1 year ago

2004 Scientific Report

Laboratory of Molecular

Laboratory of Molecular Oncology George F. Vande Woude, Ph.D. Dr. Vande Woude received his M.S. (1962) and Ph.D. (1964) from Rutgers University. From 1964–1972, he served first as a postdoctoral research associate, then as a research virologist for the U.S. Department of Agriculture at Plum Island Animal Disease Center. In 1972, he joined the National Cancer Institute as Head of the Human Tumor Studies and Virus Tumor Biochemistry sections and, in 1980, was appointed Chief of the Laboratory of Molecular Oncology. In 1983, he became Director of the Advanced Bioscience Laboratories-Basic Research Program at the National Cancer Institute’s Frederick Cancer Research and Development Center, a position he held until 1998. From 1995, Dr. Vande Woude first served as Special Advisor to the Director, and then as Director for the Division of Basic Sciences at the National Cancer Institute. In 1999, he was recruited to the Directorship of the Van Andel Research Institute in Grand Rapids, Michigan. Laboratory Members George Vande Woude, Ph.D. Rick Hay, Ph.D., M.D. Yu-Wen Zhang, M.D., Ph.D. Chongfeng Gao, Ph.D. Carrie Graveel, Ph.D. Sharon Moshkovitz, Ph.D. Staff Qian Xie, Ph.D. Dafna Kaufman, M.Sc. Meg Gustafson, B.A. Nathan Lanning, B.S. Yanli Su, A.M.A.T. Mary Beth Bruch Visiting scientists Nariyoshi Shinomiya, M.D., Ph.D. Galia Tsarfaty, M.D. Ilan Tsarfaty, Ph.D. Students Adi Laser, B.S. Marketta Hassen Research Interests T he Laboratory of Molecular Oncology seeks to develop a better understanding of the molecular basis of cancer, with special emphasis on testing and establishing new agents for diagnosis and treatment. Our research focuses on the activities of the Met receptor tyrosine kinase following interaction with its ligand, hepatocyte growth factor/scatter factor (HGF/SF), and the in vitro and in vivo biochemical, biological, and physiological consequences of Met kinase activation. Met and HGF/SF receptor-ligand signaling is required for normal development and homeostasis in animal models. However, abnormal expression of Met occurs in most types of human cancer and is associated with poor clinical outcomes. Because of this, Met is emerging as an important diagnostic and therapeutic target in cancer. HGF/SF-Met in proliferation, invasion, and angiogenesis We are studying the role of HGF/SF-Met in tumor proliferation, invasion, metastasis, and angiogenesis. Hepatocyte growth factor and scatter factor were discovered independently as a mitogen for hepatocytes and a motility factor for canine kidney cells, respectively. However, most tumor cells expressing Met display both proliferative and invasive phenotypes in response to HGF/SF, in a range from highly invasive to highly proliferative. Possessing both activities may be the most crucial aspect of malignant progression. Tumor cells must either possess both phenotypes or be able to shift from proliferative to invasive and back again in order to generate metastatic colonies. While most tumor cells we study possess both phenotypes, we can find cells with predominantly one phenotype among brain tumor cells. We characterize these phenotypic populations in vitro using invasion, branching, and soft agar assays, and in vivo using tumorigenicity and metastasis assays. The results are striking. We observe functional partitioning of proliferative and invasive cells in both in vitro and in vivo assays as well as in preferential signaling pathways. These studies support the hypothesis that HGF/SF-Met signaling can facilitate both phenotypes (malignant progression). They also suggest that Met-HGF/SF signaling in cancer mimics its role in normal development, whereby (as shown by Carmen Birchmeier’s lab) cells from the epithelial dermomyotome lip undergo an epithelial-mesenchymal transition and migrate to form limb muscles and other tissues. 48

Crucial questions surround the role of these two phenotypes in cancer. For example, are the highly invasive cells responsible for tumor micrometastasis? and, most importantly, how do we therapeutically target both pathways? We have found that certain geldanamycin drugs can inhibit invasion even at femtomolar levels by preventing the activity of a protease commonly found on the surface of invading cells. This may lead to an innovative therapy that suppresses tumor progression without serious side effects. Once invasive and metastatic cells arrive at new sites, they must establish new blood vessels, as Folkman has stated for many years. HGF/SF is a potent angiogenic factor, and we are studying how HGF/SF-Met signaling influences tumor angiogenesis. We have shown that HGF/SF signaling operates as a major angiogenic switch, turning on vascular endothelial growth factor (VEGF) and turning off thrombospondin-1 (TSP-1) expression and angiogenesis inhibition in tumor cells. HGF/SF also stimulates the proliferation and migration of vascular endothelial cells, and consequently it can promote angiogenesis by influencing the tumor environment. HGF/SF-Met activation and tumor development in animal models We have established a strain of immunocompromised transgenic mice that express human (hu) HGF/SF in several tissues, including liver, brain, lung, and kidney. These mice are robust hosts for the growth of Met-expressing human tumor cells (xenografts), which grow significantly more rapidly in the huHGF/SF mice than in nontransgenic hosts. This mouse strain should be a useful model for examining the role of Met in human tumor xenografts. To understand the relationship between tumorigenesis and inherited activating Met mutations in mice, we collaborated with Laura Schmidt and Bert Zbar at the National Cancer Institute (NCI) to characterize mutant forms of Met that had been shown to be tumorigenic in vitro and in vivo. The observation that activating mutations in Met occur as inherited and as sporadic mutations in human renal papillary carcinomas and other human cancers is compelling evidence that Met is an important oncogene in human cancer. To study how these activating mutations are involved in tumor development, we have generated mice bearing selected single mutations and introduced them into the mouse germline copy of Met. These mice develop several types of tumors, including carcinomas, sarcomas, and lymphomas, and the mutations give rise to differences in the patterns of tumors that develop. The differences in tumor types and latency may be due to signaling differences triggered by the specific mutations. We are currently looking for variations in downstream signaling among the several types of tumors. Other pathological phenotypes are observed among the mouse strains carrying different mutations, indicating that one gene can give rise to multiple pathological phenotypes. Strategies for molecular imaging and therapy In collaboration with Milton Gross at the Department of Veterans Affairs Healthcare System (DVAHS) in Ann Arbor, Michigan, we are developing radiolabeled monoclonal antibodies (mAbs) against the Met receptor as potential clinical imaging agents. Two mAbs recognizing different epitopes in the Met extracellular domain generate nuclear images of Met-expressing human tumor xenografts in immunocompromised mice. We have shown a correlation between the apparent avidity of the xenografts for radiolabeled anti-Met mAb and the level of Met expressed in vitro. Our current efforts are devoted to developing these anti-Met mAbs for clinical testing. In a collaboration involving VARI, DVAHS, Michigan State University, the Fred Hutchinson Cancer Research Center, and the Gerald P. Murphy Cancer Foundation, we are developing Met-directed strategies for imaging and treating metastatic prostate cancer. We have shown that anti-Met mAbs can be used to image human and canine prostate cancer xenografts in mice, and we are preparing to test one of these anti- Met mAbs as a diagnostic agent for dogs with spontaneously occurring prostate cancer. Here, as in humans, the tumors metastasize to bone. One major question we wish to answer is how tumor imaging and therapy will be affected by anti-Met mAb reactivity with normal Metexpressing tissues such as liver and kidney. Our program in molecular imaging of Met oncogene activation—a collaborative effort 49

Publications by Year