Prostate cancer is the most commonly diagnosed cancer in American men. Although much progress has been made in identifying the genetic alterations that underlie this disease, there remains a gap in our understanding of how these genetic changes interact to lead to cancer. Modeling human cancers in vivo using mice is a critical part of cancer biology to better understand human carcinogenesis. Most human cancers are thought to initiate from a mutation in a single cell or a few cells, which expand and become cancerous with the accumulation of additional mutations. However, current mouse prostate cancer models do not accurately mimic this process as mutations are generally induced in entire cell populations rather than in a few cells. To more accurately model human prostate cancer, we generated novel transgenic mice with focal overexpression of the oncogene c-MYC in the prostatic epithelium. Focal c-MYC activation resulted in mild pathology despite expansion of the c-MYC-positive cells. When combined with heterozygous or homozygous deletion of the Pten tumor suppressor gene, however, focal c-MYC expression promoted the development of prostate cancer. Cells with concurrent activation of c-MYC and loss of Pten were of higher grade and out-proliferated cells with mutations in Pten alone. In the prostate, Pten deletion activates the p53 pathway, which can induce either apoptosis or senescence. We found that concurrent c-MYC expression shifted the p53 pathway response from senescence in favor of apoptosis. Thus cooperativity between c-MYC and Pten leads to increased tumorigenicity due to the high rates of proliferation and reduced senescence in c-MYC/Pten mutant cells in spite of elevated rates of apoptosis. Our model of focal c-MYC expression allowed us to examine the characteristics of cells with distinct mutations (c-MYC vs. wild type or Pten-null vs. c-MYC;Pten-null cells) in the same mouse prostate to gain insight into the interaction and competition between cells in the cancer progression. This system provides an important cancer model to investigate heterogeneous and incremental nature of human prostate carcinogenesis.

The studies in this dissertation defined the molecular mechanism of prothrombin (ProT) activation by a recently-identified Staphylococcus aureus exoprotein, von Willebrand factor-binding protein (VWbp). We demonstrated that VWbp activates ProT conformationally in a mechanism requiring insertion of its NH2-terminal Val1-Val2 residues into the NH2-terminal binding cleft on ProT. Such non-proteolytic activation of ProT has been previously identified for only one other protein, staphylocoagulase (SC), which shares structural homology with VWbp. Kinetic studies of ProT activation by VWbp showed that it activates ProT by a unique substrate-dependent, hysteretic mechanism. The NH2-terminal domains of VWbp (VWbp(1-263)) bind weakly to ProT, forming an inactive complex that is fully activated through a slow conformational change mediated by binding of tripeptide substrates or the physiological substrate, fibrinogen (Fbg). The activity of VWbp(1-474) is increased compared to VWbp(1-263), supporting a contribution from the COOH-terminal region in ProT activation. The role of allostery in regulating binding and activation of ProT by VWbp was also demonstrated from the effect of the fragment 1 and 2 domains of ProT on VWbp function, as well as the cofactor effect of substrate on complex formation. Specifically, loss of fragment 1 of ProT or active-site occupation enhances affinity for both VWbp(1-263) and VWbp(1-474), suggesting a regulatory role for fragment 1 and verifying the substrate dependence of the activating conformational change. Further, the activation state of anion-binding (pro)exosite I influences ProT-VWbp binding. This pattern is consistent with that seen during the physiological pathway of ProT activation, where high-affinity exosite expression is linked to formation of a competent active site. Slow, progressive loss of the fragment domains of ProT*VWbp occurs through autocatalysis, reinforcing the potential role of these domains in ProT activation during S. aureus infections. Together, these findings reveal a pattern of regulation of ligand recognition similar to that of normal thrombin, but with strictly defined recognition of Fbg by the ProT*VWbp complex. The restricted substrate specificity provided by the hysteretic activation mechanism may allow VWbp to serve as a virulence factor in propagating the growth of fibrin-rich vegetations during acute infective endocarditis.

Cellular adhesions to the extracellular matrix (ECM) and to neighboring cells are vital interactions within metazoans. This dissertation examines cellular adhesion by utilizing the adhesive interaction between platelets and collagens that occurs at vascular wound sites and is a crucial step in hemostasis. Platelets are central elements in maintaining an intact circulatory system, but they also have a significant role in thrombotic pathologies such as heart attack and stroke. I analyze the contributions that the two platelet surface receptors, á2â1 integrin (á2â1) and glycoprotein VI (GPVI)/Fc receptor ã-chain (FcRã) complex, make toward the platelets interaction with collagens at vascular wound sites as well as the cooperation of G protein-coupled receptor (GPCR) signaling in the adhesion process. We demonstrate a novel mechanism of modulating the avidity of á2â1 for collagens that causes an increase in platelet adhesion and that is triggered through GPCR and GPVI/FcRã activation of phospholipase C (PLC). A discussion is included on platelet adhesion and signaling involved in hemostasis and thrombosis, substrate/integrin interactions, and anti-platelet therapies.

