The domain structure of the dihydrolipoyl transacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii

Sarpagan-Ajmalan-Type Indoles: Biosynthesis, Structural Biology, and Chemo-Enzymatic Significance

The biosynthetic pathway of the monoterpenoid indole alkaloid ajmaline in the genus Rauvolfia, in particular Rauvolfia serpentina Benth. ex Kurz, is one of the few pathways that have been comprehensively uncovered. Every step in the progress of plant alkaloid biosynthesis research is due to the endeavors of several generations of scientists and the advancement of technologies. The tissue and cell suspension cultures developed in the 1970s by M.H. Zenk enabled the extraction of alkaloids and crude enzymes for use as experimental materials, thus establishing the foundation for further research on enzymatic reaction networks. In vivo NMR technology was first used in biosynthetic investigations in the 1990s following the invention of high-field cryo-NMR, which allowed the rapid and reliable detection of bioconversion processes within living plant cells. Shortly before, in 1988, a milestone was reached with the heterologous expression of the strictosidine synthase cDNA, which paved the way for the application of "reverse genetics" and "macromolecular crystallography." Both methods allowed the structural analysis of several Rauvolfia enzymes involved in ajmaline biosynthesis and expanded our knowledge of the enzyme mechanisms, substrate specificities, and structure-activity relationships. It also opened the door for rational enzyme engineering and metabolic steering. Today, the research focus of ajmaline biosynthesis is shifting from "delineation" to "utilization." The Pictet-Spenglerase strictosidine synthase, strictosidine glucosidase, together with raucaffricine glucosidase, as pioneers in this area, have become useful tools to generate "privileged structures" and "diversity oriented" syntheses, which may help to construct novel scaffolds and to set up libraries of sarpagan-ajmalan-type alkaloids in chemo-enzymatic approaches.

Acetoin metabolism in bacteria

Acetoin is an important physiological metabolite excreted by many microorganisms. The excretion of acetoin, which can be diagnosed by the Voges Proskauer test and serves as a microbial classification marker, has its vital physiological meanings to these microbes mainly including avoiding acification, participating in the regulation of NAD/NADH ratio, and storaging carbon. The well-known anabolism of acetoin involves alpha-acetolactat synthase and alpha-acetolactate decarboxylase; yet its catabolism still contains some differing views, although much attention has been focused on it and great advances have been achieved. Current findings in catabolite control protein A (CcpA) mediated carbon catabolite repression may provide a fuller understanding of the control mechanism in bacteria. In this review, we first examine the acetoin synthesis pathways and its physiological meanings and relevancies; then we discuss the relationship between the two conflicting acetoin cleavage pathways, the enzymes of the acetoin dehydrogenase enzyme system, major genes involved in acetoin degradation, and the CcpA mediated acetoin catabolite repression pathway; in the end we discuss the genetic engineering progresses concerning applications. To date, this is the first integrated review on acetoin metabolism in bacteria, especially with regard to catabolic aspects. The apperception of the generation and dissimilation of acetoin in bacteria will help provide a better understanding of microbial strategies in the struggle for resources, which will consequently better serve the utilization of these microbes.

Pyruvate dehydrogenase from Azotobacter vinelandii. Properties of the N-terminally truncated enzyme

The pyruvate dehydrogenase multienzyme complex (PDHC) catalyses the oxidative decarboxylation of pyruvate and the subsequent acetylation of coenzyme A to acetyl-CoA. Previously, limited proteolysis experiments indicated that the N-terminal region of the homodimeric pyruvate dehydrogenase (E1p) from Azotobacter vinelandii could be involved in the binding of E1p to the core protein (E2p) [Hengeveld, A. F., Westphal, A. H. & de Kok, A. (1997) Eur J. Biochem. 250, 260-268]. To further investigate this hypothesis N-terminal deletion mutants of the E1p component of Azotobacter vinelandii pyruvate dehydrogenase complex were constructed and characterized. Up to nine N-terminal amino acids could be removed from E1p without effecting the properties of the enzyme. Truncation of up to 48 amino acids did not effect the expression or folding abilities of the enzyme, but the truncated enzymes could no longer interact with E2p. The 48 amino acid deletion mutant (E1pdelta48) is catalytically fully functional: it has a Vmax value identical to that of wild-type E1p, it can reductively acetylate the lipoamide group attached to the lipoyl domain of the core enzyme (E2p) and it forms a dimeric molecule. In contrast, the S0.5 for pyruvate is decreased. A heterodimer was constructed containing one subunit of wild-type E1p and one subunit of E1pdelta48. From the observation that the heterodimer was not able to bind to E2p, it is concluded that both N-terminal domains are needed for the binding of E1p to E2p. The interactions are thought to be mainly of an electrostatic nature involving negatively charged residues on the N-terminal domains of E1p and previously identified positively charged residues on the binding and catalytic domain of E2p.

The pyruvate dehydrogenase multi-enzyme complex from Gram-negative bacteria

Pyruvate dehydrogenase multi-enzyme complexes from Gram-negative bacteria consists of three enzymes, pyruvate dehydrogenase/decarboxylase (E1p), dihydrolipoyl acetyltransferase (E2p) and dihydrolipoyl dehydrogenase (E3). The acetyltransferase harbors all properties required for multi-enzyme catalysis: it forms a large core of 24 subunits, it contains multiple binding sites for the E1p and E3 components, the acetyltransferase catalytic site and mobile substrate carrying lipoyl domains that visit the active sites. Today, the Azotobacter vinelandii complex is the best understood oxo acid dehydrogenase complex with respect to structural details. A description of multi-enzyme catalysis starts with the structural and catalytic properties of the individual components of the complex. Integration of the individual properties is obtained by a description of how the many copies of the individual enzymes are arranged in the complex and how the lipoyl domains couple the activities of the respective active sites by way of flexible linkers. These latter aspects are the most difficult to study and future research need to be aimed at these properties.

