House mouse haemoglobins are encoded by two unlinked clusters of alpha- and beta-globin genes. The beta-globin polypeptides arising from three well-defined distinct adult beta-globin alleles (s, d and p) can be distinguished electrophoretically. In wild populations these exist as a discrete s/d polymorphism in Europe and a d/p polymorphism in Asia. DNA restriction site haplotypes of these beta-globin genes were investigated. A unique s haplotype was found in all European subspecies and inbred mice, but the distinct d and p haplotypes of inbred mice are likely to be just a two sample subset from a much wider spectrum of the variable diffuse- like (d and p) DNA haplotypes found in wild mice. In English mice the discrete s and d adult haplotypes appear to extend to the embryonic and beta-related sequences. Restriction mapping, cloning and DNA heteroduplex analysis of the inbred adult beta-globin genes have shown that the differences between s, d and p haplotypes point to a remarkable DNA sequence divergence within a single species. The distinct pattern of variation between all three haplotypes consists of long invariant regions (involving both coding and flanking sequences) separated by a variable region(s) that has undergone major rearrangements and DNA sequence divergence. From these results and DNA sequencing data (assuming rodent DNA molecular evolution rates) a chronology of mouse evolution can be proposed. The adult beta-globin gene was duplicated and the two copies last corrected against each other about 10-35 MA. The s and diffuse-like haplotypes began to diverge about 2-4 MA and subsequently the diffuse-like haplotype appears to have diverged in its beta-minor gene component, after perhaps a localized gene correction around 1 MA. Three models of (i) bottleneck, (ii) allopatric and (iii) sympatric evolution have been proposed to explain the distinct pattern of DNA divergence and the maintenance of the beta-globin polymorphisms.

The Aspergillus nidulans 3-phosphoglycerate kinase (PGK) gene has been isolated from a phage ? genomic library using the equivalent yeast gene as a hybridization probe and the location of the PGK gene physically mapped within the cloned DNA. The nucleotide sequence of the entire PGK gene and its 5' and 3' flanking regions has been determined. The gene was found to contain two 57 base pair introns which have splicing signals similar to those in other filamentous fungi, and codes for a 421 amino acid protein with considerable homology to the Saccharomyces cerevisiae (68%) and mammalian (64%) proteins. Almost total conservation is found in Aspergillus of residues thought to be important for the structure and function of the yeast enzyme. The codon usage shows a similar bias to that found in other filamentous fungi. The PGK gene has been shown to be constitutively expressed at a relatively high level. The transcription start site and polyadenylation sites have been determined. The PGK promoter has both the CAAT and TATA homologies generally found in more complex eukaryotes. Additionally two pyrimidine rich regions of the promoter sequence share similarities to other highly expressed genes in fungi. PGK mRNA exhibts 3' heterogeneity and the major polyadenylation site is positioned 16 bp beyond the eukaryotic consensus polyadenylation signal AAUAAA. Potential hairpin structures are also found in the 3' non-translated region of the PGK mRNA which may be important in transcription termination and polyadenylation. In parallel studies the pyruvate carboxylase protein has been identified by immunoprecipitation from the products of in vitro translations of A. nidulans mRNA as the first step of a strategy to clone a cDNA copy of the PYC mRNA. An attempt to clone the pyruvate carboxylase (PYC) gene from Aspergillus nidulans, by an immunological screen of an Aspergillus genomic library constructed in phage ?, was unproductive.

