes in a proof-of-concept translation for clinical diagnosis of orofacial cleft. The expected outcome of this project is to gain a mechanistic understanding of EMT with cell-type specific isoforms and their dynamic function during EMT. This work will have a broad impact by identifying key molecular drivers of EMT shared across biological contexts in development and disease. We will also translate isoform identification and quantification to clinical diagnosis. Moreover, we will contribute public and searchable single-cell RNA isoform database during EMT and inexpensive isoform gene panel technology, to broadly share data and tools to catalyze investigations into the fundamental biological underpinnings of EMT.