Erwin G. Van Meir - US grants

Studies examining the biological pathways used by tumors to sustain and augment their growth will further our understanding of cancer and identify novel therapeutic targets. Malignant gliomas represent 40 percent of primary brain tumors and have the highest mortality among CNS tumors, with death usually occurring in less than a year. Our long-range goal is to improve our comprehension of molecular and biological mechanisms leading to the malignant progression of human astrocytoma. The experiments proposed in this application will clarify the role of the IL-8, TSP-1 and BAI-1 angiogenesis modulators in astrocytoma progression using in vivo glioma models and will evaluate their importance as targets (IL-8) or tools (TSP-1, BAI-1) for therapy. The role of IL-8 as a growth/angiogenic stimulator and the biological basis of this activity will be functionally tested by three independent lines of research: i) overexpression of IL-8, ii) downmodulation of IL-8 and iii) antagonizing IL-8 receptor signalling. The relationship between IL-8 and VEGF will be investigated in an animal tumor model using a glioma cell line which contains a conditional VEGF gene knockout. The role of DARC and infiltrating inflammatory cells in mediating IL-8- induced angiogenesis will also be evaluated. The ability of TSP- 1 or its type 1 repeats to inhibit glioma growth/angiogenesis and the biological mechanism(s) underlying this effect will be examined by overexpression studies. In addition, the expression of TSP-1 in normal brain and during astrocytoma progression, in paired low grade/high grade tumors, will be analyzed. BAI-1 is a recently cloned brain specific inhibitor of angiogenesis which is downregulated in gliomas. The expression of BAI-1 will be examined in normal and tumoral brian using two anti-BAI-1 antibodies that were recently generated in our laboratory. In addition, the anti-angiogenic properties of BAI-1 and its ability to inhibit glioma growth will be also examined. Moreover, whether BAI-1 is the previously identified p53 regulated GD-AIF angiogenesis inhibitor will be determined. Finally, the therapeutic potential of combining these anti-angiogenic approaches will be tested and then subsequently conjugated with anti-VEGF or BCNU chemotherapy administration. Improving the treatment of malignant gliomas, an incurable disease, is an important human health goal.

DESCRIPTION (provided by applicant): Primary brain tumors (mainly malignant gliomas, medulloblastomas and ependymornas) have become the main cause of death from cancer in children and young adults. Hypoxia is a physiological difference between normal and tumor tissue. We propose to exploit this difference to construct a novel type of cancer therapy adenovirus. We will conditionally regulate the replication ability of an adenovirus by placing the adenoviral EIA gene under the control of an exogenous hypoxia-regulated promoter (HYPR-Ad). Since adenoviruses have a cytolytic cycle, the selective replication of adenoviruses within hypoxic tumor cells will lead to oncolysis. Moreover, we will augment the antitumor capability of this oncolytic: virus by having it function as a therapeutic gene delivery vehicle. We will introduce into the HYPR viral vector an expression cassette for the angiogenesis inhibitor thrombospondin-1 (HYPRA-Ad). The production of this inhibitor by infected hypoxic cells will generate a field effect that should counteract the action of the angiogenic stimulators released by these cells in response to hypoxia. In addition, it should reduce the expansion of noninfected and normoxic tumor cells since they will not be able to recruit new vascular supply. These recombinant adenoviruses will be studied for their ability to infect, replicate, and induce cytolysis of cells derived from pediatric glioma, medulloblastoma and ependymoma under normoxic and hypoxic conditions in vitro. Subsequently, the therapeutic efficacy of these viruses against xenografts of these pediatric brain tumors will be examined, in both subcutaneous and intracerebral models in immunocompromised (nulnu) mice. The tumor therapy approach presented in this proposal is novel in that these viruses can provide direct oncolytic therapy as well as deliver adjuvant gene therapy. Although these viruses have broad applicability to treat ALL cancer types which develop hypoxia, regardless of their tissue of origin and genetic composition, funding of this application will enable us to specifically develop this strategy to treat/cure pediatric brain tumors. The translation of these preclinical studies have the potential to directly benefit human health by improving the survival of children and adults with cancer.

