Eukaryotic cells tightly regulate metabolism in order to sustain normal processes. Dysregulation of cellular metabolism is associated with multiple diseases including cancer. Polyamine metabolism is a tightly regulated process that is co-opted by multiple cancers for selective growth advantages. Polyamines are small organic molecules with two or more amino groups attached, whose biosynthesis is initiated by ornithine decarboxylase (ODC). Although much is known regarding the effects of polyamine metabolism and ODC on cellular processes, little is known regarding the intracellular signaling events that are regulated by polyamines. Clinical studies demonstrated that the ODC inhibitor difluromethylornithine (DFMO) was an effective chemopreventative strategy causing a reduction of colon adenomas in patients with prior colon polyps. However, the molecular mechanisms leading to this reduction are unknown. This dissertation provides mechanistic insight into the biological roles of the polyamines and show that these amines are regulators of multiple non-coding RNAs involved in cellular responses including effects on the let-7 microRNA family. Moreover, the polyamine modified translation factor eIF5A is demonstrated to regulate the oncofetal factor LIN28. This work also indicates that polyamines regulate the mTOR pathway and suggests alternative signaling nodes for polyamine-mediated regulation of cellular processes. Overall, these findings support the notion that polyamines are oncometabolites that are targetable and serve as a promising approach to manipulate oncogenic signaling for cancer therapy.

Glioblastoma (GB) is the highest grade and most common form of primary adult brain tumors, characterized by a highly invasive cell population. GB tumors develop treatment resistance and ultimately recur; the median survival is nearly fifteen months and importantly, the invading cell population is attributed with having a decreased sensitivity to therapeutics. Thus, there remains a necessity to identify the genetic and signaling mechanisms that promote tumor spread and therapeutic resistance in order to develop new targeted treatment strategies to combat this rapidly progressive disease. TWEAK-Fn14 ligand-receptor signaling is one mechanism in GB that promotes cell invasiveness and survival, and is dependent upon the activity of multiple Rho GTPases including Rac1. Here, we show that Cdc42 is essential in Fn14-mediated Rac1 activation. We identified two guanine nucleotide exchange factors (GEFs), Ect2 and Trio, involved in the TWEAK-induced activation of Cdc42 and Rac1, respectively, as well as in the subsequent TWEAK-Fn14 directed glioma cell migration and invasion. In addition, we characterized the role of SGEF in promoting Fn14-induced Rac1 activation. SGEF, a RhoG-specific GEF, is overexpressed in GB tumors and promotes TWEAK-Fn14-mediated glioma invasion. Moreover, we characterized the correlation between SGEF expression and TMZ resistance, and defined a role for SGEF in promoting the survival of glioma cells. SGEF mRNA and protein expression are regulated by the TWEAK-Fn14 signaling axis in an NF-kB dependent manner and inhibition of SGEF expression sensitizes glioma cells to TMZ treatment. Lastly, gene expression analysis of SGEF depleted GB cells revealed altered expression of a network of DNA repair and survival genes. Thus TWEAK-Fn14 signaling through the GEF-Rho GTPase systems which include the Ect2, Trio, and SGEF activation of Cdc42 and/or Rac1 presents a pathway of attractive drug targets in glioma therapy, and SGEF signaling represents a novel target in the setting of TMZ refractory, invasive GB cells.

