Ovarian cancer is a lethal gynaecological malignancy which is known as the silent killer. It has a poor prognosis due to the lack of major symptoms in early stage disease and hence its late detection. Cancer antigen-125, the most widely used biomarker for ovarian cancer detection, lacks appropriate sensitivity and specificity. Thus, early biomarkers of the disease are urgently required. Proteomic analysis of human serum promises to be a valuable approach for the discovery of putative biomarkers for human malignancies like ovarian cancer, which could be developed into non-invasive blood tests. In this study, serum samples from a pilot study for ovarian cancer screening which were collected prior to diagnosis were processed at Memorial Sloan Kettering Cancer Research Centre, in collaboration with Prof. Tempst's group, who had developed a novel mass spectrometry (MS)-based technology platform for the high-throughput extraction and measurement of serum peptides. Several marker peaks were identified, which when used in combination with the ovarian cancer biomarker CA-125, assisted in the discrimination of case versus healthy samples at an earlier point prior to diagnosis. Work then involved the establishment and optimisation of a similar serum profiling platform at UCL. This involved the optimisation of a liquid-handling robot to provide semi-automated high-throughput sample purification and spotting, and optimisation of spectral acquisition and processing. The reproducibility of the platform was tested and the effects of different sample handling conditions on peptide profiles examined. The method was then used to search for putative markers of ovarian cancer, using identically processed samples from women diagnosed with malignant or benign ovarian cancer and healthy controls. Finally, as a complementary approach to discover protein biomarkers, the same samples were profiled using 2D Difference Gel Electrophoresis, employing different fractionation strategies to overcome the very large dynamic range of protein expression in serum. Mass spectrometry was used to identify several previously reported and some novel putative biomarkers of ovarian cancer, which warrant further validation.

The Wilms' tumour antigen 1 (WT1) is a transcription factor which is over-expressed in several leukaemias and solid tumours. Currently, there is limited information about the expression and immunogenicity of WT1 in prostate cancer. I used immunohistochemistry to analyse WT1 expression in prostate cancer samples and tetramers to detect WT1-specific T cell responses in the peripheral blood. 39% of cancer samples showed nuclear and cytoplasmic staining for WT1, whereas the staining of all normal prostate tissue was limited to the cytosol (p < 0.03). Furthermore, WT1-specific T cells which bound tetramer were detectable in the peripheral blood of 20/38 (53%) prostate cancer patients, ex vivo. Although these T cells did not expand in 20/20 patients using peptide stimulation with IL2/IL7, a population of WT1-specific CTL accumulated in 3/8 patients following culture with IL15. T cell receptor (TCR) gene transfer is a strategy to circumvent possible impairment of autologous T cell responses against tumour associated antigens such as WT1. However, exogenous and endogenous TCRs are in competition for CD3 chains, which may reduce the expression of the introduced TCR and impair the antigen specific function of transduced T cells. I developed an approach to improve the expression of exogenous TCR chains by co-transducing TCR genes together with the genes encoding the CD3 complex. Transfer of CD3 and TCR genes into primary T cells resulted in up to a 5 fold increase in TCR expression and up to a 9 fold increase in tetramer binding compared to that seen after transfer of TCR genes alone. Furthermore, T cells expressing high levels of TCR/CD3 demonstrated increased sensitivity. This data indicates that the efficacy of TCR expression, and the effector function of gene modified T cells, is substantially enhanced by the co-transfer of CD3 genes.