DESCRIPTION (PROVIDED BY APPLICANT): Malignant gliomas represent 40 percent of primary brain tumors and patients with these tumors die within 1-2 years despite current conventional therapy (surgery, radiation, and chemotherapy). Hypoxia, a physiological difference between normal and tumor tissue, is a major factor in the resistance of cancer cells to radio- and chemo-therapies. We propose to exploit this difference to construct a novel type of cancer therapy adenovirus that will target hypoxic tumor cells and, therefore complement radio- and chemotherapies. We will generate an adenovirus that selectively replicates within hypoxic tumor cells. This will lead to oncolysis of these cells because adenoviruses have a cytolytic cycle. To achieve hypoxia-specific replication, we will place the adenoviral E1A gene under the control of an exogenous hypoxia-regulated promoter (HYPR-Ad). The E1A gene encodes an early viral protein essential for the initiation of adenovirus replication. Moreover, we will augment the anti-tumor capability of this oncolytic virus by having it function as a therapeutic gene delivery vehicle. We will introduce into the HYPR viral vector an expression cassette for the angiogenesis inhibitor angiostatin (HYPRA-Ad). The production of this angiogenesis inhibitor by infected hypoxic cells will generate a field effect that should counteract the action of the angiogenic stimulators released by these cells in response to hypoxia. In addition, it should reduce the expansion of noninfected and normoxic tumor cells by preventing them from recruiting new vascular supply. These recombinant adenoviruses will be studied for their ability to infect, replicate, and induce cytolysis of cells derived from glioma under normoxic and hypoxic conditions in vitro. Subsequently, the therapeutic efficacy of these viruses will be examined using subcutaneous and intracerebral human glioma models in mice and the efficacy of their combination with standard radio- and chemo-therapy will be evaluated. This tumor therapy approach is novel in that these viruses can provide direct oncolytic therapy as well as deliver adjuvant gene therapy. Most importantly, these viruses have broad applicability to treat ALL cancer types that develop hypoxia regardless of their tissue of origin and genetic composition. The translation of these preclinical studies have the potential to directly benefit human health by improving the survival of cancer patients.

DESCRIPTION (provided by applicant): There is an urgent need to develop novel therapies for malignant solid tumors. Tumor hypoxia, a reduction In partial oxygen pressure is a characteristic of solid tumor growth and develops following insufficient oxygen supply from preexisting vasculature. This phenomenon stimulates tumor progression by activating physiological responses that permit tumor-induced angiogenesis and metabolic adaptation to growth under a hypoxic environment. It is also a major factor in the resistance of cancer cells to radio- and chemo-therapies. Hypoxia triggers activation of Hypoxia-lnducible Factor 1 (HIF-1), a transcription factor that drives transcription of genes encoding pro-angiogenic factors and glycolytic enzymes that contribute to tumor growth. Strategies that inhibit HIF-1 function through HIF-?? knockdown or knockout approaches have reduced tumor growth in experimental models including glioblastoma, the most malignant brain tumor for which there is currently no effective therapy. Here we would like to test the hypothesis that small molecules that inhibit the HIF-1 pathway will inhibit the growth of glioblastoma, either singly or in combination with other agents. We have developed a novel pipeline of such potential therapeutic agents by screening a natural product-like library of small molecular compounds using a HIF-1-responsive cell-based reporter assay. We have generated extensive preliminary data showing that two structural classes of compounds identified have potent anti-HIF activity in vitro and in vivo. Furthermore, we have found that our lead HIF inhibitor acts via a unique mechanism and is able to strongly inhibit in vivo tumor growth upon systemic administration. Here we plan to further develop these lead molecules, test them further in animal models of glioblastoma as a model for an aggressive solid tumor relying on HIF-1 activation for its growth, and determine their precise mechanism of action. These studies are innovative in that these molecules have a novel unique chemical structure and mechanism of action, and there is a pressing need for small molecule HIF pathway inhibitors. These small molecules have great potential as candidate therapeutics for a large number of solid tumors that rely on the HIF pathway for their growth. These preclinical studies have the potential to directly benefit human health by increasing the survival of cancer patients, a main goal of the National Cancer Institute.

