Cancer remains one of the leading causes of death worldwide due to late diagnosis and ineffective treatment options. To address this problem requires the elucidation of the molecular mechanisms, including the signaling pathways and transcription factors that drive cancer initiation and progression. In this regard, our laboratory has been particularly interested in the embryonically important T - box family of transcription factors which has been heavily implicated in promoting initiation and progression of a long list of cancers. For example, the overexpression of the T - box factor TBX3, has been reported to function in promoting immortalization, migration, invasion and tumour formation in a number of epithelial - derived malignancies. Furthermore, our laboratory recently reported that TB X3 is also overexpressed in a wide range of sarcoma subtypes including rhabdomyosarcomas. This suggests that TBX3 may also contribute to the development and/or progression of sarcomas and potentially may serve as a biomarker for their diagnosis and targete d therapy. This is exciting because sarcomas are diverse and heterogeneous cancers with varying clinical behaviours, high rates of metastasis and recurrence and are notoriously resistant to current chemotherapies. However, whether TBX3 is a molecular drive r of these mesenchymal - derived cancers remains to be determined. This project therefore aimed to elucidate the role of TBX3 overexpression in embryonal rhabdomyosarcomas (ERMS) which is the most common soft tissue sarcoma in children and adolescents. To this end, ERMS cell culture models were established in which TBX3 was either stably knocked down or stably overexpressed and the resulting cells were tested for several features of the cancer phenotype using in vitro and in vivo experiments. The results show that TBX3 promotes cell proliferation, anchorage independent growth and cell migration in vitro and tumour formation and invasion in vivo. This study also provides evidence that nucleolin binds to, and co - operates with, TBX3 to promote proliferation and migration of ERMS cells. Furthermore, data from initial experiments reveal that Hsc70 interacts with TBX3, to possibly increase its protein stability, and that oncogenic c - Myc and AKT 1 positively regulat e TBX3 levels in ERMS. This, albeit preliminary data, suggest that Hsc70, c - Myc and AKT1 are responsible, in part, for the overexpression of TBX3 in ERMS. Together findings from this study implicate TBX3 as an oncogene in ERMS and suggest that TBX3, nucleolin, Hsc70, c - Myc and AKT may be used in combination as biomarkers for the diagnosis and targeted therapy of ERMS.

The T-box transcription factor TBX3, plays critical roles in development including the formation of the limbs, heart and mammary glands. While haploinsufficiency of TBX3 results in ulnar mammary syndrome, its overexpression is linked to several cancers. We and others have shown that TBX3 drives tumour formation, invasion and metastasis of several sarcoma subtypes as well as melanoma, cervical cancer and breast cancer. TBX3 has thus been proposed as a novel therapeutic target to treat these cancers. Direct targeting of transcription factors for therapies however continues to represent a serious challenge and therefore an understanding of the molecular mechanisms that regulate and mediate its oncogenic activity may reveal more amenable anti-cancer drug targets. This project therefore aimed to (1) identify signalling molecules that upregulate TBX3 expression in MCF-7 breast cancer cells as well as (2) identify and characterize protein partners that cooperate with TBX3 to drive its oncogenic functions in these cells. The overexpression of the basic helix-loop-helix oncogenic transcription factor c-Myc has been widely reported in breast cancer progression and c-Myc-driven pathways are elevated in aggressive drug resistant breast cancer cells and tumours. Our laboratory has previously shown that c-Myc directly binds and activates the TBX3 promoter in several sarcoma subtypes, and it was hypothesised that c-Myc may also activate TBX3 in breast cancer. To investigate this, the impact of transiently knocking down c-Myc on TBX3 mRNA and protein levels was firstly assessed by qRT-PCR and western blotting respectively. Results show that when c-Myc is depleted, TBX3 mRNA and protein levels decrease, suggesting that c-Myc may be transcriptionally upregulating TBX3. To confirm this, c-Myc was ectopically overexpressed in MCF-7 breast cancer cells in the presence or absence of Actinomycin D, an inhibitor of de novo transcription, and TBX3 mRNA and protein levels were measured by qRT-PCR and western blotting respectively. Indeed, results show that when de novo transcription is inhibited, the c-Myc mediated activation of TBX3 expression is abolished. To identify and characterize TBX3 protein partners, MCF-7 breast cancer cells that stably overexpress FLAG-TBX3 were firstly established to enable effective immunoprecipitation for mass spectrometry. The overexpression of FLAG-TBX3 was confirmed by western blotting and immunocytochemistry and the anti-proliferative and pro-migratory roles of TBX3 overexpression in breast cancer cells was confirmed using growth curves and scratch motility assays respectively. Through affinity purifications coupled with mass spectrometry a myriad of putative TBX3 protein co-factors were identified and from this list three partners viz nucleolin, Hsc70 and HnRNP K were validated by immunoprecipitation and colocalization experiments. Importantly, results show that the interaction of TBX3 with Hsc70 is required for TBX3 protein stability and that nucleolin and TBX3 cooperate to promote MCF-7 breast cancer cell migration. Furthermore, treatment of MCF-7 cells with the nucleolin targeting aptamer, AS1411, mis localizes TBX3 and nucleolin to the cytoplasm and causes a reduction in cell viability while having no effect on the viability of normal skin fibroblasts. Together the results from this study show that c-Myc/TBX3/nucleolin/Hsc70 may be an important oncogenic pathway in breast cancer and that AS1411 may be a potentially important aptamer for the treatment of TBX3-driven breast cancer.

