Bibliography: leaves 144-163 / xi, 163 leaves : / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (M.D.)--University of Adelaide, 1985

<p>Antitumour properties of the cyanoguanidine CHS 828 and analogues were discovered in 1997. CHS 828 is presently in clinical phase I/II trials. This thesis encompasses in vitro studies of the kinetics and mode of cell death induced in the human cell line U-937 GTB, by CHS 828 and the standard antitumour drug etoposide.</p><p>Etoposide induces apoptosis in U-937 GTB within 4 h. The cells exhibited apoptotic morphology, including condensed and fragmented nuclei and formation of apoptotic bodies, activation of caspase 3 and 8, and DNA fragmentation, visualised by TdT-mediated dUTP nick end-labelling (TUNEL).</p><p>CHS 828 induced few and weak signs of apoptosis. Metabolic activity was the only parameter affected during the first 24 h of exposure. After ~30 h, proliferation (DNA synthesis) and protein synthesis ceased, and viability started to decrease towards 10% at 72 h. Morphology and ultrastructure of dying/dead cells showed predominant necrosis. The decrease in viability was postponed by protein synthesis inhibition or maintenance of ATP levels by 3-aminobenzamide. In addition, 3-aminobenzamide switched morphology towards apoptosis. </p><p>Continuous co-exposure to CHS 828 and etoposide resulted in impressive cell kill synergy in U-937 GTB cells at effect levels of 30-70%. Pre-exposure to CHS 828 for 18 h or more, on the other hand, resulted in diminished cell kill and inability to activate the apoptotic machinery upon etoposide stimulation, evaluated by morphology and caspase activity.</p><p>In summary, CHS 828 induced cell death is predominantly non-apoptotic, does not involve caspases and can be postponed by maintained protein synthesis and ATP levels.</p>

Antitumour properties of the cyanoguanidine CHS 828 and analogues were discovered in 1997. CHS 828 is presently in clinical phase I/II trials. This thesis encompasses in vitro studies of the kinetics and mode of cell death induced in the human cell line U-937 GTB, by CHS 828 and the standard antitumour drug etoposide. Etoposide induces apoptosis in U-937 GTB within 4 h. The cells exhibited apoptotic morphology, including condensed and fragmented nuclei and formation of apoptotic bodies, activation of caspase 3 and 8, and DNA fragmentation, visualised by TdT-mediated dUTP nick end-labelling (TUNEL). CHS 828 induced few and weak signs of apoptosis. Metabolic activity was the only parameter affected during the first 24 h of exposure. After ~30 h, proliferation (DNA synthesis) and protein synthesis ceased, and viability started to decrease towards 10% at 72 h. Morphology and ultrastructure of dying/dead cells showed predominant necrosis. The decrease in viability was postponed by protein synthesis inhibition or maintenance of ATP levels by 3-aminobenzamide. In addition, 3-aminobenzamide switched morphology towards apoptosis. Continuous co-exposure to CHS 828 and etoposide resulted in impressive cell kill synergy in U-937 GTB cells at effect levels of 30-70%. Pre-exposure to CHS 828 for 18 h or more, on the other hand, resulted in diminished cell kill and inability to activate the apoptotic machinery upon etoposide stimulation, evaluated by morphology and caspase activity. In summary, CHS 828 induced cell death is predominantly non-apoptotic, does not involve caspases and can be postponed by maintained protein synthesis and ATP levels.

Massively parallel measurement techniques can be used for generating hypotheses about the molecular underpinnings of a biological systems. This thesis investigates how domain knowledge can be fused to data from different sources in order to generate more sophisticated hypotheses and improved analyses. We find our applications in the related fields of cell cycle regulation and cancer chemotherapy. In our cell cycle studies we design a detector of periodic expression and use it to generate hypotheses about transcriptional regulation during the course of the cell cycle in synchronized yeast cultures as well as investigate if domain knowledge about gene function can explain whether a gene is periodically expressed or not. We then generate hypotheses that suggest how periodic expression that depends on how the cells were perturbed into synchrony are regulated. The hypotheses suggest where and which transcription factors bind upstreams of genes that are regulated by the cell cycle. In our cancer chemotherapy investigations we first study how a method for identifiyng co-regulated genes associated with chemoresponse to drugs in cell lines is affected by domain knowledge about the genetic relationships between the cell lines. We then turn our attention to problems that arise in microarray based predictive medicine, were there typically are few samples available for learning the predictor and study two different means of alleviating the inherent trade-off betweeen allocation of design and test samples. First we investigate whether independent tests on the design data can be used for improving estimates of a predictors performance without inflicting a bias in the estimate. Then, motivated by recent developments in microarray based predictive medicine, we propose an algorithm that can use unlabeled data for selecting features and consequently improve predictor performance without wasting valuable labeled data.