Program 2. Molecular Pathways and Biomarkers (MPB) The abnormal function of cancer cells can be investigated through the understanding of the changes in the activation and silencing of homeostatic cell signaling events. Identification of the functional proteome characteristic of a cancer cell permits the understanding of its underlying biology and provides unique opportunities for therapeutic targeting of critical node points as well as exploitation of biomarkers for disease prevention, detection and follow-up. The primary themes of the Molecular Pathways and Biomarkers program are in three main areas and reflect a major emphasis on basic science investigations of the cancer signalosome and its exploitation for clinical translation. The MPB Program's thematic areas include (1) tumor hypoxia and angiogenesis;(2) tumor-stroma interactions and (3) development of novel imaging technology and applications. The aims of the MPB program are: (i) to elucidate the mechanisms of the cellular signaling events that underlie angiogenesis and their targeting for therapy. (ii) To discover the key mediators of tumor-stromal interactions and identify biomarkers that predict primary tumor growth and metastasis at distant sites. (iii) To develop novel imaging applications to better detect cancer growth, help diagnose cancer type and provide more accurate tools for treatment follow-up. The MPB program is comprised of 40 core members from 15 departments within the School of Medicine. Currently there are 30 funded program core members. The total peer-reviewed funding per year of MPB is $15.23 M ($10.6 million direct), of which NCI funding represents $4.8 M ($3.3 M direct). In the P20 planning grant period of 2002-2008, MPB members published 445 articles. Intraprogrammatic collaborations accounted for 110 (24.7%) and interprogrammatic collaborations accounted for 144 (32.4%) of these publications.

PROJECT SUMMARY The Winship Cancer Institute's Cancer Cell Biology (CCB) Program studies the changes in biological function of human cells as a result of cell transformation. The CCB Program consists of 36 core members from 15 departments across Emory University, including the Schools of Medicine and Public Health and Emory College. The CCB program has two main themes: 1) Cell Survival and Death Mechanisms, and 2) Cell Adhesion, Communication and Metastasis. The CCB Program serves as Winship's scientific switchboard, a platform for discovery of novel signaling pathways, their biological validation, and an interchange of concepts among the CGE, DDT, and CPC Programs. Strategic reorganization toward increased cohesion and clarity has helped make the current project period a highly productive one for the CCB Program. CCB labs identified and targeted key molecular pathways involved in a variety of disease site-specific cancers. Its members published 300 cancer-related publications, many of which appeared in high-impact journals. The program furthered important initiatives, including the establishment of the In Silico Brain Tumor Research Center. Using start-up funding from internal pilot grants, CCB members expanded pilot science into nationally funded projects. CCB Program members currently have $23,452,748 in research grant funding (annual direct costs), of which $19,909,676 is peer-reviewed and $9,489,373 is NCI funded. As Winship's platform for scientific exchange, the CCB Program's productivity has been a win for all of the research programs and has spurred important inter-programmatic collaborations. The program has published 300 cancer-related publications in the current project period, most of which appeared in influential journals. Among these, approximately or neariy 15% represent intra-programmatic and 38% represent inter-programmatic interactions. CCB members and collaborators have contributed to the field of cancer research in a number of significant ways, and the program is poised for continued success going into the next project period.

DESCRIPTION (provided by applicant): Cancer is a major health problem worldwide and new therapies are critically needed, especially for glioblastoma the most fatal brain tumor. Unpublished studies in our lab have revealed a new bystander effect for tumor suppressor p53. Upon activation by chemo- or radiation therapies p53 induces the death of adjacent tumor cells, while sparing normal cells. We discovered that the effecter mechanism relies upon the secretion of galectin-3, a ?-galactose-recognizing lectin, which induces apoptosis. We also found that secreted galectin-3 reduced tumor formation in vivo. In this proposal we will extend these initial findings by dissecting the underlying mechanisms and determine whether Gal3 has clinical potential. We will determine the type of apoptotic signaling pathways activated in tumor cells by extracellular galectin-3 (Aim 1), whether secreted galectin-3 selectively binds to a specific cell surface receptor, with tumor-specific characteristics (Aim 2), and whether Gal-3 delivery can be used as a viable therapeutic for cancer using an in vivo mouse glioma model (Aim 3). Our working hypothesis is that p53 exerts a tumor suppressive bystander effect by stimulating exosomal secretion of Gal3, which in turn binds in a tumor-selective fashion to ? 1-integrin complexes due to unique N-glycanation in cancer, and induces a therapeutic effect by activating apoptosis. These studies are important because we identified a new p53-induced tumor suppressive mechanism mediated by soluble Gal3, which has therapeutic implications. Examining the role of extracellular Gal3 in glioma apoptosis and tumor growth in vivo is novel. These studies will provide proof-of-principle data for targeting cancer with Gal3 (or agonists such as peptidomimetics or small molecules). Successful outcome of this project will support the clinical translation of Gal3 for the treatment of malignant glioma and possibly other cancers, which is highly relevant to public health.

