To improve our understanding of LZAPs role in normal and cancer cell biology, the work in this thesis explored LZAP activities in development and carcinogenesis. This dissertation focused on understanding the role of LZAP in vertebrate development because no LZAP-/- pups were observed after crossing LZAP+/- mice. In zebrafish, we discovered that LZAP loss results in early embryonic lethality at least partially by inhibiting normal cell cycle progression while potentiating apoptosis. LZAP morphants do not initiate epiboly, the earliest developmental cell movement. Together these findings suggest that LZAP may be a critical regulator of embryonic stem cell function(s) required for their survival. Examination of mice after targeting LZAP led to the unexpected finding that p53 levels are decreased in heterozygous mice embryonic stem cells and tissues. In vitro experiments confirmed that LZAP depletion decreases both wild-type and mutant p53 protein levels by decreasing both p53 protein half-life and p53 mRNA translation. Mechanistically, loss of LZAP is associated with increased expression of nucleolin, a known p53 inhibitor, and depletion of nucleolin partially abrogates the effect of LZAP loss. Consistent with downregulation of p53, LZAP loss is associated with resistance to radiation in cells expressing wild-type p53. Remarkably, decreased mutant p53 levels are observed following LZAP depletion, which sensitize cancer cells to radiation. These data suggest that LZAP may be an ideal target for treating p53 mutant tumors, because temporary inhibition of LZAP activity toward p53 could simultaneously sensitize the tumor to DNA damaging agents (chemotherapy or radiation) while protect normal surrounding tissue. To determine if loss of LZAP is biologically significant, primary HNSCCs were examined revealing that LZAP and p53 levels significantly correlate and that cancers with downregulated LZAP have reduced pressure to inactivate p53 via mutation. This work will impact several areas of cancer research in both the relatively new LZAP field and the persistently important p53 field. In summary, these data suggest that LZAP is critical for early embryogenesis in zebrafish and mice, and that loss of LZAP is a new mechanism of p53 loss and a driver of HNSCC tumorigenesis.

In vertebrate epithelia, p120-catenin mediates E-cadherin stability and suppression of RhoA. Genetic ablation of p120 in various epithelial tissues typically causes striking alterations in tissue function and morphology. Although these effects could very well involve p120s activity towards Rho, ascertaining the impact of this relationship has been complicated by the fact that p120 is also required for cell-cell adhesion. Here, we have molecularly uncoupled p120s cadherin stabilizing- and RhoA-suppressing activity. Unexpectedly, removing p120s Rho-suppressing activity dramatically disrupted the integrity of the apical surface, irrespective of E-cadherin stability. The physical defect was tracked to excessive actomyosin contractility along the vertical axis of lateral membranes. Thus, we suggest that p120s distinct activities toward E-cadherin and Rho are molecularly and functionally coupled, and this in turn enables the maintenance of cell shape in the larger context of an epithelial monolayer. Importantly, local suppression of contractility by cadherin-bound p120 appears to go beyond regulating cell shape, as loss of this activity also leads to major defects in epithelial lumenogenesis.

Non-Hodgkin lymphomas often arise at sites of chronic inflammation, exposing them to oxidative stress, or increased levels of reactive oxygen species (ROS). Increases in ROS are associated with tumor initiation, promotion and progression. Chronic exposure to ROS may promote the transformation of lymphocytes to lymphoma. Anti-apoptotic proteins such as Bcl-2 are commonly overexpressed in lymphoid malignancies. The protective function of Bcl-2 is partially dependent on its ability to regulate the redox environment. Adaptation to oxidative stress via the upregulation of anti-oxidant defense enzymes or upregulation of anti-apoptotic proteins appear to, in part, confer resistance to chemotherapeutics through their ability to regulate the redox environment. This suggests that using an agent to disrupt redox homeostasis has potential as a therapeutic strategy to circumvent these resistance mechanisms. The following studies examine the use of the copper chelator drug, ATN-224, to modulate the redox environment and circumvent the upregulation of anti-oxidant defense enzymes and anti-apoptotic proteins. These studies demonstrate that ATN-224 inhibits the primary anti-oxidant defense enzyme SOD1 and the redox-driven proton pump CcOX (complex IV in the electron transport chain). This inhibits the ability of SOD1 and CcOX to regulate the cellular and/or mitochondrial redox environment, respectively. ATN-224 treatment increases oxidative stress and induces peroxynitrite-dependent cell death. Furthermore, ATN-224 induces the release of AIF from the mitochondria, resulting in caspase-independent cell death. Collectively, these findings suggest that modulating the redox environment with ATN-224 has therapeutic potential in the treatment of non-Hodgkin lymphoma.

