The alarmingly increasing human population needs improved food production but this aim is hampered by different abiotic stresses. Osmotic stress and salt stress are the two prominent examples of abiotic stresses and affect up to 50% of the arable land. These stresses severely affect all plants, but glycophytes (e.g. rice) are especially sensitive. During stress, nutrient uptake, such as K+, is often disturbed. Thus, better K+ nutrition and distribution play a vital role in plant abiotic stress tolerance. To improve K+ nutrition, the role of K+ transporters is likely to be essential. Loss of function and gain of function approaches could help establish the exact function of transporters involved in K+ nutrition. Rice TPKs and AKT1 are K+ channels which are localised to the tonoplast and plasma membrane respectively. The two TPK isoforms, TPKa and TPKb, are localised to the tonoplast of LV and SV respectively. They were characterized in a variety of abiotic stress conditions. The data showed better growth and higher K+ concentration for the TPKa and TPKb transgenic lines when grown in zero K+ and osmotic stress conditions suggesting their role in improving in K+ nutrition. TPKs have no direct involvement in the K+ uptake, but somehow influence K+ uptake and improve K+ nutrition. The higher K+ concentration in the leaves of overexpressor plants suggested the involvement of TPKs in the distribution of K+ within the plant body. TPKs play a role in the guard cells' movements and affect the stomatal conductance and therefore showed a better response to the osmotic stress conditions. The role of rice AKT1 was tested by comparing the knockout and overexpressing lines of AKT1 with the wild type plants. The data suggested that AKT1 is involved in the K+ uptake in a range of external K+ concentrations and osmotic stress conditions. The role of AKT1 is obvious in the K+ deficient conditions where NH4+ is present. The leaf K+ concentration suggested that AKT1 influences K+ transport into the leaves. The K+ concentration in the leaf cells showed an effect on the stomatal conductance and in turn an effect on the growth phenotype under zero K+ and osmotic stress conditions. The data revealed that AKT1 is insensitive to NH4+ toxicity.

Photodynamic therapy (PDT) is an approved but under-development modality for cancer therapy. Conventional, non-targeted PDT consists of administration of a photosensitiser to the patient, passive accumulation in the tumour, localised photosensitiser activation using visible light and subsequent cellular destruction via reactive oxygen species (ROS) and/or free radicals. A major limitation of PDT is its poor selectivity for tumours and a suboptimal pharmacokinetic/biodistribution profile. This leads to low potency and patient photosensitivity. This is being addressed with the development of antibody targeted PDT; the conjugation of multiple photosensitisers to a tumour-specific antibody. For effective antibody conjugation photosensitisers are preferable if highly water soluble. PPa (Pyropheophorbide-a) is an effective photosensitiser and more soluble versions have been prepared to aid in its conjugation. The work presented in this thesis explores the differences these novel, soluble photosensitisers exhibit in terms of toxicity, cell death mechanisms and uptake in an in vitro model of traditional PDT and also in terms of improvements in their bio-conjugation. Data presented shows chemical manipulation of PPa significantly affects the biochemical properties important for effective PDT. PEG-modified PPa (PS-1, with enhanced solubility over PPa) as an un-conjugated drug exhibits similar overall toxicity to PPa, along with similar hallmarks for caspase-independent apoptosis, however this equivalent toxicity is achieved despite a significantly reduced cellular uptake. Its toxicity therefore is higher per internalised molecule than its parent compound. This is hypothesised to be due to enhanced retention of photo-physical properties in a cellular environment. Cationic PPa (PS4, with even greater enhanced solubility over PPa) exhibits a loss in cytotoxicity in comparable levels of free photosensitiser studied; this is hypothesised to be due to decreased uptake in addition to an altered cellular localisation. Derivatives of PS1 activated for conjugation showed greatly enhanced reaction with antibody compared to PPa showing its promise for improved targeted PDT, PS4 proved too difficult to chemically activate in the time frame of this project.

