Bernard E. Weissman - US grants

Retroviruses have proven to be useful tools for investigating the processes of neoplastic transformation in mammalian cells. The majority of such studies have centered on the interactions of retroviruses with fibroblasts or hemapoietic cells. Few studies have been reported that deal with the effects of retroviral infection on cells of epithelial origin. The potential importance of such studies has been recently emphasized by the demonstration that dominant transforming genes isolated from human tumors are closely related to certain retroviral onc genes. We have recently isolated and characterized an epidermal keratinocyte cell line of murine origin designated Balb/MK, which is totally dependent on the presence of EGF for its continued proliferation. Upon exposure to extracellular levels of calcium greater than 1.0mM, these cells will terminally differentiate in a manner analogous to primary keratinocyte cultures. Transformation of these cells by Balb-, Harvey-or Kirsten-MSV results in the abrogation of the EGF requirement as well as a block in the pathway to calcium induced terminal differentiation. The research described in this proposal will investigate the interactions of other mammalian retroviruses on the growth and differentiation of Balb/MK cells. The effects of transformation by these different retroviruses on growth properties such as anchorage independent growth, growth factor requirements and transforming growth factor production will be compared. Specific stages of differentiation where these cells are blocked by various retroviruses will be identified by examining a variety of differentiated functions including production of cornified envelopes, desmosome formation, induction of transglutaminase activity and synthesis of different keratin species. Since both EGF and several retroviruses have been shown to stimulate protein kinase activity in cells, phosphorylation of tyrosine-containing proteins in these cells will also be determined.

Studies on chemical transformation in vitro have been based primarily on fibroblastic cells. While these results have led to insights into the processes of neoplastic transformation, the majority of human cancers are derived from cells of epithelial origin. Many investigators have attempted to develop a transformation assay based on epithelial cells. Although some systems have proved to be relatively successful for assaying chemical carcinogens, a reproducible assay based on a well-characterized epithelial cell line does not yet exist. Recently, an epidermal keratinocyte cell line of murine origin (Balb/MK) has been developed which possesses stable genetic properties useful for transformation studies. Balb/MK cells have an absolute requirement for epidermal growth factor (EGF) for their continued proliferation. They also will undergo terminal differentiation when exposed to high levels of extracellular calcium. Previous studies have shown that infection of these cells by a variety of mammalian retroviruses leads to an abrogation of the EGF requirement as well as a block in the calcium-induced terminal differentiation pathway. Thus, these cells are potentially useful as a general model for epithelial cell transformation. Initial studies have shown that the treatment of Balb/MK cells with the chemical carcinogen 20'-methylcholanthrene (MCA) leads to the isolation of EGF-independent cell lines in a dose-dependent manner. In addition, these chemically treated cells are also blocked in their ability to respond to calcium-induced terminal differentiation. Therefore, it is proposed that the Balb/MK cell line can be used as the basis for a chemical transformation assay for epithelial cells. The validity of the system will be examined by comparing the effects of known chemical carcinogens in vivo with their effects on the Balb/MK cells. The growth properties of the chemically transformed Balb/MK cells will be characterized by known in vitro parameters as well as assays for tumorigenic potential. The effects of carcinogen treatment on the differentiation phenotype of these cells will be examined by the use of known markers of epidermal differentiation. Finally, the interaction between viral and chemical transformation of these cells will be assessed.

Studies on chemical transformation in vitro have been based primarily on fibroblastic cells. While these results have led to insights into the processes of neoplastic transformation, the majority of human cancers are derived from cells of epithelial origin. Many investigators have attempted to develop a transformation assay based on epithelial cells. Although some systems have proved to be relatively successful for assaying chemical carcinogens, a reproducible assay based on a well-characterized epithelial cell line does not yet exist. Recently, an epidermal keratinocyte cell line of murine origin (Balb/MK) has been developed which possesses stable genetic properties useful for transformation studies. Balb/MK cells have an absolute requirement for epidermal growth factor (EGF) for their continued proliferation. They also will undergo terminal differentiation when exposed to high levels of extracellular calcium. Previous studies have shown that infection of these cells by a variety of mammalian retroviruses leads to an abrogation of the EGF requirement as well as a block in the calcium-induced terminal differentiation pathway. Thus, these cells are potentially useful as a general model for epithelial cell transformation. Initial studies have shown that the treatment of Balb/MK cells with the chemical carcinogen 20'-methylcholanthrene (MCA) leads to the isolation of EGF-independent cell lines in a dose-dependent manner. In addition, these chemically treated cells are also blocked in their ability to respond to calcium-induced terminal differentiation. Therefore, it is proposed that the Balb/MK cell line can be used as the basis for a chemical transformation assay for epithelial cells. The validity of the system will be examined by comparing the effects of known chemical carcinogens in vivo with their effects on the Balb/MK cells. The growth properties of the chemically transformed Balb/MK cells will be characterized by known in vitro parameters as well as assays for tumorigenic potential. The effects of carcinogen treatment on the differentiation phenotype of these cells will be examined by the use of known markers of epidermal differentiation. Finally, the interaction between viral and chemical transformation of these cells will be assessed.

