Search

Soluble methane monooxygenase (sMMO) is a multicomponent bacterial enzyme that catalyzes the oxidation of methane to methanol, as well as oxidizing many other adventitious substrates. A number of mutagenic studies were carried out on the sMMO enzyme of Methylosinus trichosporium OB3b in order to gain insight into sMMO and probe how structural aspects relate to function of the enzyme. Leu110 within the hydroxylase a-subunit of sMMO has been proposed as a possible gating residue, controlling access of substrate to the active site (Rosenzweig et al. 1997). A range of site directed mutants were created at the 110 position and screened for activity with a number of aromatic substrates. All mutants showed relaxed regioselectivity with all substrates assayed. However no evidence of a gating residue was found, indicating that Leu110 is more important in determining regioselectivity than substrate access to the active site. Comparison to the highly similar butane monooxygenase led to the creation of three site directed mutants: M184V F282L and C151T. M184V and C151T showed small changes in regioselectivity and reduced activity with most substrates. The M184V mutant showed relaxed regioselectivity and a novel oxidation product with the substrate mesitylene which may have implications for substrate trafficking. The F282L mutant produced a stable enzyme which had no activity with any of the substrates tested, showing Phe282 is important for the enzyme function. A random mutagenesis experiment was devised and a colorimetric screen for the oxidation of triaromatic compounds was used to screen mutant libraries for activity towards anthracene and phenanthrene. However no activity towards triaromatic compounds was detected. In order to improve the cloning strategies and to make creation of mutant libraries easier, a novel expression vector pT2ML was created. The pT2ML vector reduces the number of cloning steps required to make soluble methane monooxygenase mutants. This expression system was used to make a site directed mutants F188Aand N116G in order to complement previous site directed mutant studies, as well as a recombinant wild type mutant in order to asses the activity of the new expression system which is comparible to the wild type enzyme.