The unfolded protein response (UPR) is a cellular mechanism that detects the accumulation of misfolded proteins within the endoplasmic reticulum (ER). In mammalian cells, the UPR is mediated by three ER-transmembrane proteins: PERK, IRE1 and ATF6. Early studies in the field provided evidence for the role of BiP, the major ER Hsp70 chaperone, in UPR activation by binding to the luminal domains of PERK, IRE1 and ATF6 and maintaining them in an inactive state. However the underlying mechanism of ER stress sensing and UPR activation is not yet understood. This thesis presents (i) the novel X-ray crystal structure of PERK luminal domain and (ii) a biochemical study of the unconventional interaction between BiP chaperone and the luminal domains of PERK and IRE1 in vitro. Firstly, the structure of PERK luminal domain was solved in two oligomeric states: dimers and tetramers. Compelling evidence is provided for a role of tetramer formation in directing downstream UPR signalling. Secondly, the unprecedented and unconventional direct binding of PERK and IRE1 luminal domains to BiP is demonstrated. The binding surface was mapped to the nucleotide-binding domain (NBD) of BiP. As such, this points away from a substrate-chaperone interaction and rather implies BiP as an explicit UPR signalling component. Upon binding to CH1 unfolded protein, BiP is released from PERK and IRE1 luminal domains. BiP dissociation from PERK and IRE1 is known to lead to their activation. Based on the work presented in this thesis a novel mechanism of ER stress sensing and UPR activation by PERK and IRE1 is proposed. BiP NBD normally interacts with the luminal domains of PERK and IRE1 and represses UPR signalling. During ER stress, binding of unfolded proteins to BiP's substrate binding domain leads to the dissociation of BiP-luminal domain complexes. As such, the luminal domains are free to intertwine, mediated by an extended α-helix, and form active tetramers competent for cytoplasmic UPR signalling.

Eukaryotic chemotaxis involves distinct cell shapes, with movement in shallow gradients dominated by split pseudopods and a single, broad leading edge in steep gradients, but little is known about the significance of these modes. In this thesis, I demonstrate that the shape of aggregating Dictyostelium discoideum cells is important for chemotaxis at the fundamental limit of gradient sensing. Using Fourier shape descriptors, I show that Dictyostelium cells occupy a naturally low-dimensional space of shapes, and that these cell shapes depend on the external environment. I present evidence that this space is restricted by treatments with a phospholipase A2 inhibitor, which is known to inhibit chemotaxis. I show that biophysical simulations can recreate wild-type chemotaxis and shape behaviour, and that restrictions to the shape of these simulations alone, with no change made to their biochemistry, are sufficient to recreate the drop in chemotactic accuracy seen in drug-treated live cells. I then discuss further applications of physical principles to understanding cell shape and chemotaxis, and the application of shape analysis to other areas of cell biology, specifically to the formation of immune synapses in T cells.

Modelling haematopoietic stem cells (HSCs) mathematically allows us to probe their behaviour, test hypotheses and make predictions. HSCs are essential and elusive; despite many recent advances much remains unknown, including how the microenvironment (niche) influences HSC behaviour, and the nature of complex interactions between HSCs, the niche and disease. This thesis comprises three sections. In the first we present new methods for the analysis of ODE systems that allow us to characterise steady state properties such as the probability of a state being stable (over some parameter range). We apply these methods to models of stem cell dynamics that differ in the form of regulation (feedback) imposed on the system and study the steady states of these systems. Competition within the niche between HSCs and invading cancer cells disrupts the system. In the second section we model this using ODEs that describe the population dynamics of cellular species from an ecological perspective. From the analysis of two models differing in their treatment of the niche, using Bayesian inference, we find that maintaining a viable HSC population is necessary and almost sufficient in order to outcompete leukaemia and restore healthy haematopoiesis. In the third section we extend the analysis of interactions between HSCs, leukaemia and the niche: performing a comparison of three models that describe the dynamics of chronic myeloid leukaemia. We study heterogeneous data from a clinical trial of the disease. All models fit these data, but do so in different ways. One model is discarded due to its unrealistic predictions, suggesting that direct competition between species is important. Each of the remaining two models makes testable predictions, but validating these is at current experimental limits. In the future, the results presented will aid experimental design and provide a framework from which to identify new therapeutic targets for haematopoietic diseases.

