Thomas M. Roberts - US grants

Programmed migrations of cells during development, front-line defense against invading microbes and the development and metastasis of malignancies all depend on the capacity of certain types of cells to crawl over solid substrates. This type of cellular locomotion is complex involving interaction between the cytoplasm and the plasma membrane, the membrane and the substrate, and coordinated movement of the membrane itself. The purpose of this proposal is to study each of these related processes in a single, simplified type of crawling cell, the spermatozoon of the nematode, Caenorhabditis elegans. Monoclonal antibodies directed against membrane antigens will be used to study the membrane dynamics on crawling sperm focusing on the mechanism that propels the directed movement of membrane proteins over the cell surface, the fate of membrane components after removal from the surface, and the biochemical characteristics of cytoplasmic pools of membrane proteins that might direct their localized insertion into the plasma membrane. C. elegans sperm contain 2-nm filaments in their pseudopods but lack microfilaments and microtubules. These filaments resemble a new class of motility structures discovered in other cells. Their cellular organization will be examined by high voltage electron microscopy. A combination of immunocytochemical and biochemical techniques will be used to identify the protein that forms these filaments and the protein that links the filaments to the plasma membrane. The role of the filaments is sperm locomotion will be defined using antibodies to inhibit their function. The long term objective of this work is to understand crawling movement in molecular detail. C. elegans sperm are well-suited for this goal because they can be manipulated genetically. Eventually, it should be possible to remove individual proteins from the cell by mutation and, thereby, deduce their exact roles in locomotion.

The objective of this project is to characterize the genes for two structurally-related, functionally important cell surface antigens, Mac-1 and LFA-1. Mac-1 is associated with or identical to the complement receptor type three for C3bi on macrophages and granulocytes. LFA-1 is important in killer T lymphocyte interactions with target cells and in T helper cell responses. Mac-1 and LFA-1 comprise a family of leukocyte differentiation antigens of Alpha1Beta1 structure, which contain alternative forms of an Alpha subunit of MW = 170,000 or 180,000 noncovalently associated with an identical or highly homologous Beta subunit of MW = 95,000. Presently available conventional and monoclonal antibodies reactive with the isolated Alpha and Beta subunits, as well as further antibodies to be prepared in this study, will be used to assay mRNA by in vitro translation and immunoprecipitation, and to purify specific polysomal mRNA on immunoadsorbent columns. The cDNA will be prepared using immunopurified or size-fractionated mRNA as templates. The cDNA clones will be screened by preferential binding of appropriate probes and positive selection in translation assays. The cDNA's will be used to obtain genomic clones. The number, structure, and chromosomal location of Alpha and Beta genes will be determined. The expression of Alpha chain genes in differing lineages of leukocyte differentiation, and the induction of Mac-1 Alpha and Beta genes during maturation of the M1 myeloid cell line, will be examined at the level of the Alpha and Beta gene promoters and mRNA transcription. From nucleic acid sequences, protein sequences will be deduced of importance for understanding the structural and functional relationship between Mac-1 and LFA-1. Basic knowledge will be obtained about leukocyte differentiation and about the molecular basis of T lymphocyte and macrophage cell-mediated immunity of importance for cancer and immune diseases.

This project examines two aspects of the polyoma tumor antigens: (1) how structure is related to function within each protein and, (2) how the T antigens interact with cellular components to exert their regulatory effect on the cell. Recombinant DNA techniques are used to overproduce the T antigens and the individual proteins are purified from the resulting strains. Purified proteins can then tested for biological or biochemical function. Subsequently they can be used as reagents for raising banks of monoclonal antibodies for structural studies and as affinity reagents for studying interactios with host cell proteins. To complement these biochemical approaches, genetic studies will be done using retrovirus vectors which transduce the individual T antigens. A library of missense mutants of middle T antigens which have conditionally or absolutely lost transforming function will be made using the virus vectors. Interesting mutations will be characterized biochemically and sequenced at the DNA level to yield structure function data for the proteins. Together, the biochemical and genetic approaches will eventually allow us to overproduce mutant T-antigens so that they can be purified for further study. Such a situation should yield valuable information about the manner in which polyoma T antigens alter a cell to a transformed state.

pp60c-src is the cellular homolog of pp60v-src, the transforming protein of Rous Sarcoma Virus. The protein is located on the cytoplasmic face of the plasma membrane and is believed to function in signal transduction in response to mitogens and other extracellular signals including PDGF, NGF and, perhaps, CSF-1. The experiments proposed here make use of recently developed recombinant DNA techniques and address two different, but related, questions concerning the molecular biology of pp60c-src. The first set of experiments examines structure-function relationships in the protein. Eukaryotic viral vectors will be used to obtain milligram quantities of the protein. The purified pp60c-src should facilitate a number of experiments - we propose using it first to prepare monoclonal antibodies and to search for kinases and phosphatases believed to modify the activity of the molecule as a tyrosine kinase. To complement these biochemical studies we will construct a large library of mutant c-src genes using a recombinant murine retrovirus which we have recently constructed and the so-called GC clamp technique from the Manjatis laboratory. The second set of experiments are designed to follow up our recent finding that the level of pp60c-src is modulated some 20-fold during the differentiation of monocytes. We have designed experiments to discover if specific receptors are interacting with pp60c-src in the fully differentiated cells and to see if pp60c-src might also play a role in the differentiation process itself.