The mammalian pyruvate dehydrogenase complex (PDC) is a key multi-enzyme assembly linking the glycolytic pathway to the TCA cycle via the specific conversion of pyruvate to acetyl CoA and, as such, is responsible for the maintenance of glucose homeostasis in humans. PDC comprises a central pentagonal dodecahedral core of 60 dihydrolipoamide acetyltransferase (E2) and 12 E3 binding protein (E3BP) subunits. Presently, two conflicting models of PDC (E2+E3BP) core organisation exist: the ‘addition’ (60+12) and ‘substitution’ (48+12) models. In addition to its catalytic role, the multi-domain E2/E3BP core provides the structural framework to which 30 pyruvate decarboxylase (E1) heterotetramers and 6-12 dihydrolipoamide dehydrogenase (E3) homodimers are proposed to bind at maximal occupancy. The formation of specific E2:E1 and E3BP:E3 subcomplexes are characteristic of eukaryotic PDCs and are critical for normal complex function. Despite the availability of limited structural data, the exact subunit organisation and mechanism of operation of the mammalian E2/E3BP core remains unknown. This thesis describes the large-scale purification of tagged, recombinant human PDC cores, full-length rE2 and rE2/E3BP, truncated E2/E3BP, peripheral rE3 enzyme as well as native E2/E3BP core (bE2/E3BP) purified from bovine heart. The ability to purify large amounts of pure protein has enabled the characterisation of the individual cores as well as the E2/E3BP:E3 complex using a variety of biochemical and biophysical techniques. Full-length rE2/E3BP, rE2, bE2/E3BP, truncated E2/E3BP (tLi19/tLi30) and rE2/E3BP:E3 were analysed in solution by analytical ultracentrifugation (AUC). While AUC of the cores supported the substitution model of core organisation, the stoichiometry of interaction was determined to be 2:1 (rE2/E3BP:E3). This was further complemented by gel filtration chromatography (GFC) and small angle neutron scattering (SANS), implying the possible existence of a network of E3 ‘cross-bridges’ linking pairs of E3BP molecules across the surface of the E2 core assembly. Low resolution solution structures obtained for rE2/E3BP, bE2/E3BP and tLi19/tLi30 by small angle x-ray scattering (SAXS) and SANS revealed the presence of icosahedral cores with open pentagonal faces favouring the substitution model of core organisation. These solution structures also indicated high structural similarity between the recombinant and native cores, as well as with the crystal structure obtained previously for the truncated bacterial E2 core. In addition, homology modelling and superimpositions of high- and low-resolution structures of the core revealed conservation of the overall pentagonal dodecahedral morphology despite evolutionary diversity. Evidence for the substitution model of core organisation was further substantiated by negative stain EM of the recombinant and bovine E2/E3BP cores. SANS stoichiometry data indicated the binding of 10 E3 dimers per E2/E3BP core. Although this could correspond to approximately 1:1 stoichiometry between E2/E3BP:E3, subsequent radiolabelling studies suggested possible variation in core subunit composition between the native and recombinant E2/E3BP cores. Therefore, as opposed to the 48E2+12E3BP substitution model based on AUC and SAXS studies with the recombinant E2/E3BP core, rE2/E3BP cores produced in this study indicated a higher level of incorporation of E3BPs with a maximum core composition of 40E2+20E3BP. On the basis of this new finding we have proposed the ‘variable E3BP substitution model’, wherein the number of E3BPs within the core can range from 0 to a maximum of 20, thus resulting in variable populations of E2/E3BP cores. Despite this core variability, the highly controlled regulatory mechanisms in vivo may bias the core composition towards an average of 48E2+12E3BP. However, as the over-expression of the recombinant E2/E3BP core in our study is not as tightly regulated as in vivo, higher number of E3BPs (>12) is observed to be integrated into the core. This new level of architectural complexity and variable subunit composition in mammalian PDC core organisation is likely to have important implications for the catalytic mechanism, overall complex efficiency and tissue-specific regulation by the intrinsic PDC kinases (PDKs) in normal and disease states. The E2 cores of the PDC family are known to be highly flexible, exhibiting inherent size variability reflective of the ‘breathing’ of the core. Integration of E3BP into the E2 core assembly would then be expected to have significant consequences for the structural assembly, affecting the ‘breathing’ and in turn the function and regulation of the complex. Unfolding studies to assess core stability via circular dichroism (CD) and tryptophan fluorescence revealed lower stability of the rE2/E3BP core as compared to cores composed exclusively of rE2 subunits, thus implying the contribution of E3BP towards core destabilisation. In addition, crosslinking studies indicated weak dimerisation of rE3BP, which may be a key factor promoting core destabilisation. The lower stability of the E2/E3BP core may be of benefit in mammals where sophisticated fine tuning is required to obtain cores with optimal catalytic and regulatory efficiencies. SAXS solution structures of E2/E3BP cores obtained were unable to locate the exact positions of E3BP within the core. However, SANS in combination with contrast matching of selectively deuterated components as well as cryo-EM, EM tomography and single molecule studies could be used in future for determination of the exact locations of E3BP, and validating the importance of E2/E3BP core organisation and subunit composition for overall PDC function and regulation.

This thesis reports the characterization of two members of a superfamily of enzymes, namely CYP51 from the human pathogen Mycobacterium tuberculosis and CYP116B1 from the soil bacterium Ralstonia metallidurans . CYP51 is one of twenty P450s encoded in the M. tuberculosis genome and has been demonstrated to demethylate sterol compounds. In this study, the spectroscopic and biophysical properties of CYP51 are extensively characterized, focussing on the unusual collapse of its carbon monoxy-ferrous form to a novel, protonated species, and interactions with a proposed physiological redox partner: a ferredoxin protein found adjacent to CYP51 in the M. tuberculosis genome. Additionally, the plausibility of this enzyme as a target for novel anti-tubercular agents, particularly azole-antifungal derivatives, is assessed. The study of CYP116B1 describes initial characterization of a novel fusion of a cytochrome P450 to its flavin mononucleotide- and iron sulphur cluster-containing redox partner protein, which is of potential biotechnological interest due to its proposed role in thiocarbamate herbicide degradation. Assessment of the spectroscopic properties of this enzyme are reported, including electron paramagnetic resonance and resonance Raman investigations of the heme macrocycle, in addition to kinetic studies of the electron transfer capabilities of the reductase component. Furthermore, the search for a substrate for this P450 and subsequent LCMS product analysis assays with the thiocarbamates EPTC and vernolate are detailed. The dissection of CYP116B1 into its constituent domains, in order to facilitate studies of the P450 and reductase components in isolation and further resolve structural, spectroscopic, and thermodynamic properties of this enzyme is also reported.

