Mechanisms underlying the interactions between the proteasome inhibitor carfilzomib and HDAC inhibitors were examined in both germinal center (GC) and activated B-cell (ABC) subtypes of human diffuse large B-cell lymphoma (DLBCL). Simultaneous exposure to minimally toxic concentrations of carfilzomib and HDAC inhibitor vorinostat resulted in the release of mitochondrial pro-apoptotic proteins SMAC and cytochrome c, pro-apoptotic caspase activation, and synergistic induction in apoptosis in both ABC and GC DLBCL subtypes. These events were associated with a marked increase in the stress kinase JNK, ROS generation, G2-M cell cycle arrest, as well as induction of DNA damage. Genetic knockdown of JNK resulted in a significant decrease in carfilzomib/vorinostat induced cell death. Co-administration of the antioxidant MnTBAP significantly reduced carfilzomib/vorinostat induced cell death, and resulted in a marked decrease in caspase-3 as well as a striking decrease in JNK phosphorylation. Tumor growth reduction was also observed in animal models that were treated with a combined regimen of carfilzomib and vorinostat. Finally, the combined treatment of carfilzomib/vorinostat was able to overcome any cross-resistance to carfIlzomib in bortezomib resistant cells. Collectively, these finding indicate that the combined regimen of carfilzomib and HDAC inhibitors promote lethality in ABC and GC human DLBCL cells by a variety of mechanisms both in vitro and in vivo. Further studies are necessary for clinical development of this drug regimen.

Innate immunity provides the first line of defense against invading pathogen by recognizing and mounting a response to the pathogenic challenge. Among the cellular mechanisms of the innate immune response, Toll-like receptor 3 (TLR3) recognizes viral dsRNA and signals subsequent production of type-I interferon. The TLR3:interferon signaling cascades contains a kinase complex composed of two kinases and a scaffold protein, NAK-associated protein 1 (NAP1). The role of NAP1 in modulating kinase activation or regulation is unknown. A key inhibitory protein identified in the TLR3:interferon pathway, silencer of inhibitor of κBα kinase ε (SIKE), blocks the activity of this kinase complex through an unknown mechanism. The long term goal of this project is to define how protein:protein interactions modulate signal transduction in this pathway as mediated by host in the context of an immune response, pathogen as it attempts to subvert the host immune system and in disease states such as cancer and obesity. Objectives for this thesis were to produce recombinant SIKE and NAP1 material that could be used to elucidate the self-association pattern and hetero-interactions mediated by these components within the pathway. SIKE full-length, SIKE 72 (residues 72-207), NAP 270 (residues 1-270) and NAP 255 (residues 1-255) expression constructs were completed and recombinant protein produced using either a bacterial or baculovirus/insect cell expression system. Self-association was characterized by size exclusion chromatography and analytical ultracentrifugation. Hetero-interactions were explored via co-precipitation assays of recombinant proteins. SIKE 72 formed a primarily dimeric structure whereas NAP 255 forms a single species that appears to be monomeric at this stage of analysis. Hetero-interactions form between the kinase TBK-1 and NAP 255 and also TBK-1 and SIKE 72. TBK-1 shows a higher preference for binding NAP 255 in the prescence of both NAP 255 and SIKE 72. This work provides methodology to produce recombinant material for two components of the TLR3:interferon pathway and their initial biochemical characterization for both self-association and hetero-interactions.