Expression and characterisation of the homodimeric E1 component of the Azotobacter vinelandii pyruvate dehydrogenase complex

We have cloned and sequenced the gene encoding the homodimeric pyruvate dehydrogenase component (E1p) of the pyruvate dehydrogenase complex from Azotobacter vinelandii and expressed and purified the E1p component in Escherichia coli. Cloned E1p can be used to fully reconstitute complex activity. The enzyme was stable in high ionic strength buffers, but was irreversibly inactivated when incubated at high pH, which presumably was caused by its inability to redimerize correctly. This explains the previously found low stability of the wild-type E1p component after resolution from the complex at high pH. Cloned E1p showed a kinetic behaviour exactly like the wild-type complex-bound enzyme with respect to its substrate (pyruvate), its allosteric properties, and its effectors. These experiments show that acetyl coenzyme A acts as a feedback inhibitor by binding to the E1p component. Limited proteolysis experiments showed that the N-terminal region of E1p was easily removed. The resulting protein fragment was still active with artificial electron acceptors but had lost its ability to bind to the core component (E2p) and thus reconstitute complex activity. E1p was protected against proteolysis by E2p. The allosteric effector pyruvate changed E1p into a conformation that is more resistant to proteolysis.

Three-dimensional structure in solution of the N-terminal lipoyl domain of the pyruvate dehydrogenase complex from Azotobacter vinelandii

The three-dimensional structure of the N-terminal lipoyl domain of the acetyltransferase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii has been determined using heteronuclear multidimensional NMR spectroscopy and dynamical simulated annealing. The structure is compared with the solution structure of the lipoyl domain of the A. vinelandii 2-oxoglutarate dehydrogenase complex. The overall fold of the two structures, described as a beta-barrel-sandwich hybrid, is very similar. This agrees well with the high similarity of NMR-derived parameters, e.g. chemical shifts, between the two lipoyl domains. The main structural differences between the two lipoyl domains occur in a solvent-exposed loop close in space to the lipoylation site. Despite their high structural similarity, these lipoyl domains show a high preference for being reductively acylated by their parent 2-oxo acid dehydrogenase. Potential residues of the lipoyl domain involved in this process of molecular recognition are discussed.

Identification of a novel dihydrolipoyl dehydrogenase-binding protein in the pyruvate dehydrogenase complex of the anaerobic parasitic nematode, Ascaris suum

A novel dihydrolipoyl dehydrogenase-binding protein (E3BP) which lacks an amino-terminal lipoyl domain, p45, has been identified in the pyruvate dehydrogenase complex (PDC) of the adult parasitic nematode, Ascaris suum. Sequence at the amino terminus of p45 exhibited significant similarity with internal E3-binding domains of dihydrolipoyl transacetylase (E2) and E3BP. Dissociation and resolution of a pyruvate dehydrogenase-depleted adult A. suum PDC in guanidine hydrochloride resulted in two E3-depleted E2 core preparations which were either enriched or substantially depleted of p45. Following reconstitution, the p45-enriched E2 core exhibited enhanced E3 binding, whereas, the p45-depleted E2 core exhibited dramatically reduced E3 binding. Reconstitution of either the bovine kidney or A. suum PDCs with the A. suum E3 suggested that the ascarid E3 was more sensitive to NADH inhibition when bound to the bovine kidney core. The expression of p45 was developmentally regulated and p45 was most abundant in anaerobic muscle. In contrast, E3s isolated from anaerobic muscle or aerobic second-stage larvae were identical. These results suggest that during the transition to anaerobic metabolism, E3 remains unchanged, but it appears that a novel E3BP, p45, is expressed which may help to maintain the activity of the PDC in the face of the elevated intramitochondrial NADH/NAD+ ratios associated with anaerobiosis.

mgpS, a complex regulatory locus involved in the transcriptional control of the puc and puf operons in Rhodobacter sphaeroides 2.4.1

A new method has been developed in order to select mutants showing decreased puc operon transcription in Rhodobacter sphaeroides 2.4.1. A transcriptional fusion of a promoterless fragment derived from the sacB gene, encoding the levansucrase from Bacillus subtilis, to the upstream regulatory region of the puc operon has been constructed. With appropriate levels of exogenous sucrose, survivors of a sucrose killing challenge have been isolated. Subsequent analysis revealed the presence of both cis- and trans-acting "down" mutations in relation to puc operon expression. One of the trans-acting regulatory mutations was chosen for further study. The original mutation showed less than 2% of the level of puc operon transcription compared with the wild type under aerobic conditions and an 86% reduction under dark dimethyl sulfoxide conditions. This mutation can be complemented by a 3.9-kb BamHI DNA fragment derived from a cosmid contained within a genomic cosmid bank. DNA sequence analysis of this fragment revealed the presence of a 2.8-kb open reading frame, designated mgpS, which would encode a 930-amino-acid protein. The N-terminal portion of the putative protein product presents homologies to proteins of the RNA helicase family. Disruption of the chromosomal mgpS resulted in decreased transcription of both puc and puf, while the presence of mgpS in multicopy in the wild type, 2.4.1., increased puc expression by a factor of 2 under aerobic conditions. Structural analysis of the mgpS locus revealed that expression of mgpS was likely to be complex. A smaller protein containing the 472 C-terminal amino acids of MgpS is able to act by itself as an activator of puc transcription and is expressed independently of the large open reading frame in which it is contained.

