David M. Livingston, M.D. - US grants

Certain small and intermediate size DNA Tumor Viruses, such as the papovaviruses and the human adenoviruses, encode oncoproteins which deliver powerful signals that can seriously change the biology of mammalian cells. Among the outcomes is a state of neoplastic transformation, which can lead to tumorigenicity in a sutitable animal host. SV40 large T Ag and adenovirus E1A and two such oncoproteins, and their transforming functions are, in part, dependent upon their ability to interact with and perturb the functions of p300 and CBP, two large, closely related nuclear signal integrating and transformation suppressing proteins. In this application we will explore: i) the molecular relationships between loss of genetically defined primary murine fibroblasts; ii) the mechanisms which govern a newly discovered ability of p300/CBP to sustain mdm2-mediated p53 ubiquitination and turnover; and iii) the nature of the biochemical and biological functions of E1A/T/p53-associated p400, which turns out to be a complex of two polypeptides, TRRAP, a large C-terminal PI-3 kinase domain- containing protein, and a heretofore unrecognized member of the Swi/SNF family of chromatin remodeling proteins.

The SV40 t antigen is a 20 Kd nuclear + cytoplasmic protein which is active in the process of virus-induced neoplastic transformation. However, there is little information on how it functions in vivo. In the next grant period, we propose to search for evidence that it forms in vivo complexes with three specific cell proteins, each of which can bind specifically to t in vitro. We will also attempt to establish useful, new functional assays for t and to perform detailed genetic and in vitro chemical analyses of its structure/function relationships. Ideally, results of such studies will provide new insights into how t operates biochemically and suggest a mechanism by which it operates in the viral transforming process.

The experiments described in this application are directed toward understanding the biochemical mechanisms by which SV40 large T transforms cells. Recently, three classes of large T mutants have been isolated which are central to this investigation: transformation defective (td) mutants; supertransforming (st) mutants which fail to bind with high affinity to wt SV40 origin sequences, and mutants which contain a defective nuclear location signal and therefore cannot localize large T to the nucleus (cytoplasmic mutants). The biochemical and biological activities of the large T encoded by these mutants will be investigated. Since data with cytoplasmic mutants has shown that large T need not be in the nucleus to transform established cell lines, transformed cells will be fractionated and the non-nuclear forms of large T examined. In particular, identification of the function affected in td mutants will be sought. The phosphorylation patterns of wt and mutant large Ts will be analyzed by protein fingerprinting to determine whether aberrant phosphorylation may serve as a marker of the td phenotype. All mutants will be tested for associated protein and phosphatidylinositol kinase activities. Any activity detected will be characterized to determine whether the mechanism of transformation by large T is connected, directly or indirectly, to the complex network of second messengers generated by the turnover of phosphatidylinositol. In order to find out whether the function affected by the td mutation is required for the immortalization as well as for transformation of cells in culture, or whether immortalizing and transforming functions are separable on the large T molecule, the viral DNA will be inserted into a retroviral vector containing an independently selectible marker and transfected into primary cells. The capacity of the cytoplasmic and st mutants to immortalize will also be tested in this way. It is hoped that the results of these investigations will provide some insight into the mechanism of transformation by SV40 large T.

The long term goal of this program is to understand the biochemical mechanisms which underlie neoplastic transformation by papovaviruses. In particular, the application describes experiments (by Drs. T. Roberts, A. Smith, D. Livingston, and B. Schaffhausen) designed to decipher the mode of action of three major species of papovaviral transforming proteins - SV40 large T, polyoma middle t, and polyoma large T. In particular, emphasis will be placed on how these proteins function as perturbants of the action of known cellular oncogene and anti-oncogene products and how the resulting perturbations are converted into growth altering signals. In addition, there will be a systematic genetic analysis (by Dr. R. Garcea in collaboration with Dr. T. Benjamin) of the receptor binding function of the major viral capsid protein of polyoma, VP1. Specific receptor binding, is essential to the progression of natural infection, to the development of virus-induced tumors, and, most likely, to the appearance of a mitogenic response to viral infection. Nevertheless, little is known of how the virus interacts with its receptor or how this structure functions, once activated.