coli at different temperatures, it was observed that both types of presequence as well as the nature of mature protein affect the level of protein expression. Therefore, it was noticed that both types of presequence had no significant effect on the level of protein expression when they were linked to the mature E3 whereas the negative effect of these presequences was apparent when they were linked to mature E2 or E3BP. Secondly, comparing the solubility of precursors with their mature forms, both extended and standard presequences markedly reduced the solubility of precursor constructs by inducing the production of inclusion bodies although the effect appeared to be more marked with the former. Thirdly, on decreasing the rate of the protein synthesis by growing E. coli cultures at lower temperatures, it was possible to minimise the formation of insoluble protein aggregates and achieve partial or indeed complete solubility of precursor forms in some cases. Fourthly, these precursors appeared to retain the ability to fold correctly or at least to initiate the correct folding pathway. Thus both soluble and insoluble fractions of E2 and E3BP precursors contained lipoylated domains as judged by their ability to cross-react with PD2, an indication that these N-terminally located domains had adopted their native conformations. These observations were consistent with the view that N-terminal mitochondrial targeting sequences markedly reduced the rate of protein folding rather than suppressing the folding process completely. In this scenario, precursors would exist as nascent folding intermediates for longer periods compared to their mature equivalents and so would be more prone to aggregation and degradation as observed in this study. Further experiments are planned to test this hypothesis.

Cytochrome oxidase is the terminal proton pumping enzyme of the respiratory chain, catalysing the reduction of oxygen to water. This complex enzyme is composed of up to thirteen subunits of both nuclear and mitochondrial genetic origin, but the order in which they assemble is not fully understood. To investigate assembly I utilised blue native gel electrophoresis to analyse Saccharomyces cerevisiae strains with mutations in cytochrome oxidase causing assembly defect. I identified novel subcomplexes including, for the first time, a subcomplex containing mitochondrially encoded subunits. The respiratory chain is a target of the free radical nitric oxide (NO), which reversibly inhibits cytochrome oxidase through competition with oxygen at its active site. NO also has a myriad of other targets, and many of its actions are mediated through reactive nitrogen species (RNS) formed on reaction of NO with other species. NO is released as part of the immune response to infection and has antimicrobial action. Here, I found that prolonged incubation of yeast with an NO donor inhibited growth and caused a decrease in cytochrome oxidase content in dividing cells, which may be due to NO interfering with the assembly of the enzyme. I have also analysed gene expression in yeast after short NO exposure, to gain insight into the stress sensed by the cell and the transcription factors involved. The data suggest that NO causes a general stress response, in addition to specific effects such as repression of respiratory chain genes and activation of antioxidant/detoxification systems. Many genes known to be regulated by Haplp were repressed, suggesting that NO might interfere with this transcription factor.

The cloning of two open reading frames encoding a heterodimeric protein related in sequence to bacterial sarcosine oxidase and dimethylglycine oxidase from hyperthermophilic Pyrococcus furiosus DSM 3638 is described. the protein was overexpressed in E. coli, purified to homogeneity and identified as a flavoprotein by virtue of the enzymes UV-visible absorption spectrum showing characteristic lambdamax at 367 and 450 nm. The physiological substrate was identified as L-proline (t ½ = ∼105 s-1 for bleaching of flavoprotein spectrum at 450 nm). Additionally, the enzyme oxidises L-pipecolic acid ( t½ = ∼110.5 s-1) and, to a lesser extent sarcosine (t½ = ∼654 s-1). No reactivity with sodium sulfite was detected, consistent with the enzyme belonging to the flavoprotein dehydrogenase class. These data classified the enzyme as a novel hyperthermophilic L-proline dehydrogenase.;The crystal structure of PRODH at 3.3 A resolution shows the enzyme is a heterooctamer (alphabeta)4. The holoenzyme contains one mol each of FAD, FMN and ATP per mol of alphabeta complex. Isolation of monoflavinylated enzyme containing a single FAD cofactor permitted detailed redox potentiometry and pH-dependence studies of the reaction with L-proline. A bell-shaped dependence for kcat/K m as a function of pH was observed with macroscopic pK a values (7.0 +/- 0.2 and 7.6 +/- 0.2) attributed to residue ionisations in the free enzyme. The pH dependence of k cat is sigmoidal, with maximum activity realised in the alkaline region; the dependence is described by a macroscopic pK a value of 7.7 +/- 0.1 and by analogy with other flavoenzymes is tentatively attributed to the ionisation of L-proline in the Michaelis complex. Studies with H225A, H225Q and Y251F mutants ruled out the participation of these residues as catalytic bases.