Despite considerable progress in conventional cancer therapies, major challenges persist in the treatment of patients with advanced stage malignancies. Cancer immunotherapy (harnessing the immune system against tumors) has demonstrated limited success to date, partially due to the immunosuppressive environment generated by tumors. The mechanisms of cancer-induced immune suppression are multiple and include the promotion of immunosuppressive cells such as regulatory T cells (Treg) and myeloid-derived suppressor cells (MDSC). MDSC expand in tumor-bearing hosts and play a central role in cancer immune evasion by inhibiting adaptive and innate immunity. Different approaches have been explored to negatively impact MDSC, each associated with specific pitfalls. In this study, we demonstrated that the anthracycline doxorubicin selectively eliminates MDSC in the spleen, blood and tumor beds. Furthermore, five days after doxorubicin treatment residual MDSC exhibited impaired suppressive function, which correlated with reduced reactive oxygen species (ROS) production, and down-regulation of arginase-1 and indoleamine 2,3-dioxygenase (IDO) expression. Of therapeutic relevance, the frequency of effector lymphocytes (CD4⁺ and CD8⁺ T cells) or natural killer cells (NK) to suppressive MDSC ratios was significantly increased following doxorubicin treatment of tumor-bearing mice. Importantly, the proportion of natural killer (NK) and cytotoxic T cells (CTL) expressing perforin and granzyme B and of CTL producing IFNγ was augmented following doxorubicin administration. The mechanism of doxorubicin-mediated elimination of MDSC was partly mediated by the increase of ROS production in MDSC at earlier time points after doxorubicin treatment. Consistently, MDSC isolated from gp91-/- mice were less sensitive to doxorubicin in vitro, and doxorubicin effects on MDSC in gp91-/- tumor-bearing mice were reduced. Of clinical significance, this drug efficiently combined with Th1 or Th17 lymphocytes to suppress tumor development and metastatic disease, resulting in better overall survival. MDSC isolated from patients with different types of cancer were also sensitive to doxorubicin-mediated cytotoxicity in vitro. Our results therefore indicate that doxorubicin may be used not only as a direct cytotoxic drug against tumor cells, but also as a potent immunomodulatory agent that selectively impairs MDSC-induced immunosuppression, thereby fostering the efficacy of T cell-based immunotherapy.

Cancer is a disease that is characterized by genetic mutations that occur leading to an increased propensity for abnormal cellular growth. Chromosome 1p36.3 is lost or aberrant in several tumor types, including neuroblastoma and melanoma. In order to understand the functional significance of chromosome 1p36.3 loss in tumors, our research has been focused on elucidating the function of cyclin dependent kinase 11 (CDK11), a gene that is located within this chromosomal region. CDK11 is a serine/threonine kinase and is a member of the p34cdc2-related family of kinases. CDK11 has multiple isoforms that are involved RNA processing, apoptosis and the cell cycle.CDK11 proteins are encoded by two different genes (Cdc2L1 and Cdc2L2), which are highly homologous to each other and are evolutionarily conserved. Our studies originated from investigating the regulation of the expression of CDK11 genes by isolating and characterizing the Cdc2L1 promoter region. We identified the basal promoter region and found that Cdc2L1 is regulated by transcription factors Ets-1, Skn-1 and E2F-1. Conversely, Cdc2L2 was shown to be regulated by TCF11/Nrf2, Ets-1 and CREB, displaying differential regulation of the two different CDK11 genes.CDK11p58 is generated during the G2/M phase of the cell cycle because of an internal ribosome entry site found within the full-length CDK11 transcript. The generation of CDK11p58 during the cell cycle had previously been shown to be a regulatory event; however the actual function of CDK11p58 remained unknown. Our research focused on determining potential interaction partners of CDK11p58 and elucidating the role of these interactions in the regulation of the cell cycle. Utilizing proteomic technology, we were able to find multiple potential interacting proteins with CDK11p58, including a RNA-dependent RNA helicase protein, DDX15. DDX15 is phosphorylated by CDK11p58 and both proteins are directly involved in cell cycle regulation. Additionally, CDK11p58 was found to associate with several cytoskeletal proteins, alpha- and beta-tubulin, lamin A, and vimentin, and heat shock proteins, Hsp70 and Hsp90. These investigations are only preliminary but provide a novel insight into the possible function of CDK11p58 during mitosis.