The regulatory elements of the E. coli lactose operon have recently been found to function appropriately in mammalian cells. One result is an ability to tightly regulate the transcription of certain eukaryotic genes by varying the concentration of the specific allolactose analog, isoprophy beta-D thiogalactoside in the culture medium. The experiments described here ask multiple generic questions. First, can one adapt the lac repressor gene to the tight and reversible repression of expression of the SV40 neoplastic transforming region? Second, if so, is it hen feasible to reversibly modulate the transforming action of the two viral transforming products, large and small T/t antigen? Success is this endeavor could, in turn, lead to specific analyses of the quantitative aspects of the in vivo function of each and, thereby, to insights into their specific biochemical contributions to the viral transforming process.

We are applying for partial financial support for the 1993 FASEB Summer Meeting on Cellular and Molecular Genetics, to be held from July 11-15, 1993 at Copper Mountain, CO. The theme of the meeting is recent developments in one's understanding of the detailed molecular mechanisms which control cell growth and development. Particular emphasis will be placed on reaching an understanding of the molecular mechanisms which link the control of cell growth and the regulation of differentiation and development. Research on the specific experimental problems to be addressed will be presented by leaders in their fields, and we expect the content of the sessions to be of wide general interest to working scientists. There will be nine sessions, beginning with an analysis of the latest developments in cell cycle control, followed by a detailed look at the role of key protooncogenes in delivering growth promoting and retarding signals. Next, we will consider another topic central to understanding the basic biochemical events which lead to differentiation and development - how differentiation- affecting hormones and their receptors function. With the stage now set to look at the events controlling specific events in development and differentiation, there will be a series of sessions focusing on the following topics: neural cell communication with the environment and the control of neural cell fate; the process of induction during embryonic development; how neural cells find their specific targets; early decisions in embryonic development and differentiation; and the control of pattern formation. Woven into the program will also be a special session on the basic biology of the Zebra Fish, an especially attractive and increasingly popular object of vertebrate development research. Emphasis in this session will be placed on familiarizing the audience with the biology of the organism, the technology required to work with it, and on a discussion of the most powerful advances in understanding how early embryonic steps are controlled in this vertebrate. Sessions will be held in the morning and evening, Monday through Thursday, with a final session on Friday morning. Session chairpersons are recognized leaders in their fields and will be expected to provide an overview to their topic at the outset of each segment of the meeting. After each talk, there will be adequate time for vigorous discussion, as has been the tradition at these meetings. There will also be an opportunity to add short talks of important new results by meeting attendees, as well as late afternoon poster sessions. In this regard, meeting attendees will be encouraged to submit abstracts with their applications. The number of participants will be limited to 150. Copper Mountain has excellent facilities for lectures, small group interactions, and one to one discussions.

SV40 and the human adenoviruses and seemingly unrelated DNA tumor viruses. Each can transform a variety of cultured cell lines and strains, and each is tumorigenic in rodents. SV40 large T antigen (T) and adenovirus E1A protein carry out a significant part of the transforming work of the relevant virus. Both proteins work in transformation, in part, by forming a complex with the retinoblastoma gene product (RB) and related proteins, e.g. p107. Abundant evidence points to the ability of E1A and T to displace the transcription factor, E2F, from its binding domains on the surfaces of RB and p107. Further evidence suggests that T/E1A displacement of E2F, itself a protein which activates certain genes whose products promote exit from G0/G1, may unleash unregulated action of this protein which is suspected of contributing to the growth stimulatory effects of T and E1A. This application focuses on learning more of how E2F functions in the cell cycle, how its interaction with RB is translated into growth controlling events, Whether it has oncogene potential, whether there are multiple E2F species and, if so, why.

The neoplastic transforming function of the SV40 large T antigen (T) and the adenoviral E1A product(s) depends, in part, upon binding by T/E1A to a series of cellular proteins, two of which, pRB and p53 [binds to T only], are known tumor suppressor gene products. The results of genetic analyses strongly suggest that, in the case of pRB-T binding, T dominates pRB and perturbs one or more aspects of its growth suppression function. This application will focus on another cellular protein which also binds to T and E1A-p107. Genetic analyses indicate that binding of this protein is also linked to the performance of T/E1A transforming function. Moreover, it and pRB, in part, interact physically with the same T and E1A sequences. A third protein, p300, is a known E1A binding element, and preliminary data suggest that it may bind to T as well. Again, genetic analyses strongly suggest a link between binding to these proteins and maintenance of their transforming function. Circumstantial evidence points to the possibility that p107 and p300 are also tumor suppressor gene products. The proposed experimental plan is aimed primarily at understanding how p107 and p300 function normally and how binding of two dominant DNA tumor viral oncogene products affects them.

