Endogenous cannabinoids, such as anandamide and 2-arachidonoyl-glycerol, are compounds naturally produced in the body that act as important signalling modulators, particularly, in the brain. Endocannabinoids bind to cannabinoid receptors (the CB1 and CB2 receptors), the same receptors to which the psychoactive tetrahydrocannabinol ingredient of Cannabis saliva derivatives bind. Endocannabinoids, together with their receptors, and the protein network involved in endocannabinoid synthesis, uptake, release and degradation constitute the endogenous cannabinoid system. The endogenous cannabinoid system is an emerging target in pharmacotherapy for the treatment of several diseases. In the context of a project aimed at studying important proteins of the ECS attention was focused on a 19kDa protein called Cannabinoid Receptor Interacting Protein la (CRIP1a). CRIP1a has been recently discovered, using a yeast-two-hybrid approach, as a binding partner of the CB1 receptor, but not the CB2 receptor. CRIP1a has been suggested to bind to the C-terminal cytoplasmatic tail of the CB1 receptor and to reduce the CB1 receptor-mediated tonic inhibition of voltage-gated calcium currents. The CRIP1a rat ortholog, RnCRIP1a, was over-expressed in E. coli and purified to homogeneity. Single crystals of wild-type and selenomethionine-labelled RnCRIP1a were grown using the microseeding technique and diffracted to high resolution using synchrotron radiation. The structure of RnCRlP1a was solved using the multiple anomalous diffraction technique at a resolution of 1.8 Å. RnCRIP1a is a domain-swapped dimer in the crystal. Its C-terminal region is in an extended β-strand conformation, 'crossing over' to engage in a β-sheet symmetry-related molecule, thus resulting in the formation of a homodimer. Analytical ultracentrifugation and NMR experiments show, however, that RnCRIP1a is a monomer in solution. On the basis of homonuclear NOE signals, in solution RnCRIP1a appears to have its C-terminus folded back into the β-sandwich. In consideration that RnCRIP1a C-terminal region harbours a PDZ-binding consensus sequence, the possibility of a functional role for the opening of its C-terminal is intriguing. The RnCRIP1a monomer displays a modified immunoglobulin fold equipped with a 'lid' domain above its β-sandwich scaffold. From a topological point of view, RnCRIP1a resembles members of the Ig-fold family of DNA binding proteins. RnCRIP1a has been proposed to bind to the last 55 amino acids of the CB1 receptor. This was tested by in-vitro pull-down and NMR titration experiments and failed any interaction between RnCRIP1a and the C-terminal tail of the CB1 receptor. Subcellular localisation assessed by confocal microscopy using COS7 cells reveals that RnCRIP1a is predominantly a cytoplasmatic protein.

Wiskott-Aldrich Syndrome (WAS) is a rare X-linked recessive primary immunodeficiency characterised by eczema, thrombocytopaenia and immunodeficiency. WAS encodes for a haematopoietic restricted protein, WASp, involved in transduction of signals from the cell membrane to the actin cytoskeleton. Mutations lead to impaired actin dynamics in response to stimuli and are seen as defects of receptor capping, chemotaxis, phagocytosis and proliferation. This leads to susceptibility to pyogenic, viral and opportunistic infections and increased incidence of lymphoproliferative disease and malignancy. Due to the high morbidity and mortality associated with mismatched transplantation, WAS is considered a good target for gene therapy. We were able to reconstitute a murine model of WAS using a gamma retroviral vector, with reconstitution of specialised actin structures, podosomes, as a functional readout. Reservations concerning the safety of such vectors, following adverse events in a clinical trial, led to the development of third generation self-inactivating lentiviral vectors. Under the promotion of CMV, SFFV LTR or short sequences of the proximal endogenous Wiskott-promoter, WASp was able to restore cytoskeletal abnormalities in dendritic cells in-vitro from WASp-deficient mice. In addition we have demonstrated stable expression of WASp in T cell, B cell and myeloid lineages following transduction and engraftment of lineage negative murine bone marrow (up to 9 months) using both an SFFV LTR and endogenous promoter sequences. We have demonstrated the ability to restore cytoskeletal abnormalities and proliferative responses from the reconstituted mice. Recently a novel mutation in WAS has led to the discovery of a constitutively active WASp, with a novel monocytopaenic and neutropaenic phenotype. Here we characterised the patient phenotype of a novel constitutively active WASpn94T mutant. In addition to monocytopaenia and neutropaenia, there was abnormal cytoskeletal actin, manifesting itself as abnormal podosome distribution and an inability to phagocytose or produce an oxidative burst to physiological stimuli. There was also increased apoptosis in the bone marrow and evidence of genomic instability. Utilising lentiviral vectors, WASpI294T was expressed in a cell line to elucidate the possible mechanisms responsible for the patient phenotype. These studies demonstrate the efficacy and feasibility of a lentiviral vector-mediated gene therapy strategy for WAS using endogenous promoters where a more regulated expression level may be achieved.

