Common ancestry of Escherichia coli pyruvate oxidase and the acetohydroxy acid synthases of the branched-chain amino acid biosynthetic pathway

Molecular evolution of acetohydroxyacid synthase in bacteria

Acetohydroxyacid synthase (AHAS) is the key enzyme in the biosynthetic pathways of branched chain amino acids in bacteria. Since it does not exist in animal and plant cells, AHAS is an attractive target for developing antimicrobials and herbicides. In some bacteria, there is a single copy of AHAS, while in others there are multiple copies. Therefore, it is necessary to investigate the origin and evolutionary pathway of various AHASs in bacteria. In this study, all the available protein sequences of AHAS in bacteria were investigated, and an evolutionary model of AHAS in bacteria is proposed, according to gene structure, organization and phylogeny. Multiple copies of AHAS in some bacteria might be evolved from the single copy of AHAS, the ancestor. Gene duplication, domain deletion and horizontal gene transfer might occur during the evolution of this enzyme. The results show the biological significance of AHAS, help to understand the functions of various AHASs in bacteria, and would be useful for developing industrial production strains of branched chain amino acids or novel antimicrobials.

Crystal structure of plant acetohydroxyacid synthase, the target for several commercial herbicides

Acetohydroxyacid synthase (AHAS, EC 2.2.1.6) is the first enzyme in the branched-chain amino acid biosynthesis pathway. Five of the most widely used commercial herbicides (i.e. sulfonylureas, imidazolinones, triazolopyrimidines, pyrimidinyl-benzoates and sulfonylamino-cabonyl-triazolinones) target this enzyme. Here we have determined the first crystal structure of a plant AHAS in the absence of any inhibitor (2.9 Å resolution) and it shows that the herbicide-binding site adopts a folded state even in the absence of an inhibitor. This is unexpected because the equivalent regions for herbicide binding in uninhibited Saccharomyces cerevisiae AHAS crystal structures are either disordered, or adopt a different fold when the herbicide is not present. In addition, the structure provides an explanation as to why some herbicides are more potent inhibitors of Arabidopsis thaliana AHAS compared to AHASs from other species (e.g. S. cerevisiae). The elucidation of the native structure of plant AHAS provides a new platform for future rational structure-based herbicide design efforts.

DATABASE

The coordinates and structure factors for uninhibited AtAHAS have been deposited in the Protein Data Bank (www.pdb.org) with the PDB ID code 5K6Q.

Pyruvate decarboxylase activity of the acetohydroxyacid synthase of Thermotoga maritima

Acetohydroxyacid synthase (AHAS) catalyzes the production of acetolactate from pyruvate. The enzyme from the hyperthermophilic bacterium Thermotoga maritima has been purified and characterized (k_cat ~100 s⁻¹). It was found that the same enzyme also had the ability to catalyze the production of acetaldehyde and CO₂ from pyruvate, an activity of pyruvate decarboxylase (PDC) at a rate approximately 10% of its AHAS activity. Compared to the catalytic subunit, reconstitution of the individually expressed and purified catalytic and regulatory subunits of the AHAS stimulated both activities of PDC and AHAS. Both activities had similar pH and temperature profiles with an optimal pH of 7.0 and temperature of 85 °C. The enzyme kinetic parameters were determined, however, it showed a non-Michaelis-Menten kinetics for pyruvate only. This is the first report on the PDC activity of an AHAS and the second bifunctional enzyme that might be involved in the production of ethanol from pyruvate in hyperthermophilic microorganisms.

•

The acetohydroxyacid synthase of T. maritima has pyruvate decarboxylase activity

•

The AHAS and PDC activities share the same temperature and pH optima

•

Reconstitution of the catalytic and regulatory subunits increases both PDC and AHAS activities

Biosynthesis and Regulation of the Branched-Chain Amino Acids†

This review focuses on more recent studies concerning the systems biology of branched-chain amino acid biosynthesis, that is, the pathway-specific and global metabolic and genetic regulatory networks that enable the cell to adjust branched-chain amino acid synthesis rates to changing nutritional and environmental conditions. It begins with an overview of the enzymatic steps and metabolic regulatory mechanisms of the pathways and descriptions of the genetic regulatory mechanisms of the individual operons of the isoleucine-leucine-valine (ilv) regulon. This is followed by more-detailed discussions of recent evidence that global control mechanisms that coordinate the expression of the operons of this regulon with one another and the growth conditions of the cell are mediated by changes in DNA supercoiling that occur in response to changes in cellular energy charge levels that, in turn, are modulated by nutrient and environmental signals. Since the parallel pathways for isoleucine and valine biosynthesis are catalyzed by a single set of enzymes, and because the AHAS-catalyzed reaction is the first step specific for valine biosynthesis but the second step of isoleucine biosynthesis, valine inhibition of a single enzyme for this enzymatic step might compromise the cell for isoleucine or result in the accumulation of toxic intermediates. The operon-specific regulatory mechanisms of the operons of the ilv regulon are discussed in the review followed by a consideration and brief review of global regulatory proteins such as integration host factor (IHF), Lrp, and CAP (CRP) that affect the expression of these operons.

Generation of acetoin and its derivatives in foods

Acetoin is a common food flavor additive. This volatile compound widely exists in nature. Some microorganisms, higher plants, insects, and higher animals have the ability to synthesize acetoin using different enzymes and pathways under certain circumstances. As a very active molecule, acetoin acts as a precursor of dozens of compounds. Therefore, acetoin and its derivatives are frequently detected in the component analysis of a variety of foods using gas chromatography-mass spectrometry. Because of the increasing importance of these compounds, this paper reviews the origins and natural existence of these substances, physiological roles, the biological synthesis pathways, nonenzymatic spontaneous reactions, and the common determination methods in foods. This work is the first review on dietary natural acetoin.

