Shigella spp. are pathogenic variants of Escherichia coli that cause bacillary dysentery, resulting in over 1 million deaths per year. Across E. coli bacteria, the dynamic process of acquisition and loss of GIs, especially those associated with virulence (pathogenicity islands [PAIs]) is a driving force behind the emergence of new pathogenic strains. In Shigella only 5 GIs have been well characterised and there is currently no effective vaccine. Therefore the development of an efficient screen to detect GIs in unsequenced Shigella strains could be highly informative. Nineteen Shigella strains were probed for the presence of GIs using a high throughput PCR screen (tRIP), across 16 tRNA gene integration hotspots. Putative GIs were then investigated using a chromosome walking technique (SGSP-PCR). Representative PCR amplicons were sequenced to get a snapshot of the islands contents. Islands of particular interest were characterised further using allelic exchange and marker rescue to capture clones that harbour larger portions of the GI and subsequent sequencing and analysis. Using SGSP-PCR, 81% of the putative GI occupied tRNA loci were characterised, and sequencing analysis found they all contain island DNA, indicating that tRIP followed by SGSP-PCR is a robust strategy for GI discovery in unsequenced strains; also this method should be applicable to a broad range of microorganisms. At least 54% of the islands identified harbour phage-like integrase genes, strongly supporting the notion that many of these elements arose following acquisition of horizontally acquired integrative GIs. The frequent presence of integrase genes also highlights the potential role of bacteriophage in the original and/or ongoing dissemination of island DNA in Shigella. Only one novel GI was discovered; it has classic prophage-like features and contains completely novel DNA, indicating that while Shigella has a plastic genome, it is a highly specialised human pathogen that has undergone considerable pathoadaptive genome reduction. The major development from this study is evidence that a number of key Shigella virulence determinants are independently mobile and not only localised to a single family of islands; this significantly increases their potential to spread by HGT across Shigella and could contribute to the rapid emergence of new endemic strains.

Staphylococcus aureus is an important pathogen of both humans and animals. With the growing spread of MRSA strains in clinical environments and the wider community, it is imperative that we understand the basic physiology of this species to find new antimicrobial or vaccine targets. This project focuses on the function and regulation of glyceraldehyde-3-phosphate dehydrogenase, or GAPDH, an essential component of glucose metabolism. S. aureus contains two GAPDH homologues; GapA, a known GAPDH protein and GapB, the function of which is undefined. Using a number of complementary approaches we have shown that GapA and GapB have reciprocal functions during glucose metabolism, and that both homologues are required during infection. We have also identified novel “moonlighting” roles for both GapA and GapB and shown that both genes are regulated by divalent metal ions. Interestingly iron regulation of gapA is strain variable at the transcriptional level due to sequence variation between strains, and appears to involve a S. aureus repeat (STAR) element locus. To our knowledge this is the first indication that these repeat elements are functional. STAR elements are found across the S. aureus genome, and the number of repeats at each locus is variable between strains. Surprisingly the three STAR loci (gapR, hprK and orf0733) analysed are highly conserved and maintained in the S. aureus genome at a level similar to that of MLST loci which suggests that they may have an important function, although transcriptional analysis failed to identify a correlation between repeat number and transcript levels. However transcriptional analysis did demonstrate that a number of STAR and non-STAR associated loci in S. aureus, including housekeeping genes involved in central metabolism, known transcriptional regulators and virulence factors show strain variable levels of expression, which may play a role in the significant adaptability of S. aureus.

The Saccharomyces cerevisiae THI5 gene family consists of four highly conserved members; THIS, THI11, THI12 and THI13. Each member is located within the subtelomeric region of a different chromosome and they share these regions with members of other gene families. A detailed analysis of the genome environment of each THI5 family member has shown all four to exist with their neighbouring genes in a highly conserved arrangement. A survey of THI5 copy number among a selection of hemiascomycetes suggests that the existence of THIS as a gene family is exclusive to those yeasts of the Saccharomyces sensu stricto complex. Furthermore, the association and organisation with respect to neighbouring members of other subtelomeric gene families appears to be conserved between S. cerevisiae and other sensu stricto yeasts. This conservation of local gene order does not appear to extend to yeasts outside of this subgroup. S. cerevisiae strains have been constructed which contain deletions of THI5, THI11, THI12 and THI13 in all possible combinations to yield the single, double, triple and quadruple mutants. Phenotypic analysis of the quadruple deletion mutant has found it to exhibit thiamin (vitamin B1) auxotrophy on medium containing glucose as the carbon source. This auxotrophy can be remedied by the exogenous addition of thiamin or one of its precursors hydroxymethyl-pyrimidine (HMP). Analysis of other mutant strains has shown that the THI5 gene family members are functionally redundant, with each encoded isozyme having an apparent equal role in HMP formation from pyridoxine (vitamin B6). Physiological studies of these mutant strains have examined the regulation of each of the four genes by thiamin and its precursors, as well as investigating a recently proposed anaerobic HMP biosynthetic pathway.

No description available.

