The amount of F lac spisomal DNA present in an E. coli K12 F lac strain was measured as a percentage of total DNA at a number of growth rates by DNA/DNA hybridisation with DNA from a strain of Protsus mirabilis carrying the same F lac particle. The rate of change in average mass par call and average DNA content par call were Measured as a function of growth rats. The spisomal DNA was extracted as superhelical circles and the molecular weight measured from the contour length as seen under the electron microscope. Simple mathematical expressions were derived to interpret these data in terms of the time of replication of the episome relative to the time of initiation of chromosome repilaction, with two copies of the episome per chromosome terminus. Present models do not account than that to which the chromosome initiation is coupled end a new model is proposed.

In Escherichia coli thymidine phosphorylase, purine nucleoside phosphorylase, deoxyribomutase and deoxyriboaldolase are involved in the catabolism of nucleosides and deoxynucleosides and the sugar moiety is used as a carbon source. All four enzymes are induced by deoxynucleosides and the actual inducer appears to be dRib-5-P. Purine nucleoside phosphorylase and deoxyribomutase are also induced by purine ribonucleosides. Overwhelming evidence is available which favours the catabolic role of these enzymes. Four genes viz. tpp, dra, drm and pup specifying thymidine phosphorylase, deoxyriboaldolase, deoxyribomutase and purine nucleoside phosphorylase respectively were mapped by transduction with phage P1. All pairs showed greater than 90% co-transduction. The gene order was found to be dra-tpp-drm-pup and the gene cluster lies between the hsp and serB loci on the chromosome map of E.coli. A regulatory mutant which leads to constitutive synthesis of at least three of the four enzymes (and perhaps the fourth, deoxyribomutase) was analysed. The regulatory gene is designated nucR and is located near galE. The amount of thymine required for growth (colony formation) of thy- strains is affected by the nucR mutation. The amount required by the thy- drm- strain is reduced about four fold if it carries the constitutivity mutation. The amount required by a thy- drm+ strain is increased at least two fold. These differences in nutritional requirement provide a method for selecting constitutive from non-constitutive strains and vice versa. By this method a thymidine phosphorylase constitutive mutant was selected. The data obtained from the analysis of this second regulatory mutant indicate that it is similar to an Oc type mutation of the lac operon. The mutation of the putative operator, which is located to the left of the dra, renders only thymidine phosphorylase and deoxyriboaldolase constitutive. Purine nucleoside phosphorylase is not affected by this mutation and therefore it is believed that the four genes constitute at least two operons, one operon containing dra and tpp and the other pup and drm. A double mutant of E.coli was produced which lacks cytosine deaminase activity and requires pyrimidine for growth. This mutant is not able to grow on cytosine as sole pyrimidine source. This observation indicates that the only pathway involved in the incorporation of exogenous cytosine is via its first step conversion to uracil. A gene, udp, specifying uridine phosphorylase and a gene, cod, specifying cytosine deaminase were located at 74-75 and around 87 minutes respectively on the E.coli linkage map. Fluorouracil was used to isolate various mutants lacking nucleoside and deoxynucleoside phosphorylase activities. Using these mutants, it was shown that there are three routes via which FU can be converted to compounds which inhibit cell growth.

The 2Um plasmid of Saccharomyces cerevisiae is a stable multi-copy plasmid encoding copy number amplification and partitioning functions dedicated solely to its own maintenance. Although the role of these regions has been identified, the mechanism of partitioning and the precise relationship between amplification and partitioning is still unknown. At cell division plasmids may be distributed equally between mother and daughter cells, or there may be an unequal distribution resulting in a broad range of copy numbers. The amplification mechanism may play an important role in correcting any partitioning error. Investigations into the nature of partitioning and the co-ordination of partitioning and amplification require the measurement of copy number in individual cells. The aim of this work was to develop both indirect and direct methods to measure copy number in single cells, and to use these methods to analyse the distribution of plasmid molecules amongst individual cells. A feasibility study of the use of a gene product as a measurement of copy number suggested that an indirect enzyme assay would not be sensitive enough to detect low copy numbers. A direct measurement of nucleic acid sequences by in situ hybridization was developed for single yeast cells using both radio-labelled probes and non-radioactive probes detected by fluorescence. The method was specific, sensitive and provided the first direct evidence of a nuclear location for the 2Um plasmid. A significant difference was detected in the pattern of hybridization between budded and unbudded cells containing the wild-type 2Um plasmid and between wild-type cells and those containing a variant plasmid defective in the amplification mechanism. This indicates that the amplification mechanism may have a role in plasmid maintenance. Use of non-radioactive in situ hybridization revealed a distribution of fluorescence intensity amongst cells, suggesting a range of plasmid copy numbers amongst the population. In view of these results a proposed segregation method is discussed.

