Michael A. Dyer - US grants

DESCRIPTION (provided by applicant): The coordination of cell division with differentiation is critical for appropriate retinal formation. If this process is disrupted, vision may be compromised, and in some cases, malignant retinoblastoma may form. The Rb family of proteins (Rb, p107, and p130) lies at the heart of the regulatory network that controls proliferation during development and the suppression of tumorigenesis. Based on our preliminary data, we hypothesize that these proteins play distinct roles regulating proliferation and differentiation during mouse retinal development. This hypothesis stands in contrast to the long-standing assumption that Rb and p107 play redundant roles in the murine retina. In humans, we suggest that the complete absence of any role for p107 and a minimal role for p130 in regulating cell cycle exit during retinal development results in a unique susceptibility to early childhood retinal tumors. In Specific Aim 1, we will study the roles played by each Rb family member in regulating retinal progenitor cell proliferation using mutant mice, recombinant retroviral vectors and retinal explant cultures specifically optimized for such studies. Particular attention will be directed toward elucidating redundant vs. compensatory mechanisms among family members. In Specific Aim 2, gain- and loss-of-function studies will be performed to determine the specific requirement of each Rb family member in the specification and differentiation of the seven classes of retinal cell types. Emphasis will be placed on distinguishing changes in retinal development due to proliferation alterations from direct effects on developmental processes. In Specific Aim 3, we will compare our data on retinal development in the mouse to developmental data obtained from normal human and monkey samples, along with human retinoblastoma samples to address the question of human susceptibility to retinal tumors. Specific mechanisms underlying retinoblastoma formation will be tested in experiments designed to faithfully model retinoblastoma in the mouse. These studies bring together an extensive knowledge base and reagents from the fields of cell cycle control and retinal development to address fundamental questions regarding retinal development and disease.

DESCRIPTION (provided by applicant): During retinal development, the decision to exit the cell cycle must be precisely coordinated with cell fate specification and differentiation to ensure that the correct proportion of each cell type is generated. The Rb family of proteins (Rb, p107 and p130) regulate cell cycle exit, cell fate specification, differentiation and survival during development. In the previous funding period, we focused on the unique and overlapping roles of the Rb family in regulating retinal progenitor cell proliferation during development in mice and humans. These studies moved the field forward and allowed us to develop some of the first knockout mouse models of retinoblastoma. We used these and other preclinical models of retinoblastoma to test new therapies and our research has directly impacted an ongoing clinical trial (RET-5) at St. Jude. In this grant proposal, we will extend our previous studies and focus on the role of the Rb family in neuronal cell fate specification and differentiation. Our preliminary data suggest that Rb regulates these two processes through distinct mechanisms. Specifically, we propose that Rb regulates rod photoreceptor cell fate specification by repressing aE2Fs on the Chx10 and Pax6 promoters and Rb regulates rod photoreceptor differentiation through rE2Fs and histone modification to activate the Nrl, Nr2e3 and Crx promoters. This is the first example of such complex regulation of neurogenesis by a tumor suppressor. In contrast to rod photoreceptors, horizontal neurons in the developing retina do not require the Rb family for cell fate specification, migration or differentiation. However, in the absence of the Rb family, mature horizontal cells re-enter the cell cycle and form metastatic retinoblastoma while maintaining their differentiated state. These data challenge the widely held belief in developmental biology that differentiation and proliferation are incompatible in neurons. More importantly they suggest that cells that rely upon Rb for their normal development such as rod photoreceptors are resistant to tumor formation following Rb family inactivation but cells that do not rely upon the Rb family for their normal development such as horizontal cells are more susceptible to tumorigenesis following Rb family gene inactivation. We will determine here if the role of the Rb family during normal development in different cell types directly influences their susceptibility to become tumorigenic.