Chaperone proteins such as heat-shock proteins 70, 90 and 110, glucose-related protein 94 and calreticulin have been reported to be effective anti-tumor vaccines when purified from a tumor source. We have developed a procedure utilizing a free-solution-isoelectric focusing technique to obtain vaccines from tumor or normal tissue sources that are rich in multiple immunogenic chaperone proteins, called Chaperone-Rich Cell Lysate (CRCL). Tumor-associated peptides are presumed to be the currency of T-cell mediated anti-cancer immunity, and tumor-derived chaperone vaccines are believed to be purveyors of such peptides. As a novel anti-cancer strategy, we have examined the extent to which the peptide repertoire of CRCL can be manipulated. Here, we explored the concept of creating a designer CRCL, utilizing the adjuvant properties and the carrying capacity of CRCL to deliver exogenous antigenic peptides for DC-based presentation and ultimately demonstrate the anti-tumor efficacy of the designer vaccine in vivo. Designer CRCL allows for the development of personalized vaccines to those afflicted with cancer expressing known antigens.Growing evidence indicates that the stress response, specifically involving HSPs, has a profound impact on tumor immunogenicity. Enhancement of T-cell-mediated immunogenicity correlates with the expression of inducible heat shock protein 70 (iHSP70), the major heat-inducible member of the HSP70 family. In addition, studies have shown tumor-specific cell surface localization of iHSP70 correlates with an increased sensitivity to lysis mediated by human natural killer (NK) cells. Given these findings, investigating novel and effective means of modulating the heat shock response within tumor cells may bear great therapeutic potential and result in potent anti-tumor immune activity. Withaferin A (WA) is a compound isolated from the plant Withania somnifera that has been shown to induce a robust transcriptional heat shock response. In our studies, we found that WA treatment resulted in increased surface expression of iHSP70 in several tumor types leading to significant immunostimulatory effects. These findings indicated that WA-dependent modulation of the heat shock response may enhance tumor immunogenicity. Given the potent immunomodulatory and anti-tumor effects of WA as well as the adjuvanticity and specificity of peptide-complexed CRCL against tumors, these therapies individually have shown profound anti-cancer activity.

Murine model organisms are an essential tool in the scientific community quest to decipher the molecular etiology of human diseases. Currently, several methods are used to induce or reproduce human diseases in mouse models using advanced genetic engineering techniques to mutate the wild-type genes. We utilized the Baygenomics gene-trap method to study the effects of two mammalian genes: FOXN3 and CHD2. The Forkhead Box (FOX) family of transcription factors shares a common DNA-binding domain and has been associated with organ development, differentiation, cell growth and proliferation, and cancer. Meanwhile, the CHD (Chromodomain helicase DNA binding protein) family of proteins is known to be involved in chromatin remodeling and regulation of gene expression. Phenotypic analysis of Foxn3 mutant animals revealed its indispensible role in craniofacial and embryonic development, embryonic lethality, expression of bone morphogenetic proteins, and spontaneous development of cancers in heterozygous and homozygous mutant mice. Preliminary evaluation of molecular mechanisms of FOXN3 signifies deregulation of cell-cycle checkpoint proteins Cyclin-B1 and CDK2 as the underlying etiology of tumors. Chd2 mutant mice exhibit spontaneous thymic and splenic lymphomas and reduced lifespan which can be restored through Chd2 re-expression in the thymus. At the molecular level, CHD2 deficiency reduces Puma (p53-upregulated modulator of apoptosis) induction after DNA damage in mouse thymocytes and HCT116 cells. Additionally, CHD2 is enriched at the Puma locus after DNA damage. CHD2-deficient cells also exhibit global reduction of active transcription markers H3K9-Acetylated and H4K8-Acetylated.

Heat shock proteins (HSPs) play important roles in the process of maintaining proteostasis in a cell. HSP72, the inducible form of the HSP70 family, is expressed in response to stress on the cell or tissue, including those stresses caused by tumor growth. Increasing evidence suggests that HSP72 is necessary for a cancerous cell to survive under the stresses of a tumor microenvironment. This has naturally raised interest in identifying an inhibitor selective for HSP72. The Haystead Laboratory at Duke University identified such a small-molecule inhibitor, referred to as HS-72, and proposed the scaffold as an ideal starting point to develop a family of therapeutic agents targeting HSP72. This work examines the potency and effectiveness of HS-72 and a number of its analogs developed by the Haystead Laboratory. These results suggest that HS-159 is a more effective inhibitor of HSP72 on a range of human tumor cell lines than HS-72. Further studies are needed to quantify how much more potent HS-159 is than HS-72 and potentially identify even more potent compounds.

