No description available.

No description available.

No description available.

Accessory hydrogen uptake genes have been identified in a region of the Azotobacter chroococcum genome about 5 kb downstream of the hydrogenase structural genes (hupSL). DNA sequencing has revealed six genes (hupABYCDE) in this region. These genes are probably transcribed in the same direction as hupSL but are probably in a different operon. Mutational analysis had shown that disruption of the hupB, hupY, hupD and hupE genes gives a Hup$ sp-$ phenotype. In the present work additional mutational analysis, using Tn5, a Tn5 -derivative containing a promoterless lacZ gene, and a kanamycin resistance gene, confirms the direction of transcription and the separate nature of the hupABYCDE operon, and extends the region known to be necessary for Hup activity to hupA and possibly to 1.6 kb upstream of hupA.

V-ATPases are enzymes found in all eukaryotic cells. They are organized into a peripheral membrane complex (V1) and an integral membrane complex (V0). VI is responsible for ATP hydrolysis and generates the energy used by Vo to pump protons from the cytosol into the vacuole. Subunit d is a component of Vo possibly located at the interface between V 1 and V. in the V-ATPase complex. We hypothesize that subunit d could be involved in the structural and functional coupling of VI and Vo. This was tested by generating point mutations along the open reading frame of subunit d from yeast. The mutations F94A, H128A, D173A, D217A, D261A, E317A, W325A, E328A and C329A, all in conserved regions of the protein sequence, were characterized by examining their growth phenotype and by assessing their ATPase specific activity, proton transport and V1Vo assembly in purified vacuolar membranes. The mutations E317A, W325A, E328A and C329A had reduced ATPase and proton transport activities. In addition, V1Vo assembly was compromised by the mutation W325A. Our results suggest that residues at the carboxyl-end of subunit d are important for ATPase activity, proton pumping and V1Vo assembly at the membrane. / Department of Chemistry

The mannose-binding lectin from snowdrop (Galanthus nivalis agglutinin; GNA) was produced in Escherichia coli and purified as a functional protein after denturation/renaturation. Incorporation of the four extra C-terminal residues recently revealed from X-ray crystallographic data demonstrated that these residues increase binding to the glycoprotein carboxypeptidase Y. However, no differences in activities were observed in haemagglutination assays when compared to native GNA and toxicity towards rice brown planthopper (Nilaparvata lugens', BPH) in artificial diet bioassays was unaltered. Site-directed mutagenesis of the carbohydrate-binding site of GNA provided evidence of a direct correlation between the binding potential of GNA to BPH gut glycoprotein 'receptors' and the toxicity levels of GNA towards BPH nymphs. Functional recombinant plant lectins GNA and PHA (Phaseolus vulgaris agglutinin) were expressed in Pichia pastoris using native signal peptides or the Saccharomyces a-factor prepro-sequence to direct secretion. The a-factor prepro-sequence was inefficiently processed unless Glu-Ala repeats were added at the C-terminal end. In the latter case, removal of the Glu-Ala repeats was itself inefficient leading to recombinant lectins with heterogenous N-termini. In contrast, PHA expressed with the native signal peptide was secreted, correctly processed and fully functional. No expression of GNA from a construct containing the native GNA signal peptide was observed. The PHA-E signal peptide directed correct processing and secretion of both GNA and green fluorescent protein (GFP) when used in expression constructs in Pichia. A fusion protein containing both GNA and GFP (GNA-GFP) was expressed in Pichia pastoris. Simultaneous dual activities (i.e. carbohydrate binding and fluorescence) of recombinant GNA-GFP were demonstrated. Partial cleavage in the linker region resulted in co-purification of GNA which increased the binding activity of the fusion protein. Selective binding of GNA-GFP to haemocytes in the haemolymph of Lacanobia oleracea was observed, both in vitro and when the protein was fed to insects in diet.

Inorganic phosphate is an essential nutrient for all organisms. It is required for many cellular components as nucleic acids and phospholipids, and as energy-carrying compounds such as ATP. Thus, a regulated uptake of this pivotal nutrient is of outermost importance. Depending of the availability of phosphate in the surroundings the yeast Saccharomyces cerevisiae make use of two different systems for transporting phosphate into the interior of the cell: a low-affinity system that is active during surplus phosphate conditions and a high-affinity system that is active when the availability becomes limited. This thesis focuses on the high-affinity system, which is comprised of the Pho84 and Pho89 transporters. Of the two transporters, Pho84 is the predominant one, responsible for almost all phosphate uptake during low phosphate conditions, and the contribution of Pho89 is of minor importance. Hence Pho84 is by far the most well characterized phosphate transporter. Even though much is known about phosphate transporters in yeast little in known about how phosphate is transported. The work in this thesis aims to broaden the knowledge about the transport mechanism by the means of site-directed mutagenesis and functional characterization. Also the similarity of Pho84 to glucose sensors and the potential role of conserved residues in phosphate signaling are investigated. By the use of a high-affinity system deletion strain (∆Pho84 ∆Pho89), we also managed to investigate the functional importance of well conserved residues in Pho89. In summary: the work presented in this thesis has contributed to increase the knowledge about transport mechanisms in phosphate transporters.