The objective of the project is to obtain full length genomic DNA clones and cDNA clones for both the alpha and beta-chains of the functionally important cell surface antigen, MAC-1, and to characterize these clones in detail. The complete DNA sequence of the cDNA clones will be determined and from this the protein sequence of MAC-1 deduced. Extensive studies on the relationship between structure and function of the protein will be undertaken using a variety of mutagenesis protocols. Mutants in the protein will be assayed for function using a very efficient protocol based upon an easily rescued retrovirus vector. The promoters for both MAC-1 genes will be mapped on the genomic clones and further characterized by genetic analysis. Changes in the level of methylation of the gene and of the DNAse reactivity of the gene's chromatin will be studied as a function of macrophage differentiation. Proteins which bind to the genes control region will be characterized by Crothers' gel analysis.

The capacity of cells to migrate over surfaces plays a key role in embryonic development, wound healing, defense against microbial infection, homing to target organs, and metastasis of tumors. The spectrum of cells that display this type of locomotion encompasses a broad range of phenotypes but each is thought to be propelled by cytoplasmic contraction of actin-myosin complexes. The amoeboid spermatozoa of nematodes are an exception. These cells exhibit the same motile behavior as other amoeboid cells but lack both actin filaments and myosin. Their pseudopods, instead, contain arrays of filaments, composed of a 15,500 protein, that exhibit patterns of organization and movement consistent with a central role in sperm locomotion. The purpose of this proposal is to analyze this new motility system in detail using sperm from two species of nematodes, Caenorhabditis elegans and Ascaris lumbricoides. Filaments will be isolated from sperm for analysis of their structure and organization by high resolution electron microscopy. The capacity of these filaments to self-assemble in vitro will be exploited to define the chemical requirements and kinetics of their polymerization. In order to assess the role of the filaments in locomotion, computer-enhanced video microscopy will be used for a stepwise dissection of motility, starting with intact sperm and proceeding to isolated filament complexes. In addition, accessory proteins involved in organizing the filaments into three- dimensional arrays, linking these arrays to the plasma membrane, and generating force within the filament system will be identified and isolated and their specific interactions with filaments defined by in vitro assays. The long-term objective of this work is to understand the machinery of sperm locomotion in molecular detail and determine how this new system of motility relates to other types of cellular motors.

DESCRIPTION (provided by applicant): Polyoma middle T antigen has been a workhorse for the study of signaling pathways critical to cancer. MT is the oncogene of polyoma virus, but is itself devoid of biochemical activity. It transforms by stealing host cell components. More importantly, the virus has been quite judicious in its choices of host pathways to attack, and consequently the study of MT binding proteins has proved fruitful. Notably PI3 Kinase, perhaps the leading candidate for tumor therapy at this time, was actually introduced to the world via the work of the Cantley, Schaffhausen and Roberts labs on MT. Notably what we call PI3 Kinase is actually composed of a small family of closely related lipid kinases. My colleagues and I have continued our studies of MT and PI3K, and have recently used conditional knockouts of the key catalytic subunits of PI3K, p110? and p110?, to discover a previously unknown subtlety in the roles of these enzymes in signaling and cancer. Both the p110? and p110? isoforms appear to play distinct roles in oncogenic transformation, and, interestingly, isoform functionality varies according to tumor type. The roles of the 2 isoforms in insulin signaling are also quite distinct suggesting that inhibiting individual isoforms could have fewer side-effects than the pan inhibitors now entering the clinic. Thus we are excited that the differences in the roles of the isoforms may be exploited to make safer second-generation drugs for PI3Ks. In the next grant period we propose to exploit the known strengths of MT to study the roles of PI3K isoforms in much greater detail. We want to use the well known ability of MT to transform cells of multiple tissue types to study the isoform requirements for MT mediated transformation. This will allow us to decipher whether isoform dependence on tumor initiation is determined by the tissue type or the nature of the oncogenic lesion(s). We plan to examine the isoform dependence of MT tumors in the breast and prostate. We will also follow up on preliminary data, which suggests that isoform specific PI3K inhibitors might be used as chemopreventatives to ameliorate familial cancers arising from PI3K pathway activation. Finally we will examine mechanisms by which cells become resistant to PI3K inhibition. PUBLIC HEALTH RELEVANCE: The first generation of PI3 Kinase inhibitors are just now entering clinical trials. The experiments proposed in this grant use murine tumor models driven by the viral oncogene MT antigen to (1) speed the use of second generation isoform specific PI3K inhibitors by determining how best to target individual PI3 Kinase isoforms in various tumor types, (2) investigate the potential of PI3K inhibitors as chemopreventatives and (3) determine how tumors may become resistant to PI3K inhibitors. Each of these aims has considerable potential to facilitate therapy for a number of important human tumor types.