A-kinase anchoring proteins (AKAPs) are signaling scaffolds which provide spatial and temporal organization of signaling pathways in discrete subcellular compartments. Through tethering the cyclic-AMP dependent protein kinase A (PKA), AKAPs target PKA activity to distinct regions in the cell, bringing PKA in close proximity to its target proteins. This provides a high level of specificity and regulation of PKA and its role in mediating a number of biological processes, one of which is cell migration. Cell migration is a highly dynamic and fundamental process, when misregulated can lead to a number of pathologies. The process of cell migration requires integration and coordination of actin cytoskeletal dynamics, adhesion turnover, and contractility. The important role of PKA in regulating the cellular processes involved in cell migration has been extensively studied. Our lab has shown that PKA activity and spatial distribution through AKAPs are localized to the leading edge of migrating cells and are required for effective cell migration, yet the specific AKAPs responsible remain unknown. Traditional methods for identifying AKAPs suffer from a number of limitations. Therefore the objective of the enclosed work is to establish and characterize a novel approach for the identification of cytoskeletal and adhesion-associated AKAPs. We show for the first time, an in vitro approach to identify cytoskeletal AKAPs which may be responsible for localizing PKA to the leading edge of migrating cells.

Cyclic AMP is an intracellular messenger that regulates numerous processes in various types of cells. In pancreatic β-cells, cAMP potentiates the secretion of insulin by promoting Ca2+ signals and by amplifying Ca2+-triggered exocytosis. Whereas Ca2+ signals have been extensively characterized, little is known about the kinetics of cAMP signals. To enable measurements of the cAMP concentration beneath the plasma membrane ([cAMP]pm) of individual cells, a translocation biosensor was created based on fluorescent-protein-tagged subunits of protein kinase A (PKA). Evanescent wave microscopy imaging of biosensor-expressing clonal β-cells revealed that the insulinotropic hormones glucagon and GLP-1 triggered pronounced oscillations in [cAMP]pm. Simultaneous measurements of the intracellular Ca2+ concentration showed that cAMP and Ca2+ oscillations were synchronized and interdependent. [cAMP]pm oscillations were also triggered in clonal and primary mouse β-cells by an elevation of the glucose concentration from 3 to 11 mM. These oscillations were preceded and enhanced by elevations of Ca2+. However, conditions raising cytoplasmic ATP could trigger cAMP elevations also without accompanying Ca2+ changes, indicating that adenylyl cyclase activity may be directly controlled by the substrate concentration. Experiments with 3-isobutylmethylxanthine (IBMX) and various family-selective phosphodiesterase (PDE) inhibitors indicated that [cAMP]pm oscillations are generated by periodic formation of the messenger by adenylyl cyclases. PDE1 and PDE3 as well as IBMX-insensitive mechanisms shape [cAMP]pm, but no single PDE isoform was required for glucose generation of [cAMP]pm oscillations. Recordings of single-cell insulin secretion kinetics with a fluorescent biosensor that reports formation of the phospholipid PIP3 in the plasma membrane in response to autocrine insulin receptor activation showed that [cAMP]pm oscillations were paralleled by pulsatile insulin release. Whereas adenylyl cyclase inhibition