To defend against pathogen challenge, multi-cellular organisms mount an immune response that recognizes, sequesters and eradicates invading infectious agents. Critical to this safeguard is the receptor-mediated detection of pathogens. Pathogen recognition then initiates a variety of signaling cascades that lead to the modulation of genes orchestrating an immune response. Toll-like receptor 3 (TLR3), a transmembrane receptor found in endosomes, is vital to the innate immune response against viruses. Double-stranded RNA (dsRNA) stimulation of TLR3 initiates a signaling cascade that leads to the production of type-I interferons and proinflammatory cytokines necessary to trigger the protective defenses of the immune system. Critical to this pathway is the activation of a kinase, TANK binding kinase 1 (TBK1), which phosphorylates the downstream transcription factors, IRF3 and IRF7, and leads to the production of IFN-beta. Interestingly, TBK1 function has been implicated in a number of other signaling cascades ranging from the insulin response and vesicle transport to xenophagy and anti-viral immunity. Increasingly, however, TBK1 dysregulation has been linked to autoimmune disorders and cancers, heightening the need to understand regulatory controls of TBK1. As a result, this dissertation investigates three components of the TLR3 signaling cascade in an attempt to further advance our understanding of the innate immune response. First, investigations into the adjuvant potential of dsRNA reveal that a 139bp dsRNA molecule is a viable candidate for vaccine adjuvant studies. Next, structural and functional studies of TLR3 with neutralizing antibodies provide evidence for a new TLR-signaling model in which dsRNA:TLR3 signaling units laterally cluster to achieve efficient signaling. Finally, cell-based assays, biophysical experiments and kinetic investigations into the mechanism by which an endogenous regulator of the TLR3 response, SIKE, functions, reveal that SIKE not only inhibits TBK1-mediated IRF3 phosphorylation, but is also a high affinity substrate. Findings from this study further suggest that SIKE regulates a critical catalytic hub not by direct repression of activity, but by redirection of catalysis through substrate affinity. Taken together, the results presented in this dissertation establish a foundation for building long-term studies on the function, regulation and viral subversion of the innate immune response.

Ovarian cancer is the most lethal of all gynecological cancers. Current ovarian cancer drug regimens, including taxanes and platinum-based agents, are susceptible to chemoresistance necessitating the development of novel chemotherapeutics. Within tumors pathogen-derived ligands, such as dsRNA, can activate pattern recognition receptors (PRRs) that are capable of inducing apoptosis. In this dissertation we have found that in ovarian cancer cell lines (DOV-13, SKOV-3, CAOV-3, and OVCAR-3), dsRNA treatment alters cell survival. When treated with dsRNA, ovarian cancer cell lines and patient samples could be divided into two categories, responsive which undergo significant levels of apoptosis (CAOV-3 and OVCAR-3) or non-responsive which are unaffected (DOV-13 and SKOV-3). Following dsRNA treatment, dsRNA receptor expression levels increase in responsive cell lines and patient samples only. This suggests a potential role for dsRNA receptors as biomarkers to identify dsRNA-responsive patients. Detailed investigation of the mechanism by which cell lines succumb to or avoid dsRNA-induced cell death showed that in responsive cell lines, NF-kappaB, IFN-beta and caspase 3 activation occurred. Cell death was caspase and IFN-dependent. In non-responsive cell lines, increased c-IAP2 levels and RIP1 kinase ubiquitination occurred, as well as, an increase in basal level autophagy with dsRNA stimulation. However, individual blockade of these pathways did not restore dsRNA-induced apoptosis. In a non-responsive cell line, dsRNA enhanced the action of paclitaxel, carboplatin, and vorinostat through an as yet undetermined mechanism. In a responsive cell line, dsRNA produced a synergistic effect when combined with these drugs. These novel dual therapies, innate immune ligand plus cytotoxic agent, may find application in chemoresistant ovarian cancers.

In the past two decades, the field of protein engineering has evolved rapidly to include new genetic and chemical techniques to alter protein function. Protein engineering seeks to improve enzyme properties through powerful methods that specifically incorporate novel or improved function in proteins. One such method is protein ligation, which is used to selectively link synthetic and recombinant polypeptides. Due to the limitations of current protein labeling techniques, simple site-specific modification methods remain in high demand. Use of enzyme-based labeling has been the focus of various studies because of its substrate specificity. Sortase-mediated transpeptidation is one approach that has been well documented. Staphylococcus aureus sortase A (SrtAstaph), a membrane-anchored cysteine transpeptidase present in gram-positive bacteria, covalently anchors virulence-associated surface proteins to the peptidoglycan cross bridge of the cell wall. SrtAstaph, one of the most characterized sortases, has found numerous applications in the semi-synthesis of protein and peptide conjugates. While current studies have demonstrated the growing range of applications for sortase A, the enzyme itself has seen very few improvements. In steady-state kinetic analysis, the calculated K cat value of SrtAstaph was 2.27 × 10−5 s−1 indicative of its slow in-vitro turnover rate. Due to sortase’s relative inefficiency, several studies documented the use of excessive amounts of the enzyme in vitro (>30μM) or reactions were incubated for long periods. Through the use of directed evolution, we aimed to improve the catalytic activity of sortase A. Using random mutagenesis and an in vivo bacterial-based screen we isolated a variant that showed a 13-fold increase in its catalytic efficiency when compared to wild-type. This sortase mutant will enable more efficient labeling of LPETG-tagged substrates and will provide further insight into the enzyme’s molecular mechanism of catalysis, which is currently limited.