The successful treatment of cancer is dependent upon an accurate diagnosis of the tumor. It has become clear that while many tumors appear indistinguishable at the morphological level, they are in fact molecularly distinct, and such molecular distinctions can be predictive of clinical outcome. The present research proposal lays out a strategy for developing a molecular classification system for two of the most common human tumors: adenocarcinoma of the lung and prostate. The classification system will be based upon gene expression profiles obtained using DNA microarray technologies. There are three phases to the proposed project: 1) gene expression data collection for 42,000 genes and ESTs using oligonucleotide arrays for a series of lung and prostate adenocarcinoma patients with known clinical outcome, 2) classification model building using both supervised and unsupervised learning techniques, and 3) testing of the validity of these models on an independent set of lung and prostate adenocarcinoma samples. It is hoped that the development of a molecular classification system for these common tumors will help to optimize the use of existing anti-cancer therapies, and may also lay the groundwork for the development of new therapeutic strategies targeted to patients with particular subsets of these diseases.

This Program proposes to continue its interactive efforts aimed at elucidating the central molecular mechanisms which underlie the transformation of cultured cells and the development of a broad array of malignant tumors by the papova/papilloma families of DNA tumor viruses. During the past grant period, progress was made in understanding certain key events in the biochemical function of SV40 and PY large T Ag, PY mT and st Ags, as well as BPV and HPV E6. Among the most surprising discoveries made in this program was that the N-terminal SV40 and PY T Ag common region(s) are DNA J domains which support both the replicative and elements of the transforming properties of these viruses. Moreover, fresh insights into the degree of complexity of PY tumorigenesis emerged from studies of the behavior of PY mT mutants defective in the binding of selected signaling proteins. In the next grant period, we propose to continue to work across a wide spectrum of complementary approaches, Our goal is to take advantage of extraordinary, new opportunities for drawing a more detailed and comprehensive picture of the major biochemical links connecting T, mT, St, and E6 biochemical function to key cellular perturbations that result in the evolution of a fully transformed and tumorigenic phenotype. The overriding goal of this work is to define the relationships between papovaviral transforming mechanisms and those processes which -give rise to spontaneous human neoplasms.

brca gene; breast neoplasms; neoplasm /cancer genetics; gene expression; molecular oncology; protein protein interaction; gene mutation; oncoproteins; binding proteins; protein structure function; ovary neoplasms; genetically modified animals; laboratory mouse;

The main goal of this work is to understand how SV40 large T Ag(T), adenovirus EIA, and HPV E7 function as neoplastic transforming elements. They operate, in part, through strong, specific interactions with the nuclear pocket proteins (pRB, p107, p130). The latter function as biochemical regulators of the E2F family of transcription factors. pRB participates as a major GIIS progression control element, in part, through specific interactions with selected members of the E2F family. In the next granting period, we propose to investigate the significance of mounting evidence that pRB and the other pocket protein(s) also have DNA damage-associated cell cycle checkpoint control function(s) which contribute to their transformation repression function. The contribution of pocket protein/ E2F interactions to these processes will also be explored, along with the mechanisms by which RB and E2F-l serve their tumor suppression function. Finally, we will continue our efforts to understand how E2F-4,5, and a newly discovered species, E2F-6, a transcriptional repression, operate with special emphasis on understanding whether they normally function as proliferation controlling proteins.

