The epithelial SAA1 protein regulates Epithelial-Mesenchymal-Transition (EMT) in normal esophageal epithelium and esophageal carcinoma

Esophageal Squamous Cell Carcinoma (ESCC) is a highly prevalent cancer type in China and Hong Kong, and is derived from thin and flat squamous cells lining the inner wall of the esophagus. ESCC is considered as one of the most aggressive squamous cell carcinomas. The understanding of molecular basis of how a normal epithelium is transformed into an invasive ESCC phenotype will certainly enhance drug development against the advanced disease and identification of more specific/sensitive prognostic biomarkers. Our preliminary data has indicated that the Serum Amyloid A1 (SAA1) protein was predominantly expressed in normal epithelium of the esophagus, and frequent down-regulated/loss of SAA1 expression was observed in the ESCC tumor tissues. Interestingly, the expression levels of SAA1 were inversely associated with the Epithelial-Mesenchymal Transition (EMT) marker N-cadherin. The gene knockdown of SAA1 resulted in the induction of N-cadherin and the EMT-associated integrin V expression and acquisition of the mensenchymal morphology in two normal esophageal epithelial cell lines. Alternatively, when SAA1 expression was restored in ESCC cell lines, the integrin V protein expression was significantly reduced and Transforming Growth Factor- (TGF-)-induced EMT was almost completely abolished. Similarly, the integrin V knockdown could also counteract the effects of TGF-in the ESCC cells. As can be seen, our preliminary data implies that SAA1 could be a strong negative regulator of EMT in ESCC as well as the normal epithelial cells, and its expression seems to be inversely associated with the integrin V and N-cadherin. We want to determine whether SAA1 can suppress EMT through inhibition of integrin V in the normal epithelium and its neoplastic counterpart in the esophagus. We have recently established the mouse organoid and the human organotypic models which mimic the 3D organization of the normal esophageal epithelium. The functions and the expression of SAA1, integrin V, N-cadherin, and other EMT markers in different stratified cell layers (with various degree of differentiation) within the epithelium will be carefully examined. On the other hand, we will investigate if the restoration of SAA1 in the aggressive ESCC cell lines could suppress in vivo tumor invasion, an EMT-associated event. The ESCC cell lines will be transplanted into the esophagus of a mouse (orthotopic ESCC model), the tumor invasion and the expression of SAA1, integrin V, N-cadherin, and other EMT markers at the invasion edge will be examined. If inhibition of integrin V is essential for SAA1 to suppress EMT/tumor invasion, integrin V might represent a drugable target for treating the tumor progression in ESCC. To obtain clinical relevance, we aim to determine if the loss of SAA1 expression will be associated with the increased integrin V and N-cadherin expression from benign tumor to advanced metastatic carcinoma. We want to determine if the change of expression patterns of these three proteins in normal versus tumor tissues can be used as a prognostic biomarker to predict tumor progression and the patient overall survival. The results of this proposed project will indicate that SAA1 is a novel epithelial marker as well as a candidate gatekeeper for epithelial cells to enter the EMT program in the esophagus, the expression pattern of SAA1 and its associated mesenchymal proteins will help to understand the sophisticated structure of the normal esophageal epithelium.