YKL-40 is a secreted glycoprotein, which is a shared biomarker of chronic inflammation and oncogenic transformation. Indeed, YKL-40 expression is up-regulated in many diseases including multiple sclerosis, Alzheimer disease, viral encephalitis, HIV-associated dementia, brain infarction, and traumatic brain injury. YKL-40 is also expressed by several solid tumors, such as osteosarcoma, ovarian carcinoma and glioblastoma multiforme (GBM). It promotes the migration and invasion of astrocytes as well as GBM cells. Serum YKL-40 levels have been shown to directly correlate with tumor grade and potentially tumor burden in GBM. In contrast to the numerous reports documenting elevated expression of YKL-40, relatively little is known about inflammatory mediators and specific molecular mechanisms that control its expression. For my PhD project, I decided to elucidate the mechanism regulating YKL-40 expression in inflammation and to understand how YKL-40 mediates the migration and invasion of GBM cells in vivo. As per my first project, I found that YKL-40 expression is up-regulated in many inflammatory models such as irritant induced inflammation (turpentine), EAE (experimental autoimmune encephalomyelitis), irradiation induced brain inflammation and in the HIV-TAT overexpression model. YKL-40 expression is also up regulated in oligodendroglioma patient samples. We show that YKL-40 expression is activated by major pro-inflammatory cytokines including IL-1, IL-6, and OSM in astrocytes, and that the activation of YKL-40 expression by cytokines depends on the STAT and NF-κB regulatory elements located within the YKL-40 promoter. Additionally, we discovered that STAT3 is a major regulator of YKL-40 expression in response to the IL-6 family cytokines, including OSM. Indeed, both depletion of STAT3 expression or overexpression of dominant-negative STAT3 abolishes activation of YKL-40 expression. In contrast, although IL-1 activates NF-κB (p65/p50) in astrocytes, IL-1-induced YKL-40 expression is p65/p50 independent, and instead regulated by the RelB/p50 complexes. Interestingly, OSM promotes formation of p50/RelB complexes. In conclusion, we found that YKL-40 is up regulated by major pro-inflammatory cytokines during inflammation associated with various disease models. Thus, we uncovered the regulatory mechanism that governs the expression of YKL-40 during sterile inflammation. Next, I studied the role of YKL-40 in GBM in vivo. The major obstacle for treatment of GBM is the unique ability of GBM cells to diffusely infiltrate healthy brain tissue. GBM tumors express high levels of YKL-40, and we showed that administration of YKL-40 increases migration and invasion of U1242 cells in vitro. We used the U1242 cell line for our studies, due to its unique ability to form infiltrative tumors, which reflects the real brain pathology in GBM. Additionally, knock-down of YKL-40 suppresses the migration and invasion of GBM cells in vitro. To study this phenomenon in vivo, we utilized an inducible lentiviral vector containing shRNA to YKL-40. We generated stable GBM cells containing this shYKL-40 vector and used them in a mouse xenograft model for glioma to evaluate the effect of YKL-40 in invasion and migration of GBM cells. RelB expression is increased in GBM patients, and it is associated with up-regulation of a distinct subset of gene. Here we show that RelB expression is increased during inflammation in response to IL-1, where it can enhance the expression of YKL-40, which can modulate the inflammatory response. Thus, we found that YKL-40 is up regulated by major inflammatory cytokines during inflammation associated with various diseases. In addition, YKL-40 secreted by brain cells other than GBM cells may have an effect on invasion of GBM cells in vivo.