This proposal is intended to advance knowledge concerning the role(s) of the src family of tyrosine kinases in transformation and differentiation. Specific aims for the next project period have been chosen from the many possible experiments based on two criteria. They are designed (1) to provide information on fundamental gaps in current understanding of the src family and (2) to make use of the considerable expertise and supply of reagents which has been built in the previous grant period. Understanding of the mechanism of action of a kinase necessarily requires a knowledge of its physiological substrates. Thus, the excellent anti-phosphotyrosine monoclonal antibody which was developed will be utilized to find previously uncharacterized src substrates. Of the hundreds of possible substrates attention will be focused on substrates shared with growth factor receptors (an area which we pioneered in our work on the PI3 kinase) and those unique to the src mediated differentiation of M1 cells. Intense effort will also be trained on structure function studies of the src family. Work will be concentrated on the region of the molecule which has been heretofore neglected, the kinase domain itself, once again using special reagents which have been developed in the previous grant period. Key questions to be answered are: (1) how important conserved residues in the kinase domain really are, (2) how the kinase picks out its substrates. The latter point seems particularly important now that the role of the N-terminus in substrate selection is being questioned. An attempt will also be made to create a very useful class of mutants, those which are resistant to kinase inhibitors. This type of mutant is essential to understanding inhibitor based experiments. A foray will also be made into the more contested turf of genetic studies of the SH2 region. Finally the large variety of baculovirus overexpression vectors available in the lab will be used to explore in detail the mechanism by which v-src activates the Raf-1 kinase.

Amoeboid motility plays a vital role in such diverse processes as embryonic development, wound healing, cellular immunity, homing to target organs, and tumor metastasis. In most amoeboid cells, locomotion is mediated by a complex, actin-based apparatus. Nematode sperm are an exception. These cells move like other crawling cells by extending a motile pseudopod, but contain a well-ordered, dynamic system of filaments composed of major sperm protein (MSP). Thus, these sperm offer a valuable complement to actin-based systems for investigation of the mechanism of amoeboid movement. This novel motility system will be investigated using sperm of the parasitic nematode, Ascaris suum. In these cells, MSP filaments are organized into well-ordered arrays called fiber complexes, that span the length of the pseudopod and flow rearward at the same rate as the cell crawls forward. Thus, sperm locomotion is closely related to both continuous assembly of MSP filaments at the extreme leading edge of the pseudopod and the arrangement of those filaments into motile fiber complexes. This study will focus on the molecular interactions underlying these key events in motility. The capacity to form crystals and filaments of MSP in vitro will be exploited to examine the molecular structure of the protein and its subunit arrangement in filaments to high resolution using a combination of X-ray diffraction, electron microscopy and computer image processing. Several lines of evidence suggest that the precise control of filament formation in vivo is due to pH-sensitive interaction of MSP with plasma membrane. To test this hypothesis sperm membranes will be isolated and fractionated to identify the components that bind to MSP and analyze quantitatively the effects of membranes and pH on the polymerization process. Proteins that crosslink MSP filaments together to establish the distinctive, three-dimensional architecture of the fiber complexes will be identified and characterized by determining their biochemical properties, their sequence, and their MSP-binding domain. Three complementary strategies, microinjection of labeled analogs of motility proteins to examine their localization and dynamics, disruption of function by incorporation of modified proteins or inhibitory antibodies into intact cells, and reconstitution of function by introduction of motility proteins into pseudopods emptied by hypotonic lysis, will be used to determine the roles of components of the motile apparatus in vivo and deduce the molecular interactions necessary for locomotion. The long-term objective of this proposal is to define the molecular mechanism of sperm locomotion so that comparison of MSP- and actin-based systems can be used to obtain a fuller understanding of the principles of amoeboid motility.

This proposal is intended to advance knowledge concerning the role(s) of the src family of tyrosine kinases in transformation and differentiation. Specific aims for the next project period have been chosen from the many possible experiments based on two criteria. They are designed (1) to provide information on fundamental gaps in current understanding of the src family and (2) to make use of the considerable expertise and supply of reagents which has been built in the previous grant period. Understanding of the mechanism of action of a kinase necessarily requires a knowledge of its physiological substrates. Thus, the excellent anti-phosphotyrosine monoclonal antibody which was developed will be utilized to find previously uncharacterized src substrates. Of the hundreds of possible substrates attention will be focused on substrates shared with growth factor receptors (an area which we pioneered in our work on the PI3 kinase) and those unique to the src mediated differentiation of M1 cells. Intense effort will also be trained on structure function studies of the src family. Work will be concentrated on the region of the molecule which has been heretofore neglected, the kinase domain itself, once again using special reagents which have been developed in the previous grant period. Key questions to be answered are: (1) how important conserved residues in the kinase domain really are, (2) how the kinase picks out its substrates. The latter point seems particularly important now that the role of the N-terminus in substrate selection is being questioned. An attempt will also be made to create a very useful class of mutants, those which are resistant to kinase inhibitors. This type of mutant is essential to understanding inhibitor based experiments. A foray will also be made into the more contested turf of genetic studies of the SH2 region. Finally the large variety of baculovirus overexpression vectors available in the lab will be used to explore in detail the mechanism by which v-src activates the Raf-1 kinase.