Sequential 1H and 15N nuclear magnetic resonance assignments and secondary structure of the lipoyl domain of the 2-oxoglutarate dehydrogenase complex from Azotobacter vinelandii. Evidence for high structural similarity with the lipoyl domain of the pyruvate dehydrogenase complex

A 79-amino-acid polypeptide, corresponding to the lipoyl domain of the succinyltransferase component of the 2-oxoglutarate dehydrogenase multienzyme complex from Azotobacter vinelandii, has been sub-cloned and produced in Escherichia coli. Complete sequential 1H and 15N resonance assignments for the lipoyl domain have been obtained by using homo- and hetero-nuclear NMR spectroscopy. Two antiparallel beta-sheets of four strands each were identified from characteristic NOE connectivities and 3JHN alpha values. The lipoyl-lysine residue is found in a type-I turn connecting two beta-strands. The secondary structure of the lipoyl domain very much resembles the secondary solution structure of the N-terminal lipoyl domain of the A. vinelandii pyruvate dehydrogenase complex, despite the sequence identity of 25%. A detailed comparison of the NMR-derived parameters of both lipoyl domains, i.e. chemical shifts, NH-exchange rates, NOEs, and 3JHN alpha values suggests a high structural similarity in solution between the two lipoyl domains. Preliminary tertiary-structure calculations confirm that these lipoyl domains have very similar overall folds. The observed specificity of the 2-oxo acid dehydrogenase components of both complexes for these lipoyl domains is discussed in this respect.

Sequential 1H and 15N nuclear magnetic resonance assignments and secondary structure of the N-terminal lipoyl domain of the dihydrolipoyl transacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii

The N-terminal lipoyl domain (79 residues) of the transacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii has been sub-cloned and produced in Escherichia coli. Over-expression exceeds the capacity of E. coli cells to lipoylate all expressed lipoyl domain, but addition of lipoic acid to the growth medium results in expression of fully lipoylated domain. A two-dimensional homo- and heteronuclear NMR study of the lipoyl domain has resulted in sequential 1H and 15N resonance assignments of the unlipoylated form of the protein. Small differences in chemical shift values for protons of residues in the vicinity of the lipoyl-lysine residue are observed for the lipoylated form of the domain, suggesting that the conformation of the lipoyl domain is not altered significantly by the coupled cofactor. From nuclear Overhauser effects, backbone coupling constants and slowly exchanging amide protons, two antiparallel beta-sheets, each containing four strands, were identified. The lipoyl-lysine residue is exposed to the solvent and located in a type-I turn between two strands. The N- and C-terminal residues of the folded chain are close together in the other sheet. Preliminary data on the relative three-dimensional orientation of the two beta-sheets are presented. Comparison with the solution structure of the lipoyl domain of the Bacillus stearothermophilous pyruvate dehydrogenase complex shows resemblance to a large extent, despite the sequence identity of 31%.

Identification and molecular characterization of the aco genes encoding the Pelobacter carbinolicus acetoin dehydrogenase enzyme system

Use of oligonucleotide probes, which were deduced from the N-terminal sequences of the purified enzyme components, identified the structural genes for the alpha and beta subunits of E1 (acetoin:2,6-dichlorophenolindophenol oxidoreductase), E2 (dihydrolipoamide acetyltransferase), and E3 (dihydrolipoamide dehydrogenase) of the Pelobacter carbinolicus acetoin dehydrogenase enzyme system, which were designated acoA, acoB, acoC, and acoL, respectively. The nucleotide sequences of acoA (979 bp), acoB (1,014 bp), acoC (1,353 bp), and acoL (1,413 bp) as well as of acoS (933 bp), which encodes a protein with an M(r) of 34,421 exhibiting 64.7% amino acid identity to the Escherichia coli lipA gene product, were determined. These genes are clustered on a 6.1-kbp region. Heterologous expression of acoA, acoB, acoC, acoL, and acoS in E. coli was demonstrated. The amino acid sequences deduced from acoA, acoB, acoC, and acoL for E1 alpha (M(r), 34,854), E1 beta (M(r), 36,184), E2 (M(r), 47,281), and E3 (M(r), 49,394) exhibited striking similarities to the amino acid sequences of the components of the Alcaligenes eutrophus acetoin-cleaving system. Homologies of up to 48.7% amino acid identity to the primary structures of the enzyme components of various 2-oxo acid dehydrogenase complexes also were found. In addition, the respective genes of the 2-oxo acid dehydrogenase complexes and of the acetoin dehydrogenase enzyme system were organized very similarly, indicating a close relationship of the P. carbinolicus acetoin dehydrogenase enzyme system to 2-oxo acid dehydrogenase complexes.