MDM2, the human homologue of the Mouse Double Minute 2 gene product, has been shown to be over-expressed in many cancers and to induce tumorigenesis. The role of MDM2 in oncogenesis was thought to be p53 dependent. However recent years have shown MDM2 to be a key player in a complex network of interactions that affect cell cycle, apoptosis, and tumorigenesis in a p53 independent manner. Here we report a novel p53 independent role for the multidimensional protein MDM2; its ability to induce phosphorylation of Akt at serine 473 residue. Transient and stable over-expression of MDM2 in cultured cell lines induces Akt phosphorylation. Silencing of MDM2 expression in cancer cells that over express MDM2 inhibits Akt phosphorylation suggesting endogenous MDM2 participates in Akt phosphorylation. Stable up-regulation of MDM2 expression reduced sensitivity of cells to chemotherapeutic drugs such as Etoposide, Carboplatin or Paclitaxel. The domain of MDM2 responsible for drug resistance overlaps with the Akt phosphorylation domain. An inhibitor of Akt phosphorylation abrogated MDM2-mediated Akt phosphorylation and reduction of Etoposide sensitivity indicating that MDM2 reduces Etoposide sensitivity of cancer cells by activating the Akt phosphorylation. A PI-3 kinase inhibitor Wortmannin inhibits the ability of MDM2 to induce Akt phosphorylation and silencing of Rictor, a known kinase of Akt, does not hamper the ability of MDM2 to induce phosphorylation of Akt. MDM2-mediated Akt phosphorylation does not require p53, and the p53-interaction domain of MDM2 is dispensable for Akt phosphorylation. The presence of MDM2 enhances the Insulin like Growth Factor 1 mediated activation of Akt. Further cells harboring MDM2 show enhanced Interleukin 8( IL8) activation, which could be a possible mechanism of Akt activation. Downstream of Akt activation we show increased events that have been correlated with Akt activation like increased Bcl-2 levels increased processing of NF-κB2, and GSK3α/β phosphorylation among others. Our observation reveals a novel signaling function of MDM2 important for regulation of cell growth and chemotherapeutic sensitivity through Akt phosphorylation.

Lysophosphatidic acid (LPA) is a naturally occurring bioactive lysophospholipid that mediates a broad range of cellular processes such as cell proliferation, survival, migration and invasion. LPA also plays a potential role in human oncogenesis as suggested by elevated expression of its receptors and its producing enzymes in malignant tissues. In the current study, we demonstrated that LPA is a potent mediator of interleukin-6 (IL-6) production in ovarian cancer. IL-6 is a pleiotropic cytokine which is thought to be an important mediator of ovarian cancer development and progression. Here, we demonstrated that IL-6 levels are indeed increased in the plasma of ovarian cancer patients as compared to normal women. The IL-6 concentrations in ascites of ovarian cancer patients are even higher than those present in the plasma samples. These results suggest that increased IL-6 are expressed and secreted by ovarian cancer cells, forming a gradient from the ascites to the blood. Ovarian cancer cells indeed produce IL-6 in culture. However, when these cells are starved in serum-free medium, they cease producing IL-6, suggesting that IL-6 is not constitutively expressed, but rather in response to exogenous factors present in serum. We showed that IL-6 expression is not driven by peptide growth factors such as insulin-like growth factor I or epidermal growth factor. Instead, IL-6 expression is most potently induced by the lysophospholipid growth factor LPA. Treatment of ovarian cancer cells with LPA leads to transcriptional activation of the IL-6 gene promoter through activation of the NF-kB and C/EBP transcription factors. LPA also induces tyrosine phosphorylation and activation of Stat-3, a well known intracellular effector of IL-6. However, blockade of IL-6 with a neutralizing antibody only slightly reduced Stat-3 phosphorylation in response to LPA, suggesting that LPA may induce Stat-3 directly or through secondary mediators other than IL-6. Together, these studies demonstrate the role of LPA in regulation of IL-6 production and the underlying mechanism in ovarian cancer.

