Due to its role in regulation of mitochondrial function, PGC1α is emerging as an important player in ageing and neurodegenerative disorders. PGC1α exerts its neuroprotective effects by promoting mitochondrial biogenesis (MB) and functioning. However, the precise regulatory role of PGC1α in the control of mitochondrial dynamics (MD) and neurotoxicity is still unknown. Here we elucidate the role of PGC1α ?in vitro and in vivo in the regulatory context of MB and MD in response to lead (II) acetate as a relevant model of neurotoxicity. We show that there is an adaptive response (AR) to lead, orchestrated by the BAP31-calcium signalling system operating between the ER and mitochondria. We find that this hormetic response is controlled by a cell-tolerated increase of PGC1α expression, which in turn induces a balanced expression of fusion/fission genes by binding to their promoters and implying its direct role in regulation of MD. However, dysregulation of PGC1α expression through either stable downregulation or overexpression, renders cells more susceptible to lead insult leading to mitochondrial fragmentation and cell death. Our data provide novel evidence that PGC1α expression is a key regulator of MD and the maintenance of tolerated PGC1α expression may offer a promising strategy for neuroprotective therapies.

Mumps virus is the causative agent of mumps disease, of which humans are the only natural host. Mumps infection is vaccine-preventable. Mumps vaccines are live viruses attenuated by serial passage on cell substrates or in embryonated eggs. They are known to vary in their effectiveness, degree of attenuation and adverse event profile. The Jeryl Lynn mumps virus is the most commonly used vaccine strain. Two related commercial mumps vaccines, which originate from the same attenuated strain (Jeryl Lynn-5) but have different efficacies, were investigated. Jeryl Lynn-Canine Kidney (JL-CK), produced on primary canine kidney cells is less effective than RIT 4385 which is produced on chicken embryo fibroblasts. JL-CK and RIT 4385 could be distinguished by a number of in vitro and in vivo properties. JL-CK produced heterogeneous, generally smaller plaques than RIT 4385, but gave 100 fold higher viral titres when grown on Vero or MDCK cells and produced a higher degree of hydrocephalus in the neonatal rat brain. Sequencing of the JL-CK virus identified 14 regions of heterogeneity throughout the genome. Plaque isolation demonstrated the presence of five mumps virus variants encompassing these mutations in the JL-CK vaccine. One mutation was associated with a small plaque phenotype, the effects of the others either in vitro or in vivo were less clear. Only 4 % of the JL-CK vaccine population corresponded with the parent Jeryl Lynn-5 strain. Deep sequencing of JL-CK and virus replicating in cell lines and neonatal rat brains showed that propagation in vitro or in vivo altered the population dramatically. The data presented in this study suggest that growth of JL-CK in primary canine kidney cells resulted in the selection of a mixture of mumps virus variants that have different biological properties compared to the parent Jeryl Lynn-5 virus.