In the last grant period this laboratory identified two of the cellular proteins bound by small and middle T antigens of polyoma virus as the 36kd catalytic and 63kd regulatory subunits of PP2A. We now propose to follow up our initial discovery by carrying out a careful genetic study of the structural requirements of the interaction. We will then proceed to use our mutants to attempt to gain insights into molecular details of the function of PP2A in vivo in both mammalian and insect systems. Finally we will attempt to push our understanding of the interaction to the atomic level via crystallography.

DESCRIPTION: The goal of the proposed studies is to understand how the structure of the Src tyrosine kinase regulates its function as a signaling molecule. Src is a member of a family of tyrosine protein kinases which play critical roles in both normal and malignant cell growth. The proposal has two specific aims. Aim 1 will focus on signaling by members of the Src family using three lines of investigation. First, a novel Src SH3 binding protein identified by the PI during the previous grant period, termed Dif-1, will be examined for its role in adipose differentiation and intracellular signaling. Second, the role of tyrosine phosphorylation of Raf-1 will be probed. Third, a modified version of the yeast 2-hybrid system, in which a tyrosine kinase is also introduced, will be exploited in an attempt to identify new substrates for Src kinases. Aim 2 will examine the structure of the Lck tyrosine kinase in great detail, using a synthetic lck gene generated by the PI. A specific region within the N-terminus of kinase subdomain VII will be targeted for mutagenesis, in an attempt to identify Lck mutants with altered substrate specificity. Mutants will be initially screened by identifying alterations in tyrosine phosphorylation patterns in E. coli, followed by peptide library screening. A second set of experiments will examine the function of the "activation loop", a region downstream from the tyrosine autophosphorylation site of Lck. The PI will look for SH2 containing proteins which bind to this region.

In the last grant period this laboratory was able to play a major role in the discovery that the amino terminus common to all T antigens in the SV40/polyoma family of viruses is a functional dnaJ domain. This discovery, which was the work of a group of labs in the papova virus field, stands as a milestone in the study of these viruses. The viral J domain "function" plays a role in viral replication, regulation of the host cell cycle, and transformation. It is possible that the J domain may underlie other viral processes such as virus assembly as well. It is now time to bring our studies of the J domain to a more mechanistic level. In the coming grant period the molecular mechanisms by which the J domain of SV40 virus enhances DNA replication and frees the E2F complex from inhibition by the Rb family of tumor suppressor proteins will be elucidated.

DESCRIPTION (provided by applicant): Amoeboid cell motility, a property of many eukaryotic cells, plays a key role in physiological processes such as inflammation, wound healing, neuronal targeting, and metastatic invasion. The purpose of this proposal is to investigate the molecular mechanism of cell crawling using the simple, specialized sperm of the nematode, Ascaris suum, as an experimental system. These cells display the same motile behavior as conventional crawling cells but lack the actin machinery usually associated with cell migration. Instead, the motility apparatus of sperm is based on major sperm protein (MSP) filaments that assemble along the leading edge and disassemble at the base of the lamellipod. These unique filaments have no structural polarity indicating that molecular motor proteins are not required for sperm motility. The coupling of MSP cytoskeletal dynamics to locomotion suggests a "push-pull" mechanism for movement in which forces for leading edge protrusion and cell body retraction are produced at opposite ends of the lamellipod and linked reciprocally to the assembly status of the cytoskeleton. The push-pull model will be evaluated by characterizing the components of the motility apparatus and integrating this information to define how the cell machinery produces movement. Structural studies will be extended to determine the orientation of the MSP subunits in filaments and to define the interactions that promote intrinsic bundling of filaments into larger arrays. Based on this information MSP mutants will be constructed to study the contributions of filament polymerization and bundling to generating the forces for movement. Biochemical and molecular methods will be used to analyze the membrane and cytosolic proteins required to nucleate MSP polymerization at the leading edge and explore the roles of pH and phosphorylation in regulating this process. The hypothesis that the force for cell body retraction is produced by deswelling of the MSP cytoskeleton will be tested by defining conditions that induce shrinkage of MSP filament gels in vitro and examining the organization of the cytoskeleton at the base of the lamellipod. The long term goal of this project is to define the mechanism of sperm locomotion so that comparison of actin- and MSP-based systems can be used to understand the basic principles of amoeboid cell motility.