DESCRIPTION (provided by applicant): Cells in the ciliary epithelium (CE) of the eye can clonally expand in culture to produce spheres of cells that differentiate into retinal neurons and glia. It is believed that these spheres are produced by retinal stem cells (RSCs) and hold promise for cell-based therapies to treat degenerative retinopathies. To improve our understanding of RSC expansion and differentiation, we have carried out a series of preliminary studies on human and mouse CE-derived spheres. Our data have led us to reconsider the current retinal stem cell model and to propose a new hypothesis on the expansion and differentiation of CE-derived cells. We propose that the pigmented CE cells rather than RSCs, expand to form spheres and subsequently transdifferentiate into rods, bipolar cells and Muller glia. This proposal will test if CE-derived spheres form by proliferative expansion of pigmented CE cells or by an RSC mechanism (Aim 1). Culture experiments and in vivo transplantation studies will also be done to test whether pigmented CE cells transdifferentiate into retinal neurons and glia, or if those cells are produced from retinal stem/progenitor cells (Aim 2). Distinction between these 2 hypotheses must be made, because the pathways regulating proliferation and differentiation in retinal stem/progenitor cells differ from those in epithelial cells, and efforts to optimize expansion and differentiation of CE-derived cells must focus on the pathway that reflects that system's biology. The successful completion of these experiments may substantially affect future development of cell-based therapies for millions of people worldwide who suffer from retinal degeneration. This research proposal will also move the RSC field forward by resolving several outstanding questions regarding the proliferation and differentiation of CE-derived cells in culture and in vivo.

DESCRIPTION (provided by applicant): The decision to exit the cell cycle during retinal development must be precisely regulated to ensure that each cell type is generated in the correct proportion. In addition, when cell cycle regulation becomes perturbed during retinogenesis it can lead to microphthalmia, retinal degeneration or retinoblastoma. The Rb family of proteins (Rb, p107 and p130) lie at the heart of the cell cycle machinery that regulates retinal progenitor cell proliferation. Interestingly, we found that, distinct from its role in regulating retinal progenitor cell proliferation, Rb is required for rod photoreceptor development. In addition, Rb family proteins appears to be required in differentiated horizontal cells to prevent them from re-entering the cell cycle and our preliminary data suggest that Rb may be required for the reorganization of horizontal cell processes that occurs during the later stages of differentiation. Specifically, in Rb-deficient retina, horizontal cells extend their processes into the outer nuclear layer (ONL), while in the normal mature retina these processes are restricted to the outer plexiform layer (OPL). In this research proposal, our first aim is to define the role of Rb in horizontal neuron development, by determining whether Rb is required cell autonomously for horizontal cell maturation, or if defects in synaptogenesis in Rb-deficient retina are triggered secondarily by the absence of rods. In addition, we will visualize the maturation of horizontal cells in the absence of rods to determine whether the defects observed in Rb-deficient retinae are caused by a failure of horizontal cells to form appropriate laterally-arranged processes during development, or if they reflect reorganization of the processes after they extend laterally along the outer plexiform layer (OPL). This research will be done primarily in Brazil, at the Biophysics Institute of the Federal University of Rio de Janeiro (IBCCF/UFRJ) in collaboration with Dr. Rodrigo Martins, as an extension of the NIG Grant R01EY014867-06. The project period will be from 07/01/2009 until 06/30/2012. We believe the findings from the proposed experiments will provide essential data for Aim 3 of the parent grant (R01EY014867-06) and will directly contribute to a better understanding of the roles of Rb tumor suppressor in retinogenesis. PUBLIC HEALTH RELEVANCE: The role of the tumor suppressor Rb in horizontal neuron development This proposal aims to establish a scientific collaboration between the Biophysics Institute of the Federal University of Rio de Janeiro, Brazil (IBCCF/UFRJ) and Developmental Neurobiology of St Jude Children's Research Hospital, USA. The research proposed here is essential for the parent grant (R01EY014867-06), but was beyond the scope of the original grant. The LMIC collaborator (Dr. Martins) has worked as a postdoctoral research fellow under the supervision of the PI (Dr. Dyer) for the last 4 years and was recently appointed an Associate Professor of the IBCCF/UFRJ. In Brazil, Dr. Martins will have access to all of the resources of the Laboratory of Neurogenesis led by Dr. Rafael Linden and to all the facilities of the IBCCF/UFRJ. The approval of this project will be instrumental to the continuation of the scientific collaboration between Dr. Martins and Dr. Dyer. We believe that advanced training in sophisticated technologies and access to innovative perspectives and concepts will allow the development of high merit scientific projects in Brazil and will help decreasing the disparity in the research performed in this country as compared to most US-based research centers. Our goal is to build on the acquired training and experience to develop this innovative research project that will lead to a significant advancement of the aims of the parent grant and directly contribute to a better understanding of the roles of Rb tumor suppressor in retinogenesis. In the parent grant aim 3 is focused on elucidating the role of the entire Rb family (Rb, p107 and p130) in horizontal cell maturation. One essential question that is not addressed in the parent grant is whether the Rb family is required cell autonomously for the later stages of horizontal neuron development (e.g. synaptogenesis). Because traditional Rb-knockout mice die in utero, we conditionally inactivated Rb by mating RbLox/- mice14 with Chx10-Cre mice. The transgenic mouse model to be used in this study (Chx10-Cre;RbLox/-) is the only one in which the rod photoreceptors fail to form and do not adopt another cell fate or die. The pattern of recombination of RbLox alleles by the Chx10-Cre transgene is also unique. This retina is chimeric, having apical-basal stripes of Rb-deficient retina, in which Rb has been inactivated, flanked by wild-type stripes of retina. This mosaic distribution of Rb-deficient and wildtype retinal cells facilitates distinction between Rb's cell autonomous roles and non-cell autonomous roles in retinogenesis. We propose a unique combination of classic approaches to retinal synaptogenesis, including electron microscopy, and sophisticated genetic mosaic analysis. We also plan to perform long-term 2-photon live imaging of Rb-deficient retinae, because this cutting-edge approach will allow us for the first time to monitor the continuum of horizontal cell maturation in the absence of rods. These experiments will advance our understanding of fundamental process that lie at the heart of retinal development and are required for vision. They may also provide valuable insight into the mechanisms of synaptogenesis that may ultimately benefit efforts to replace photoreceptors lost to retinal degeneration using cell-based assays.