Monoclonal anti-CHO chromosome scaffold, and polyclonal anti-centromere autoimmune antibodies were employed to investigate the molecular organisation of the centromeres of Chinese hamster and human chromosomes. The presence of a novel ring-like structure was revealed, which was conserved amongst the metaphase centromeres of two rodents, namely Chinese hamster and mouse, and in humans. In metaphase the antigen had a molecular weight of 170kD, this polypeptide may be processed from, or the stable degradation product of, a 200kD interphase precursor full-length polypeptide. The antigen could not be detected in interphase centromeres, but the ring-like structure was detected in chromosome scaffold preparations, indicating that the ring antigen resisted the 2M NaCl extraction, and hence, tightly bound the DNA of metaphase chromosomes. The possible relationship of the centromere antigen with another high molecular weight and major chromosome scaffold protein, DNA topoisomerase II, was also explored. The centromere antigen was also localised apparently to the interdigitating microtubules of the midzone, in anaphase; seemingly to the contractile ring at cytokinesis, and also to the interphase centrioles. The antigen was shown to be distinct from the microtubule-associated proteins and had non-identity with another high molecular weight protein component of the contractile ring, namely myosin. The dramatic relocation of the antigen, however, at the metaphase/anaphase transition suggests that it falls into a class of recently identified dual chromosomal and cytoplasmic components, termed the "passenger proteins". The ring antigen is postulated to be involved in holding the sister chromatids together by a strong interaction with the centromeric DNA through prometaphase to the metaphase-anaphase transition. It is suggested that the ring antigen is disengaged from the centromere prior to anaphase, and may be involved in the regulation of the metaphase-anaphase transition.

delta-Aminolaevulinate synthase (delta-ALAS) is the first enzyme in the haem biosynthetic pathway, catalysing the condensation of succinyl CoA and glycine to form delta-aminolaevulinic acid (delta-ALA). The enzyme delta-ALAS has been studied in a number of species including yeast, a facultative anaerobe, and multicellular eukaryotes. The primary structure of the protein from a number of species exhibit regions of highly conserved amino acids, and a nucleotide probe based on one region was designed and used to clone the structural gene for delta-ALAS from the filamentous fungus Aspergillus nidulans. The cloned gene, hemA, was sequenced and found to possess a single intron, of 64bp, located between 355 and 419nt in the gene; between amino acids 119 and 120 of the protein. The deduced protein sequence, of 648 amino acids, shows 64% identity over the carboxyl domain of the protein derived from the yeast HEM1 gene. The amino-terminal sequence consists of basic amino acids, believed to be involved in mitochondrial targeting, and is consistent with observations on other delta-ALAS proteins. One copy of the hemA gene was disrupted in a diploid strain of A. nidulans following the transformation with a suitable disruption vector. The recessive hemA-mutant strain was isolated by mitotic haploidisation and shown to have an absolute requirement for delta-ALA, not replaceable by haem supplied in the medium. Genetic analysis by mitotic haploidisation of a heterozygous diploid constructed between a hemA-mutant and a suitable mapping strain, located the hemA gene to chromosome VII of A. nidulans.

