INTRODUCTION: Crohn's disease (CD) is a, relapsing and remitting inflammatory disorder of the gastrointestinal tract. The pathogenesis of CD is not entirely understood, but may represent a combination of immune, genetic, and environmental stressors. Recently, a number of genes have been discovered that confer susceptibility to CD, many of these functioning in the innate immune system and novel therapies that act on molecules associated with the innate immune system are being developed. The anti-tumour necrosis factor (TNF) therapies, infliximab and adalimumab, are two such treatments, however data relating to the longer-term efficacy and safety of these therapies in CD, along with their cost implications, are limited. NLRP3 is a pathogen recognition receptor which, amongst other ligands, recognises muramyl dipeptide (MDP), the major ligand sensed by NOD2, the first susceptibility gene identified in CD. The NLRP3 inflammasome has recently been associated with inflammation in rheumatoid arthritis (RA) and has been shown to be modulated following infliximab therapy in this condition. AIMS: Firstly, to determine the long-term efficacy, safety and cost implications of anti-TNF therapies in CD, and also to assess the efficacy of switching to a second anti-TNF upon failure or non-response to the initial anti-TNF treatment. Secondly, to investigate the NLRP3 inflammasome and other associated inflammatory molecules in patients with CD at both gene and protein level to analyse if the NLRP3 inflammasome is modulated in CD patients compared to healthy controls, and to determine if the inflammasome is modified following treatment with the anti-TNF therapy, infliximab.

Mumps virus (MuV) is an enveloped, non-segmented, negative-sense, RNA virus, genus Paramyxoviridae. MuV causes an acute, systemic infection in humans characterised by parotitis, orchitis and aseptic meningitis. Despite widespread vaccination, outbreaks continue to occur. The biology of MuV infection and pathogenesis in humans remains poorly understood, in part due to a lack of physiologically-relevant virus and host cell model systems. Of particular interest is how MuV interacts with epithelial cells during infection. Therefore, a reverse genetics system was generated based on the consensus genomic sequence of unpassaged, clinical material from a genotype G MuV infection from the New York 2009 outbreak. The resulting viruses were referred to as recombinant (r) MuVG09. Establishment of this reverse genetics system allowed the generation of rMuVG09 expressing the enhanced green fluorescent protein (EGFP) in the third position, rMuVG09EGFP(3) to allow real -time tracking of virus replication in vitro. Surprisingly, recovered viruses had mutations, suggesting genetic instability of MuVG09 during growth in vitro. However, using this system, it was demonstrated that MuV could infect transformed and primary polarised epithelial cells from the apical and basolateral surfaces. Infection of differentiated, normal, human, bronchial, epithelial (dNHBE) cells showed that MuV could infect ciliated and non-ciliated, non-goblet cells. MuV was released from the apical and basolateral surfaces. In dNHBE cells this was associated with cell -to-cell spread and cell -to-cell fusion. MuV cell -to-cell fusion was controlled by haemagglutinin-neuraminidase (HN) and fusion (F) glycoproteins. Mutation of a single amino acid (I278T) in the F glycoprotein ablated cell -to-cell fusion while preserving virus-to-cell fusion, which was associated with increased virion stability. Analysis of growth kinetics in vitro showed that the non-fusogenic virus was growth restricted in interferon (IFN) competent cell lines. This was correlated with an increase in interferon stimulated gene (ISG) expression during infection, suggesting a role for fusogenicity in evading the human IFN response.