DESCRIPTION (provided by applicant): Retinoblastoma is a childhood cancer of the developing retina that begins in utero and is diagnosed in the first years of life. It is the third most common form of cancer in infants after leukemia and neuroblastoma. If left untreated, retinoblastoma is virtually always fatal, and even with the best treatment, the 10% survival rate for metastatic retinoblastoma is among the worst for any pediatric cancer. In the U.S., most patients survive because the cancer is detected early before it metastasizes, but saving the patient's life often requires surgical enucleation. This is particularly debilitating for children ith the most advanced form of bilateral retinoblastoma; approximately half of those patients lose at least 1 eye and are permanently vision impaired. The goal of this translational research is to identify new treatments to save the vision of children with retinoblastoma and the lives of children with metastatic progression. We have identified 3 novel, exciting molecular-targeted therapies that show promise in laboratory studies. In this application, we will perform comprehensive preclinical testing of these 3 agents. The results from these studies will be used to design the next retinoblastoma clinical trial at St. Jude Children's Research Hospital. Therefore, the successful completion of the Aims presented here will directly impact human health. One unique and innovative aspect of this application is the multidisciplinary team that we have assembled, which is made up of ocular oncologists, pediatric oncologists, developmental neurobiologists, chemists, and pharmacologists. This team has a proven record of moving basic science discoveries into clinical trials and then sharing our findings with the national and international communities to affect the greatest change in global human health. Another innovative aspect of the application is our expertise in retinoblastoma translational research and development of preclinical models, which we have optimized and characterized for these studies. No other center has the team, resources, expertise, or tools available to perform the studies presented here and move the most promising findings directly into a clinical trial.