Invariant natural killer T (iNKT) cells are a specialized subset of immune regulatory cells that recognize glycolipid antigens and are thought to be pro-atherogenic under hyperlipidemic conditions. We previously reported that hyperlipidemic apolipoprotein E-deficient (apoE-/-) mice have decreased iNKT cell-mediated cytokine production in vitro and in vivo in response to alpha-galactosylceramide (a-GalCer), a prototypic iNKT cell glycolipid antigen. These data suggested changes in endogenous circulating lipids can affect normal iNKT cell functions. In the current study, we investigated whether dyslipidemia-associated perturbed iNKT cell function is due to intrinsic changes in iNKT cells or defects in the ability of antigen presenting cells to activate iNKT cells. Our data reveal that iNKT cells from dyslipidemic apoE-/- mice exhibit a phenotype similar to those rendered anergic due to chronic stimulation. We also tested the ability of B6 and apoE-/- splenic dendritic cells (DCs) to present a-GalCer to purified iNKT cells. Although DCs from apoE-/- mice were able to activate B6 iNKT cells, iNKT cells from apoE-/- mice were not able to respond to B6 DCs. These data suggest that chronic hyperlipidemia induces an iNKT cell phenotype that is unresponsive to further simulation by exogenous glycolipid, and that sustained unresponsiveness appears to be iNKT cell intrinsic. Additionally, our results indicate that increased circulating lipids exert direct effects upon iNKT cells which lead to decreased responsiveness.

Characterization of fibrinogen-binding surface protein B (FbsB) in human blood coagulation and fibrinolysis, and of the NH2-terminal dipeptide of staphylocoagulase (SC) in conformational prothrombin activation are presented in this thesis. FbsB was hypothesized to bind fibrinogen and to be a member of the Zymogen Activator and Adhesion Protein (ZAAP) family. Kinetic assays, affinity and size-exclusion chromatography experiments, and Western blots performed in these studies determined that FbsB neither binds fibrinogen nor belongs to the ZAAP family. However, it was concluded that FbsB binds human plasmin and plasminogen, significantly inhibits the rate of plasmin substrate hydrolysis, and enhances plasminogen activation in the presence of tissue-type plasminogen activator. FbsB may be only the second characterized Streptococcus agalactiae protein capable of contributing to the spread of infection through its interaction with human fibrinolytic proteins. <p> There remains a gap in knowledge about the importance of the sequence of the NH2-terminal dipeptide of SC in conformational prothrombin activation and the impact these conserved residues have in physiological serine protease activation. Kinetic assays with 29 different NH2-terminal SC(1-246) mutants revealed the relative promiscuity of these residues in prothrombin activation. Seven of the mutants screened activate prothrombin with 100-160% of the activity of wild-type. This data, along with future studies, may provide vital information about the mechanisms of prothrombin activation, and serine protease zymogen activation in general. <p>

In this dissertation, I examined the function of the alpha2beta1 integrin in epithelial tumor cells as well as in the tumor microenvironment. The alpha2beta1 integrin is a surface heterodimeric receptor found on epithelial, endothelial, fibroblasts, platelets, and several immune system cell subsets where it is responsible for cell-cell and cell-matrix interactions. Using the K14-HPV16 mouse model of inflammation-driven skin carcinogenesis on wild-type or alpha2beta1 integrin-null backgrounds, as well as an orthotopic murine breast cancer model, I elucidated the roles of this integrin during the multi-step progression towards cancer and eventual metastasis. My studies have implicated a role of the alpha2beta1 integrin in determining the growth and invasive behavior of tumor cells that is independent of the tumor microenvironment, yet influenced by the malignant cell type involved. My findings demonstrate an important function of the alpha2beta1 integrin on maintaining lymphatic vasculature integrity, thus providing novel insights into the function of this integrin in normal physiology as well as in tumor-associated lymphatics and metastasis. Through my examination of the multiple compartments known to drive cancer, it has been possible to ascertain this integrins contribution on the tumor cells, inflammatory cells, and additional cells of the tumor microenvironment towards mediating neoplastic progression and hematogenous as well as lymphatic metastasis.

