Strategic manipulation of the substrate specificity of Saccharomyces cerevisiae flavocytochrome b2

Direct Electron Transfer of Enzymes Facilitated by Cytochromes

The direct electron transfer (DET) of enzymes has been utilized to develop biosensors and enzymatic biofuel cells on micro- and nanostructured electrodes. Whereas some enzymes exhibit direct electron transfer between their active-site cofactor and an electrode, other oxidoreductases depend on acquired cytochrome domains or cytochrome subunits as built-in redox mediators. The physiological function of these cytochromes is to transfer electrons between the active-site cofactor and a redox partner protein. The exchange of the natural electron acceptor/donor by an electrode has been demonstrated for several cytochrome carrying oxidoreductases. These multi-cofactor enzymes have been applied in third generation biosensors to detect glucose, lactate, and other analytes. This review investigates and classifies oxidoreductases with a cytochrome domain, enzyme complexes with a cytochrome subunit, and covers designed cytochrome fusion enzymes. The structurally and electrochemically best characterized proponents from each enzyme class carrying a cytochrome, that is, flavoenzymes, quinoenzymes, molybdenum-cofactor enzymes, iron-sulfur cluster enzymes, and multi-haem enzymes, are featured, and their biochemical, kinetic, and electrochemical properties are compared. The cytochromes molecular and functional properties as well as their contribution to the interdomain electron transfer (IET, between active-site and cytochrome) and DET (between cytochrome and electrode) with regard to the achieved current density is discussed. Protein design strategies for cytochrome-fused enzymes are reviewed and the limiting factors as well as strategies to overcome them are outlined.

Trifluorosubstrates as mechanistic probes for an FMN-dependent l-2-hydroxy acid-oxidizing enzyme

A controversy exists with respect to the mechanism of l-2-hydroxy acid oxidation by members of a family of FMN-dependent enzymes. A so-called carbanion mechanism was initially proposed, in which the active site histidine abstracts the substrate α-hydrogen as a proton, followed by electron transfer from the carbanion to the flavin. But an alternative mechanism was not incompatible with some results, a mechanism in which the active site histidine instead picks up the substrate hydroxyl proton and a hydride transfer occurs. Even though more recent experiments ruling out such a mechanism were published (Rao & Lederer (1999) Protein Science 7, 1531-1537), a few authors have subsequently interpreted their results with variant enzymes in terms of a hydride transfer. In the present work, we analyse the reactivity of trifluorolactate, a substrate analogue, with the flavocytochrome b2 (Fcb2) flavodehydrogenase domain, compared to its reactivity with an NAD-dependent lactate dehydrogenase (LDH), for which this compound is known to be an inhibitor (Pogolotti & Rupley (1973) Biochem. Biophys. Res. Commun, 55, 1214-1219). Indeed, electron attraction by the three fluorine atoms should make difficult the removal of the α-H as a hydride. We also analyse the reactivity of trifluoropyruvate with the FMN- and NAD-dependent enzymes. The results substantiate a different effect of the fluorine substituents on the two enzymes compared to their normal substrates. In the discussion we analyse the conclusions of recent papers advocating a hydride transfer mechanism for the family of l-2-hydroxy acid oxidizing FMN-dependent enzymes.

QM/MM study of l-lactate oxidation by flavocytochrome b2

Free energy surfaces calculated from a state-of-the-art computational methodology highlight the role of active site residues in l-lactate oxidation by flavocytochrome b₂.

The Ala95-to-Gly substitution in Aerococcus viridans l-lactate oxidase revisited - structural consequences at the catalytic site and effect on reactivity with O2 and other electron acceptors

Aerococcus viridansl-lactate oxidase (avLOX) is a biotechnologically important flavoenzyme that catalyzes the conversion of L-lactate and O₂ into pyruvate and H₂O₂. The enzymatic reaction underlies different biosensor applications of avLOX for blood L-lactate determination. The ability of avLOX to replace O₂ with other electron acceptors such as 2,6-dichlorophenol-indophenol (DCIP) allows the possiblity of analytical and practical applications. The A95G variant of avLOX was previously shown to exhibit lowered reactivity with O₂ compared to wild-type enzyme and therefore was employed in a detailed investigation with respect to the specificity for different electron acceptor substrates. From stopped-flow experiments performed at 20 °C (pH 6.5), we determined that the A95G variant (fully reduced by L-lactate) was approximately three-fold more reactive towards DCIP (1.0 ± 0.1 × 10(6) M(-1) ·s(-1) ) than O₂, whereas avLOX wild-type under the same conditions was 14-fold more reactive towards O₂(1.8 ± 0.1 × 10(6) m(-1) ·s(-1)) than DCIP. Substituted 1,4-benzoquinones were up to five-fold better electron acceptors for reaction with L-lactate-reduced A95G variant than wild-type. A 1.65-Å crystal structure of oxidized A95G variant bound with pyruvate was determined and revealed that the steric volume created by removal of the methyl side chain of Ala95 and a slight additional shift in the main chain at position Gly95 together enable the accomodation of a new active-site water molecule within hydrogen-bond distance to the N5 of the FMN cofactor. The increased steric volume available in the active site allows the A95G variant to exhibit a similar trend with the related glycolate oxidase in electron acceptor substrate specificities, despite the latter containing an alanine at the analogous position.

