Corticosterone is the main glucocorticoid in rodents. Its secretion follows a circadian rhythm and is increased in response to stress. Corticosterone is thought to possess rewarding effects. For example, enhanced corticosterone secretion is associated with the exposure to natural rewards such as food, sex and running, and it has been reported that corticosterone is implicated in drug-seeking behaviour. In addition, rats orally self-administer the steroid hormone and are motivated to nose-poke for stress-like levels of corticosterone, suggesting that the stress hormone has rewarding effects on its own, although this proposal is not supported by conditioned place preference studies. Corticosterone is also found to be critical for learning and memory performance, but it is unclear whether cognitive effects of the hormone are modulated by its rewarding properties. Elevated glucocorticoid levels have been reported to accompany certain psychiatric and cognitive disorders, but it is not clear whether this is an underlying cause or a consequence of these diseases. The two broad aims of the present thesis were to investigate potential rewarding properties of corticosterone in C57BL/6J male mice and to examine their long-term impacts on cognition. In the first experiment, mice were subjected to a prolonged two-bottle choice test to measure their preference for the stress hormone. Results indicated that mice were motivated to self-administer corticosterone at high doses and this was accompanied by enhanced dopamine turnover in the striatum, suggesting that the rewarding effects of corticosterone involve the mesolimbic dopamine pathway; the activity of which is enhanced by the majority of addictive drugs. Corticosterone self-administration did not impair recognition memory in most groups, but it did improve spatial memory. Importantly, corticosterone self-administration induced neuroplasticity changes in the hippocampus and frontal cortex, indicating the involvement of these memory processing areas in the development of corticosterone-seeking behaviour. These results suggest that corticosterone self-administration is associated with spatial memory improvement in mice. Therefore, the remainder of the studies aimed to use a stressful natural reward (voluntary wheel running) to extend our understanding of the role of corticosterone in stress-seeking behaviour and its memory-related effects. The preliminary results revealed that single or repeated 1-hour running sessions increased plasma corticosterone levels. Moreover, metyrapone (a corticosterone synthesis inhibitor) attenuated the running performance of trained mice. In another pilot study, repeated running sessions prevented the improvement of the spatial memory of mice but did not impair their recognition memory compared to non-exercising mice. The final study was undertaken to determine the influences of pharmacological blockade of glucocorticoid or dopamine D2 receptors on the running performance and the exercise-related effects on memory in mice. In contrast to metyrapone, mifepristone (a glucocorticoid receptor antagonist) did not decrease running distances of mice, but it did induce recognition memory impairment in non-exercising mice. These memory deficits were reversed by running, suggesting that exercise-evoked corticosterone increase is involved in the cognitive effects of wheel running. On the other hand, a dopamine D2 receptor antagonist (sulpiride) attenuated the running performance of mice and induced exercise-dependent opposite effects on memory. Overall, the results of this thesis support the hypothesis that corticosterone is rewarding in mice, and that these effects probably enhance certain aspects of cognition.

Cannabis has been used as a medicinal and natural product for thousands of years. Whether it has been used to make rope or paper, or been used to treat pain or depression, cannabis has always had a place in human civilisation. With the isolation of the psychoactive compounds responsible for cannabis’ effects, the discovery of two human cannabinoid receptors and an expanding knowledge of the therapeutic uses of cannabis, interest in the development of novel cannabinoids grew. The CB2 cannabinoid receptor has gained particular attention, as the often unwanted central and psychoactive effects of cannabinoids has been attributed to the CB1 cannabinoid receptor. Development of CB2 receptor selective ligands offers treatment opportunities in many areas, but most especially for pain, multiple sclerosis and immunomodulation. The preparation of fluorescently labelled ligands for a variety of receptors has improved compound screening techniques, as well as allowing use as biomolecular probes for aiding our understanding of the receptor in situ. The aim of this work is to design, synthesise and evaluate novel fluorescently labelled cannabinoids, with a particular interest in CB2 selective compounds. Focusing on the CB2 receptor selective alkylindole JWH-015, targeted substitutions were made to its naphthyl ring to identify sites that might be suitable for fluorophore attachment. With a site chosen, a series of fluorescent JWH-015 analogues was synthesised and evaluated for their CB2 receptor binding affinities. Though none of the evaluated compounds showed sufficient binding affinity for them to be used as biomolecular tools, the structure activity relationships gained suggested that improved design of fluorescent JWH-015 analogues in future could lead to the first ever active fluorescent cannabinoid.