Changes in the regulatory sequences of the genes involved in development are thought to be important in the evolution of morphology. However, molecular coevolution between functionally interacting genetic elements allows sequence divergence to be tolerated whilst the functional interaction is maintained. Molecular coevolution can lead to species-specificity in the sequence basis underlying molecular interactions. The concentration-dependent activation of hunchback (hb) expression in the anterior half of the Drosophila melanogaster embryo by the gradient of bicoid (bcd) protein represents a primary step in the elaboration of pattern along the anterior- posterior axis, and this interaction is conserved in the housefly, Musca domestica. In order to investigate the possibility that the molecular basis of this interaction may have coevolved, the bed and hb genes have been partially sequenced from M. domestica and compared to those of D. melanogaster. Analysis of the putative M. domestica hb regulatory region identified three candidate bed binding sites, with a consensus sequence of TTTAATCC, rather than the TCTAATCC of D. melanogaster. Comparison of the bed sequences revealed 5 changes within the 60 amino acids of the homedomain. Hence, it is possible that M. domestica bcd may have a subtly altered binding specificity, pointing towards the possibility that the coordinated changes in the binding site sequences have elicited compensatory changes in the M. domestica bed homeodomain. Preliminary analyses have been made of the functional significances of the observed differences. Although the functional significance of the observed differences in the bed and hb genes is not fully understood, the possibility remains that the molecular nature of the interaction between bed and hb has diverged between M. domestica and D. melanogaster.

Staphylococcus aureus is an important opportunistic pathogen, capable of causing a wide range of diseases. This ability to colonise and infect a variety of different tissues is due to the number of virulence factors it can produce. These factors are tightly regulated so they are only expressed when required and allow rapid adaptation to changing environments. In other bacteria the RNA binding protein Hfq is important for growth, resistance to stresses and virulence. Hfq regulation occurs through RNA stability, processing and translation. However the role of Hfq in S. aureus is controversial as conflicting reports on the subject have been published. Preliminary work in our laboratory indicated a role for Hfq in the positive regulation by the DNA binding protein Fur but the mechanisms involved are unknown. Therefore the aim of this study is to identify targets for Hfq regulation and investigate how Fur is involved in this regulation. This work demonstrates that S. aureus Hfq, along with Fur, has a key role in the resistance to oxidative stress and the regulation of several important virulence genes including the immune evasion factor, Eap. Hfq was found to regulate eap expression at the post transcriptional level. In addition, both Hfq and Fur were found to regulate eap expression transcriptionally, possibly through regulation of sae, an important virulence gene regulator. Fe-Fur was shown to directly bind the sae promoters suggesting direct positive regulation of sae by Fur. But the mechanisms involved in Hfq regulation remain unclear as Hfq did not have a major affect eap or sae mRNA stability. The regulation by Hfq and Fur shows some strain variation indicating that other factors are involved. Therefore Hfq and Fur play key roles in S. aureus virulence regulation. Further understanding of this complex regulatory network may reveal new targets for antimicrobial development to combat this important pathogen.

The IncIl conjugative plasmid ColIb-P9 carries a psiB gene that prevents induction of the SOS response in host bacteria. This locus was found to be located 2.5 kb downstream of the ssb (single-stranded DNA- binding protein) gene in the leading region of ColIb. This portion of the plasmid is transferred first to the recipient cell during conjugation and is strikingly similar to part of the leading region of the otherwise distinct F plasmid. Determination of the nucleotide sequence of ColIb psiB demonstrated that the gene has 84% identity with the psiB gene of F. Promoterless lacZ fusions on ColIb were created to leading region genes ssb and psiB and to sog, a representative transfer gene. It was found that expression of all three genes is increased when the ColIb transfer system is derepressed, but ssb and psiB are expressed at a much lower level than sog. Expression of psiB and ssb is increased when the host cell is exposed to UV- irradiation or mitomycin C treatments. The DNA-damage inducibility of ssb and psiB is recA and lexA-independent showing that neither gene is a component of the SOS regulon. Expression of both psiB and ssb is strongly enhanced in conjugatively infected recipient cells. No enhanced synthesis of Sog polypeptides was detected following conjugation showing that the zygotic induction of ssb and psiB is not a general property of plasmid genes. The implication is that PsiB and SSB proteins function in the transconjugant cell, rather than in the primary donor. It has been proposed that PsiB acts to prevent triggering of the SOS response during conjugation by transferring single-stranded DNA. Consistent with this hypothesis, carriage of the psiB gene by ColIb was shown to prevent a low level of SOS induction following conjugation. Plasmids that carry ssb and psiB genes have replicons belonging to the RepFIC family. It is postulated that the trigger for SOS induction during conjugation may be generated during the initial replication of the plasmid in the newly infected recipient cell rather than by the process of single-stranded DNA transfer.