DESCRIPTION (provided by applicant): Pediatric cancers of neural ectodermal origin are the most common solid tumor in children and are now the most common cause of cancer-related death in children. The goal of this grant is to define drugable targets for two of the most common pediatric cancers of the central nervous system - malignant medulloblastoma and low-grade astrocytoma. Neither of these tumors exhibits gross chromosomal instability that characterizes malignant astrocytomas in adults. Both tumors are generally wild type for the tumor suppressor genes most commonly mutated in adult CNS cancers - p53, RB and PTEN. On the other hand, upregulated kinase activity has been associated with poor outcome or increased metastatic potential in these tumors. The hypothesis of this grant is that the malignant phenotype for medulloblastoma and also for low-grade astrocytoma reflects a gain-of-function mutation within a single protein kinase (or a small number of kinases) unique to each tumor type. Recent insights into the genetics of malignant melanoma suggest that this hypothesis is reasonable. We have assembled a tripartite team in 1) Pediatric Oncology/Neuro Pathology, 2) Molecular Biology/Bioinformatics and 3) Signal Transduction/Drug Discovery that makes the hypothesis testable and our goal achievable. Our study plan has three aims: (1) to isolate DNA for mutation analysis from at least 15 medulloblastoma and 15 low-grade astrocytomas a year for three years. (2) to identify mutations in all tyrosine kinases and all serine/threonine and lipid kinases involved in oncogenic signaling pathways. Towards this end, we will sequence roughly 4000 exons, covering the entire coding sequences of tyrosine kinases and of all oncogenic serine/threonine kinases as well as key portions of the remaining serine/threonine kinases and type 1 PI3 kinases, (3) to characterize the biochemical activity and biological activity of the mutant kinases. Mutant kinases will be subjected to a battery of analytical tests to see if the mutation increases the specific activity of the kinase in vitro or in cells. Finally, we will determine if the mutation increases the transformation potential of the kinase. The research will lead to a new generation of selective therapeutics for children with brain cancer. Since cancers of children are often "informative" in a larger context, it is likely that these medicines will find use for more frequent adult cancers.

We have detected a functional interaction between SV40 large T antigen and the mitotic regulator Bub 1. The protein kinase Bub1, a member of the family of checkpoint proteins that monitor the assembly of the mitotic spindle, has been found to be mutated in certain human cancers characterized by aneuploidy. Notably, T antigen can also cause genomic instability by inducing chromosomal aberrations and aneuploidy. T antigen co-immunoprecipitates with endogenous Bub1, as well as with Bub3, another component of the checkpoint complex at the kinetochore. While we have come to think of T antigen as an inactivator of its target proteins as is the case with p53 and pRb, T actually enhances Bub1 kinase activity. Bub1 phosphorylates p53 on ser37, previously identified as a site of phosphorylation by DNA-PK. Genetic analysis demonstrates that interaction of T antigen with Bub1 is not required for immortalization but appears to be necessary for T antigen to deregulate the spindle assembly checkpoint and transform. This interaction with Bub1 suggests a novel role for T antigen, which may provide new insight into its ability to regulate p53, cause chromosome aberrations, and transform. We propose to study the interaction in detail. It is our intent to determine: (1) What are the the mechanistic consequences of Bub1 mediated phosphorylation of p53? (2) Does T antigen direct Bub1 toward other substrates in addition to p53? And (3) Is T antigen's interaction with Bub1 required for T antigen to destabilize the chromosomes in its target cells?

DESCRIPTION (provided by applicant): The class IA phosphatidylinositol 3 kinase (PI3K) signaling axis is perhaps the most frequently activated pathway in human cancer. In response to the activation of receptor tyrosine kinases (RTKs), G-protein coupled receptors (GPCRs) or Ras, class IA PI3Ks, consisting of three catalytic isoforms termed p110?, p110? and p110?, are activated to generate the primary intracellular lipid signal, phosphatidylinositol 3,4,5-trisphosphate (PIP3), which is essential for multiple cellular processes. The tumor suppressor PTEN, a lipid phosphatase, dephosphorylates PIP3, thereby antagonizing the actions of PI3K and regulating the PI3K pathway activity. Pathway activation in tumors is most commonly achieved through activating mutations in p110? isoform or via loss of the PTEN tumor suppressor. Importantly, PI3K enzymes are highly suited for pharmacological intervention, making them attractive targets for cancer therapy. In fact, there are a number of PI3K inhibitors from major pharmaceutical companies that have entered clinical trials for cancer treatment, but most of these inhibitors target all p110 isoforms, which may cause side effects arising from the essential roles of PI3K in normal physiology. While isoform specific inhibitors are being further developed, most of which are directed toward p110? (for solid tumors) or p110? (Hematological malignancies). We believe that the drug companies have blundered by failing to develop p110?-specific inhibitors. We and others have recently demonstrated that tumors driven by PTEN loss are specifically dependent of p110? not p110?. The broad goal of this project is to generate p110? -specific inhibitors for use as new, targeted therapeutics in diverse cancers featuring PTEN mutations. To this end we have assembled a team of scientists optimized to achieve this goal. Our team's unique reagents for assessing PI3K signaling, coupled with and our expertise in protein chemistry, X-ray crystallography, medicinal chemistry and animal models, position us to effectively develop p110? inhibitors over a two-year time period for future clinical trials. Our specific goals are to generate cell-based systems and genetic models to determine the role of p110? in tumorigenesis driven by PTEN in different tissue types and to test p110? -specific inhibitors, to purify large amounts of active p110? for enzyme assays and crystallography and to pursue a chemistry campaign to design and evaluate new scaffolds for p110? inhibition and optimize 2 of these scaffolds using both cell and animal models and structural information from a complex of p110? and an inhibitor. PUBLIC HEALTH RELEVANCE: Statement of the relevance of the project to public health: The PTEN tumor suppressor is genetically or epigenetically silenced at high frequency in may types of human cancer. Recent data suggests that tumors driven by PTEN loss are dependent on the activity of a particular isoform of PI3K termed p110?. This project is designed to speed development of p110? inhibitors and hence could have a positive impact on many cancer patients.