Molecular evolution of biotin-dependent carboxylases

Amino-acid sequences of three functional units from various biotin-dependent carboxylases, biotin carboxylase, biotin-carboxyl-carrier protein and carboxyl transferase, were investigated by computer-assisted sequence comparison to obtain information about the structure, function, and molecular evolution of the enzymes. Biotin-dependent carboxylases, except transcarboxylase and oxaloacetate decarboxylase which lack biotin carboxylase, exert their catalytic activities through the three functional units. The three functional units correspond with functional domains or subunits of the enzymes, and the genetic information for the units is encoded in different ways from enzyme to enzyme. It is known that biotin carboxylase is homologous to carbamoyl-phosphate synthetase, and that the biotin-carboxyl-carrier protein is homologous to lipoic-acid-binding domain. The evolutionary relationships between the functional units and their homologues were described. A model for the evolutionary history of the enzymes was proposed by molecular phylogenetic analysis, which shows how a wide variety of domain and/or subunit structures for the enzymes may have been established. A repeated structure was found in biotin-carboxyl-carrier protein, and the secondary structure of the protein was predicted using the observed sequence similarity with a lipoic-acid-binding domain.

Structure/function relationships in the pyruvate dehydrogenase complex from Azotobacter vinelandii. Role of the linker region between the binding and catalytic domain of the dihydrolipoyl transacetylase component

The role of the hinge region between the binding domain and the catalytic domain in dihydrolipoyl transacetylase (E2p) from Azotobacter vinelandii was addressed by deletion mutagenesis. Mutated dihydrolipoyl transacetylase proteins were constructed with a deletion of 11 amino acids in the hinge region between the binding domain and the N-terminal part of the catalytic domain of E2p [E2p(pAPE1)] and with a further deletion of 9 amino acids into the N-terminal sequence protruding from the globular structure of the catalytic domain [E2p(pAPE2)] and found to take part in the intratrimer interaction. Both proteins behaved as wild-type E2p with respect to catalytic activity and quaternary structure. The interaction of the peripheral components pyruvate dehydrogenase (E1p) and lipoamide dehydrogenase (E3) with the mutated E2p proteins was studied. E2p(pAPE1) assembles to a trimeric pyruvate dehydrogenase complex (PDC) with 15% decreased complex activity. No difference in affinity towards the peripheral components was detected. Upon binding of E3, E2p(pAPE2) dissociates into trimers and monomers. At saturation, two dimers of E3 were bound/E2p monomer instead of one dimer/E2p chain in trimeric wild-type E2p or E2p (pAPE1). The monomeric E2p species was catalytically inactive. Upon binding of excess E1p, some monomer formation of the E2p mutant took place. E1p however can prevent monomerization by E3. It is concluded that E1p is bound between two different E2p chains in the trimer. The substrates CoA and acetyl-CoA also prevent monomerization because they are bound by amino acid residues of two different E2p chains. In the presence of CoA no difference in affinity with respect to E1p and E3 binding was observed. CoA (and acetylCoA) also prevent dissociation of the 24-subunit core structure of wild-type E2p when added before addition of E1p or E3. Therefore, it seems likely that in vivo A. vinelandii PDC is based on a 24-subunit E2p core, like Escherichia coli PDC. A functional difference between complexes based on a trimer or a 24-subunit core has not been observed. A role of the hinge region as a spacer to allow binding of E1p or E3 seems unlikely. The results are discussed on the basis of the three-dimensional structure of the catalytic domain.

Reconstitution of pyruvate dehydrogenase multienzyme complexes based on chimeric core structures from Azotobacter vinelandii and Escherichia coli

Two unique restriction sites were introduced by site-directed mutagenesis at identical positions in the DNA encoding the dihydrolipoyltransacetylase (E2p) components of the pyruvate dehydrogenase complex from Azotobacter vinelandii and from Escherichia coli. In this manner each DNA chain could be cut into three parts, coding for the lipoyl domain, which consists of three lipoyl subdomains, the binding domain and the core-forming catalytic domain, respectively. Chimeric E2p components were constructed by exchanging the three domains between E2p from A. vinelandii and E. coli on gene level. The six chimeric E2p proteins were expressed and purified from E. coli TG2. All chimeras were catalytically active, 24-subunit E2p proteins. Interactions of the peripheral components E1p and E3 with the wild-type enzymes from A. vinelandii and E. coli and with the chimeric proteins were studied by gel-filtration experiments, analytical ultracentrifugation and reconstitution of the overall activity of the complex. A. vinelandii E3 interacts only with those chimeras that contain the A. vinelandii binding domain, whereas E. coli E3 interacts with all chimeras. Exchange of the lipoyl or catalytic domain did not influence the binding properties of E3. Recognition of E1p depends on the origin of both the binding domain and the catalytic domain. E. coli E1p interacts strongly with those chimeras in which both the binding domain and the catalytic domain were derived from E. coli E2p and weakly with chimeras that contained either the binding domain or the catalytic domain from E. coli E2p. No binding of E. coli E1p was observed when both domains were of A. vinelandii origin. A. vinelandii E1p recognizes E2p from A. vinelandii and E. coli, but strong interaction required that the binding and catalytic domain were of the same origin. Exchange of lipoyl domains had no effect on the binding properties of the E1p component. These observations confirm previous conclusions, based on site-directed mutagenesis of A. vinelandii E2p [Schulze, E., Westphal, A. H., Boumans, H., and de Kok, A. (1991) Eur. J. Biochem. 202, 841-848], that the binding site for E1p consists of amino acid residues derived from both the binding and the catalytic domain and extend these conclusions to E. coli E2p. Dissociation of the 24 subunit E2p core was only detected when the chimeric E2p proteins contained the catalytic domain from A. vinelandii E2p. Dissociation depends on the binding of peripheral components to the E1p-binding sites, pointing to differences in the inter-trimer contacts between the E2p proteins from both species.(ABSTRACT TRUNCATED AT 400 WORDS)

