miR-944 Suppresses EGF-Induced EMT in Colorectal Cancer Cells by Directly Targeting GATA6

Background: miR-944 belongs to the MicroRNAs family, as shown in our previous study, and is essential in the colorectal cancer (CRC) progression. It is negatively associated with invasion depth and lymph node status. Epithelial-mesenchymal transition (EMT) is essential in tumor invasion and metastasis. However, the relationship between miR-944 and EMT in CRC is unknown and should be further investigated. Methods: Epithelial-mesenchymal transition (EMT) progression in CRC cell lines was detected with Cell morphology and Western blotting. CRC cell migration and invasion were examined using Transwell assays. Transcriptome and clinical data were obtained from The Cancer Genome Atlas (TCGA) database. The potential pathway of miR-944 and GATA6 were predicted using KEGG analysis. Colocalization was validated using immunofluorescence and Immunohistochemistry. Nuclear and Cytoplasmic Protein Extraction assays were conducted to determine the effects of miR-944 on Wnt/beta-catenin signaling. Results: We found that miR-944 influences EGF-induced EMT malignant phenotype in vitro. KEGG analyses showed that miR-944 and GATA6 are associated with EMT related pathways, wnt signaling pathways. On the other hand, Western Blot analyses showed that miR-944 can regulate EMT and wnt-beta-catenin pathway-related protein, including beta-catenin, ZEB1, snail1 via GATA6 regulation. miR-944 also abrogates E-ca after EGF induction. Immunohistochemistry (IHC) and Immunofluorescence (IF) co-expression showed that GATA6 expression is positively associated with beta-catenin and ZEB1. GATA6 silencing can reverse EMT malignant phenotype and alterations of related protein induced by miR-944. Quantitative polymerase chain reaction analysis results showed that miR-944 is negatively associated with the UICC stage (P= 0.02), lymph nodes (p=0.04), and liver metastasis (p=0.03). Moreover, patients with high miR-944 expression have better survival (p=0.045). We finally combined miR-944 and GATA6 and found that miR-944/GATA6 ratio could be a novel prognostic biomarker in the TCGA dataset and it is an independent risk prognosis factor (p=0.045). Conclusion: Our results suggest that miR-944 suppresses the aggressive biological processes by directly repressing GATA6 expression and could be a potential candidate for therapeutic applications in CRC.