suppressed both [cAMP]pm oscillations and pulsatile insulin release, elevation of [cAMP]pm enhanced secretion. Investigation of the effects of different temporal patterns of [cAMP]pm showed that brief [cAMP]pm elevation is sufficient to trigger cytoplasmic responses, whereas sustained elevation is required to induce translocation of the PKA catalytic subunit into the nucleus. In conclusion, these studies demonstrate for the first time in mammalian cells that [cAMP]pm oscillates in response to physiological stimuli. The glucose-induced [cAMP]pm oscillations are generated by periodic cAMP production mediated by interplay between ATP and Ca2+ in the sub-membrane space, and may contribute to both triggering and amplifying pathways of insulin secretion. Apart from regulating the precise kinetics of insulin exocytosis, temporal encoding of cAMP signals might constitute a basis for differential regulation of downstream cellular targets.

Background and aims: Type 2 diabetes mellitus results from complex interplay between genetic predisposition and environmental factors that together promote impairment of insulin-producing β-cells. Elevated levels of glucose, fatty acid palmitate and apolipoprotein CIII (apoCIII) are implicated in this process. To delineate effects of these factors, the role of enhanced carnitine palmitoyltransferase 1 (CPT1) expression in glucolipotoxic cells, glucose-dependency of the unfolded protein response (UPR) in palmitate-induced apoptosis and activation of mitogen activated protein kinases (MAPKs) in apoCIII-induced apoptosis were evaluated. In addition, protein profiles of β-cell exposed to elevated levels of glucose or palmitate were generated to identify proteins regulated by these nutrients. Methodology: INS-1E cells were cultured at different glucose concentrations in the absence or presence of palmitate or apoCIII for up to 48 hours. CPT1 was over-expressed with a Tet-ON regulated adenovirus. In cells exposed to apoCIII, inhibitors of MAPKs p38 or ERK1/2 were included during culture. After culture, apoptosis, insulin secretion, expression of UPR-markers and MAPKs and protein profiles were determined. Results: INS-1E cells exposed to elevated levels of glucose and palmitate showed deranged insulin secretion with increased insulin secretion at non-stimulatory glucose level, enhanced apoptosis and induced expression of UPR-markers. Over-expression of CPT1 reduced basal insulin secretion and attenuated apoptosis. Palmitate-induced apoptosis was accentuated by increasing the culture glucose concentration. Markers of UPR were not modulated by the glucose concentration in INS-1E cell exposed to palmitate, however. ApoCIII-induced apoptosis in INS-1E cells was accompanied by activation of p38 and ERK1/2. Protein profiling of INS-1E cells exposed to elevated levels of glucose or palmitate revealed changes in expression of multiple β-cell proteins implicated in glucose metabolism, defence against reactive oxygen species, protein translation/folding/degradation and insulin granular trafficking. Conclusions: Over-expression of CPT1 counteracts β-cell glucolipotoxicity. Activation of UPR is not a major determinant for palmitate-induced β-cell apoptosis. ApoCIII-induced β-cell apoptosis involves activation of MAPKs. The identified differentially expressed proteins indicate a central role of altered glucose metabolism and protein synthesis in gluco- and lipotoxic β-cells and may provide specific molecular mechanisms offering new ways of treating the disease.