A loss of genetic information has been postulated to play a key role in the development of human cancer. This concept has been supported by several types of investigations. Somatic cell hybrids between tumorigenic human cancer cells and their normal cellular counterparts are totally suppressed for tumor-forming ability indicating that tumorigenicity behaves as a recessive genetic trait. Furthermore specific chromosomal deletions have been identified in certain types of pediatric cancers implying the loss of certain critical genes during development of these tumors. Despite this strong circumstantial evidence for the existence of recessive cancer genes, little is known about the functions, number or locations of these genes. One appraoch to characterizing these recessive cancer genes is to study whole cell hybrids between different pediatric tumor cells and microcell hybrids into which normal human chromosomes have been introduced. A complementation analysis using whole cell hybrids would determine how many different cancer genes are invoved in the development of human cancer. The location of these genes can be revealed by the introductin of individual chromosomes into pediatric cancer cells. Further localization of the sites of these genes can be accomplished by the use of chromosomal transloations and RFLP analyses. The hybrid cell will also be characterized for in vitro growth parameters associated with transformation including serum growth factor requirements, anchorage independent growth, fibronectin expression and production of plasminogen activator. Tumorigenic potential will be assayed by the injection of these cells into nu/nu mice. Cellular protein synthesis will be characterized by two- dimensional polyacrylamide electrophoresis. These studies will provide new insights into the functions of recessive cancer genes in human cells and serve as the basis for future molecular cloning studies.

tissue /cell culture; biomedical facility; oncogenic virus; cryopreservation; restriction endonucleases; growth media; biopsy; diagnosis quality /standard; antiseptic sterilization;

The identification of human tumor suppressor genes has led to new insights into the mechanisms of human cancer development. Most of the known tumor suppressor genes were isolated by determining their chromosomal location using molecular markers or cytogenetics. We have taken the opposite approach by using a biological assay, tumor suppression, to map the locations of functional tumor suppressor genes via monochromosome transfer. In this manner, we have mapped a tumor suppressor gene for a Wilms' tumor cell line to a region of chromosome 11p15.5 flanked by two anonymous markers. Our preliminary physical map of this area suggests that this region contains less than 1000 kilobases (kb). Based on similarities between our functional studies and those concerning the RB and p53 genes, we propose that this gene represents another cell cycle control gene. To test this hypothesis, we will use positional cloning techniques to identify candidates for the functional tumor suppressor gene. We have gathered both cosmids and yeast artificial chromosomes (YACs) from this area for probing a fetal kidney cDNA library. To sort through this group, we will look for genes with abnormalities in the Wilms' tumor cell line. Alternatively, we will chose genes which show increased expression in the microcell hybrids relative to the parental cell line. The ultimate identification of this tumor suppressor gene will depend on its ability to suppress tumorigenicity in the Wilms' tumor cell line. Once we have isolated the tumor suppressor gene, we will search for sequence homologies with other known genes to further delineate its functions. After developing immunological reagents for the gene product, we will characterize expression of the gene in Wilms' tumor cell lines, normal cells and non-tumorigenic microcell hybrids. We will also determine its patterns of expression in developing human fetal material and in different types of normal human tissues. Finally, we will search for abnormalities in this gene in tumor samples from Wilms' tumor patients as well as patients with other malignant and/or genetic diseases which map to this same region of chromosome 11. The isolation of this tumor suppressor gene will represent the first such gene identified by a functional assay. The availability of another cell cycle control gene would broaden our understanding of tumor suppressor gene functions and may provide important clues about the process of normal mammalian tissue development.