Microbial lactate utilization: enzymes, pathogenesis, and regulation

Lactate utilization endows microbes with the ability to use lactate as a carbon source. Lactate oxidizing enzymes play key roles in the lactate utilization pathway. Various types of these enzymes have been characterized, but novel ones remain to be identified. Lactate determination techniques and biocatalysts have been developed based on these enzymes. Lactate utilization has also been found to induce pathogenicity of several microbes, and the mechanisms have been investigated. More recently, studies on the structure and organization of operons of lactate utilization have been carried out. This review focuses on the recent progress and future perspectives in understanding microbial lactate utilization.

MD and QM/MM studies on long-chain L-α-hydroxy acid oxidase: substrate binding features and oxidation mechanism

Long-chain L-α-hydroxy acid oxidase (LCHAO) is a flavin mononucleotide (FMN)-dependent oxidase that dehydrogenates l-α-hydroxy acids to keto acids. There were two different mechanisms, named as hydride transfer (HT) mechanism and carbanion (CA) mechanism, respectively, proposed about the catalytic process for the FMN-dependent L-α-hydroxy acid oxidases on the basis of biochemical data. However, crystallographic and kinetic studies could not provide enough evidence to prove one of the mechanisms or eliminate the alternative. In the present studies, theoretical computations were carried out to study the molecular mechanism for LCHAO-catalyzed dehydrogenation of L-lactate. Our molecular dynamics (MD) simulations indicated that L-lactate prefers to bind with LCHAO in a hydride transfer mode rather than a carbanion mode. Quantum mechanics/molecular mechanics (QM/MM) calculations were further carried out to obtain the optimized structures of reactants, transition states, and products at the level of ONIOM-EE (B3LYP/6-311++G(d,p)//B3LYP/6-31G(d,p):AMBER). Quantum chemical studies indicated that LCHAO-catalyzed dehydrogenation of L-lactate would be a stepwise catalytic reaction in a hydride transfer mechanism but not a carbanion mechanism. MD simulations, binding free energy calculations, and QM/MM computations were also implemented on the complex between L-lactate and Y129F mutant LCHAO. By comparing the Y129F mutant system with the wild-type system, it was further confirmed that the key residue Tyr129 in the active site of LCHAO would not affect L-lactate's binding to LCHAO but play an important role on the catalytic reaction process through an H-bond interaction.

Rationally re-designed mutation of NAD-independent L-lactate dehydrogenase: high optical resolution of racemic mandelic acid by the engineered Escherichia coli

Background

NAD-independent l-lactate dehydrogenase (l-iLDH) from Pseudomonas stutzeri SDM can potentially be used for the kinetic resolution of small aliphatic 2-hydroxycarboxylic acids. However, this enzyme showed rather low activity towards aromatic 2-hydroxycarboxylic acids.

Results

Val-108 of l-iLDH was changed to Ala by rationally site-directed mutagenesis. The l-iLDH mutant exhibited much higher activity than wide-type l-iLDH towards l-mandelate, an aromatic 2-hydroxycarboxylic acid. Using the engineered Escherichia coli expressing the mutant l-iLDH as a biocatalyst, 40 g·L^-1 of dl-mandelic acid was converted to 20.1 g·L^-1 of d-mandelic acid (enantiomeric purity higher than 99.5%) and 19.3 g·L^-1 of benzoylformic acid.

Conclusions

A new biocatalyst with high catalytic efficiency toward an unnatural substrate was constructed by rationally re-design mutagenesis. Two building block intermediates (optically pure d-mandelic acid and benzoylformic acid) were efficiently produced by the one-pot biotransformation system.