Cyclohexane-1,2-dione hydrolase from denitrifying Azoarcus sp. strain 22Lin, a novel member of the thiamine diphosphate enzyme family

Alicyclic compounds with hydroxyl groups represent common structures in numerous natural compounds, such as terpenes and steroids. Their degradation by microorganisms in the absence of dioxygen may involve a C-C bond ring cleavage to form an aliphatic intermediate that can be further oxidized. The cyclohexane-1,2-dione hydrolase (CDH) (EC 3.7.1.11) from denitrifying Azoarcus sp. strain 22Lin, grown on cyclohexane-1,2-diol as a sole electron donor and carbon source, is the first thiamine diphosphate (ThDP)-dependent enzyme characterized to date that cleaves a cyclic aliphatic compound. The degradation of cyclohexane-1,2-dione (CDO) to 6-oxohexanoate comprises the cleavage of a C-C bond adjacent to a carbonyl group, a typical feature of reactions catalyzed by ThDP-dependent enzymes. In the subsequent NAD(+)-dependent reaction, 6-oxohexanoate is oxidized to adipate. CDH has been purified to homogeneity by the criteria of gel electrophoresis (a single band at ∼59 kDa; calculated molecular mass, 64.5 kDa); in solution, the enzyme is a homodimer (∼105 kDa; gel filtration). As isolated, CDH contains 0.8 ± 0.05 ThDP, 1.0 ± 0.02 Mg(2+), and 1.0 ± 0.015 flavin adenine dinucleotide (FAD) per monomer as a second organic cofactor, the role of which remains unclear. Strong reductants, Ti(III)-citrate, Na(+)-dithionite, and the photochemical 5-deazaflavin/oxalate system, led to a partial reduction of the FAD chromophore. The cleavage product of CDO, 6-oxohexanoate, was also a substrate; the corresponding cyclic 1,3- and 1,4-diones did not react with CDH, nor did the cis- and trans-cyclohexane diols. The enzymes acetohydroxyacid synthase (AHAS) from Saccharomyces cerevisiae, pyruvate oxidase (POX) from Lactobacillus plantarum, benzoylformate decarboxylase from Pseudomonas putida, and pyruvate decarboxylase from Zymomonas mobilis were identified as the closest relatives of CDH by comparative amino acid sequence analysis, and a ThDP binding motif and a 2-fold Rossmann fold for FAD binding could be localized at the C-terminal end and central region of CDH, respectively. A first mechanism for the ring cleavage of CDO is presented, and it is suggested that the FAD cofactor in CDH is an evolutionary relict.

Protein-protein interactions in assembly of lipoic acid on the 2-oxoacid dehydrogenases of aerobic metabolism

Lipoic acid is a covalently attached cofactor essential for the activity of 2-oxoacid dehydrogenases and the glycine cleavage system. In the absence of lipoic acid modification, the dehydrogenases are inactive, and aerobic metabolism is blocked. In Escherichia coli, two pathways for the attachment of lipoic acid exist, a de novo biosynthetic pathway dependent on the activities of the LipB and LipA proteins and a lipoic acid scavenging pathway catalyzed by the LplA protein. LipB is responsible for octanoylation of the E2 components of 2-oxoacid dehydrogenases to provide the substrates of LipA, an S-adenosyl-L-methionine radical enzyme that inserts two sulfur atoms into the octanoyl moiety to give the active lipoylated dehydrogenase complexes. We report that the intact pyruvate and 2-oxoglutarate dehydrogenase complexes specifically copurify with both LipB and LipA. Proteomic, genetic, and dehydrogenase activity data indicate that all of the 2-oxoacid dehydrogenase components are present. In contrast, LplA, the lipoate protein ligase enzyme of lipoate salvage, shows no interaction with the 2-oxoacid dehydrogenases. The interaction is specific to the dehydrogenases in that the third lipoic acid-requiring enzyme of Escherichia coli, the glycine cleavage system H protein, does not copurify with either LipA or LipB. Studies of LipB interaction with engineered variants of the E2 subunit of 2-oxoglutarate dehydrogenase indicate that binding sites for LipB reside both in the lipoyl domain and catalytic core sequences. We also report that LipB forms a very tight, albeit noncovalent, complex with acyl carrier protein. These results indicate that lipoic acid is not only assembled on the dehydrogenase lipoyl domains but that the enzymes that catalyze the assembly are also present "on site."

Protein families reflect the metabolic diversity of organisms and provide support for functional prediction

Comparative genomics has shown that protein families vary significantly within and across organisms in both number and functional composition. In the present work, we tested how the diversity at the family level reflects biological differences among organisms and contributes to their unique characteristics. For this purpose, we collected sequence-similar proteins of three selected families from model bacteria: Escherichia coli, Bacillus subtilis, and Pseudomonas aeruginosa. Protein relationships were identified using a phylogenomic approach to connect the functional diversity of enzymes to the metabolic capabilities of these organisms. All protein families studied have distinct functional compositions across the selected bacteria as supported by our Bayesian analysis. Some conserved functional features among family members included a shared reaction mechanism, cofactor usage, and/or ligand specificity. Many observations of the presence/absence of protein functions matched current knowledge of the physiology and biochemistry of the bacteria. In some cases, new functional predictions were made to family members previously uncharacterized. We believe that genome comparisons at the protein family level would also be useful in predicting metabolic diversity for organisms that are relatively unknown or currently uncultured in the laboratory.

The diverse roles of flavin coenzymes--nature's most versatile thespians

Flavin coenzymes play a variety of roles in biological systems. This Perspective highlights the chemical versatility of flavins by reviewing research on five flavoenzymes that have been studied in our laboratory. Each of the enzymes discussed in this review [the acyl-CoA dehydrogenases (ACDs), CDP-6-deoxy-l-threo-d-glycero-4-hexulose-3-dehydrase reductase (E3), CDP-4-aceto-3,6-dideoxygalactose synthase (YerE), UDP-galactopyranose mutase (UGM), and type II isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IDI-2)] utilizes flavin in a distinct role. In particular, the catalytic mechanisms of two of these enzymes, UGM and IDI-2, may involve novel flavin chemistry.

Acetohydroxyacid synthase and its role in the biosynthetic pathway for branched-chain amino acids

The branched-chain amino acids are synthesized by plants, fungi and microorganisms, but not by animals. Therefore, the enzymes of this pathway are potential target sites for the development of antifungal agents, antimicrobials and herbicides. Most research has focused upon the first enzyme in this biosynthetic pathway, acetohydroxyacid synthase (AHAS) largely because it is the target site for many commercial herbicides. In this review we provide a brief overview of the important properties of each enzyme within the pathway and a detailed summary of the most recent AHAS research, against the perspective of work that has been carried out over the past 50 years.