The common liver fluke, Fasciola hepatica, infects both human and animals causing fasciolosis. The World Health Organization estimates that ~2 .4 million people are infected with fasciolosis and another 180 million are at risk of being infected. It is thought to be the most economically damaging livestock disease compared to any other helminth parasite diseases. An important influence on the biochemistry of the parasite is its changing environment throughout its life cycle, requiring its metabolism to move from aerobic to anaerobic as it matures. The fluke's metabolic adaptations permit energy extraction to occur even in an anaerobic environment using a variation on the CA cycle. is thesis explores five enzymes from Fasciola hepatica concerned witl~ energy production - citrate synthase (FhCS), triose-phosphate isomerase (FhTPI), glyceraldehyde-3-phosphate dehydrogenase (FhG3PDH), and galactokinase (FhGALK), and UDP-galactose 4-epimerase (FhGALE). FhCS was cloned and sequenced; however, expression and purification proved difficult. Nevertheless, extracts from bacterial cells expressing the protein showed additional CS activity. FhTPI, FhG3PDH, FhGALK and FhGALE have been successfully cloned, sequenced, expressed, purified and characterized. TPI from the related helminth, Schistosoma mansoni, (SmTPI) was also characterised in a comparative study. FhTPI and SmTPI show remarkable thermal and proteolytic stability. FhG3PDH shows increased stability in the presences of substrate; furthermore the addition of ligands results in the change of the enzymes conformation and oligomierc state. FhGALK shows similarity to both galactokinase and N-acetylgalactoasmine from other organisms; however no activity was displayed with either of heir substrates. FhGALE has been characterized kinetically; early inhibition studies have identified two compounds (found through an in silico screen) which show selectivity for the fluke enzyme over the human one. these biochemical studies, combined with molecular modelling, were used to assess whether these metabolic enzymes from F. hepatica may make plausible targets for the development of novel anthelmintics.

Obligate anaerobic bacteria such as Prevotella are present in large numbers in the lower airways of Cystic Fibrosis (CF) patients. However, their contribution to the development and/or perpetuation of bronchiectasis is unknown. In CF, unopposed Matrix Metalloproteinase (MMP) activity drives loss of extracellular matrix proteins, leading to loss of elastic tissue and cartilage destruction and its high activity reported in CF sputum inversely correlates with lung function (FEV1). In our lab it has been previously shown that Pseudomonas aeruginosa (PA01) drives unopposed production of MMP-9 in infected leukocytes and can further drive unopposed MMP-9 by non CF and CF cell lines regulated by mitogen activated protein kinase (MAPK) dependent mechanisms via an inflammatory-epithelial cell network. PA01 also degraded MMP-9's natural in vivo inhibitor, Tissue Inhibitor of Metalloproteinase-1 (TIMP-1) in vitro. Prevotella have been implicated in driving net-MMP-dependent proteolytic activity and tissue destruction in periodontal disease, however their role in CF has yet to be determined. Direct Prevotella infection of mononuclear cells, drove heightened pr?-inflammatory responses coupled with an increase in MMP-9 and TIMP-1. These effects were also observed when Prevotella-infected mononuclear conditioned media was used to stimulate epithelial cell lines, CF and non-CF via an inflammatory-epithelial cell network. Direct mononuclear infection of Prevotella was shown to activate MAPK and NF-KB signalling pathways. In addition Prevotella were shown to destroy TIMP-1 protein through secretion of a cysteine protease, which may suggest an indirect contribution to net proteolytic activity. In summary, Prevotella may contribute to the pro-inflammatory cascade in CF but may not directly drive net MMP-dependent proteolytic activity through cellular mechanisms. However it may favour a proteolytic phenotype by the degradation of TIMP-1, indirectly contributing to tissue destruction in CF.

Coeliac disease is a common small intestinal disorder affecting approximately 1% of the Northern European population. It is caused by an inappropriate immune response to dietary gluten found in wheat, rye, barley and, for a small minority of patients, oats. There is mounting evidence to suggest that the immune response of coeliac disease is not purely driven by gluten-sensitive T lymphocytes but also by an innate immune response mediated through interleukin-15. Dietary gluten comprises of two major protein fractions, gliadins and glutenins. Considerable research has been carried out on the immunostimulatory components in wheat gliadin. Glutenins, newly discovered to be toxic to coeliac individuals, have limited evidence describing their immunostimulatory potential in coeliac disease. Glutenins can be further differentiated by their molecular weight into low and high molecular weight fractions. Small intestinal T-cell studies suggest that an in vitro response to certain high molecular weight glutenin proteins may occur in some, but not all, coeliac individuals. There is even less evidence available concerning the coeliac disease toxicity of low molecular weight glutenin proteins. The stimulating epitopes within these proteins are not fully understood. This thesis describes the immunostimulatory potential of peptides contained in high molecular weight glutenins, glut 04 p721-735, and in low molecular weight glutenins, glt 156 p44-59, that have been implicated in a small number of coeliac individuals. The adaptive immune response to these peptides has been measured by proliferation assays using gluten-sensitive small intestinal lymphocytes and measurement of their interferon-γ secretion. The innate immune response has been assessed by morphometric measurement of small intestinal biopsies following overnight incubation with these peptides, followed by interleukin-15 assessment via secretion into the tissue culture supernatant. A number of attempts have been made to raise monoclonal antibodies to these peptides. A variety of immunisation schedules have been attempted in host mice. A number of cell fusion experiments have been performed without success. The possible reasons for this lack of success are discussed. Results generated by in vitro studies of coeliac small intestinal explants and isolated cells have improved our understanding of the pathogenesis of coeliac disease.