The human beta-globin gene family, situated on chromosome 11, consists of five functional genes (epsilon Ggamma, Agamma, delta and beta) and a non-processed pseudogene, Psibeta1. The major work undertaken in this thesis consisted of a phylogenetic analysis of the non-processed pseudogene of the human globin cluster in order to establish the tempo and mode of evolution of such sequences, their suitability as examples of more general non-coding DNA sequence evolution and their possible influences on eukaryotic multigene family evolution. This study of contemporary Psibeta1 pseudogene sequences has shown that this gene has been a stable component of the beta-globin gene cluster during the evolution of the primate, and other, mammalian orders. The pseudogene was probably functional early in primate evolution and silenced probably before the basal primate radiation -70 million years ago. The presence of a functional orthologue to the human Psibeta1 gene in the goat (the ?II gene) supports the view that the human Psibeta1 gene was functional prior to its silencing early in primate evolution. After silencing, the primate Psibeta1 pseudogene has evolved randomly in terms of base substitution and insertion/deletion at a mean rate thought to be representative of non-functional non-coding DNA sequences throughout the primates. These conclusions are supported by the mode and tempo of non-coding DNA sequence evolution observed within the functional brown lemur beta-globin gene. However, the tempo of primate Psibeta1 gene evolution conflicts with views concerning the universal constant rate of neutral evolution, the rate of non-coding DNA evolution having apparently decreased to varying extents within the different lineages of this mammalian order. The consequences for primate beta-globin gene cluster evolution of the presence of a non-processed pseudogene are discussed. The distinct nature of the gene in the human beta-globin gene cluster, the history of the Psibeta1 gene in the primates and the presence of sequences related to Psibeta1 in various other mammalian orders suggests an additional ancient genetic locus was present in the ancestral beta-globin gene cluster prior to the mammalian radiation. In order to acknowledge the distinct nature of this locus the human Psibeta1 gene has therefore been renamed n and other contemporary n-related sequences renamed accordingly. The evolution and gene orthologies of the other non-primate beta-globin gene families are discussed in the light of the phylogenetic analysis of the human Psin pseudogene. The simplest interpretation of the evolution of contemporary mammalian beta-globin gene clusters is that they resulted from a common minimal ancestral cluster composed of proto epsilon-, gamma-, eta-, delta- and beta- like sequences. The generality of the conclusions drawn from this work concerning pseudogene longevity and sequence evolution after silencing await the phylogenetic analysis of other pseudogene sequences. It is apparent however that pseudogenes may constitute another potential source of genetic variation on which the processes of natural selection can act in the evolution of both eukaryotic multigene families and the genome in general.

The first three steps in the utilization of quinic acid in Aspergillus nidulans are catalyzed by highly inducible enzymes encoded by a gene cluster containing three enzyme structural genes and two trans-acting regulatory genes. The gene cluster has been mapped meiotically between the ORNB and FWA genes in linkage group VIII. The regulatory gene QUTD codes for an activator protein, whose function seems to depend upon interaction with the product of the QUTA gene. The QUTA protein thus has a complex role, acting both in repression of gene expression in the absence of the inducer quinic acid and in mediating the role of the QUTD activator protein, upon induction. To test models for the regulation of expression of the QUT enzyme structural genes, the gene cluster has been physically isolated from an A. nidulans genomic DNA library in an X phage replacement vector, using DNA fragments from the equivalent genes in Neurospora crassa as heterologous hybridization probes. An unique 13.2 Kb DNA sequence from A. nidulans was thus isolated and the three structural genes mapped within a 3.5 Kb BamHI fragment. An adjacent region of the cloned DNA shows hybridization to the N. crassa activator gene, QA1F, thus possibly identifying the QUTD gene. No hybridization could be detected with the N. crassa repressor gene, QA1S, which was initially suggested to be equivalent to the QUTA gene of A. nidulans. Evidence against the hypothesis that the QUTA gene encodes a protein with only properties of a repressor is discussed. The QUTE gene which codes for dehydroquinase has been sequenced. Major rearrangements in the coding sequence are found in comparison with the equivalent gene of Neurospora. Analysis of the promoter region of QUTE reveals that it contains a CAAT sequence exhibiting perfect homology with the consensus sequence for eukaryotes, and in close proximity 5' to a sequence showing a dyad of symmetry. The dyad exhibits homology to sequences in other fungal genes identified as binding activator proteins. The importance of these two promoter elements in the control of the expression of the QUTE gene is discussed.

The Sp diastereoisomer of adenosine 5'-O-(l-thiotriphosphate) (ATPaS) was found to be incorporated into RNA transcribed by T7 RNA polymerase with an apparent KM similar to that for ATP; the Rp diastereoisomer was neither a substrate nor a competitive inhibitor. The configuration of the thiophosphodiester link in the RNA produced was analysed with stereospecific nucleases. Surprisingly, the nucleases exhibited reduced discrimination compared to their activity on dinucleotides. The results show that phosphorothioate linkages in T7 RNA polymerase transcripts are in the Rp configuration. Thus, the transcription reaction proceeds with inversion of configuration at phosphorus. In vitro transcription by T7 RNA polymerase in the presence of a nucleoside 5'-O-(1-thiotriphosphate) has been used to prepare pre-mRNA analogues of the small intron of a rabbit ?-globin gene and flanking exon sequences. Incubation of transcripts prepared with ATPaS in a HeLa cell nuclear extract showed that the presence of the thionucleotide in a transcript inhibited splicing, but a novel product (termed E2*) was formed by cleavage three nucleotides upstream of the 3' splice site. This product was formed with the same kinetics as the intermediates of a normal splicing reaction, and its formation depended on the presence of ATP, Mg2+, and intact small nuclear RNAs U1, U2, and U6. Hence, E2* formation would seem to be a splicing related phenomenon. However, E2* formation does not require all the components necessary for in vitro splicing. E2* formation was supported by S100 extracts and mildly heat treated nuclear extracts, both of which were inactive in splicing. Retardation gel asssays and combined RNase T1 digestion and immunoprecipitation experiments indicated that large spliceosome-like complexes did not form on transcripts prepared with ATPaS. Furthermore, neither the absence of a functional 5' splice site nor polypyrimidine tract sequence prevented E2* formation, despite both these sequences being required for splicing. In addition, a nuclease activity has been identified in HeLa nuclear extract which is dependent on intact small nuclear RNAs U1 and U2. This activity may be involved in E2* formation.