The entire adult central nervous system (CNS) derives from an initially small population of apparently homogeneous neuroepithelial precursor cells (NEPs) which produce specific differentiating cell types in a highly organised fashion with respect to both the time and place at which they are generated. A unifying phenomenon throughout the CNS is that neurons are always generated before glia. To better understand what mechanisms might be involved, the dynamics of precursor lineages need to be described. Here, single NEPs from the murine dorsal embryonic neocortex were cultured at clonal density and filmed using time-lapse microscopy to monitor their divisions over time. The progeny they gave rise to were identified by immunocytochemical methods and expression of the oligodendrocyte lineage-affiliated transcription factor, olig2 was directly observed by using a transgenic mouse that expressed enhanced green fluorescent protein (EGFP) in olig2-expressing cells. (The transgenic mouse line was created by phage artificial chromosome (PAC) transgenesis). This data enabled lineage trees for individual clones to be retrospectively drawn to include the timing of olig2 expression alongside the final identification of the daughter cells produced. In this way, the effects of different growth factors with respect to the induction of glial in preference to neuronal phenotypes were assessed. Using this system it was possible to determine what cell types could be derived from a single precursor and with what pattern within a lineage olig2 might be expressed under different culture conditions. Both FGF-2 and a Sonic Hedgehog agonist were seen to produce mixed clones in which olig2 was transcribed at early branch points within a lineage and later down-regulated in a selection of daughter cells. This means that olig2 expression does not denote commitment to the oligodendrocyte lineage and, furthermore, that induction is a sporadic event which seems to be dictated at the level of the individual progenitor cells rather than by an intrinsic cell-timer dictated within the original NEP.

Flavin-containing monooxygenase 3 (FM03) is a hepatic, microsomal enzyme important in the Phase I metabolism of xenobiotics. Numerous single nucleotide polymorphisms (SNPs) in FM03 have been identified and shown to affect the catalytic activity of the enzyme, with some destroying the activity altogether, causing the distressing disorder, Trimethyalminuria (TMAU), in humans. Genotyping was performed at the FM03 locus on several individuals displaying symptoms of TMAU. Two new FM03 variants, R238Q and R492Q, were discovered in this investigation, with kinetic studies suggesting that R238Q destroys FM03 catalysis. Kinetic studies of FM03 common polymorphisms showed that the E158K/E308G displayed significantly reduced catalytic activity compared with either single variant alone. E158K/V257M also displayed significantly reduced catalytic activity, but not compared to either single variant alone. The pharmacogenetic implications are discussed with particular reference to the recent finding that E158K/E308G has been associated with reduced polyp formation in patients with familial adenomatous polyposis who were treated with sulindac sulphide, an FM03 substrate. The evolutionary history of FM03 was probed using sequence data of genomic DNA in a Japanese cohort of potential TMAU sufferers and control individuals. Mutational relationships among haplotypes were inferred and time depth of the variation and ages of individual mutations were estimated by coalescent analysis, with test statistics used to detect departure from neutral evolution. A case of balancing selection is proposed at the FM03 locus. In an attempt to understand the structural and biophysical consequences of FM03 variants, a homology model of FM03 was generated, refined and validated. Flavin adenine dinucleotide (FAD), an FM03 cofactor, was also modeled into the FM03 model and the interactions between the enzyme and cofactor predicted.

