Abstract 48: Activation of c-Met receptor signaling by carcinoma-associated fibroblasts in the tumor microenvironment

Abstract

Abstract Objective: Carcinoma-associated fibroblasts (CAFs) play a key role in tumorigenesis by promoting tumor growth, invasion, and metastasis. At sites of intra-abdominal ovarian cancer metastases, more than half of the tumor microenvironment is composed of CAFs, however, the mechanism for how CAFs promote metastatic disease in ovarian cancer is not known. Because of the prominent role that the receptor tyrosine kinase c-Met plays in metastasis, we hypothesized that CAFs within the metastatic tumor microenvironment may activate c-Met signaling in ovarian cancer cells leading to tumor progression at a metastatic site. Methods: Human cancer cell lines and primary human carcinoma-associated fibroblasts isolated from patients’ ovarian cancer metastases were used in multiple biochemical assays to analyze the role of CAFs in c-Met receptor signaling. Results: Using a receptor tyrosine kinase array to analyze phosphorylation of 42 different receptor tyrosine kinases (RTKs), we found that c-Met was the only RTK that was phosphorylated in cancer cells upon co-culture with CAFs. Further analysis of the co-culture revealed that direct contact between cancer cells and CAFs was required for activation of c-Met. Interestingly, the unique ligand for c-Met, hepatocyte growth factor (HGF), could not be detected in conditioned media from CAFs by ELISA and addition of HGF neutralizing antibodies to the co-culture did not prevent activation of the receptor. Furthermore, CAFs transiently transfected with HGF siRNA were still able to activate c-Met in the human ovarian cancer cell line SKOV3ip1 suggesting that c-Met activation by CAFs is HGF-independent. This effect was not limited to ovarian cancer cell lines as co-culture of CAFs with the cervical cancer cell line, HeLa, which expresses wild-type c-Met, and the breast cancer cell line, T47D, showed the same results as the two ovarian cancer cell lines, SKOV3ip1 and OVCAR-5. Remarkably, membrane fragments isolated from CAFs were sufficient to stimulate phosphorylation of c-Met in cancer cells in an HGF-independent manner. Several candidate factors were identified in the CAF membrane proteome which can activate c-Met signaling. Conclusions: Collectively, these results suggest a potentially novel mechanism whereby CAFs activate c-Met in cancer cells through direct contact which provides an additional rationale for the therapeutic targeting of c-Met in cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 48. doi:1538-7445.AM2012-48