Atrial fibrillation (AF) is the commonest disorder of cardiac rhythm. Following coronary artery surgery approximately 1 in 3 patients will develop de novo post-operative AF (POAF) leading not only to prolonged hospital stay but also to increased morbidity and mortality. The pathophysiology of this arrhythmia is highly complex, and combines factors such as ion channel dysfunction, inflammation, oxidative stress, fibrosis and autonomic dysfunction that through electrical and structural remodelling promote both triggering and maintenance of this arrhythmia. For many years POAF has been regarded as a reactive phenomenon, occurring in response to post surgical inflammatory stressors and electrolyte imbalance. However, it is also possible that in a proportion of patients, prior cardiomyocyte remodelling predisposes to atrial arrhythmogenesis when exposed to surgical stress. Understanding the genomic and metabolic mechanisms that underlie this substrate may not only provide novel diagnostic biomarkers to identify at risk patients, but also isolate previously unrecognised therapeutic targets for prevention and treatment. In this work, a clinical observational study was utilised to obtain microRNA, gene expression and metabolic profiles of patients developing POAF after coronary artery bypass graft (CABG) surgery. Based on these results, a network of genomic and subsequent downstream pathway interactions was established to characterise the atrial substrate of post-operative AF. Furthermore, analysis of pre-operative blood was performed in order to identify novel microRNA that may provide a platform for biomarker development. Finally, the metabolic signature of the atrial myocardium and its relationship with the surrounding epicardial adipose were investigated to complete a comprehensive overview of the central mechanisms thought to underlie POAF pathogenesis. This work demonstrates that prior to surgery and the onset of arrhythmia, several distinct genomic and post-translational characteristics are evident in the both the myocardium and circulating blood of patients going on to develop POAF. Analysis of right atrial biopsies highlights a characteristic microRNA profile associated with POAF, and identifies target genes regulating intracellular signalling pathways, leukocyte recruitment, and ion channel remodelling. Furthermore, selected gene expression analysis demonstrates a de-differentiated phenotype similar to that seen in chronic AF, whilst at the same time establishing that disordered cardiomyocyte calcium handling is apparent at the time of surgery. Finally, analysis of the pre-operative circulating blood serum highlights microRNA selectively upregulated in POAF and establishes the potential for future biomarker development. In summary, the results presented here support the presence of a pre-existing atrial substrate in POAF, suggesting the potential exists for high-risk patients to be identified prior to surgery and the onset of arrhythmia. Furthermore, for the first time a number of similarities have been made apparent between post-operative and chronic AF, implying a common mechanistic spectrum of structural and electrical remodelling. As a consequence, the results presented in this thesis have not only improved our understanding of the complex interplay of factors leading to the pathogenesis of AF, but also provide a platform for both the development of a unique clinical biomarker and the identification of novel therapeutic targets.

The cohesin complex is essential for proper sister chromatid segregation during cell division and post-replicative DNA repair. Cohesin is also important for the regulation of gene expression. However, the mechanisms by which cohesin impacts gene expression remain incompletely understood. Owing to its vital role in cell division and DNA repair, loss of cohesin can indirectly impact gene expression programme in cycling cells. Thus, in order to investigate cohesin's role in gene regulation, a conditional knockout system was used which allowed rapid depletion of the cohesin subunit RAD21 and avoided secondary stress-induced effects on gene expression. Acute depletion of cohesin in mouse embryonic stem (ES) cells did not lead to a global collapse in pluripotency. Instead, the impact of cohesin depletion was limited to about 600 genes and was locus-specific in terms of direction of deregulation. A subset of deregulated genes was selected based on the positioning of cis-regulatory elements and relevance to the pluripotent state and the role of cohesin in mediating long-range interactions was analysed using chromosome conformation capture (3C). Interestingly, cohesin binding, DNA looping and transcriptional changes were not always correlated. At some of the loci tested, these interactions were maintained after removal of cohesin, questioning models where cohesin regulates gene expression solely by mediating long-range interactions. One of the pluripotency factors affected by cohesin depletion in ES cells was Myc. Experiments analysing the expression of Myc showed that it was post-transcriptionally upregulated, specifically in cohesin deficient ES cells growing in defined media supplemented with ERK and GSK3β inhibitors (2i media). Further investigation revealed that contrary to the previously reported downregulation of Myc upon cohesin depletion, cohesin was not essential for Myc expression in various cell types. In separate experiments, I investigated if cohesin was required for the transcriptional activation of a silent gene in response to extracellular stimuli. Results from IFNγ induction of cohesin deficient non-cycling mouse embryonic fibroblasts (MEFs) showed that cohesin was important for the activation of MHC class II genes and their master regulator Ciita. The expression of MHC class I genes and the associated regulatory factors remained unaffected by cohesin depletion. Further evidence is provided for the involvement of cohesin in regulating transcription by modulating RNA polymerase processivity and through the action of PTIP subunit of the MLL complexes. Altogether my work gives an insight into the role of cohesin in mediating long-range DNA interactions important for regulation of gene expression and explores novel mechanisms of gene activation by cohesin.