TTQ-dependent aromatic amine dehydrogenase (AADH) from Alcaligenes faecalis has been expressed heterologously in Paracoccus denitrificans via the cloning of the aromatic amine dehydrogenase utilizing genes (aau) under the regulation of the mauF promoter, another TTQ-dependent methylamine dehydrogenase (MADH). The recombinant AADH obtained during this study is completely active indicating complete biosynthesis of the TTQ cofactor by restricted cloning of only five genes (aauBEDA and orf-2) of the nine known genes which encode AADH. Chemical and physical characterization as well as kinetic analysis, show that the recombinant AADH expressed in Paracoccus denitrificans is indistinguishable from the native AADH expressed in Alcaligenes faecalis. The catalytic small subunit of AADH has been isolated and studied in the absence of the large subunit. Separation from the holoenzyme was achieved by chemical denaturation. The separation small subunit was refolded, characterized and studied using a number of different kinetic approaches. The small subunit shows a fully integrated and completely redox active TTQ centre. The ability of the separated small subunit to oxidase amine substrate was demonstrated, despite the very slow reaction rates when compared to the holoenzyme. The altered catalytic potency of the TTQ-containing small subunit revealed the significant role of the large subunit in providing the enzyme's catalytic power. Structural X-ray crystallographic studies were performed on crystalline enzyme in an attempt to elucidate further details about the reaction mechanism of enzyme and specifically the mode of C-H bond breakage. Structures of AADH in complex with different para substituted benzylamines were obtained. These structures provide valuable datasets for future studies using computational methods to address specific aspects of the chemistry of TTQ reduction by amine substrates.

Cytochrome P450 BioI is isolated from the soil bacterium Bacillus subtilis. It is located in an operon that encodes genes involved in the synthesis of the vitamin biotin. From knock-out studies it was concluded that BioI catalyses an early step in the biosynthetic pathway, forming a precursor to the isolatable intermediate pimelic acid. This thesis covers the characterisation of this P450 and its various endogenous redox partners. The data presented here reports upon the bio-physical properties of the P450 including information on the redox potentials, how its environment affects its properties and the factors that govern substrate binding. To produce a catalytically active BioI redox system, two flavodoxins were cloned from B. subtilis and their over-expression and preparation is described. These flavodoxins have been characterised according to their redox and structural properties. The ability of the two flavodoxins to reconstitute the oxidase activity of BioI was examined by kinetic and spectroscopic techniques culminating in the analysis of the monooxygenation of various fatty acids. To complete the redox chain two NADPH-oxidoreductase candidates and a di-flavin reductase from B. subtilis were assessed and their ability to complement the P450 BioI-flavodoxin system examined.

This work presents the crystal structure determination of the transcriptional regulator CprK and of individual domains of the multidomain cytochrome P450-BM3. The crystal structure of the A264E mutant heme domain was determined with and without substrate present. Surprisingly, the structures reveal the protein to exhibit a substrate bound conformation regardless of the presence of substrate. This has provided further evidence that substrate binding leads to a dramatic shift in the equilibrium of conformational states available to the protein. In addition, the crystal structure of the C773A mutant flavin binding domain has been determined both in presence and absence of NADP+. Together with the already available structures of the other domains, this now allows both modelling and further solution studies of the full length cytochrome P450-BM3 structure.;Dehalogenans sp. are capable of using a range of chlorophenolic compounds as terminal electron acceptors in a respiratory metabolism known as halorespiration. This process is under transcriptional control by CprK, a member of the CRP-FNR family of transcriptional regulators. The crystal structure of D. hafniense CprK in complex with o-chlorophenolacetic acid (OCPA) reveals tightly bound effector molecules. Binding of OCPA is analysed through both mutagenesis and fluorescence quenching binding studies. The results have led to the hypothesis that CprK uses the bound phenolic compound pKa as an additional mechanism to sense the presence of the chloride atom. The structure presents a structural framework for further studies of the mechanism of this family of transcriptional regulators and of CprK homologues in particular.