Background: Cystic fibrosis is the most common inheritable genetic disease affecting 1 in 2500 new born Caucasian infants. The primary cause of morbidity and mortality in these individuals is respiratory failure resulting from chronic airway infections. Previous studies have shown that many bacterial species colonise the airways at any one time. These results have suggested the need for understanding the ecological mechanisms occurring within these communities. This thesis uses ecological experiments to assess the impact of biodiversity on the activity of bacterial assemblages. Methods: Bacterial species and P. aeruginosa ecotypes, isolated from expectorated sputum samples and identified using 16S rRNA gene sequence variation, were assembled into combinations of increasing diversity. These assemblages were inoculated into different environments to assess their effect on respiration. Total respiration was monitored using the MicroResp™ or the BIOLOG EcoPlate™ systems. A general linear model approach was used to analyse the data and investigate the ability of three ecological processes (species richness, composition and interactions) to account for the observed variance. Results: Richness was generally shown to be statistically significant in all environments tested with mixed species present (p < 0.005). Conversely, when P. aeruginosa ecotypes were present no effect of richness was observed (p > 0.272). Composition was found to account for significant variation in the data in all environments and species combinations (p < 0.001). Altering the environment was shown to affect the significance of productivity and interactions among the species. Conclusions: The experiments reported within show a novel application of established ecological models. These identify a complex interplay of mechanisms within the bacterial assemblages that are dependent on the environment and relatedness of the bacteria present.

The GIPIO (Gastrointestinal Peptides in Obesity) collaboration was set up with EU (FP7) funding in 2008, with the aim to develop a series of peptide analogues for the treatment of obesity. Based on the dual Y2/Y4 agonist, Obinepitide ([Q34]hPP), and the Y4 agonist, TM30339 (hPP2-36), novel peptides were developed and subsequently modified by either PEGylation or lipidation to improve their relatively short half-life and pharmacokinetic profile. This thesis presents a detailed study of the in vitro pharmacological data performed to aid the progression of the most promising peptide candidates. An assessment of Y2Y4 potency and activity was made, using optimised Y4 (human colonic epithelial monolayers) and Y2Y4 (human colon mucosae) bioassays, where preparations were mounted in Ussing chambers and changes in short-circuit current (Isc) recorded. In addition, receptor specificity of the most promising agonists was dissected using Y1 and Y2 antagonists. Furthermore, an assessment of Y4 receptor desensitisation was made by application of a subsequent addition of the native hormone, hPP. With a lysine linker at position 30 or 22, chemical modification of Obinepitide with either PEGylation or Palmitoylation resulted in prolonged reductions in Isc in both preparations compared with the transient response seen with their respective predecessors. Increasing the PEG size resulted in greater reductions in Isc, though a PEG22 modification resulted in a loss of functional activity. Mid to long chain lipidation of TM30339 (lauric acid and palmitic acid) also resulted in sustained reductions in Isc. Interestingly, lipidation with a short chain fatty acid (caprylic acid) caused a biphasic response, with an initial transient drop in Isc followed by a sustained anti-secretory response. Importantly, PEG and lipid side chains had no effect upon Isc. Pre-treatment of either human mucosa or epithelial monolayers with PEG2-, or 5-ylated Obinepitide did not cause significant subsequent Y4 desensitisation, an effect that was greater with the larger PEGylated peptide. Contrary to this, long chain lipidated peptides seemed to facilitate rapid and sustained desensitisation, revealed by the loss of further Y4 signalling even after nM lipidated agonist. This divergence in receptor signalling was also observed in fluorescent imaging studies performed by our collaborators. The data presented herein provides the first functional evidence of prolonged activity at the Y2 and Y4 receptors and these modified peptides have the potential to act for longer in vivo as anti-obesity treatments.