Children with Down syndrome (DS; trisomy 21, T21) have a 500-fold risk of developing acute megakaryoblastic leukaemia (AMKL) in early childhood compared to children without DS. DS-AMKL is preceded by a transient neonatal leukaemia, unique to DS, caused by acquired mutations in the GATA1 gene in fetal liver (FL) haematopoietic stem/progenitor cells (HSPC). Previous work has shown that T21 itself perturbs FL haematopoiesis prior to acquisition of GATA1 mutations providing a cellular substrate for leukaemic transformation. Similar changes are not reported in fetal bone marrow (FBM). The mechanism(s) that perturb FL haematopoiesis in DS and the role of the unique fetal microenvironment in DS are unknown. My project investigated the role of the environment in abnormal fetal haematopoiesis in DS using both an unbiased approach to evaluate the global gene expression by fetal stromal cells and a candidate pathway approach, investigating the specific role of the insulin-like growth factor (IGF) signalling pathway, a key regulator of cell proliferation previously linked to AMKL development in DS. To examine the role of the IGF system in fetal haematopoiesis, I measured expression of the IGF signalling proteins in stromal cells. In comparison with adult BM MSC, fetal MSC expressed marked differences in components of the IGF signalling pathway. In particular, IGF1 was selectively expressed by adult BM MSC whilst fetal MSC expressed IGF2 with very little IGF1. I next assessed the expression of IGF receptors (R) on FL haematopoietic progenitors. Both normal and DS FL CD34+ cells expressed high levels of IGF1R, confirming potential for IGF responsiveness, while IGF2R was slightly reduced in DS. IGF2 promoted the growth of DS FL megakaryocyte colonies in serum-free clonogenic assays and of megakaryocytes in liquid culture, implicating the IGF pathway in the perturbation of DS fetal haematopoiesis. To further investigate the DS fetal microenvironment, I derived mesenchymal stromal cells (MSCs) from normal and DS FL and FBM and performed gene expression profiling using microarray. Analysis identified marked differences between normal FL and FBM and between normal and DS populations. In particular, several genes which encode secreted proteins, including IGFBP1, showed differential expression in DS, suggesting that they may play a role in the abnormal haematopoiesis. Furthermore, there were remarkably few differences between DS FL and DS FBM MSC indicating strong similarities in the transcriptomes of these two microenvironments. Finally, I established a co-culture system to assess the effects of primary MSC on the differentiation of primary FL HSPC. Intriguingly, preliminary experiments indicate that DS MSC, but not normal FL MSC, promote erythroid differentiation of normal FL HSPC supporting the hypothesis that differences in the transcriptome of DS fetal liver MSC are functionally important. This finding provides strong evidence that the FL microenvironment is likely to play a significant role in the disruption of fetal liver haematopoiesis in DS.

This thesis evaluated the clinical utility of cardiopulmonary computed tomography (CT) in patients presenting with chest pain and dyspnoea. Studies within this thesis confirmed the following. Firstly, there is a requirement for improved diagnostic pathways to minimise patients being discharged without a diagnosis, which currently occurs in 30-40% of patients admitted with chest pain and dyspnoea. Historically, CT has been utilised in 32% of admissions with chest pain and 10% of admissions with dyspnoea. Secondly, challenges exist to the wider adoption of cardiopulmonary CT. These include patient-related factors, institutional capabilities and guideline restrictions. In acute admissions, 11% of patients with dyspnoea and 7% of patients with chest pain and a low to moderate likelihood of CAD are suitable for CT. In the RACPC setting, including patients across the entire spectrum of CAD likelihood, 18% of patients are suitable for CT. NICE CG95 would recommend only 1% of acute chest pain admissions and 2% of RACPC attenders for CT. Thirdly, NICE CG95 would recommend 51% of acute chest pain admissions and 66% of RACPC attenders for discharge without cardiac investigation. In the RACPC population, significant CAD is identified in 10% of these patients and a major adverse cardiac event in 2%. Fourthly, in selected patients with suspected cardiac chest pain, cardiac CT has a diagnostic yield of 21% in acute admissions and 13% in RACPC attenders for significant CAD. In acute admissions with dyspnoea, cardiopulmonary CT has a diagnostic yield of 20% for CAD, 20% for pulmonary embolism, nil for aortic dissection and 89% for non-vascular chest pathology. Fifthly, inclusion of CT in diagnostic pathways for chest pain result in fewer patients discharged without a diagnosis, fewer invasive angiography procedures and reduced diagnostic costs. In patients with dyspnoea, CT provides value to clinicians making diagnoses and supports early discharge without detrimental outcomes.