Pyruvate:quinone oxidoreductase in Corynebacterium glutamicum: molecular analysis of the pqo gene, significance of the enzyme, and phylogenetic aspects

Corynebacterium glutamicum recently has been shown to possess pyruvate:quinone oxidoreductase (PQO), catalyzing the oxidative decarboxylation of pyruvate to acetate and CO2 with a quinone as the electron acceptor. Here, we analyze the expression of the C. glutamicum pqo gene, investigate the relevance of the PQO enzyme for growth and amino acid production, and perform phylogenetic studies. Expression analyses revealed that transcription of pqo is initiated 45 bp upstream of the translational start site and that it is organized in an operon together with genes encoding a putative metal-activated pyridoxal enzyme and a putative activator protein. Inactivation of the chromosomal pqo gene led to the absence of PQO activity; however, growth and amino acid production were not affected under either condition tested. Introduction of plasmid-bound pqo into a pyruvate dehydrogenase complex-negative C. glutamicum strain partially relieved the growth phenotype of this mutant, indicating that high PQO activity can compensate for the function of the pyruvate dehydrogenase complex. To investigate the distribution of PQO enzymes in prokaryotes and to clarify the relationship between PQO, pyruvate oxidase (POX), and acetohydroxy acid synthase enzymes, we compiled and analyzed the phylogeny of respective proteins deposited in public databases. The analyses revealed a wide distribution of PQOs among prokaryotes, corroborated the hypothesis of a common ancestry of the three enzymes, and led us to propose that the POX enzymes of Lactobacillales were derived from a PQO.

Herbicide-binding sites revealed in the structure of plant acetohydroxyacid synthase

The sulfonylureas and imidazolinones are potent commercial herbicide families. They are among the most popular choices for farmers worldwide, because they are nontoxic to animals and highly selective. These herbicides inhibit branched-chain amino acid biosynthesis in plants by targeting acetohydroxyacid synthase (AHAS, EC 2.2.1.6). This report describes the 3D structure of Arabidopsis thaliana AHAS in complex with five sulfonylureas (to 2.5 A resolution) and with the imidazolinone, imazaquin (IQ; 2.8 A). Neither class of molecule has a structure that mimics the substrates for the enzyme, but both inhibit by blocking a channel through which access to the active site is gained. The sulfonylureas approach within 5 A of the catalytic center, which is the C2 atom of the cofactor thiamin diphosphate, whereas IQ is at least 7 A from this atom. Ten of the amino acid residues that bind the sulfonylureas also bind IQ. Six additional residues interact only with the sulfonylureas, whereas there are two residues that bind IQ but not the sulfonylureas. Thus, the two classes of inhibitor occupy partially overlapping sites but adopt different modes of binding. The increasing emergence of resistant weeds due to the appearance of mutations that interfere with the inhibition of AHAS is now a worldwide problem. The structures described here provide a rational molecular basis for understanding these mutations, thus allowing more sophisticated AHAS inhibitors to be developed. There is no previously described structure for any plant protein in complex with a commercial herbicide.

Two-Carbon Compounds and Fatty Acids as Carbon Sources

This review concerns the uptake and degradation of those molecules that are wholly or largely converted to acetyl-coenzyme A (CoA) in the first stage of metabolism in Escherichia coli and Salmonella enterica. These include acetate, acetoacetate, butyrate and longer fatty acids in wild type cells plus ethanol and some longer alcohols in certain mutant strains. Entering metabolism as acetyl-CoA has two important general consequences. First, generation of energy from acetyl-CoA requires operation of both the citric acid cycle and the respiratory chain to oxidize the NADH produced. Hence, acetyl-CoA serves as an energy source only during aerobic growth or during anaerobic respiration with such alternative electron acceptors as nitrate or trimethylamine oxide. In the absence of a suitable oxidant, acetyl-CoA is converted to a mixture of acetic acid and ethanol by the pathways of anaerobic fermentation. Catabolism of acetyl-CoA via the citric acid cycle releases both carbon atoms of the acetyl moiety as carbon dioxide and growth on these substrates as sole carbon source therefore requires the operation of the glyoxylate bypass to generate cell material. The pair of related two-carbon compounds, glycolate and glyoxylate are also discussed. However, despite having two carbons, these are metabolized via malate and glycerate, not via acetyl-CoA. In addition, mutants of E. coli capable of growth on ethylene glycol metabolize it via the glycolate pathway, rather than via acetyl- CoA. Propionate metabolism is also discussed because in many respects its pathway is analogous to that of acetate. The transcriptional regulation of these pathways is discussed in detail.

Structural basis for activation of the thiamin diphosphate-dependent enzyme oxalyl-CoA decarboxylase by adenosine diphosphate

Oxalyl-coenzyme A decarboxylase is a thiamin diphosphate-dependent enzyme that plays an important role in the catabolism of the highly toxic compound oxalate. We have determined the crystal structure of the enzyme from Oxalobacter formigenes from a hemihedrally twinned crystal to 1.73 A resolution and characterized the steady-state kinetic behavior of the decarboxylase. The monomer of the tetrameric enzyme consists of three alpha/beta-type domains, commonly seen in this class of enzymes, and the thiamin diphosphate-binding site is located at the expected subunit-subunit interface between two of the domains with the cofactor bound in the conserved V-conformation. Although oxalyl-CoA decarboxylase is structurally homologous to acetohydroxyacid synthase, a molecule of ADP is bound in a region that is cognate to the FAD-binding site observed in acetohydroxyacid synthase and presumably fulfils a similar role in stabilizing the protein structure. This difference between the two enzymes may have physiological importance since oxalyl-CoA decarboxylation is an essential step in ATP generation in O. formigenes, and the decarboxylase activity is stimulated by exogenous ADP. Despite the significant degree of structural conservation between the two homologous enzymes and the similarity in catalytic mechanism to other thiamin diphosphate-dependent enzymes, the active site residues of oxalyl-CoA decarboxylase are unique. A suggestion for the reaction mechanism of the enzyme is presented.

The enzymes of oxalate metabolism: unexpected structures and mechanisms

Oxalate degrading enzymes have a number of potential applications, including medical diagnosis and treatments for hyperoxaluria and other oxalate-related diseases, the production of transgenic plants for human consumption, and bioremediation of the environment. This review seeks to provide a brief overview of current knowledge regarding the major classes of enzymes and related proteins that are employed in plants, fungi, and bacteria to convert oxalate into CO(2) and/or formate. Not only do these enzymes employ intriguing chemical strategies for cleaving the chemically unreactive C-C bond in oxalate, but they also offer the prospect of providing new insights into the molecular processes that underpin the evolution of biological catalysts.

