Crystal structure of D-amino acid oxidase: a case of active site mirror-image convergent evolution with flavocytochrome b2

A conserved aspartate is essential for FAD binding and catalysis in the D-amino acid oxidase from Trigonopsis variabilis

To evaluate the possible contribution of Asp206 of Trigonopsis variabilis D-amino acid oxidase (DAO) to its flavin adenine dinucleotide (FAD) binding and catalytic function, six mutant enzymes were constructed by site-directed mutagenesis. Western immunoblot analysis revealed that a protein with an apparent molecular mass of about 39.2 kDa was present in the cell-free extracts of wild-type and mutant strains. Replacement of Asp206 with Leu, Gly, and Asn resulted in the loss of DAO activity and characteristic absorption spectrum for flavoenzyme, while the other mutant DAOs, Asp206Glu, Asp206Ser, and Asp206Ala, exhibited a similar spectral profile to that of wild-type enzyme and retained about 6-90% of the enzyme activity. These results suggested that Asp206 of T. variahilis DAO might play an important role in the binding of FAD.

The soluble alpha-glycerophosphate oxidase from Enterococcus casseliflavus. Sequence homology with the membrane-associated dehydrogenase and kinetic analysis of the recombinant enzyme

The soluble flavoprotein alpha-glycerophosphate oxidase from Enterococcus casseliflavus catalyzes the oxidation of a "non-activated" secondary alcohol, in contrast to the flavin-dependent alpha-hydroxy- and alpha-amino acid oxidases. Surprisingly, the alpha-glycerophosphate oxidase sequence is 43% identical to that of the membrane-associated alpha-glycerophosphate dehydrogenase from Bacillus subtilis; only low levels of identity (17-22%) result from comparisons with other FAD-dependent oxidases. The recombinant alpha-glycerophosphate oxidase is fully active and stabilizes a flavin N(5)-sulfite adduct, but only small amounts of intermediate flavin semiquinone are observed during reductive titrations. Direct determination of the redox potential for the FAD/FADH2 couple yields a value of -118 mV; the protein environment raises the flavin potential by 100 mV in order to provide for a productive interaction with the reducing substrate. Steady-state kinetic analysis, using the enzyme-monitored turnover method, indicates that a ping-pong mechanism applies and also allows the determination of the corresponding kinetic constants. In addition, stopped-flow studies of the reductive half-reaction provide for the measurement of the dissociation constant for the enzyme. alpha-glycerophosphate complex and the rate constant for reduction of the enzyme flavin. These and other results demonstrate that this enzyme offers a very promising paradigm for examining the protein determinants for flavin reactivity and mechanism in the energy-yielding metabolism of alpha-glycerophosphate.

About the pKa of the active-site histidine in flavocytochrome b2 (yeast L-lactate dehydrogenase)

Flavocytochrome b2 or L-lactate dehydrogenase from yeasts catalyzes the oxidation of L-lactate at the expense of monoelectronic acceptors such as cytochrome c, its physiological partner. When incubated in the presence of both L-lactate and a keto acid, the enzyme catalyzes a transhydrogenation reaction wherein only the flavin is involved. During this reaction, the substrate alpha-hydrogen is transferred not only to the solvent but also in part to the keto acid, which acts as reverse substrate. Thus, when bound to the reduced enzyme, this hydrogen is sticky. In the context of a carbanion mechanism, it resides on Nepsilon of His373, the active site base. We have shown before that a correlation between the amount of intermolecular hydrogen transfer from [2-3H] lactate and the keto acid reverse substrate concentration enables the determination of the first-order rate constant, kHe, for exchange of the substrate-derived protein-bound hydrogen with bulk solvent (Urban P, Lederer F, 1985, J Biol Chem 260:11115-11122). In this work, we show that the exchange with the solvent appears to be independent of the phosphate buffer concentration in the range from 40 to 500 mM. It is thus probable that exchange occurs directly with water molecules. The second-order rate constant for exchange is then 0.16 (+/-0.03) M(-1) s(-1). Using the Eigen equation, this figure yields a pKa of 9.1+/-0.1 for His373 in the reduced enzyme, compared to a probable value of 6.0 or less in the oxidized enzyme (Suzuki H, Ogura YC, 1970, J Biochem 67:291-295). The mechanistic significance of these results is discussed.

