|Ph.D Student||Boyango Ilanit|
|Subject||Role of Heparanase in the Early Phases of Tumor Initiation:|
in vitro and in vivo Model Systems
|Department||Department of Medicine||Supervisor||Professor Israel Vlodavsky|
|Full Thesis text|
While heparanase up-regulation and its pro-malignant features are well documented, the timing of its induction in the course of tumor development is less investigated. In order to explore heparanase function in the early phases of tumor initiation and progression, we have utilized non-transformed human mammary epithelial cell line (breast MCF 10A) and genetic (Hpa-Tg mice) approaches. Once plated on and overlaid with reconstituted extracellular matrix (Matrigel), MCF10 cells produced acinar structures that displayed many of the properties of normal mammary structures in vivo. Notably, over-expression of heparanase or its C-terminal domain (8C) resulted in significantly bigger and asymmetrical acinar structures, indicating increased cell proliferation and decreased organization. These features far more dramatic when heparanase variants were over-expressed in MCF10AT1 cells that acquired the ability for xenograft growth after transfection with T24 H-Ras. Once inoculated into the mammary fat pad of immunodeficient mice, these cells produced lesions depicting mild-atypical hyperplasia and ductal carcinoma in situ. Over-expression of heparanase or its 8C variant resulted in far greater amount of highly cellular lesions. These lesions exhibited a high cell proliferation index, decreased recruitment of myoepithelial cells, and decreased cell-cell contacts (E-cadherin staining), together indicative of progressive disease. Indeed, lesions formed by heparanase- or 8C- over expressing MCF10AT1 cells were diagnosed as invasive carcinoma. In order to substantiate co-operation between Ras and heparanase, we exposed Hpa-Tg and control mice to two-steps DMBA/TPA skin carcinogenesis model because more than 90% of skin cancer initiated by DMBA contain Ha-Ras activating mutations. Hpa-Tg mice were far more sensitive to DMBA/TPA treatment, exhibiting a 10-fold increase in the number and size of tumor lesions compared with control mice. Remarkably, tumor lesions were not developed in heparanase KO mice subjected to DMBA/TPA treatment, clearly depicting a critical role of heparanase in skin tumorigenesis.
We further generated transgenic mice in which expression of heparanase and the 8C variant is driven by the mouse mammary tumor virus (MMTV). In these mice, mammary-specific heparanase over expression resulted in more developed mammary tissue accompanied by increased STAT5 phosphorylation. In order to examine if heparanase over expression sensitize the mammary gland to carcinogen, we administrated Hpa-Tg mice with DMBA considered to predominantly affect the mammary epithelium. Importantly, Hpa-Tg mice were far more sensitive to this carcinogen and survived significantly less compared with control mice, implying that heparanase co-operates with oncogenes such as Ras or Myc to drive carcinogenesis that leads to lethal disease.