Inflammatory diseases of the bowel and liver form an important part of my clinical work as a gastroenterologist. Absent or reduced clinical response to steroids is detrimental to patient outcome in these conditions. This thesis serves to investigate the correlation between in vitro and in vivo steroid response in these inflammatory diseases and to elaborate further on the intracellular mechanisms by which a patient may fail to respond clinically to steroids with particular emphasis on the role of IL-2 and its effect on the glucocorticoid receptor. A greater understanding of these mechanisms may allow the development of diagnostic techniques and targeted therapies to enhance glucocorticoid response in steroid resistant patents and thus improve clinical outcome

An exciting therapeutic goal would be to use agents in clinical practice which could increase the size of a patient's functioning hepatic tissue in times of liver disease such as in the recovery phase from liver transplantation. An increase in the size and function of the liver at times such as this would ensure a more rapid recovery and less associated liver related morbidity and mortality. As a group, the primary hepatomitogens might be employed to this end. Tri-iodothyronine (T3) is a non carcinogenic agent and a very plausible candidate to be used in this setting. T3 is one of a number of drugs which can bring about proliferation of hepatocytes in the adult liver without any preceding cell loss i.e. a process of direct hyperplasia as opposed to the compensatory regeneration one sees following hepatectomy. The work undertaken in this thesis aims to form a better understanding of the mechanism of action of the hepatomitogen tri-iodothyronine, T3. Thus far the mitogenic response is reliably modelled in vivo in the rodent but not in hepatocyte culture. This research attempts to demonstrate directly for the first time a mitogenic effect in human liver and starts with exploring T3 effects in perfused liver and then in vivo in the rodent model where gene array studies are performed to identify novel mediators of the hepatic T3 proliferative response. Potential mediators have their serial expression profiles studied over time after T3 stimulation by Real Time PCR. Finally, the role of mTOR (important in mediating cell growth and proliferative signalling pathways) is identified in mediating the hepatic mitogenic T3 response.

Aims: Columnar metaplasia in the oesophageal remnant occurring after subtotal oesophagectomy (neo-Barrett’s) has been proposed as a human model for the development of Barrett’s oesophagus. This study aimed to assess the incidence of this phenomenon and it’s accuracy as a model as well as looking for evidence of field cancerisation in the oesophagus. Methods: Patients underwent prospective endoscopic evaluation having previously undergone oesophagectomy. The presence or absence of columnar epithelium above the surgical anastomosis was noted and biopsies taken. Specimens were stained using H&E and, where consent was granted, with immunohistochemical stains for proteins which have a well described expression pattern in Barrett’s oesophagus. Tumours and adjacent Barrett’s oesophagus from patients who subsequently developed neo-Barrett’s were screened for genetic mutations. Where these were present, subsequent neo-Barrett’s samples were evaluated for the presence of these mutatations. Results: Of 126 eligible patients, 45 (36%) had confirmed neo-Barrett’s. Median time from surgery was greater for patients with neo-Barrett’s (5.7 vs 2.2yrs, p<0.001). There were no cases of dysplasia. Non-intestinalised columnar epithelium occurred earlier than neo-Barrett’s with specialised intestinal metaplasia. Surgery for dysplastic Barrett’s or adenocarcinoma was associated with a similar prevalence of neo-Barrett’s to other indications (41% vs 27%, p=0.157). 37 samples underwent molecular analysis. Typical, Barrett’s like CK7/20 staining pattern was present in 23 cases (62%). Chromogranin A and trefoil factors 1 and 2 were were present in all cases. TFF3 expression was significantly associated with increasing time from surgery (median 8.1yrs vs 3.4yrs, p=0.004). Genetic mutations identified in the resection specimen were not present in the neo-Barrett’s tissue. Conclusions: Columnar metaplasia is common following oesophagectomy. Cellular protein expression is similar to that of sporadic Barrett’s suggesting this is an accurate model. Presence of intestinal metaplasia and TFF3 expression appear to represent later stages in the development of Barrett’s. No evidence of field cancerisation was found.