The crystal structure of phenol hydroxylase in complex with FAD and phenol provides evidence for a concerted conformational change in the enzyme and its cofactor during catalysis

BACKGROUND

The synthesis of phenolic compounds as by-products of industrial reactions poses a serious threat to the environment. Understanding the enzymatic reactions involved in the degradation and detoxification of these compounds is therefore of much interest. Soil-living yeasts use flavin adenine dinucleotide (FAD)-containing enzymes to hydroxylate phenols. This reaction initiates a metabolic sequence permitting utilisation of the aromatic compound as a source of carbon and energy. The phenol hydroxylase from Trichosporon cutaneum hydroxylates phenol to catechol. Phenol is the best substrate, but the enzyme also accepts simple hydroxyl-, amino-, halogen- or methyl-substituted phenols.

RESULTS

The crystal structure of phenol hydroxylase in complex with FAD and phenol has been determined at 2.4 A resolution. The structure was solved by the MIRAS method. The protein model consists of two homodimers. The subunit consists of three domains, the first of which contains a beta sheet that binds FAD with a typical beta alpha beta nucleotide-binding motif and also a fingerprint motif for NADPH binding. The active site is located at the interface between the first and second domains; the second domain also binds the phenolic substrate. The third domain contains a thioredoxin-like fold and is involved in dimer contacts. The subunits within the dimer show substantial differences in structure and in FAD conformation. This conformational flexibility allows the substrate to gain access to the active site and excludes solvent during the hydroxylation reaction.

CONCLUSIONS

Two of the domains of phenol hydroxylase are similar in structure to p-hydroxybenzoate hydroxylase. Thus, phenol hydroxylase is a member of a family of flavin-containing aromatic hydroxylases that share the same overall fold, in spite of large differences in amino acid sequences and chain length. The structure of phenol hydroxylase is consistent with a hydroxyl transfer mechanism via a peroxo-FAD intermediate. We propose that a movement of FAD takes place in concert with a large conformational change of residues 170-210 during catalysis.

The glutathione dependence of inorganic sulfate formation from L- or D-cysteine in isolated rat hepatocytes

The GSH dependence of the metabolic pathways involved in the conversion of cysteine to sulfate in intact cells has been investigated. It was found that hepatocyte-catalysed sulfate formation from added L-cysteine did not occur if hepatocyte GSH was depleted beforehand, but was restored when GSH levels recovered. Furthermore, sulfate formation did not recover in GSH-depleted hepatocytes if GSH synthesis was prevented with buthionine sulfoximine. Thiosulfate formation was, however, markedly enhanced in GSH-depleted hepatocytes. These results suggest that thiosulfate is an intermediate in the formation of inorganic sulfate from L-cysteine and that GSH was required for the conversion of thiosulfate to inorganic sulfate. Much less sulfate was formed if the cysteine was replaced with cysteinesulfinate. Furthermore, sulfate formation from L-cysteine was markedly inhibited by the addition of the transaminase inhibitor DL-cycloserine or the gamma-cystathionase inhibitor DL-propargylglycine. The major routes of sulfate formation from L-cysteine therefore seems to involve pathways that do not involve L-cysteinesulfinate. Similar amounts of sulfate were formed from D-cysteine as L-cysteine. Thiosulfate instead of sulfate was also formed in GSH-depleted hepatocytes. However, sulfate formation from D-cysteine differed from L-cysteine in that it was inhibited by the D-aminoacid oxidase inhibitor sodium benzoate and was not affected by transaminase or gamma-cystathionase inhibitors. These results suggest that thiosulfate is an intermediate in sulfate formation from D-cysteine and involves the oxidation of D-cysteine by D-amino acid oxidase to form beta-mercaptopyruvate.