The main aim of this thesis was to characterize mammalian chitotriosidase in terms of its function in innate immunity. This was achieved using several different approaches. In terms of gene expression, it was demonstrated that chitotriosidase is highly expressed in human peripheral blood monocyte-derived macrophages and can be induced further by β-glucan-containing PAMPs. This suggests that the enzyme may be involved specifically in anti-fungal immunity. Furthermore, chitotriosidase activity was detected in neutrophil pellets and the supernatants of cells treated with prolactin (consistent with its putative role in immune modulation) and the degranulating compound fMLP. The presence of chitotriosidase activity was also demonstrated in human tears; previously it has been shown that there are high levels of expression in the lachrymal glands of the eye, that also express lysozyme. As lysozyme has been shown to have antimicrobial activity in combination with other compounds and has additional chitinase activity, it was postulated that these enzymes may produce synergistic effects. The addition of recombinant human chitotriosidase, however, had no additional effect against Gram-positive or Gram-negative bacteria, when added with sub-optimal doses of lysozyme. It has been suggested that the loss of the mutant chitotriosidase genotype in two meso-endemic West African countries demonstrates that the enzyme plays a critical role in malaria prognosis. Furthermore, the homozygous mutant genotype has been linked with susceptibility to filarial infection. However in the current study of Papua New Guinean populations, it was shown that there is no correlation between genotype and the extent of hookworm infection. Furthermore, the frequency of the mutant allele was maintained at a level similar to those in non meso-endemic countries, suggesting that the enzyme does not play a critical role in the outcome of malaria. Phylogenetic analyses of the mammalian chitinase/chi-lectin gene cluster demonstrated that these genes have a highly conserved gene signature and may have evolved from a common AMCase ancestor gene. This would be consistent with the requirement for a chitin-digesting enzyme in primitive mammals. Through duplication and mutations events, this gene most likely produced subsets of other chitinases and chi-lectins, the latter evolving non-enzymatic functions, courtesy of the conserved α/β TIM-barrel domain.

ER accessory proteins are a novel class of endoplasmic reticulum (ER) proteins that facilitate the exit of polytopic membrane proteins from the ER. They are important for the correct targeting of their cognate polytopic membrane proteins to the plasma membrane (PM) and their absence leads to abnormal accumulation of their target in the ER. Until recently, it was not clear if such proteins exist in plants. However, work by Dharmasiri et al (2006) and Gonzales et al (2005) suggest that such proteins exists in plants too. Polytopic membrane proteins such as nutrient transporters, hormone transporters and sugar transporters are a very important class of proteins as they regulate many important physiological and biochemical processes. Better understanding of the targeting of these proteins to the PM is of considerable agronomic interest due to the importance of efficient use of resources in sustainable agriculture. One of the projects aims is to identify novel ER accessory proteins in Arabidopsis. Using a bioinformatics approach, 40 novel ER resident proteins were identified from a protein localisation database (LOPIT) generated by Dunkley et al (2006) as potential candidates for ER accessory proteins. Genetic, phenotypic and molecular approaches have been used to assess their role as potential ER accessory proteins. A few promising candidates have been identified, one of which AtBPL1 and related family. The AtBPL1 family has similarity to mammalian BAP31 which has been shown to function as an ER accessory protein (Ladasky et al, 2006). To determine if AtBPL1 family plays a similar role in plants a detailed molecular characterisation was carried out, this involved detailed expression analysis using reporter genes and in situ immunolocalisation and characterisation of miRNA lines. Smart screens suggest that BPL1 family members may be involved in the targeting of a nitrate transporter, however its precise target is currently unknown. A key focus of this present investigation have been on further characterisation of AXR4, which is required for the correct targeting of AUX1 to the plasma membrane (Dharmasiri et al, 2006). AUX1 belongs to a multi-gene family, involving three other members, LAX1, LAX2 and LAX3. Using genetic and cell biology approaches, AXR4 has been shown to be necessary for the correct localisation of at least two other members of this family LAX2 and LAX3. AXR4 mutants show defects in targeting of LAX2 and LAX3 to the plasmamembrane and show weak lax2 and lax3 phenotypes. Co-Immunoprecipitation studies revealed that AXR4 and AUX1 interact directly when co-expressed in insect cells. Finally molecular, bioinformatics and protein modelling approachs were used to probe the function of alpha beta hydrolase domain in AXR4 function. AXR4 appears to be tolerant to amino acid subsitition even at highly conserved amino acids, suggesting that the alpha beta hydrolase domain may not be important for its function.