The matricellular protein SP ARC (secreted protein acidic and rich in cysteine) is a master regulator of tissue stroma; it is expressed during tissue remodeling and repair and its impaired regulation has a role in fibrotic responses, in determining the composition of the tumor-associated stroma and, less expected, in regulating the immune response. The complexity of studying SP ARC in cancer stems from its concomitant or alternative expression in tumor and/or stroma cells depending on the tumor histotypes. Extracellular matrix (ECM) gene profile of human breast carcinomas correlates SP ARC expression with prognosis and response to therapy. In particular, a subgroup of breast tumors can be identified based on ECM gene signature headed by SP ARC that correlates with tumor grade, bad prognosis and poor response to therapy. Epithelial to mesenchymal transition (EMT) has been associated to drug resistance and SPARC has been implicated in EMT. Nevertheless the source of SP ARC and its path toward EMT and the associated drug resistance have not been identified yet. I have used a well-defined model of transgenic mouse mammary carcinoma expressing the mutated rat oncogene c-erB2 (HER-2/neu), under the mouse mammary turn or virus promoter, backcrossed with Sparc-I - mice to establish mammary carcinoma cell lines devoid of SP ARC expression, in which SP ARC can be restored by retroviral gene transduction. With this isogenic cell lines I have been able to show that SP ARC expression in tumor cells induces EMT only in vivo, indicating that EMT is a micro environmental and not a cell autonomous process. Moreover, SP ARC expressing tumors became resistant to treatment with Doxil (Doxil, a pegylated form of Doxorubicin). Investigating the environmental players responsible of EMT in SP ARC transduced tumors, I found that myeloid cells and particularly the so called myeloid derived suppressor cells (MDSC) have a role in EMT and drug resistance depending on SP ARC. SP ARC does not change the number or ratio between the Ly60high and the Ly6010w fractions ofCD11b MDSC rather it detelmines the immunosuppressive phenotype of recruited myeloid cells. The data show that SPARC-forced expression increased the activation of pro-turn oral (CCL2, CCLS) and immunosuppressive (ArgI, NOS2, Cox-2) genes in monocyte-MDSC (the Ly6010w fraction). Furthermore, the role of myeloid cells recruited in Doxil resistance and EMT has been proven adding bisphosphonate to Doxil during treatment. , Bisphosphonates have been shown to inhibit induction and function of myeloid suppressor cells. Indeed, Bisphosphonates addition reverted EMT, the immunosuppressive phenotype of myeloid cells and rendered SP ARC-producing tumors sensitive to Doxil administration.

The non-invasive diagnosis of cancer and assessment of disease severity are both important challenges facing urologists managing patients with kidney, bladder and prostate cancer. Both challenges can potentially be met through the use of molecular markers of cancer that are present in blood, urine and biopsy material from tumours. In this thesis I explore the use of DNA replication licensing proteins, key components of the DNA replication machinery, as diagnostic, prognostic and predictive factors for urological cancers. The main markers studied, Mcm2, McmS and geminin are crucially important as their expression is closely linked with cell cycle activation and control in normal and malignant cells. Loss of growth control rapidly leads to abnormal expression of these proteins. Furthermore, by combined analysis with the well-known cell cycle marker Ki67, we have learned more about the cell cycle kinetics in urological tumours. The studies in this thesis include the examination of DNA replication licensing protein expression in kidney and prostate cancer tissue. In the study of kidney cancer, my coworkers and I examined a series of tumours from patients who had undergone nephrectomy for renal cell carcinoma. We showed that increased Mcm2 expression is associated with reduced disease-free survival time and that Ki67 is an independent predictor of disease free survival in this disease. In prostate cancer we investigated linkages between the MEKS/ERKS pathway and DNA replication licensing during prostate carcinogenesis. We confirmed that dysregulation of upstream pathways leads 8 to increased expression of DNA replication licensing proteins. Importantly, we showed that Mcm2 was an independent prognostic marker in prostate cancer. The latter two studies examine the use of Mcm5 as a urinary diagnostic marker for prostate and bladder cancer. We were able to demonstrate that in a series of men at a variety of stages of prostate cancer, elevated urinary Mcm5 levels were present in a very large proportion of men with prostate cancer and mostly absent in patients with a benign diagnosis (82% sensitivity, 73-93% specificity). In a much larger study, the largest to date to evaluate a commercialized comparative marker NMP22, we determined that the academic Mcm5 assay was a highly accurate diagnostic marker for bladder cancer. Interestingly, we found that the combination of Mcm5 with NMP22 improved the detection of bladder cancer and allowed the identification of95% of patients with clinically significant disease. In the discussion section ofthis thesis I conclude that the DNA replication licensing proteins offer researchers a number opportunities to develop diagnostic and predictive markers for urological cancer. Clinical trials of a commercialized Mcm5 assay (UrosensTM) are underway and may lead to novel minimally invasive approaches to urological cancer detection.