The PHBH fold: not only flavoenzymes

p-Hydroxybenzoate hydroxylase, D-amino acid oxidase, cholesterol oxidase and glucose oxidase form a family of structurally related flavoenzymes. Comparison of their three-dimensional structures reveal how the same FAD-binding scaffold has been employed to implement diverse active-site architectures, suited for different types of catalytic reactions. The substrate binding mode differs in each of these enzymes, with the catalytically relevant residues not located on homologous positions. A common feature is provided by the ability of these enzyme to bury their substrates beneath the protein surface. In D-amino acid oxidase and cholesterol oxidase, a loop forms a 'lid' controlling the active site accessibility, whereas in p-hydroxybenzoate hydroxylase is the flavin itself, which swings out to allow substrate binding. The crystallographic analysis has revealed that the GTP-dissociation inhibitor of RAB GTPases has a folding topology remarkably similar to p-hydroxybenzoate hydroxylase. This finding highlights the versatile nature of this folding topology, which in addition to flavin-dependent catalysis, is suited for diverse functions, such as the regulation of GTPases.

Limited proteolysis and site-directed mutagenesis reveal the origin of microheterogeneity in Rhodotorula gracilis D-amino acid oxidase

When analysed by isoelectric focusing, D-amino acid oxidase from the yeast Rhodotorula gracilis normally consists of three molecular isoforms (pI 7.8, 7.4 and 7.2, respectively) all with the same N-terminal sequence. However, only a single band of pI 7.8 is detected with the recombinant wild-type protein expressed in E. coli. To determine whether the molecular basis of this heterogeneity is due to proteolysed forms of the protein, we treated R. gracilis D-amino acid oxidase with various proteases. Limited proteolysis by chymotrypsin and thermolysin produced truncated and nicked monomeric holoenzymes containing two polypeptides of approximately 34 kDa (Met1-Leu312) and one of approximately 5 kDa (Ala319-Arg364 with chymotrypsin or Ala319-Ala362 with thermolysin). On the other hand, treatment with endoproteinase Glu-C gave a dimeric holoenzyme lacking the C-terminal SKL tripeptide. This cleavage of Glu365-Ser366 peptide bond caused the disappearance of the three isoelectric bands and a single homogeneous band (pI 7.2) appeared. To study this protein form, we used site-directed mutagenesis to produce a mutant form of R. gracilis D-amino acid oxidase lacking the SKL C-terminal tripeptide (which is the targeting sequence PTS1 for peroxisomal proteins). As expected, the SKL-deleted mutant gave a single band (pI 7.2) in isoelectric focusing. The three-band pattern of native yeast enzyme was generated by in vitro experiments using an equimolar mixture of the wild-type (pI 7.8) and the SKL-deleted recombinant (pI 7.2) DAAOs. The microheterogeneity of yeast DAAO thus stems from the association of two polypeptide chains differing in the C-terminal tripeptide, giving three different holoenzyme dimers.

Rat D-amino-acid oxidase cDNA: rat D-amino-acid oxidase as an intermediate form between mouse and other mammalian D-amino-acid oxidases

Nucleotide sequence of cDNA encoding rat D-amino-acid oxidase (DAO) was determined. Two species of DAO mRNA were present in rat kidney, liver, and brain. They were probably produced by alternative splicing. Rat DAO cDNA encodes 346 amino acid residues, indicating that rat DAO is an intermediate form between mouse DAO (345 amino acids) and DAOs (347 amino acids) of human, rabbit, and pig. Deduced amino acid sequence indicates 93% identity between rat and mouse DAO. Northern hybridization and western blotting supported the sequence data.

