Esophageal cancer, which frequently exhibits coordinated loss of E-cadherin (Ecad) and transforming growth factor beta (TGFbeta) receptor II (TbetaRII), has a high mortality rate. In a three-dimensional organotypic culture model system, esophageal keratinocytes expressing dominant-negative mutant versions of both Ecad and TbetaRII (ECdnT) invade into the underlying matrix embedded with fibroblasts. We also find that cathepsin B induction is necessary for fibroblast-mediated invasion. Furthermore, the ECdnT cells in this physiological context activate fibroblasts through the secretion of TGFbeta1, which, in turn, is activated by cathepsin B. These results suggest that the interplay between the epithelial compartment and the surrounding microenvironment is crucial to invasion into the extracellular matrix.

E-cadherin is frequently lost during epithelial-mesenchymal transition and the progression of epithelial tumorigenesis. We found a marker of epithelial-mesenchymal transition, CD44, upregulated in response to functional loss of E-cadherin in esophageal cell lines and cancer. Loss of E-cadherin expression correlates with increased expression of CD44 standard isoform. Using an organotypic reconstruct model, we show increased CD44 expression in areas of cell invasion is associated with MMP-9 at the leading edge. Moreover, Activin A increases cell invasion through CD44 upregulation after E-cadherin loss. Taken together, our results provide functional evidence of CD44 upregulation in esophageal cancer invasion.

Although the etiology of squamous cell carcinomas of the oral mucosa is well understood, the cellular origin and the exact molecular mechanisms leading to their formation are not. Previously, we observed the coordinated loss of E-cadherin (CDH1) and transforming growth factor beta receptor II (TGFBR2) in esophageal squamous tumors. To investigate if the coordinated loss of Cdh1 and Tgfbr2 is sufficient to induce tumorigenesis in vivo, we developed two mouse models targeting ablation of both genes constitutively or inducibly in the oral-esophageal epithelium. We show that the loss of both Cdh1 and Tgfbr2 in both models is sufficient to induce squamous cell carcinomas with animals succumbing to the invasive disease by 18 months of age. Advanced tumors have the ability to invade regional lymph nodes and to establish distant pulmonary metastasis. The mouse tumors showed molecular characteristics of human tumors such as overexpression of Cyclin D1. We addressed the question whether TGFβ signaling may target known stem cell markers and thereby influence tumorigenesis. From our mouse and human models, we conclude that TGFβ signaling regulates key aspects of stemness and quiescence in vitro and in vivo. This provides a new explanation for the importance of TGFβ in mucosal homeostasis.

No abstract

We previously observed that 70% of esophageal tumors demonstrated coordinated loss of E-cadherin and TGFβRII. When grown in three-dimensional organotypic reconstruct cultures, cells lacking E-cadherin and TGFβRII demonstrate fibroblast-dependent invasion into the underlying matrix. Therefore, we hypothesized that coordinated loss of CDH1 (E-cadherin) and TGFBR2 (TGFβRII) will induce tumorigenesis in vivo. We developed a mouse model targeting CDH1 and TGFBR2 loss in the oral-esophageal epithelium using the Epstein-Barr virus L2 promoter, ED-L2. For spatio-temporal control of CDH1- and TGFBR2 gene expression, we also generated an inducible mouse model using Cre-ERT(tam) under the control of the keratin 14 promoter. Few mouse models focusing on genetic alterations of oral and head-and-neck cancer exist. We show that the loss of E-cadherin and TGFβ receptor II without carcinogen treatment is sufficient to induce invasive HNSCC and forestomach tumors. Double knock-out animals succumb to the disease between 1 and 1.5 years of age and show invasive tumors in the oral cavity and tongue, as well as the forestomach. Advanced tumors metastasize to the lung. The tumors are characterized by Ki67-positive and p63-positive staining and show disruption of adherens junctions with loss of β-catenin and mislocalisation of p120 to the cytoplasm. Additional genetic modifications frequently described for HNSCC are the upregulation of c-myc and cyclin D1. We could show that the oral mouse tumors are positive for these markers allowing us to conclude that this animal model recapitulates HNSCC at the pathologic and molecular levels. Furthermore, the forestomach is an extension of the squamous epithelium of the esophagus in the mouse. Known genetic alterations causing esophageal squamous cancer are amplification of EGFR, cyclin D1 and mutations in p53 in addition to the loss of E-cadherin and TGFβRII. The analysis of the forestomach tumors recapitulates these events in the mouse model as a result of CDH1 and TGFBR2 loss. This tumor model will provide us with a unique tool for testing therapeutic approaches. Citation Format: Thomas Andl, Gregoire F. Le Bras, Nicole F. Richards, Gillian L. Allison, M. Kay Washington, R. Katie Lee, Claudia D. Andl. A genetic mouse model of head-and-neck squamous cell carcinoma using targeted deletion of E-cadherin and TGFβ receptor II. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 335. doi:10.1158/1538-7445.AM2013-335