Introduction: Coronary Wave Intensity Analysis (WIA) provides an invasive measure of energy transfer within the coronary circulation. I set out to derive a non-invasive measure of the backward expansion wave (BEW) responsible for coronary flow and assess it during exercise and in aortic stenosis (AS). Methods: 17 patients (mean age 60, 11 male) with normal cardiac function underwent invasive LAD WIA calculation using a pressure- and flow-tipped wire. Non-invasive WIA was calculated immediately after angiography from simultaneous PW Doppler of the LAD and a suprasystolic-cuff derived measure of central pressure. Non-invasive WIA was then assessed in 9 healthy volunteers whilst exercising on an exercise bike, 25 patients with varying degrees of AS (AVmax range: 2.41-5.43m/s) and 29 patients before, after and at 6 and 12 months following aortic valve intervention for severe AS. Results: Mean peak BEW was -14.7 ± 8.7x104 Wm-2s-2 invasively and -14.4 ± 8.2 Wm-2s-2 non-invasively and increased with exercise (at peak: -20.5±6.8Wm-2s-2, p=0.02) along with a rise in coronary flow (28.8cm/s to 42.1cm/s, p 0.06). A significant correlation was noted with the BEW and AS severity, strongest when valvulo-arterial impedence was assessed (r=-0.66, p<0.001). In severe AS, a reduction in coronary flow (0.41 to 0.33m/s, p<0.01) and the BEW (-22.1 vs 10.9x104Wm-2s-2, p<0.01) was seen after intervention. With LVH regression BEW increased (-21.6±12.6x104 Wm-2s-2 at 6 months) without a significant change in coronary flow. Conclusion: It is possible to construct a non-invasive measure of coronary WIA thus markedly increasing its applicability. Using this technique, the BEW is seen to increase during progressive levels of exercise accounting for the increase in coronary flow. The BEW progressively climbs with increasing AS, falls to sub-normal levels after aortic valve intervention but then increases to normal levels with LVH regression.

Background: The relationship between living in rural areas and the acquisition of protective environmental factors against the development of asthma and atopy gave rise to the hygiene hypothesis. Between the environmental factors attributed to asthma, particular airways microbiota patterns have been encountered in adult asthmatics using molecular independent techniques (Hilty et al., 2010) and in children using conventional culture methods (Bisgaard et al., 2007). Here a retrospective time series study has been performed using samples from infants at different ages to study the microbiome variations in a population with very low antibiotic use history and no corticosteroid usage. Methods: Pyrosequencing of amplicons of the bacterial 16S rRNA gene was performed from oropharyngeal samples from 134 infants with episodic wheezing versus 200 healthy infants (total of 334 infants examined) sampled at different ages (7, 12 and 24 months). Bioinformatic analyses were conducted using QIIME 1.7 software and Phyloseq package on R. Additionally a new culture-independent pyrosequencing approach using the map gene was successfully developed to enable discrimination of streptococci at species level. Results: Significant abundance differences between infants with wheezing history and healthy controls were found for the Fusobacteria, Proteobacteria and Actinobacteria phyla. At genera level a significant increase in potential pathogenic bacteria (Neisseriacea, Haemophilus, Staphylococcus) was found in wheezers whilst a higher prevalence of Veillonella spp. was seen in controls. In addition, using map gene pyrosequencing, Streptococcus salivarious was found to be statistically significantly related with wheezing syndrome whilst Streptococcus mitis was more prevalent in controls. When age was considered differences were found in the microbiota displayed as species numbers increased (alpha diversity). Conclusions: The respiratory microbiota is different at phyla, genera and Operational Taxonomic Unit levels when comparing between wheezing and healthy children. A progressive more complex respiratory microbial community develops with age.