Molecular cloning of TvDAO1, a gene encoding a D-amino acid oxidase from Trigonopsis variabilis and its expression in Saccharomyces cerevisiae and Kluyveromyces lactis

The DAO1 gene of Trigonopsis variabilis encoding a D-amino acid oxidase (EC 1.4.3.3) was isolated from genomic clones selected for their specific hybridization to synthetic oligodeoxyribonucleotide probes based on regions of the enzyme that have been conserved through evolution. The nucleotide sequence of the gene predicts a protein with similarities to human, pig, rabbit, mouse and Fusarium solani D-amino acid oxidases. The open reading frame of the T. variabilis DAO1 gene was interrupted by an intron. The Dao1p sequence displays two regions, one in the N-terminal section--the FAD binding site--and the other near the C-terminal region that contains conserved signatures found in all the D-amino acid oxidases. The three C-terminal amino acids suggest that the enzyme may be located in peroxisomes. Northern blot experiments showed that no transcriptional activation occurred in the presence of D-methionine. The cDNA encoding Dao1p was expressed in Saccharomyces cerevisiae and Kluyveromyces lactis. Both yeast species are able to synthesize a functional enzyme under the control of the GAL1 promoter. In K. lactis, up to six times more enzyme units per gram of dry weight are produced with a multicopy plasmid in comparison with the wild-type strain of T. variabilis. The yeast expression system we describe may constitute an alternative source for the production of D-amino acid oxidases at industrial level.

Structure of D-amino acid oxidase: new insights from an old enzyme

D-amino acid oxidase is the prototype of flavin-dependent oxidases. The recent resolution of its 3D structure has provided an explanation for several of its properties and has led to a substantial revision of the mechanism of D-amino acid dehydrogenation, with significant implications for the general understanding of flavin-dependent catalysis.

Expression of thiamin biosynthetic genes (thiCOGE) and production of symbiotic terminal oxidase cbb3 in Rhizobium etli

In this paper we report the cloning and sequence analysis of four genes, located on plasmid pb, which are involved in the synthesis of thiamin in Rhizobium etli (thiC, thiO, thiG, and thiE). Two precursors, 4-methyl-5-(beta-hydroxyethyl)thiazole monophosphate and 4-amino-5-hydroxymethylpyrimidine pyrophosphate, are coupled to form thiamin monophosphate, which is then phosphorylated to make thiamin pyrophosphate. The first open reading frame (ORF) product, of 610 residues, has significant homology (69% identity) with the product of thiC from Escherichia coli, which is involved in the synthesis of hydroxymethylpyrimidine. The second ORF product, of 327 residues, is the product of a novel gene denoted thiO. A protein motif involved in flavin adenine dinucleotide binding was found in the amino-terminal part of ThiO; also, residues involved in the catalytic site of D-amino acid oxidases are conserved in ThiO, suggesting that it catalyzes the oxidative deamination of some intermediate of thiamin biosynthesis. The third ORF product, of 323 residues, has significant homology (38% identity) with ThiG from E. coli, which is involved in the synthesis of the thiazole. The fourth ORF product, of 204 residues, has significant homology (47% identity) with the product of thiE from E. coli, which is involved in the condensation of hydroxymethylpyrimidine and thiazole. Strain CFN037 is an R. etli mutant induced by a single Tn5mob insertion in the promoter region of the thiCOGE gene cluster. The Tn5mob insertion in CFN037 occurred within a 39-bp region which is highly conserved in all of the thiC promoters analyzed and promotes constitutive expression of thiC. Primer extension analysis showed that thiC transcription in strain CFN037 originates within the Tn5 element. Analysis of c-type protein content and expression of the fixNOQP operon, which codes for the symbiotic terminal oxidase cbb3, revealed that CFN037 produces the cbb3 terminal oxidase. These data show a direct relationship between expression of thiC and production of the cbb3 terminal oxidase. This is consistent with the proposition that a purine-related metabolite, 5-aminoimidazole-4-carboxamide ribonucleotide, is a negative effector of the production of the symbiotic terminal oxidase cbb3 in R. etli.

