Functional inactivation of E-cadherin by Trop-2 drives cancer metastasis

Background: Metastatic disease is the dominant cause of death in cancer patients. Hundreds of proteins/genes have been linked to the metastatic phenotype. However, neither proteomic analysis nor large-scale genome sequencing succeeded in identifying consistent markers of tumour aggressiveness and metastatic potential in cancer patients. Methods: In vitro cellular assays were utilized to identify signal transduction networks and for dynamic imaging of fluorescent signal transducer chimeras. In vivo colon cancer xenotransplants were utilized to capture drivers of metastatic diffusion to the liver. Patients case series of breast, colon, uterus, ovary, stomach, lung, and pancreatic cancers, were utilized for large-scale analysis of primary tumor impact of metastasis drivers and downstream signal transducers. Results: We discover that Trop-2 Ca2+ signaling is carried out through the ion pump Na+/K+-ATPase, for release of Ca2+ from the endoplasmic reticulum through junctional microdomain IP3 receptors, mTORC2-dependent activation of PKC and its recruitment at the cell membrane, and activatory phosphorylation of the Trop-2 cytoplasmic tail. These events lead to the activation of a universally expressed, growth-control protein super-complex, that overrides the basal growth program of cancer cells. The Trop-2/PKC feed-forward loop induces remodeling of the -actin/-actinin/myosin II cytoskeleton through cofilin-1, annexins A1/A6/A11 and gelsolin, and funnels into Akt and ERK to stimulate cancer cell growth, and to determine response to Akt inhibitors. This then leads to E-cadherin inactivation and release from the cytoskeleton, with loss of cell-cell adhesion and activation of β-catenin. This global, pro-metastatic program was shown to impact on breast, colon, ovary, uterus, stomach cancer metastasis. Conclusions: We identify functional inactivation of E-cadherin by Trop-2 as driver of metastatic diffusion in human cancer. Clustered parameters showed stronger prognostic power than any individual determinant, indicating such clusters as targets for next-generation personalized therapies.