We have discovered an interesting set of functional interactions among three proteins: SV40 Large T antigen (LT), the p53 tumor suppressor protein and the mitotic kinase Bub1. Bub1 is a member of the family of checkpoint proteins that monitor the assembly of the mitotic spindle, and has been found to be mutated in certain human cancers characterized by aneuploidy. LT antigen can also cause genomic instability by inducing chromosomal aberrations and aneuploidy. Genetic analysis demonstrates that interaction of T antigen with Bub1 is not required for immortalization but is necessary for T antigen to drive viral replication and transform. Notably LT appears to act as a scaffold bringing Bub1 to p53, which then is phosphorylated on ser37 directly by Bub1 and on ser15, indirectly by a second kinase, leading to the stabilization of p53. Preliminary data suggests that this stabilization fo p53 is necessary for transformation by LT at least in certain circumstances. Also of note is finding that downregulating Bub1 function either by LT expression or by RNAi against Bub1 in the absence of LT, results in p53 dependent cellular senescence, accompanied by phosphorylation of ser37. Recently another group has demonstrated that ras induced senescence, which we can block with a dominant negative allele of Bub1, is also dependent on phosphorylation of p53 at serine 37. The key questions in the context of this grant are how Bub1 contributes to LT mediated replication and transformation and to suppressing tumor formation via senescence in the absence of LT. We can imagine several possible mechanisms: LT may use Bub1 to regulate replication and transformation via its direct phosphorylation of p53, via Bub1 mediated phosphorylation of other targets in the LT complex or more indirectly via by perverting Bubl's role in regulating genome stability. These mechanisms need not be mutually exclusive. Obviously similar mechanisms may be in play as Bub1 guards normal cells from transformation by promoting senescence. In this application we propose to probe each of these mechanisms in turn.

Although HER2-directed therapies are effective in the subset of breast cancers characterized by amplification of the HER2 gene (HER2+), resistance to these therapies remains an important clinical problem and the mechanisms of this resistance are not well defined. Recent data demonstrate that at least 40% of HER2+ breast cancers have activating mutations in the PI3-kinase gene (PIK3CA) or other alterations in the PI3K pathway. Furthermore, preclinical and clinical investigations, including our own, have implicated these PI3K pathway alterations as potential mediators of resistance to anti-HER2 therapy and have demonstrated that combining PI3K inhibitors with anti-HER2 agents can overcome this resistance. In Project 2, we aim to optimize the application of PI3K-directed therapies, both in novel genetically-engineered mouse (GEM) models and subsequently in a clinical trial. Since a combination of anti-HER2 therapy and PI3K inhibition will likely be ineffective for some patients due to additional mutations bypassing the HER2/PI3K pathway, we will utilize recently developed methods to study mechanisms of resistance that lie outside the PI3K pathway. The aims of this project are to identify and overcome both PI3K-dependent and -independent mechanisms of resistance to targeted therapy of HER2+ breast cancer. Specifically we will: 1) Use GEM models to optimize PI3K-targeted treatment strategies for each subset of HER2+ breast cancer, with an emphasis on comparing pan-PI3K inhibitors with isoform specific agents, 2) Identify novel resistance mechanisms to HER2- and PI3K-targeted therapies in GEM models, and 3) Evaluate the role of PI3K inhibition in conjunction with HER2-targeted therapy in a preoperative clinical trial of patients with HER2+ breast cancer. Tumor tissue from that trial and others will be analyzed with next generation sequencing techniques in order to validate resistance mechanisms identified in Aim 2. Together these studies will strengthen our ability to overcome therapeutic resistance in HER2+ breast cancer and thereby improve outcomes for patients with this disease.

? DESCRIPTION (provided by applicant): Loss of the tumor suppressor phosphatase and tensin homolog (PTEN), the key negative regulator of Phosphatidylinositol 3-Kinase (PI3K) activity, is one of the most common genetic events in primary prostate cancer (PCa). Moreover, its frequency further increases in metastatic PCa, indicating that hyperactivation of the PI3K pathway plays an important role in the pathogenesis of PCa and implicating PI3K as an attractive target in this tumor type. However, early clinical trials with PI3K inhibitors in multipe cancer types, including PCa, have been disappointing. The clinical efficacy of these early PI3K inhibitors, which were largely pan-PI3K inhibitors that block the action of all Class I PI3K isoforms, may have been limited by their lack of isoform specificity. Indeed, preclinical work and emerging clinical trials suggest that inhibitors of individual isoforms may achieve greater efficac with fewer side effects. For example, Cal101, an inhibitor of the p110d isoform of PI3K, has demonstrated remarkable clinical efficacy in certain B-cell malignancies. Early unpublished clinical results also suggest that p110a inhibitors are outperforming pan-PI3K inhibitors in luminal breast cancer. Because the PIK3CA gene encoding p110a is frequently mutated in tumors, p110a has garnered the bulk of the attention from pharmaceutical and basic researchers. However, using a murine genetic model we identified p110ß as a key target in PTEN-null prostate tumors. We and others have subsequently shown that human PTEN-null prostate cancer cell lines are selectively dependent on p110ß. We have further identified Kin-193, also known as AZD6482, as a potent and specific p110ß inhibitor suitable for studies in vitro and in mice. Notably, a new p110ß inhibitor, GSK2636771, is now in clinical trials in patients with PTEN-deficient advanced solid tumors (NCT01458067). Therefore, it is both imperative and timely to carefully investigate p110ß inhibitors in preclinical settings. In this gant we propose to evaluate the therapeutic potential of this novel class of inhibitors in prostate cancer in vitro and in vivo using human cancer cell lines, genetic mouse models as well as primary human prostate tumor explants, to develop biomarkers predictive of response to p110ß inhibition, and to identify optimal combination partner agents for p110ß-based therapy. The studies proposed in this grant will likely provide important information that will help to optimize the clinical impact of this class of inhibitors in PTEN-deficient tumors.

