A Cell Line and Mouse Model for Studying Pathological TGFß Signaling during Cancer Progression and Metastasis in vivo

Description:

Princeton Invention # 07-2351

Cancer metastasis requires the intricate interactions between cancer cells and the host tissue microenvironment. Although TGFb signaling pathway has been implicated in breast cancer metastasis to bone and other target organs, the temporal-spatial requirement and the in vivo dynamics of TGFb signaling in organ-specific metastasis has not been vigorously investigated. Here, we engineered a MDA-MB-231 human breast cancer cell line in which endogenous Smad4 expression is eliminated by a shRNA vector and replenished by inducible expression of an RNAi-insensitive exogenous Smad4. In order to analyze the kinetics of TGFb signaling during bone metastasis in vivo, we further engineered the cell line to harbor a dual-reporter system in which the size and location of metastastic lesions are indicated by a renilla-luciferase reporter and the strength of TGFb signaling is indicated by a firefly luciferease reporter under the control of a highly responsive TGFb responsive promoter. Using a xenograft animal model, we controlled TGFb signaling activity at various time points before or after the inoculation of tumor cells into recipient nude mice. Furthermore, we used this in vivo system to study the efficacy and real-time dynamics of a small molecule TGFb receptor kinase inhibitor and bisphosphonate, the only currently available treatment for bone metastasis. Strong TGFb signaling activities are detected in osteolytic bone lesions and are susceptible to genetic ablation of Smad4, inhibition of TGFb receptor kinase, and interference of osteoclast function by bisphosphonate. Importantly, our results indicated that the therapeutic benefits of suppressing TGFb signaling is most significant at the early stage of bone metastasis formation, and become less obvious after bone metastases become well-established. Additionally we found that bisphosphonate treatment is more effective in reducing tumor burden in certain metastasis sites, such as femur bones and skull, where TGFb signaling activity is much stronger than in the mid-spine and soft tissues. Our in vivo system for real-time manipulation and detection of TGFb signaling in bone metastasis provided proof of principle for using similar strategies to analyze the in vivo dynamics of metastasis-associated signaling pathways that mediate tumor-stroma interactions. The cell lines and mouse model that we developed in this study will be a valuable tool for analyzing differential TGFb signaling activities in common metastasis target sites as well as evaluating the therapeutic impacts of anti-metastasis agents that can directly or indirectly interfere with the TGFb pathway in vivo.