In this application are proposed experiments aimed at a better understanding of the breast cancer suppression function of the tumor suppressor gene product, BRCA1. In particular, we have found that the protein plays a significant role in the processes that allow post-mitotic chromatin to replicate in the next S phase, and the proposal describes experiments using Xenopus BRCA1 that are aimed at understanding the biochemical basis for this phenomenon. Furthermore, the BRCA1 gene is large and encodes multiple, alternatively spliced mRNAs. The product of one of them, a 1399 residue chromatin associated protein called IRIS is of special interest. Unlike the full length BRCA1 protein, p220, IRIS appears to facilitate S phase progression, and the proposal describes experiments aimed at understanding how this function is performed and also at whether IRIS plays a role in the evolution of certain BRCA1 -/- human tumors. Finally, we have discovered a novel DNA helicase, BACH1, as a directly interacting BRCAl-associated partner protein. BACH1/BRCA1 complex formation is essential for the development of proper double strand break repair in cells exposed to ionizing radiation, and it may well be the product of a human cancer gene in its own right. The proposal is aimed at a better understanding of how BACH 1 functions in vivo and at gaining insights into the possibility that BACH1 operates as a breast cancer suppressing element through its interaction with BRCA1.

We have accumulated evidence pointing to a distinctive role of the un(der)phosphorylated form of the Rb protein during S phase that is blocked by SV40 small t Ag. Specifically, Rb can suppress the rereplication of DNA segments after DNA damage, at least in part by homing to and forming complexes with multiple genomic replication initiation elements. In preparation for post-damage replication site localization, Rb must also localize in the chromatin fraction, a process that we have shown is pp2A-dependent. Thus, small t blocks Rb/chromatin localization and, thereby, its ability to home to replication initiation sites after DNA damage. One end product of small t action in this regard was endoreduplicafion of cellular DNA. In addition, we have found that the product of the Nijmegen Break Syndrome gene, Nbs1, is a heretofore unappreciated large TAg binding protein; that Nbs1, like Rb, participates in rereplication suppression-both of cellular replicons and the viral genome; and that TAg can, upon Nbs1 binding, override this activity. The latter effect may well contribute to the established ability of T in inducing genomic rereplication as well as fully autonomous replication of circular DNA structures that contain an SV40 replication origin. In this application, we propose to search for: 1) biochemical evidence that sheds light on how Rb and Nbs1 suppress genomic rereplication; 2) for insights into where during the cell cycle the pp2A Rb/chromatin entry process is applicable; 3) for information on how pp2A operates biochemically in this process, and 4) for evidence indicating whether or not Rb and/or other pocket proteins serve as key small t targets during the performance of its neoplastic co-transforming function in human cells.

DESCRIPTION (provided by applicant): BRCA1 is a tumor suppressor gene, germ line mutation of which can result in inherited breast and ovarian cancer. How BRCA1 delivers its disease-preventing effects and why they are clinically apparent only in selected estrogen target cells of females is a mystery. In this regard, the protein is expressed in a wide spectrum of proliferarting male and female cells. Recently, we have observed that a fraction of the BRCA1 in female somatic cells is concentrated on the inactive X chromosomes (Xi). Its presence there is necessary for the colocalization of the noncoding RNA, Xist and the specialized histone, Macrohistone H2A 1.2. Both of these elements play a key role in the X inactivation process during embryogenesis. Furthermore, BRCA1 interacts with Xist, although whether or not this is a direct interaction remains unclear. Although neither X chromosome in BRCA1- deficient tumor cells carried associated Xist or MH2A, the abundance of these elements in such cells was the same as in their BRCA1- producing derivatives or counterparts. Hence, where studied, BRCA1 appears to promote the localization of Xist and MH2A on Xi rather than directing their synthesis or controlling their stability. Finally, multiple human and murine BRCA1 -/- tumors lack a readily detectable Xi structure despite the presence of 2 X chromosomes. This proposal is aimed at understanding the biochemical bases for these phenomena. Since both BRCA1 tumor suppression and the BRCAI->Xist/Xi effects are female-specific processes, the proposal is also aimed at learning whether these phenomena are linked.