Role of a Conserved Arginine in the Mechanism of Acetohydroxyacid Synthase

The thiamin diphosphate (ThDP)-dependent bio-synthetic enzyme acetohydroxyacid synthase (AHAS) catalyzes decarboxylation of pyruvate and specific condensation of the resulting ThDP-bound two-carbon intermediate, hydroxyethyl-ThDP anion/enamine (HEThDP(-)), with a second ketoacid, to form acetolactate or acetohydroxybutyrate. Whereas the mechanism of formation of HEThDP(-) from pyruvate is well understood, the role of the enzyme in control of the carboligation reaction of HEThDP(-) is not. Recent crystal structures of yeast AHAS from Duggleby's laboratory suggested that an arginine residue might interact with the second ketoacid substrate. Mutagenesis of this completely conserved residue in Escherichia coli AHAS isozyme II (Arg(276)) confirms that it is required for rapid and specific reaction of the second ketoacid. In the mutant proteins, the normally rapid second phase of the reaction becomes rate-determining. A competing alternative nonnatural but stereospecific reaction of bound HEThDP(-) with benzaldehyde to form phenylacetylcarbinol (Engel, S., Vyazmensky, M., Geresh, S., Barak, Z., and Chipman, D. M. (2003) Biotechnol. Bioeng. 84, 833-840) provides a new tool for studying the fate of HEThDP(-) in AHAS, since the formation of the new product has a very different dependence on active site modifications than does acetohydroxyacid acid formation. The effects of mutagenesis of four different residues in the site on the rates and specificities of the normal and unnatural reactions support a critical role for Arg(276) in the stabilization of the transition states for ligation of the incoming second ketoacid with HEThDP(-) and/or for the breaking of the product-ThDP bond. This information makes it possible to engineer the active site so that it efficiently and preferentially catalyzes a new reaction.

A unique catalytic mechanism for UDP-galactopyranose mutase

The flavoenzyme uridine 5'-diphosphate (UDP)-galactopyranose mutase (UGM) catalyzes the interconversion of UDP-galactopyranose (UDP-Galp) and UDP-galactofuranose (UDP-Galf). The latter is an essential precursor to the cell wall arabinogalactan of Mycobacterium tuberculosis. The catalytic mechanism for this enzyme had not been elucidated. Here, we provide evidence for a mechanism in which the flavin cofactor assumes a new role. Specifically, the N5 of the reduced anionic flavin cofactor captures the anomeric position of the galactose residue with release of UDP. Interconversion of the isomers occurs via a flavin-derived iminium ion. To trap this putative intermediate, we treated UGM with radiolabeled UDP-Galp and sodium cyanoborohydride; a radiolabeled flavin-galactose adduct was obtained. Ultraviolet-visible spectroscopy and mass spectrometry indicate that this product is an N5-alkyl flavin. We anticipate that the clarification of the catalytic mechanism for UGM will facilitate the development of anti-mycobacterial agents.

The crystal structures of Klebsiella pneumoniae acetolactate synthase with enzyme-bound cofactor and with an unusual intermediate

Acetohydroxyacid synthase (AHAS) and acetolactate synthase (ALS) are thiamine diphosphate (ThDP)-dependent enzymes that catalyze the decarboxylation of pyruvate to give a cofactor-bound hydroxyethyl group, which is transferred to a second molecule of pyruvate to give 2-acetolactate. AHAS is found in plants, fungi, and bacteria, is involved in the biosynthesis of the branched-chain amino acids, and contains non-catalytic FAD. ALS is found only in some bacteria, is a catabolic enzyme required for the butanediol fermentation, and does not contain FAD. Here we report the 2.3-A crystal structure of Klebsiella pneumoniae ALS. The overall structure is similar to AHAS except for a groove that accommodates FAD in AHAS, which is filled with amino acid side chains in ALS. The ThDP cofactor has an unusual conformation that is unprecedented among the 26 known three-dimensional structures of nine ThDP-dependent enzymes, including AHAS. This conformation suggests a novel mechanism for ALS. A second structure, at 2.0 A, is described in which the enzyme is trapped halfway through the catalytic cycle so that it contains the hydroxyethyl intermediate bound to ThDP. The cofactor has a tricyclic structure that has not been observed previously in any ThDP-dependent enzyme, although similar structures are well known for free thiamine. This structure is consistent with our proposed mechanism and probably results from an intramolecular proton transfer within a tricyclic carbanion that is the true reaction intermediate. Modeling of the second molecule of pyruvate into the active site of the enzyme with the bound intermediate is consistent with the stereochemistry and specificity of ALS.

Crystal structure of yeast acetohydroxyacid synthase: a target for herbicidal inhibitors

Acetohydroxyacid synthase (AHAS; EC 4.1.3.18) catalyzes the first step in branched-chain amino acid biosynthesis. The enzyme requires thiamin diphosphate and FAD for activity, but the latter is unexpected, because the reaction involves no oxidation or reduction. Due to its presence in plants, AHAS is a target for sulfonylurea and imidazolinone herbicides. Here, the crystal structure to 2.6 A resolution of the catalytic subunit of yeast AHAS is reported. The active site is located at the dimer interface and is near the proposed herbicide-binding site. The conformation of FAD and its position in the active site are defined. The structure of AHAS provides a starting point for the rational design of new herbicides.

The hydroxynitrile lyase from almond: a lyase that looks like an oxidoreductase

BACKGROUND

Cyanogenesis is a defense process of several thousand plant species. Hydroxynitrile lyase, a key enzyme of this process, cleaves a cyanohydrin into hydrocyanic acid and the corresponding aldehyde or ketone. The reverse reaction constitutes an important tool in biocatalysis. Different classes of hydroxynitrile lyases have convergently evolved from FAD-dependent oxidoreductases, alpha/beta hydrolases, and alcohol dehydrogenases. The FAD-dependent hydroxynitrile lyases (FAD-HNLs) carry a flavin cofactor whose redox properties appear to be unimportant for catalysis.

