No description available.

Quinolones are a clinically-useful class of antibacterial agents known to target DNA gyrase, a bacterial type II topoisomerase. Gyrase is unique among topoisomerases in its ability to introduce negative supercoils into DNA using the energy derived from ATP hydrolysis. The active enzyme is composed of two GyrA and two GyrB subunits, forming an A2B2 tetramer of molecular weight 374 kDa. The mechanism of supercoiling by gyrase involves the ATP-driven passage of one segment of DNA through a gyrase-stabilised double-stranded break in another. Tyrosine 122 of E. coli GyrA becomes covalently attached to DNA when gyrase breaks the phosphodiester bonds of DNA during supercoiling. When this residue is mutated to serine or phenylalanine, gyrase can no longer cleave or supercoil DNA, but can bind DNA normally. Rapid-gel filtration experiments have shown that quinolones can still bind to proteins bearing these mutations, suggesting that DNA cleavage by gyrase is not required for quinolone binding. Transcription by T7 and E. coli RNA polymerases is blocked by the presence of a gyrase-quinolone-DNA complex. Mapping of the transcription termination sites in the presence of gyrase and quinolones shows that blocking occurs about 10 to 20 base-pairs upstream of the gyrase cleavage site. Blocking of transcription by T7 RNA polymerase by a gyrase-quinolone complex on DNA does not occur when the active-site tyrosine of gyrase is mutated to serine, which indicates that the polymerase blocking requires DNA cleavage. Analysis of transcription in the absence of drug suggest that RNA polymerase does not displace gyrase from the template. DNA gyrase is also the target of the CcdB protein which is encoded by the F plasmid. When its action is not prevented by CcdA protein, CcdB is a potent cytotoxin. Using in vitro transcription by T7 RNA polymerase, it has been shown that CcdB complexed with gyrase can block transcription in a similar manner to the gyrase-quinolone complex. Furthermore, in the presence of CcdA, CcdB can no longer induce gyrase to block transcription.

A growing weight of experimental evidence indicates that all four secreted protein members of the cysteine-rich Dkk family, Dkk1, Dkk2, Dkk3 and Dkk4, can act as inhibitors of the Wnt pathway. To date, no structural and little functional information has been reported for Dkk4, which has been the focus of this thesis. Full length human Dkk4, including a C-terminal His6 tag, was successfully expressed in E.coli. An effective refolding and purification protocol was developed giving a final yield of 3.0 – 5.0 mg/L of homogenous, correctly folded protein. The activity of Dkk4 was assessed by investigating both its ability to bind the Wnt co-receptor LRP6, and to inhibit Wnt signalling using a variety of experiments including immunoprecipitation, FACS based cell surface binding and Wnt activated luciferase activity. Refolded Dkk4 showed well-dispersed 3D NMR spectra, which allowed the essentially complete, sequence specific assignment of the backbone atoms. The NMR data clearly indicated that both cysteine-rich domains adopt folded structures and strongly suggests a well-defined interface between the two. In addition, a novel interaction between Dkk4 and heparin was demonstrated using chemical shift perturbation experiments. The generation of models of the N-terminal region and C-terminal domain of Dkk4 allowed the perturbation data to be mapped, and revealed that a conformational change is likely to occur upon heparin binding. It is proposed that cell membrane proteins displaying heparan sulphate, a closely related molecule to heparin, may sequester Dkk4 yielding a high local concentration at the cell surface in order the control the Wnt cascade. Interestingly, Dkk4 may be able to interact with both heparin and LRP6 simultaneously rather than competitively.

