In this thesis I aimed to provide base-line data to inform conservation of neotropical otters (Lontra longicaudis ) at both the range-wide and local (Mexico) scale. In Chapter 2, I compared three commonly used preservation methods for faecal DNA in order to identify the best method for neotropical otter faeces under challenging field conditions and long-term storage: 1) ambient-temperature drying, 2) a two-step protocol involving incubation in 95% ethanol and posterior silica desiccation, and 3) RNAlater. The results of this experiment showed that that RNAlater provides the highest mtDNA amplification success. In Chapter 3, I looked into the demographic history, genetic diversity and genetic structure of L. Longicaudis in Mexico using mtDNA. I found high genetic structure among North and South regions of the country, potentially due to geographic formations. Analyses of demographic history in Mexico indicated a recent expansion coinciding with the end of the Pleistocene. Given that recent evidence supports the existence of three subspecies of L. longicaudis across its range, I combined mtDNA haplotypes identified in this study with available Central and South American haplotypes in order to examine phylogeographic patterns; as a result, a distinct lineage distributed in North and Central America (NCAM) was identified. Due to the monophyly of this lineage, I propose to consider it a distinctive Evolutionary Significant Unit (ESU). In Chapter 4, I used landscape genetics to identify landscape features that affect otter geneflow in Mexico by means of microsatellites. I looked into the effect of elevation, slope, river networks and land cover on geneflow at a country-wide scale and two regional scales (North and South Pacific). I used Bayesian clustering to examine country-wide genetic structure. In terms of landscape genetics, elevation and slope were the only variables that explained genetic distance among individuals at the range-wide and North Pacific scale, respectively. The results of Bayesian clustering indicated two population clusters roughly distributed in the North and South of Mexico. The results of this thesis suggest that non-invasive methods can be applied to inform conservation efforts for Neotropical otters. I suggest that the NCAM lineage should be considered a distinct Evolutionary Significant Unit (ESU) throughout the range of L.longicaudis. Within Mexico, it is recommended to plan conservation corridors for the species where naturally low elevations and slopes allow genetic connectivity.

Since its discovery over a decade ago, DISC1 has become one of the most promising candidate genes for Schizophrenia and associated chronic mental disorders. This notion has been supported by a wealth of evidence from genetic and biochemical studies. With multiple interacting partners, DISC1 acts as a scaffold protein, orchestrating vital signalling pathways that underpin neurodevelopment and signalling. While the aetiology of Schizophrenia is poorly understood, loss of DISC1 protein function remains one of the proposed disease mechanisms. Furthermore, its tendency to form aggregates is reminiscent of neurodegenerative illnesses such as Alzheimer’s and Parkinson’s disease. C-terminal truncation of DISC1 (TrDISC1) is known to decrease neurite outgrowth and number in the PC12 cell line, abolish protein interaction with proteins such as Ndel1 and also disrupt vital physiological process such as mitochondrial transport. However, very little is known about the underlying disease mechanism at the molecular level. In order to gain insight in to the role of DISC1 pathway in Schizophrenia and associated mental illnesses, I studied novel post translational modifications of DISC1. The main conclusion of my thesis is that these modifications affect DISC1 turnover and its scaffold function. The work described in this thesis has uncovered 2 novel post translational modifications and identified the E3 ligase involved in regulating DISC1 turn over. My work has also laid the foundation for the design and discovery of both peptide and non-peptide, small molecule inhibitors of the DISC1 and its cognate E3 ligase interaction. These inhibitors can serve as both pharmacological tools and for further investigation of the role of this novel interaction in DISC1 pathway and the vital physiological functions it is involved in. Furthermore, this work also indicates the feasibility of controlled and directed differentiation of patient specific iPS cells in to neurons, which act as a useful tool for disease modelling.