The adaptor protein Shb and the non-receptor tyrosine kinase Frk have been implicated in intracellular signalling in insulin-producing beta cells. In this thesis, knockout mice are used to further elucidate the role of Shb and Frk for beta cell number, cytokine-induced cell death, and glucose homeostasis. In addition, the effect of Shb deficiency upon tumour growth is studied in a mouse model of endogenous tumourigenesis. Previously, overexpression of Frk has been associated with increased beta cell replication, and increased susceptibility to cytokine induced beta cell destruction. To test whether Frk has a non-redundant role in regulating beta cell mass, beta cell number in Frk-/- mice was assessed at different stages of life. The results showed that Frk is involved in regulating beta cell number during embryonal and early postnatal life, but is probably redundant in the adult. An earlier study had suggested that Shb participates in cytokine-induced beta cell death, a model of autoimmune diabetes. To test this further, Shb-/- islets were exposed to cytokines, or to an ER-stress inducing agent. Shb knockout islets exhibited decreased cell death, and this effect appeared to be independent of NO, JNK, p38 MAP kinase, FAK and c-Abl, but may involve an augmented induction of Hsp70. Furthermore, glucose homeostasis in Shb-/- mice was impaired, with elevated basal blood sugar concentration and reduced glucose-induced insulin secretion. Previously Shb deficient mice had showed an impaired ability to sustain growth of implanted tumour cells, due to reduced angiogenesis. Tumour growth and angiogenesis were here assessed in an inheritable tumour model. Shb deficient mice exhibited fewer tumours, and reduced vessel density in small tumours, indicating impaired angiogenesis. However, a few large tumours developed in Shb-/- mice, suggesting that tumours can escape the angiogenic restriction caused by the absence of Shb.

Type 1 diabetes mellitus is a severe life-long disease with a pronounced risk of developing secondary complications. One way to avoid the latter is to restore the fine tuning of blood glucose homeostasis by transplantation of pancreatic islets. However, isolated islets need to be properly engrafted and to re-establish a vascular network in order to regain function. Earlier studies have shown that pancreatic islets experimentally transplanted to e.g. the liver or the kidney become poorly revascularized. In the present thesis, mice deficient of the angiostatic factor thrombospondin-1 (TSP-1) were found to have an impaired beta-cell function. Development of this beta-cell dysfunction was prevented by treatment of TSP-1 deficient mice from birth with the TGFbeta-1 activating sequence of TSP-1. TSP-1 in islets was predominantly expressed in the endothelial cells. Isolated islet endothelial cells was observed to have a low proliferatory and migratory capacity towards angiogenic stimuli, but this could be reversed by neutralizing antibodies to the angiostatic factors alpha1-antitrypsin, endostatin or TSP-1. Transient downregulation of TSP-1 expression in mouse islet cells prior to transplantation improved graft revascularization, blood perfusion, oxygenation and function when evaluated one-month post-transplantation. The same result was achieved when islets or recipients of islets were pre-treated with the hormone prolactin one-month post-transplantation. The present study illustrates the importance of the angiostatic factor TSP-1 for islet beta-cell function and engraftment of islets following transplantation. Interference with TSP-1 can possibly be used to improve the results of clinical islet transplantation.