Cloning, sequencing and expression in E. coli of a D-amino acid oxidase cDNA from Rhodotorula gracilis active on cephalosporin C

We have cloned the cDNA coding for the Rhodotorula gracilis D-amino acid oxidase (DAAO), an enzyme that performs with high catalytic efficiency biotechnologically relevant bioconversions, by PCR amplification. The first strand cDNA was synthesised from the total mRNA fraction isolated from R. gracilis cells grown under DAAO-inducing conditions. The R. gracilis DAAO cDNA consists of 1104 bp encoding a protein of 368 amino acids. The insertion of the cDNA into the pKK223-3 plasmid allowed the expression of recombinant DAAO in Escherichia coli as a wholly soluble and catalytically active holoenzyme (approximately 0.5 U mg-1 protein) with a fermentation yield, in terms of DAAO units, of 800 U l-1. This level of expression allowed the purification, in homogeneous form and high yield (50%), of the recombinant enzyme which showed a high catalytic activity on cephalosporin C as substrate. The nucleotide sequence reported in this paper will appear in the nucleotide sequence databases under accession number.

On the reaction mechanism of L-lactate oxidase: quantitative structure-activity analysis of the reaction with para-substituted L-mandelates

The rate constants for reduction of the flavoenzyme, L-lactate oxidase, and a mutant (in which alanine 95 is replaced by glycine), by a series of para-substituted mandelates, in both the 2-1H- and 2-2H- forms, have been measured by rapid reaction spectrophotometry. In all cases, significant isotope effects (1H/2H = 3-7) on the rate constants of flavin reduction were found, indicating that flavin reduction is a direct measure of alpha-C-H bond breakage. The rate constants show only a small influence of the electronic characteristics of the substituents, but show a good correlation when combined with some substituent volume parameters. A surprisingly good correlation is found with the molecular mass of the substrate. The results are compatible with any mechanism in which there is little development of charge in the transition state. This could be a transfer of hydride to the flavin N(5) position or a synchronous mechanism in which the alpha-C-H is formally abstracted as a H+ while the resulting charge is simultaneously neutralized by another event.

Glycerol-assisted restorative adjustment of flavoenzyme conformation perturbed by site-directed mutagenesis

The replacement of histidine 307 with leucine in pig kidney D-amino acid oxidase perturbs its active site conformation accompanied by dramatic losses in protein-flavin interactions and enzymatic activity. However, the negative effect of this mutation on the holoenzyme structure is essentially eliminated in the presence of glycerol, resulting in up to 50% activity recovery and greater than 16-fold increase in the flavin affinity. Further analysis revealed that glycerol assists in the rearrangement of the protein toward its holoenzyme-like conformation together with reduction in the solvent-accessible protein hydrophobic area as demonstrated by limited proteolysis and use of affinity and hydrophobic probes. A substantial decrease in the protein-flavin interactions was demonstrated at a low temperature, but this reversible process was completely blocked in the presence of 40% glycerol. We suggest that the perturbation of the D-amino acid oxidase active site is due to the nonpolar nature of the mutation whose negative impact on the holoenzyme structure can be overcome by glycerol-induced strengthening of protein internal hydrophobic interactions.

Catalytic mechanism of the oxidative demethylation of 4-(methoxymethyl)phenol by vanillyl-alcohol oxidase. Evidence for formation of a p-quinone methide intermediate

The catalytic mechanism for the oxidative demethylation of 4-(methoxymethyl)phenol by the covalent flavoprotein vanillyl-alcohol oxidase was studied. Using H218O, it was found that the carbonylic oxygen atom from the product 4-hydroxybenzaldehyde originates from a water molecule. Oxidation of vanillyl alcohol did not result in any incorporation of 18O. Enzyme-monitored turnover experiments revealed that for both substrates a process involving flavin reduction is rate determining. During anaerobic reduction of vanillyl-alcohol oxidase by 4-(methoxymethyl)phenol, a relatively stable spectral intermediate is formed. Deconvolution of its spectral characteristics showed a typical pH-independent absorption maximum at 364 nm (epsilon364 nm = 46 mM-1 cm-1). A similar transient species was observed upon anaerobic reduction by vanillyl alcohol. The rate of flavin reduction and synchronous intermediate formation by 4-(methoxymethyl)phenol is 3.3 s-1 and is fast enough to account for turnover (3.1 s-1). The anaerobic decay of the intermediate was too slow (0.01 s-1) to be of catalytical relevance. The reduced binary complex is rapidly reoxidized (1.5 x 10(5) M-1 s-1) and is accompanied with formation and release of product. Oxidation of free-reduced enzyme is an even faster process (3.1 x 10(5) M-1 s-1). The kinetic data for the oxidative demethylation of 4-(methoxymethyl)phenol are in accordance with a ternary complex mechanism in which the reduction rate is rate-limiting. It is proposed that, upon reduction, a binary complex is produced composed of the p-quinone methide of 4-(methoxymethyl)phenol and reduced enzyme.

