It is now appreciated that the tumor stroma contributes to tumor development, and can facilitate migration of tumor cells from the primary tumor to metastatic sites. Histologically, tumors have an activated stroma which is also a hallmark of wound healing. Indeed, tumors have been described as wounds that do not heal. Wounding activates fibroblasts involved in the repair process, and wound fibroblasts are known to share several features also found to be present in peritumor fibroblasts including profiles of secreted growth factors (cytokines, chemoattractants) and matrix proteins. Interestingly, the wound microenvironment is highly permissive for tissue regeneration and activation of stromal stem and progenitor cells and may therefore also foster expansion of tumor stem cell pools. In this project we use tissue culture and animal models to study the crosstalk between tumor and stromal cells. We established a syngeneic orthotopic mouse model to investigate the influence of wounding on breast cancer progression. During fiscal year 2007 we explored the influence of the T-cell compartment using athymic nude mice, and of stromal TGF-beta signaling on wound promoted tumor growth by using a Smad3 knockout model. We are currently investigating the contribution of Smad3 mediated angiogenesis to tumor growth in this model. To investigate the mechanism of stromal cell reprogramming in human cancers we established a collaboration with the Division of Gynecologic Oncology at the University of Virginia. From them we have received fresh tissue from endometrial cancer and surrounding normal endometrium. We use those samples to grow fibroblasts from both the normal and cancerous tissue with the goal of characterizing the differences between normal and cancer stromal cells using RNA expression profiling, proteomics, microRNA analysis, and the effect on cancer cell growth in mouse xenograft models. This year we accumulated seven pairs of normal/tumor-derived fibroblasts, determined the purity of the cell population by using epithelial/mesenchymal markers and performed microRNA profiling on the samples. The preliminary data show that indeed some microRNAs were differentially expressed in normal versus cancer fibroblasts. We are now preparing for further validation of these results by quantitative real-time PCR technique. We established a collaboration with Dr. Tim Veenstras group (NCI-FCRF) and performed preliminary experiments to optimize the conditions for mass spectrometry analysis of secreted proteins. At present we are preparing cells labeled with the SILAC reagent that will be used for the proteomics study of secreted and intracellular proteins from fibroblasts derived from normal or cancer tissue. In collaboration with Dr. Melinda Holinshead we are developing an assay which will allow us to test the role of endometrial stromal cells in supporting tumor growth in mice. This assay will be a valuable tool to study the function of microRNA or other differentially expressed genes that may contribute to tumor growth promoting functions of stromal fibroblasts.