DESCRIPTION (provided by applicant): Retinal degeneration affects millions of people around the world each year. For the past decade, my lab has studied the coordination of proliferation and differentiation in the developing retina and in proliferative diseases of the retna such as retinoblastoma. Recently, we made a startling discovery that has fundamentally altered our understanding of the molecular and cellular mechanisms of retinal development and may also have a major impact on efforts to restore vision in some patients with retinal degeneration. We discovered that individual retinoblastoma tumor cells express multiple developmental programs simultaneously. This occurs through deregulation of the epigenetic programs that are directly or indirectly regulated by the RB1 protein. To explore this exciting finding further, we developed a novel experimental system to quantify the epigenetic reprogramming of individual retinal neurons by using 4 factors (Oct4, Klf4, Sox2, and Myc) and somatic cell nuclear transfer. We discovered that the epigenetic barriers to reprogramming dramatically differ across retinal cell types, and they are developmental stage-specific. Moreover, we have used the Sasai 3-dimensional culture system to show for the first time that mouse iPSCs can form the optic cup and differentiated retinae. One of the most exciting results from these experiments is that our iPSC lines derived from retinal neurons bypass the normal transition through anterior neuroectodermal specification. Instead, they retain retinal epigenetic memory and form exclusively retinal progenitor cells that differentiate into laminated retinae. The iPSCs derived from retinal neurons retain their epigenetic retinal memory for at least 50 passages, whereas iPSCs generated from genetically identical MEFs rarely produce retinae in this system. We have now shown that the photoreceptor precursors derived from retinal iPSCs can integrate into the retina; thus, iPSCs generated from mature retinal neurons may provide a renewable source of photoreceptor precursors for cell-replacement therapies to restore vision in those who suffer from retinal degeneration. This innovative research proposal will advance our understanding of the role of epigenetics in retinal development and begin to elucidate the molecular mechanisms and cell type-specific target genes involved in that process. It will also provide crucial preclinial data on the use of retinal-derived iPSCs for future clinical trials of photoreceptor-replacement therapy to treat retinal degeneration.

Rhabdomyosarcoma (RMS) is the most common soft-tissue sarcoma of childhood and adolescence. Using contemporary multimodal therapies, more than 75% of patients with localized RMS are cured However, patients with recurrent disease experience 5-year survival rates of only 30%, and this outcome has not significantly improved in more than 20 years. Over the past 6 years, comprehensive genomic, epigenomic, proteomic, and drug-sensitivity profiling of RMS has been completed and integrated into a central database (iRDb). In addition, a multidisciplinary RMS translational research team (RMS-TRT) has been assembled to use the iRDb to identify tumor vulnerabilities that can be exploited therapeutically. Three signaling pathways have been identified that appear to be vulnerable to novel molecular-targeted regimens, which have shown promise in laboratory studies. In this proposal, comprehensive preclinical testing of those regimens (e.g., pharmacokinetics, pharmacodynamics, imaging and efficacy studies) will be performed to determine the most effective treatment. The results from these studies will be used to design the next RMS clinical trial at St. Jude Children's Research Hospital. Therefore, the successful completion of the Aims presented here will directly affect human health. One unique and innovative aspect of this proposal is the Childhood Solid Tumor Network (CSTN), a website (http://www.stjude.org/CSTN/) through which all of the iRDb data are shared freely with the international biomedical research community with no obligation to collaborate. The majority (>95%) of the data available through the CSTN is unpublished, but all of it is shared to accelerate discovery and advance cures for pediatric cancer. To date, 314 requests from 122 investigators at 46 institutions across 11 countries have been fulfilled. I have a proven record of moving basic science discoveries into clinical trials and then sharing findings with the national and international research communities to affect the greatest change in global human health. No other center has the team, resources, expertise, or tools available to perform the studies presented here and efficiently move the most promising findings directly into a clinical trial.

PROJECT SUMMARY/ABSTRACT?Developmental Biology and Solid Tumor Program Despite recent advances in understanding the etiology of pediatric solid tumors, the overall survival rate has not significantly improved in over 2 decades. For children with recurrent solid tumors, the survival rate is below 30%, and long-term survivors have an increase in the burden of disease associated with the administration of curative therapies. Therefore, the goal of the Developmental Biology and Solid Tumor Program (DBSTP) is to improve the survival and quality of life of children with solid tumors by integrating basic, translational, and clinical research. Michael Dyer, Ph.D. (laboratory lead) and Alberto Pappo, M.D. (clinical lead) are responsible for the overall leadership, academic themes, and direction of the DBSTP. They meet weekly to coordinate the activities of the program, identify opportunities for clinical translation and plan clinical trials. In addition, they mentor faculty within the Division of Solid Tumors, identifying and removing barriers to successful collaboration among SJCCC members and investigators around the world who share their goal to improve survival and quality of life for children with solid tumors. DBSTP holds a weekly seminar series to provide a forum for every Program member to present their research as well as monthly working group meetings focused on thematic research areas. The DBSTP has 21 Full Members and 4 Associate (junior mentored) Members representing the Departments of Oncology, Chemical Biology and Therapeutics, Developmental Neurobiology, Radiation Oncology, Diagnostic Imaging, Global Pediatric Medicine, Cancer Center Administration, Pathology, Pharmaceutical Sciences, Surgery, Computational Biology and Immunology. Among the full members, 11 are MDs, 7 are PhDs and 3 are MD/PhDs. Thirteen new members have been recruited to the DBSTP over the past 5 years to strengthen the basic, translational, and clinical research programs aligned with the 4 thematic Working Groups (Immunotherapy, Precision Medicine, Rare Tumors and Recurrent Disease). Peer-reviewed cancer focused funding has increased by 60% from $2.5M to $4M and the DBSTP also has $1.3M in annual funding from foundations. Research from the DBSTP has resulted in 475 publications, of which 22.9% are intra-programmatic, 40.4% are inter- programmatic, and 68.2% are inter-institutional (with other NCI-designated Cancer Centers). Overall, interactions both within the Program and with other Programs increased substantially since 2013. Program members serve as principal investigators on 18 clinical trials and 26 COG or industry-sponsored trials. They contributed a total of 801 interventional enrollments (51 national, 730 institutional, and 20 industry).

