Surgery (St Vincent's) - Theses
Permanent URI for this collection
Search Results
1 - 7 of 7
-
Item
-
ItemCharacterisation of endogenous repair mechanisms following endothelin-1 induced stroke in rats and the effects of human adult stem cell transplant to support brain reconstruction( 2014)Brain injury from stroke often results in permanent damage and disability due to neurons failing to re-establish lost connections. Potential for brain regeneration relies heavily on the surrounding microenvironment. Contributions from inflammatory cells, angiogenesis and stem cells all require collective consideration when investigating treatment options. Understanding this paradigm is critical to developing therapies that promote recovery. Cell-based therapies offer hope in rescuing the stroke affected brain and restoring function. However, differentiation of transplanted cells into significant neuronal populations is yet to be realised. The work presented in this thesis investigates cellular responses to brain injury and repair mechanisms activated following stroke using the endothelin-1 model of focal cerebral ischemia in conscious rats. Additionally these studies explore the potential of human adult neural progenitor cells to support brain repair. First, we investigated aspects of brain remodelling initiated following stroke, including the impact of lesion size on angiogenesis, cell responses within the subventricular zone (SVZ), inflammation, and scar formation. Immunohistochemical analysis revealed a positive correlation between stroke severity and the degree of pathological responses to recovery after stroke. Stroke severity was found to increase cell proliferation and migration from the SVZ, with many of these cells positive for GFAP and incorporated into the glial scar. Therefore, we highlight this as an important factor to consider when developing treatment strategies that stimulate cell responses within the neurogenic niche. Long term survival and success of non-autologous stem cell transplants requires use of immunosuppressive agents such as cyclosporine A (CsA). The influence of CsA on stroke outcome required investigation prior to commencing the intended stem cell transplant studies. We explored the effects of CsA administration on neurological and histological outcomes 7 days after stroke. Findings indicated CsA treatment significantly reduced the development of neurological deficits after stroke but did not affect infarct volume, activation of microglia/macrophages, or events within the neurogenic niche. CsA treatment did however attenuate reactive gliosis after stroke and retained pro-survival astrocytic phenotypes important for supporting neuronal rescue. Finally, we investigated the use of cell-based therapies to promote brain repair. SVZ-derived human neural progenitor cells (hNPCs) were isolated, characterized, and differentiated into GABAergic neurons. Pre-differentiated GABAergic neurons, undifferentiated SVZ-hNPCs or media alone were transplanted into the rat brain 7 days after stroke during angiogenesis as it was hoped exogenous transplants would benefit from a redeveloped microvascular bed. GABAergic cell transplants were observed to accelerate functional recovery, showed evidence of maturation and promoted endogenous neurogenesis 28 days post-transplant. Undifferentiated hNPC transplants were predominantly GFAP positive and incorporated into the glial scar. These results suggest techniques aimed at differentiating cells into a neural lineage prior to transplant may be a favourable alternative for stroke treatment. Furthermore, targeting angiogenesis for cell-based treatments may offer greater survival of cells and support graft maturation for functional recovery. In conclusion, this work contributes significantly towards characterising endogenous cellular responses to stroke injury and recovery, and provides preclinical evidence for the benefits of directing exogenous stem cell differentiation towards neural lineage cells prior to transplant, resulting in accelerated recovery.