RESULTS

We have determined the crystal structure of a 61 kDa hydroxynitrile lyase isoenzyme from Prunus amygdalus (PaHNL1) to 1.5 A resolution. Clear electron density originating from four glycosylation sites could be observed. As concerns the overall protein fold including the FAD cofactor, PaHNL1 belongs to the family of GMC oxidoreductases. The active site for the HNL reaction is probably at a very similar position as the active sites in homologous oxidases.

CONCLUSIONS

There is strong evidence from the structure and the reaction product that FAD-dependent hydroxynitrile lyases have evolved from an aryl alcohol oxidizing precursor. Since key residues implicated in oxidoreductase activity are also present in PaHNL1, it is not obvious why this enzyme shows no oxidase activity. Similarly, features proposed to be relevant for hydroxy-nitrile lyase activity in other hydroxynitrile lyases, i.e., a general base and a positive charge to stabilize the cyanide, are not obviously present in the putative active site of PaHNL1. Therefore, the reason for its HNL activity is far from being well understood at this point.

Thiamin auxotrophy in yeast through altered cofactor dependence of the enzyme acetohydroxyacid synthase

The THI1 gene of Saccharomyces cerevisiae has been identified and found to be allelic with the previously characterized gene ILV2 that encodes acetohydroxyacid synthase (AHAS). This enzyme catalyses the first step in the parallel biosyntheses of the branched-chain amino acids isoleucine and valine, using thiamin pyrophosphate (TPP) as a cofactor. The ilv2-thi1 allele encodes a functional AHAS enzyme with an altered dependence for the cofactor TPP resulting in the thiamin auxotrophic phenotype. Nucleotide sequence analysis and site-directed mutagenesis revealed that the thi1 mutation is a single base substitution which causes the conserved amino acid substitution D176E in the AHAS protein. This study therefore implicates aspartate 176 as another amino acid residue important either for the efficient binding of TPP by AHAS or for the functional stability of the holoenzyme.

Divergence of function in sequence-related groups of Escherichia coli proteins

The most prominent mechanism of molecular evolution is believed to have been duplication and divergence of genes. Proteins that belong to sequence-related groups in any one organism are candidates to have emerged from such a process and to share a common ancestor. Groups of proteins in Escherichia coli having sequence similarity are mostly composed of proteins with closely related function, but some groups comprise proteins with unrelated functions. In order to understand how function can change while sequences remain similar, we have examined some of these groups in detail. The enzymes analyzed in this work include representatives of amidotransferases, phosphotransferases, decarboxylases, and others. Most sequence-related groups contain enzymes that are in the same classes of Enzyme Commission (EC) numbers. We have concentrated on groups that are heterogeneous in that respect, and also on groups containing more than one enzyme of any pathway. We find that although the EC number may differ, the reaction chemistry of these sequence-related proteins is the same or very similar. Some of these families illustrate how diversification has taken place in evolution, using common features of either reaction chemistry or ligand specificity, or both, to create catalysts for different kinds of biochemical reactions. This information has relevance to the area of functional genomics in which the activities of gene products of unknown reading frames are attributed by analogy to the functions of sequence-related proteins of known function.

Conversion of Escherichia coli pyruvate oxidase to an 'alpha-ketobutyrate oxidase'

Escherichia coli pyruvate oxidase (PoxB), a lipid-activated homotetrameric enzyme, is active on both pyruvate and 2-oxobutanoate ('alpha-ketobutyrate'), although pyruvate is the favoured substrate. By localized random mutagenesis of residues chosen on the basis of a modelled active site, we obtained several PoxB enzymes that had a markedly decreased activity with the natural substrate, pyruvate, but retained full activity with 2-oxobutanoate. In each of these mutant proteins Val-380 had been replaced with a smaller residue, namely alanine, glycine or serine. One of these, PoxB V380A/L253F, was shown to lack detectable pyruvate oxidase activity in vivo; this protein was purified, studied and found to have a 6-fold increase in K(m) for pyruvate and a 10-fold lower V(max) with this substrate. In contrast, the mutant had essentially normal kinetic constants with 2-oxobutanoate. The altered substrate specificity was reflected in a decreased rate of pyruvate binding to the latent conformer of the mutant protein owing to the V380A mutation. The L253F mutation alone had no effect on PoxB activity, although it increased the activity of proteins carrying substitutions at residue 380, as it did that of the wild-type protein. The properties of the V380A/L253F protein provide new insights into the mode of substrate binding and the unusual activation properties of this enzyme.

Identification of the gene encoding sulfopyruvate decarboxylase, an enzyme involved in biosynthesis of coenzyme M

The products of two adjacent genes in the chromosome of Methanococcus jannaschii are similar to the amino and carboxyl halves of phosphonopyruvate decarboxylase, the enzyme that catalyzes the second step of fosfomycin biosynthesis in Streptomyces wedmorensis. These two M. jannaschii genes were recombinantly expressed in Escherichia coli, and their gene products were tested for the ability to catalyze the decarboxylation of a series of alpha-ketoacids. Both subunits are required to form an alpha(6)beta(6) dodecamer that specifically catalyzes the decarboxylation of sulfopyruvic acid to sulfoacetaldehyde. This transformation is the fourth step in the biosynthesis of coenzyme M, a crucial cofactor in methanogenesis and aliphatic alkene metabolism. The M. jannaschii sulfopyruvate decarboxylase was found to be inactivated by oxygen and reactivated by reduction with dithionite. The two subunits, designated ComD and ComE, comprise the first enzyme for the biosynthesis of coenzyme M to be described.

Catalytic promiscuity and the evolution of new enzymatic activities

Several contemporary enzymes catalyze alternative reactions distinct from their normal biological reactions. In some cases the alternative reaction is similar to a reaction that is efficiently catalyzed by an evolutionary related enzyme. Alternative activities could have played an important role in the diversification of enzymes by providing a duplicated gene a head start towards being captured by adaptive evolution.