Mad3p is a component of the spindle checkpoint in <i>Saccharomyces cerevisiae.</i> Mad3p is known to be in a constitutive complex with another checkpoint component Bub3p and upon checkpoint activation it interacts with Mad2p and Cdc20p to form the mitotic checkpoint complex. Recent analysis also suggests that the Mad proteins, in addition to the direct inhibition of Cdc20p through formation of the mitotic checkpoint complex, have an additional role in the down-regulation of Cdc20p levels in the cell. The results presented in this thesis show Mad3p to be a protein that is phosphorylated during every mitosis and consequently upon spindle checkpoint activation. Such phosphorylation does not require any other checkpoint components. Experiments also show that Mad3p is phosphorylated on serine 337 by Ipllp kinase <i>in vitro </i>and the phosphorylation of Mad3p <i>in vivo, </i>due to an absence of tension at the kinetochore, requires Ipllp. Mad3p contains two KEN box motifs that are conserved between species. Evidence in this thesis demonstrates that mutation of either of these KEN boxes renders cells sensitive to microtubule depolymerising drugs. Further analysis of these mutants shows that Cdc20p binding is reduced in both but they can both still bind Bub3p. Mutation of the N-terminal KEN box increases the half-­life of Cdc20p and consequently the total levels of Cdc20p in the cell. Cdc20p turnover is also reduced in the absence of Bublp and Bub3p, but to a lesser extent. Mutation of the second KEN box in Mad3p leads to a dominant phenotype that perturbs mitosis even in the presence of wild­type Mad3p.

Base pair mismatches may be formed during replication or genetic recombination. Not all mismatched base pairs are repaired with the same efficiency, in particular the A.G. mismatch is consistently under repaired. Why the A.G. mismatch is poorly repaired is not clear, although previous work has suggested the mismatch is conformationally flexible. In the present study the A.G. mismatch in the sequence dCGCAAATTGGCG was examined by X-ray crystallography and pH dependant UV melting studies. Crystals were grown at pH 6.6. The base pair has an A (<i>anti</i>).G(<i>syn</i>) conformation. From the pH dependant UV melting studies it is likely this is a protonated AH<SUP></SUP>&43 (<i>anti</i>).G(<i>syn</i>) base pair formed by protonation of the N(1)- atom of adenine. The G (<i>syn</i>) conformation is stabilised by hydrogen bonding to a network of solvent molecules in the major and minor grooves. A parallel investigation of the A.l mismatch was performed. The oligonucleotide dCGCAAATTIGCG was crystallised. An A (<i>anti</i>).l(<i>syn</i>) base pair may be indicated, pH dependent UV melting profiles suggest the existence of a protonated base pair. 5-Bromouracile is strongly mutagenic forming mismatches with G. The G.BrU mismatch has been reported to form a wobble base pair although other forms are possible. The oligonucleotide dCGCGAATTBrUGCCG was synthesised and used in pH dependent UV melting studies. These suggest that there is little or no contribution to the base pair from non-wobble, ionised or enolised, forms. Frame shift mutations may arise from nonmatched bases in DNA. A series of tridecameras was synthesised for analysis by X-ray crystallography; dCGCGTAATTCGCG, dCGCGBrUAATTCGCG, dCGCGTAATBRUCGCG and dCGCGTAATTBrUCGCG. It was hoped to determine the looped out, stacked or nonmatched nature of the additional bases. 2-Amino-2'-deoxyadenosine (dA') is an analogue of dA. Base pairs formed between A' and T contain three hydrogen bonds and are A.T like in the major groove and G.C like in the minor groove. The additional hydrogen bonding affords extra stability to A'. T base pairs relative to A.T base pairs. Z-DNA is formed by d(GC)<SUB>n</SUB> sequences but not by d(AT)<SUB>n</SUB> sequences. The mixed sequence oligonucleotide dCA'TA'TG was made for X-ray crystallographic investigation and UV melting studies. It was hoped this would crystallise as Z-DNA. Recently X-DNA has been investigated. This can be formed by poly(dA-dT) under stringent conditions due to the effects of salt or alcohols. X-DNA was first identified by its CD spectrum. Similar CD spectra have been observed for poly(dA'-dT) closer to physiological conditions. The oligonucleotides d(TA')<SUB>n</SUB> n = 2,3,4 were synthesised for X-ray crystallography, d(TA')<SUB>4</SUB> was also investigated by CD spectroscopy. This showed the ability of the oligonucleotide to form X-DNA. The additional stability of dA' oligonucleotides gives them potential used as stable short genetic probes. A series of oligonucleotides was made, these hybridised successfully under conditions where the native sequences failed to. A new phosporamidite for the synthesis of dA' oligonucleotides was developed. This overcomes the problem of acid catalysed depurination of dA' residues during solid phase phosphoramidite DNA synthesis, di-n-butylformamide dimethylacetal was used to protect the N<SUP>6</SUP> function. This was shown to be stable to the detritylation conditions encountered in DNA synthesis.