-
ItemRole of Epithelial Mesenchymal Plasticity associated cancer subpopulations in mammary tumourigenisis and chemoresistance( 2014)Tumour heterogeneity is a key characteristic of cancer and has significant implications relating to tumour response to chemotherapy as well as patient prognosis and potential relapse. It is increasingly accepted that tumours are clonal in origin, suggestive of a tumour arising from a deregulated or mutated cell. Cancer stem cells (CSC) possess/propagate these capabilities, and with appropriate intracellular triggers and/or signalling from extracellular environments, can ‘differentiate’ to initiate tumour formation. Additionally through epithelial mesenchymal plasticity (EMP), where cells gain and maintain characteristics of both epithelial and mesenchymal cell types, epithelial-derived tumour cells have been shown to ‘de-differentiate’ to acquire cancer stem attributes, which also imparts chemotherapy resistance. This new paradigm places EMP centrally in the process of tumour formation, propagation, progression and metastasis, as well as modulating drug response to current forms of chemotherapy. Furthermore, EMP and CSCs have been identified in cancers arising from different tissue types making them a possible generic therapeutic target in cancer biology. In this study, we expand on the relationship between tumour heterogeneity, EMP and CSC in BrCa through the identification and characterisation of epithelial and mesenchymal subpopulations within two BrCa cell lines. In addition, we demonstrate the plasticity that allows these cell populations to effectively regenerate the other cell populations with a particular emphasis on the CSC phenotype. Through a functional genomics screen, the importance of the mesenchymal phenotype in tumour initiation is demonstrated. Taken together, this study demonstrates that heterogeneity exists at a cell line level and this heterogeneity differs in different cellular systems. We also find evidence to suggest that BrCa cell lines can use multiple mechanisms to achieve an outcome such as tumour initiation or mammosphere formation, and subsequently emphasize the importance of phenotype specific drugs. This ideology of drug repurposing to identify phenotype specific drugs is explored through the use of the connectivity map database to identify new uses for previously established drugs to target these subpopulations find preliminary evidence for the role of HDACi to affect these EMP associated subpopulations in BrCa cell lines.
-
ItemAntioxidants as potential anti-osteolytic therapies for breast cancer bone metastasis( 2013)Reactive oxygen species (ROS) are known to play a crucial role in different stages of breast cancer progression, in particular in the maturation and activation of bone-resorbing osteoclast cells. In support of this, several recent reports have shown that antioxidants suppress bone resorption in mouse models of osteoporosis. On this basis, antioxidants may have therapeutic potential in inhibiting breast cancer-induced osteolysis. While measurement of antioxidant activity in cell culture is relatively easy, the assessment of antioxidant activity and ROS levels in human and animal tissue is far more problematic. The constant challenge of antioxidant studies is a) to demonstrate treatment effect and b) providing evidence that the observed effect was indeed caused by antioxidant activity. To investigate the impact of antioxidant treatment on breast cancer primary tumour, metastasis to secondary organs and cancer-induced osteolysis, we employed the orally available antioxidant N-acetylcysteine (NAC) in appropriate pre-clinical mouse models, using grafted mouse (4T1.2) and human (MDA-MB-231, MDA-MB-468) cell lines. Fluorescent-tagged (mCherry) breast cancer cells were inoculated into the mammary fat pad of Balb/C wild type (syngeneic) or Balb/C nu/nu (xenograft) mice. Tumour growth rates for the three cell lines grafted were similar for Placebo (water)-treated mice as well as the doses of NAC examined (0.1 to 2 g/kg/day). Tumour architecture and ratio of necrotic to viable cell mass in the MDA-468-mCherry tumours was unaffected by NAC treatment. Metastatic foci in the lungs of 4T1.2-mCherry and MDA-468-mCherry inoculated mice formed within 30 and 100 days respectively. However, overall tumour burden was not reduced by NAC treatment in the 4T1.2-mCherry model and sub-optimal duration of tumour growth limited the detection of metastatic lung lesions in the MDA-468-mCherry and MDA-231-mCherry models. Contrary to recent reports in the literature, antioxidant treatment did not promote metastasis in our pre-clinical breast cancer models. NAC, when combined with the chemotherapeutic agent doxorubicin (DOX) had no impact on the efficacy of the anthracycline. Supporting in vitro studies demonstrated that NAC at sub-lethal concentration (10 mM) has no effect on DOX-induced inhibition of clonogenicity in MDA-231-mCherry cells, but the antioxidant has an additive, positive impact on the DOX-induced inhibition of clonogenicity when the same assessment is conducted with murine 4T1.2-mCherry cancer cells. Breast