DESCRIPTION: (adapted from the investigator's abstract) The identification of human tumor suppressor genes has led to new insights into the mechanisms of human cancer development. Some of the first tumor suppressor genes were identified through studies of pediatric malignancies including the RB and WTI genes. In the case of Wilms' tumors, further investigations have led to the discovery of other potential tumor suppressor genes on chromosomes 11, 16 as well as an unmapped familial form. In a complementary fashion, he has taken a functional approach by using a biological assay, tumor suppression, to map the locations of functional tumor suppressor genes via monochromosome transfer. He has now narrowed the location of a second Wilms' tumor suppressor gene, WT2, to an approximately 350 kb region on 11p15.5. Other studies have placed translocations in Beckwith-Weidemann Syndrome patients and loss of heterozygosity for Wilms tumor samples in this same region. In addition, this region of the human genome contains genes subject to inactivation by genomic imprinting. Intriguingly, many Wilms' tumors show a loss of imprinting for genes in 11p15.5 leading to either their inactivation or increased expression. The role of these epigenetic events such as imprinting in Wilms' tumor and other human cancer development remains unknown. Therefore, the isolation and characterization of the WT2 tumor suppressor gene would constitute a major advance in understanding these influences. During the last funding period, he developed a PAC/BAC/PI contig across the WT2 tumor suppressor gene region and began the identification of candidate genes in the area. In this competitive renewal application, he proposes to isolate the WT2 gene and characterize its status in the development of Wilms' tumor. In Specific Aim A, he will identify as many genes as possible for the WT2 tumor suppressor region by solution hybrid capture and analysis of genomic DNA sequence. In Specific Aim B, he will screen each candidate gene for correlative expression with tumor suppression in a microcell hybrid model system. He also will search for genomic alterations in primary tumor samples and examine the expression pattern in normal tissues. He hopes to limit the number of strong candidate genes by these criteria to five or less. In the last specific aim, he will identify the WT2 gene by screening for mutations and loss of expression in primary tumor samples. He will also transfer the gene into the G401 cell line to demonstrate functional tumor suppressor activity. The availability of the WT2 gene will broaden the understanding of tumor suppressor gene functions, provide important clues about the process of normal mammalian tissue development including genomic imprinting and impact upon treatment and detection of human cancer.

The development of human cancer proceeds in a multistep fashion involving both the loss of tumor suppressor gene function and the activation of oncogenes. While greater than one hundred oncogenes have been identified, only a modest number of tumor suppressor genes have been isolated. The discovery of most tumor suppressor genes has relied on the knowledge of their physical location on a particular chromosome; few functional assays for these genes exist. We have used the technique of microcell hybridization to map the location of tumor suppressor genes for adult and pediatric cancers. Demonstration of tumor suppression by transferring a single normal human chromosome into a human tumor cell line maps tumor suppressor gene(s) to that genetic unit. This method provides a powerful tool for identifying these genes in malignancies where extensive chromosome rearrangements and deletions mask their locale. Using this approach, we have mapped a functional tumor suppressor gene for a human squamous cell carcinoma (SCC) cell line to chromosome 11. By using a translocated chromosome T(X;11), we have localized the gene to the region between 11q13 to 11qter. We and others have also demonstrated tumor suppressor gene activity for renal cell carcinoma, lung carcinoma and cervical carcinoma on this same chromosome. Thus, our results may indicate that many types of epithelial malignancies result from the inactivation of the same tumor suppressor gene. In this grant application, we plan to localize the side of the tumor suppressor information to an approximately 1 mb region on chromosome 11. A. We have switched to the use of an in vitro raft assay which significantly decreases the time required for determination of tumor suppressor gene activity. B. We will begin to identify candidate tumor suppressor genes from this region. The availability of epidemiology data for these samples will allow us to gauge the relative importance of this tumor suppressor gene in the process of squamous cell carcinoma development. Squamous cell carcinomas appear to develop from a complex interaction of genetic, epigenetic and environmental factors. Exposure of environmental agents including alcohol, tobacco and food carcinogens may especially impact the development of oral squamous cell carcinomas. Identification and characterization of tumor suppressor genes affected by these events will allow investigators to determine the relative importance of each of them in the process of carcinogenesis. Recent reports linking other tumor suppressor genes with a genetic susceptibility to cancer emphasize the importance of these proposed studies.