The thesis describes mechanistic studies of the bacterial protein morphinone reductase (MR) using mutagenesis, kinetic and spectroscopic methods. The roles of seven active site residues were investigated. Cys-191 was initially predicted to be an active site acid; however, mutagenesis studies revealed that Cys-191 does not act in such role. His-186 and Asn-189 were shown to be involved in ligand binding. Mutation of His-186 to alanine has dramatic effects on the enzyme in the steady-state and in the oxidative half-reaction suggesting that His-186 could be an active site acid; however, pH dependence and NMR spectroscopic studies proved that His-186 could not play such role. No charge-transfer complex was observed between the H186A mutant and b-NADH in the reductive half-reaction. Mutation of Asn-189 affects enzyme kinetics and results in a slower rate of flavin oxidation. Faster rates for flavin reduction in the N189A mutant suggest that mutation of Asn-189 favours orientation of the flavin and b-NADH for hydride transfer. Neither Tyr-72 nor Tyr-356 is an active site acid. Mutation of Tyr-72 does not substantially affect enzyme kinetics. Mutations of Tyr-356 and Trp-106 had pronounced affects on enzyme kinetics, suggesting that each mutation alters the active site geometry. In the seven MR mutants studied, the flavin redox potentials are not altered compared to wild-type enzyme except for the T32A mutant where a 50 mV decrease is observed. Mutation of Thr-32 produced enhanced activity in the oxidative half-reaction and a slower activity in the reductive half-reaction effect of copper on MR was investigated. His-186 is shown to be involved in copper binding. 1-Nitrocyclohexene is a potential substrate for MR and the underlying mechanism requires further study.

The NADPH oxidase generates O2- in phagocytes such as neutrophils. The system is dormant in resting cells and activated on exposure to stimuli following the assembly of the cytosolic components p40phox, p47phox, p67phox and rac2 with cytochrome b558 in the membrane. A region of homology of about 100-110 residues in length exists between the N-termini of p40phox and p47phox and has been termed the PX domain, the role of which is unknown. To further understand the role of this domain with respect to the cytosolic factor proteins p40phox and p47phox of the NADPH oxidase, we have carried out expression, purification, characterisation and crystallisation trials of the PX domain of p40phox and p47phox, and full-length p47phox. We also expressed and purified the full-length p40phox and truncated p67phox (302-460). A complex of full-length p40phox and p67phox (302- 460) has been made and the binding has been characterised. Numerous attempts to crystallise this complex have been carried out and protein crystals have been obtained. We also identified a PX domain binding partner in neutrophils using affinity column made from recombinant PX domain of p40phox. Actin was found to bind to PX domains of p40phox and p47phox. In vitro binding studies confirmed PX domains bound to the side of F-actin. Significantly, PX domains co-localised with F-actin when PX domain of p40phox and p47phox constructs were microinjected into fibroblasts nuclear, suggesting a positive role of actin filaments in the NADPH oxidase activation. Recent evidence shows that specific lipid microdomains, "rafts", are involved in signal transduction and in activation of many cells. Here we have examined the contribution of lipid rafts to NADPH oxidase activation in neutrophils. We found that the NADPH oxidase components in neutrophils associated with membrane rafts. Stimulation of neutrophils with PMA or via Fc receptors induced cytosolic components and PKC isotypes to be recruited to membrane rafts. Treatment of neutrophils with a cholesterol depletion drug-MβCD resulted in less efficient activation of the NADPH oxidase by PMA and IgG coated S. aureus, hence pointing to a function of membrane rafts as a "trap" to concentrate molecules and increasing the efficiency of cross-talking between molecules.