Biosynthesis of 2-aceto-2-hydroxy acids: acetolactate synthases and acetohydroxyacid synthases

Two groups of enzymes are classified as acetolactate synthase (EC 4. 1.3.18). This review deals chiefly with the FAD-dependent, biosynthetic enzymes which readily catalyze the formation of acetohydroxybutyrate from pyruvate and 2-oxobutyrate, as well as of acetolactate from two molecules of pyruvate (the ALS/AHAS group). These enzymes are generally susceptible to inhibition by one or more of the branched-chain amino acids which are ultimate products of the acetohydroxyacids, as well as by several classes of herbicides (sulfonylureas, imidazolinones and others). Some ALS/AHASs also catalyze the (non-physiological) oxidative decarboxylation of pyruvate, leading to peracetic acid; the possible relationship of this process to oxygen toxicity is considered. The bacterial ALS/AHAS which have been well characterized consist of catalytic subunits (around 60 kDa) and smaller regulatory subunits in an alpha2beta2 structure. In the case of Escherichia coli isozyme III, assembly and dissociation of the holoenzyme has been studied. The quaternary structure of the eukaryotic enzymes is less clear and in plants and yeast only catalytic polypeptides (homologous to those of bacteria) have been clearly identified. The presence of regulatory polypeptides in these organisms cannot be ruled out, however, and genes which encode putative ALS/AHAS regulatory subunits have been identified in some cases. A consensus sequence can be constructed from the 21 sequences which have been shown experimentally to represent ALS/AHAS catalytic polypeptides. Many other sequences fit this consensus, but some genes identified as putative 'acetolactate synthase genes' are almost certainly not ALS/AHAS. The solution of the crystal structures of several thiamin diphosphate (ThDP)-dependent enzymes which are homologous to ALS/AHAS, together with the availability of many amino acid sequences for the latter enzymes, has made it possible for two laboratories to propose similar, reasonable models for a dimer of catalytic subunits of an ALS/AHAS. A number of characteristics of these enzymes can now be better understood on the basis of such models: the nature of the herbicide binding site, the structural role of FAD and the binding of ThDP-Mg2+. The models are also guides for experimental testing of ideas concerning structure-function relationships in these enzymes, e.g. the nature of the substrate recognition site. Among the important remaining questions is how the enzyme suppresses alternative reactions of the intrinsically reactive hydroxyethylThDP enamine formed by the decarboxylation of the first substrate molecule and specifically promotes its condensation with 2-oxobutyrate or pyruvate.

Novel keto acid formate-lyase and propionate kinase enzymes are components of an anaerobic pathway in Escherichia coli that degrades L-threonine to propionate

An immunological analysis of an Escherichia coli strain unable to synthesize the main pyruvate formate-lyase enzyme Pfl revealed the existence of a weak, cross-reacting 85 kDa polypeptide that exhibited the characteristic oxygen-dependent fragmentation typical of a glycyl radical enzyme. Polypeptide fragmentation of this cross-reacting species was shown to be dependent on Pfl activase. Cloning and sequence analysis of the gene encoding this protein revealed that it coded for a new enzyme, termed TdcE, which has 82% identity with Pfl. On the basis of RNA analyses, the tdcE gene was shown to be part of a large operon that included the tdcABC genes, encoding an anaerobic threonine dehydratase, tdcD, coding for a propionate kinase, tdcF, the function of which is unknown, and the tdcG gene, which encodes a L-serine dehydratase. Expression of the tdcABCDEFG operon was strongly catabolite repressed. Enzyme studies showed that TdcE has both pyruvate formate-lyase and 2-ketobutyrate formate-lyase activity, whereas the TdcD protein is a new propionate/acetate kinase. By monitoring culture supernatants from various mutants using 1H nuclear magnetic resonance (NMR), we followed the anaerobic conversion of L-threonine to propionate. These studies confirmed that 2-ketobutyrate, the product of threonine deamination, is converted in vivo by TdcE to propionyl-CoA. These studies also revealed that Pfl and an as yet unidentified thiamine pyrophosphate-dependent enzyme(s) can perform this reaction. Double null mutants deficient in phosphotransacetylase (Pta) and acetate kinase (AckA) or AckA and TdcD were unable to metabolize threonine to propionate, indicating that propionyl-CoA and propionyl-phosphate are intermediates in the pathway and that ATP is generated during the conversion of propionyl-P to propionate by AckA or TdcD.

Cloning and phylogenetic analysis of the genes encoding acetohydroxyacid synthase from the archaeon Methanococcus aeolicus

The gene for acetohydroxyacid synthase (AHAS) was cloned from the archaeon Methanococcus aeolicus. Contrary to biochemical studies [Xing, R. and Whitman, W.B. (1994) J. Bacteriol. 176, 1207-1213] the enzyme was encoded by two open reading frames (ORFs). Based on sequence homology, these ORFs were designated ilvB and ilvN for the large and small subunits of AHAS, respectively. A putative methanogen promoter preceded ilvB-ilvN, and a potential internal promoter was found upstream of ilvN. ilvB encoded a 65-kDa protein, which agreed well with the measured value for the purified enzyme. ilvN encoded a 19-kDa protein, which fell within the range of M(r) of small subunits from other sources. Phylogenetic analysis of the deduced amino acid sequence of ilvB showed a close relationship between the AHAS of Bacteria and Archaea, to the exclusion of other enzymes in this family, including pyruvate oxidase, glyoxylate carboligase, pyruvate decarboxylase, and the acetolactate synthase found in fermentative Bacteria. Thus, this family of enzymes probably arose prior to the divergence of the Bacteria and Archaea. Moreover, the higher plant AHAS and the red algal AHAS were related to the AHAS II of Escherichia coli and the cyanobacterial AHAS, respectively. For this reason, these genes appear to have been acquired by the Eucarya during the endosymbiosis that gave rise to the mitochondrion and chloroplast, respectively. One of the ORFs in the Methanococcus jannaschii genome possesses high similarity to the M. aeolicus ilvB, indicating that it is an authentic AHAS.

Evolution of the biosynthesis of the branched-chain amino acids

The origin of the biosynthetic pathways for the branched-chain amino acids cannot be understood in terms of the backwards development of the present acetolactate pathway because it contains unstable intermediates. We propose that the first biosynthesis of the branched-chain amino acids was by the reductive carboxylation of short branched chain fatty acids giving keto acids which were then transaminated. Similar reaction sequences mediated by nonspecific enzymes would produce serine and threonine from the abundant prebiotic compounds glycolic and lactic acids. The aromatic amino acids may also have first been synthesized in this way, e.g. tryptophan from indole acetic acid. The next step would have been the biosynthesis of leucine from alpha-ketoisovaleric acid. The acetolactate pathway developed subsequently. The first version of the Krebs cycle, which was used for amino acid biosynthesis, would have been assembled by making use of the reductive carboxylation and leucine biosynthesis enzymes, and completed with the development of a single new enzyme, succinate dehydrogenase. This evolutionary scheme suggests that there may be limitations to inferring the origins of metabolism by a simple back extrapolation of current pathways.