Natural Killer (NK) cells interact with other cells through a structured interface, the immune synapse (IS). A balance of signals controls NK cell activity through ligation of activating and inhibitory receptors. If signaling favours activation, NK cells mediate the directed secretion of cytotoxic mediators, such as perforin (PFN). To test whether the IS also functions as a gasket to extracellular molecules, fluorescently labeled nanometer-scale dextrans of varying sizes were co-incubated with effector-target cell-cell conjugates. Quantitative fluorescence microscopy of synapses revealed that dextrans with hydrodynamic diameters ≥32 nm were excluded from activating synapses, whilst smaller dextrans could enter. Size-dependent exclusion required an intact filamentous actin scaffold, but not continuous reorganisation following synapse formation. Time-lapse microscopy further revealed that the synapse assembles in a zipper-like manner, clearing larger dextran from the synapse. In addition, monoclonal antibodies and low-density lipoproteins were also excluded from the IS, whereas smaller domain antibodies could penetrate. NK cells can lyse more than one target in series. Whilst it is known that, among other proteins, PFN is secreted to lyse diseased cells, the amount of PFN secreted by NK cells is currently unknown. To quantify PFN release following stimulation through NKG2D or CD16, NK cells were plated on protein-coated surfaces that could capture PFN. Quantification using fluorescence microscopy revealed that PFN secretion was analogue, varying with increased ligand density. Simulating serial killing, repeated stimulation decreased the amount of PFN secreted with sequential activation. Unexpectedly, CD16 stimulation following serial NKG2D ligation recovered this decrease in secretion, however a similar recovery was not seen under reciprocal conditions. These data show that the activating IS clears and excludes extracellular molecules, including antibodies, in a size-dependent manner. Further, NK cell PFN secretion is an analogue response that varies with both ligand density and the receptors ligated in series.

Pesticides remain a necessary component of many agricultural systems and used judiciously they can play an important role in Integrated Pest Management (IPM) programmes. The aim of the present study was to investigate factors influencing the differential toxicity of insecticides against a cosmopolitan insect pest of crucifer crops, the diamondback moth, Plutella xylostella, and its respective hymenopteran parasitoid, Cotesia vestalis. Such knowledge can help in the effective use of insecticides with biological control agents in IPM. Three insecticides regarded as being compatible with some natural enemies (abamectin, spinosad, indoxacarb) and a compound generally regarded as harmful to natural enemies (lambda-cyhalothrin) were examined. Similar tests were also carried out with the peach potato aphid Myzus persicae and its parasitoid Aphidius colemani due to the loss of the Cotesia vestalis culture. A comparative measure of the intrinsic toxicity of fresh deposits (Day 0) of insecticides on Chinese cabbage was determined for both pest and parasitoid species. Lambda-cyhalothrin and abamectin were the most toxic compounds against both pests and their parasitoids, while indoxacarb and spinosad were less toxic. Residual bioassays were conducted using sprayed plants maintained under glasshouse conditions for 0-28 days after insecticide application. Results indicated lambda-cyhalothrin was the most persistent compound and abamectin and spinosad the least persistent. A leaf wax stripping technique was used with bioassays to compare the distribution of insecticide residues between the epicuticular wax layer and underlying leaf tissues. Wax removal significantly reduced the toxicity of all insecticides. No-choice and choice behavioural assays were conducted for both parasitoid species with leaf discs treated with LC5 and LC50 levels of insecticides. Both parasitoids tended to avoid insecticide-treated leaves, giving preference to untreated leaves or the arena. Emergence of adult parasitoids from cocoons/mummies on insecticide-treated leaves was not significantly different from untreated controls. The results are discussed in terms of the bioavailability of insecticides to phytophagous and non-phytophagous insect species.

