HGF mediates breast tumor cell migration towards blood vessels

Abstract

In the United States, around 40,000 women are expected to die every year from breast cancer and one of eight women will develop breast cancer through their lifetime. The major reason for these deaths from cancer is due to the metastasis of the tumor cells. Breast cancer metastasis is a complex process involving several steps where tumor cells must breach the basement membrane, migrate towards a blood vessel, and then enter the vasculature to disseminate elsewhere in the body. Specifically in breast cancer, tumor cells have a specialized mode of migration towards the blood vessels called streaming, where tumor cells migrate rapidly on collagen-fibronectin fibers in a highly directional manner. Often associated with the breast tumor cells during streaming are macrophages. The macrophages and tumor cells participate in a paracrine loop where tumor cells release CSF-1 to attract the macrophages and macrophages, in turn, release EGF to attract tumor cells. This paracrine loop attraction causes tumor cells to pair with macrophages. Previous in vitro reconstitution experiments have successfully led to the formation of tumor cell and macrophage pairs. However, the ability of tumor cells to stream directionally as an ensemble seen in vivo was unsuccessful. Since breast tumor cells stream towards blood vessels in vivo, I hypothesized that endothelial cells from the blood vessel secreted a factor that would promote tumor cell streaming.;To test this hypothesis, I had to develop an in vitro 1 D assay would mimic the in vivo conditions found during tumor cell streaming. Utilizing the PyMT and MTLn3 tumor mouse models, I found that breast tumor cells prefer to stream on 2-3 micron diameter collagen fibers in vivo. Further characterization of these collagen fibers showed that the collagen fibers are often associated with fibronectin. Using these findings, I was able to create an in vitro 1 D assay that mimicked the in vivo conditions for tumor cell streaming. By using endothelial cells in the transwell assay, I was able to demonstrate that endothelial cells attract breast tumor cells through a soluble cue. More careful characterization of the signaling mechanism showed that endothelial cells secrete hepatocyte growth factor (HGF) and breast tumor cells have the receptor for HGF (CMET). This HGF/C-MET signaling is necessary for breast tumor cell streaming to occur both in vitro and in vivo as disruption of this signaling cascade reduced the ability of tumor cells to stream directionally. My work has also demonstrated that the EGF/CSF1 paracrine loop alone is insufficient to cause directional tumor cell migration. However, the EGF/CSF-1 paracrine loop plays an important role in mediating streaming of tumor cells towards endothelial cells when tumor cells are further way from the endothelial cells.;Collectively, my work was able to successfully define the signaling mechanisms and the conditions required for tumor cell streaming to occur. Understanding the signaling mechanisms that drive tumor cell streaming during breast cancer metastasis will contribute to the development of more specific and effective pharmaceutical therapies in the prevention of metastasis.