Biology and treatment of BRCA1 mutant and sporadic Basal-like cancers The cancers that arise in women harboring BRCA1 mutations have a characteristic phenotype, being mostly poorly differentiated, high grade invasive ductal carcinomas that do not express HR or PR and do not have amplifiation of erbB2. Gene expression array analysis has shown that these BRCA1-mutant cancers display the same gene expression profile as the basal-like cancers (BLC), a distinct subset of high grade KR(-), PR (-), Her2(-) tumors that account for -15% of human breast cancers. These observations suggest that the BRCA 1-mutant and sporadic BLC share a similar functional molecular phenotype. As BRCA1 -mutant tumor cells show clear defects in DNA repair and epigcnetic stability of the inactive X chromosome, and are characterized by heightened sensitivity to DNA cross linking agents such as cis-platinum, the BLC] may also share these features. The goal of our project is to use our knowledge of BRCA 1 biology to gain insight into the biology of BLC and identify specific physiologic vulnerabilities that will guide the development of more effective treatment strategies for this group of aggressive breast cancers. The first two aims arc to assay specific aspects of genomic stability, DNA repair and heterochromatin maintenance in BLC and BRCA 1 - mutant, cancer samples to characterize their functional similarities and differences. The third aim is to evaluate the response of a cohort of women with ER(-), PR(-), Her2(-) breast cancers, of which the majority should be BLC, to cis-platinum in a phase II pre-operative treatment trial, correlate the results of any informative biological assays validated in the first two aims with treatment response.

In this project we propose to continue our ongoing work relating to how defective function of the BRCA1 gene results in breast cancer development. Specifically, we propose to ask the following questions. 1) Do specific molecular defects in the response to DNA double DNA strand breaks (DSB) exist in the tumor cells of sporadic basal like breast cancers(BLC)? BLC is a relatively common, BRCA 1 wt, phenocopy of BRCA1 mutant disease. If such defects exist, do any affect the operation of DSB response pathways to which BRCA1 normally makes a major contribution? 2) Is the BRCA1-IRIS protein, an alternatively spliced BRCA1 gene product, a protooncoprotein? If so, does its oncoprotein function contribute to the development of certain subtypes of sporadic breast cancer and, if so, how? 3) In a normal setting, BRCA 1 p220 is present and readily detectable in the nucleus both during the G1 and S/G2 cell cycle phases. It is less abundant in G1 than in S in some cell lines. Moreover, although it localizes after ionizing radiation (IR) in special nuclear foci (so called IRIF) in S/G2, it fails to do so in G1, even though these foci contain a protein (Rap80) that normally participates in tethering BRCA 1 to these structures. We seek to understand the molecular basis for this difference between G1 and S/G2 and, if deciphered, to probe its physiological significance, especially with respect to how it might relate to the normal regulation of BRCA1 DNA repair function.

Acute infection by the small DNA tumor virus, SV40, can drive quiescent mammalian cells into the cell cycle. At least part of this work is performed by SV40 LT, one of the two viral oncoproteins. The rest is performed by the other oncoprotein, SV40 ST. The role of ST in this process depends upon its ability to interfere with the function of the protein phosphatase, PP2A, the product of a known human.tumor suppressing gene. Unexpectedly, at least part of the mechanism underlying SV40 ST perturbation of quiescence control is directed at events that occur in G2 (or G2/M). The process targeted by ST during G2 is PP2A- dependent and must be intact for cells to quiesce following mitogen deprivation initiated during the next G1 phase. Also involved is the p130/E2F4/DREAM complex, the integrity of which is PP2A dependent, This complex has been shown to operate in the maintenance of quiescence. Whether G2-centered PP2A function and/or PP2A function operating at another time(s) during the cell cycle communicates with this complex to license quiescence is unclear. The thrust of the proposed research is aimed at understanding 1) how, in molecular terms, the G2 quiescence control process operates, 2) whether its perturbation contributes to SV40 STdependent neoplastic transformation and tumorigenesis and 3), if this is the case, how its perturbation elicits these effects.

? DESCRIPTION (provided by applicant):BRCA1 is a multi-functional breast and ovarian cancer suppression gene that is suspected of operating in this manner by suppressing the development of genome disorder. But how it delivers signals that elicit a clinical cancer suppressing effect is unknown. During the current granting period, we reported that the BRCA1 protein (aka B1), itself, executes at least one of its multiple biochemical functions under tight control afforded bya B1-bound, multi-subunit protein complex, called RAP80. This larger complex is called the RAP80/B1 complex, and it allows B1 to perform at normal amplitude error-free double strand DNA break (DSB) repair by homologous recombination(HR). Thus, RAP80/B1 physiologically regulates B1 HR function, which is a known contributor to B1 cancer suppression. This process is termed `HR tuning'. We have also reported that there is another protein present in RAP80/B1 complexes, i.e. the enzyme PARP1. In these complexes, PARP1 selectively poly-ADP ribosylates (parsylates) the DNA binding domain of B1 to form a parsylated derivative, aka B1pars. RAP80/B1 complex formation and proper B1 parsylation contribute to B1 HR and B1 genome stability control. All of these events take place in and require the operation of specialized, DSB- containing nuclear focal structures (aka IRIF) that form after DNA damage. In this proposal, we will determine: a) whether PARP1-catalyzed B1 parsylation regulates any of the other, known functions of B1, including cancer suppression; b) how the specific PARP1-driven parsylation of B1 is directed to its DNA binding domain; c) whether defects in B1 parsylation and RAP80/B1pars complex formation elicit clinically important contributions to sporadic breast and ovarian cancer development; and d) whether they dictate a specific strategy for achieving effective breast and ovarian cancer therapy.