cancer-induced osteolysis was initiated by inoculation of 4T1.2-mCherry and MDA-231-mCherry into the tibial marrow of syngeneic and xenograft mice respectively. Treatment with NAC inhibited the onset of osteolysis in the xenograft model and showed mild suppression of osteolysis in the syngeneic mice. To assess impact of NAC treatment on anti-osteolytic therapy with the bisphosphonate Zometa, mice were inoculated with the 4T1.2-mCherry and MDA-231-mCherry breast cancer cell lines following combination treatment with the two agents. Zometa inhibited osteolysis in the MDA-231-mCherry model and supplementation with NAC appeared to increase the inhibition, although the added bone protection did not reach statistical significance. Zometa therapy reduced bone resorption in the 4T1.2-mCherry model, but this effect was only statistically significant when Zometa and 30 mM NAC treatment were combined. This finding suggests that supplementing bisphosphonate therapy with antioxidants like NAC may increase efficacy in anti-osteolysis therapies, including therapies against breast cancer-induced bone resorption. An important part of our studies was the evaluation of pharmacodynamic assays that directly or indirectly measure antioxidant activity in vivo. A modified prooxidant/antioxidant balance (PAB) assay proved useful to measure antioxidant capacity in the NAC drug formulation and in plasma from NAC-treated, non-tumour challenged mice. The dihydroethidium (DHE) assay was employed to detect superoxide formation in kidneys from NAC-treated and control mice. We were able to demonstrate that the fluorescent oxidation product 2-hydroxyethidium can be detected by other means other than high performance liquid chromatography (HPLC) and mass spectrometry, which supports the aim to establish reliable pharmacodynamic assays that can be employed by research and clinical laboratories without expensive instrumentation or specialist skills. The thiobarbituric acid reactive substances (TBARS) assay identified higher malondialdehyde (MDA) levels in tumour challenged mice compared to non-tumour-challenged mice, indicating breast cancer-induced oxidative stress in this model. Furthermore, the assay identified significantly lower MDA levels after treatment with the antioxidant Trolox, in comparison with other antioxidants, including NAC. An assessment of 8-OH-dG urine levels in the same mice revealed significantly higher antioxidant activity in the Olive leaf extract- and Curcumin-treated mice when compared with the control group and NAC-treated mice. This offers the possibility to assess antioxidant impact on breast cancer progression and in particular breast cancer-induced osteolysis with potentially stronger antioxidants.
-
ItemMolecular phenotypes of ascites cells in drug resistant ovarian carcinomas( 2013)Epithelial ovarian cancer (EOC) is an aggressive form of cancer diagnosed at an advanced-stage. A manifestation of this disease is the accumulation of ascites fluid in the abdominal cavity. Surgery and systemic administration of platinum/taxol based combination therapy eliminates most tumour cells, however, resistant residual tumour cells eventually emerge. It is hypothesised that cancer stem cells (CSC) together with epithelial-mesenchymal transition- (EMT) and mesenchymal-epithelial transition (MET)-associated mechanisms contribute to drug resistance. The molecular characteristics accompanying these phenotypes have yet to be fully determined. Hence, the aims of this study were to: • separate and characterise the epithelial and mesenchymal ascites cells prior to and following chemotherapy, • apply gene expression microarray technology to chemonaive (CN) and chemoresistant (CR) samples to determine the gene expression signatures of the epithelial tumour and stromal cells, • determine the role of EMT and CSCs marker profiles in facilitating cisplatin resistance, • establish the role of the JAK2/STAT3 pathway in cisplatin resistant EOC cells by utilising a novel inhibitor (CYT387) which inhibits JAK2/STAT3 pathway and sensitises EOC cells to cisplatin by suppressing CSC-like phenotype and • determine the tumourigenicity of cisplatin treated and untreated EOC cells by inoculating nude mice with cisplatin-treated surviving cells exhibiting features of CSC-like cells. Tumourigenic and non-tumourigenic ascites cells were separated by a novel method developed during this study and were characterised by in vitro assays including wound healing, cell growth and chemosensitivity assays. The expression of genetic markers was assessed using immunofluorescence, flow cytometry, Western blot, qPCR and gene expression microarray. Tumourigenicity of EOC cells was assessed by xenograft studies in mouse. Hematoxylin and Eosin staining performed on histology section of mice organs was used to determined invasive tumour behaviours. The significant findings were; • Separation of the epithelial and mesenchymal ascites cells isolated from high-grade EOC patients prior to and following chemotherapy. • Molecular and gene expression profiles of epithelial cells, which were tumourigenic and expressed markers such as