The aim of this thesis is to investigate the spatial extent of brain tumours by developing new methods combining the metabolic information provided by low resolution 1 H magnetic resonance spectroscopic imaging (MRSI) data with high resolution structural information from diffusion tensor imaging (DTI). In chapter 3 and chapter 4, the spectral analysis tool LCModel is used to decompose single voxel (SV) MRS data of common adult and childhood brain tumours into the most likely tissue class respectively. Classification according to the highest estim,ated tissue proportion suggested comparable performance to published specialised pattern recognition methods and emphasises the flexibility of the method. In chapter 5, the LCModel tissue type analysis is refined and extended to decompose short echo MRSI data of glioma patients into normal and abnormal tissue proportions. Spatial assessment revealed metabolic low grade characteristics around most grade IV glioblastomas as potential tumour infiltration. Several visualisation techniques explored reveal heterogeneous infiltration patterns varying across patients. In chapter 6 a radial choline-to-N-acetyl-aspartate index (rCNI) is presented as an alternative method for the delineation of brain tumour MRSI exams using the choline to NAA ratio. Both simulations and analysis of real 1.5T and 3T glioma data suggest a higher specificity at similar sensitivity for rCNI over the previously published CN!. The final chapter 7 presents a novel method for the combination of the LCModel tissue type information from chapter 5 with DTI and conventional MRI data. The tissue type information is used to sample high confidence regions of tumour and normal brain and extract the underlying DTI and MRI information and create tissue specific probability density distributions. Additional distributions from areas of tumour infiltration and vasogenic oedema were sampled from glioma and metastasis data respectively. Bayes' theorem was then used to calculate probability maps of the different tissue types which show promise for tumour grading and the differentiation of vasogenic oedema and tumour infiltration.

Zerumbone, a sesquiterpene extracted from rhizomes of ginger (zingiber zerumbet Smith) is reported to have anti-proliferative activities and can induce toxicity in human cancer cells. However, its molecular mechanisms' are still poorly understood. In this study, in vitro antioxidant (DPPH, H20 2, Fe2+ chelating and reducing power), apoptotic and anti-proliferative activities of zerumbone were investigated in human cancer cells. The specific objective was to identify whether zerumbone-induced cell death occurs through apoptosis, autophagy, necrosis or another fmID of cell death by undeliaking morphological and biochemical characterisation. Human cancer cell line (Caco-2, Huh-7 and EA.hy926) viability and activity with time and in the presence of different concentrations of zerumbone were investigated using LDH. In addition, characterisation of cell death induced by different concentrations of zerumbone including changes in cell sizerphosphatidylserine externalization, caspase activation and P ARP-l involvement were studied. The results showed that cancer cell death occulTed in the absence of DNA fragmentation and caspase activation at (5 Ilg/ml). Additionally, cancer cell death was characterised by cell shrinkage and an absence of necrotic cell death pathway. Anti-proliferative activity of zerumbone (5 and 10 Ilg/ml) on human cancer cells was also investigated by changes in the DNA content using flow cytometry.

Deficits in social interaction, communication and repetitive patterns of behaviour are common characteristics of autistic spectrum disorders (Autism). Visuo-motor neurons (known as mirror neurons), that fire both when observing or executing goal directed actions, have been shown by research to play a role in motor action and intention understanding. This has led to a link suggesting that a dysfunction of mirror neurons may be involved in some aspects of Autism difficulties. Frequent research updates are providing a greater understanding and clarification for the important specification of this hypothesis. This leads the way to suggesting the possibility of the further development and support for novel interventions such as neurofeedback for this condition. This review represents a selective overview of core documents on the topic area. Declaration of Position My interest in the potential link between Autism and a dysfunctional mirror nemon system (MNS) was inspired by reading literature on the topic 'while I was completing a Masters in Neuropsychology. What became clear to me, both from the literature and from my limited clinical exposure, is the considerable heterogeneity of Autism. Furthermore, in the absence of distinct and consistent, nemo or physiological markers, accurate identification and diagnosis of autistic spectrum disorders can be problematic and consequently the ability of clinicians and researchers to detect this condition early in development is often restricted.