Many micro-organisms catablise gluconate via the Entner-Doudoroff pathway. The key enzymes of this pathway are gluconate kinase, 6-phosphogluconate dehydratase (Edd) and 2-keto 5-deoxy 6-phosphogluconate (KDPG) aldolase. Biochemical and genetic techniques have been used to study gluconate utilisation in Escherichia coli K12. Mutants of the pathway have been isolated, their genetic lesions mapped and their physiological effects studied. The first step in gluconate utilisation by E. coli is its entry into the cells. It has been established that uptake is an active process induced by growth on gluconate, and appears to be the rate-limiting step in gluconate utilisation. Gluconate thus taken up is then phosphorylated to 6-phosphogluconato by ATP, catalysed by gluconate kinase: mutants completely devoid of this enzyme have not been obtained and it may be that there are two gluconate kinases in E. coli. 6-phosphogluconate can only be metabolised via the Entner-Doudoroff and pentose-phosphate pathways since Edd- mutants that also lack 6-phosphogluconate dehydrogenase (Gnd-) do Dot grow on gluconate but the Entner-Doudoroff pathway plays the predominant role. The importance of the KDPG aldolase that catalyses the cleavage of KDPG to pyruvate and glyceraldehyde 3-phosphate has been studied with mutants devoid of this enzyme. It has been established that KDPG is a very effective competitive inhibitor of 6-phosphogluconate for Gnd. The enzymes involved in the catabolism of gluconate are inducible. The likely inducer for the uptake system and gluconate kinase(s) is gluconate itself; 6-phosphogluconate probably induces Edd and Kga. Genetic analysis shows that at least three regions of coli chromosome (at 36, 66 and 85 min) contain genes involved in gluconate utilisation; only some of these genes are linked.

The products formed during transport in excised plant segments of radioactively labelled indolyl-3-acetic acid and 2,4-dichlorophenoxyacetic acid have been investigated. The techniques employed were extraction of the tissue followed by radiochromatography. It was found that indolyl-3- acetic acid is readily converted to indolyl-3-acetyl aspartate in Coleus tissue but not in Helianthus tissue. 2,4-dichlorophenoxyacetic acid appears to undergo little change. The distribution of radioactive material which results from the transport of labelled auxin in plant segments and small plants was also investigated by means of oxidation and subsequent scintillation counting. It was found that the system has a definite capacity and that much of the transported auxin is immobilised but not necessarily conjugated and that amounts of auxin reaching receiving systems is not a reliable basis for estimating auxin which is transported within tissue. Auxin transport into root primordia, lateral buds, abscission zones and through developmental transition zones is also briefly considered and it is reported that no barrier to the transport is encountered in these regions.

1. The phosphoenolpyruvate (PEP): fructose phosphotransferase system plays a key role in the utilization of fructose by Escherichia coli. Mutants devoid of, or altered in, this system are not only modified in their ability to phosphorylate fructose but are also altered in their ability to take up this hexose from solutions. 2. The major product of fructose phosphorylation in E. coli is fructose 1-phosphate, though fructose 6-phosphate can also be formed by a PEP-dependent mechanism. The relationship between PEP: fructose 1- and PEP: fructose 6-phosphotransferase activities is discussed. 3. PEP-dependent fructose phosphorylation, fructose uptake by whole cells and ATP-dependent fructose 1-phosphate phosphorylation to fructose 1, 6-diphosphate are activities that, in E. coli, are induced by fructose. A regulatory mutant that exhibits these activities constitutively has been isolated; its properties indicate that the expression of fructose-induced proteins is controlled as a regulon. 4. The utilization of fructose by E. coli is subject to control at the level of fructose entry into whole cells. Fructose uptake is inhibited by intracellular hexose phosphates or by carbon sources that can readily give rise to hexose phosphates, such as glucose or galactose. 5. Fructose 1-phosphate can serve as a sole carbon source for the growth of E. coli. Fructose 1-phosphate, like other hexose-phosphates such as glucose 6-phosphate and fructose 6-phosphate, is a substrate of an inducible hexose phosphate transport system. However, unlike glucose 6-phosphate, fructose 1-phosphate, when present in the growth medium of E. coli, does not give rise to the induction of this uptake system.

L-serine dehydratase from Arthrobacter globiformis has been purified over 1000-fold, to a point approaching homogeneity. An activity stain for the enzyme on poly-acrylamide gels was developed to follow this purification. The enzyme was characterised by a non-linear time course of activity in which the rate of pyruvate production slowly and progressively increased to a linear rate as the enzyme was activated. It is therefore a hysteretic enzyme (Frieden, 1970). The enzyme also had a sigmoidal substrate saturation curve and showed a requirement for cations. The concentration of L-serine affected the extent of activation whereas cations [e.g. Mg2+ or K+] affected both the rate and the extent of activation, but were not required for the catalytic activity of the enzyme. Two compounds, L-cysteine and D-serine, have been found to substitute for L-serine in the activation of L-serine dehydratase. Enzyme activated by these compounds or by L-serine showed a hyperbolic substrate saturation curve, had a linear time course of activity, and no requirement for cations. On activation, the molecular weight of the enzyme doubled. If the concentration of the activator was then reduced, the enzyme dissociated to the monomer form. These studies show that the activated enzyme was always in the dimer form and the non-activated enzyme was in the monomer form. An examination of the activation process showed it to be a second order reaction with respect to protein concentration, thereby identifying dimerisation as the slow step and the cause of the hysteretic response. A scheme, based on the equilibrium model for 'allosteric' proteins, has been presented to describe the roles played by L-serine, cations and protein in the activation of the enzyme. Investigations of the activated enzyme showed it to have the same properties as described by Bridgeland (1968) for L-serine dehydratase in toluene-tested whole cells. This suggests that the differences observed in the properties of the enzyme in vivo and in vitro are due to the maintenance of the enzyme in the activated form within the cell.

