Introduction and methods Asymmetric dimethylarginine (ADMA) is a potent endogenous competitive inhibitor of nitric oxide synthases, which has attracted considerable attention as a marker and mediator of atherosclerotic disease and as a potential mediator of multiple organ failure in critical illness due to endothelial dysfunction. However, data regarding basic aspects of its biology such as biological variation and its response to acute inflammation are lacking. Moreover, significant methodological variability has been a barrier to collating the burgeoning data available. Therefore, this thesis describes the development and validation of a reliable assay for measurement of ADMA and related compounds in plasma, urine and other biological fluids based on isocratic reverse phase high performance liquid chromatography (HPLC). This method was used to determine the biological variation of ADMA in human plasma, and its response to acute inflammation using a model of elective knee arthroplasty. Further HPLC methods for measurement of dimethylamine (DMA), the main metabolite of ADMA, and nitrate were developed and used to determine excretion of these compounds in acute inflammation to complement the observed changes in plasma ADMA concentration. Results Complete chromatographic separation of arginine, homoarginine, monomethyl-arginine, ADMA and its structural isomer SDMA was achieved, permitting their accurate quantification using a novel, non-endogenous, internal standard. The intra-individual biological variation of ADMA was found to be low at 7.4%, imposing a tight imprecision goal for analytical methods. Plasma ADMA concentration decreases rapidly during the acute inflammatory response, with a median decrease of around 30%, and a significant change already evident as little as 12 hours following the onset of inflammation. No similar change was seen in the concentration of the closely related compound SDMA. No significant increase in the urine excretion of DMA was noted during the early phase of the response, with a significant increase seen 5 days following the insult by which point the plasma ADMA concentration had returned to baseline levels. A small, but significant, decrease in nitrate excretion during the inflammatory response was seen, mirroring the observed changes in plasma ADMA. Conclusion The low biological variation of ADMA suggests physiological regulation. The rapid and significant decrease in plasma concentration during inflammation does not appear due to increased catabolism, but rather is more likely to represent increased cellular partitioning. This may be associated with an impairment in NOS activity. It is unclear whether this is of pathological significance, or represents a physiological response to regulate NO production in inflammation. Further study is warranted in relevant models, particularly with attention to intracellular concentrations.

A crucial and challenging aspect of the drug development process is the requirement to measure the distribution of a pharmaceutical compound and its metabolites in tissue. Industry-standard methods used to look at total localisation of drug-related material are limited due to their dependence on labels. These labelled techniques can have difficulty in distinguishing between the drug of interest and its metabolites. Imaging mass spectrometry is a technique that has the potential to spatially distinguish between drug and metabolites, due to its high chemical specificity and sensitivity. A number of imaging mass spectrometry approaches have been described for localisation of drug compounds in tissue, most notably matrix-assisted laser desorption/ionisation (MALDI) imaging, which can provide data complementary to existing imaging techniques. Two imaging mass spectrometry approaches have been evaluated and compared for use in the localisation of a range of drug compounds in target tissues. The techniques used were MALDI imaging and a recently described electrospray ionisation-based technique, liquid extraction surface analysis (LESA). Both techniques have been successfully used for the detection of drug compounds in dosed tissue sections. A major challenge associated with imaging techniques is the required selectivity of the experiment for the compound of interest, due to the complex nature of tissue sections. Combining the shape-selective method of ion mobility separation with MS/MS fragmentation has been shown to improve the selectivity of both imaging approaches for the compound of interest. Results obtained using LESA-MS have demonstrated the suitability of this technique as a rapid and sensitive profiling technique for the detection of drugs and metabolites in tissue, but with a lower achievable spatial resolution than MALDI imaging. Higher spatial resolution was achieved with MALDI imaging; however data acquisition times were longer and required higher dosing levels for successful detection of drug compounds in tissue. A biological application of MALDI imaging was also evaluated. Mobility-enabled MALDI imaging was used to assess differences in the localisation of important adenine nucleotides between control and metabolically stressed mouse brain sections. Tissue fixation methods were evaluated to overcome rapid post-mortem degradation of adenine nucleotides such that biologically relevant localisation images can be obtained. These studies highlight the crucial importance of appropriate biological sample preparation in MALDI imaging experiments.

