ABSTRACT: The overall description of the project: Epithelial-mesenchymal transition (EMT) is a fundamental cellular process frequently hijacked by cancer cells to promote tumor progression, especially metastasis formation. Molecularly, EMT is orchestrated by various molecular networks acting at different layers of gene regulation. Compared to transcriptional regulation that has been extensively studied in the context of EMT, post-transcriptional mechanisms remain relatively underexplored. Here, by taking advantage of pooled CRISPR screen, we analyzed the influence of 1547 RNA binding proteins (RBPs) on cell motility and identified multiple core P-body (PB) components as negative modulators of cancer cell migration. Further experiments demonstrated that depletion of the PB via silencing DDX6 or EDC4 could activate hallmarks of EMT thereby enhancing cell migration in vitro as well as metastasis formation in vivo. Integrative multi-omics analysis revealed that the PB could repress the translation of its target genes, including an EMT-driver gene, HMGA2. Furthermore, we demonstrated that endoplasmic reticulum (ER) stress is an intrinsic signal that can induce PB disassembly and translational derepression of HMGA2. Finally, we used mouse genetics to demonstrate that knockout of Ddx6 resulted in EMT-related defects in embryonic development. Taken together, our study has put forward a novel function of the PB as an EMT regulator in both pathological and physiological conditions. The description of the MS part: Recent studies have suggested that the PB is mainly involved in translational regulation. Therefore, we performed mass spectrometry analysis on DDX6- and EDC4-KO clones and parental cells to measure changes at the protein level upon PB perturbation. After filtering and normalization, 3,762 and 3,000 peptides corresponding to 3,721 and 2,991 proteins were identified in DDX6- and EDC4-KO cells, respectively, and their abundance was then compared to parental HCT116 cells. A good correlation between two independent gene KO clones was observed in terms of protein level changes. In total, DDX6-KO induced up- and down-regulation of 230 and 291 proteins, respectively, while EDC4-KO induced up- and down-regulation of 73 and 22 proteins, respectively. To assess whether the PB mainly contributes to the RNA degradation or translational repression of its mRNA targets in HCT116 cells, we compared the RNA level and translation efficiency changes of DDX6-bound genes with other genes upon PB loss. As a result, both DDX6 and EDC4 KO led to a lower RNA, but higher translational efficiency of DDX6-bound genes compared to other unbound genes, suggesting PBs mainly function as a translational repressor of stored mRNAs in HCT116 cells.
Mengran Wang 