Structures of the G81A mutant form of the active chimera of (S)-mandelate dehydrogenase and its complex with two of its substrates

(S)-Mandelate dehydrogenase (MDH) from Pseudomonas putida, a membrane-associated flavoenzyme, catalyzes the oxidation of (S)-mandelate to benzoylformate. Previously, the structure of a catalytically similar chimera, MDH-GOX2, rendered soluble by the replacement of its membrane-binding segment with the corresponding segment of glycolate oxidase (GOX), was determined and found to be highly similar to that of GOX except within the substituted segments. Subsequent attempts to cocrystallize MDH-GOX2 with substrate proved unsuccessful. However, the G81A mutants of MDH and of MDH-GOX2 displayed approximately 100-fold lower reactivity with substrate and a modestly higher reactivity towards molecular oxygen. In order to understand the effect of the mutation and to identify the mode of substrate binding in MDH-GOX2, a crystallographic investigation of the G81A mutant of the MDH-GOX2 enzyme was initiated. The structures of ligand-free G81A mutant MDH-GOX2 and of its complexes with the substrates 2-hydroxyoctanoate and 2-hydroxy-3-indolelactate were determined at 1.6, 2.5 and 2.2 A resolution, respectively. In the ligand-free G81A mutant protein, a sulfate anion previously found at the active site is displaced by the alanine side chain introduced by the mutation. 2-Hydroxyoctanoate binds in an apparently productive mode for subsequent reaction, while 2-hydroxy-3-indolelactate is bound to the enzyme in an apparently unproductive mode. The results of this investigation suggest that a lowering of the polarity of the flavin environment resulting from the displacement of nearby water molecules caused by the glycine-to-alanine mutation may account for the lowered catalytic activity of the mutant enzyme, which is consistent with the 30 mV lower flavin redox potential. Furthermore, the altered binding mode of the indolelactate substrate may account for its reduced activity compared with octanoate, as observed in the crystalline state.

Active site and loop 4 movements within human glycolate oxidase: implications for substrate specificity and drug design

Human glycolate oxidase (GO) catalyzes the FMN-dependent oxidation of glycolate to glyoxylate and glyoxylate to oxalate, a key metabolite in kidney stone formation. We report herein the structures of recombinant GO complexed with sulfate, glyoxylate, and an inhibitor, 4-carboxy-5-dodecylsulfanyl-1,2,3-triazole (CDST), determined by X-ray crystallography. In contrast to most alpha-hydroxy acid oxidases including spinach glycolate oxidase, a loop region, known as loop 4, is completely visible when the GO active site contains a small ligand. The lack of electron density for this loop in the GO-CDST complex, which mimics a large substrate, suggests that a disordered to ordered transition may occur with the binding of substrates. The conformational flexibility of Trp110 appears to be responsible for enabling GO to react with alpha-hydroxy acids of various chain lengths. Moreover, the movement of Trp110 disrupts a hydrogen-bonding network between Trp110, Leu191, Tyr134, and Tyr208. This loss of interactions is the first indication that active site movements are directly linked to changes in the conformation of loop 4. The kinetic parameters for the oxidation of glycolate, glyoxylate, and 2-hydroxy octanoate indicate that the oxidation of glycolate to glyoxylate is the primary reaction catalyzed by GO, while the oxidation of glyoxylate to oxalate is most likely not relevant under normal conditions. However, drugs that exploit the unique structural features of GO may ultimately prove to be useful for decreasing glycolate and glyoxylate levels in primary hyperoxaluria type 1 patients who have the inability to convert peroxisomal glyoxylate to glycine.

Crystal Structure Analysis of Recombinant Rat Kidney Long Chain Hydroxy Acid Oxidase,

Long chain hydroxy acid oxidase (LCHAO) is a member of an FMN-dependent enzyme family that oxidizes L-2-hydroxy acids to ketoacids. LCHAO is a peroxisomal enzyme, and the identity of its physiological substrate is unclear. Mandelate is the most efficient substrate known and is commonly used in the test tube. LCHAO differs from most family members in that one of the otherwise invariant active site residues is a phenylalanine (Phe23) instead of a tyrosine. We now report the crystal structure of LCHAO. It shows the same beta8alpha8 TIM barrel structure as other structurally characterized family members, e.g., spinach glycolate oxidase (GOX) and the electron transferases yeast flavocytochrome b2 (FCB2) and Pseudomonas putida mandelate dehydrogenase (MDH). Loop 4, which is mobile in other family members, is visible in part. An acetate ion is present in the active site. The flavin interacts with the protein in the same way as in the electron transferases, and not as in GOX, an unexpected observation. An interpretation is proposed to explain this difference between GOX on one hand and FCB2 and LCHAO on the other hand, which had been proposed to arise from the differences between family members in their reactivity with oxygen. A comparison of models of the substrate bound to various published structures suggests that the very different reactivity with mandelate of LCHAO, GOX, FCB2, and MDH cannot be rationalized by a hydride transfer mechanism.