Eukaryotic initiation factor 4A (eIF4A) is an ATP-dependent RNA helicase responsible for unwinding the secondary structure of mRNAs. In humans, eIF4A exists as three separate paralogs: eIF4AI and eIF4AII possess a high degree of homology while eIF4AIII is distinct. Knockdown of eIF4AII had no effect on the expression of a reporter construct containing a structured RNA hairpin. Knockdown of eIF4AI and treatment with hippuristanol (an eIF4A inhibitor) caused a dramatic reduction in the hairpin-mediated gene. This reporter system was developed as part of this project to act as a screen for eIF4A activity along with an in vitro screening approach. The activity of eIF4A is suppressed in vivo by the tumour suppressor PDCD4. The fact that loss of PDCD4 function increases the severity of DNA damage is probably attributable its eIF4A-suppressive activity. Based on previous microarray data, it was supposed that eIF4A inhibition may be therapeutically beneficial in the treatment of Alzheimer's disease. As part of this project, it was demonstrated that eIF4A suppression significantly reduced the expression of reporter genes preceded by the 5’ UTRs of genes predicted to play harmful roles in Alzheimer’s disease. The expression of reporter genes preceded by the 5’ UTR sequences of genes predicted to be beneficial in Alzheimer's were not affected by this suppression. Reporter plasmids containing the 5’ UTR sequences of the oncogenes ODC1, EGFR and VEGFA have high requirements for eIF4A as estimated using hippuristanol. eIF4A inhibition did not significantly affect the reporters containing the 5’ UTRs of non-pathogenic genes. The EGFR 5’ UTR was found to contain an IRES which explains why EGFR is upregulated in response to hypoxia.

Nanosensors offer the opportunity to measure intracellular domains with minimal chemical or physical perturbation. Typically only 60 nm in diameter and synthesised from polymer matrices they entrap chemical sensing elements and can be surface functionalised allowing for further chemical modification. Upon intracellular localisation the cellular environment can be monitored using conventional techniques such as confocal laser scanning microscopy. Reported here is one use of nanosensors to investigate the mechanisms of intracellular delivery mediated via the cell penetrating peptide, Tat. It is shown that information obtained from the nanosensors reveals that the post-delivery environment is representative of a lysosome in terms of both pH and morphology. The delivery mechanism of Tat, however, is shown to be dependent upon the cargo being delivered, corresponding to the absence or presence of a body of polymer matrix; thus nanosensors have been used to further the understanding of the cell penetrating mechanisms of Tat peptide. Technological aspects of nanosensor development have been investigated including polymer matrix modification and different methods of incorporating fluorophores into the nanosensor body. Addressing nanosensors located in an intracellular domain has historically been achieved with epi-fluorescent and confocal microscopy acquiring data from individual or low numbers of cells only. Reported here is the combination of nanosensors with flow cytometry as a technique for en masse investigation into entire cell populations.