PROJECT SUMMARY This application seeks the mechanisms underlying the roles of YAP/TAZ in cancer. A wide variety of human cancers have altered YAP/TAZ signaling that is generally connected to prognosis. YAP and its cousin TAZ are the major effectors of Hippo pathway that controls organ size, cell movement, proliferation and survival. This pathway also regulates other important signaling pathways (Hedgehog, Notch, TGF? and Wnt). We will use murine polyomavirus (MuPyV) to probe YAP function. MuPyV causes a broad range of tumors. Its study has illuminated the roles of tyrosine phosphorylation and PI3 kinase in cancer. It continues to point to important issues, such as the role of PP2A A? isoform, in cancer. Recently, we reported that MuPyV's small T antigen (ST) directly binds YAP and alters its phosphorylation, stability, and function by promoting its association with PP2A. This interaction allows ST to block differentiation in several systems. Extensive preliminary data on MT, the oncogene of MuPyV, show that MT also associates with YAP. This association is important for MT driven neoplastic transformation. MT also affects YAP phosphorylation, but MT's effects are different than those of ST. Our first aim is to study ST regulation of the differentiation of stem-like cells. Since we know that ST also interacts with TAZ, we will test whether it contributes to the phenotype. This aim will include a determination of the transcriptional mechanisms involved and the YAP effector protein associations that ST uses to achieve these alterations. Our second aim will be to study the roles of YAP in MT driven transformation. Here too we will be interested in changes in gene expression and protein-protein associations. In our last aim we seek to examine the roles of YAP in MuPyV MT mediated transformation in vivo using transgenic and knockout mice. In the context of this Program Project, we have a great opportunity to exploit the virus interactions to gain a deeper insight into Hippo signaling in human oncogenesis. By comparing our results on the effects of MuPyV ST/MT on PP2A/YAP/TAZ with the results of our colleagues working with SV40 ST and Merkel ST and those of the Hahn lab working on YAP in K-Ras driven human cancers, we should be able to determine which of YAP's effects on cellular signaling are truly important, greatly enhancing the impact of our data. Ideally, these studies will lead us to our ultimate goal of translating our work into human cancer therapy, as we have previously done for PI3 kinase.

Project Summary/Abstract Triple-negative breast cancers (TNBCs) account for 15 to 20% of breast cancer cases, are often clinically aggressive, and typically exhibit high rates of recurrence and mortality. Approximately one third of TNBC cases feature genetic inactivation of the PTEN tumor suppressor, and a considerably larger percentage exhibit PTEN deficiency due to epigenetic and post-transcriptional mechanisms. Since PTEN is the major negative regulator of PI3K activity in cells, PTEN-deficient cancers are characterized by hyper-activated PI3K signaling and were therefore expected to respond to therapeutic PI3K inhibition. However, pan-PI3K inhibitors have demonstrated poor activity against PTEN-deficient cancers in clinical studies. Notably, we have previously shown that tumors driven by loss of PTEN are uniquely dependent on the p110? isoform of PI3K. This finding likely explains why PI3K inhibitors first tested on PTEN null tumors failed in the clinic, as they primarily targeted p110? and did not adequately inhibit p110?. New p110?-specific compounds, on the other hand, are now showing clinical promise in PTEN-deficient solid tumors. In attempting to understand the molecular mechanisms that uniquely couple PTEN loss to p110? activation, we have uncovered a set of molecular mechanisms that not only explain how p110? is activated in response to loss of PTEN, but also suggest potential mechanisms of resistance and drug targets that could be targeted in combination with p110? inhibition. In this project, we will rigorously evaluate therapeutic strategies combining p110? inhibitors and targeted therapies to the targets we have identified using multiple genetically-engineered mouse models (GEMMs) of TNBC with PTEN deficiency, as well as in a panel of patient-derived xenograft (PDX) models of TNBC and ex vivo primary organoid cultures. Our preliminary data and pilot in vivo studies have provided encouraging results indicating that targeting these pathways in combination with p110? inhibition results in better response. Moreover, we found that p110? is crucial for mediating immune evasion in PTEN-deficient TNBC and that p110? inhibition strongly synergizes with anti-PD-1 immune checkpoint blockade to inhibit tumor growth. These findings revealed a previously unexplored role for p110? signaling in immune modulation, and provide a strong rationale for this combination treatment in PTEN-deficient TNBC. We will therefore investigate combined p110? inhibition and immunotherapy in GEMMs and organoid models of PTEN-null TNBC. Furthermore, we will research the molecular mechanisms by which p110? mediates tumorigenesis and immune evasion. As we move forward, we expect that the results from this project will provide important pre-clinical information for the design of future clinical trials of these combination therapies.