Therapeutic options are currently limited for Neurofibromatosis type 1 (NF1) associated-malignant peripheral nerve sheath tumours (MPNSTs). MPNSTs are characteristically aggressive and the major cause of death in NF1 patients. Clinical trials using single drug agents to treat MPNSTs have so far been unsuccessful, which could be attributed to high levels of intra-tumoural molecular heterogeneity. To explore common cellular migratory and invasive signalling properties within the heterogeneous NF1-MPNST population, we utilised four different MPNST-derived cell lines, ST8814, S462, S1844.1 and S1507.2. MET has previously been shown to be elevated in MPNST cells and is thought to promote their cellular migration and invasion. Interestingly, we report variation in MET gene expression and protein levels in a variety of MPNST derived cell lines. MET inhibitors were effective at suppressing the migration and invasion of cell lines with elevated MET protein levels but not those without. Importantly, targeted inhibition of STAT3 suppressed cell migration, invasion and tumour formation in all cell lines tested, regardless of MET expression levels. Herein, we demonstrate that STAT3 functions as a common driver of tumourigenesis in multiple NF1-MPNST cell lines with varying signalling profiles. STAT3 is activated downstream of a variety of receptor tyrosine kinases which are associated with NF-1-tumourigenesis, including MET, IL-6 and EGFR, making it an appealing therapeutic target for the heterogeneous NF1-MPNST population. We also demonstrate that cellular migration, invasion and tumour formation through STAT3 is highly dependent on the transcription factor HIF-1α, where knockdown of HIF-1α ablated these oncogenic facets of STAT3. Our research demonstrates that aberrant signalling through STAT3 and HIF-1α drives tumour progression within MPNSTs, indicating that inhibition of the STAT3/HIF-1α /VEGF-A signalling axis could be a viable therapeutic strategy in this context.

Hepatocellular carcinomas carrying Axin1 mutations belong to a subset of tumours with an especially poor prognosis. Data obtained from an Axin1 mutant mouse line, challenged the traditional idea of Axin function; as simply a component of the β-Catenin-destruction complex. Axin1 deletion led to the development of highly proliferative HCC in the absence of an obvious Wnt/β-Catenin signature. In order to uncover the mechanism(s) leading to Axin dependent tumourigenesis, this study focused on the role of Axin in two systems. Firstly, we generated an allelic series of Axin mutant ES cell lines to analyse the role of Axin1 and 2 in ES cells. We could show, that single Axin mutants had a largely normal ES cell phenotype. In Axin double mutant ES cells, Wnt target gene expression was slightly upregulated, but cell proliferation stayed at normal levels. By contrast, upon differentiation into embryoid bodies, multiple readouts of the Wnt pathway were increased and a G2/M and cell cycle related gene expression profile was activated, accompanied by severe differentiation defects. In the second system, we developed Axin1 mutant 3D liver cultures, which allow fate tracing of Axin mutant and wt cells in real time. We could show tightly regulatable gene deletion in vitro and produced preliminary evidence that Axin1 loss in culture closely mimics the in vivo situation in respect to G2/M gene expression in the absence of Wnt activation. Overall, the effects of Axin loss on Wnt signalling and cell cycle regulation appeared to be tissue and cell cycle specific. Future use of the 3D culture system, together with the data obtained in Axin mutant ES cells and embryoid bodies, will not only advance our understanding of the involvement of Axin1 in hepatocellular carcinogenesis and cell cycle regulation; but may also be the starting point in the development of new therapeutic strategies.

Genomic analysis and DNA sequencing is a mature field with many established techniques. Developments in micro- and nanofabrication over the last decade or more have brought advances in a number of areas of chemical and biological analysis and genomics has been of particular interest. This thesis presents the development of two new techniques in microfluidic DNA manipulation that are directly applicable to the fabrication of next-generation genomic analysis devices. A key area for any miniaturised device is the ‘world-to-chip’ connection. Even the most highly integrated micro total analysis systems require efficient sample loading and for sensitive detection as well as device reuse it is essential that molecules of interest are not lost in the world to chip transition. A new type of microfluidic interconnect is described in this work, which uses a shallow slope to make an inplane connection between a microfluidic channel and a glass capillary. The microfluidic channel is fabricated in silicon capped with glass. The phenomenon of deep reactive-ion etch (DRIE) lag was applied to fabricate the slope and DRIE was also used to fabricate the microfluidic channels. The interconnect has been demonstrated to provide low loss loading of 48.5kbp DNA molecules from world to chip and is fabricated using standard silicon processing equipment. The second new development reported in this thesis is the design, fabrication and characterisation of microfluidic DNA ‘preconditioning’ channels which are used to improve the efficiency of a downstream taper in the channel used for fluidic DNA extension. Extension of DNA in elongational flow has the potential to become a high-throughput genomic mapping technology. A common limitation to all systems previously demonstrated for stretching DNA with an elongational flow is that the stretching efficacy is strongly dependent upon the initial conformation of the DNA strand. The devices described in this thesis demonstrate an improvement in DNA preconditioning using non-contact microfluidic shearing flow in order to deliver extended molecules to a microfluidic channel in which optical interrogation of fluorescent markers takes place using a bespoke confocal spectroscopy system.