Long DNA palindromes are sites of genome instability (deletions, gene amplification and translocations) in eukaryotic cells. In both prokaryotic and eukaryotic cells they are sites of DNA breakage that can be repaired by homologous recombination. Genetic evidence suggests that in <i>E. coli</i> breakage is mediated by the SbcCD complex (Rad50/Mre11) that can cleave DNA-hairpin structures. Here an arabinose inducible-<i>sbcDC (p_BAD-sbcDC) </i>system has been constructed and used to obtain genetic and <i>in vivo</i> physical evidence of DNA double-strand breaks (DSBs) at a 246 bp interrupted palindrome. These breaks are shown to have two ends suggesting that they occur behind the replication fork probably on the lagging strand. A two-ended break is not compatible with the palindrome blocking DNA replication and cleavage causing the fork to collapse. Repair of the breaks requires homologous recombination facilitated by the RecA and RecBCD proteins as well as the Holliday junction resolves RuvABC and RecG. Contrary to previous reports this study rules out a significant role of the RecFOR proteins in the repair of breaks and implicates a role for the PriA protein – probably to establish replication forks at the break site. Removing the XerCD/<i>dif</i> system (chromosome dimer resolution) has a minor negative effect on cell viability suggesting that chromosome crossover events occur at a detectable (but unquantifiable) frequency. There is also a requirement for cleavable LexA protein demonstrating that the SOS response is required for efficient repair.

The role of ornithine decarboxylase (ODC) was investigated in muscle and embryos of different chicken lines selected and unselected for increased growth and weight-for-age. Both lines showed a peak in muscle ODC activity at 5-6 days post hatch which can be up to 4-fold higher in the selected line. Use of α-difluoromethylornithine (DFMO) to try to inhibit this activity in the selected line produced birds which had significantly lower body weight, but not muscle weight or muscle ODC activity, than control birds. It seems that treatment may have been started too late (4 days post-hatch) to have a significant effect on muscle ODC activity, but still affected overall growth rate. ODC activity in embryos showed a peak at 4-5 days. This was found to be highest in embryos from strains which showed fastest growth rates at this particular stage, but this did not necessarily reflect the pattern found in post-hatch growth. In all strains examined at this stage the ODC activity was found to be higher in the head region than in the rest of the embryo. A cDNA clone was isolated from a chicken embryo cDNA library using a mouse ODC probe. The DNA sequence of this clone was found to be 78% homologous to the human ODC sequence and almost 90% homologous at the amino acid level. This amount of sequence conservation suggests an extremely important role for this enzyme. This cDNA hybridized to several bands in chicken genomic DNA in a way which suggests that there is only a single ODC gene in chickens. It also detected a Haelll polymorphism between broiler and layer strains which is at the 3' end of the gene. Probing Northern blots with this probe indicated the presence of two transcripts, which closely correspond both to the size predicted from the cDNA clone and to the size of mRNAs found in other species. It is suggested that many pathways of investigation are opened up by the work presented here. Further investigation might involve use of the chicken ODC cDNA to investigate the cause of the differences in ODC activity between strains, cloning of the chicken ODC gene, and the use of transgenics to study the role of ODC in growth and development. Myoblasts and satellite cells in culture will also provide a useful system for studying ODC activity and expression and its involvement in muscle growth.

