6772

STAT1

BACKGROUND: Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide. Recurrence and metastasis after curative resection remain critical obstacles in HCC treatment. CD146 predicted poor prognosis of a variety of cancers including melanoma, breast tumors, prostate cancer, and gastric cancer. However, the role of CD146 in HCC has not yet been systematically explored. METHODS: To investigate the role of CD146 in HCC, we evaluated its expression in HCC tissues and HCC cell lines using real-time PCR and western blotting (WB). Second, we established HCC cell lines that stably overexpressed and interfered CD146 and explored the function of CD146 in HCC in vitro and in vivo. Third, we conducted microarray analysis to investigate the potential mechanism by identifying differentially expressed genes. Last, follow ups were conducted to help uncover the connection of CD146 expression and the prognosis of HCC patients. RESULTS: We found that CD146 was overexpressed in HCC tissues and that high CD146 expression predicted poor overall survival time and shorter recurrence period in HCC patients. In vitro and in vivo experiments indicated that CD146 promoted migration and invasion of HCC cell lines. Further study indicated that CD146 promoted epithelial mesenchymal transition (EMT), IL-8 upregulation, and STAT1 downregulation. CD146 was upregulated in HCC tissues and cell lines. CONCLUSIONS: CD146 promoted metastasis of HCC cells and predicted poor prognosis of HCC patients. CD146 induced EMT, and IL-8 upregulation and STAT1 downregulation may be the potential underlying mechanism. The exact mechanism still needs further investigation.

Our previous studies have shown that the novel oncogene, cancer upregulated gene 2 (CUG2), activates STAT1, which is linked to anticancer drug resistance, induces epithelialmesenchymal transition (EMT) and cancer stem celllike phenotypes as determined by MTT, migration and sphere formation assays. We thus aimed to ascertain whether the activation of STAT1 by CUG2 is involved in these malignant phenotypes besides drug resistance. Here, we showed that STAT1 suppression decreased the expression of Ncadherin and vimentin, biomarkers of EMT, which led to inhibition of the migration and invasion of human lung A549 cancer cells stably expressing CUG2, but did not recover Ecadherin expression. STAT1 siRNA also diminished CUG2induced TGFbeta signaling, which is critical in EMT, and TGFbeta transcriptional activity. Conversely, inhibition of TGFbeta signaling reduced phosphorylation of STAT1, indicating a crosstalk between STAT1 and TGFbeta signaling. Furthermore, STAT1 silencing diminished sphere formation, which was supported by downregulation of stemnessrelated factors such as Sox2, Oct4, and Nanog. Constitutive suppression of STAT1 also inhibited cell migration, invasion and sphere formation. As STAT1 acetylation counteracts STAT1 phosphorylation, acetylation of STAT1 by treatment with trichostatin A, an inhibitor of histone deacetylases (HDACs), reduced cell migration, invasion, and sphere formation. As HDAC4 is known to target STAT1, its role was investigated under CUG2 overexpression. HDAC4 suppression resulted in inhibition of cell migration, invasion, and sphere formation as HDAC4 silencing hindered TGFbeta signaling and decreased expression of Sox2 and Nanog. Taken together, we suggest that STAT1HDAC4 signaling induces malignant tumor features such as EMT and sphere formation in CUG2overexpressing cancer cells.