Acyl-CoA:diacylglycerol acyltransferase (EC 2.3.1.20, DGAT or DAGAT) is a membrane protein found mainly in the endoplasmic reticulum (ER). It catalyzes the final step in the biosynthesis of triacylglyerol (TAG or TG), which is the principal repository of fatty acids for energy utilization and membrane formation. Several lines of evidence have indicated that DGAT has a substantial effect on carbon flux into TAG. DGAT has at least two discrete family members (DGAT1 and DGAT2) with different physiological roles. High-resolution structures of both DGATs, however, are absent due to difficulties in purification. In order to gain insight into structural and functional relationships of DGATs, a functional DGAT2 protein from the yeast Saccharomyces cerevisiae (ScDGAT2, also known as Dgalp) was selected. The structural and functional role of cysteine residues in ScDGAT2 was studied using site-directed mutagenesis (SDM) in combination with chemical modification. Although ScDGAT2 is susceptible to thiol-modifying reagents, none of the cysteines are essential for the catalytic activity or involved in structure support though disulfide linkages. Inhibition of DGAT activity by thiol-specific modification was localized to cysteine314, which is in the proximity of a highly conserved motif in DGAT2s. Thus, cysteine314 may reside in a crucial position near a possible active site or related to proper protein folding. The functional importance and topological orientation of signature motifs in ScDGAT2 were also studied using the same methods. Both the N- and C-termini of ScDGAT2 are oriented toward the cytosol. A highly conserved motif, 129YFP131, and a hydrophilic segment exclusive to ScDGAT2, reside in the ER and play essential roles in enzyme catalysis. In addition, the strongly conserved H195, which may be part of the active site of DGAT2, is likely embedded in the membrane. Although ScDGAT2 has a topology similar to that of murine DGAT2, there are striking differences which suggest that the topological organization of DGAT2 is not ubiquitously conserved. / Plant Science

Doctor of Philosophy (PhD) / The field of epigenetics emerged primarily from studies in Drosophila, and is now being studied intensively by mammalian biologists. In order to increase our knowledge of epigenetic gene control in the mouse, I have studied modifiers of epigenetic gene silencing. My main method of investigation involved the characterisation of mutants from a sensitised ENU mutagenesis screen performed previously in our laboratory. The screen was carried out in an FVB/NJ strain carrying a variegating GFP transgene expressed in erythrocytes. To date we have recovered 12 dominant (D) and seven recessive (R) mutant mouse lines from this screen that display altered transgene expression. We have named these Mommes (Modifiers of murine metastable epialleles). I investigated the phenotype and attempted to identify the underlying causative mutation of two of these Momme mutants. MommeD6 is a semi-dominant, homozygous lethal mutation that acts as a suppressor of variegation with respect to the GFP transgene. This mutation has a large effect on the level of expression of the transgene in expressing cells, but little effect on the percentage of cells expressing the transgene. MommeD6 is linked to a 2.5 Mbp interval on chromosome 14. MommeD9 is a semi-dominant, homozygous lethal mutation that acts as an enhancer of variegation with respect to the GFP transgene. Mutants have a tendency to become obese as they age, show abnormal haematology profiles, and females develop infertility. MommeD9 is linked to a 17.4 Mbp region on chromosome 7. I produced and studied a strain carrying the same GFP transgene but in a new strain background, C57BL/6J. This strain provided an opportunity to look for strain-specific modifiers of expression of the GFP transgene. Several regions were mapped to chromosomal locations. Further work will be needed to identify the genes involved. This mouse will be useful in future mutagenesis screens of this type.

Research Doctorate - Doctor of Philosophy (PhD) / Machine learning methods usually represent knowledge and hypotheses using attribute-value languages, principally because of their simplicity and demonstrated utility over a broad variety of problems. However, attribute-value languages have limited expressive power and for some problems the target function can only be expressed as an exhaustive conjunction of specific cases. Such problems are handled better with inductive logic programming (ILP) or relational reinforcement learning (RRL), which employ more expressive languages, typically languages over first-order logic. Methods developed within these fields generally extend upon attribute-value algorithms; however, many attribute-value algorithms that are potentially viable for RRL, the younger of the two fields, remain to be extended. This thesis investigates an approach to RRL derived from the learning classifier system XCS. In brief, the new system, FOXCS, generates, evaluates, and evolves a population of ``condition-action'' rules that are definite clauses over first-order logic. The rules are typically comprehensible enough to be understood by humans and can be inspected to determine the acquired principles. Key properties of FOXCS, which are inherited from XCS, are that it is general (applies to arbitrary Markov decision processes), model-free (rewards and state transitions are ``black box'' functions), and ``tabula rasa'' (the initial policy can be unspecified). Furthermore, in contrast to decision tree learning, its rule-based approach is ideal for incrementally learning expressions over first-order logic, a valuable characteristic for an RRL system. Perhaps the most novel aspect of FOXCS is its inductive component, which synthesizes evolutionary computation and first-order logic refinement for incremental learning. New evolutionary operators were developed because previous combinations of evolutionary computation and first-order logic were non-incremental. The effectiveness of the inductive component was empirically demonstrated by benchmarking on ILP tasks, which found that FOXCS produced hypotheses of comparable accuracy to several well-known ILP algorithms. Further benchmarking on RRL tasks found that the optimality of the policies learnt were at least comparable to those of existing RRL systems. Finally, a significant advantage of its use of variables in rules was demonstrated: unlike RRL systems that did not use variables, FOXCS, with appropriate extensions, learnt scalable policies that were genuinely independent of the dimensionality of the task environment.