Insensitivity to chemotherapy is an ongoing issue in cancer treatment, one that appears to be highly dependent on patient-specific variations. It has been shown clinically that while a subset of patients will successfully respond to a particular chemotherapeutic regimen, there exists another subset of patients who when exposed to the same course of therapy will remain resistant to treatment or exhibit signs of relapse after treatment has been administered. This discrepancy raises interesting questions regarding the role that patient-specific variations play in controlling the efficacy of chemotherapy treatment regimens. Doxorubicin (Dox) is a common chemotherapeutic agent used in the treatment of a variety of solid tumors and leukemias and resistance to Dox treatment is a major issue in cancer chemotherapy, oftentimes leading to patient relapse. To gain a deeper understanding of the processes that influence Dox resistance, we must first understand the mechanisms that underlie and contribute to Dox's toxicity. To this end, the metabolic reactions that activate Dox have been implicated as major determinants of Dox cytoxicity and as possible factors that control Dox resistance in cancer cells. There are several lines of evidence that redox-dependent metabolism plays a large role in Dox toxicity. The Dox bioactivation network is comprised of a system of reduction/oxidation (redox) reactions that lead to the formation of toxic Dox metabolites and reactive oxygen species (ROS). Moreover, multi-drug resistant acute lymphoblastic leukemia cells derived from relapsed patients have elevated levels of the antioxidant glutathione and show insensitivity to Dox treatment. The redox dependence of Dox bioactivation, the understanding that Dox treatment generates ROS, and the evidence that Dox resistant cells exhibit increased antioxidant capacity, suggest the possibility that redox pathways modulate the efficacy of Dox treatment in cancer cells. The overall objectives of the proposed dissertation, therefore, were to investigate how the redox properties of the Dox bioactivation network influence Dox toxicity in acute lymphoblastic leukemia cells, and to provide evidence that cell-specific variations in the intracellular levels of these redox components influences the degree to which Dox treatment will induce cancer cell death. The significant findings of this study are that the redox reactions involved in Dox metabolism are dual-natured, containing a toxicity-generating module characterized by nicotinamide adenine dinucleotide phosphate (NADPH)-dependent Dox reductive conversion, as well as an ROS signal-generating module characterized by NADPH- and oxygen-dependent Dox redox cycling. The balance between the coupled redox reactions that comprise the toxicity- and ROS signal-generating modules of Dox bioactivation determines the sensitivity-phenotype of leukemia cells and phenotypic changes in the Dox-sensitivity of leukemia cells can be induced by the successful modulation of the Dox bioactivation network through the pharmacological inhibition of NADPH in a concentration- and cell type-dependent manner. This study highlights the importance of the intracellular redox network in controlling chemotherapy-induced ROS. The unequal distribution in antioxidant burden across the various intracellular antioxidant enzymes suggests a significant role for NADPH supply, as controlled by the enzyme glucose-6-phosphate dehydrogenase (G6PD), to the intracellular ROS buffering capacity of cells during instances of oxidative stress. Changes in G6PD activity were shown to promote protein-S-glutathionylation during oxidative stress conditions, thereby implicating G6PD in the modulation of redox-sensitive signal transduction pathways. The intracellular glutathione redox balance, a measure of the intracellular redox environment, can effectively regulate Dox-induced NF-κB signal transduction in leukemia cells. The systematic modulation of intracellular glutathione redox balance in leukemia cells by N-acetylcysteine (NAC) revealed an important role for protein S-glutathionylation mechanisms in the control of NF-κB signal transduction induced by Dox treatment. These findings identify the glutathione redox network as a potential therapeutic target for the systematic modulation of Dox sensitivity in cancer cells and elucidate the complex role that antioxidants such as NAC can play in modulating the effectiveness of Dox chemotherapy treatment regimens. Lastly, this study highlights the need for and the capacity of computational models to accurately describe the complex redox-reactions that contribute to Dox metabolism in leukemia cells. This study is groundbreaking in its use of computational modeling to analyze reversible electron transfer events between proteins using mass-action kinetics. The models developed in this study can accurately explain cytosolic doxorubicin bioactivation, intracellular hydrogen peroxide clearance, and kinase-specific S-glutathionylation, thereby showing that the use of comprehensive and/or relatively simple computational models can provide semi-quantitative predictions about the behavior of redox systems in mammalian cells as they relate to Dox-induced toxicity and Dox-induced cell signaling.