DESCRIPTION (Adapted from investigator's Abstract): The ectodermal dysplasias consist of diseases with developmental abnormalities of ectodermal tissue primarily the hair, skin, teeth, and nails. Many forms of this disease present with other types of developmental disorders suggesting a common genomic alteration among them. One form, EEC, involves individuals who show evidence of ectodermal dysplasia, ectrodactyly and cleft palate. Previous work from this laboratory and others have mapped a form of ectrodactyly to human chromosome 7q21-22. Several families with EEC also show chromosomal abnormalities in this region. Thus, the Principal Investigator proposes that this area of the human genome contains the genes responsible for these developmental disorders. A YAC/cosmid/phage contig across this region of chromosome 7 has been developed and used to identify several candidate EEC genes. In this application, studies are proposed to further refine the EEC locus by a molecular analysis of the chromosome 7q21-22 target region in sporadic and familial EEC patients. To accomplish this aim, the investigator will look for submicroscopic deletions by PFGE and Southern analyses. At the same time, the technique of solution hybrid capture will be used to isolate candidate EEC genes from the target region. Patient material will then be screened for mutations and altered expression as well as look for mutations in these genes by SSCP analysis and loss of expression of one allele by polymorphic mRNA markers. Finally, patient samples will continue to be collected from sporadic and familial EEC patients to support these ongoing studies. This repository will be expanded to include other multiple phenotype disorders such as Rapp-Hodgkin and LADD which also include ED. It will then be determined if markers in the EEC critical region demonstrate linkage to these other types of ED-related families. These proposed studies offer a unique opportunity to isolate and characterize genes responsible for several well-characterized human developmental disorders. Furthermore, an understanding of the functions of these genes will eventually allow investigation of the molecular bases of reduced penetrance and variable expressivity, poorly understood genetic phenomena common to many inherited diseases. Finally, the determination of the relationship among the various birth defects associated with EEC will allow accurate genetic counseling to individuals with this disorder.

The identification of human tumor suppressor genes has led to new insights into the mechanisms of human cancer development. Isolation of the first tumor suppressor genes resulted from studies of pediatric malignancies including the RB and WT1 genes. In the case of rhabdoid tumors, frequent LOH on chromosome 22 has led to the discovery of a novel tumor suppressor gene designated INI1/hSNF5/BAF47. This gene codes for the human homolog of the yeast SNF5 gene, a member of the SWI/SNF chromatin remodeling complex. The SWI/SNF complex acts as a global transcriptional activator that alters nucleosome positioning on DNA in an energy-dependent manner. The role of altered chromatin remodeling during neoplastic progression has gained increasing recognition over the last several years. Recent reports strongly support the notion that INI1/hSNF5/BAF47 acts as a prototypical tumor suppressor gene. These include demonstrations that mutations and deletions occur in rhabdoid tumors, choroid plexus tumors and rhabdomyosarcomas, that LOH drives the removal of the remaining wild-type allele, that families carrying germline mutations develop these tumors at a high frequency and that germline inactivation in mice leads to the development of rhabdoid-like tumors. We have found that re- expression of SNF5 in rhabdoid tumor cell lines causes growth inhibition accompanied by a dramatic rise in p16INK4A protein levels. Based on these preliminary studies as well as the known functions of the SWI/SNF complex and other relevant scientific literature, we hypothesize that alterations in the INI1/SNF5 component of the hSWI/SNF complex contribute to human tumor development by blocking the induction of p16INK4A and disrupting normal cell cycle control. In this application, we will test this hypothesis by determining the mechanism by which loss of activity of this gene affects p16INK4A protein levels using biochemical, biological and animal model assays. In Specific Aim number 1, we will determine the cell cycle control pathways regulated by INI1/SNF5 and the relevant domains for these activities. In Specific Aim number 2, we will ascertain the biochemical effects of INI1/SNF5 loss on SWI/SNF function and whether the protein directly interacts with the p16INK4A promoter. In Specific Aim number 3, we will develop a mouse model for choroid plexus carcinomas by crossing TAg transgenic mice to SNF5+/- mice. The characterization of the INI1/SNF5 gene role in regulation of gene expression will broaden our understanding of tumor suppressor gene functions, provide important clues about the role of chromatin remodeling complexes in normal and neoplastic development and impact upon treatment and detection of these devastating pediatric cancers.