The re-emergence of plague as a world-wide health concern and the potential risk posed by bioterrorism has led to an increased interest in available treatments for the disease. The bacterial DNA adenine-N6 methyltransferase, Dam, is involved in the regulation of a range of pathogenic bacteria and has been validated as a target for the development of antimicrobial agents with activity against Yersinia pestis, the causative agent of plague. The lack of a functionally similar enzyme in mammals suggests that highly selective Dam inhibitors could be developed. A coupled, real-time break light Dam activity assay has been optimised for HTS, and assays for the validation and characterisation of screening hits have also been developed. Screening of random and in silico enriched compound libraries, and the subsequent application of counter-screening and hit confirmation assays, resulted in the identification of a single viable lead, (4-(N-(2-hydroxyethyl)sulfamoyl)phenyl) stibonic acid (13776). Screening of compounds analogous to 13776 identified a series of arylstibonic acids with activity against Dam. Kinetic characterisation of the most potent arylstibonic acid, 4-stibonobenzenesulfonic acid (13746), revealed a DNA-competitive mode of action, and a Ki of 6.46 ± 0.07 nM. However, selectivity assays have revealed a potentially non-specific mode of action for the stibonic acids, which have shown activity against a range of DNA and protein binding enzymes. Yersinia cell culture experiments have shown a single compound, (3-((2-hydroxyethyl)carbamoyl)phenyl)stibonic acid (13782), to be capable of penetrating Yersinia cells and partially inhibiting methylation, and mRNA profiling experiments have shown 13782 to induce a statistically significant change in several genes involved in the pathogenicity of Y. pestis. Attempts at resynthesising 13782 have proved challenging, with only a fraction of the activity of the original sample reproduced. HPLC analysis of the original and resynthesised samples has shown the former to comprise two components, with only one present in both samples. The in vitro evaluation of a series of bisubstrate analogues designed to mimic both the methyl donor S-adenosylmethionine (AdoMet), and the methylation target (adenine) has shown that substitution of the AdoMet sulfur for nitrogen results in a significant but not total loss of activity. Furthermore, the addition of a bicyclic heteroaromatic adenine analogue mimic to this scaffold led to an increase in potency and selectivity for Dam over the human cytosine methyltransferase DNMT1 but a reduction in selectivity for Dam over the restriction enzyme DpnI. These results suggest that a selective and potent Dam inhibitor can be obtained by carefully modifying both components of the bisubstrate analogue inhibitor.