Purification and cellular localization of wild type and mutated dihydrolipoyltransacetylases from Azotobacter vinelandii and Escherichia coli expressed in E. coli

Wild type dihydrolipoyltransacetylase(E2p)-components from the pyruvate dehydrogenase complex of A. vinelandii or E. coli, and mutants of A. vinelandii E2p with stepwise deletions of the lipoyl domains or the alanine- and proline-rich region between the binding and the catalytic domain have been overexpressed in E. coli TG2. The high expression of A. vinelandii wild type E2p (20% of cellular protein) and of a mutant enzyme with two lipoyl domains changed the properties of the inner bacterial membrane. This resulted in a solubilization of A. vinelandii E2p after degradation of the outer membrane by lysozyme without any contamination by E. coli pyruvate dehydrogenase complex (PDC) or other high-molecular-weight contaminants. The same effect could be detected for A. vinelandii E2o, an E2 which contains only one lipoyl domain, whereas almost no solubilization of A. vinelandii E2p with one lipoyl domain or of E2p consisting only of the binding and catalytic domain was found. Partial or complete deletion of the alanine- and proline-rich sequence between the binding and the catalytic domain did also decrease the solubilization of the E2p-mutants after lysozyme treatment. Immunocytochemical experiments on E. coli TG2 cells expressing A. vinelandii wild type E2p indicated that the enzyme was present as a soluble protein in the cytoplasm. In contrast, overexpressed A. vinelandii E2p with deletion of all three lipoyl domains and E. coli wild type E2p aggregated intracellularly. The solubilization by lysozyme is therefore ascribed to excluded volume effects leading to changes in the properties of the inner bacterial membrane.

Site-directed mutagenesis of the dihydrolipoyl transacetylase component (E2p) of the pyruvate dehydrogenase complex from Azotobacter vinelandii. Binding of the peripheral components E1p and E3

Site-directed mutagenesis was performed in the protease-sensitive region, between the lipoyl and catalytic domains and in the catalytic domain, of the dihydrolipoyl transacetylase component (E2p) of the pyruvate dehydrogenase complex from Azotobacter vinelandii. The interaction of the mutated enzymes with the peripheral components pyruvate dehydrogenase (E1p) and lipoamide dehydrogenase (E3) was studied by gel filtration experiments, analytical ultracentrifugation and reconstitution of the pyruvate dehydrogenase complex. Upon binding of peripheral components, the 24-subunit core of A. vinelandii wild-type E2p dissociates into tetramers. Four E1p or E3 dimers can bind to a tetramer. Binding is mutually exclusive, resulting in an active complex containing one E3 and three E1p dimers. Large deletions of the protease-sensitive region of E2p resulted in a total loss of the E1p and E3 binding. A small deletion (delta P361-R362) or the point mutation K367Q in the protease-sensitive region did not influence E3 binding, but affected E1p binding strongly, although with excess E1p almost complete reconstitution was reached. For E2p with the point mutation R416D in the N-terminal region of the catalytic domain only 16% overall activity could be measured in reconstituted complexes. This is due to a very weak E1p/E2p interaction, whereas the E3 binding was not affected. The point mutation R416D did not influence the catalytic activity of E2p, although a function for this residue in the formation of the active site was predicted from amino acid similarities with chloramphenicol acetyltransferase type III from Escherichia coli. Deletion of the complete Ala + Pro-rich sequence between the protease-sensitive region and the catalytic domain did not affect the enzymological properties of E2p, nor the affinity for E1p or E3. A further deletion of 20 N-terminal residues from the catalytic domain destroyed the E2p activity. From gel filtration experiments it was concluded that the quaternary structure was unaffected, as was E3 binding. E1p binding was lost and, in contrast to the wild-type enzyme, no dissociation of the core upon addition of E3 was observed. This mutant enzyme possesses, like E. coli E2p, six E3 binding sites and clearly shows that interaction of E3 or E1p with the E1p sites and dissociation are linked processes. It is concluded that the binding site for E3 is located on the N-terminal part of the protease-sensitive region. In contrast, the binding site for E1p consists of two regions, one located on the protease-sensitive region and one of the catalytic domain. These regions are separated by a flexible sequence of about 20 amino acids.

The catalytic domain of the dihydrolipoyl transacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii and Escherichia coli. Expression, purification, properties and preliminary X-ray analysis

Partial sequences of the dihydrolipoyl transacetylase component (E2p) of the pyruvate dehydrogenase complex from Azotobacter vinelandii and Escherichia coli, containing the catalytic domain, were cloned in pUC plasmids and over-expressed in E. coli TG2. A high expression of a homogeneous protein was only detectable for E2p mutants consisting of the catalytic domain and the alanine-proline-rich sequence between a putative binding region for the peripheral components and the catalytic domain (apa-4). Most of the catalytic domain from A. vinelandii without the apa-4 sequence was degraded intracellularly, probably due to incorrect folding. Fusion proteins of six amino acids from beta-galactosidase, the apa-4 region and the catalytic domains of A. vinelandii or E. coli E2p could be highly purified. Both catalytic domains were assembled in 24-subunit structures with a molecular mass of approximately 670 kDa. The expression of catalytic domain from A. vinelandii E2p is more than twice as high as found for wild-type E2p. This can be explained by intracellular degradation of over-expressed wild-type E2p, whereas the catalytic domains are stable against proteolysis in vivo and in vitro. The interaction of the peripheral components pyruvate dehydrogenase (E1p) and dihydrolipoamide dehydrogenase (E3) with the catalytic domains was studied, using gel filtration on Superose-6 and sedimentation velocity experiments. No binding of either E1p or E3 to the catalytic domain of either organism was detectable. Crystals of the catalytic domain of A. vinelandii E2p could be grown to a maximum size of 0.6 x 0.6 x 0.4 mm. They diffract up to a resolution of 0.28 nm.