To provide further insight into the relationship between the structure of hydrocolloid solutions and gels and perception of taste and flavour, solutions of gelatin and locust bean gum, and gels prepared from mixtures of (a) high acyl and low acyl gellan (b) carrageenan and locust bean gum were studied. Both solutions contained sodium chloride and the gels were flavoured with ethyl butyrate. The gels were classified from rheological measurements into 3 categories: strong/brittle, intermediate and soft/elastic. Volatile release was measured by monitoring nose space volatile concentration during consumption using Atmospheric Pressure Chemical Ionisation-Mass Spectrometry (APCI-MS). In addition headspace measurements were performed with APCI-MS. The headspace concentrations did not exhibit significant differences between the gels systems but the release of ethyl butyrate in-nose was affected by the matrix, showing a higher intensity for the more brittle gels containing high levels of low acyl gellan. The release of Na' following a two bite compression was monitored by the use of an ion specific electrode. The more brittle gels containing high levels of low acyl gellan and high amount of K-carrageenan exhibited significantly higher release of Na'. Strain at break correlated inversely with salt release (r2=-0.87) and nose space volatile concentration (r2=-0.55). In a later stage gelatin was added (1-5%) in the previous mixtures of HA-Gellan and LA-Gellan (constant polysaccharide concentration of 0.6%). The rheological analysis of the gels yielded different behaviour of the gels. At low levels of LA-Gellan the rheological data can be explained by polymer blending laws. At higher levels of LA-Gellan, development of elastic behaviour from the previous brittle gels observed does not fit polymer blending law theory. Flavour release during diffusion experiment showed that at 37°C the gels containing gelatin exhibited higher salt release. Temperature sweeps have shown that a drop of G' is observed around 27-28° C indicating that the gelatin present in the mixture is melting. However the level of the drop of G' indicates that the continuous phase of the gel composite was the gellan system. Volatile release was measured by monitoring nose space volatile concentration during consumption using Atmospheric Pressure Chemical Ionisation-Mass Spectrometry (APCI-MS) but showed no significant differences between the different gels. Headspace experiments performed at different temperatures showed that the gels containing high amounts of gelatin when compared to control gels that contained 0% gelatin exhibited higher release of ethyl butyrate. To mimic the mixing of gelatin with saliva after melting, gelatin solutions at 50°C containing salt were mixed with water. Even at high concentrations (30%) of gelatin mixing efficiency and release was very efficient. In contrast when locust bean gum solutions containing salt were mixed with distilled water it was found that both salt release and mixing efficiency decreased at polysaccharide concentrations above c*. It is concluded that the intensity of flavour perception in hydrocolloid solutions and gels is dominated by the release of the tastant. In solutions this is favoured by good mixing behaviour between the hydrocolloid solution and saliva and in gels by a low strain at break. A gelatine replacement should not only show melt in the mouth behaviour but good mixing between the melted gelatin and saliva. It was shown that thermal processing at 121°C induced deacetylation of HA-gellan at much lower pH than would normally be needed in a typical deacetylation process. Therefore new textures can be achieved through deacetylation through heat processing. When gels which were prepared by deacetylation by heat processing were compared with blends that had an equivalent acyl content different textures were obtained even though the Young's modulus was very similar at the same total acyl content. The implications of this work for gelled petfood products is discussed.

Body condition score has been considered an indirect measure of nutritional status and nutrition has been demonstrated to interact with reproduction. Adiponectin and leptin, because of its insulin sensitising actions and their association with body condition could be potential regulators of metabolism during the transition from pregnancy to lactation. The work described in this PhD project was designed to investigate the role of metabolic, body condition, and dietary factors in regulation of productive and reproductive performance in dairy cows, with particular regard to glucose homeostasis and adipokines. A special focus was directed to circulating adiponectin and its association with metabolic and hormonal status, and reproduction. Interestingly, adiponectin was found to be present in bovine milk at concentrations similar those of plasma. Body condition score, GH, insulin, leptin, and adiponectin showed a potential to modulate glucose homeostasis and reproductive output. Results from this study demonstrated a negative effect of long-term moderately elevated insulin concentration on reproductive performance. Additionally, BCS at calving and ΔBCS are determinant factors influencing postpartum reproductive recovery and they should be basic components of farming monitoring schemes. Although further investigation is needed to address the proposed negative relationship between adiponectin and GH, and to identify other dietary factors that may modulate circulating adiponectin, data from this study support regulatory roles of adiponectin in metabolism and reproduction. Moreover, adiponectin holds great promise to serve as a putative target molecule that integrates metabolism, reproduction and feeding behaviour. More importantly, hypoadiponectinemia could be another alternative mechanism that contributes to poor fertility in dairy cows. The incorporation of this new knowledge into the development of better nutritional strategies is a key point that is being considered to improve the welfare, and reproductive and productive performance in dairy cows.