Crystal structures and inhibitor binding in the octameric flavoenzyme vanillyl-alcohol oxidase: the shape of the active-site cavity controls substrate specificity

BACKGROUND

Lignin degradation leads to the formation of a broad spectrum of aromatic molecules that can be used by various fungal micro-organisms as their sole source of carbon. When grown on phenolic compounds, Penicillium simplicissimum induces the strong impression of a flavin-containing vanillyl-alcohol oxidase (VAO). The enzyme catalyses the oxidation of a vast array of substrates, ranging from aromatic amines to 4-alkyphenols. VAO is a member of a novel class of widely distributed oxidoreductases, which use flavin adenine dinucleotide (FAD) as a cofactor covalently bound to the protein. We have carried out the determination of the structure of VAO in order to shed light on the most interesting features of these novel oxidoreductases, such as the functional significance of covalent flavinylation and the mechanism of catalysis.

RESULTS

The crystal structure of VAO has been determined in the native state and in complexes with four inhibitors. The enzyme is an octamer with 42 symmetry; the inhibitors bind in a hydrophobic, elongated cavity on the si side of the flavin molecule. Three residues, Tyr108, Tyr503 and Arg504 form an anion-binding subsite, which stabilises the phenolate form of the substrate. The structure of VAO complexed with the inhibitor 4-(1-heptenyl)phenol shows that the catalytic cavity is completely filled by the inhibitor, explaining why alkylphenols bearing aliphatic substituents longer than seven carbon atoms do not bind to the enzyme.

CONCLUSIONS

The shape of the active-site cavity controls substrate specificity by providing a 'size exclusion mechanism'. Inside the cavity, the substrate aromatic ring is positioned at an angle of 18 degrees to the flavin ring. This arrangement is ideally suited for a hydride transfer reaction, which is further facilitated by substrate deprotonation. Burying the substrate beneath the protein surface is a recurrent strategy, common to many flavoenzymes that effect substrate oxidation or reduction via hydride transfer.

Identification of a reactive cysteine in the flavin-binding domain of Rhodotorula gracilis D-amino acid oxidase

The holoenzyme form of Rhodotorula gracilis D-amino acid oxidase, an 80-kDa homodimer, reacted only to a limited extent with general thiol reagents (2,2'-dithiodipyridine, 5,5'-dithiobis(2-nitrobenzoic acid), and N-[7-(dimethylamino)-4-methylcoumarinyl]maleimide) (60% residual activity), whereas the monomeric apoprotein was completely inactivated and denatured by these reagents. To investigate the presence of thiol residue(s) in the active site of the enzyme, the apoprotein was reconstituted with the 8-(methylsulfonyl)-FAD chemical-affinity probe. Competitive inhibition between this analogue and FAD for apoprotein binding was observed. The covalent attachment of the flavin analogue to the apoprotein was complete after approximately 20 h of incubation and the flavinylated enzyme, containing 8-(cysteinyl)-FAD, was monomeric and inactive. After HPLC isolation of the flavin-labeled tryptic peptides, Cys208 was identified as the only cysteine to react with the FAD analogue. These results show that a single cysteine of R. gracilis D-amino acid oxidase reacts with the flavin analogue and that this is located near or at the FAD-binding domain.