Detection by site-specific disulfide cross-linking of a conformational change in binding of Escherichia coli pyruvate oxidase to lipid bilayers

Escherichia coli pyruvate oxidase, a peripheral membrane homotetrameric flavoprotein, exposes its C-terminal lipid binding site in the presence of substrate pyruvate and co-factor thiamine pyrophosphate Mg2+ and binds tightly to phospholipid bilayers during catalysis. Using site-specific disulfide cross-linking, we demonstrate that disulfide cross-links are formed between C termini of D560C pyruvate oxidase and that the degree of cross-linking is greatly increased by the presence of substrate and co-factors indicating a conformational change that results in juxtaposition of two subunit C termini. The cross-linked oxidase is enzymatically active and remains able to associate with lipid micelles. These results argue strongly that lipid bilayer binding of pyruvate oxidase involves pairing of the C termini of two subunits.

On the levels of enzymatic substrate specificity: implications for the early evolution of metabolic pathways

The most frequently invoked explanation for the origin of metabolic pathways is the retrograde evolution hypothesis. In contrast, according to the so-called "patchwork" theory, metabolism evolved by the recruitment of relatively inefficient small enzymes of broad specificity that could react with a wide range of chemically related substrates. In this paper it is argued that both sequence comparisons and experimental results on enzyme substrate specificity support the patchwork assembly theory. The available evidence supports previous suggestions that gene duplication events followed by a gradual neoDarwinian accumulation of mutations and other minute genetic changes lead to the narrowing and modification of enzyme function in at least some primordial metabolic pathways.

The oxygenase reaction of acetolactate synthase detected by chemiluminescence

In addition to the synthesis of ketolacids the enzyme acetolactate synthase shows an oxygen-consuming side reaction. Partially purified acetolactate synthase from corn (Zea mays L.) and barley (Hordeum vulgare L.) exhibits chemiluminescence in the presence of oxygen, Mn2+ and low concentrations of pyruvate. Light emission is inhibited by azide, but not by catalse or superoxide dismutase. The data suggest the formation of singlet oxygen during the catalytic cycle, and provides a basis for a highly sensitive assay for the oxygenase reaction of acetolactate synthesis. Both synthase activity and chemiluminescence are inhibited by sulfonylurea herbicides. The results add a new aspect to the irreversible inhibition of acetolactate synthase by these herbicides which may be enhanced by the presence of reactive oxygen species.

Purification and characterization of the oxygen-sensitive acetohydroxy acid synthase from the archaebacterium Methanococcus aeolicus

Acetohydroxy acid synthase (EC 4.1.3.18) of the archaebacterium Methanococcus aeolicus was purified 1,150-fold to homogeneity. The molecular weight of the purified enzyme was 125,000, and it contained only one type of subunit (M(r) = 58,000). The amino-terminal sequence had 46 to 57% similarity to those of the large subunits of the eubacterial anabolic enzymes and 37 to 43% similarity to those of the yeast and plant enzymes. The methanococcal enzyme had a pH optimum of 7.6. The pI, estimated by chromatofocusing, was 5.6. Activity required Mg2+ or Mn2+ ions, thiamine pyrophosphate, and a flavin. Flavin adenine dinucleotide, flavin mononucleotide, and riboflavin plus 10 mM phosphate all supported activity. However, activity was strongly inhibited by these flavins at 0.3 mM. The Michaelis constants for pyruvate, MgCl2, MnCl2, thiamine pyrophosphate, flavin adenine dinucleotide, and flavin mononucleotide were 6.8 mM, 0.3 mM, 0.16 mM, 1.6 microM, 0.4 microM, and 1.3 microM, respectively. In cell extracts, the enzyme was sensitive to O2 (half-life = 2.7 min with 5% O2 in the headspace), but the purified enzyme was less sensitive to O2 (half-life = 78.0 min with 20% O2). Reconstitution of the enzyme with flavin adenine dinucleotide increased the sensitivity to O2. Moreover, in the assay the homogeneous enzyme was rapidly inactivated by O2, and the concentration required for 50% inhibition (I50) was obtained with an atmosphere of 0.11% O2. The methanococcal enzyme has similarities to the eubacterial and eucaryotic enzymes, consistent with the ancient origin of the archaebacterial enzyme.

Expression of Escherichia coli pyruvate oxidase (PoxB) depends on the sigma factor encoded by the rpoS(katF) gene

The activity of Escherichia coli pyruvate oxidase (PoxB) was shown to be growth-phase dependent; the enzyme activity reaches a maximum at early stationary phase. We report that PoxB activity is dependent on a functional rpoS(katF) gene which encodes a sigma factor required to transcribe a number of stationary-phase-induced genes. PoxB activity as well as the beta-galactosidase encoded by a poxB::lacZ protein fusion was completely abolished in a strain containing a defective rpoS gene. Northern and primer extension analyses showed that poxB expression was regulated at the transcriptional level and was transcribed from a single promoter; the 5' end of the mRNA being located 27 bp upstream of the translational initiation codon of poxB. The poxB gene was expressed at decreased levels under anaerobiosis; however, the anaerobic regulatory genes arcA, arcB or fnr were not involved in anaerobic poxB gene expression. Expression of the rpoS(katF) gene has been reported to be affected by acetate, the product of PoxB reaction. However, we found that poxB null mutations had no effect on rpoS(katF) expression. Inactivation of two genes involved in acetate metabolism, ackA and pta, had no effect on either poxB or rpoS(katF) expression.