This thesis describes the development of expression-plasmid vectors for use in the mouse, myeloma cell-line, J558L. Following introduction of the plasmids into J558L using spheroplast fusion, the vectors allow selection for stably-transfected cells. Selective survival depends on expression from a gpt gene in the plasmid, which encodes the enzyme, xanthine-guanine phosphoribosyltransferase (XGPRT). Expression of XGPRT confers, on J558L, a dominant selectable-resistance to mycophenolic acid in the presence of xanthine and hypoxanthine. The resultant stable gpt+-transfectants can then be screened for expression of a protein encoded by a non-selected gene in the expression plasmid. Initially, a systematic study was carried out to compare the effect on XGPRT expression of different expression elements placed upstream of the gpt coding sequence, in single-gene plasmids. Comparative expression levels were estimated by comparing the stable gpt+-transfection frequencies of J558L obtained with each plasmid. This study demonstrated the importance of the IgH-gene enhancer for obtaining high-level expression from a transfected gene. A combination of an IgH enhancer and a promoter element, upstream of the gpt coding-sequence, resulted in the highest levels of expression, but the type of promoter used was of only secondary importance. Identified combinations of effective upstream elements were then incorporated into plasmid vectors, which were constructed for expression from heterologous genes encoding proteins of interest. Stable gpt+- transfectants containing a non-selected gene were only obtained following transfection with plasmids which also contained the gpt gene. This was in contrast to situations in which a non-selected gene was cotransfected with a gpt gene which resided on a different plasmid; here, none of the stable gpt+-transfectants screened had cointegrated the non-selected gene with the gpt gene. Different classes of the two-gene expression plasmids were constructed, which differed in the relative orientation and position of the transcription units. The efficiency of the various plasmids was estimated by measuring the expression levels from a model cDNA, encoding chicken lysozyme. Transfectants were isolated which express and secrete biologically-active chicken lysozyme, at levels greater, in molar terms, than the typical level of secretion of endogenous Ig from myeloma cells. In some cases, the stable gpt+-transfection frequencies obtained with two-gene plasmids were low, compared with the transfection frequencies obtained with plasmids containing only the gpt gene. It was proposed that this was due to transcriptional interference between the two genes on the same plasmid. As this phenomenon might also restrict the expression from the non-selected gene, overcoming this effect was considered important in optimising expression. However, attempts to identify and overcome the phenomenon were inconclusive and, consequently, a rational strategy for overcoming this potential limitation on expression levels was not obtained.

A gene family encoding msp36, a major product of the mouse submaxillary gland, was shown by Southern analysis and genomic cloning to contain approximately ten related genes. Restriction and heteroduplex mapping, allied to hybridisation of the cDNA to cloned genomic fragments, have elucidated the overall structures of some of the msp36-related genes, and in one case have produced a precise gene arrangement. The use of specific oligonucleotide probes has further refined these structures and demonstrated the close physical linkage of six of the genes in "head-to-head", "head-to-tail" and "tail-to-tail" configurations. They have also implied the presence of an intron-less msp36-related sequence which is probably a pseudogene. Investigations at the mRNA level have led to the conclusions that the expression of the msp36 genes is tissue-specific and not androgen-regulated. There appear to be two classes of mature transcript which may originate by differential splicing or, more likely, be the products of diferent genes. The function of msp36 is not known, but from the cDNA-derived primary protein sequence it seems likely that it is a secreted glycoprotein. Several possible identities are discounted on the basis of amino-acid composition, immunoprecipitation and the tissue-specificity of the expression.