DESCRIPTION (provided by applicant): The identification of tumor suppressor genes has led to new insights into the mechanisms of human cancer development. The normal functions of these genes often lie in the control of gene expression, especially in the realm of cell cycle control and cellular differentiation. Several recent studies have implicated aberrant activity of chromatin remodeling complexes in the development of human cancer. Mutations in the INI1/SNF5 gene, a component of the SWI/SNF chromatin remodeling complex, occur in the majority of malignant rhabdoid tumors. The SWI/SNF complex acts as a global transcriptional activator that alters nucleosome positioning on DNA via an energy-dependent mechanism. Others and we have also demonstrated the loss and/or mutations of both human SWI2 homologs, BRG1 and BRM, in human tumor cell lines and primary tumors. Loss of expression of both proteins abrogates Rb-mediated cell cycle arrest. However either protein appears to regulate the expression of key cancer progression genes including Ecadherin and CD44. Furthermore, while BRG1v/- mice develop adenocarcinomas, BRM-/- mice do not show an increased tumor incidence. Therefore, the mechanism by which loss of expression of either or both proteins contributes to the etiology of human cancers remains unresolved. We have observed an association between BRG1-induced gene expression and loss of promoter methylation in human tumor cells, especially Non-Small Cell Lung Carcinomas (NSCLC). Based on these studies, we hypothesize that loss of SWI/SNF complex activity through inactivation of BRG1 and/or BRM fuels genomic instability during human tumor progression by facilitating gene silencing through DNA methylation. To test this hypothesis, we require a better understanding how loss of expression of BRG1 and BRM proteins alters the biological properties of tumor cells and the chromatin structure of cancer related genes. In this application, we will identify overlapping and independent biological and biochemical functions of these proteins in the first specific aim. In the second specific aim, we will determine how the loss of these proteins contributes to lung tumor development using a human tracheobronchial cell culture model. In the last specific aim, we will characterize the effects of BRG1 and BRM loss on the chromatin structure and DNA methylation status of important target genes. The dissection of the role of these genes in human cancer development will broaden our understanding of their normal biological and biochemical functions, provide new insights into the control of DNA methylation and impact upon treatment and detection of this clinically important tumor.

The goal of the Animal Procedures Core Facility is to provide a central facility staffed by two skilled individuals to promote reproducible experimentation in laboratory animals. Services include drug screening, tumorigenicity testing and reagent production and small animal colony management. The Core is led by Dr. Bernard Weissman and Ms. Natalie Edmunds. The Core adds value to the Center by providing the animal tumor model expertise lacking in most laboratories, reducing the overall cost of such studies, ensuring a high level of reproducibility for animal studies and fostering interactions between Cancer Center members. Highlights of research supported by the Core include: highly-skilled maneuvers to characterize novel antineoplastic reagents including: intra-tibial inoculations, oral garage, and new treatments of NPC tumors with adenoviral reagents. Future plans for the Core include the addition of new Cancer Center users, expansion of colony management services including genotyping in conjunction with the Animal Models Core Facility and integration of small animal imaging with the new Small Animal Imaging Core Facility.

