Effect of hydrogen peroxide on d-amino acid oxidase from Rhodotorula gracilis

Enzyme and Microbial Technology 27 (2000) 234 –239

www.elsevier.com/locate/enzmictec

Effect of hydrogen peroxide on D-amino acid oxidase from Rhodotorula gracilis Isabel de la Mata, Fernando Ramo´n, Virginia Obrego´n, Ma Pilar Castillo´n, Carmen Acebal* Departamento de Bioquı´mica y Biologı´a Molecular, Facultad de Biologı´a, Universidad Complutense de Madrid, 28040 Madrid, Spain Received 19 February 1999; received in revised form 21 December 1999; accepted 25 February 2000

Abstract D-amino acid oxidase from Rhodotorula gracilis is a FAD-containing enzyme that belongs to the oxidase class that is characterized by the ability of the reduced flavin to react quickly with oxygen, yielding hydrogen peroxide and the oxidized cofactor. Hydrogen peroxide, necessary for the production of glutaryl-7-ACA from cephalosporin C had a deleterious effect on the enzyme. H2O2 induced the oxidation of tryptophan and cysteine residues of the protein that could be involved in the dimerization process, required for the attainment of a fully competent enzyme. H2O2 had also a kinetic effect on the reaction catalyzed by D-amino acid oxidase. It was a pure noncompetitive inhibitor; the corresponding inhibition constants were Kis ⫽ 0.52 mM and Kii ⫽ 0.70 mM. © 2000 Elsevier Science Inc. All rights reserved.

Keywords: Hydrogen peroxide; D-amino acid oxidase; Rhodotorula gracilis

1. Introduction D-amino acid oxidase (DAAO) (EC 1.4.3.3) is a flavoenzyme containing FAD as the prosthetic group. It catalyzes the oxidation of D-amino acids to the corresponding ␣-keto acid and ammonia; the coenzyme FAD, which is reduced during the course of the reaction, is reoxidized by molecular oxygen to yield hydrogen peroxide, according to the following scheme:

RCHNH2COOH ⫹ E-FAD 3 RC ⫽ NHCOOH ⫹ E-FADH2 E-FADH2 ⫹ O2 3 E-FAD ⫹ H2O2

(1) (2)

RC ⫽ NHCOOH ⫹ H2O 3 RCOCOOH ⫹ NH3

(3)

The reductive half reaction (1), in which FAD cofactor is reduced, is followed by FAD re-oxidation by molecular oxygen to give hydrogen peroxide (2). The amino acid hydrolyzes spontaneously to the corresponding keto acid and ammonia, in a nonenzymatic process (3). D-amino acid oxidase from the yeast Rhodotorula graci-

* Corresponding author. Tel.: ⫹34-91-394-41-50; fax: ⫹34-91-39442-76. E-mail address: [email protected] (C. Acebal).

lis converts cephalosporin C into ␣-ketoadipyl-7-aminocephalosporanic acid, which is further decarboxylated by the hydrogen peroxide formed as a product in the reaction, to give glutaryl-7-aminocephalosporanic acid (glutaryl-7ACA). Then, glutaryl-7-ACA can be transformed by glutaryl-7-ACA acylase to 7-ACA, an important starting material in the production of semisynthetic cephalosporins [1–3]. Although hydrogen peroxide is very important for the production of glutaryl-7-ACA, it can induce enzyme deterioration [3,4]. Actually, treatment of proteins with hydrogen peroxide is known to oxidize several amino acids, including cysteine, tryptophan, tyrosine, histidine, and methionine [5,6]. This fact is of great importance to design enzymatic bioreactors for the continous or semicontinous industrial production of 7-ACA. Szwajcer Dey et al. [3] show that, during the conversion of cephalosporin C to ketoadipyl-7-aminocephalosporanic acid by DAAO from Trigonopsis variabilis, with the simultaneous production of equimolecular amounts of hydrogen peroxide, an incomplete nonenzymatic conversion of the keto form into the glutaryl form occurs, where cephalosporin C as well as DAAO are partially destroyed in the presence of hydrogen peroxide. They also show that the presence of catalase coimmobilized with DAAO improved the operation stability of the enzyme, allowing both removal