The adult mammalian brain hosts two regions of quiescent neural stem cells that continually generate new neurons throughout life. One of these regions, the subventricular zone (SVZ), primarily produces immature neuroblasts that migrate and integrate into the olfactory bulb during neurogenesis. Stem cells in the SVZ are of interest because of their potential utility in brain repair. Strikingly, stem cell location within this germinal region determines the type of neuronal progeny that a given stem cell generates, and stem cells retain this positional identity upon culture or transplantation. However, the molecular mechanisms that regulate fate specification and proliferation in neural stem cells remain largely unknown, and it is not yet clear whether the intrinsic properties of particular stem cells affect their response to oncogenic events or damage. My lab has shown that active sonic hedgehog (Shh) signaling occurs in the ventral SVZ and is necessary and sufficient for the regulation of neuronal fate in a specific group of ventral stem cells. The dorsal region of the SVZ does not have high shh pathway activation; despite expression of most pathway components, these cells appear to be resistant to stimulation by Shh ligand. I hypothesize that Shh pathway signaling is actively repressed in the dorsal SVZ through expression of the Gli3 transcription factor, and that this repression is necessary to generate dorsal neural progeny. My thesis seeks to determine which downstream components of the shh pathway control proliferation and fate specification in dorsal stem cells.

Perturbation of physiological processes stimulate a cascade of events resulting in adverse pathogical conditions, the most predominant being cancer. Colorectal cancer (CRC) is the third most common cancer in the United States and the third leading cause of cancer deaths among men and women. CRC is most commonly initiated by mutation of Adenomatous polyposis coli (APC), a tumor suppressor gene of the Wingless/Wnt pathway, which leads to constitutive activation of Wnt signaling responsible for the majority of CRC cases. Although significant progress has been made in understanding the sequential genetic events leading to cancer development, the precise genes and associated molecular events underlying tumor progression are still poorly understood. Our lab has previously reported a correlation between increased expression of claudin-1, a tight junction protein, in primary and metastatic tumor samples of CRC patients. Additionally, in adenomas of the APCMin mouse, an experimental model for CRC, and human cell lines harboring mutations in APC, claudin-1 expression is highly increased and mislocalized. Taken together, these studies suggest existence of a potential interplay between dysregulated claudin-1 expression and APC, thus supporting an essential role in the regulation of sporadic colonic neoplasia. In addition to genetic stimuli, CRC can be initiated by an inflammatory stimulus. Patients with inflammatory bowel disease, a chronic disorder of the gastrointestinal tract, are at high risk for developing CRC. This may be attributed, in part, to repeated injury and repair, which induces transformation of the colonic epithelium. As such, increased Claudin-1 expression is also reported to be associated with active IBD and colitis-associated cancer, yet its distinct role in inflammation has yet to be elucidated. This study investigates the specific contribution of increased claudin-1 expression to inflammation and CRC. These results identify a role for Claudin-1 in epithelial differentiation through a novel interaction with Notch signaling. Additionally, claudin-1 enhances tumorigenesis through upregulation of inflammation, IL-23/IL17 signaling, when combined with mutant APC. Once understood how overexpression claudin-1 contributes to tumor progression, we can better investigate its potential as a target for therapy.