Epigenetic changes in gene expression play an important role in the development and progression of human non-small cell lung carcinoma (NSCLC). Recent studies have shown that patients with tumors possessing one or more epigenetically silenced genes often show a poorer overall survival. Most reports have focused upon two major mechanisms to account for epigenetic modifications, DNA methylation and alterations in histone modifications. However, altered nucleosome positioning at gene promoters represent another important mechanism by which genes can be epigenetically regulated. Indeed, several studies have now implicated chromatin-remodeling complexes in the genesis of epigenetic silencing in human tumor development. In particular, the SWI/SNF (mating type switch/sucrose nonfermenting) chromatin remodeling complex appears like a strong candidate for contributing to epigenetic alterations in NSCLC. Originally identified in yeast, the SWI/SNF complex alters chromatin structure by remodeling nucleosomes through an ATP-dependent process. Loss or expression of the SWI2 ATPase homologs, BRG1 and BRM, by promoter methylation and/or mutations of either or both genes occurs in ~25% of human NSCLC cell lines and ~10% of primary human NSCLCs. Importantly, several groups including our own have shown a correlation between loss of expression of BRG1/BRM and poor prognosis in NSCLC patients. Our published reports and preliminary results demonstrate that reexpression of BRG1 or BRM in deficient NSCLC cells induces expression of many epigenetically silenced genes. Therefore, we hypothesize that loss of SWI/SNF complex activity represents a novel mechanism for gene silencing during NSCLC development. To test this hypothesis, we propose a synergistic research plan using cell culture and genetically engineered mouse models. Specifically, we will determine which gene promoters are activated by BRG1 and/or BRM after reexpression in deficient NSCLC cell lines, discover gene promoters that undergo silencing after loss of BRG1 expression in NSCLC cell lines and assess the effects of BRG1 and/or BRM loss on tumor development in a genetically engineered mouse model for NSCLC. The successful completion of the proposed studies will provide valuable insights into the mechanisms of epigenetic silencing during NSCLC development and of SWI/SNF chromatin remodeling as well as generate a novel genetically engineered animal model for further basic and translational studies. Furthermore, if DNMT or HDAC inhibitors are not effective in reversing gene silencing in the subset of NSCLCs that lack BRG1 and BRM expression, they may require a novel approach for treatment and prevention of progression. By identifying the unique chromatin changes that occur in these tumors and how they differ from BRG1 and/or BRM-positive tumors, we can initiate rational drug design studies to find the reagents to treat this deadly disease.

Animal Models Core Facility We are combining two previous cores dealing with mouse cancer nnodels into a single Animal IVIodeis Core that functions to help LCCC members with all aspects of mouse-related research. Core staff assists with animal handling, xenograft tumor models, colony management, therapeutic trials, imaging studies, allele phenotyping and the design and production of genetically engineered mice (GEM). This core adds value to the Cancer Center by enabling cost-efficient murine testing even by members who do not have significant infrastructure and expertise for animal work. The core is co-directed by Chariene Ross and Dale Cowley, with faculty co-advisors Norman Sharpless and Bernard Weissman. Dr. Sharpless and Dr. Cowley have been added to the leadership since the last cycle to add expertise in GEM models and expand core capabilities. Experimental help spans the spectrum from transgenic / knockout allele production and design to allele phenotyping, animal imaging and therapeutic testing of novel anti-cancer agents in xenograft and genetically engineered tumor models. The core has 39 users (97% use by peer reviewed members for Animal Studies). Significant growth has occurred in the animal studies component of the core during the last year with core staff operating beyond overall capacity since August 2009. We request an increased budget of $283,591 that will represent 15% of the total Animal Models Core budget to promote expanded use. The Animal Models Core will grow significantly during the next cycle, driven largely by increased NIH funding to UNC investigators, the 50% increase in campus animal space and a strategic plan featuring cancer genetics and preclinical therapeutics testing. The Genetic Medicine building has recently opened with space for over 40,000 additional cages. The Imaging Research Building will open in 2013 with 2,000 additional cages for longitudinal mouse therapy and Imaging protocols. The Animal Models Core is well positioned to become the nexus for Cancer Center members exploiting the power of GEMs and xenograft tumor models for basic and translational cancer research.