0141-0229/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 2 2 - 2

I. de la Mata et al. / Enzyme and Microbial Technology 27 (2000) 234 –239

235

2.2. Methods 2.2.1. Enzyme purification D-amino acid oxidase was isolated and purified from R. gracilis (American Type Culture Collection, strain number 26217) as previously described [7]. Apoenzyme was prepared by dialysis against potassium bromide, as described by Casalin et al. [8], using a strategy originally proposed for preparation of mammalian apo-D-amino acid oxidase [9]. 2.2.2. Enzyme assay The activity of D-amino acid oxidase was determined at 35°C in 100 ␮l air-saturated incubation mixtures containing 0.7 ␮M enzyme and 10 mM D-alanine in 50 mM potassium phosphate buffer at pH 8.0. The released pyruvic acid was determined after 10 min by reacting with 2,4-dinitrophenylhydrazine, and the corresponding hydrazone was monitored at 450 nm. The product formation proceeded linearly with time for this period under all the conditions used in kinetic experiments. For product inhibition studies where pyruvate was added to reaction mixtures as an inhibitor, the activity was determined by quantification of the produced hydrogen peroxide with a coupled horseradish peroxidase/O-phenylenediamine system. Besides DAAO and D-Ala, reaction mixtures contained 40 ␮g/ml peroxidase and 50 ␮g/ml O-phenylenediamine, and H2O2 formation was followed at 411 nm.

of hydrogen peroxide and the recycling of O2. Therefore, hydrogen peroxide should be added, in a second stage, after the enzymatic transformation to ketoadipyl-7-ACA is completed. D-amino acid oxidase from R. gracilis seems to be more resistant to H2O2 deterioration than D-amino acid oxidases from other sources. This feature, together with the tight binding of FAD and its high catalytic activity, are relevant for its possible exploitation in industrial bioreactors. In this paper, we report the effect of hydrogen peroxide on D-amino acid oxidase from R. gracilis and also its kinetic effect as a final product of the reaction catalyzed by the enzyme.

2. Materials and methods 2.1. Materials D-alanine, D-serine, hydrogen peroxide, pyruvic acid, dithio-bis-nitrobenzoic acid (DTNB), FAD, and 2,4-dinitrophenylhydrazine were from Sigma Chemical (St. Louis, MO, USA). Iodo-[14C]acetic acid was from Amersham Life Science (Buckinghamshire, England). All other reagents and solvents were of analytical grade and were purchased from Merck (Darmstadt, Germany).

2.2.3. Enzyme modification with H2O2 Solutions of 0.7 ␮M apo- and holoenzyme were incubated with several concentrations of H2O2 in 50 mM potassium phosphate buffer, pH 8.0, at 30°C at different times. Then H2O2 was removed from the mixtures by adding 0.2 mg/ml catalase in the assay buffer, and enzyme activity was assayed under standard conditions. In the case of the apoenzyme, before enzyme assay, reconstituted holoenzyme was obtained by addition of FAD. Concentration of H2O2 in stock solutions was checked spectrophotometrically at 240 nm using an extinction coefficient of 43.6 M⫺1 cm⫺1. Protection experiments were performed by pre-incubating 0.7 ␮M of DAAO (both apo- and holoenzyme forms) with 10 mM D-alanine or 10 mM benzoate in 50 mM potassium phosphate buffer at pH 8.0 during 10 min. Inactivation with H2O2 was then carried out. 2.2.4. Product inhibition studies When the product inhibition by H2O2 and pyruvic acid was studied, the rate of the reaction was determined in reaction mixtures containing up to 10 mM D-alanine and up to 10 mM inhibitor. Kinetic analysis of data were performed using SigmaPlot 4.0 (SPSS Inc.). The absence of oxidative decarboxilation by H2O2 of the ␣-oxoacid formed was checked by incubation of 250 ␮M pyruvate solutions with 5 and 50 mM H2O2 at room temperature. The comparison of the spectra of the corresponding 2,4-dinitrophenylhydrazones was made.