<p> Patients with the autoimmune disease systemic lupus erythematosus (SLE) have an increased risk of developing premature cardiovascular disease, most notably atherosclerosis. While the mechanisms behind this increased risk are currently unknown, it is widely thought that immune dysregulation and inflammation contribute to SLE-accelerated cardiovascular disease (SACVD). <p> Our laboratory recently developed a model of SLE-accelerated atherosclerosis. This model uses a triple congenic mouse model harboring three lupus susceptibility loci derived from the NZM2410 mouse strain. Each locus confers phenotypes associated with SLE pathogenesis. Sle1 is associated with chronic lymphocyte activation and anti-nuclear antibody production. Sle2 is mainly associated with B cell hyperactivity while Sle3 mediates CD4+ T and antigen presenting cell hyperactivity. While having one or two intervals leads to various phenotypes associated with SLE, mice with all three intervals display an SLE phenotype similar to human disease. <p> We demonstrated that radiation chimeras of SLE-susceptible B6.Sle1.2.3 and low density lipoprotein receptor-deficient mice (LDLr.Sle1.2.3) have increased atherosclerosis, associated with increased T cell burden in lesions. In the current study, we take advantage of this mouse model in order to understand the mechanism of SACVD, specifically the role of T cells in mediating disease. Our data reveal that the increased atherogenesis observed in LDLr.Sle1.2.3 mice is independent of high fat diet feeding and that these mice have T cell phenotypes commonly observed in SLE patients. We find that transfer of individual lupus susceptibility loci associated with T cell hyperactivity (Sle1 and Sle3) is not sufficient to accelerate atherosclerosis; however, adoptive transfer of CD4+ T cells that contain all three lupus susceptibility loci into immunodeficient mice enhances atherogenesis. Moreover, we observe that an imbalance in regulatory and inflammatory T cell populations potentially contributes to disease pathogenesis. Taken together, our results indicate that T cells significantly contribute to SACVD and that their role is multifaceted. Furthermore, these studies enhance our scientific knowledge of how immune dysregulation mediates disease progression in autoimmunity and atherosclerosis and will hopefully facilitate the development of therapeutics designed to treat both diseases.

This goal of this dissertation was to investigate the molecular mechanisms of prostate tumorigenesis initiated by loss of the tumor suppressor NKX3.1. In human prostate cancer, NKX3.1 is lost during disease progression. Deletion of Nkx3.1 in mouse models of prostate cancer have indicated that Nkx3.1 loss promotes initiation but require additional mutations for progression. Human clinical data and mouse models suggest that loss of NKX3.1 and overexpression of the oncogene c-MYC collaborate to promote prostate tumorigenesis. To evaluate cooperation between Nkx3.1 and c-Myc, I generated transgenic mice with concurrent, prostate specific loss of Nkx3.1 and activation of c-MYC. Cooperation was demonstrated by the development of high grade prostate intraepithelial neoplasia lesions with signs of micro-invasive cancer and increased proliferation. Using a combination of gene expression analysis, chromatin immunoprecipitation and immunohistochemistry, I demonstrate that Nkx3.1 and c-Myc cross-regulate cancer relevant target genes which facilitates tumor progression. These findings significantly improved our understanding of the role of NKX3.1 loss in tumor initiation and progression. In addition, these studies identified tumor contributing gene targets that may provide new molecular therapeutic targets.

Coordination of viral replication requires successful and often complex interactions of viral proteins and cellular factors. Although much progress has been made in understanding the formation of specialized sites of replication where viral proteins and repurposed cellular factors direct genome multiplication and particle assembly, much remains unknown about the specific mechanisms of post-entry determinants of viral replication efficiency and cell tropism. Reoviruses are non-enveloped, double-stranded (ds) RNA viruses that display broad cell, tissue, and host tropism and serve as tractable models for the study of virus-cell interactions that dictate unique tropic properties. Reovirus replication protein μ2 has many functional and biochemical properties that are essential to post-entry steps in the viral life cycle. The goal of my dissertation research has been to determine how reovirus μ2, and its cooperation with other viral proteins, support progression of the viral replication cycle and contribute to reovirus cell tropism. Toward that goal, I characterized the genetic basis of strain-dependent reovirus replication in Madin Darby canine kidney (MDCK) cells. μ2 controls efficiency of reovirus replication in these cells, and amino acid residue 347 is the primary determinant of this phenotype. Furthermore, polymerase protein λ3 is a co-regulator of viral replication in a strain-dependent manner. I systematically tested post-entry steps in the reovirus life cycle to identify the critical step in reovirus tropism for MDCK cells. The replication block occurs following inclusion formation but prior to dsRNA synthesis, indicating that inclusions formed during non-permissive infection are morphologically normal but functionally abnormal. I also analyzed inclusion ultrastructure under permissive and non-permissive conditions of infection and found that particle assembly is inhibited in non-productively infected cells. These studies further an understanding of the role of μ2 in the viral life cycle to include initiation of particle assembly within viral inclusions and indicate that this step is subject to regulation by cellular factors.

The work in this thesis involves characterization of a novel Streptococcus agalactiae-secreted protein, skizzle, and its interactions with key proteins of the human fibrinolytic system. Skizzle binds human plasminogen (Pg) with high affinity and acts as a cofactor of Pg activation to form the clot-dissolving protease, plasmin. As a cofactor, skizzle uses two different mechanisms to enhance Pg activation by the endogenous Pg activators, urokinase and tissue-type plasminogen activator. Skizzle-enhanced Pg activation by urokinase is specific for the circulating, unmodified form, [Glu]Pg, and involves a skizzle-induced Pg conformational change to a more-easily activated conformation. Enhanced activation of both unmodified [Glu]Pg and modified [Lys]Pg by tissue-type plasminogen activator involves formation of a skizzle-containing ternary or quaternary complex with Pg and tissue-type plasminogen activator, resulting in enhanced Pg activation. To our knowledge, skizzle is the first S. agalactiae-secreted cofactor of human Pg activation. Skizzle has the potential to be a virulence factor in the pathogenesis of life-threatening S. agalactiae infections of newborns and immune-compromised adults.