Breast cancer is the most common malignancy in women and represents one of the major causes of death worldwide. The PI3K-Akt-FOXO3a signalling pathway has been shown to play a crucial role in tumorigenesis and the development of drug resistance in many cancer types. However, previous studies on FOXO3a using breast cancer tissues were controversial. So this study aims at better understanding of the role of FOXO3a in the development of drug resistance, especially endocrine resistance and anthracycline resistance in breast cancer. Examination of FOXO3a and phosphorylated-Akt (P-Akt) expressions in breast cancer tissue microarrays revealed nuclear FOXO3a was significantly associated with poor prognosis (p=0.014) and lymph node positivity (p=0.052) in invasive ductal carcinoma. Using the tamoxifen and anthracycline-sensitive and -resistant breast cancer cell lines as models, we found that the nuclear accumulation of FOXO3a was associated with enhanced anthracycline-resistance but not tamoxifen-resistance. This was consistent with the finding that sustained nuclear FOXO3a was associated with poor prognosis, as cytotoxic chemotherapy resistance is linked to limited therapeutic options and poor prognosis. We demonstrated a possible feedback mechanism in which induction of FOXO3a activity in the anthracycline-sensitive MCF-7 cells induced Akt phosphorylation and promoted cell proliferation arrest. Using MDA-MB-231-FOXO3a(A3):ER cells in which FOXO3a activity could be induced by 4-hydroxytamoxifen, we showed that FOXO3a induction could up-regulate PI3K-Akt activity but had little effect on cell proliferation, which indicates impaired Akt-FOXO3a axis in chemoresistant cell models. To further uncover the precise mechanism of Akt-FOXO3a deregulation in the development of chemoresistance, we have explored the post-translational regulation of FOXO3a by miRNAs. Through a series of Gain-and-Loss functional experiments and luciferase reporter assays in vitro, three miRNAs, including miR-222, miR-221 and miR-29a, were found to suppress FOXO3a protein expression through binding directly to FOXO3a 3’UTR. Moreover, the aberrant expressions of the miR-222/221 cluster and miR-29a in drug resistant cell lines could confer a proliferation advantage to cancer cells through suppressing FOXO3a expression. We further demonstrated that FOXO3a as a transcription factor could transactivate the oncogenic miR-222 and miR-221 expressions under certain chemotherapy stimulation. This suggests the existence of a feedback regulatory loop composed of the miR-222/221 cluster and FOXO3a which may not only play a self-protective role under drug treatment in chemosensitive cells, but also partially explain the tolerated nuclear FOXO3a in the breast cancer with poor prognosis. Taken together, our study suggested that lymph node metastasis and poor survival in invasive ductal breast carcinoma are linked to an uncoupling of the Akt-FOXO3a signalling axis, as in these breast cancers the nuclear-located FOXO3a was unable to induce cell death or cell cycle arrest. We also demonstrated post-translational regulation of FOXO3a by miR-222/221 and miR-29a, while aberrant expressions of miR-222/221 and miR-29a may promote cell resistance to therapy through directly suppressing FOXO3a. FOXO3a could further contribute to the deregulation of the miR-222/221 cluster as a transcription factor in breast cancer. Studying this Akt-FOXO3a-miRNAs signalling circuit will provide us better understanding in predicting and monitoring treatment response in breast cancer and other malignancies. / published_or_final_version / Pathology / Doctoral / Doctor of Philosophy

