In no tissue is the interplay between host and tumor as obvious or as destructive as in bone metastases, where tumor cells can stimulate host cells to destroy bone tissue. In understanding the role of host cells and proteins in these processes, it would be useful to take advantage of the many genetically modified mouse strains as hosts. With this in mind, transformed murine mammary epithelial cells in the 129 strain background have been developed: these cells (r3T), like the well characterized human MDA-MB-231 cells, form osteolytic bone metastases after intracardiac injection, but they do so in inbred 129 host mice, a strain commonly used for gene knockouts. Transformation of these cells to the full metastatic phenotype required several manipulations, so there are several parental cell lines at intermediate stages of transformation which will be useful in understanding the molecular changes required for the conversion of normal mammary epithelial cells to those able to form metastases in the bone. In this proposal, this novel cell system together with genetically modified host mice will be used to elucidate the role of both tumor and host proteins in bone metastasis. Two proteins that are of especial interest here since they likely function to enhance tumorigenesis both as host and as tumor proteins are TGF-beta which has multiple tumor-enhancing effects, and osteopontin (OPN), which is required for bone resorption - the metastatic r3T cells are OPN deficient. First, since the MDA-MB- 231 cells are the benchmark system for studying bone metastasis, the different cell lines will be characterized in terms of gene expression and response to TGF-beta, asking if the molecular mechanism of bone metastatic tumor growth is comparable in these cells and the MDA-MB-231 cells. Then the ability of the cells to form osteolytic bone lesions in animals lacking OPN or with reduced expression of TGF-beta will be evaluated, and bone-specific effects evaluated by comparison of bone and liver metastasis formation. In addition, the effects of tumor cell expression of OPN will be evaluated, and the mechanism of OPN effects explored through the use of mutant forms of the protein. The effect of the metastatic cells on bone cell differentiation and function in vitro will be examined, making use of the less transformed parental cell lines to evaluate how activities that influence bone cell function are induced during the process of transformation. Finally, array analysis will be used to compare genes expressed at the different stages of transformation, using the results of the in vitro experiments to help identify genes that are likely to be important in defining the bone metastatic phenotype. Together, these experiments will provide important data to establish the mechanism of bone metastasis by a novel cell line and define the role of host OPN and TGF-beta as potential therapeutic targets.