Project Summary/Abstract In studying the PI3 Kinase isoform dependence of different tumor types, we made an extremely surprising finding that now turns out to have considerable clinical importance. This finding forms the basis for this OIA application. We discovered that tumors driven by the loss of the PTEN tumor suppressor are uniquely dependent on the p110? isoform of PI3 Kinase. This finding likely explains why PI3K inhibitors first tested on PTEN null tumors failed in the clinic, as they were poor p110? inhibitors. New p110b specific compounds are now showing clinical promise. In attempting to understand the molecular mechanisms that uniquely couple PTEN loss to p110? activation, we have uncovered a set of molecular mechanisms, which not only explains how p110? is activated in response to PTEN loss but also suggests why the same tumors might quickly become partially or even totally resistant to PI3K inhibition. Notably the same mechanisms clearly suggest other drug targets, which can and should be attacked in combination with PI3K in PTEN null tumors. Our very recent data identify 2 proteins that uniquely interact with p110?, and not with p110?, form a positive feedback loop in the absence of PTEN. One of these proteins is the small ras family GTPase known as Rac, which interacts with p110? but not p110?. We have recently shown that Rac localizes p110b to the lipid rafts where it is activated. Thus Rac is an upstream activator of p110?. However Rac family members are unique in that their activators, the Rac GEFs, are activated by the phosphoinositide products of PI3Ks. Thus Rac is also a downstream effector of p110?. The interactions of Rac and p110? constitute the very definition of a positive feedback loop. However, this leaves open how the Rac/p110? feedback loop is initiated- what activates p110?/Rac in the first place. We have found that the activation event is dependent the small adapter protein CRKL which is also a p110? specific binding protein. Activation of CRKL occurs via a SRC/p130Cas signaling cascade that is also activated by PTEN loss. Notably SRC signaling renders cells resistant to PI3K inhibition. Finally and most exciting we have found that p110b inhibitors synergize with immune checkpoint blockade. We have generated data already showing the inhibitors of SRC RAC PAK (another downstream target of RAC) lipid raft formation and immune checkpoints can all combine well with p110b inhibitors in vitro, and in some cases, in vivo. This grant will focus more rigorous testing of new drug combinations on the one hand and on the other hand, further refining our mechanistic understanding of the effects of PTEN loss to generate even better combination therapy.

Project Summary/Abstract The faculty of the Dana-Farber Cancer Institute (DFCI) is requesting funds to support a new, integrative, postdoctoral training program in Cancer Chemical Biology and Metabolism (CCBM) ? two synergistic and conceptually related research disciplines grounded in small molecule chemistry and biochemistry that are of growing importance from both basic and translational research perspectives. The goals of this program are to: 1) provide training in a wide range of topics relevant to cancer chemical biology and metabolism such as preclinical cancer target validation, chemical screens, structure-based drug discovery, tumor metabolism, cross-talk between tumor and host metabolism, cellular bioenergetics and metabolomics, mechanistic biochemistry, proteomics and chemoproteomics, and animal tumor modelling; 2) familiarize trainees with state- of-the-art technologies and approaches in chemical biology and metabolic science, 3) train fellows to identify important questions and approaches that will move the field forward and provide translational opportunities to impact cancer treatment; 4) provide trainees with opportunities to develop and enhance their grantsmanship and scientific communication skills; 5) coach fellows on their paths to independent careers in science by implementing Individual Development Plans (IDPs), individual postdoctoral mentoring committees, and customized opportunities for trainees to acquire experience in mentoring and teaching depending on their career goals. The CCBM Training Program will support 8 postdoctoral fellows per year for 2-year appointments. It will be administered by a Director (Dr. Thomas Roberts, Contact PD/PI) and a co-Director (Dr. Nika Danial, PD/PI), and will include 26 mentors and 6 co-mentors from 4 departments at DFCI. The program Director and co- Director will work closely with a Training Oversight Committee (TOC), an External Advisory Board (EAB), and a trainee-led advisory group consisting of current and past program fellows who will help identify ways to optimize the program. Additional strategies to measure trainee satisfaction, quantify outcomes, and evaluate the effectiveness of the program will include annual surveys from trainees and mentors, as well as exit and alumni interviews. Institutional support through state-of-the-art core facilities and technology platforms, as well as funds to support T32-specific training activities and yearly EAB meetings will significantly enhance the training program. In addition to offering robust training in a diverse range of scientific and career development skills, a key goal of the program is to recruit and support a more diverse group of trainees. To further ensure the program's effectiveness in enhancing diversity, our EAB and Training Oversight Committee include leaders with deep experience in diversity and inclusion. We believe the integrative CCBM Training Program will prepare fellows for a range of research careers and train the next generation of scientists who will thrive in the 21st century biomedical workforce that will make breakthroughs in cancer research and treatment. !