This thesis investigates the effects of various inflammatory stimuli, including cytokines and air pollution particles, on the expression and secretion of various proinflammatory proteins in the lung epithelial cell line A549. The pro-inflammatory proteins investigated were C-reactive protein (CRP), fibrinogen, and heat shock protein 70 (Hsp70), all of which are known to be produced during inflammation and are also known risk factors for cardiovascular disease. These proteins were investigated because inhalation of particulate air pollution has been associated with increased cardiovascular disease and mortality. The production of these proteins in the lung may be involved in a systemic inflammatory response which increases the plasma levels of these proteins and other cardiovascular risk factors therefore increasing the susceptibility for cardiovascular events. Alternatively, localised CRP functions as an opsonin, while fibrinogen and its degradation products are involved in the recruitment of neutrophils and leukocytes to the lung. Both CRP and fibrinogen are known to be produced in hepatocytes in response to cytokines. Chapter 3 investigates the effect of cytokine treatment of the A549 cells which shows that CRP and fibrinogen can be induced in this cell line by various cytokines. CRP was found to be induced by the cytokines ILI P, EL6, IL8, TNF(x and IFN7 with increased effects shown by simultaneous treatment with two cytokines. Fibrinogen was found to be induced mainly by IL6, but fibrinogen levels were also slightly increased by ILlp; simultaneous treatments of IL6 with either ILlP or TNF(x reduced the effect of IL6 treatment. Treatment of A549 cells with IL6 and IL8 simultaneously induced a synergistic effect. After establishing that cytokines induce the expression of CRP in A549 cells, the effects of particulate air pollution on pro-inflammatory protein expression were then investigated (chapter 4). The cells were treated with carbon black (CB), ultrafine CB (ufCB), and iron chloride (FeC13) to find out what effect these air pollution components and PM10 would have on the expression of CRP, fibrinogen, Hsp70 and the transcription factor NFKB and its inhibitor IKB. NFKB is known be activated by PM10 and is involved in the signalling of pro-inflammatory responses. All the particulate treatments induced a pro-inflammatory response with expression and secretion of CRP, fibrinogen, and Hsp70, whereas the soluble metal treatment had little effect. The metal salt FeC13w as used to treat the cells since it has been suggested that PM10 may mediate its effect through the presence of transition metals which are implicated in oxidant generation. The particulate exposures were also associated with the activation and nuclear translocation of NFKB indicating the involvement of NFKB and IxB in the induction of the pro-inflammatory response. Investigations into the mechanisms by which the particles induced the proinflammatory response are discussed in chapter 5. Firstly, investigations into the effect of particles on the secretion of the pro-inflammatory proteins were carried out; these indicated that particles were capable of inducing the secretion of CRP, flbrinogen and Hsp70. Investigations into the transcriptional mechanisms of proinflammatory protein expression were carried out using specific inhibitors. CRP and fibrinogen were induced in an NFKB dependent manner, while Hsp70 was produced as a result of activation of the JAK/STAT pathway. The effect of oxidative stress being induced as a result of particle exposure was investigated using antioxidants. This showed a reduction in the amount of intracellular and secreted CRP, fibrinogen and Hsp70 in the presence of antioxidants, indicating that oxidative stress was involved. This correlated with a reduction in the levels of intracellular ATP. Finally in chapter 6 the effects of proteins secreted from A549 cells on the monocytelike cell line MM6 were examined. This was carried out as a model of the effects of secreted pro-inflammatory proteins from the lung epithelium on other cells. It was found that CRP and fibrinogen were able to induce the activation of NFkB in MM6 cells, indicating that secretion of these proteins during lung inflammation could have an effect on other resident lung cells. Conditioned medium from A549 cells was also found to induce the expression of pro-inflammatory proteins including CRP, fibrinogen, Hsp70, IL6, IL8 and TNFcc. The A549 conditioned medium also appeared to induce a stress response and cellular damage in MM6 cells, thereby potentially exacerbating the inflammatory response. These results indicate that pro-inflammatory proteins can be produced in lung epithelial cells in response to particle exposure as a result of oxidative stress. They may be involved in the progression of localised lung inflammation and in the systemic response to particulate air pollution.

This thesis describes the use of enzymes to partially deprotect or deprotect sugars. The advantages of this technique are that mild conditions may be used for the protections and that enzymatic catalysis may allow derivatives to be made which involve multi-step procedures or cannot be made using standard methods. These protected sugars can then be used to synthesise interesting derivatives. The particular aim of this project at the outset was to use biotransformations to make partially protected derivatives for the synthesis of a novel artificial sweetener, l',4,6'-trichloro-r,4,6'-trideoxy-ga/acto- sucrose, known as sucralose. The analysis of the products from a deacetylation reaction is a potential problem due to the number of possible products that may be formed. A new approach to this problem is discussed in Chapter Two. A combination of n.m.r. and mass spectrometry is used to analyse the products after they have first been perdeuterioacetylated by treatment with d6-acetic anhydride. The analysis is therefore carried out on one compound, the starting material, now containing deuteriated acetate groups in place of those that were hydrolysed during the reaction. The technique was used initially to analyse the deacetylation of sucrose octaacetate catalysed by yeast esterase. The selectivity of the enzyme for certain positions of the sugar may be determined in this way but little information can be found about the individual species that are formed. The technique can be considerably enhanced by the introduction of a chromatographic separation step. The separation of the deacetylation mixture into classes, according to the number of acetate groups, allows a much more detailed analysis of the individual components to be carried out. If the reaction shows a certain amount of selectivity then it is possible to determine the quantity of each of the individual species. This technique is used to analyse the deacetylation of glucose pentaacetate catalysed by Aspergillus niger lipase. The deacetylation of sucrose octaacetate catalysed by yeast esterase is also analysed in the same way. Chapter Three describes the conversion of N-acetyl- glucosamine to N-acetyl-galactosamine. This is of interest due to the importance of this sugar in biological systems and its high cost relative to the starting material. The synthesis involves the use of an enzyme catalysed deesterification to make a partially protected intermediate, demonstrating the practical application of biotransformations in the synthesis of sugar derivatives.