CA125, EpCAM, STAT3, Oct4 and E-cadherin, while non-tumourigenic stromal cells exhibited a mesenchymal phenotype and were enriched for CD44, MMP2, MMP9, FSP, CD90, CD105, CD73, N-cadherin and vimentin. • Genetic signatures of the epithelial tumours and stromal cells obtained from gene expression microarray technology applied to CN and CR samples. • EMT and CSCs marker profiles were determined to play an important role in facilitating cisplatin resistance. • The JAK2/STAT3 pathway was implicated in cisplatin resistance EOC cells as the CYT387 inhibitor sensitised EOC cells to cisplatin by suppressing CSC-like phenotype. • Tumourigenicity of cisplatin treated and untreated EOC cells in nude mice - surviving cells exhibited features of EMT- and CSC. Unique populations of EOC cells exhibiting characteristics of CSC-like cells are capable of undergoing cellular changes in line with EMT-like (during cisplatin treatment) and enhanced epithelial phenotypes (in in vivo settings). Together these traits contribute to drug resistance and recurrence, and targeting pathways that facilitate these phenotypes may improve treatment.
-
ItemPrediction of chemotherapy-induced hepatic injuries on clinical, imaging and genetic parameters following treatment of colorectal carcinoma( 2012)Chemotherapy-induced hepatic injuries (CIHI) are an increasingly common and serious problem facing clinicians. Peri-operative outcomes following hepatic resection are inferior in patients with severely injured livers. Development of injury however remains unpredictable. Similar injury phenotypes are associated with other disease states such as diabetes and obesity. It may be possible to predict risk of CIHI by combining an analysis of clinical features with genetic factors known to be involved in the metabolism of chemotherapeutics. Similarly, functional imaging such as hepatic iminodiacetic acid (HIDA) scanning may provide early warning of deteriorating liver function secondary to hepatic injury. A combined model incorporating clinical, imaging and genetic parameters may be able to be developed to provide a risk level for individual patients based on these features. Firstly, 233 samples of non-cancerous hepatic tissue stored in a tissue bank and pathology archive were retrieved and scored for histological hepatic injury (steatosis, steatohepatitis and sinusoidal injury [SI]) using previously validated methods. Clinical features regarding these patients were extracted and correlated with degree of injury. Next, applying a candidate gene approach employing genes known to be relevant regarding tumour response or host toxicity, genetic features were correlated with hepatic injury and peri-operative outcomes. mRNA expression data, genetic sequencing for known polymorphisms and probing for presence of absence of other relevant isoforms of relevant genes was undertaken. A pilot study was then developed to prospectively assess hepatic injury comparing pre-chemotherapy samples of non-cancerous hepatic tissue with tissue collected at hepatic resection following treatment. HIDA scanning was performed pre- and post-chemotherapy, and genetic features were analyzed using microarray mRNA expression techniques. Injury rates were 18%, 4% and 19% for steatosis, steatohepatitis and SI, respectively. On multivariate analysis, high-grade steatosis was more common in diabetics [odds ratio (OR)=3.24, p=0.01] and patients with higher weight (OR/kg=1.04, p=0.02) and steatohepatitis was increased with metabolic syndrome (OR=5.54, p=0.02). Chemotherapy overall demonstrated a trend towards approximately doubled risk of high-grade steatosis and steatohepatitis while not affecting SI. Pre-operative chemotherapy was however associated with increased SI on multivariate analysis (OR=3.39, p<0.001). Operative morbidity was not increased with chemotherapy, but was increased with steatosis (OR=2.38, p<0.001). Low-level DPD mRNA expression was associated with steatosis on multivariate analysis (OR=1.53 p=0.03). Presence of GSTT1 was associated with sinusoidal injury (regardless of chemotherapy administration status) OR=5.31 (p=0.01). Preliminary results suggest changes in HIDA clearance rate and excretion half-life may herald hepatic injury. It appears there is a specific microarray expression signature in non-cancerous liver tissue indicative of severe hepatic injury following treatment with chemotherapy. Predisposition to development of CIHI may be predictable based upon individual patient characteristics, such as diabetes and higher weight (for steatosis), metabolic syndrome (for steatohepatitis) or pre-operative chemotherapy use (SI). Into this equation, low expression of DPD mRNA needs to be considered regarding steatosis, and GSTT1 presence for SI. HIDA parameters may contribute baseline characteristics predictive of injury, but this requires further validation. Similarly, microarray expression analyses also appear promising regarding identifying an injury-prone subgroup pre-treatment. It is the presence of high-grade injury (particularly steatosis) that increases peri-operative morbidity and not administration of chemotherapy as such.