The aim of this work was to express the Ca2+ sensitive photoprotein, apoaequorin, in filamentous fungi to a high level in order to allow routine and simple measurement of changes in cytosolic Ca2+ ([Ca2+]c) concentration in living hyphae. Analysis of the codon bias of the available apoaequorin genes showed a large discrepancy with the codon usage of the target fungi, Neurospora crassa, Aspergillus niger and A. awamori. To overcome this discrepancy, a synthetic apoaequorin gene with optimal codon usage for Neurospora and Aspergillus was designed and synthesised. An expression vector was generated for the production of apoaequorin from the synthetic gene using a constitutive Aspergillus promoter and suitable selectable markers for transformation. An expression vector was also produced for use in N. crassa to allow comparison of the aequorin levels obtained between the synthetic gene and the native gene. Strains of A. niger and A. awamori were transformed with the Aspergillus apoaequorin expression vector, and transformants expressing large amounts of active aequorin were selected and purified. Transformants of N. crassa were also produced using the Neurospora expression vector, and the highest expression transformants isolated and purified. Comparisons of the aequorin expression levels obtained in these transformants showed a 200-fold higher level of expression in A. awamori compared with that obtained in N. crassa. Also, the comparison between native and synthetic aequorin production showed a 280-fold higher level in the synthetic aequorin transformant. One high expression transformant from each Aspergillus species was chosen for investigation of [Ca2+]c responses to a variety of external stimuli. A successful method of growing cultures in small volume, still liquid culture in 96-well microtitre plates was developed which allowed efficient analysis of [Ca2+]c in growing cultures. Using such cultures, [Ca2+]c was monitored during germination, growth and conidiation. A suitable age of culture (18-36 h) was determined for investigation of changes in [Ca2+]c in response to external stimuli. The first stimulus investigated was the application of high extracellular concentration (50 mM) of Ca2+, which elicited a large [Ca2+]c increase.

In many animals female meiotic spindles lack centrosomes. In the absence of centrosomes the chromosomes drive spindle formation. This process is not well understood so I examined two mutants from a screen for female sterile mutations with acentrosomal spindle defects. I found that the first mutant, <i>remnants (rem), </i>disrupted spindle morphology, chromosome alignment and microtubule dynamics in non-activated oocytes. <i>rem </i>encodes Cks30A a conserved subunit of Cdc2. Essential pole proteins, msps and D-TACC were often mislocalised to the equator and Cks30A is involved in modification of D-TACC. The second mutant, <i>msps-like</i>, had tripolar spindles in non-activated oocytes. I found that <i>msps-like</i> encodes Weel, a negative regulator of Cdc2 activity. Wee1 is also required for modification of D-TACC in a Cdc2 dependent manner. Twine, the positive regulator of Cdc2, is essential for spindle unification/metaphase arrest of non-activated oocytes and Cdc2 itself is required for correct spindle morphology in female meiosis. Cdc2, its subunit, Cks30A, and its regulators, Weel and Twine, are all required for ancentrosomal spindle formation. Cdc2 is a central figure in a kinase pathway required for multiple facets of acentrosomal spindle formation. To gain new insights in the formation of female meiotic spindles I took part in a screen for metaphase I arrested spindle mutants in female meiosis on the X chromosome of <i>Drosophila</i>. I found mutants defective in chromosome alignment, spindle morphology, spindle unification and spindle formation around individual chromosomes. I characterised several of the mutants found in this screen as well as established complementation groups for all of the mutants discovered to aid in future studies.