DESCRIPTION (provided by applicant): There is an urgent need to develop novel therapies for patients with highly malignant uveal melanomas in the eye. Patients with uveal melanoma die within 1-2 years of diagnosis despite current conventional therapies, including eye enucleation, brachotherapy and chemotherapy. Hypoxia drives tumor progression by activating angiogenesis, cell motility and metastasis, as well as metabolic adaptation to growth under a hypoxic environment and is a major factor in the resistance of cancer cells to radio- and chemotherapies. Hypoxia activates transcription factors of the Hypoxia-Inducible Factor (HIF) family that induce the expression of genes that encode pro-angiogenic factors and glycolytic enzymes essential for tumor growth and favor tumor invasion. Based on these findings, we formulated the central hypothesis that development of hypoxia and activation of the HIF pathway play a critical role in ocular cancer growth and spread, and that therapeutic targeting of this pathway using small molecule inhibitors will inhibit ocular tumor growth and metastasis. We have generated substantial preliminary data validating this concept. We show that our lead probe (KCN1) is a potent inhibitor of the in vivo growth of uveal melanoma in the eye (70% tumor size reduction) and its metastasis to the liver (50% reduction in number of metastases), while being extremely well tolerated. The overall goals of this proposal are to refine the structur of the novel HIF pathway inhibitor (HPI) chemical probes we developed, optimize their potency and pharmacological properties, leading to the identification of 1- 2 clinical lead probes that wil be ready to undergo IND-directed pharmacology and toxicology towards phase 1 clinical testing in patients with malignant uveal melanoma through the NCI NExT program. Our multi- disciplinary team has expertise in major aspects of chemical probe development and will divide the project tasks into the following aims: screening analogs of the parent compound in primary and secondary assays to identify and confirm chemical probes with improved potency and solubility (Aim 1); screening analogs of the optimized probes for improved pharmacology and formulation development (Aim 2); and determine the anti- tumor efficacy of the optimized lead probe(s) in orthotopic uveal melanoma models in mice (Aim 3).

DESCRIPTION (provided by applicant): Ocular melanoma is the most common primary eye cancer. Although the primary tumor in the eye can be controlled, frequent cancer spread to the liver results in significant mortality. There are currently no effective treatments for metastastic ocular melanoma in the liver. In the early/pre-metastatic stage of the disease, hypoxia induces the focal expression of chemokine/growth factor receptors in the eye tumor, rendering single melanoma cells responsive to activation by their respective paracrine ligands, stromal derived factor (SDF) and hepatocyte growth factor (HGF) produced in the liver. After extravasation into the circulation these cells home to the liver where they initially form micrometastatic foci that progress to dormant avascular colonies. In the late disease stage, an angiogenic switch leads to the formation of large hepatic macrometastases, which cause patient demise. We have discovered and characterized the anti-tumor properties of novel small molecule arylsulfonamides (ASAs), and obtained exciting preliminary data demonstrating that KCN1, our lead molecule, can potently decrease primary tumor growth, and the establishment and progression of hepatic metastases of uveal melanoma in an orthotopic mouse model we developed. We hypothesize that KCN1 alters the pro-tumorigenic signaling mediated by CXCR4/SDF and cMet/HGF that initiates metastasis, blocks STAT3 signaling involved in early progression in the liver and VEGF pro-angiogenic signaling that leads to macrometastasis, because Hypoxia Inducible Factor (HIF) can regulate these processes and KCN1 blocks HIF transcription. The goal of our proposal is to define the mechanism(s) underlying the anti-tumor effect of KCN1 at the different stages of disease progression. We will determine whether KCN1 inhibits i) the extravasation and survival of primary uveal melanoma cells into the circulation, and their homing to the liver (Aim 1), ii) the progression of micrometastatic foci to avascular melanoma cell colonies in the liver (Aim 2), and iii) the progression of avascular melanoma micrometastatic colonies to macrometastases in the liver by blocking micrometastases-induced angiogenesis (Aim 3). Our preliminary findings support our working hypothesis, as we demonstrate that KCN1 inhibits cMet cell surface receptor activation, STAT3 phosphorylation, and VEGF-mediated tumor angiogenesis in vivo. This work is important as it will better define the mechanisms of uveal melanoma metastases, identify therapeutic targeting points, and help the translation of the small molecules we identified towards becoming novel therapeutic agents for the control of metastasis of uveal melanoma.