Alternative splicing of tropomyosin pre-mRNA: control in non-muscle cells. Ian R. Graham The human tropomyosin gene hTMnm contains a pair of mutually exclusive exons, NM and SK, which are used in non-muscle and skeletal muscle cells, respectively. I have undertaken an analysis of the factors affecting the splicing of these exons in the non-muscle cell line COS-1. I used a strategy involving mutation of the gene, followed by recloning of the appropriate region into a mammalian expression vector containing a tropomyosin cDNA clone. The wild-type and mutant mini-genes were transfected into the cell line, and the RNA produced after 48 hours' expression was isolated, then analysed by S1 nuclease protection mapping, and by a reverse transcriptase-polymerase chain reaction (RT-PCR) process. The results showed that exons NM and SK are not in competition in this non-muscle cell line; rather, I have shown that exon SK is not recognised as a splicing substrate when any other exons are present that can be used instead. Improvement, by mutation, of the branchpoint associated with exon SK restored use of that exon, as did replacement of the extreme 5' and 3' ends of the exon with the corresponding sequences from exon NM. The observation that exon SK is still overlooked by the cell's splicing apparatus, when it is placed in the exact context normally occupied by exon NM, strongly suggests that the exon itself is contributing to its deficiency. I have proposed a model in which the poor branchpoint sequence and elements within exon SK are responsible for preventing its recognition in the non-muscle cell, which is overcome, in skeletal muscle, by stimulation of the exon 4 to exon SK splice. Additionally, by studying the alternative splicing of the chick a-actinin gene, I have attempted to compare the regulation of splicing in smooth and skeletal muscle.

Vinculin and talin are components of adherens type cell/ECM junctions (adhesion plaques) where they are thought to form part of the linkage between the cytoplasmic domain of the beta-subunit of integrins and the actin cytoskeleton. Vinculin has been shown to interact with the adhesion plaque proteins talin, paxillin and beta-actinin, and talin has been shown to interact with vinculin, integrins beta1 and beta3, and actin. This study has examined the domain structure of vinculin and talin with regard to the interaction between them. Saturation binding analysis using purified chicken vinculin and talin demonstrated that the interaction was biphasic, with two dissociation constants of 3x10.;-8Mand 5.5x10.;-7M. The lower affinity interaction indicated that multiple bindingsites existed, with three moles of vinculin binding per mole of talin. The high affinity interaction was substoichiometric, and this remains to be explained. To localise the talin binding domain in vinculin, contiguous regions of the molecule were expressed as fusion proteins in E. coli and tested for binding iodinated talin in solid phase assays. Binding was localised to the N-terminal 258 amino acids of vinculin, and no further truncation of this region left detectable talin binding activity. This region was found to bind talin as effectively as intact vinculin, and no evidence was obtained to suggest that short linear sequences within this region could account for the interaction. Two independently isolated vinculin cDNA clones, one of which lacked 123bp of coding sequence, had suggested that alternative splicing of the vinculin mRNA within the region encoding the talin binding domain could be a mechanism for regulating talin binding. Though this region was found to be contained on a single exon within the chicken vinculin gene, no evidence was found to authenticate the existence of such a spliced message using a reverse transcriptase/PCR protocol to map vinculin transcripts. Vinculin binding sites were mapped in talin by expressing a series of overlapping talin fusion proteins and investigating their ability to bind vinculin in solid phase assays. This identified three non-overlapping regions of talin within its 190kDa domain which bound vinculin with dissociation constants around 10-7M. Two of these were within residues 498-950, whilst the most C-terminal mapped to between residues 1554-2268. Further experiments on the most C-terminal of these sites suggested that a 40 amino acid sequence was able to bind vinculin.

No description available.