Sequence similarities within the family of dihydrolipoamide acyltransferases and discovery of a previously unidentified fungal enzyme

A composite protein sequence database was searched for amino acid sequences similar to the C-terminal domain of the dihydrolipoamide acetyltransferase subunit (E2p) of the pyruvate dehydrogenase complex of Escherichia coli. Nine sequences with extensive similarity were found, of which eight were E2 subunits. The other was for a putative mitochondrial ribosomal protein, MRP3, from Neurospora crassa. Alignment of the MRP3 and E2 sequences showed that the similarity extends through the entire MRP3 sequence and that MRP3 is most closely related to the E2p subunit of the pyruvate dehydrogenase complex from Saccharomyces cerevisiae, with 54% identical residues and a further 36% that are conservatively substituted. Other features of the MRP3 gene and protein are also consistent with it being the acyltransferase subunit of a 2-oxo acid dehydrogenase complex. A multiple alignment of 13 E2 sequences indicated that 120 (34%) of 353 equivalenced residues are identical or show some degree of conservation. It also identified residues that are potentially important for the structure, catalytic activity and substrate-specificity of the acyltransferases.

Purification and characterization of acetoin:2,6-dichlorophenolindophenol oxidoreductase, dihydrolipoamide dehydrogenase, and dihydrolipoamide acetyltransferase of the Pelobacter carbinolicus acetoin dehydrogenase enzyme system

Dihydrolipoamide dehydrogenase (DHLDH), dihydrolipoamide acetyltransferase (DHLTA), and acetoin: 2,6-dichlorophenolindophenol oxidoreductase (Ao:DCPIP OR) were purified from acetoin-grown cells of Pelobacter carbinolicus. DHLDH had a native Mr of 110,000, consisted of two identical subunits of Mr 54,000, and reacted only with NAD(H) as a coenzyme. The N-terminal amino acid sequence included the flavin adenine dinucleotide-binding site and exhibited a high degree of homology to other DHLDHs. DHLTA had a native Mr of greater than 500,000 and consisted of subunits identical in size (Mr 60,000). The enzyme was highly sensitive to proteolytic attack. During limited tryptic digestion, two major fragments of Mr 32,500 and 25,500 were formed. Ao:DCPIP OR consisted of two different subunits of Mr 37,500 and 38,500 and had a native Mr in the range of 143,000 to 177,000. In vitro in the presence of DCPIP, it catalyzed a thiamine pyrophosphate-dependent oxidative-hydrolytic cleavage of acetoin, methylacetoin, and diacetyl. The combination of purified Ao:DCPIP OR, DHLTA, and DHLDH in the presence of thiamine pyrophosphate and the substrate acetoin or methylacetoin resulted in a coenzyme A-dependent reduction of NAD. In the strictly anaerobic acetoin-utilizing bacteria P. carbinolicus, Pelobacter venetianus, Pelobacter acetylenicus, Pelobacter propionicus, Acetobacterium carbinolicum, and Clostridium magnum, the enzymes Ao:DCPIP OR, DHLTA, and DHLDH were induced during growth on acetoin, whereas they were absent or scarcely present in cells grown on a nonacetoinogenic substrate.

Time-resolved fluorescence studies on mutants of the dihydrolipoyl transacetylase (E2) component of the pyruvate dehydrogenase complex from Azotobacter vinelandii

Fluorescence anisotropy decays were measured for the wild-type dihydrolipoyl transacetylase (E2) component of pyruvate dehydrogenase complex from Azotobacter vinelandii and E. coli and for E2-mutants from A. vinelandii in which the alanine-proline-rich sequence between the binding domain and the catalytic domain is partially or completely deleted. In both E2-mutants the rotational mobility of the lipoyl domain and the overall activity after reconstitution of the complex are significantly decreased indicating the important role of the deleted sequence for the movement of the lipoyl domain and the transfer of substrates between the different active sites within the complex.

The 2-oxoglutarate dehydrogenase complex from Azotobacter vinelandii. 2. Molecular cloning and sequence analysis of the gene encoding the succinyltransferase component