Many patients with emphysema remain breathless despite optimal medical therapy. Non-pharmacological approaches to reduce the volume of hyperinflated lungs include lung volume reduction surgery (LVRS) which is effective in selected patients with upper lobe predominant emphysema and low exercise capacity. Bronchoscopic techniques to reduce lung volume are also being developed. Studies of two bronchoscopic techniques to achieve lung volume reduction (LVR) are presented in this thesis; LVR coils (LVRCs) and endobronchial autologous blood instillation. In a trial of LVRCs we demonstrate for the first time in a randomised controlled setting, that treatment with LVRCs results in statistically and clinically meaningful improvements in quality of life, lung function and exercise capacity compared with controls, and that benefits are maintained up to 12 months following treatment compared to baseline. In two pilot studies, we used autologous blood instilled endobronchially aiming to achieve lung volume reduction by inducing parenchymal scarring and fibrosis. Instilling 180-240 mls of autologous blood withdrawn from patients during the bronchoscopic procedure directly into a giant bullae resulted in significant reduction in bulla size over subsequent months in three of five patients, with associated improvements in lung function, exercise capacity and quality of life. However a randomised controlled trial of instilling 60 mls of autologous blood into three segments of one lobe in patients with heterogeneous emphysema was ineffective. In addition, I investigated the use of a novel 3-dimentional measurement system, optoelectronic plethysmography (OEP), to track abdominal and chest wall movements during respiration. This showed that successful lung volume reduction approaches were associated with significant improvements in lower rib cage paradoxical inspiratory movements after lung volume reduction. Improvements in chest wall asynchrony were larger the worse the asynchrony was at baseline, and those with larger improvements in asynchrony derived greater benefits in lung function and other clinical outcomes following LVR.

Accumulation of ectopic fat in liver and pancreas are seen as key indicators of the metabolic syndrome affecting over a third of the population in the western world. Deposition of these ectopic fat depots is linked to insulin resistance, a major risk factor in the development of type 2 diabetes. Quantifying ectopic fat deposition, without the use of biopsy, has always been a challenge. Magnetic resonance (MR) techniques have emerged as the gold-standard for estimating overall fat deposition. However, little is known on the regional variation of fat deposition in liver and pancreas in human subjects. The aim of this project was to determine, using an in vivo Multi-Echo MR imaging technique, regional variations in ectopic fat deposition in liver and pancreas and assess the potential impact of environmental factors, including BMI, age, and gender. The overall ranges in liver and pancreas fat ranged from 0.69- 32.27 % and 0.40 - 24.18% respectively. There was a strong correlation between liver fat, body weight, waist-to-hip ratio and waist circumferences. A significant correlation between pancreatic fat and waist-to-hip ratio, particularly in men, was also observed. The relationship between liver and pancreatic fat was surprisingly weak, suggesting different mechanism of accumulation. Regional variations in liver and pancreatic fat were assessed by using surgical segmentations as anatomical markers, resulting in 8 anatomical segments for the liver and 3 for the pancreas. Intra-abdominal adipose tissue and age had the largest effect on liver and pancreatic fat deposition. Relative physical activity influenced the overall levels of liver ectopic fat content but did not significantly affect regional distribution in the pancreas. No gender differences in fat distribution in the liver were detected but a strong correlation with pancreas and men was detected. Life-style interventions, such as weight loss programs, showed significant reduction in overall liver and pancreatic fat content. As for bariatric interventions, more significant reduction of overall ectopic fat in liver was seen with an increase in total and regional pancreas fat content. In conclusion, I have shown, utilising an anatomical compartmentalisation of the liver and pancreas, that there are significant heterogeneity in fat deposition in the pancreas, and that different regions respond variably to lifestyle interventions. There is a more homogeneous fat deposition in the liver, which is more consistently modulated in response to lifestyle changes.