No description available.

No description available.

No description available.

No description available.

No description available.

Prostate cancer is the most common male cancer in the UK. A healthy diet and lifestyle may have the potential to impact upon cancer progression, treatment related side effects, and comorbidities as well as on quality of life in prostate cancer patients. Nutrition and lifestyle behaviours are therefore important aspects to consider in the treatment of prostate cancer patients. The broad aims of this thesis were to explore the importance of health behaviours in prostate cancer patients and to identify and test a suitable intervention specifically for those who are treated with androgen deprivation therapy (ADT). To achieve these aims this thesis has documented the findings of the following studies investigating nutritional and life-style behaviours in prostate cancer patients. 1) Using the cross-sectional population-based data from California Health Interview Survey (CHIS), diet and health behaviours in men with prostate cancer were compared to men with cancer other than prostate and to non-cancer controls. No differences were noted between prostate cancer patients and controls with respect to their dietary intake and physical activity levels. 2) A systematic review was conducted to determine the effects of ADT on body composition in prostate cancer patients. Significant increases in fat and declines in lean mass were observed in prostate cancer patients treated with ADT. 3) A systematic review of the literature related to lycopene intake and prostate cancer progression suggested that evidence available to date was insufficient to draw any firm conclusions regarding the benefits of lycopene supplementation in prostate cancer patients. 4) Preliminary analysis of a RCT of a 6-month dietary and physical activity intervention in prostate cancer patients treated with ADT showed that the intervention group had significant improvements in their body composition compared with controls. Physical activity, functional capacity and dietary quality also improved in the intervention group compared with controls. Recruitment to this trial is ongoing.

The cochaperone CDC37 promotes recruitment of a subset of client proteins, predominantly protein kinases, to the HSP90 chaperone, thereby regulating the activity of several signalling pathways including those altered in malignancy. Here, an investigation of CDC37's role in chaperoning clients in cancer cells was undertaken to test the hypothesis that targeting CDC37 may be therapeutically beneficial in cancer. To study the function of CDC37, expression or activity was modulated. Overexpression of CDC37 did not affect the steady state levels or activity of kinase clients such as CRAF and AKT; however, CDK4 expression was consistently elevated compared with controls.CDK4 protein half-life and CDK4-HSP90 association were also increased, suggesting that CDC37 is rate-limiting for CDK4 stability. Following overexpression of CDC37, CDK4 was depleted to a lesser extent following 17-AAG treatment; however, the sensitivity to HSP90 inhibitors, assessed by GI50, was unchanged. Conversely, siRNA-mediated CDC37 silencing caused depletion of clients such as ERBB2, CRAF, CDK4 and CDK6 by a proteasome-dependent mechanism and inhibited both activation of downstream signalling pathways and cell proliferation. Furthermore, CDC37 silencing sensitised cells to HSP90 inhibitors by potentiating kinase client depletion, cell cycle arrest and apoptosis. Expression of a CDC37 mutant deficient in binding to HSP90 caused an accumulation of endogenous CDC37 and client proteins in high molecular weight complexes corresponding to HSP90. Increased CDK4-HSP90 association without any effect on CDK4 expression was also observed, suggesting that CDC37 is required in the chaperone complex to correctly load client proteins. Additionally, it was found that the phosphatase PP5 can dephosphorylate CDC37 and influence the maturation of kinase client proteins, suggesting that it may regulate CDC37 activity in the chaperone cycle. In conclusion, by modulation of CDC37, its critical role in concert with HSP90 to stabilise kinase clients has been demonstrated, supporting the therapeutic potential of targeting CDC37 in cancer.

Colorectal cancer (CRC) is the second most common malignancy in developed countries. Germline mutations in a number of genes have been shown to cause CRC, but the Mendelian syndromes associated with these known genes (APC, DNA mismatch repair genes, MYH, SMAD4, ALK3 and STK11) account for less than 5% of the familial clustering. Colorectal tumor families which show evidence against linkage to known disease loci, provide direct evidence for the existence of uncharacterized high/moderate-penetrance CRC genes. Such observations strongly support the continued search for novel CRC predisposition genes through genome-wide linkage. To identify novel CRC susceptibility genes through linkage analyses a series of CRC families in which involvement of known susceptibility genes has been excluded were ascertained. Genome-wide genotyping was performed using high-density SNP arrays to provide maximal power to detect linkage. This study provided statistically significant evidence for a novel predisposition locus for CRC mapping to 3q21-24. Using a series of microsatellite markers the critical interval of linkage was refined to 1 4.4Mb region at 3q22. To further investigate Mendelian susceptibility to CRC, a search for a disease gene at 3q22 was performed by linkage disequilibrium (LD) mapping in 620 familial CRC cases and 960 controls, genotyping 1,676 tagging SNPs and sequencing 30 candidate genes from the region. Signal intensity data from the SNP markers was further evaluated for structural variation within 3q22. On the bases of LD mapping, chromosomal profiling and mutational analyses of genes mapping to 3q22, no causal mutation was identified. Findings are consistent with the hypothesis that variation at 3q22 contributes to a significant proportion of the unexplained genetic risk to CRC, however this is unlikely to be mediated through a restricted set of alleles and is refractory to identification by current analyses strategies.