? DESCRIPTION (provided by applicant): Inactivation of tumor suppressor genes constitutes a driving mechanism in oncogenesis. In normal cells, tumor suppressors act as central hubs coordinating cell cycle, cell death, and cell fate via regulation of gene expression. The loss of tumor suppressor activities disrupts these processes, permitting aberrant cell growth. Recent work has shown that ~20% of human cancers carry mutations in at least one subunit of the SWI/SNF chromatin-remodeling complex. While genetic, animal and cell culture studies have established that several SWI/SNF subunits function as tumor suppressors, few studies have comprehensively established the mechanisms by which they act. We recently identified inactivating mutations in the SWI/SNF ATPase subunit, SMARCA4, in the majority of small cell carcinoma of the ovary of hypercalcemic type (SCCOHT), a poorly- differentiated and aggressive tumor, with concomitant protein loss in nearly 90% of primary tumors. We also found loss of expression of the alternative SWI/SNF ATPase subunit, SMARCA2, in all tumors lacking SMARCA4. In contrast to most adult cancers, SCCOHT genomes remain predominantly diploid with rare secondary mutations in other cancer genes. Therefore, we hypothesize that 1) SMARCA4 mutations drive tumor development through disruption of normal differentiation; and 2) that these mutations will define a global therapeutic vulnerability in cancer. To test this hypothesis, we propose to 1) Investigate the epigenetic consequences of SMARCA4/A2 re-expression; 2) Model the histogenesis of SCCOHT; and 3) Identify and validate targets for tumor treatment. To carry out the proposed studies, we take advantage of the cutting-edge technologies available at the University of British Columbia (UBC), the Translational Genomics Research Institute (TGen) and the University of North Carolina (UNC) and the complementary expertise of the 3 multiple principal investigators including ovarian cancer pathology and biology (UBC), systems biology and translational applications (TGen) and chromatin biology and cancer epigenetics (UNC). We will also generate novel cell culture and animal models to parse the mechanisms by which inactivation of SMARCA4 and/or SMARCA2 drive SCCOHT development and to identify and validate new reagents for treatment of this deadly disease. In addition, the predominance of SMARCA4 mutations along with the scarcity of secondary mutations in other cancer genes yields a unique model for understanding the specific contributions of SWI/SNF complex mutations to human cancer development. Thus, the results of these studies will significantly impact the diagnosis and treatment of SCCOHT by providing seminal insights into the mechanisms that drive its development, including the disruption of SWI/SNF complex function. Furthermore, the ubiquitous nature of SWI/SNF complex mutations in human malignancies broadens the scope of our findings to other deadly human cancers including lung cancer, renal cell carcinoma and brain tumors.

ABSTRACT- Animal Models (AM) shared resource The Animal Models Shared Resource (AM) assists LCCC members with all aspects of mouse-related research. It consists of three arms that seamlessly interact to provide a broad range of experimental approaches and expert advice. The SR staff assists with complex surgical techniques, animal handling, xenograft tumor models, colony management, therapeutic trials, imaging studies, allele phenotyping and the design and production of genetically engineered mice (GEM) and provides access to a colony of immunocompromised animals for PDX studies. This SR adds value to the Cancer Center by enabling cost- efficient murine testing for all members including those without significant infrastructure and expertise for animal work. The SR is co-directed by Charlene Ross, Dale Cowley and David Darr, with faculty co-advisors. Ms. Santos has served as a Facility Director providing expertise in mouse xenograft models while Dr. Cowley has led efforts in the design and generation of GEM models since 2005. Mr. Darr has been added to the leadership since the last cycle to add expertise and administrative oversight in pre-clinical testing in GEM models and expand SR capabilities. The AM has grown significantly during the last cycle, driven largely by increased NIH funding to UNC investigators, the 50% increase in campus animal space and a strategic plan featuring cancer genetics and preclinical therapeutics testing. The SR has 61 LCCC users, the remaining 22 Non-member users are collaborators located at Cancer Centers across the country. Importantly, Marsico Hall opened in late 2014 with 2,000 additional cages for longitudinal mouse therapy and imaging protocols. The availability of this space has also enhanced the interactions between this SR and the Imaging SR. The AM has become the nexus for Cancer Center members exploiting the power of GEMs and xenograft tumor models for basic and translational cancer research. The result is publications in high impact journals such as Cell, JNCI, JCI, Nature, Proc. Natl. Acad. Sci. USA., Cancer Cell, Cancer Discovery and CCR. We request an increased budget of 226,107 that will represent 10% of the total AM budget to promote expanded use.

