Colorectal cancer (CRC) has a US incidence rate of almost 150,000 cases encompassing over 10% of new cancer diagnoses. In 2006 Within the Veterans Administration Healthcare System there were 4500 new cases and of those 650 were at advanced stage with very limited treatment options. Understanding the basic biology of underlying malignant transformation is critical in identifying new therapeutic targets and implementable strategies. Epithelial junctional pathology is common in malignancy. Decreased expression or mislocalization of E-cadherin, an adherens junction protein, has long been implicated in CRC. Claudin -3, -4, and -7, all transmembrane tight junction proteins, are overexpressed in ovarian, CRC, and gastric cancers and knock-down of claudin-3 and -4 in ovarian cancer cell lines leads to reduced invasiveness. Thus, both tight and adherens junctional dysfunction is implicated in CRC. BVES is a tight junction associated protein which when suppressed induces mesenchymal transformation in human corneal epithelial cells. Because mesenchymal transformation occurs in late stage CRC, facilitating metastasis, we postulated that BVES may be altered in CRC. We found decreased expression of BVES in CRC. BVES was underexpressed in adenomatous polyps indicating that loss of expression is an early event, likely implicating BVES in regulating additional pro- tumorigenic programs in addition to potentially contributing to metastasis. We determined the mechanism of underexpression was via transcriptional silencing via promoter hypermethylation, occurring in a large fraction of clinical samples, and virtually all CRC cell lines surveyed. We tested the functional relevance of BVES loss by restoring its expression in LIM2405 cells. These cells are weakly adherent, highly proliferative, migratory, and invasive. BVES strikingly reversed all of these phenotypes causing "epithealization" of the line with conversion to epithelial morphology and increased expression of cytokeratin and reciprocal loss of vimentin. These phenotypic changes were associated with increased E-cadherin expression with a shift in ?-catenin distribution to the cell membrane with associated decreased WNT reporter activity, implicating BVES in regulating a known cancer-signaling pathway. Additionally, in investigating the mesenchymal phenotype, we observed increased GEF-H1, BVES, Zo-1 membrane accumulation with decreased RHOA activity;indicating BVES could regulate Rho signaling. Furthermore, BVES expression in LIM2405 cells attenuated tumor growth as nude mouse xenografts and blocked SW620 metastasis. Collectively, this data suggests that BVES functions as a tumor suppressor in epithelial malignancy, thus BVES may represent a new diagnostic marker and/or therapeutic target in cancer. Much remains to be known about BVES function in epithelial malignancy. In this proposal we structure three specific aims designed to understand the role of BVES in tumor biology. First we will use murine genetic approaches to test for BVES cooperation with APC in promoting tumorigenesis and to test for a role for BVES in inflammatory carcinogenesis. We also propose mechanistic structure function studies to map BVES functional domains contributing to BVES dependent phenotypes. We have performed a yeast 2-hybrid screen and identified a panel of BVES interacting proteins. We propose characterizing the interaction between BVES and BCAR3 a breast protooncogene with GEF activity known to regulate RHO dependent processes. Lastly, we will determine if restoration of BVES expression using epigenetic modifying agents reverses the malignant phenotype of CRC cells. This could provide support for using BVES methylation as a diagnostic marker to guide therapy. Through these studies we will gain fundamental insight into how loss of BVES contributes to malignant progression, potentially providing evidence that supports developing BVES as a therapeutic target or diagnostic marker in colorectal carcinoma and addressing Provocative Question #20 from the NCI "...can biomarkers or signatures be identified that can serve as predictors or surrogates of therapeutic efficacy?".
The VA healthcare system is one of the largest healthcare providers in the United States, with over million veterans subscribed in 2009. Colorectal cancer remains the second leading cause of cancer related mortality within the United States. Because of the burden of disease associated with colorectal cancer, the VHA has issued specific directives (VHA Directive 2007-004) describing colorectal cancer prevention and control best practices for the VA. Despite these aggressive screening programs, which are clearly intended to identify disease at treatable stages, colorectal cancer is diagnosed in 4500 veterans and 650 of these are stage IV disease. We have identified a tight junction associated protein that appears to suppress cancer function. We are just barely beginning to understand the biology of this protein. Understanding its function may illuminate unrealized aspects of tumor biology. We hope to develop this protein as a diagnostic marker or druggable target in colorectal cancer, and possibly other tumor types.