The enzymes dehydroquinase and shikimate dehydrogenase catalyse the third and fourth steps of the biosynthetic shikimate pathway, respectively. The work in this thesis investigates the active sites and overall structure of bacterial type II dehydroquinases, shikimate dehydrogenases and related proteins using biochemical and biophysical techniques. The active site of Streptomyces coelicolor type II dehydroquinase (SCDHQase) was characterised by site-directed mutagenesis of the residues: Arg113, His106, Glu104, Seri 08 and Tyr28, proposed to be important in previous structural studies of the enzyme. None of the mutations significantly altered the secondary or tertiary structure of the enzyme as shown by circular dichroism. All five mutantions significantly decreased the catalytic activity; the most effective mutation was the Y28F mutation with a 4700-fold reduction in activity. This has provided strong evidence of the role of each of the residues within the active site. To date Bacillus subtilis is the only organism known whose genome encodes proteins that resemble both types of dehydroquinase. The type I dehydroquinase was shown to be the active form in cell extracts of B. subtilis by sodium borohydride trapping of the Schiff base involved in the mechanism of all type I dehydroquinases. The YqhS protein from B. subtilis has 47% identity to the S. coelicolor type II dehydroquinase. However, YqhS has been found to have an extremely low level of dehydroquinase activity (kcat < 10-4 s-1). YqhS has a phenylalanine in place of the highly conserved tyrosine residue found to be catalytically important and implicated in proton abstraction at C2 of the substrate. Site-directed mutagenesis of the phenylalanine to a tyrosine (F23Y) effected a 2000-fold increase in Growth experiments on the disruption mutant of the yqhS gene (YqhSd) produced by the Japanese Bacillus Consortium show that the gene is essential for growth in minimal medium. Detailed studies of the pH dependence, effects of salts on catalytic activity and the inhibitory effect of polyanions were performed on the type II dehydroquinases of S. coelicolor, Mycobacterium tuberculosis (MTDHQase), Helicobacter pylori (HPDHQase) and B. subtilis (F23Y mutant). pH-dependence studies on all the dehydroquinases tested show that there is a seven-fold increase in kcat between pH 6.5 and 8.0 indicating that all of these type II DHQases share the same catalytic mechanism. Low concentrations of chloride cause a moderate increase the catalytic activity of MTDHQase and a 4-fold increase in kcat for HPDHQase. In contrast, the kcat of both SCDHQase and B. subtilis (F23Y mutant) decreases in the presence of low concentrations of chloride. The differences in the effect of chloride are thought to arise from variations in the flexibility of dehydroquinase structures. Polyanions (sulphate and phosphate) are competitive inhibitors of SCDHQase and HPDHQase. The interactions are more complex in MTDHQase and B. subtilis YqhS (F23Y). Analysis of the X-ray structures of SCDHQase and MTDHQase reveals contrasting modes of binding of polyanions between the two enzymes. The differences between the interactions and binding of polyanions in type II dehydroquinases may help to explain the relative potencies of rationally designed inhibitors. The Haemophilus influenzae paralogs; shikimate dehydrogenase (HIAroE) and shikimate dehydrogenase related protein (HIYdiB) have been investigated using steady-state kinetics and detailed analysis of their sequences in relation to the published X-ray structures of Escherichia coli shikimate dehydrogenase and YdiB protein. Both HIAroE and HIYdiB are NADPH-dependent dehydrogenases and catalyse the oxidation of shikimate to form dehydroshikimate, with kcat values of 194 and 19 s-1, respectively. HIAroE is also able to oxidise quinate as part of the catabolic quinate pathway (kcat of 50 s-1). Fluorescence quenching studies have demonstrated that shikimate is able to bind to HIAroE in the absence of cofactor. HIAroE is likely to follow a random sequential mechanism. The function of HIYdiB is still unclear as the catalytic activity of the HIYdiB protein is significantly lower than the HIAroE enzyme for all the substrates tested. The N-terminal amino acid sequence of the HIYdiB protein is significantly different from the E.coli YdiB and other members of the shikimate dehydrogenase family and it is proposed that the N-terminal region defines the function of this protein.