Most secreted proteins are hydrophilic in character, so the question arises of how a cell exports a hydrophilic secreted protein, (e.g. Immunoglobulin), through the hydrophobic lipid bilayer of the cell membrane. Two approaches to the problem were undertaken in this study. Firstly, to generate and analyse mutants which were defective in some aspect of secreted or membrane bound IgG production, and secondly, to identify and characterise membrane bound IgG to determine what differences in the membrane bound molecule directed it to the membrane and not to secretion. A number of variants of the IgG producing cell line X63.Ag8 were isolated and tested for secreted or membrane bound IgG. Of 698 individual clones tested, 55 were found to be variant in some aspect of secreted or membrane bound IgG expression. These mutant cell lines were analysed, and an attempt was made to explain the findings within the context of the data obtained. Work done on the identification and characterisation of the membrane IgG revealed that the molecule was assymetric with respect to the Ig heavy chain. The molecule was anchored in the membrane by a hydrophobic extension at the C-terminal end of one heavy chain. The other heavy chain was of the same type as is found in the secreted IgG. Inhibition of glycosylation of the membrane bound IgG had no adverse effect on its expression at the cell surface. An attempt was made to rationalise all of these data within the confines of current theories concerning the mechanisms of protein secretion.

Colonization of gut mucosal surfaces by enteropathogenic Escherichia coli (EPEC) elicits a severe persistant diarrhoea in infants and young children without elaboration of enterotoxins or tissue invasion. The formation of a distinct ultrastructural lesion involving intimate adherence to cell surfaces, loss of microvilli and membrane perturbations, however, has for some time been recognized as an important component of pathogenesis, but the processes involved in lesion formation and its relevance to the disease state remained obscure. In this study intimate adherence of EPEC to surfaces of cultured cells has been shown to elicit specific changes reminiscent of cellular activation by various hormones and growth factors via signal transduction pathways. The formation of two second messengers, 1,4,5-inositol trisphosphate and diacylglycerol by the activity of phospholipase C (PLC) on membrane phosphatidylinositol lipids is known to promote release of calcium from intracellular stores and to activate protein kinase C (PKC) respectively. Therefore the observed intracellular calcium elevation, calcium-dependent actin accretion at sites of bacterial attachment, release of glycosyl-phosphatidylinositol (GPI) anchored proteins, eventual loss of host cell viability, and phosphorylation of target cell proteins on EPEC infection of cultured cells, are consistent with a mechanism of diarrhoeagenesis that involves activation of a host cell signal transduction pathway terminating in PLC activity. I propose that the rapid onset of severe secretory diarrhoea associated with EPEC infections is induced by phosphorylation of particular ion transport proteins in the microvillus membrane, mediated by calcium-dependent kinases and PKC. In addition reduction in the surface area of infected regions of the gut, which may itself enhance hypersecretion by preventing absorption, can be explained by calcium-dependent breakdown of the microvillus cytoskeleton.

Tandemly repeated tracts of DNA are a ubiquitous feature of eukaryote genomes. One class of tandem repeats, 'minisatellites', have been shown to be highly variable both in overall length and in the internal arrangement of variant versions of the repeating unit along the array. Consequently, both length and internal variation at these loci can be exploited to generate individual-specific profiles of use in forensic science and the establishment of family relationships. Recently it has been demonstrated that short dinucleotide repeats, or 'microsatellites', and other simple tandem repeat arrays can also show length variation. This work describes the isolation and characterization of simple tandem repeat arrays, and their application in a forensic science case. The evolutionary persistence of variability at tandem repeat loci is also explored. Simple tandem repeats isolated were frequently associated with other tandem repeats and interspersed repetitive elements, a phenomenon previously described for minisatellites and perhaps indicative that certain genomic regions show relaxed fidelity in the maintenance of large-scale DNA structure, allowing tandem array expansion and retroposon insertion. Although variability is low relative to minisatellites, microsatellites, owing to limited length, can readily be amplified from highly degraded DNA using the polymerase chain reaction. Consequently it was possible to identify positively the skeletal remains of a murder victim by comparing microsatellite profiles of the skeleton with those of the presumptive parents. Comparative studies of microsatellite and minisatellite loci between man and other primates indicate that in evolutionary terms, the variable state is reasonably persistent at microsatellite loci, whereas highly variable minisatellites show extreme evolutionary transience. Such transience was also demonstrated for two large 'midisatellite' loci, suggesting that highly variable tandem repeat loci are extremely unstable and transient, whereas lower variability leads to evolutionary persistence of the variable state.