-
ItemThe role of RECK in sarcoma( 2011)Sarcomas are malignant tumours of connective tissue, and arise in bone, muscle, adipose, or fibrous tissue. Osteosarcoma and chondrosarcomas are the most common primary bone tumours, with treatment consisting of wide-margin surgery and, in osteosarcoma, neoadjuvant chemotherapy. Chemotherapy has improved survival, but not without systemic side effects, and it is not effective in chondrosarcoma. Liposarcoma is a common soft tissue sarcoma, which also has a variable response to chemotherapy. All sarcomas demonstrate angiogenesis, growth and invasion of surrounding tissues as part of their natural, destructive progression. Targeting the molecular pathways underlying these processes is likely to improve outcomes and also avoid systemic cytotoxicity. The RECK protein has shown promising therapeutic properties in this regard. In the present study, RECK was studied in human sarcoma to examine the expression pattern and relationship with prognosis. RECK was downregulated in osteosarcoma cells, and the average survival for RECK negative patients was below average. No relationship between RECK and prognosis was found for chondrosarcoma and liposarcoma, although larger studies are needed. A consistent finding across all sarcomas studied was the presence of RECK in sarcoma vessels, and this expression was up-regulated compared with non-tumour vessels. RECK was overexpressed in sarcoma cell lines by transfection. This inhibited the invasion of SaOS-2 (osteosarcoma), JJ012 (chondrosarcoma), and SW872 (liposarcoma). Furthermore, RECK inhibited JJ012 proliferation and supported adhesion to collagen I in SaOS-2, while colony formation was suppressed in both cell lines. Overexpressing RECK in HMEC-1 (microvascular endothelial cell line), supported mature tube formations and decreased HMEC-1 invasion. RECK is likely to exert these effects through inhibition of MMP-2. RECK-transfected SaOS-2 cells were co-cultured with RAW cells (pre-osteoclasts) and this inhibited osteoclast mineralised bone-resorbing capacity. RECK overexpression in an orthotopic animal model of osteosarcoma significantly inhibited tumour formation and lung metastasis. These finding were noted on serial assessment of tumour size, x-ray changes, micro CT, and macroscopic specimens. An orthotopic mouse model of chondrosarcoma, utilising periosteal and intratibial sites for JJ012, was established for testing RECK, although increased transfection toxicity in the JJ012 cell line precluded a definitive analysis. To assess therapeutic RECK delivery, recombinant partial and full-length rhRECK proteins were first tested in vitro, and this demonstrated reduced SaOS-2 invasion when treated with the partial-length protein, while SaOS-2 cell viability was significantly reduced when treated with the full-length protein. A vascularised model of osteosarcoma growing within a polycarbonate chamber inside a rat limb was developed to test delivery of rhRECK proteins via an osmotic pump and catheter. An initial study delivering partial-length RECK was hampered by high infection rates and mechanical disruption of the catheters. Revisions to the model allowed more adequate testing of full-length rhRECK delivery, which resulted in increased percentage tumour necrosis in comparison with control chambers. These collective findings indicate that RECK has important anti-tumour roles in sarcoma and may provide prognostic information for clinicians in the future. Further studies evaluating intravascular delivery of RECK to sarcoma tumours, within the newly described model may provide a new avenue of treatment for sarcoma patients.