We have recently discovered that Src mediates bone metastasis through potentiating the survival response of breast cancer cells to CXCL12 and IGF1. Both of these cytokines are enriched in the bone metastasis microenvironment. This finding provided insights into the mechanisms of metastatic latency, and suggested Src inhibition as a potential strategy to eliminate disseminated cancer cells in the bone or bone marrow. In this project, our original aims were to further test the role of Src in latent bone metastasis, and to elucidate how the enhanced Src activity is acquired in different subtypes of breast tumors. Based on knowledge from the literature, we postulated that Src activation was connected to ERBB2 and ER signaling in ERBB2+ and ER+ breast cancer, respectively. We have confirmed these connections during the K99 phase. In the ROO phase, we will further investigate how Src mediates bone metastasis and resistance to anti-ERBB2 and anti- ER therapies in these two subtypes on breast cancer (Aim 2 and 3). Regarding ER-/ERBB2- tumors, our preliminary analyses indicated that Src activity associates with CXCL12 and IGF1 enrichment in primary tumors, which represents a resemblance to the microenvironment of bone metastasis. Although CXCL12 and IGF1 do not directly activate Src, they promote cell survival in a Src dependent manner. We therefore hypothesized a model of "metastasis seed pre-selection", postulating that cancer cells with enhanced Src activity are enriched in CXGL12/IGF1-hlgh primary tumors due to the survival advantages conferred by Src. These tumors are predisposed for bone colonization because ofthe resemblance of primary tumor microenvironment to that ofthe bone metastases. During the K99 phase, we have obtained multiple lines of evidence supporting the seed pre-selection model. In addition, we demonstrated that fibroblasts derived from mesenchymal stem cells are a major source of CXCL12 and IGF1 in primary tumors. These results not only provided insights into how cancer cells acquire enhanced Src activity, but also depicted an intriguing course of metastasis progression. We are therefore going to extend the original aim (Aim 1) by corroborating and generalizing this model. |