Role of glycine 81 in (S)-mandelate dehydrogenase from Pseudomonas putida in substrate specificity and oxidase activity

(S)-Mandelate dehydrogenase from Pseudomonas putida belongs to a FMN-dependent enzyme family that oxidizes (S)-alpha-hydroxyacids. Despite a high degree of sequence and structural similarity, this family can be divided into three subgroups based on the different oxidants utilized in the second oxidative half-reaction. Only the oxidases show high reactivity with molecular oxygen. Structural data indicate that the relative position of a peptide loop and the isoalloxazine ring of the FMN is slightly different in the oxidases compared to the dehydrogenases; the last residue on this loop is either an alanine or glycine. We examined the effect of the G81A, G81S, G81V, and G81D mutations in MDH on the overall reaction and especially on the suppression of activity with oxygen. G81A had a higher specificity for small substrates compared to that of wtMDH, though the affinity for (S)-mandelate was relatively unchanged. The rate of the first half-reaction was 20-130-fold slower for G81A and G81S; G81D and G81V had extremely low activity. Redox-potential measurements indicate that the reduction in activity is due to the decrease in electrophilicity of the FMN. The affinity for oxygen increased 10-15-fold for G81A and G81S relative to wtMDH; the rate of oxidation increased 2-fold for G81A. The increased reactivity with molecular oxygen did not correlate with the redox potentials and appears to primarily result from a higher affinity for oxygen. These results suggest that one of the ways the oxidase activity of MDH is controlled is through steric effects because of the relative positions of the FMN and the Gly81 loop.

The catalytic role of tyrosine 254 in flavocytochrome b2 (L-lactate dehydrogenase from baker's yeast). Comparison between the Y254F and Y254L mutant proteins

Flavocytochrome b2 catalyses the oxidation of L-lactate to pyruvate in yeast mitochondrial intermembrane space. Its flavoprotein domain is a member of a family of FMN-dependent 2-hydroxy-acid-oxidizing enzymes. Numerous solution studies suggest that the first step of the reaction consists of proton abstraction from lactate C2, leading to a carbanion that subsequently yields electrons to FMN. The crystal structure suggests that the enzyme base is His373, and that Tyr254 may be hydrogen bonded to the substrate hydroxyl. Studies carried out with the Y254F mutant [Dubois, J., Chapman, S.K., Mathews, F.S., Reid, G.A. & Lederer, F. (1990) Biochemistry 29, 6393-6400] showed that Tyr254 does not act as a base but stabilizes the transition state. As the mutation did not induce any change in substrate affinity, the question of the existence of the hydrogen bond in the Michaelis complex remained open. Similar results with glycolate oxidase, mutated at the same position, led to the suggestion that these enzymes actually operate via a hydride transfer mechanism [Macheroux, P., Kieweg, V., Massey, V., Soderlind, E., Stenberg, K. & Lindqvist, Y. (1993) Eur. J. Biochem. 213, 1047-1054]. In the present work, we have re-investigated the matter by analysing the properties of a Y254L mutant flavocytochrome b2, as well as the behaviour of the Y254F enzyme with two substrates other than lactate, and a series of inhibitors. The Y254L protein is less efficient with L-lactate than the wild-type enzyme by a factor of 500, but the substrate affinity is unchanged. In contrast, L-phenyllactate and mandelate, poor substrates (the latter acting more as an inhibitor), exhibit an increased affinity. In addition, the Y254L mutant enzyme is more efficient with phenyllactate than lactate as a substrate. In order to rationalize these observations, we have modelled phenyllactate and mandelate in the active site, using previously described modelling experiments with lactate as a starting point. The results indicate that mandelate cannot bind in an orientation allowing proton abstraction by His373, due to steric interference by the side chains of Ala198 and Leu230. It might possibly adopt a binding mode as proposed previously for lactate, which leads to a hydride transfer and with which the 198 and 230 side chains do not interfere. However, other researchers [Sinclair, R., Reid, G.A. & Chapman, S.K. (1998) Biochem. J. 333, 117-120] showed that A198G, L230A and A198G/L230A mutant enzymes exhibit a strongly improved mandelate dehydrogenase activity. These results indicate that relief of the steric crowding facilitates catalysis by enabling a better mandelate orientation at the active site, suggesting that its productive binding mode is similar to that proposed for lactate in the carbanion mechanism. The modelling studies therefore support the hypothesis of a carbanion mechanism for all substrates. In addition, we present the effect of the two mutations at position 254 on the binding of a number of competitive inhibitors (such as sulfite, D-lactate, propionate) and of inhibitors that are known to bind at the active site both when the flavin is oxidized and when it is in the semiquinone state (propionate, oxalate and L-lactate at high concentrations). Unexpectedly, the results indicate that the integrity of Tyr254 is necessary for the binding of these inhibitors at the semiquinone stage.