Therapies targeting the oestrogenic stimulation of tumour growth have been a major success in reducing mortality from oestrogen receptor (ER)-positive breast cancer. However, resistance remains a major clinical problem. Various cellular signalling pathways, including the pro-angiogenic Vascular Endothelial Growth Factor (VEGF) pathway, have been implicated in the mechanism of therapeutic resistance. The aim of this project was to investigate the relationship between oestrogen- and VEGF-mediated signalling in breast cancer. The clinical relevance of the interaction between the ER and VEGF pathways was evaluated in two clinical studies: (i) Circulating VEGF and sVEGFR-1 levels were measured both before and after administration of endocrine therapy in women with primary breast cancer. The results provided evidence for possible cross-talk between ER- and VEGF-signalling. However, the aromatase inhibitor (AI) anastrozole and the ER-antagonist tamoxifen had differential effects on VEGF and sVEGFR-1. (ii) The expression of VEGF, VEGFR-2 and pVEGFR-2 and assoications with ER, PgR, pp38 and HER2 were investigated in paired samples of breast cancers before and after relapse following tamoxifen treatment. Positive correlations between ER and PgR with VEGF were noted in the relapse specimens which were not present in the primary cancers, providing support for an interaction between the ER- and VEGF-signalling pathways that may be relevant to endocrine therapy resistance. The role of ER in the regulation of VEGF expression was investigated in cell lines modelling resistance to tamoxifen and Als. Oestradiol(bound)-ER increased VEGF promoter activity and VEGF secretion and HER2 knockdown reduced VEGF secretion. This suggested that ER regulates VEGF in resistance and showed that an imperfect oestrogen response element (ERE) dependent mechanism is likely to be involved. The possibility of targeting ER- and VEGF- pathways together was investigated, initially by combining an Al with a VEGFR inhibitor (PTK787). Unexpectedly, this inhibitor was shown to directly inhibit aromatase activity in vitro. PTK787 decreased the number of vessels, cell turnover index, uterine weight, PgR and TFF1 in vivo suggesting that dual targeting of both angiogenic and endocrine signalling in a single agent is possible and may be worth exploring further. This series of studies has provided new insight into the significance of VEGF in ER-positive breast cancer, particularly that which has progressed to endocrine resistance. In addition, the ability of a single small molecule inhibitor to inhibit two distinct biological pathways may allow the rational development of agents designed to inhibit cross class targets.

The classical ERK/MAPK signalling pathway regulated proliferation, differentiation and survival. At the core of this pathway is the serine/threonine-specific protein kinase RAF, whose activation downstream of RAS leads to the sequential activation of the kinases MEK and ERK. There are three RAF proteins (ARAF, BRAF, CRAF) and three RAS proteins (HRAS, KRAS, NRAS) in humans. In melanoma, ERK is hyperactivated because BRAF is mutated in 44% and RAS is mutated in 24% of the cases. Importantly, downstream of oncogenic RAS, signalling to MEK and ERK occurs exclusively through CRAF. CRAF is regulated at the level of phosphorylation but until recently, controversy surrounded the identity of its upstream kinases. Here, a semi-high-throughput screen was developed in which NRAS mutated melanoma cells were screened using a human siRNA kinoma library in 96-well format. The rationale was to identify kinases that are required to activate CRAF using ERK phosphorylation as a readout. Primary hits were validated in secondary and tertiary screens by western blotting and finally by quantitative PCR to ensure knockdown of target proteins at the RNA level. On close examination, the majority of the hits were found to have their inactivating effects by down-regulating NRAS at both the level of protein and RNA rather than by depleting their specific targets. One kinase (casein kinase 2), which I had identified as a confirmed hit, was published during the course of this work. This study demonstrates how off-target effects can confound high-throughput siRNA screening campaigns. In melanoma cells with BRAF mutations, MEK, pan-RAF and BRAF inhibitors block ERK signalling. The second aspect of my project was to examine the effects of these inhibitors on signalling in cells with different genetic lesions. Intriguingly, I found that in cells which harbour oncogenic RAS, inhibition of BRAF drives BRAF into a complex with CRAF, resulting in BRAF hyperphosphorylation. Furthermore, under these conditions, CRAF is strongly activated, leading to an increase in ERK signalling. Notably, these effects are not seen when oncogenic BRAF is inhibited. When RAS mutant cells are treated with a pan-RAF inhibitor CRAF undergoes paradoxical activation whereby it is activated as a consequence of the drug binding to BRAF and complex formation, but is inhibited by direct binding of the pan-RAF inhibitor. Importantly, I also show that a kinase-dead mutant of BRAF constitutively binds to CRAF in RAS mutant cells and simulates ERK activation. Crucially, through collaboration with another student in the Marais laboratory, we show that the interaction I have elucidated causes accelerated tumourigenesis in mouse models. My data show that BRAF-selective inhibitors will only be effective in BRAF mutated cells and should not be administered to patients with RAS mutant tumours. This study demonstrates the importance of determining the underlying genetic cause of disease when selecting treatment options for patients.