Isoleucine synthesis in Corynebacterium glutamicum: molecular analysis of the ilvB-ilvN-ilvC operon

Acetohydroxy acid synthase (AHAS) and isomeroreductase (IR) catalyze subsequent reactions in the flux of metabolites towards isoleucine, valine, leucine, and pantothenate. A 4,705-bp DNA fragment from Corynebacterium glutamicum known to code for AHAS and IR was sequenced and analyzed by Northern (RNA blot) analysis. As in other bacteria, the AHAS of this gram-positive organism is encoded by two genes, ilvB and ilvN. Gene disruption verified that these genes encode the single AHAS activity in C. glutamicum. The start of ilvB was determined by amino-terminal sequencing of a fusion peptide. By Northern analysis of the ilvBNC cluster, three in vivo transcripts of 3.9, 2.3, and 1.1 kb were identified, corresponding to ilvBNC, ilvNC, and ilvC messages, respectively. The ilvC transcript (encoding IR) was by far the most abundant one. With a clone from which the ilvB upstream regions had been deleted, only the ilvNC and ilvC transcripts were synthesized, and with a clone from which the ilvN upstream regions had been deleted, only the smallest ilvC transcript was formed. It is therefore concluded that in the ilv operon of C. glutamicum, three promoters are active. The amounts of the ilvBNC and ilvNC transcripts increased in response to the addition of alpha-ketobutyrate to the growth medium. This was correlated to an increase in specific AHAS activity, whereas IR activity was not increased because of the relatively large amount of the ilvC transcript present under all conditions assayed. Therefore, the steady-state level of the ilvBNC and ilvNC messages contributes significantly to the total activity of the single AHAS. The ilvC transcript of this operon, however, is regulated independently and present in a large excess, which is in accord with the constant IR activities determined.

Properties of subcloned subunits of bacterial acetohydroxy acid synthases

The acetohydroxy acid synthase (AHAS) isozymes from enterobacteria are each composed of a large and small subunit in an alpha 2 beta 2 structure. It has been generally accepted that the large (ca. 60-kDa) subunits are catalytic, while the small ones are regulatory. In order to further characterize the roles of the subunits as well as the nature and the specificities of their interactions, we have constructed plasmids encoding the large or small subunits of isozymes AHAS I and AHAS III, each with limited remnants of the other peptide. The catalytic properties of the large subunits have been characterized and compared with those of extracts containing the intact enzyme or of purified enzymes. Antisera to the isolated subunits have been used in Western blot (immunoblot) analyses for qualitative and semiquantitative determinations of the presence of the polypeptides in extracts. The large subunits of AHAS isozymes I and III have lower activities than the intact enzymes: Vmax/Km is 20 to 50 times lower in both cases. However, for AHAS I, most of this difference is due to the raised Km of the large subunit alone, while for AHAS III, it is due to a lowered Vmax. The substrate specificities, R, of large subunits are close to those of the intact enzymes. The catalytic activity of the large subunits of AHAS I is dependent on flavin adenine dinucleotide (FAD), as is that of the intact enzyme, although the apparent affinities of the large subunits alone for FAD are 10-fold lower. Isolated subunits are insensitive to valine inhibition. Nearly all of the properties of the intact AHAS isozyme I or III can be reconstituted by mixing extracts containing the respective large and small subunits. The mixing of subunits from different enzymes does not lead to activation of the large subunits. It is concluded that the catalytic machinery of these AHAS isozymes is entirely contained within the large subunits. The small subunits are required, however, for specific stabilization of an active conformation of the large subunits as well as for value sensitivity.

Early steps of isoprenoid biosynthesis in Escherichia coli

The incorporation of 2H- and 13C-labelled precursors into ubiquinone-8 (Uq-8) by strains of Escherichia coli was measured in order to define the pathway for the early steps in the biosynthesis of isoprenoids in these eubacteria. Cells grown with DL-[methyl-2H6]valine were found to label both the alpha-oxoisovaleric ('alpha-ketoisovaleric') acid alpha-oxoisohexanoic ('alpha-ketoisocaproic') acid, but not the Uq-8. Since these acids are required for the biosynthesis of isoprenoids by the acetolactate pathway, the operation of this pathway in the biosynthesis of Uq-8 is excluded. Cells grown with [1,2-13C2]acetate and non-labelled glucose readily incorporated 13C2 units into fatty acids, but failed to incorporate any label into the Uq-8. Cells grown with [U-13C6]glucose and non-labelled acetate, however, were found to label both the fatty acids and the Uq-8. Oxidative cleavage with periodate/permanganate of the Uq-8 isolated from cells grown with U-13C6-labelled glucose produced laevulinic acid, which was shown to be derived from two C2 units and one C1 unit of the labelled glucose by mass-spectral analysis of the 4,5-dihydro-6-methyl-2-phenylpyridazin-3(2H)-one derivative. The results of this work indicate that the C-2 and C-3 carbon unit of pyruvate, not acetyl-CoA, is the precursor to isopentenyl pyrophosphate (IPP) in these cells; however, the labelling pattern observed is consistent with the established acetoacetate pathway of isoprenoid biosynthesis. These data, coupled with the observed lack of inhibition of the growth of E. coli by mevinolin, a specific inhibitor of 3-hydroxy-3-methylglutaryl-CoA, can be best rationalized by the biosynthesis of IPP occurring in E. coli through a series of bound intermediates.

Acetohydroxy acid synthase activity from a mutation at ilvF in Escherichia coli K-12

Examination of the ilvF locus at 54 min on the Escherichia coli K-12 chromosome revealed that it is a cryptic gene for expression of a valine-resistant acetohydroxy acid synthase (acetolactate synthase; EC 4.1.3.18) distinct from previously reported isozymes. A spontaneous mutation, ilvF663, yielded IlvF+ enzyme activity that was multivalently repressed by all three branched-chain amino acids, was completely insensitive to feedback inhibition, was highly stable at elevated temperatures, and expressed optimal activity at 50 degrees C. The IlvF+ enzyme activity was expressed in strains in which isozyme II was inactive because of the ilvG frameshift in the wild-type strain K-12 and isozymes I and III were inactivated by point mutations or deletions. Tn5 insertional mutagenesis yielded two IlvF- mutants, with the insertion in ilvF663 in each case. These observations suggest that the ilvF663 locus may be a coding region for a unique acetohydroxy acid synthase activity.

A common structural motif in thiamin pyrophosphate-binding enzymes

The amino acid sequences of a wide range of enzymes that utilize thiamin pyrophosphate (TPP) as cofactor have been compared. A common sequence motif approximately 30 residues in length was detected, beginning with the highly conserved sequence -GDG- and concluding with the highly conserved sequence -NN-. Secondary structure predictions suggest that the motif may adopt a beta alpha beta fold. The same motif was recognised in the primary structure of a protein deduced from the DNA sequence of a hitherto unassigned open reading frame of Rhodobacter capsulata. This putative protein exhibits additional homology with some but not all of the TPP-binding enzymes.