Multiple receptor tyrosine kinases (RTK) have been shown to be over-expressed in the malignant Hodgkin Reed-Sternberg (HRS) cells of Hodgkin lymphoma (HL). However, the activation status of many of these RTKs has not been studied. Furthermore, the contribution of aberrant RTK activation to the pathogenesis of HL is currently unknown. In chapter three, I have shown using a human phospho-receptor tyrosine kinase array that HL cells are characterised by the activation of multiple RTK. I have confirmed the over-expression and activation in HRS cells of two of these RTK, MET and RON and provided preliminary evidence that MET is negatively regulated by LRIG1 in these cells. In chapter four, I have shown for the first time that DDR1 is over-expressed in primary HRS cells. Furthermore, I have shown that in many cases, DDR1-expressing HRS cells are intimately associated with collagen, the ligand for DDR1. However, knockdown of DDR1 in a HL cell line in which DDR1 appeared to be constitutively phosphorylated revealed no detectable change in phenotype and few transcriptional changes. While exploring possible reasons for this, I identified that HL cells express multiple DDR1 isoforms including several novel transcripts. Finally, in chapter five, I have shown that HL cells are sensitive to the RTK inhibitor, dasatinib. Furthermore, consistent with the aberrant activation of multiple RTKs in HL cells, I observed that these cells were also sensitive to lestaurtinib and dovitinib, two next generation multiple-target RTK inhibitors.

The growing problem of insecticide resistance is jeopardising current pest control strategies and current insecticide development pipelines are failing to provide new alternatives quickly enough. Metabolomics offers a potential solution to the bottleneck in insecticide target discovery. As a proof of concept, metabolomics data for permethrin exposed Drosophila melanogaster was analysed and interpreted. Changes in the metabolism of amino acids, glycogen, glycolysis, energy, nitrogen, NAD+, purine, pyrimidine, lipids and carnitine were observed along with markers for acidosis, ammonia stress, oxidative stress and detoxification responses. Many of the changed metabolites and pathways had never been linked to permethrin exposure before. A model for the interaction of the observed changes in metabolites was proposed. From the metabolic pathways with the largest changes, candidate genes from tryptophan catabolism were selected to determine if the perturbed pathways had an effect on survival when exposed to permethrin. Using QPCR it was found that all genes in the entire pathway were downregulated by permethrin exposure with the exception of vermilion suggesting an active response to try and limit flux through tryptophan catabolism during permethrin exposure. Knockdown of the tryptophan catabolising genes vermilion, cinnabar and CG6950 in Drosophila using whole fly RNAi resulted in changes in susceptibility to permethrin for both topical and oral routes of exposure. Knockdown of the candidate genes also caused changes in susceptibility when the insecticides fenvalerate, DDT, chlorpyriphos and hydramethylnon were orally administered. These results show that tryptophan catabolism knockdown has an effect on surviving insecticides with a broad range in mode of action. Symptoms that occur in Drosophila during exposure to the different insecticides were also noted. To gain further understanding into the mechanisms affecting survival, tissue specific knockdown was performed revealing tissue and gender specific changes in survival when vermilion, cinnabar and CG6950 are knocked down. Metabolomics was performed on the knockdown strains to determine the efficacy of the knockdowns on tryptophan catabolism and to identify any knock-on effects. The results indicate that tryptophan metabolite induced perturbations to energy metabolism and glycosylation also occur in Drosophila along with apparent changes in the absorption of ectometabolites. As the knockdown of vermilion, cinnabar and CG6950 tended to result in reduced susceptibility to insecticides, they would make poor targets for insecticidal compounds, however, they may be the first examples of genes that are not directly involved in insecticide metabolism or cuticle synthesis that increase insecticide tolerance in Drosophila. As the first metabolomics data set showed evidence for oxidative stress during permethrin exposure, preliminary work was begun for identifying the tissue specificity and timing of oxidative stress in both Dipterans and Lepidopterans using Drosophila and Bombyx mori as models. In Drosophila oxidative stress did not begin immediately suggesting that the insecticide itself is not a cause, however, a rapid increase in oxidative stress occured over a six hour period after a day of oral exposure implicating catabolites of permethrin. Bombyx were highly susceptible to permethrin showing oxidative stress in the Malpighian tubule and silk gland when exposed. This study has shown that metabolomics is highly effective at identifying pathways which modulate survival to insecticide exposure. It has also brought insight into how insecticide induced pathology may cause death. Data has also been generated which could help characterize the putative transaminase CG6950.