Until 1983, third generation cephalosporins (3GC) were thought to be resistant to hydrolysis by all plasmid-mediated β-lactamases. However, TEM and SHV derived β-lactamases have recently evolved which can confer transferable resistance to 3GC. Six novel plasmid β-lactamases, which confer transferable resistance to 3GC have been identified and characterised. They have been compared directly to the other 3GC hydrolysing β-lactamases discovered elsewhere. The criteria implemented to distinguish these different β-lactamases were: Km and Vmax values for the hydrolysis of different substrates, molecular weight, isoelectric focusing point (pi) and susceptibility to inhibitors. The β-lactamases TEM-E1, TEM-E2, TEM-E3, and TEM-E4, were all TEM derived and, although they conferred a greater transferable resistance to ceftazidime than cefotaxime, they hydrolysed both these substrates with similar efficiencies. TEM-E2 was produced by an organism which was isolated in Liverpool in 1982 and is, consequently, the first example of transferable 3GC resistance. TEM-E3 was produced by clinical isolates from two different London hospitals, and found to be the same as IBM-10 which was discovered in the USA after the characterisation of TEM-E3. When ceftazidime was used as a selecting agent, TEM-E1 and TEM-E2, could be obtained spontaneously from a TEM-1 producing E.coli and, in the same way, TEM-E4 could be obtained from a TEM-2 producing organism. Utilising the same methodology a β-lactamase, which conferred transferable resistance to ceftazidime, was obtained from PSE-4, although such an enzyme has not yet been reported in clinical isolates. The fifth novel β-lactamase, E8825 (pi 7.9), was produced by an organism isolated in India which also produced TEM-1 and CAZ-6. This was the first example of transferable 3GC resistance in Asia and the first report of two transferable 3GC hydrolysing β-lactamases being encoded by the same plasmid. The characteristics of the sixth novel enzyme, DJP-1, suggest that it was originally an E.coli chromosomal β-lactamase. Analysis of this strain revealed that the β-lactamase (pi greater than 8.2) conferred transferable resistance to all β-lactam antibiotics and clavulanic acid. Finally, two novel methods for the isolation and purification of (3-lactamases were developed during these studies. Electro-dialysis of 0-lactamases from polyacrylamide gel allowed rapid purification of TEM-E2 from TEM-1. Fast Liquid Protein Chromatography (FPLC System) was employed to separate β-lactamase E882S from the other β-lactamases produced by the host strain. In addition it was illustrated that different β-lactam molecules can induce a variety of different β-lactamase satellite bands which can be visualized by isoelectric focusing. Moreover, it was also verified that these satellite bands were variants of the main β-lactamase protein.

The copper-containing protein superoxide dismutase is required for the virulence of C. albicans in a mouse model. Previous work in our laboratory has shown that copper uptake and regulation in C. albicans has some similarities to Saccharomyces cerevisiae, including the activation of the copper transporter gene CaCTR1 in low copper conditions by the transcription factor CaMac1p. However, further analysis in this study has demonstrated that the actual mechanism of regulation by CaMac1p is different from that of S. cerevisiae Mac1p.;This thesis demonstrates for the first time that the CaMAC1 gene is transcriptionally autoregulated in a copper-dependent manner. This is in contrast to the S. cerevisiae MAC1 homologue, which is constitutively transcribed. The presence of one binding site for CaMac1p in the promoters of CaCTR1, CaMAC1 and the ferric/cupric reductase gene CaFRE7 is sufficient for copper-responsive regulation. In contrast, two promoter elements are essential for normal levels of copper-dependent activation by S. cerevisiae Mac1p. CaMac1p is also involved in the regulation of the iron-responsive transcriptional repressor gene SFU1 and the alternative oxidase gene AOX2. This work describes key features of the copper uptake system in the human pathogen C. albicans that distinguishes it from similar processes in the model yeast S. cerevisiae. Transcriptional autoregulation of the CaMAC1 gene could enable C. albicans to respond more precisely to environmental changes, conferring an adaptation to the human host that may be an advantage in the disease process.

No description available.

No description available.

No description available.

The over-expression of the Saccharomyces cerevisiae pyruvate kinase gene (PYK1) is limited at the level of transcription and translation (Moore et al, 1990a). High levels of PYK1 mRNA are not tolerated by the transformants which grow up to 5.2-fold slower than the untransformed host strain. In addition, the transformants are genetically unstable, reverting at high frequency to fast growing cells via plasmid loss (Moore et al., 1990c). A novel experimental system was designed to facilitate the analysis of the translational regulation of the PYK1 gene. This system had two components designed to overcome the problems associated with the use of a multi-copy PYK1 plasmid. The first component was a PYK-lacZ reporter gene under the transcriptional control of the PGK1 promoter and terminator sequences. The translational regulation of the PYK1 gene could therefore be studied independently of transcription via -galactosidase assays. The PYK1 5'-leader regulated the translation of the reporter gene. A control reporter gene was also constructed containing a 'neutral' 5'-leader sequence. The reporter genes were used to confirm that the PYK1 5'-leader was both necessary and sufficient to mediate the translational regulation. The second component was a GAL-regulable multi-copy PAL-PYK plasmid. This gene is strongly induced by galactose but was found not to be tightly repressed by glucose. This lack of tight repression restricts the use of this plasmid in analyses. The over-expression of the PAL-PYK mRNA, however, did not impose the predicted dosage limitation on the PYK-lacZ reporter gene. It was shown that the sequence at the 5'-end of the PYK1 mRNA and/or the position of the 5'-cap relative to the translation start codon is critical to the function of the PYK1 5'-leader in mediating the translational regulation. The implications of this result on the mechanism of translational regulation are discussed.