Background: Access to genetic cancer risk information can be highly dependent on whether familial risks are discussed within the family. Despite its essential role in ensuring family members have access to genetic services, there are a number of gaps in the knowledge available on people’s experiences regarding talking to their relatives about genetic testing for BRCA1/2 and its risk implications. In particular, research to date has focused far more on with whom and why (motivations) family communication regarding genetic testing occurs, rather than when or how it is occurring. Method: The study is qualitative in nature, employing in-depth interviews and constructing eco-maps as a method of identifying relevant family members and guiding the researcher through the family structure and relationships. These methods were chosen in line with an interpretive description methodology to ensure depth and richness in analysis and reporting of findings. Results: The Key Findings are as follows: 1. Communication between emotionally close relatives is different to communication with emotionally distant relatives; with emotionally close family and friends it is about sharing and supporting; whereas with emotionally distant family it is about gaining and imparting information. 2. A family’s engagement in communication regarding genetic testing is implicitly linked to their experiences of cancer burden, and how openly this is discussed in the family. 3. There is a lack of understanding of risks to men and their offspring based on perceptions of hereditary breast and ovarian cancer being a female disease. 4. Emotionally distant and male relatives are only contacted selectively. Those undergoing genetic testing for BRCA1/2 are not good at identifying all at-risk family members in order to share the implications of the genetic test with them. 5. As far as the family are concerned, members do not have the right to make an informed decision to decline. 6. Plans for telling people in the future, especially children, is a cause of worry and concern for those undergoing testing and needs further support, especially in the longer term. Conclusions: Developing interventions to help manage problems associated with family communication regarding genetic testing for cancer risk should be a top research priority, especially as the numbers of people affected by these issues is set to rise as more genes are discovered. The longitudinal view identified gives deep insight into how and when genetic testing for BRCA1/2 are discussed within these families, allowing future interventions to be targeted where they are most helpful.

Many chemotherapy drugs act by inducing DNA damage leading to cell death, and the platinum analogue class of anticancer drugs are the most commonly used DNA damaging chemotherapeutic drugs. Despite extensive analysis of platinum-DNA interactions, particularly characterising the individual adducts and their effects on DNA replication, transcription and cell survival, measurement of these adducts in cells with higher sensitivity and precision is necessary. Previous work studying platinum-DNA adduct formation has been performed using DNA damage assays such as immuno-slot-blots to detect whole genome DNA damage, or with combinations of chromatography and mass spectrometry to characterise each adduct individually. The ability to measure platinum-induced DNA damage genome-wide with high resolution in human cells could have profound implications for basic mechanistic research, as well as clinical translational research and treatment stratification, by providing a tool with the potential for predicting clinical response to these agents. The achievement of this PhD was developing an assay to measure platinuminduced DNA damage induction at high resolution, density and precision within the genome of human cells. This was achieved using DNA immunoprecipitation coupled with analysis using DNA microarrays, allowing measurements of platinum-induced DNA damage to be made at high resolution throughout the human genome for the first time. This assay was initially developed to measure cisplatin and oxaliplatin induced DNA damage in the genome of the yeast model organism Saccharomyces cerevisiae and experimental profiles of cisplatin and oxaliplatin-induced DNA damage were validated by demonstrating close correlation with mathematically generated predicted profiles for platinum-induced DNA damage. The assay was then applied and validated to measure cisplatin, oxaliplatin and ultraviolet-induced CPD formation in human fibroblast cells, and again, experimental profiles of cisplatin and oxaliplatin-induced DNA damage and UV-induced CPD formation were shown to correlate well with predicted profiles of DNA damage. Novel comparative analytical approaches for studying microarray data from these genome-wide DNA damage datasets are demonstrated and further validation of the assay is provided by demonstrating the contrast between platinum-induced DNA damage and UV-induced CPD formation genome-wide and in the context of repeat sequences of DNA. Finally, cisplatin and oxaliplatin-induced histone H3 acetylation changes are examined and histone H3 K14 acetylation is demonstrated to be a prominent histone modification following exposure to platinum analogues. Novel analysis is performed to investigate the influence of chromatin on platinum adduct formation, and greater platinum-induced DNA damage is demonstrated in DNA samples treated in vitro compared with DNA samples taken from cells treated in culture. Comparisons were also performed between histone H3 acetylation in yeast cells following exposure to cisplatin or UV irradiation and this comparison revealed very similar patterns of histone acetylation following exposure to these two different genotoxins.