PROJECT SUMMARY During retinal development, more than 8,000 genes change in their expression as multipotent retinal progenitor cells produce each of the 7 classes of cell types in an evolutionarily conserved birth order. Although it has been well established that changes in the covalent modifications to the DNA and histones and higher-order DNA looping accompany changes in gene expression, little is known about how those processes are coordinated during retinal development. Over the past 5 years, we developed a detailed map of the structure and accessibility of the human and mouse retinal genome during development. Specifically, we performed a multifaceted integrated analysis that included profiling of the covalent modifications to the DNA and histones, promoter structure, chromatin accessibility, looping interactions, and euchromatin/heterochromatin localization. All these published and unpublished data are shared freely with the biomedical research community through our integrated retinal nucleome database (iRNDb) (https://pecan.stjude.cloud/retinalnucleome). One of the most significant discoveries to come from the iRNDb was the identification of a series of core regulatory circuit super-enhancers (CRC-SEs) adjacent to genes having important roles in retinal development, including Vsx2, Crx, Six3, Otx2, Fgf15, and Ascl1. The CRC-SE upstream of the Vsx2 gene was particularly exciting because it had activity consistent with bipolar cell development. We deleted the Vsx2-CRC-SE in mice and showed that bipolar neurons are absent yet all other cell types develop normally. Importantly, retinal progenitor cell proliferation was normal, indicating that we had separated the bipolar cell regulatory elements from that of retinal progenitor cells. In this proposal, we will elucidate the structure and organization of the Vsx2 CRC-SE, identify other transcription factors that may cooperate with Vsx2 to regulate bipolar cell type?specific expression and test the consequences of loss of bipolar cells on other cell types in the retina. The results of these studies will be important for filling a fundamental gap in our knowledge about the role of CRC-SEs in retinal development and will set the stage for characterization of CRC-SEs in other genes required for retinogenesis. All published and unpublished data are shared through the iRNDb to accelerate discovery on retinal development and disease.

PROJECT SUMMARY Several landmark discoveries in cancer genetics have come from studies on a rare childhood cancer of the developing retina called retinoblastoma. In parallel, advances in preclinical testing and clinical research has led to improvements in outcome for children with this devastating disease. Despite these advances, there are still fundamental questions in the retinoblastoma field that remain unanswered. Are retinal cells fully transformed once they sustain biallelic inactivation of the RB1 gene or is retinoblastoma tumorigenesis a multistage process? Why do some family members with the same germline RB1 mutation have bilateral multifocal retinoblastoma at a young age while others have no evidence of disease? Can treatment induce a process of tumor cell clonal evolution and selection that leads to tumor progression and enucleation? These questions have been impossible to answer because retinoblastomas are not biopsied and enucleation is only performed for advanced stage eyes. In order to overcome this barrier in the field, we have developed the first spontaneous human retinoblastoma tumor model using 3D retinal organoids produced from patients with germline RB1 mutations. I have assembled a multidisciplinary team with expertise in computational and stem cell biology, oncology, pathology, epidemiology and biostatistics to use this innovative new model of retinoblastoma to answer fundamental questions in 3 specific aims. We will determine if retinoblastoma progresses through a multistep process (Aim 1), if molecular, cellular or genetic factors contributes to differences in penetrance and expressivity (Aim 2) and if there is clonal selection with treatment (Aim 3). I have a proven record in retinoblastoma genetics and of moving basic science discoveries into clinical trials. This proposal will impact patients with retinoblastoma through preclinical testing of a novel maintenance therapy (Aim 3) to prevent new tumors from forming in the peripheral retina in the first few months after completion of chemotherapy. It may also help to identify a subset of retinoblastoma survivors with germline RB1 mutations that have an ultra-high risk of developing a 2nd malignancy and require more extensive cancer screening (Aim 2). No other center has the team, resources, expertise, or tools available to perform the studies presented here and efficiently move the most promising findings directly into a clinical trial.