Non-melanoma skin cancer (NMSC) is the most common malignant tumor, representing more than a third of all malignant tumors combined and the incidence is increasing every year. Ultraviolet (UV) radiation from the sun is the most dominant factor contributing to tumor initiation and progression. The condition is most prevalent in populations with lighter skin and older age. Current pharmaceutical molecular research targets the inhibition of the Epidermal Growth Factor Receptor (EGFR), a receptor which is commonly over-expressed or dysregulated in skin malignancies. This study evaluates the content and location of the damage marker p21 within keratinocytes that were incubated in sphingomyelin (SM) and later exposed to UV. Confocal microscopy and automated image processing provided the tools to assess large populations of keratinocytes in the effort to accurately identify the photoprotective qualities of sphingomyelin. Classification of individual cells into subpopulations yielded results suggesting SM may be involved in the inhibition of EGFR, and could potentially be a more naturally derived treatment.

No description available.

DNA replication is a highly complex part of cell metabolism that ensures safe propagation of the genome through tight regulation of the expression, localization, and activity of a large number of factors. Replication starts from distinct sites in the genome and initiation events are temporally ordered in a manner that is, on average, highly reproducible across cell populations. The specific order with which different parts of the genome are replicated has been proposed to be important to processes such as gene expression, cell differentiation, development, and genome evolution. Nevertheless, the fundamental mechanisms that are responsible for establishing these timing programs remain elusive. Unlike in higher eukaryotes, DNA replication in budding yeast initiates at sequence-specific loci called origins of replication. The timing of initiation at these loci is determined by the activation of the main replicative helicase Minichromosome Maintenance (MCM) complex. Recent results have placed MCM in a key role in establishing a replication timing program that is reproducible but arises from stochastic activation of origins, as has been observed in yeast and higher eukaryotes. One particular model posits that the loading of multiple MCMs at individual origins increases the chances that origins will be activated earlier in S phase by a limited amount of initiation factors. To further test this model, we set out to examine the consequences of modulating MCM levels in budding yeast in order to ascertain their effects on the dynamics of helicase loading during G1 and subsequent replication timing. Overexpression of MCM2-7 had no effects on cell viability, cell cycle progression, MCM abundance at origins, or replication timing. On the other hand, depletion of Mcm4, one of the six obligate components of the MCM helicase, caused reduced viability, slower progression through S phase, and increased sensitivity to replication stress. Importantly, Mcm4 depletion led to differential reduction in MCM loading at origins during G1, with low MCM origins being disproportionately affected by reduced MCM pools. Finally, reduced MCM loading at origins of replication led to delayed replication during S phase. Our data support a model where the loading activity of origins, controlled by their ability to recruit ORC and compete for MCMs, determines the number of helicases loaded, which in turn has strong implications for replication timing.

Genetic changes acquired in the pigment producing cells of the skin, called melanocytes, can lead to formation of the deadly cancer melanoma. Mutations or amplifications leading to the activation of the RAS/MAPK pathway occur in more than 90% of melanomas. Melanocyte development and survival requires the stimulation of this pathway by the receptor tyrosine kinase (RTK) KIT. In ~2% of melanomas, oncogenic KIT mutations drive tumor formation; however, the majority of melanomas lose wild-type KIT expression, suggesting that KIT could suppress melanoma formation. In human melanoma patients of The Cancer Genome Atlas (TCGA), we found an association between BRAFV600E mutations and low KIT mRNA expression, so we tested whether KIT loss would affect BRAFV600E-driven tumor onset by crossing a kit(lf) mutant allele into melanoma-prone Tg(mitfa:BRAFV600E); p53(lf) zebrafish. We observed that kit(lf)-mutant zebrafish experienced accelerated tumor onset and their tumors had increased RAS/MAPK pathway activation. In BRAFV600E-mutant melanoma cells, KIT activity reduced RAS/MAPK signaling by promoting activation of wild-type BRAF (BRAFWT). Furthermore, we found that overexpression of BRAFWT delayed tumor onset in Tg(mitfa:BRAFV600E); p53(lf); mitfa(lf) zebrafish, but had no effect in kit(lf); Tg(mitfa:BRAFV600E); p53(lf); mtifa(lf) zebrafish and a cohort of TCGA BRAFV600E-mutant melanoma patients with high KIT expression and high BRAFWT allele ratios experienced a reduced likelihood of metastasis and extended overall survival. These studies indicate that wild-type KIT acts to suppress melanoma formation through activation of BRAFWT, causing reduced signaling output of BRAFV600E-mutant cells.

