Scientific Overview: Molecular Oncology

Research Interests

The Laboratory of Molecular Oncology is focused on understanding the numerous and diverse roles that Met and HGF/SF play in malignant progression and metastasis. Our work involves a wide variety of cancers, animal models, and drug therapies. The combination of studies, coupled with our examination of Met signaling, will lead to a greater understanding of tumor progression and new knowledge for developing and delivering novel targeted therapies.

Clonal Selection of Proliferative and Invasive Cells in Metastasis

Malignant progression leads to metastasis, which is the primary cause of death due to cancer. Metastasis begins with proliferating tumor cells that become invasive and detach from the primary tumor mass, invading the extracellular matrix, entering the bloodstream or lymphatic vessels, and establishing metastases or secondary tumors as proliferating colonies at distant sites. Because phenotypic switching between proliferation and invasion is critical to malignant progression, we use in vitro and in vivo methods to select subclones of glioblastoma tumor cells that are either highly proliferative or highly invasive.

The molecular signaling pathways that accompany phenotypic switching change dramatically in response to HGF/SF. We discovered that invasive cells signal through the Ras/MAPK pathway, while the c-myc pathway is highly expressed in proliferative clones. Using gene expression analysis, spectral karyotyping (SKY), and fluorescent in situ hybridization (FISH), we observed that subtle and specific changes in chromosome content ratio are virtually the same as the changes in the chromosome transcriptome ratio, showing that major changes in gene expression are mediated by gains or losses in chromosome content. Importantly, a significant number of the genes whose expression change is greater than twofold are functionally consistent with changes in the proliferative or invasive phenotypes. Our results imply that chromosome instability can provide the diversity of gene expression that allows a tumor to switch between proliferative and invasive phenotypes during tumor progression.

Met Induces Mammary Tumors in Mice and is Associated with Human Basal Breast Cancers

We are also investigating the role that the MET oncogene plays in breast cancer progression and metastasis through a novel mouse model of mutationally activated Met. We discovered that mutationally activated Met induces a high incidence of mammary tumors in mice. These mammary tumors have several unique pathological characteristics and contain high levels of extrachromosomal Met amplification. In addition, all of the tumors lack progesterone receptor expression and only half express ErbB2. These characteristics are similar to those of aggressive forms of human breast cancer and led us to examine how Met is associated with the various human breast cancer subtypes.

Recently, gene expression studies have identified several distinct breast cancer subtypes that correlate with clinical outcome. These molecular subtypes include three main groups of estrogen receptor (ER)–negative tumors (basal, ErbB2, and normal-like/unclassified) and at least two types of ER-positive tumors (luminal A and luminal B). A study of Met expression data f1rom existing human breast cancer datasets indicated that Met was significantly expressed in basal-like cancers relative to nonbasal cancers. To further examine Met expression patterns in human breast cancer, we used a human breast cancer tissue microarray containing 139 patient samples (in collaboration with Dr. Matthew Ellis and associates, Washington University). High Met staining was associated with the basal and ErbB2 subtypes and was inversely associated with the luminal subtypes. To confirm this observation, Met expression was compared to ER status and was found to negatively correlate with ER expression. These results show that Met protein levels are increased in the majority of breast cancer cases, but that the protein levels are highest in the more aggressive basal subtypes. Therefore, Met can be a novel therapeutic target for those patients with the most aggressive tumors and, currently, the fewest therapeutic options.

Noninvasive Imaging of Glioblastoma Progression in a Novel Mouse Model

One major deficiency of the existing glioblastoma tumor cell lines used in mouse orthotopic models is in their lack of invasiveness. We have determined that the invasive phenotype of human glioblastoma cells is greatly enhanced in cells that develop extensive metastatic foci in the lungs, skeletal muscle, and lymph nodes after tail vein injection. Importantly, all individual infiltrative cells express Met, indicating that Met would be an effective target for inhibiting glioblastoma growth and invasion.

One of the cell lines, when inoculated orthotopically, displays extensive infiltrative growth into normal mouse brain tissue. The brain tumor growth generates necrosis with pseudopalisades and closely resembles malignant glioblastoma in humans. In this model, osteolysis occurs at the inoculation site and, as a result, the tumor grows both intra- and extracranially. This growth pattern provides a transcranial acoustic window, allowing observation of tumor growth and vascularization with high-resolution micro-ultrasound. Such observation allows real time monitoring of orthotopic brain tumor growth, for assessing intracranial tumor vascularity and for evaluating the therapeutic efficacy of antitumor agents. We determined that increases in the opening of the skull are proportional to tumor growth, and therefore ultrasound provides a surrogate measurement of tumor growth. With this cell line, we can measure tumor growth orthotopically in the brain, subcutaneously as tumor xenografts, and as metastatic growth in experimental lung metastases assays. We have shown that an anti-HSP90 drug, the geldanamycin derivative 17-(allylamino)-17-demethoxygeldanamycin (17AAG), inhibits tumor growth in all three model systems.

The Role of Mig-6 in Met Signaling and Tumor Suppression

Mig-6 is one of several feedback regulators that we have found is rapidly induced by HGF/SF-Met signaling, as well as by other receptor tyrosine kinases such as EGFR. Mig-6 is a scaffolding adaptor protein that upon induction can negatively regulate EGFR and Met signaling. Mig-6 is located on human chromosome 1p36, a locus that is frequently associated with many human cancers. We have discovered that Mig-6 may function as a tumor suppressor, because mutations in the MIG-6 gene have been observed in human lung cancers, and disruption of Mig-6 in mice leads to lung, gallbladder, and bile duct cancers. Mig-6 may also play an important role in stress response and tissue homeostasis, as mice having a Mig-6 deficiency develop degenerative joint diseases that might be triggered by mechanical joint stress. We are currently investigating how Mig-6 regulates EGFR and Met signal transduction and what role Mig-6 may play in the development and progression of cancer and of degenerative joint disease.

External Collaborators

• Donald Bottaro and Benedetta Peruzzi, National Cancer Institute, Bethesda, Maryland
• Sandra Cottingham, Spectrum Health Hospitals, Grand Rapids, Michigan
• Francesco DeMayo, Baylor College of Medicine, Houston, Texas
• Ermanno Gherardi, MRC Center, Cambridge, England
• Sherri Davies and Matthew Ellis, Washington University, St. Louis, Missouri
• Beatrice Knudsen, Fred Hutchinson Cancer Research Center, Seattle, Washington
• Ernest Lengyel and Ravi Salgia, University of Chicago, Illinois
• Patricia LoRusso, Karmanos Cancer Institute, Detroit, Michigan
• Alnawaz Rehemtulla, Brian Ross, and Richard Simon, University of Michigan, Ann Arbor
• Ilan Tsarfaty, Tel Aviv University, Israel
• Robert Wondergem, East Tennessee State University, Johnson City