Aggressive cancers often have activating mutations in growth-controlling oncogenes and inactivating mutations in tumor-suppressor genes. In neuroblastoma, amplification of the MYCN oncogene and inactivation of the ATRX tumor-suppressor gene correlate with high-risk disease and poor prognosis. Here we show that ATRX mutations and MYCN amplification are mutually exclusive across all ages and stages in neuroblastoma. Using human cell lines and mouse models, we found that elevated MYCN expression and ATRX mutations are incompatible. Elevated MYCN levels promote metabolic reprogramming, mitochondrial dysfunction, reactive-oxygen species generation, and DNA-replicative stress. The combination of replicative stress caused by defects in the ATRX?histone chaperone complex and that induced by MYCN-mediated metabolic reprogramming leads to synthetic lethality. Therefore, ATRX and MYCN represent an unusual example, where inactivation of a tumor-suppressor gene and activation of an oncogene are incompatible. In this proposal, we will exploit this synthetic lethality to improve outcomes for patients with high-risk and/or recurrent neuroblastoma. Specifically, we will use molecular targeted therapeutics to perturb ATRX-dependent processes in MYCN amplified neuroblastomas and to disrupt MYCN-dependent processes in ATRX deficient tumors. In this translational research proposal, we will target both the tumor cells and the tumor microenvironment. All published and unpublished data are freely shared through the Childhood Solid Tumor Network to accelerate discovery on neuroblastoma. The results of these studies will be used to design the next neuroblastoma clinical trials at St. Jude.

PROJECT SUMMARY Despite the advances made in our understanding of the etiology of pediatric soft tissue sarcomas (STS), the overall survival of those diseases has not significantly improved in over 2 decades. For children with recurrent disease, survival is below 30%, and long-term survivors have an increased burden of disease associated with the curative therapies they received. Therefore, the goal of our research team is to improve the survival and quality of life of children with STS by integrating basic, translational, and clinical research. For the past 10 years, we have consented STS and other solid tumor patients to donate tissue for orthotopic implantation into immunocompromised mice to develop orthotopic patient derived xenografts (O-PDXs). Our O-PDXs have been used for ex vivo high- throughput drug screening and in vivo testing using a standardized preclinical phase I, II, III paradigm. Rhabdomyosarcoma (RMS) is the most common STS in children and genomic studies have shown that rare subsets of clonally related cells can survive treatment and contribute to disease recurrence. Subsequent integrated analyses using transcriptomic, epigenetic and proteomic data showed that RMS tumors retain lineage-specific transcriptional and epigenetic signatures of their developmental origins. More recently, single cell and single nucleus RNA-seq (sc/snRNA-seq) and in vivo lineage-tracing showed that clones of cells can transition through their normal developmental programs. Indeed, single- cell ATAC-seq demonstrated that the cell- and developmental stage?specific super-enhancer activity is correlated with those clonal changes in gene expression. Chemotherapy eliminates the most proliferative tumor cell populations, and the surviving dormant tumor cells rapidly expand and re- establish their developmental hierarchy, which leads to disease recurrence. This is a striking example of the complex cell-intrinsic and -extrinsic signaling within STS and the intricate connection between developmental and oncogenic pathways in childhood cancer. In this proposal, we will perform in vivo testing for 8-10 drugs per year using our STS O-PDX models. The most compelling pathways are developmental kinase pathways (Aim 1), cell stress pathways (Aim 2) and G2/M cell cycle checkpoints (Aim 3). Novel drug combinations will be tested as well as those that include conventional chemotherapy for standard of care. Particular emphasis will be placed on eliminating all the clones in the tumor to improve survival by reducing disease recurrence.