The nucleotide sequence encoding the succinyltransferase component (E2o) of the 2-oxoglutarate dehydrogenase complex from Azotobacter vinelandii has been determined. Previously the cloning in Escherichia coli of the gene encoding lipoamide dehydrogenase from A. vinelandii was reported [Westphal, A.H. & de Kok, A. (1988) Eur. J. Biochem. 172, 299-305]. The 3.2-kb fragment used for the sequence determination contained the main part of the gene encoding succinyltransferase. The complete E2o gene, as well as the gene encoding the 2-oxoglutarate dehydrogenase component, resided on a 14.7-kb fragment from which the 3.2-kb fragment was subcloned. The protein-coding sequence of the gene consists of 1200 bp (400 codons, including the AUG start codon and the UGA stop codon). It is separated from the gene encoding the 2-oxoglutarate dehydrogenase component by 42 bp. No E. coli-like promoter sequence was found. A putative ribosome-binding site is located 9-15 bp upstream from the start codon. No terminator sequences were found downstream of the stop codon. This makes it likely that the three genes of the oxoglutarate dehydrogenase complex are transcribed as a single mRNA transcript analogous to the pyruvate dehydrogenase complex in E. coli. The intact gene was subcloned from the 14.7-kb fragment and brought to high expression under the influence of the vector-encoded lacZ promoter. The similarity with the E. coli enzyme is high with 63% identity. Like the enzyme from E. coli, it consists of a single lipoyl-binding domain, a putative E1- and E3-binding domain and a catalytic domain. The main difference is found in a 31-residue sequence rich in alanine and proline located between the lipoyl domain and the putative E1- and E3-binding domain. This sequence, usually found in acetyltransferases and there identified as a highly mobile region by 1H-NMR, is replaced by a more polar, charged region in the E. coli enzyme.

The gene encoding dihydrolipoyl transacetylase from Azotobacter vinelandii. Expression in Escherichia coli and activation and isolation of the protein

The gene encoding the dihydrolipoyl transacetylase (E2) component from Azotobacter vinelandii has been cloned in Escherichia coli. High expression of the gene was found when the cells were grown for more than 14 h. The E2 produced was partially active, varying 10 and 90% in different experiments. By limited proteolysis of the protein it was shown that the catalytic domain was incorrectly folded, caused by formation of intermolecular or intramolecular S-S bridges. The enzyme was fully activated after unfolding in 2.5 M guanidine hydrochloride containing 2 mM dithiothreitol, followed by refolding by dialysis. Active E2 was isolated in a simple three-step procedure. It possessed a specific activity in the same order as that found after isolation of E2 from purified pyruvate dehydrogenase complex from A. vinelandii. Active E2 comprises about 7% of the total soluble cellular protein in the E. coli clone. By genetic manipulation, deletion mutants of E2 were created, one encoding the lipoyl domain and the N-terminal half of the pyruvate-dehydrogenase (E1)- and lipoamide-dehydrogenase (E3)-binding domain, the other encoding the catalytic domain and the C-terminal half of the E1- and E3-binding domain. In E. coli expression of both mutants was observed.

The quaternary structure of the dihydrolipoyl transacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii. A reconsideration

After limited proteolysis of the dihydrolipoyl transacetylase component (E2) of Azotobacter vinelandii pyruvate dehydrogenase complex (PDC), a C-terminal domain was obtained which retained the transacetylase active site and the quaternary structure of E2 but had lost the lipoyl-containing N-terminal part of the chain and the binding sites for the peripheral components, pyruvate dehydrogenase and lipoamide dehydrogenase. The C-terminus of this domain was determined by treatment with carboxypeptidase Y and shown to be identical with the C-terminus of E2. Together with the previously determined N-terminus and the known amino acid sequence of E2, a molecular mass of 27.5 kDa was calculated. From the molecular mass of the native catalytic domain, 530 kDa, and the symmetry of the cubic structures observed on electron micrographs, a 24-meric structure is concluded instead of the 32-meric structure proposed previously. From the effect of guanidine hydrochloride on the light-scattering of intact E2 it was concluded that dissociation occurs in a two-step reaction resulting in particles with an average mass 1/6 that of the original mass before the N----D transition takes place. Cross-linking experiments with the catalytic domain indicated that the multimeric E2 is built from tetramers and that the tetramers are arranged as a dimer of dimers. A model for the quaternary structure of E2 is given, in which it is assumed that the tetrameric E2 core of PDC is formed from each of the six morphological subunits located at the lateral face of the cube. Binding of peripheral components to a site that interferes with the cubic assembly causes dissociation, resulting in the unique small PDC of A. vinelandii.

Mobile sequences in the pyruvate dehydrogenase complex, the E2 component, the catalytic domain and the 2-oxoglutarate dehydrogenase complex of Azotobacter vinelandii, as detected by 600 MHz 1H-NMR spectroscopy

600 MHz 1H-NMR spectroscopy demonstrates that the pyruvate dehydrogenase complex of Azotobacter vinelandii contains regions of the polypeptide chain with intramolecular mobility. This mobility is located in the E2 component and can probably be ascribed to alanine-proline-rich regions that link the lipoyl subdomains to each other as well as to the E1 and E3 binding domain. In the catalytic domain of E2, which is thought to form a compact, rigid core, also conformational flexibility is observed. It is conceivable that the N-terminal region of the catalytic domain, which contains many alanine residues, is responsible for the observed mobility. In the low-field region of the 1H-NMR spectrum of E2 specific resonances are found, which can be ascribed to mobile phenylalanine, histidine and/or tyrosine residues which are located in the E1 and E3 binding domain that links the lipoyl domain to the catalytic domain. In the 1H-NMR spectrum of the intact complex, these resonances cannot be observed, indicating a decreased mobility of the E1 and E3 binding domain.

Time-resolved fluorescence studies on the dihydrolipoyl transacetylase (E2) component of the pyruvate dehydrogenase complex from Azotobacter vinelandii

The dihydrolipoyl transacetylase (E2) component of A. vinelandii PDC and its lipoyl domain shows similar dynamic properties as revealed with fluorescence anisotropy decay of lipoyl-bound IAANS. The lipoyl domain (32.6 kDa), containing three almost identical subdomains shows a mode of rotation characteristic for a protein of about 30 kDa. A similar rotation is found in E2, indicating an independent rotational mobility of the whole domain in the multimeric E2 core (1.6 MDa). No independent rotation of a single lipoyl subdomain (10 kDa) is observed. The E1 component, in contrast to the E3 component, shows interaction with the lipoyl domain.