The Principal Investigator (D Livingston) oversees the entire operation of the Program. The chief administrative assistant (Ann Desai) reports to the PI, who provides direction as she carries out her financial and other logistical planning and oversight responsibilities while coordinating among the three institutions involved in the Program (DFCI, HMS, Tufts Med School). The director of our Gene Function Manipulation Core facility (W. Hahn) will also report to the PI on the ongoing operations of that unit.

This proposal is based on results from our laboratory showing that primary skin fibroblasts and primary normal- appearing mammary epithelial cells {MECs) from healthy BRCA1 {B1)+/- women are defective in the repair of stalled replication forks (stalled replication fork repair=SFR). Thus, they are haploinsufficient for this function. The same cells do not appear to be defective in homologous recombination-based double strand break repair, a canonical BRCA1 biochemical function. These cells will be tested for multiple, other B1 functions with the goal of determining whether they perform the other, established B1 functions normally. Preliminary indications are that they are intact for at least 1 other B1 function- centrosome proliferation control. Defective repair of stalled replication forks is a common human cancer- promoting force. The experiments to be performed here are aimed at learning whether a B1-SFR defect and any other haploinsufficient B1 functions present in ostensibly normal B1+/-mammary epithelium represent early drivers of B1 breast cancer development. Positive results would open new avenues to the development of disease mechanism-based B1 breast cancer prevention and therapeutic strategies.

Project Summary/Abstract Inherited BRCA1 mutations very often result in high frequency breast and ovarian cancer in affected families. This proposal is aimed at illuminating key steps that occur after menarche in tumor-free, BRCA1-mutated mammary epithelium and contribute in a major way to future BRCA1 breast cancer (BrCa) development. Our published data and work in progress suggest the existence of a series of interrelated biochemical steps that, when interrupted, result in the conversion of ostensibly normal, BRCA1- mutated breast epithelial cells into an abnormal, dedifferentiated, unusually primitive, and likely pre-malignant state. This change involves the loss of function of any member of a group of 6 particular genes which translates into a breakdown in DNA damage control which, in turn, promotes a breakdown in mammary epithelial differentiation. In breast cells these genes normally communicate with one another, functionally, as if they were components of a biologically important pathway dedicated, in part, to BrCa suppression. Indeed, their protein products form a distinct molecular complex which also contains the BRCA1 tumor suppressor protein, p220, and this complex operates in ways that suggest a major role for it in such a pathway. Among the 6 other complex members (BRG1, FancD2, CtIP,NUMB, HES-1, and BARD1), some (e.g. FancD2, BRG1) are somatically mutated in sporadic BrCa. Another (Numb) is a known BrCa suppressor in animals and a supporter of normal stem cell proliferation, and at least 5 of the 6 must be present and bind to p220 for the entire complex to form and mammary epithelial differentiation to proceed normally. Our work in progress also suggests that the integrity of this complex is required to prevent BRCA1 BrCa from developing. The proposed research will test the validity of this hypothesis along with the physiological importance of such a pathway, using a combination of cutting edge methods. Included will be an effort to a) test for lineage relationships between failed mammary epithelial cell (MEC) DNA damage repair and MEC dedifferentiation and between MEC dedifferentiation and BRCA1 tumor development and b) to decipher the mechanisms governing the interrelationships of these events. A major component of these efforts will be the application of single cell RNAseq to assess the above-noted lineage relationships and mammary epithelium- focused genomic CRISPR and protein overexpression screens to detect individual proteins that participate in these important steps in BRCA1 BrCa development. We will use a newly developed mouse model in these experiments.!