In eukaryotes, sophisticated mechanisms have been evolved to ensure that key events in the cell cycle take place at the correct time. The triggering of these events is largely controlled by transient increases in the concentration of intracellular free calcium ions [Ca2+ ]i. If these transient increases m [Ca2+ ]i are prevented by the use of drugs that either block the voltage operated calcium channels (VOCCs), inhibit the release of Ca2+ from the cells' internal reserves, or block the receptor proteins that bind intracellular Ca2+ ions, the cell cycle will stop at specific points. The cell cycle will only recommence following removal of the Ca2+ blockade. In bacteria, many genes have now been identified that are essential for specific cell cycle events. However, no regulator has been identified that actually controls the timing of these events. Many theoretical models have been developed to explain bacterial cell cycle control, but there is little solid evidence to support them. A hypothesis has been developed that states that Ca2+ ions regulate the bacterial cell cycle in a manner analogous to eukaryotes. To test this hypothesis, temperature sensitive mutants of the E. coli strain, N43, resistant to the eukaryotic VOCC inhibitors verapamil and diltiazem were isolated, although detailed characterisation was confined to a verapamil resistant mutant, N43verA1. All the mutants are affected at a specific locus on the E. coli chromosome consisting either of deletions or major genetic rearrangements of three genes, hns, galU and hnrG (renamed rrx in this work). These genes do not constitute the expected receptor, but nevertheless, detailed examination of N43verA1 demonstrated that it appears to be unable to regulate its [Ca2+]I, since it is hypersensitive both to the concentration of Ca2+ in the medium and the Ca2+-chelator, EGTA. Moreover, when N43verA1 was transferred from the permissive to non-permissive temperature, it formed filaments, minicells and "chains of sausages", demonstrating some defect in cell division. An examination of the wild-type and mutant's responses to treatment with verapamil, EGTA or Ca2+ by labelling cell proteins with [35S-] methionine, indicated an extremely complex response. N43verA1 constitutively expresses a number of heat-stable proteins, which are further induced following treatment with verapamil or EGTA. Preliminary data suggests that some of the EGTA inducible proteins possess calmodulin-like properties. The three genes, hns, galU and rrx appear to interact, either at the genetic or protein level. Measurements of hns expression indicated that it is induced by verapamil, but not by EGTA. Overexpression of galU results in hyper-resistance to verapamil and Ca2+, which is modulated by hns and rrx, respectively. It is proposed that these genes in some way regulate the [Ca2+]i, in addition to, or in relation to their role in nucleoid organisation and control of the stationary phase sigma factor, RpoS.

Campylobacter jejuni is now recognised as a major cause of food-borne enteritis. Characteristics of Campylobacter spp. make them less amenable to genetic analysis using standard molecular biology techniques. These problems can prevent the isolation and characterisation of campylobacter loci. In the absence of an efficient method for the generation of random mutations, the only mutagenesis strategy available is that which targets specific genes. The two approaches described were developed specifically for the analysis of specific genes. The polymerase chain reaction (PCR) with degenerate oligonucleotide primers (PCRDOP/inverse PCR mutagenesis (IPCRM)) and inverse PCR strategies are designed to isolate and utilise small regions of DNA for the defined mutagenesis of targeted campylobacter genes. Moreover with the development of the IPCR mutagenesis, a small amount of sequence is the only requirement for the application of this approach. To evaluate the potential of both of these approaches, they were used to isolate and characterise potential virulence genes from C.jejuni. This study focused on the role of two classes of putative virulence determinants, stress response proteins of the HtrA family and response regulator proteins which are involved in the adaptation of bacteria to differing environmental stresses. A defined htrA mutant was constructed and phenotypic analysis established that this gene is not involved in the response to either temperature or oxidative stress, unlike many other homologues of HtrA. Three response regulator genes were isolated during this study, one of which is thought to encode CheY. Attempts to construct regX2 and regX3 mutants were unsuccessful. The flanking regions of htrA, regX2 and regX3 were isolated to facilitate the further characterisation of these loci. The work described in this study has verified that the PCRDOP/IPCRM and IPCR methodologies represent powerful approaches for the genetic analysis of Campylobacter spp.