Although the reason for insufficient release of insulin in diabetes mellitus may vary depending on the type and stage of the disease, it is of vital importance that an amplified insulin biosynthesis can meet the increased need during periods of hyperglycemia. The insulin mRNA is highly abundant in beta cells and changes in insulin mRNA levels are, at least in part, controlled by altered rates of mRNA degradation. Since the mechanisms behind the control of insulin messenger stability and translation are still largely obscure, the work presented in this thesis therefore aimed to further investigate the role of insulin mRNA binding proteins in the control of insulin mRNA break-down and utilization for insulin biosynthesis. To clarify how glucose regulates insulin mRNA stability and translation we studied the correlation between polypyrimidine tract binding protein (PTB) gene expression and insulin mRNA levels. It was found that an increase in PTB mRNA and protein levels is paralleled by an increase in insulin mRNA levels. It was also found that PTB binds to the 5’-untranslated region of the insulin mRNA and that insulin mRNA can be translated through a cap-independent mechanism in human islets of Langerhans, possibly due to the interaction with PTB. Further it was discovered that the suppressed insulin biosynthesis in human islets during glucotoxicity is partly due to an induction of the microRNA miR-133a. This induction leads to decreased levels of PTB and insulin biosynthesis rates in human islets. Finally, we were able to identify two proteins, hnRNP U and TIAR, that bind specifically to the insulin mRNA in vitro, and show that the stability and translation of insulin mRNA is oppositely affected by these proteins. In conclusion, insulin producing cells seem to be able to regulate insulin messenger stability and translation by a control mechanism in which the binding of specific proteins to the insulin messenger dictates the outcome. A better understanding of the events leading to insulin production may in the future aid in optimal diagnosis and treatment of type 2 diabetes.

<p>The transcription factor interferon (IFN) regulatory factor 3 (IRF-3) plays a central role in the innate immune response to viral stimulation. IRF-3 participates in both the type I IFN-dependent and -independent signalling pathways that result in the induction of an antiviral state. The work presented in this thesis characterizes the central role of IRF-3 in the IFN-independent response to virus particle entry. In addition, novel splice variants of human IRF-3 are identified and characterized, implying a role for splice-mediated regulation of IRF-3-mediated antiviral signalling. Finally, a role for reactive oxygen species in the activation of IRF-3 following virus particle entry is described, with virus particle entry inducing danger associated molecular patterns associated with IRF-3 activation and IFN-independent antiviral gene expression. Taken together, this thesis characterizes the role of IRF-3 in the innate antiviral signalling pathways activated following viral stimulation, and highlight the importance of danger-associated molecular patterns as important mediators of antiviral signalling.</p> / Doctor of Philosophy (PhD)

<p>Epithelial-mesenchymal transition (EMT) is a pathological process leading to the formation of anterior subcapsular cataract (ASC). Mediated by transforming growth factor beta (TGFβ), EMT involves the transformation of the monolayer of lens epithelial cells (LECs) into spindle-shaped myofibroblasts, which manifest as plaques directly beneath the lens capsule. TGFβ-induced EMT leading to ASC has been associated with the upregulation of two specific matrix metalloproteinases (MMPs), MMP-2 and MMP-9. Having identified MMP-2 and MMP-9 as participants in the formation of cataracts, the specific roles of either of these MMPs have yet to be determined.</p> <p>The current study utilized MMP-2 and -9 knockout (KO) mice to determine their unique roles in TGFβ-induced EMT. First, adenoviral injection of active TGFβ1 into the anterior chamber of MMP-2 KO mice led to the formation of distinct αSMA-positive anterior subcapsular plaques, in contrast to treated MMP-9 KO eyes, which were resistant. Additionally, an <em>ex vivo </em>mouse LEC explant system was established in these KO lines. In the isolated lens epithelial explants, TGFβ triggered a transformation of LECs from a tightly packed cuboidal monolayer to an elongated mesenchymal phenotype. This process involved a disruption in epithelial cell contacts indicated by a loss of E-cadherin, and an acquisition of myofibroblast marker, αSMA. In the absence of MMP-2, TGFβ was still able to induce EMT with E-cadherin loss and concurrent αSMA expression. In contrast, LEC explants from MMP-9 KO mice treated with TGFβ did not acquire a characteristic spindle-like phenotype and showed substantially less αSMA expression. Results from both of these approaches were consistent; MMP-2, but not MMP-9, KO mice stimulated with TGFβ exhibited phenotypic changes typical of those described in ASC formation, namely a loss in cell attachments, multilayering of previously epithelial-like cells, and αSMA reactivity. Therefore, while MMP-2 is not necessary, MMP-9 is critical to TGFβ-induced EMT in LECs.</p> / Master of Science (MSc)