Comparison of the amino acid sequences of the transacylase components of branched chain oxoacid dehydrogenase of Pseudomonas putida, and the pyruvate and 2-oxoglutarate dehydrogenases of Escherichia coli

The nucleotide sequence of bkdB, the structural gene for E2b, the transacylase component of branched-chain-oxoacid dehydrogenase of Pseudomonas putida has been determined and translated into its amino acid sequence. The start of bkdB was identified from the N-terminal sequence of E2b isolated from branched-chain-oxoacid dehydrogenase of the closely related species, P. aeruginosa. The reading frame was composed of 65.5% G + C with 82.3% of the codons ending in G or C. There was no intergenic space between bkdA2 and bkdB. No codons requiring minor tRNAs were utilized and the codon bias index indicated a preferential codon usage. The bkdB gene encoded 423 amino acids although the N-terminal methionine was absent from E2b prepared from P. aeruginosa. The relative molecular mass of the encoded protein was 45,134 (45,003 minus methionine) vs 47,000 obtained by SDS/polyacrylamide gel electrophoresis. There was a single lipoyl domain in E2b compared to three lipoyl domains in E2p, and one domain in E2o, the transacylases of pyruvate and 2-oxoglutarate dehydrogenases of Escherichia coli respectively. There was significant similarity between the lipoyl domain of E2b and of E2p and E2o as well as between the E1-E2 binding domains of E2b, E2p and E2o. There was no similarity between the E3 binding domain of E2b to E2p and E2o which may reflect the uniqueness of the E3 component of branched-chain-oxoacid dehydrogenase of P. putida. The conclusions drawn from these comparisons are that the transacylases of prokaryotic pyruvate, 2-oxoglutarate and branched-chain-oxoacid dehydrogenases descended from a common ancestral protein probably at about the same time.

The dihydrolipoyltransacetylase component of the pyruvate dehydrogenase complex from Azotobacter vinelandii. Molecular cloning and sequence analysis

The gene encoding the dihydrolipoyltransacetylase component (E2) of the pyruvate dehydrogenase complex from Azotobacter vinelandii has been cloned in Escherichia coli. A plasmid containing a 2.8-kbp insert of A. vinelandii chromosomal DNA was obtained and its nucleotide sequence determined. The gene comprises 1911 base pairs, 637 codons excluding the initiation codon GUG and stop codon UGA. It is preceded by the gene encoding the pyruvate dehydrogenase component (E1) of pyruvate dehydrogenase complex and by an intercistronic region of 11 base pairs containing a good ribosome binding site. The gene is followed downstream by a strong terminating sequence. The relative molecular mass (64913), amino acid composition and N-terminal sequence are in good agreement with information obtained from studies on the purified enzyme. Approximately the first half of the gene codes for the lipoyl domain. Three very homologous sequences are present, which are translated in three almost identical units, alternated with non-homologous regions which are very rich in alanyl and prolyl residues. The N-terminus of the catalytic domain is sited at residue 381. Between the lipoyl domain and the catalytic domain, a region of about 50 residues is found containing many charged amino acid residues. This region is characterized as a hinge region and is involved in the binding of the pyruvate dehydrogenase and lipoamide dehydrogenase components. The homology with the dihydrolipoyltransacetylase from E. coli is high: 50% amino acid residues are identical.

Lipoamide dehydrogenase from Azotobacter vinelandii. Molecular cloning, organization and sequence analysis of the gene

The gene encoding lipoamide dehydrogenase from Azotobacter vinelandii has been cloned in Escherichia coli. Fragments of 9-23 kb from Azotobacter vinelandii chromosomal DNA obtained by partial digestion with Sau3A were ligated into the BamHI site of plasmid pUC9. E. coli TG2 cells were transformed with the resulting recombinant plasmids. Screening for clones which produced A. vinelandii lipoamide dehydrogenase was performed with antibodies raised against the purified enzyme. A positive colony was found which produced complete chains of lipoamide dehydrogenase as concluded form SDS gel electrophoresis of the cell-free extract, stained for protein or used for Western blotting. After subcloning of the 14.7-kb insert of this plasmid the structural gene could be located on a 3.2-kb DNA fragment. The nucleotide sequence of this subcloned fragment (3134 bp) has been determined. The protein-coding sequence of the gene consists of 1434 bp (478 codons, including the AUG start codon and the UAA stop codon). It is preceded by an intracistronic region of 85 bp and the structural gene for succinyltransferase. A putative ribosome-binding site and promoter sequence are given. The derived amino acid composition is in excellent agreement with that previously published for the isolated enzyme. The predicted relative molecular mass is 50223, including the FAD. The overall homology with the E. coli enzyme is high with 40% conserved amino acid residues. From a comparison with the three-dimensional structure of the related enzyme glutathione reductase [Rice, D. W., Schultz, G. E. & Guest, J. R. (1984) J. Mol. Biol. 174, 483-496], it appears that essential residues in all four domains have been conserved. The enzyme is strongly expressed, although expression does not depend on the vector-encoded lacZ promoter. The cloned enzyme is, in all the respects tested, identical with the native enzyme.