1 year ago

2004 Scientific Report

Laboratory of Tumor

Laboratory of Tumor Metastasis and Angiogenesis Craig P. Webb, Ph.D. Dr. Webb received his Ph.D. in cell biology from the University of East Anglia, England, in 1995. After receiving his degree, Dr. Webb served as a postdoctoral fellow in the laboratory of George Vande Woude in the Molecular Oncology Section of the Advanced BioScience Laboratories–Basic Research Program at the National Cancer Institute–Frederick Cancer Research and Development Center, Maryland (1995–1999). Dr. Webb joined VARI as a Scientific Investigator in October 1999. Staff Jennifer Bromberg-White, Ph.D. Jeremy Miller, Ph.D. David Monsma, Ph.D. Emily Eugster, M.S. Laboratory Members Sujata Srikanth, M.Phil. Stephanie Benton, B.Sc. Erika Briegel, B.A. Meghan Sheehan, B.S. Sabbatical John Ubels, Ph.D. Research Interests T umor metastasis, the process by which cancer spreads throughout a host to secondary tissues, accounts for the majority of cancer-related mortalities. The active recruitment of tumor vasculature, generally termed angiogenesis, is integral to both tumor growth and metastasis. Our laboratory focuses on identifying the key cellular and molecular determinants of metastatic progression. In addition to improving our conceptual understanding of the metastatic process, these studies may lead to the development of diagnostic and therapeutic strategies that target this most lethal aspect of cancer. Our laboratory currently uses both in vitro and in vivo systems to study metastasis and angiogenesis. Working closely with several clinical collaborators, we have collected normal and tumor tissues together with blood plasma and urine from a number of patients presenting with widespread metastatic disease. Using a variety of molecular technologies, including single nucleotide polymorphism (SNP) chips, gene expression arrays, and proteomic analyses, we have identified genomic and proteomic correlates of metastatic disease that may be future biomarkers or molecular targets for accurate diagnosis and treatment. Using our proprietary informatics solution (described below), we have identified several genes that distinguish normal from abnormal colon tissue and predict the metastatic outcome of patients with colorectal cancer. The potential diagnostic applications of these data are currently being pursued in a larger cohort of colorectal cancer patients; our findings to date suggest we can accurately diagnose colon cancer in pathological samples and, moreover, predict the likelihood of metastatic relapse well in advance of clinical presentation. In addition, we are using laser capture microdissection in conjunction with genomic and proteomic technologies to identify key tumor-host interactions during metastatic progression, with particular emphasis on identifying the molecular factors that contribute to metastatic dormancy in the liver. Through these analyses, a number of candidate genes are now being pursued as potential targets for the future treatment of metastatic disease. For this purpose, we have developed a novel retroviral system for the delivery of small interfering RNA (siRNA) molecules that target candidate genes. Using this high-throughput approach, we are knocking out the expression of several potential mediators of the metastatic phenotype in human tumor cell lines and assessing the effect on their metastatic propensity in orthotopic xenograft murine models. Collectively, we are striving towards the early and accurate diagnosis of malignant disease and its successful treatment. Multiple myeloma The most recent statistics from the American Cancer Society predict that approximately 15,000 new cases of multiple myeloma will be diagnosed within the United States this year. Some 40,000 Americans are living with multiple myeloma, and over 11,000 deaths are predicted per year, usually within three years of diagnosis. The patients endure prolonged pain associated with the spread of this cancer to multiple sites 52

within bone, leading to bone wasting, fractures, and spinal compressions. At present, treatment options for this highly aggressive and devastating cancer are extremely limited, predominantly due to our lack of understanding of its underlying causes. Recent studies show that the incidence of multiple myeloma is rapidly increasing, likely due to an aging population and unknown environmental factors. Despite these facts, multiple myeloma research is underfunded at both the national and state levels. Without in-depth study of the molecular causes of multiple myeloma, it is unlikely that significant advances in the diagnosis and treatment of this disease will soon be forthcoming. At the end of 2002, with the generous support of the McCarty Foundation and Ralph Hauenstein, we initiated the development of a dedicated multiple myeloma research laboratory (MMRL). Our specific goal for this laboratory is to use our unique integrated approach to identify optimal treatments for patients. In collaboration with Keith Stewart, Director of the McLaughlin Centre for Molecular Medicine at the Princess Margaret Hospital, Toronto, we have performed genomic and proteomic analyses on a panel of 20 human multiple myeloma cell lines that display varying responses to the drug melphalan. Using a combination of gene expression profiling and proteomic analysis, we have identified a handful of candidate genes that appear to mediate melphalan resistance. Using shRNA and pharmaceutical agents, we are attempting to reverse the drug-resistant phenotype and restore the cytotoxic response to melphalan. In addition, we have recently initiated a larger collaborative effort with several local hematologists, oncologists, pathologists, and other medical specialists to collect and process bone marrow aspirates/core biopsies, blood plasma, and urine from consenting patients having either monoclonal gammopathy of undetermined significance (MGUS) or multiple myeloma. Through the collection of detailed clinical information such as treatment response (efficacy and toxicity), coupled with single nucleotide polymorphism, gene expression, and proteomic analysis of collected samples, we aim to identify molecular Human Animal (eg. Mouse) Cell Line Discovery ➤ Biomarkers ➤ Targets ➤ Diagnostic correlates of therapeutic response and disease progression. We believe that these studies could be used to determine optimal treatments for patients with multiple myeloma in the future. XenoBase XenoBase (patent pending) is a fully integrated genomic/proteomic/medical informatics database with associated analysis and annotation tools (Fig. 1); it was designed in the Laboratory of Tumor Metastasis and Angiogenesis at the Van Andel Research Institute. Raw data generated from a variety of platforms (comparative genomic hybridization, Affymetrix SNP chips, cDNA microarrays, Affymetrix GeneChips, 2D-gel/mass spectrometry) can be associated with specimens and subjects of interest (clinical samples, animal models, cell lines), and comparative analysis can be performed on data across platforms and species. Moreover, XenoBase allows for direct correlation between subject, sample and experimental parameters, and molecular data (Fig. 2). Literature-based and gene ontology annotation software have been incorporated, along with specific metrics for biomarker and target discovery. Currently available therapeutics that specifically target molecular aberrations of interest can also readily be identified. Thus, XenoBase represents an integrated system for basic biomolecular research, clinical diagnostics, and/or new pharmacogenomic strategies for the future. HL-7 (Clinical Data) BioStore / SCION Frostbite Internal or external data DNA Genotype (SNP) RNA Gene Expression Proteomics (2D Gels, mass spectrometry) Subject • Birthdate • Sex • Condition • Images • Treatments Sample • Tracking & Storage • Images • Detailed Information Molecular Data • Protocols • Virtual Notebook • QC (MIAME) • Sample Usage Analysis • Standard statistics • Annotation • Functional predicting • Filtering Expression builder Figure 1. Schematic of XenoBase Common ID (Unigene/Homologene) Normalization Hypothesis Generator Test / Validate Develop/ Test/Validate Hypothesis 53

Publications by Year