236

I. de la Mata et al. / Enzyme and Microbial Technology 27 (2000) 234 –239

2.2.5. Determination of methionine residues Methionine residues were determined in both apo- and holo-D-amino acid oxidase before and after oxidation with hydrogen peroxide using 14C-labeled iodoacetic acid according to the method of Caldwell et al. [5] slightly modified. Solutions of 12.5 ␮M of holo- and apoenzyme were subjected to oxidation with 50 mM H2O2 for 12 h and 20 mM for 30 min, respectively, at 30°C in order to achieve a complete inactivation in both cases. As controls, identical samples were kept in the same conditions substituting H2O2 with H2O. Resulting solutions were passed through PD-10 gel filtration columns, and the isolated protein samples were vacuum-dried. These samples (5.8 ␮M) were dissolved using a solution with 8 M urea and 2.32 mM (7.57 Ci/mol) iodo-[14C]acetic acid adjusted with HCl to pH 3.0. Alkylation was driven at 40°C for 24 h in the dark, and solutions were then filtered through Millipore filters with an exclusion limit of 5000 Da. Filters were then washed three times with water and once with ethanol. After drying, the filters were placed in vials containing 10 ml of Ready Safe scintillation cocktail and counted in a Tricarb (Packard) liquid scintillation counter. 2.2.6. Determination of cysteine residues Cysteine residues were determined in both apo- and holo-D-amino acid oxidase before and after oxidation with hydrogen peroxide using Ellman’s reagent [10]. Solutions of 200 ␮l of 1 ␮M apo- and holoenzyme were oxidized with 20 mM H2O2 for 30 min and 50 mM H2O2 for 12 h, respectively, at 30°C. A reference mixture was prepared for each oxidation reaction in which H2O2 was substituted by H2O. Reactive thiols were then determined by addition of 5 ␮l of a 10 mM DTNB solution in potassium phosphate, pH 8.0, and incubation for 5 min at room temperature. The color developed was stable for at least 1 h. Spectra were recorded in the 300 –520 nm range. 2.2.7. Fluorescence spectra Fluorescence spectra of both apo- and holo-D-amino acid oxidase before and after oxidation with hydrogen peroxide were recorded with a MPF-44E Perkin–Elmer spectrofluorometer. Solutions of 1 ␮M apo- and holoenzyme were oxidized with 20 mM H2O2 for 30 min and 50 mM H2O2 for 12 h, respectively, at 30°C. A reference mixture was prepared for each oxidation reaction in which H2O2 was substituted by H2O. After incubation, H2O2 was removed using columns of semi-dry Sephadex G-25. After excitation at 280 nm, the emission spectra at room temperature were recorded. 2.2.8. Characterization of binding of FAD to oxidized apo-D-amino acid oxidase Kinetic studies of binding of FAD to apo-D-amino acid oxidase oxidized with H2O2 were carried out by recording the FAD fluorescence emission [11]. Upon excitation at 450

nm, the fluorescence emission was recorded at 530 nm. Dissociation constant of FAD-enzyme complex was calculated by fitting the experimental data in Fig. 3 to the following equation [12]: 1/(1 ⫺ a) ⫽ (1/Kd[FAD/a]) ⫺ (1/Kd[DAAO])

(1)

where “a” is the fraction of the total FAD binding sites and Kd is the dissociation constant for the FAD-D-amino acid oxidase complex. UV/VIS absorption and fluorescence spectra were recorded after reconstitution of the oxidized apo-D-amino acid oxidase with FAD. 2.2.9. Gel filtration studies Solutions of 2.37 ␮M of apoenzyme or holoenzyme before and after treatment with H2O2 and also oxidized apoenzyme reconstituted with FAD were applied on an HPLC Biosep-Sec 3000 column in 50 mM potassium phosphate buffer with 2 mM EDTA at pH 7.5.