Crystallization and preliminary X-ray analysis of the flavoenzyme vanillyl-alcohol oxidase from Penicillium simplicissimum

Vanillyl-alcohol oxidase catalyses the oxidation of several 4-hydroxybenzyl alcohols by using 8-alpha-(N3-histidyl)-FAD as a covalently bound prosthetic group. Crystals of vanillyl-alcohol oxidase from Penicillium simplicissimum have been grown by using the vapor diffusion technique. The space group was found to be I, with cell dimensions a = b = 140.5 A, c = 132.9 A. Diffraction data have been recorded to 3.2 A resolution by using a laboratory source and to 2.5 A resolution on flash freezing the crystal at the ELETTRA Synchrotron X-ray diffraction beam line.

cDNA cloning and expression of the flavoprotein D-aspartate oxidase from bovine kidney cortex

The isolation and sequencing of the complete cDNA coding for a d-aspartate oxidase, as well as the overexpression of the recombinant active enzyme, are reported for the first time. This 2022 bp cDNA, beside the coding portion, comprises a 5' untranslated tract and the whole 3' region including the polyadenylation signal and the poly(A) tail. The encoded protein comprises 341 amino acids, with the last three residues (-Ser-Lys-Leu) representing a peroxisomal targeting signal 1 (PTS1), hitherto unknown for this protein. The overexpression of recombinant d-aspartate oxidase was achieved in a prokaryotic system, and a soluble and active enzyme was obtained which accounted for about 10% of total bacterial protein. Comparisons with the known cDNAs for mammalian d-amino acid oxidase, another peroxisomal enzyme, are also made. The close structural and functional similarities shared by these enzymes at the protein level are not reflected at the nucleic acid level.

On the mechanism of D-amino acid oxidase. Structure/linear free energy correlations and deuterium kinetic isotope effects using substituted phenylglycines

The kinetic mechanism of the reaction of D-amino acid oxidase (EC 1.4.3.3) from Trigonopsis variabilis with [alpha-1H]- and [alpha-2H]phenylglycine has been determined. The pH dependence of Vmax is compatible with pKa values of approximately 8.1 and >9.5, the former of which is attributed to a base which should be deprotonated for efficient catalysis. The deuterium isotope effect on turnover is approximately 3.9, and the solvent isotope effect approximately 1.6. The reductive half-reaction is biphasic, the first, fast phase, k2, corresponding to substrate dehydrogenation/enzyme flavin reduction and the second to conversion/release of product. Enzyme flavin reduction consists in an approach to equilibrium involving a finite rate for k-2, the reversal of k2. k2 is 28.8 and 4.6 s-1 for [alpha-1H]- and [alpha-2H]phenylglycine, respectively, yielding a primary deuterium isotope effect approximately 6. The solvent deuterium isotope effect on the apparent rate of reduction for [alpha-1H]- and [alpha-2H]phenylglycine is approximately 2.8 and approximately 5. The rates for k-2 are 4.2 and 0.9 s-1 for [alpha-1H]- and [alpha-2H]phenylglycine, respectively, and the corresponding isotope effect is approximately 4.7. The isotope effect on alpha-H and the solvent one thus behave multiplicatively consistent with a highly concerted process and a symmetric transition state. The k2 and k-2 values for phenylglycines carrying the para substituents F, Cl, Br, CH3, OH, NO2 and OCH3 have been determined. There is a linear correlation of k2 with the substituent volume VM and with sigma+; k-2 correlates best with sigma or sigma+ while steric parameters have little influence. This is consistent with the transition state being structurally similar to the product. The Bronsted plot of DeltaG versus DeltaG0 allows the estimation of the intrinsic DeltaG0 as approximately 58 kJ.M-1. From the linear free energy correlations, the relation of DeltaG versus DeltaG0 and according to the theory of Marcus it is concluded that there is little if any development of charge in the transition state. This, together with the recently solved three-dimensional structure of D-amino acid oxidase from pig kidney (Mattevi, A., Vanoni, M.A., Todone, F., Rizzi, M., Teplyakov, A., Coda, A., Bolognesi, M., and Curti, B. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 7496-7501), argues against a carbanion mechanism in its classical formulation. Our data are compatible with transfer of a hydride from the substrate alphaC-H to the oxidized flavin N(5) position, although, clearly, they cannot prove it.