The epidemiology of D. septosporum in Britain, from its behaviour and effects in a forest stand to its spread and population biology was investigated in this thesis. The peak infection period for the pathogen was determined to be spring and summer (June to September) albeit with low levels of infection occurring throughout the year. Contrary to previous belief the pathogen was found to have more than one life cycle per year in England. This is undoubtedly a factor contributing to the severity of the disease seen in this country. Such high disease levels across multiple years reduce the length of needles on affected trees. This reduced needle length compounds the effects necrosis and premature defoliation have on photosynthetic capacity and contributes to the reduced main stem increment seen in DNB affected trees. Once needles are shed from the tree the fungus does not survive for extended periods of time. In needles that reach the forest floor 90% of infective propagules die within three months whereas this figure rises to six months in needles that remain lodged in the canopy. Nonetheless, the pathogen is successfully dispersed far greater distances than previously believed. Infection was detected over 1,400 m from an inoculum source, which is over five times the greatest distance previously reported. Microsatellite analysis of isolates from across Britain revealed six distinct population clusters: two restricted to Scotland, two predominantly in England and Wales and two occurring throughout Britain. The two England and Wales populations had considerable overlap with two populations from Brittany, France, suggesting that some degree of pathogen exchange had occurred between the two countries. Furthermore, one of the Scottish population clusters was found to have links with western North America, the origin of its dominant host, lodgepole pine, indicating its introduction from this area into Scotland. Together the findings presented in this thesis can be used to guide improved management practices of dothistroma needle blight in Britain.

The successful development of consolidated bioprocessing requires microorganisms capable of degrading lignocellulosic biomass and fermenting the resulting sugars. Commercial cellulases and hemicellulases are currently being used to access these sugars, adding to the cost of producing useful products from lignocellulose. This study reports the enrichment of thermophilic, miscanthus degrading bacterial cultures from a municipal composting facility. The detected and isolated bacteria were characterized by 16S rRNA gene sequence analysis and were mostly Chitinophagaceae family, Meiothermus spp. and Geobacillus spp. Other isolated species included Cohnella spp., Brevibacillus sp., Chelatococcus spp., Thermobacillus spp., Thermoanaerobacterium spp., Thermobispora bispora, Bacillus spp., Staphylococcus sp. and Micrococcus sp. After enrichment, the mixed population was able to degrade greater than 50% of an ammonium hydroxide pre-treated Miscanthus x giganteus sample (1 g) over a six week incubation period at 55oC, with a reduction in the amounts of all components, including acid soluble and acid insoluble lignin. The glycoside hydrolases and other enzymes identified in the culture supernatants included endo-1,4-β-glucanase A, glucoamylase, xylan 1,4-β-xylosidase, xylose isomerase, xylulokinase, superoxide dismutase, transaldolase, Mn-catalase, Δ-1-pyrroline-5-carboxylate dehydrogenase and endo-β-N-acetylglucoseaminidase H. The HPLC analysis showed that fermentation products formate and lactate were present in the culture supernatant. Expression of an endoglycoside hydrolase (Csac_0137 from Caldicellulosiruptor saccharolyticus) gene in Geobacillus thermoglucosidasius strains, NCIMB 11955 and DL33, improved their β-glucosidase specific activity on cellobiose, and improved glycoside hydrolase activities of recombinant DL33 strain when grown on pre-treated M. x giganteus. Co-culturing of either transformed or wild-type NCIMB 11955 and DL33 with some of the isolated strains improved their glycoside hydrolase activity and growth on pretreated M. x giganteus.

G-protein-coupled receptors (GPCR) are transmembrane proteins that play a crucial role in the communication of cells with their external environment. In the last few years, several GPCR crystal structures have been solved using genetically engineered protein. The turkey β1-adrenergic receptor, the human neutrotensin 1 receptor and the adenosine A2A receptor (A2AR) structures involved the introduction of stabilizing mutations. The engineered mutant can be stabilized in an agonist or an antagonist bound conformation making the GPCR less flexible and therefore easier to crystallize. The aim of this study was to use functional characterization of the key thermostabilising mutants of the A2AR in order to understand the molecular basis of the thermostabilisation. The different mutants were characterized using a yeast-based growth assay, which measures down-stream signaling in response to agonist and radioligand binding analysis using both an agonist and an antagonist. Point mutations leading to a reduction/loss of constitutive receptor activity have been identified. In addition, a single point mutation abolishing the ability of receptor to bind the agonist NECA has also been identified. Conformational stabilization of the receptor is thus achieved by reducing basal activity along with modifying the ligand-binding pocket leading to inability to bind agonist. Such markers can be used to screen for stable mutants for structural characterization. Since thermostabilising mutations are not directly transferable across receptors, the yeast based growth assay could serve as a quick and inexpensive way to screen for mutations for a wide range of GPCRs.