Re-design of Saccharomyces cerevisiae flavocytochrome b2: introduction of L-mandelate dehydrogenase activity

Flavocytochrome b2 from Saccharomyces cerevisiae is an l-lactate dehydrogenase which exhibits only barely detectable activity levels towards another 2-hydroxyacid, l-mandelate. Using protein engineering methods we have altered the active site of flavocytochrome b2 and successfully introduced substantial mandelate dehydrogenase activity into the enzyme. Changes to Ala-198 and Leu-230 have significant effects on the ability of the enzyme to utilize l-mandelate as a substrate. The double mutation of Ala-198-->Gly and Leu-230-->Ala results in an enzyme with a kcat value (25 degrees C) with L-mandelate of 8.5 s-1, which represents an increase of greater than 400-fold over the wild-type enzyme. Perhaps more significantly, the mutant enzyme has a catalytic efficiency (as judged by kcat/Km values) that is 6-fold higher with l-mandelate than it is with L-lactate. Closer examination of the X-ray structure of S. cerevisiae flavocytochrome b2 led us to conclude that one of the haem propionate groups might interfere with the binding of L-mandelate at the active site of the enzyme. To test this idea, the activity with l-mandelate of the independently expressed flavodehydrogenase domain (FDH), was examined and found to be higher than that seen with the wild-type enzyme. In addition, the double mutation of Ala-198-->Gly and Leu-230-->Ala introduced into FDH produced the greatest mandelate dehydrogenase activity increase, with a kcat value more than 700-fold greater than that seen with the wild-type holoenzyme. In addition, the enzyme efficiency (kcat/Km) of this mutant enzyme was more than 20-fold greater with L-mandelate than with l-lactate. We have therefore succeeded in constructing an enzyme which is now a better mandelate dehydrogenase than a lactate dehydrogenase.

L-Mandelate dehydrogenase from Rhodotorula graminis: cloning, sequencing and kinetic characterization of the recombinant enzyme and its independently expressed flavin domain

The l-mandelate dehydrogenase (L-MDH) from the yeast Rhodotorula graminis is a mitochondrial flavocytochrome b2 which catalyses the oxidation of mandelate to phenylglyoxylate coupled with the reduction of cytochrome c. We have used the N-terminal sequence of the enzyme to isolate the gene encoding this enzyme using the PCR. Comparison of the genomic sequence with the sequence of cDNA prepared by reverse transcription PCR revealed the presence of 11 introns in the coding region. The predicted amino acid sequence indicates a close relationship with the flavocytochromes b2 from Saccharomyces cerevisiae and Hansenula anomala, with about 40% identity to each. The sequence shows that a key residue for substrate specificity in S. cerevisiae flavocytochrome b2, Leu-230, is replaced by Gly in L-MDH. This substitution is likely to play an important part in determining the different substrate specificities of the two enzymes. We have developed an expression system and purification protocol for recombinant L-MDH. In addition, we have expressed and purified the flavin-containing domain of L-MDH independently of its cytochrome domain. Detailed steady-state and pre-steady-state kinetic investigations of both L-MDH and its independently expressed flavin domain have been carried out. These indicate that L-MDH is efficient with both physiological (cytochrome c, kcat=225 s-1 at 25 degrees C) and artificial (ferricyanide, kcat=550 s-1 at 25 degrees C) electron acceptors. Kinetic isotope effects with [2-2H]mandelate indicate that H-C-2 bond cleavage contributes somewhat to rate-limitation. However, the value of the isotope effect erodes significantly as the catalytic cycle proceeds. Reduction potentials at 25 degrees C were measured as -120 mV for the 2-electron reduction of the flavin and -10 mV for the 1-electron reduction of the haem. The general trends seen in the kinetic studies show marked similarities to those observed previously with the flavocytochrome b2 (L-lactate dehydrogenase) from S. cerevisiae.