Single nucleotide polymorphisms (SNPs) are one of the forensic markers used to resolve the problem of DNA typing from degraded samples. It has been found in previous studies that when profiling heavily degraded forensic samples the small amplicon required for SNP analysis has an advantage over the larger STR loci, which are routinely used in forensic case work. A total of 66 SNPs from the non-coding region of the 22 pairs of autosomal chromosomes were identified and SNP assays developed. Instead of selecting the SNPs from the available GenBank® sites, SNPs were typed from Arab individuals from Kuwait and United Arab Emirates (UAE) to identify polymorphic SNPs. In order to obtain SNP data from Arab populations, a total of 10 unrelated Arab individuals from Kuwait and UAE were typed. The Affymetrix GeneChip® Mapping 250 K Array Sty І was employed to generate profiles for approximately 238,000 SNPs. Only autosomal SNPs were selected from the data. Following selection, allele frequencies were estimated using the SNaPshot™ technique (Applied Biosystems) with 25 UAE individuals. For this technique, PCR forward and reverse primers were designed to generate PCR products less than 150 bp. The single base extension primers were designed to hybridise 1 bp upstream from the target SNP. SNP characterization, including Hardy¬Weinberg equilibrium and pair wise linkage disequilibrium, was carried out using the software package Arlequin v 3.1. Allele frequencies were calculated using Excel spreadsheets. PowerStats v.12 software used for discrimination power and match probability estimation. All the 66 SNPs were polymorphic with average heterozygosity levels of 47%. A high heterozygosity level is very valuable for forensic application improving the individualization of forensic samples (Vallone et al. 2005). The probability that two individuals having identical genotype profile was found to be very low, 3.058 x 10-25. The combined power of discrimination was found to be 0.999999999. This indicated that the selected SNPs met the parameters needed for forensic application. The SNPs genotype sensitivity gave profiles from minute amounts of DNA template as little as 100 pico grams (pg) and optimal and reproducible results at 300 pg of DNA template. The profiling of DNA from forensic samples is not always possible. This can be due to insufficient amount of samples being recovered and in many cases, DNA degradation. Biological materials that are recovered from the scene of the crime have often been exposed to sub-optimal environmental conditions such as high temperature and humidity. SNPs performance on degraded samples was tested on artificially degraded saliva and semen samples. Controlled temperature and humidity experiments were performed to study the effect of these environmental factors on the samples. Also uncontrolled experiments on samples being subjected to different weather conditions (UK summer and UAE winter and summer) was performed in order to study and compare both weather effects on saliva samples. The triplex sets of SNPs that were developed for such study showed full allele profiles when compared to STRs, the current method used in forensic labs. In addition, SNPs produced a higher success rate than STRs when tested with samples obtained from human teeth remains and on samples subjected to DNase 1 digestion. The small size of SNPs, between 90 and 147 base pair (bp), showed more resistance to degradation than the STRs size ranging between 100 and 360 bp. This study demonstrated that the 66 SNPs selected are useful markers when the typing of degraded samples by STRs fails to produce complete or partial profiles.

Parkinson’s disease (PD) is the most common neurodegenerative movement disorder, currently affecting 6.3 million people worldwide. Although it is associated, in the longterm, with severe complications (dyskinesias), L-DOPA remains the gold standardtreatment. An alternative approach to the treatment of PD is the replacement of the lost striatal dopaminergic innervation by transplantation of foetal ventral mesencephalon (VM) dopaminergic precursor cells. Opened trials have provided the proof of concept that intrastriatal VM transplant can survive, integrate and in some cases, restore motor functions. Nevertheless, later double blind studies reported inconsistent benefit of the therapy and the development of dyskinesias remaining after withdrawal of L-DOPA medication. The failure of the animal models in predicting these problems raises concern about their reliability. Therefore, the global aim of this PhD work was to identify some of the critical factors that can influence the functional outcome of cell therapy for PD, and on the basis of this, to develop an improved 6-OHDA unilaterally lesioned rat model for transplantation. The first step was to determine the most reliable method to assess dyskinesias in rats. The second part of this thesis was set out to determine the effect that chronic L-DOPA treatment, administered at different time could had on the survival and function of immunologically incompatible foetal VM transplant. The results demonstrated that L-DOPA administered chronically post-grafting increases the host immune response around the xenogeneic transplant. Therefore, the last set of experiments were designed to create a model of mixed donors graft to better reproduce the patient situation, where each transplant required up to 8 donors from unknown immunological background. All of these experiments come together to help to develop a rat model that more accurately represents all aspects of patients undergoing transplantation for PD.