Epithelial ovarian cancer is a leading fatal malignancy in women. The high mortality rate of this cancer is due to the poor prognosis and that the majority of patients are diagnosed at late stage. Therefore, a combination of cytoreduction and adjuvant chemotherapy is the only choice for this disease at late stage. Platinum-based chemotherapy regimens are the first line treatment for ovarian cancer. However, the repetitive challenges of platinum-based agents and the frequent relapse cause acquired chemoresistance which is the major obstacle for clinical management of this disease. The underlying molecular mechanism for acquired chemoresistance remains largely unclear. Therefore, understanding the molecular mechanism in acquired chemoresistance of ovarian cancer is urgently needed. Emerging evidence has suggested that dysregulation of miRNAs is significantly involved in acquired chemoresistance in human cancers. In this study, the expression status and functional roles of miR-199b-5p were characterized in the chemoresistance of ovarian cancer. By miRCURY LNA™ miRNA array, miR-199b-5p was one of the downregulated miRNAs identified in acquired chemoresistant ovarian cancer cells. Further Q-PCR analysis found that miR-199b-5p exhibited a pattern of progressive reduction from early to advanced stage and from low to high grade ovarian cancer tissue samples (N=79). Interestingly, the expression of miR-199b-5p could be restored upon 5-Aza-2’-deoxycytidine (5-Aza-dC) treatment in ovarian cancer cells. Methylation-specific PCR and bisulfite genomic sequencing revealed that the promoter region of miR-199b-5p (Chromosome9q34) was frequently hypermethylated in chemoresistant ovarian cancer cells. Functionally, enforced expression of miR-199b-5p remarkably diminished the cisplatin-resistance in miR-199b-5p low expressing ovarian cancer cells, whereas depletion of miR-199b-5p by siRNA approach augmented cisplatin-resistance in miR-199b-5p high expressing ovarian cancer cells. A tumor xenograft mouse model further confirmed that miR-199b-5p could sensitize ovarian cancer cells to cisplatin-mediated tumor growth inhibition in vivo. On the other aspect, computational analysis plus luciferase reporter assay and western blotting identified JAG1, a key ligand of Notch1 signaling frequently associated with chemoresistance of human cancers, as a direct target of miR-199b-5p in ovarian cancer cells. Intriguingly, an inverse relationship between the downregulation of miR-199b-5p and the upregulation of JAG1 was found by in situ hybridization (ISH) and immunohistochemical (IHC) analyses in a commercial ovarian cancer tissue array (OVC1021). Functionally, enforced expression of JAG1 increased cisplatin resistance of ovarian cancer cells. In contrast, depletion of JAG1 by siRNA approach abrogated the cisplatin-resistance of ovarian cancer cells. This finding was in consistent with the effect of enforced expression of miR199b-5p in cisplatin-resistant ovarian cancer cells. Mechanistically, the Notch1 luciferase reporter assay and western blotting analysis demonstrated that the Notch1 signaling activity could be activated by JAG1 or conversely be inhibited by enforced expression of miR-199b-5p. Further investigation using the Notch specific inhibitor γ-secretase showed that JAG1-Notch1 signaling is a crucial pathway associated with chemoresistance of ovarian cancer in vitro and in vivo. Taken together, this is the first study showing that the epigenetic silencing of miR-199b-5p leads to aberrant activation of JAG1-Notch1 signaling and such signaling cascade plays a critical role in tumor progression and chemoresistance of ovarian cancer. / published_or_final_version / Obstetrics and Gynaecology / Doctoral / Doctor of Philosophy

published_or_final_version / Chemistry / Doctoral / Doctor of Philosophy

Esophageal squamous cell carcinoma (ESCC) is the predominant type of esophageal cancer in Asia. Cisplatin is commonly used as an agent for treating ESCC patients undergoing chemotherapy. However establishment of resistance over the course of treatment diminishes the clinical usefulness and is one reason explaining poor prognosis of ESCC patients. In order to gain insights into the mechanism of cisplatin resistance in ESCC, HPLC/nESI-MS/MS proteomic profiling was employed to examine the global protein alterations of cisplatin-resistant ESCC cell line HKESC2/CDDP comparing with its parental cisplatin-sensitive cell line HKESC2. Stable over-expression and knocked-down cell lines were established for pathway analysis and functional studies. Seventeen proteins were identified with more than 2-fold difference in expression levels. These proteins are involved in endoplasmic reticulum stress response, metabolic processes, DNA replication and repair, nucleotide binding and cell cycle control, while some of them are components of cytoskeletal proteins. Among them, 14-3-3σ was one of the most significantly upregulated proteins found in HKESC2/CDDP cells and its differential expression levels were validated using western blotting and real-time quantitative polymerase chain reaction. Pathway analysis revealed that ectopic overexpression of 14-3-3σ caused a general upregulation in DNA repairing genes. Furthermore, functional validation showed that elevated 14-3-3σ expression contributed considerably to the observed cisplatin resistance in HKESC2/CDDP cells. While knocking down 14-3-3σ expression reversed the above situations in SLMT1 cells. I conclude that up-regulation in 14-3-3σ, together with DNA repairing genes, contributes to the establishment of cisplatin resistance in HKESC2/CDDP cells. Knocking down 14-3-3σ expression sensitized ESCC cells to cisplatin treatment, and hence, opens a therapeutic opportunity for ESCC cisplatin resistance. / published_or_final_version / Surgery / Doctoral / Doctor of Philosophy

published_or_final_version / abstract / toc / Molecular Biology / Doctoral / Doctor of Philosophy