? DESCRIPTION (provided by applicant): There is an urgent need to develop novel therapies for patients with medulloblastoma (MB), the most common malignant central nervous system (CNS) tumor in children. Current treatments include surgery, radiotherapy, and chemotherapy and result in 5-year survival rates of 40-90% depending on subtype. Moreover, children suffer important morbidity secondary to treatment, including neurological, intellectual and physical disabilities. The overall purpose of the present project is to investigate the role of the Brain-specific Angiogenesis Inhibitor 1 (BAI1) in cerebellar development and susceptibility to transformation, and explore new therapies for MB based on the related mechanisms. BAI1 is an orphan seven transmembrane G protein-coupled receptor (GPCR) specifically expressed in the brain, and belonging to the adhesion-type sub-family. Our new preliminary data show that BAI1 expression is significantly reduced in patients with MBs, and the promoter is epigenetically silenced, suggesting that BAI1 loss may facilitate MB formation. To test this in the physiological setting, we generated Bai1 knockout (KO) mice and found haploinsufficiency of Bai1 dramatically accelerates MB tumorigenesis in Ptch1+/- transgenic mouse models of MB, the first demonstration that a reduction in Bai1 dosage can promote MB formation in vivo. Interestingly, we detected enhanced Gli1/2 expression and a thicker external granule layer (EGL) during early postnatal cerebellum development in the Bai1 KO mice. Therefore, our preliminary studies link BAI1 with cerebellar development and neoplastic transformation. Based on these results, we hypothesize that BAI1 is a tumor suppressor in the cerebellum and that restoration of its expression with epigenetic therapy may represent a novel therapeutic intervention for MB. To test our hypothesis, we propose the following aims: (i) determine whether Bai1 loss accelerates MB formation in mice through abnormal activation of a growth-signaling pathway in the developing cerebellum, (ii) determine how BAI1 restoration in human MB cells can inhibit their growth, and alter their tumorigenic properties, and (iii) define the mechanisms of BAI1 inactivation in MB, and determine whether epigenetic reactivation of BAI1 expression has therapeutic effects in vivo. These studies are important as they increase our knowledge about developmental neurobiology in the CNS, and may lead to the development of novel therapeutic approaches for patients with medulloblastoma.

Project Summary/Abstract There is an urgent need to develop novel therapies for patients with medulloblastoma (MB), the most common malignant central nervous system (CNS) tumor in children. Current treatments include surgery, radiotherapy, and chemotherapy and result in 5-year survival rates of 40-90% depending on subtype. However, children suffer important morbidity secondary to treatment, including neurological, intellectual and physical disabilities. The overall purpose of the present project is to investigate the role of the ADGRB3 receptor in susceptibility of cerebellar transformation, and explore new therapies for MB based on the related mechanisms. ADGRB3 is an orphan seven transmembrane G protein-coupled receptor (GPCR) specifically expressed in the brain, and belonging to the adhesion-type sub-family. Our new preliminary data show that ADGRB3 expression is significantly reduced in patients with MBs of the WNT group, and the promoter is epigenetically silenced, suggesting that ADGRB3 loss may facilitate WNT-MB formation. We present evidence for the involvement of methylated CpG binding protein MBD2 and histone methyltransferase EZH2 in switch to a silent chromatin. Moreover, we show that reactivation of ADGRB3 can reduce cell proliferation and tumor growth, supporting a tumor suppressive role. To test this in the physiological setting, we generated ADGRB3 knockout (KO) mice, which we plan to cross with mice expressing mutant b-catenin in neural progenitors of the rhombic lip and dorsal brainstem, which are the cells of origin of WNT-MB. Based on these results, we hypothesize that ADGRB3 is a tumor suppressor in the cerebellum and that restoration of its expression with epigenetic therapy may represent a novel therapeutic intervention for children with WNT-MB. To test our hypothesis, we propose the following aims: (i) identify and target the epigenetic mechanism(s) underlying ADGRB3 gene silencing in WNT-MB, (ii) determine whether and how restoration of ADGRB3 expression can inhibit MB cell growth, oncogenic signaling and tumorigenic properties, and (iii) determine whether loss of ADGRB3 gene expression in the background of oncogenic Ctnnb1 activation predisposes mice to cerebellar transformation and MB tumor development. These studies are important as they increase our knowledge about developmental neurobiology in the CNS, and may lead to the development of novel therapeutic approaches for patients with medulloblastoma.