3. Results In native conditions, D-amino acid oxidase from R. gracilis is a homodimer of 79 kDa, while the monomer has a molecular mass of 39 kDa. The holoenzyme contains one molecule of tightly noncovalently bound FAD per monomer. By, contrast the apoenzyme is entirely present as a monomeric protein. The apoprotein is inactive and it regains its activity when assayed in the presence of exogenous FAD. Reconstituted holoenzyme elutes in gel filtration experiments as a 79 kDa dimer, indicating that a conformational change that promotes dimerization when FAD binds to the monomer occurs. To analyze the effect of hydrogen peroxide on D-amino acid oxidase, holo- and apoenzyme were pre-incubated with 1–50 mM hydrogen peroxide at the indicated times. Fig. 1 shows the time course of inactivation of D-amino acid oxidase holoenzyme in the presence of different concentrations of hydrogen peroxide. The inactivation followed pseudo-first order kinetics with a second-order rate constant of 6.96 ⫻ 10⫺5 mM⫺1 min⫺1 (from the intercept of the replot in Fig. 1). It can be observed that the holoenzyme is rather resistant to the deleterious effect of hydrogen peroxide because the activity was unchanged after 13 h of pre-incubation of the enzyme with 1–5 mM of the hydrogen peroxide. However, the apoenzyme was very sensitive to hydrogen peroxide. Fig. 2 shows also a pseudo-first order kinetics, although the second-order rate constant of inactivation was 3.12 ⫻ 10⫺2 mM⫺1 min⫺1, three orders of magnitude higher than in the preceeding case. In both cases, the order of the reaction was close to 1, indicating that 1 mol of reactant reacts with 1 mol of the enzyme. Hydrogen peroxide is a relatively nonspecific oxidizing agent that reacts with a wide variety of organic compounds.

I. de la Mata et al. / Enzyme and Microbial Technology 27 (2000) 234 –239

Fig. 1. Time-dependent inactivation of holo-D-amino acid oxidase. Solutions of 0.7 ␮M DAAO were pre-incubated with H2O2 in 50 mM potassium phosphate buffer, pH 8, at 30°C. Residual enzyme activity was measured in the presence of 10 mM D-alanine as substrate. Substrate solution contained 0.2 mg/ml catalase to remove H2O2. Inset, log of the apparent first order inactivation rate (Kobs) versus log H2O2 concentration.

Nevertheless, under relatively mild conditions, this reagent is highly specific for a small number of amino acid side chains. Under acidic conditions, the primary reaction is the conversion of methionine residues to the corresponding sulfoxide [13]. Protein-S-CH3 ⫹ H2O2 3 Protein-S-CH3 ⫹ H2O 㛳 O Standard procedure used to hydrolyze proteins (6 N HCl, 110°C, 24 h) reduces the sulfoxide to methionine. However, methionine can be indirectly determined after successive

Fig. 2. Time-dependent inactivation of apo-D-amino acid oxidase. Solutions of 0.7 ␮M apoenzyme were pre-incubated with H2O2, as described in Fig. 1. Before the enzymatic assay, the holoenzyme was reconstituted by adding FAD, and residual activity was measured as described in Fig. 1. Inset, log of the apparent first order inactivation rate (Kobs) versus log H2O2 concentration.

237

treatment of the oxidized protein with iodo[14C]acetate, which alkylates the methionine residues but not the methionine sulfoxide residues. It was possible to demonstrate, that after treatment of D-amino acid oxidase with H2O2, none of the methionine residues in holo- or apoenzyme were modified. Taking into account that other residues as cysteine, tryptophan, or tyrosine may also be attacked by H2O2, experiments were carried out in order to test the possible modification of other residues that could be responsible for enzyme inactivation. Treatment of holoenzyme with H2O2 led the oxidation of two cysteines (each monomer contains six cysteines [14]); the same treatment on the apoenzyme gave one oxidized cysteine. Several researchers have suggested the involvement of a sulfhydryl group of DAAO from pig kidney and R. gracilis in the binding of FAD to the apoenzyme [8,15, 16]. Very recently [17], the presence of a reactive cysteine in the flavin binding domain of R. gracilis DAAO has been shown. This cysteine, identified as Cys 208, was the only residue covalently reacting with a FAD analogue. The flavinylated enzyme was inactive and was unable to dimerize. On the other hand, treatment of the holoenzyme with sulfhydryl reagents [17] resulted in a limited inactivation that was not affected by the presence of benzoate, a competitive inhibitor of DAAO, or exogenous FAD; however, the apoprotein lost all its activity in the presence of the same sulfhydryl reagents. The rapid loss of enzyme activity observed for the apoprotein when treated with H2O2 could be due to the oxidation of one cysteine, being different to that modified in each monomer of the holoenzyme. Nevertheless, oxidized apoenzyme binds FAD. A dissociation constant of 5.26 ⫻ 10⫺9 has been determined for the FAD-oxidized apoenzyme complex. The corresponding dissociation constant for the native FAD-apoenzyme complex was 9.2 ⫻ 10⫺9, indicating that FAD binds both forms of the enzyme with similar affinity. According to the results of Pollegioni et al. [17], mentioned above, some other cysteine residue, if any, different from cysteine 208 might be responsible for the loss of enzyme activity induced by H2O2. Because it has been shown that the oxidation of tryptophan by H2O2 occurs optimally between pH 8 and 10 [18], titration of tryptophans of holo- and apoenzyme before and after treatment with H2O2 has also been carried out. The fluorescence emission spectra for different forms of the enzyme are shown in Fig. 3. There was an important change in the fluorescence emission between the apoenzyme and the H2O2-treated apoenzyme. indicating that H2O2 oxidized tryptophans in the apoenzyme (Fig. 3A). Also, a minor change was observed between the holoenzyme and the corresponding H2O2-treated holoenzyme, indicating that some of the tryptophans oxidized in the apoenzyme were protected in the holoenzyme. On the other hand, the holoenzyme showed a more structured spectrum (Fig. 3B) probably because binding of FAD and dimerization