PROJECT SUMMARY This proposal seeks support for 6 postdoctoral fellows for a training program focused upon cancer epigenetics, a critical area of cancer research. Demand for training in this area has steadily increased, highlighting the crucial need for robust training programs. Surprisingly, no NCI-funded training program focused on this topic currently exists. UNC-Chapel Hill has recently emerged as a ?hub? for epigenetics and cancer research, with over 35 faculty who actively perform epigenetics-focused research. Thus, we propose to create the first Cancer Epigenetics Training Program (CETP), leveraging our expertise in the area of chromatin and cancer biology with the resources of the University and the Lineberger Comprehensive Cancer Center (LCCC). This pool of preceptors will allow fellows to engage in interdisciplinary studies across the subspecialties of cancer epigenetics: Histone Biology, DNA Methylation, Chromatin Remodeling, RNA and Chemical Biology. The CETP program will be administered by a Director and Associate Director, with advice from a Training Oversight Committee and External and Internal Advisory Boards. Program evaluation will also include preceptor and trainee input via an annual, anonymous survey. Access to the LCCC?s core facilities, supported by the NCI Core Grant (rated ?exceptional?), will further enhance the training environment. The LCCC will also provide significant support by funding of training events, yearly EAB visits, travel and lodging costs for interviews and for trainee travel to scientific meetings and salary support for program directors and staff. Fellows will jointly apply to the CETP and one or more mentor laboratories and undergo a rigorous selection process. Upon appointment, each fellow will develop a training plan approved by the preceptor, the Director, and Associate Director and establish a mentorship committee. During their first 2 years in the CETP, each fellow will participate in the following training events: (i) CETP colloquia that cover the breadth of epigenetics in the context of cancer development and treatment, (ii) a yearly CETP symposium to present current research projects, (iii) training in ethics in research (iv) a monthly research club, and (v) a grant writing and grant review program. Based on their individualized training plans, fellows may also participate in relevant training events sponsored by the LCCC. Progress will be monitored annually by the Director and Associate Director, along with the Training Oversight Committee, before renewal for Year 2 support. Although fellows are supported for two years, our trainees will be active in the CETP program and participate in multiple events throughout their tenure at UNC. The CETP will recruit trainees from underrepresented minorities through outreach to historically black and Hispanic universities and our yearly symposium on cancer epigenetics. The trainees who complete the CETP will be poised to become independent investigators focused upon this vital new area of cancer research.

Project Summary Despite the dismal prognosis for patients with advanced lung squamous cell cancer (LUSC), few effective treatments of only limited benefit exist. Cytotoxic therapy with radiation remains the mainstay for most patients. However, it carries a relatively narrow therapeutic index and patients invariably suffer treatment-related systemic toxicities. The development of new targeted therapies for LUSC remains a high priority. One of the most significant discoveries from lung cancer genome sequencing is the frequent (~30%) alterations of the KEAP1- NRF2 signaling pathway. The NRF2 antioxidant signaling pathway constitutes the primary cellular defense system against oxidative stress. Several mechanisms responsible for governing NRF2 activity are known, however they have proven largely intractable for therapeutic intervention (eg. transcription factors). Recent studies have revealed that several protein kinases functionally impact NRF2, although a global evaluation of how the kinome instructs NRF2 biology remains untested. Being among the most druggable of protein classes, kinases and phosphatases offer attractive targets for NRF2-directed treatment intervention. Our preliminary data have established reciprocal communication between NRF2 and protein kinases, including upstream modifiers and downstream effectors. Therefore, we hypothesize that the kinome encompasses key regulators of NRF2 signaling and holds novel therapeutic targets for NRF2-active lung cancer. We have assembled a unique multidisciplinary team of investigators with experience in LUSC molecular signaling, cancer cell biology, animal and cell culture models, proteomics, and clinical therapeutics to decipher the interactions between the kinome and NRF2 signaling, identify novel therapeutic targets and analyze them in pre-clinical models. Our specific aims include: (1) IDENTIFY KINASES THAT REGULATE NRF2; (2) IDENTIFY NRF2-RESPONSIVE KINASES AND PHOSPHATASES; and (3) EVALUATE KINOME FUNCTION IN NOVEL MODELS OF NRF2 ACTIVE LUSC. This project shows strong innovation through kinome proteomic profiling of LUSC tumor samples and cell lines, high-throughput chemical screens, gain- and loss-of-function genetic screens, and the application of unique 2- and 3-dimensional cell culture models. Our experiments employ gene targeted-transformation of human bronchial epithelial cells and novel genetically engineered mouse models and derived cell lines. The results of this work will reveal protein kinases that functionally impact NRF2 biology, and in doing so may lead to new effective treatments for LUSC and other NRF2-active tumors, including head and neck cancer, bladder cancer, and ovarian cancer. Ultimately, the successful completion of our proposed studies will provide a roadmap for similar efforts on targeted therapeutic discovery in these other human malignancies.