The Jak/Stat signaling cascade mediates cell proliferation, differentiation, survival, apoptosis and immune responses. Aberrant activation of this pathway mediates neoplastic transformation and abnormal growth of many malignancies including breast cancer, the most common cancer among women, and the second leading cause of cancer deaths in women in United States. The mechanism by which the Jak/Stat pathway modulates the pathogenesis of breast cancer is unclear. This dissertation elucidates roles of Jak/Stat members that mediate the pathogenesis of breast cancer. For these studies, we used 4T1 mouse mammary tumor cells as a model which mimics human breast cancer. First, we investigated the role of Tyk2 tyrosine kinase in the pathogenesis of breast cancer. Here we show for the first time that compared with wild type mice, Tyk2 -/- mice show increased tumor growth rate as well as metastatic disease and splenomegaly when inoculated with 4T1 breast cancer cells. Such increased tumorigenicity was associated with a significant decrease of IFNg production in 4T1 tumor-bearing Tyk2 deficient mice T cells compared with wild type (WT) mice. We demonstrated that NK cells or CD8+ T cells control tumor growth in both Tyk2-/- and WT mice, but neither Tyk2-/- NK cells alone nor Tyk2-/- CD8+ T cells alone do not contribute to enhanced tumor growth and metastatic disease of Tyk2-/- mice. Tumor-bearing Tyk2-/- mice have increased level of myeloid-derived suppression cells than tumor-bearing mice. Tyk2-/- MDSCs have a slight increase in suppression of T cell proliferation. Since elevated phosphorylation of Stat3 has been seen in human and murine breast cancer, and expression of Stat3 in the mitochondria (mitoStat3) appears to have important affects on cell growth, we studied the ability of Stat3 targeted to the mitochondria (MLS Stat3) to influence growth and metastasis of 4T1 cells. We show that a serine mutant of Stat3 expressed in the mitochondria (Stat3 S727A) inhibits the ability of 4T1 tumor cells to grow and metastasize. In contrast, a serine to aspartic acid mutant of Stat3 (S727D) enhances tumorigenesis. We found that expression of mitochondrial-targeted Stat3 does not affect cell growth rate in cell culture under normal conditions, however in low glucose, the serine to alanine mutant shows reduced growth rate and ability to invade. Moreover, we found that expression of mitochondrial-targeted Stat3 protects cells from hypoxia. Our data indicate that serine phosphorylation of mitochondrial-localized Stat3 is required for cell transformation. In summary, our studies provided new insights into the role of Stat3 in breast cancer and suggest new therapeutic targets for the treatment of this disease.

Rett syndrome (RTT) is an X-linked neurological disorder primarily caused by mutations in the MECP2 gene. The majority of RTT mutations disrupt the interaction of MeCP2 with DNA or TBL1X/TBL1XR1, which forms the scaffold of NCoR/SMRT co-repressor complex. Patients with RTT show no signs of neuronal death, although they have abnormal neuronal morphology, indicating that it is a neurodevelopmental rather than a neurodegenerative disease. It has been shown that reactivation of silenced MeCP2 in mice rescues the RTT phenotype, which implies that the disease is treatable. The RTT mutations in MeCP2 cluster to two regions - the methyl-CpG-binding domain (MBD) and NCoR/SMRT Interaction Domain (NID). While the interaction between MBD and DNA has been biochemically and structurally characterised, there are no structural data about the interaction between MeCP2 NID and TBL1XR1. The aim of this work was to understand how mutations in the NID cause RTT by characterising the interaction between MeCP2 and TBL1XR1. I have solved the structure of MeCP2 NID bound to TBL1XR1 WD40 domain. I show that a small region of the MeCP2 NID makes extensive contacts with TBL1XR1, and that these contacts are mediated primarily by MeCP2 residues known to be mutated in RTT. I also measured the affinities between TBL1XR1 and MeCP2-derived peptides using fluorescence anisotropy and surface plasmon resonance assays. I determined the affinity between MeCP2 NID peptide and TBL1XR1 to be around 10- 20 μM, and show that mutations in either MeCP2 or TBL1XR1 can abolish this interaction. Taken together, these data strongly suggest that the abolition of the interaction between MeCP2 NID and TBL1XR1 WD40 domain is sufficient to cause RTT. This knowledge can help with the rational design of small drug-like molecules that might be able to mediate the interaction between mutated MeCP2 and TBL1XR1, potentially helping to treat the disease.