In this thesis molecular dynamics simulations, in conjunction with the complementary methods of docking and QM-MM, are used, and further developed, to study two unusual polypeptide systems: the conformational preferences of isomers of an antibiotic peptide and the binding behaviour of a human transporter protein. The antibiotic peptides are analogues of a naturally occurring antibacterial called nisin which has a biological function dependent on the formation of five macrocyclic rings closed by a thioether bond between modified L-amino and D-amino residues. We propose analogues where the thioether bond is replaced by a disulfide bond between cysteine residues and the chirality of the cysteines is altered. The conformational preferences of the nisin analogues, and the dependence of ring formation on cysteine chirality, are characterised using molecular dynamics. An analogue (D-Cys3-D-Cys7-L-Cys8-L-Cys11) is identified that favours the simultaneous formation of the S3-S7 and S8-S11 disulfide bonds and has an RMSD of 0.6 Å to 1.7 Å between the centroids from clustering the MD trajectories and an NMR structure of wt-nisin. The nisin analogues contain unusual D-amino residues and using explicit solvent MD simulations of four polypeptides, it is shown that the (φ, ψ) → (-φ, -ψ) transformation of the CMAP term in the CHARMM potential energy function leads to sampling of conformations which are closest to X-ray crystallographic structures for D-amino residues and that the standard CMAP correction destabilises D-amino β-sheets and β-turns. The ileal lipid binding protein (ILBP) shows cooperative binding comparable to haemoglobin and unusual site selectivity where one ligand will completely displace another from a binding site, despite both sites having an affinity for each ligand type and the ligands only differing by a single hydroxyl group. A probable location of the third binding site of ILBP is identified which has a role in the allosteric binding mechanism. MD simulations indicate that binding to this exterior site induces changes in the orientation of the α-helices with respect to the β-barrel by ~10°. An energetic mechanism of site selectivity for ILBP is proposed using evidence from MD simulations. The higher hydrophobicity of chenodeoxycholic acid leads it to sit deeper in the binding cavity and interact with Gln-51. This causes the cholic acid ligand to be deeper and induces the helices to move closer to the β-barrel, preventing further ligand exchange.

In Chapter 2, bis- and tris- arylsulfonamides, were investigated as possible inhibitors of the p53-MDM2 protein-protein interaction (PPI). The lead compound, 19, inhibited the PPI, in a fluorescence polarisation (FP) based competitive binding assay with ICso 26.4 pM and the most potent analogue, 66, with ICso 3 μM. The active compounds in this series, possess a 5-chloro-4-nitro-2-sulfonamoyl substituted thiophene ring that is very susceptible to SNAr reactions at the 5-position. Analogues of 19 and 66 were prepared to investigate the SAR of these inhibitors. No improvements in activity or structural activity relationship (SAR) consistent with MDM2 binding were observed and no active analogues without the reactive functionality were found. These compound are no longer being investigated. Chapter 3 describes a 3-D shape-based virtual screening campaign to find new lead compounds. Using queries based on the established Nutlin, benzodiazepine and spiro-oxindole inhibitors, the ZINC database was screened using the program ROCS to find compounds that have good shape similarity (measured by 3D Tanimoto) and similar functional group overlap to the query molecules. 155 compounds were purchased and tested 16 of which inhibited the MDM2-p53 PPI in the FP assay at IC50 ranging between 48.22 and 140.42 1tM. Three analogues, 156,168 and 180, induce low levels of p53 induction in cells using a Luciferese based reporter gene assay with most the potent compound, 180, showed 5.75 fold induction at 8.89 μM. A number of the hit compounds warrant further investigation. Chapter 4 describes the investigation 1.5-benzodiazepiene-2-ones and 1,3-dihydrobenzimidazolin-2-ones as novel scaffolds on which to base potential p53 inhibitors. A small series of analogues of each class were prepared and their ability to disrupt the MDM2-p53 PPI determined using an FP assay. None of the 1.5-benzodiazepiene-2-ones showed any inhibition of the PPI at concentrations up to 500 μM. Some of the 1,3 dihydrobenzimidazolin-2-one based compounds showed low levels of inhibition with the most potent analogue, 214 having IC5o 196.18 PM. These inhibitors showed some SAR based on the size of substituents and the presence of a 6-chloro substituent that has been shown to considerably enhance the activity in other classes of inhibitor. Compounds of this type warrant further investigation using a more diverse compound library.