Dissertation under the direction of Professor Harold L. Moses Transforming Growth Factor Beta (TGF-β) is acts as both a tumor suppressor and promoter in the context of epithelial tumor progression. In epithelial cells, active TGF-β signaling promotes cell cycle arrest to slow tumor growth but also promotes EMT and cell motility to enhance tumor cell metastasis. We sought to address TGF-β signaling role in tumor progression through its effects on stromal cells and microenvironmental changes. Interestingly, as with epithelial TGF- β signaling, we found active TGF-β signaling in fibroblasts to present with the same dichotomous relationship. We found that TGF-β can promote fibroblast activation which in turn drives the expression of numerous matrix remodeling genes. The phenotypic outcome of this matrix remodeling promotes tumor cell metastasis. Alternatively, abrogation of TGF-β signaling in fibroblasts also promotes tumor cell metastasis. Albeit, this occurs through a different mechanism involving altered chemokine expression to promote chemotaxis of tumor promoting myeloid cells. Gene expression analysis of stroma from reduction mammoplasty as well as breast cancer patients suggests that TGFβR2 is lost as breast cancer progresses from normal mammary tissue to invasive ductal carcinoma. Importantly, this loss of TGFβR2 in the stroma of breast cancer patients is associated with poor patient outcome and rick of recurrence. This work shows the importance of TGF-β signaling as a central player in numerous biological process which effects tumor progression through the tumor microenvironment. As this signaling axis in the tumor microenvironment is certainly important in the progression of human breast cancer, the therapeutic targeting of the TGF-β pathway will need to consider stromal effects for efficacious use.

Inactivation of STK11/LKB1 is one of the most common genetic events in lung cancer, and understanding the cellular phenotypes and molecular pathways altered as a consequence will aid the development of therapeutic strategies targeting LKB1-deficient cancers. We report the comprehensive analysis of gene and protein expression patterns associated with LKB1 loss in lung adenocarcinomas, through which we identify hallmarks of altered tumor metabolism and downregulation of the PI3K/AKT pathway. Significant differences are observed between human tumors and those derived from a genetically engineered mouse model of LKB1 loss. A 16-gene signature is predictive of both mutational and non-mutational LKB1 loss in human tumors. Cell lines expressing this signature show increased sensitivity to MEK inhibition, independent of mutations in RAS and RAF family members. Restoration of LKB1 in lung cancer cell lines downregulates the gene expression pattern, attenuates FOXO3, and induces resistance to MEK inhibition. These findings identify characteristic phenotypic features of LKB1-deficient tumors and identify LKB1 loss as a novel determinant of MEK sensitivity.

E-cadherin is widely recognized as a tumor- and/or metastasis suppressor, with its activity as a cell-cell adhesion receptor is dependent on tightly coupled interactions with cytoplasmic cofactors p120-, α-, and β-catenins. However, how the catenins, particularly p120, influence tumor initiation and metastasis is not clearly established. Here, we use conditional p120 ablation along with the Apc1638 and ApcMin mouse models to directly examine the role of p120 in intestinal tumorigenesis. Surprisingly, limited p120 knockout in less than 10% of the intestinal epithelium increased tumor development by up to 10-fold. The tumors, however, did not exhibit the expected complete loss of p120 and were found instead to be monoallelic, suggesting that p120 functions as a haploinsufficient tumor suppressor. Indeed, further studies show loss of one p120 allele is sufficient to induce the changes observed in total tumor number and that these results were not due to interaction with p120s nuclear binding partner, Kaiso. Surprisingly, retention of the remaining p120 allele was evidently obligatory in adenomatous cells, as complete loss of p120 was selectively incompatible with tumor cell viability. To better appreciate the impact of p120 haploinsufficiency, we turned to a Sleeping Beauty mutagenesis system that enables semi-quantitative analyses of relative tumorigenic potency through mutation frequency. Surprisingly, p120 and α-catenin are among the highest ranked genes in this system, with E-cadherin close behind. Remarkably, p120, α-catenin, and E-cadherin all ranked within the top 4% of 919 identified mutations in the Apc-sensitized intestine. As loss of any of these three components destabilized the cadherin complex, monoallelic disruption at any of these loci likely impacts functionality and accelerates tumorigenesis. Thus, although E-cadherin is widely recognized as a metastasis suppressor in colorectal cancer, our data show that the integrity of the cadherin complex is critical at the earliest stages of tumor initiation and progression.