Hepatocellular carcinoma (HCC) is an aggressive subtype of liver cancer with a poor prognosis. Currently, prognosis for HCC patients remains poor as few therapies are available. The clinical need for more effective HCC treatments remains unmet partially because HCC is genetically heterogeneous and HCC driver genes amenable to targeted therapy are largely unknown. Mutations in the TP53 gene are found in ~30% of HCC patients and confer poor prognosis to patients. Identifying genes whose depletion can inhibit HCC growth, and determining the mechanisms involved, will aid the development of targeted therapies for HCC patients. Therefore, the first half of this thesis focuses on identifying genes that are required for cell growth in HCC independent of p53 status. We performed a kinome-wide CRISPR screen to identify genes required for cell growth in three HCC cell lines: HepG2 (p53 wild-type), Huh7 (p53-mutant) and Hep3B (p53-null) cells. The kinome screen identified 31 genes that were required for cell growth in 3 HCC cell lines independent of TP53 status. Among the 31 genes, 8 genes were highly expressed in HCC compared to normal tissue and increased expression was associated with poor survival in HCC patients. We focused on TRRAP, a co-factor for histone acetyltransferases. TRRAP function has not been previously characterized in HCC. CRISPR/Cas9 mediated depletion of TRRAP reduced cell growth and colony formation in all three cell lines. Moreover, depletion of TRRAP reduced its histone acetyltransferase co-factors KAT2A and KAT5 at the protein level with no change at the mRNA level. I found that depletion of KAT5, but not KAT2A, reduced cell growth. Notably, inhibition of proteasome- and lysosome-mediated degradation failed to rescue protein levels of KAT2A and KAT5 in the absence of TRRAP. Moreover, tumor initiation in an HCC mouse model failed after CRISPR/Cas9 depletion of TRRAP due to clearance via macrophages and HCC cells depleted of TRRAP and KAT5 failed to grow as subcutaneous xenografts in vivo. RNA-seq and bioinformatic analysis of HCC patient samples revealed that TRRAP positively regulates expression of genes that are involved in mitotic progression. In HCC, this subset of genes is clinically relevant as they are overexpressed compared to normal tissue and high expression confers poor survival to patients. I identified TOP2A as one of the mitotic gene targets of the TRRAP/KAT5 complex whose inhibition greatly reduces proliferation of HCC cells. Given that this was the first time the TRRAP/KAT5 complex has been identified as a therapeutic target in HCC, the second half of this thesis focuses on identifying the mechanism via which depletion of this complex inhibits proliferation of HCC cells. I discovered that depletion of TRRAP, KAT5 and TOP2A reduced proliferation of HCC cells by inducing senescence. Typically, senescence is an irreversible state of cell cycle arrest at G1 that is due to activation of p53/p21 expression, phosphorylation of RB, and DNA damage. Surprisingly, induction of senescence after loss of TRRAP, KAT5 and TOP2A arrested cells during G2/M and senescence was independent of p53, p21, RB and DNA damage. In summary, this thesis identifies TRRAP as a potential oncogene in HCC. I identified a network of genes regulated by TRRAP and its-cofactor KAT5 that promote mitotic progression. Moreover, I demonstrated that disruption of TRRAP/KAT5 and its downstream target gene TOP2A result in senescence of HCC cells independent of p53 status. Taken together, this work suggests that targeting the TRRAP/KAT5 complex and its network of target genes is a potential therapeutic strategy for HCC patients.