The p53 research area represents an ever expanding field in which new effectors and new biological functions contribute to broaden the cellular activities in which p53 is involved in both physiological and pathological conditions. Among these biological functions, p53 is acknowledged to be a key factor in controlling genomic stability, apoptosis, metabolism, response to stress and DNA damage. Thus, the search for new p53-target genes is crucial as these could shed new light into the mechanisms through which p53 controls cell integrity and response to damage, leading to the identification new perspective therapeutic targets. The aim of the present work was to complete the characterization of DRAGO (DRug Activated Gene Overexpressed)IKIAA0247, a highly conserved gene. Data previously obtained in the lab supported the hypothesis of a potential tumour-suppressive role for DRAGO since: the gene was demonstrated to be up-regulated in vitro in response to treatment with chemotherapeutic compounds; p53 and p73 were shown to be involved in gene expression regulation; DRAGO ectopic overexpression caused cell death in host cell lines; stage III ovarian cancers displayed significantly reduced levels ofDRAGO expression compared to stage I samples. However DRAGO knockout mice had a normal phenotype and did not develop spontaneous tumours. The experiments described in this thesis work confirmed the tumour-suppressor role of DRAGO by demonstrating that in p53-1- or p53+1- mice, the deletion of DRAGO alleles significantly accelerated tumour development and shortened lifespan compared to p53-1- or p53 +1- mice bearing wild-type DRAGO alleles. DRAGO was also proved to be responsive to DNA damaging treatments in vivo displaying a tissue-specific expression pattern, similarly to other p53-downstream targets. In vitro experiments showed that DRAGO expression is regulated both at transcriptional level - through p53, p73, and promoter-methylation mechanisms - and at post-transcriptional levels by miRNAs. The generation of GFP fusion protein confirmed that DRAGO expression induced cell death but neither apoptosis or senescence seemed to be involved in DRAGO-mediated cell toxicity. Overall these data contributed to the characterization of a new p53-downstream gene and confirmed its tumour suppressive properties. As regards the biological fi1nction of the gene, preliminary experiments performed on mouse macrophages seemed to suggest a possible role ofDRAGO in the immune system functionality, even though further work is required to define how the gene mediates its tumour protective functions in this immune cell subpopulation

The cancer stem cell hypothesis suggests that tumour growth is initiated and maintained by a subset of tumour cells that display stem cell- like properties. Stem cell markers such as Leucine-rich repeat-containing G-protein coupled receptor 5 (LGRS), a regulator and downstream target of Wnt signalling, have been identified in the normal intestinal stem cells and also in colorectal tumours. This project aimed to investigate the regulation of LGRS by epigenetic mechanisms and in particular by sodium butyrate, a dietary fibre derivative, histone deacetylase (HDAC) inhibitor and candidate chemopreventive agent. It is important to study dietary compounds as many have been reported to play a role in carcinogenesis. This investigation may therefore lead to improved chemoprevention and the development of new cancer therapies. Initially the role of DNA methylation in regulating LGRS expression in normal t issue and cancer cell lines was investigated, however, although LGRS methylation was observed in some cell lines, no evidence for the functional regulation of LGRS by methylation was found. This study hypothesised that butyrate acts as a chemopreventive agent by targeting cancer stem cells and therefore aimed to investigate the effect of butyrate on the expression of LGRS and other cancer stem cell markers. Butyrate at physiological concentrations was found to downregulate LGRS expression in colorectal cancer cell lines through inhibition of LGR5 gene transcription. This effect occurred within four hours of treatment and was reversible upon removal of butyrate. LGRS expression was also decreased by other pan and class I HDAC inhibitors such as Trichostatin A and Mocetinostat, leading to the conclusion that HDAC inhibit ion was the critical mechanism by which butyrate regulated LGRS. siRNA knockdown of individual HDACs showed that it was the inhibition of HDACl specifically that mediated this effect. Additional work in to the mechanism of butyrate action suggested that Wnt signalling is not involved in the downregulation of LGRS and further work in to other potential intermediate factors will be interesting. Finally the role of LGRS overexpression on cell growth and response to butyrate was investigated. These results suggested that in some cell lines and culture conditions LGRS may enhance cell growth and also reduce the growth inhibitory effects of butyrate. These results present a novel mechanism for the regulation of LGRS, an important cancer stem cell marker, by a well-known dietary derived chemopreventive agent and hint at a role for this effect in regulating stem cell dynamics

Continuous advances in science and technology, and developments in clinical and public health are leading to improvements in the quality of life and longevity. Many diseases have been found to be curable or controlled very well by treatments. However, there are still some diseases which are not treatable or are age-related degenerative diseases such as heart disease and cancer. Cancer is a disease of complex aetiology and there is still much to be studied and understood. In this study the impact of inflammation on melanoma migration, invasion and survival was studied in a simple 2D model and in an engineered human skin model.