238

I. de la Mata et al. / Enzyme and Microbial Technology 27 (2000) 234 –239

Fig. 3. Fluorescence emission spectra. A) Apoenzyme: [1] native (gain 3), and [2] H2O2-treated (gain 10). B) Holoenzyme: [1] native (gain 10) and [2] H2O2-treated (gain 10).

bury some indole groups avoiding the contact with the polar surrounding. Contact of the indole nucleus with polar solvents molecules is known to quench the structured fluorescence emission and to broaden the absorption spectrum. The apoenzyme also showed a broadening of its absorption spectrum when compared to that for the holoenzyme (data not shown). Although the H2O2-treated apoenzyme bound FAD, it was unable to dimerize. In gel filtration experiments, on a HPLC Biosep-Sec 3000 column, native holoenzyme eluted as a single peak, corresponding to a molecular mass of 78 kDa, whereas native apoenzyme eluted as a 39 kDa monomer. When holo- and apoenzyme were oxidized with hydrogen peroxide (see Figs. 1 and 2), the peak corresponding to a molecular mass of 78 kDa disappeared, indicating that the oxidation of some amino acid of the protein prevents the dimerization process. On the other hand, the inactivation of holo- and apoenzyme by H2O2 was not affected by the presence of benzoate, a competitive inhibitor of D-amino acid oxidase [9], nor did D-alanine protect the enzyme from the inactivation. This result indicates that the modified residues are not near to the catalytic center. Recently, the kinetic mechanism of DAAO from R. gracilis was elucidated [19] which involves a ternary complex mechanism being the reductive half reaction, the rate-limiting step. To determine the order of product release, product inhibition studies were carried out. Fig. 4 shows the double reciprocal plot 1/v versus 1/S at several fixed concentrations of pyruvic acid. Pyruvic acid competitively inhibited the reaction when D-alanine was the variable substrate, indicating that both of them combine with the same form of the enzyme (E); slope replot (Fig. 4, inset) was linear, indicating full competitive inhibition with a Ki of 1.81 mM. H2O2 was a noncompetitive inhibitor when D-alanine was the variable substrate and O2 was at not saturating level, indicating that H2O2 binds to a form of the enzyme different from that D-alanine binds. The inhibition was noncompetitive (Fig. 5); the family of reciprocal plots intersect on the 1/[S] axis; slope and the 1/v axis intercept are linear (see Fig. 5, inset). The corresponding inhibition constants were:

Fig. 4. Double reciprocal plots of product inhibition by pyruvic acid. Experimental conditions as described Section 2. Enzyme activity was measured using the peroxidase coupled system in assay mixtures containing 0.5–3 mM inhibitor. Inset, Replot of slopes versus pyruvic acid concentration.

Kis ⫽ 0.52 mM and Kii ⫽ 0.70 mM. These results indicate that H2O2 is the first product released and pyruvic acid the second one [20] according to the hypothesis of Pollegioni et al. [19]. 4. Discussion D-amino acid oxidase from R. gracilis was fairly resistant to the harmful effect of hydrogen peroxide. The very different behavior of holo- and apoenzyme toward this reagent reveals the different conformations of both forms of the enzyme [21].

Fig. 5. Double reciprocal plots of product inhibition by H2O2. Experimental conditions as in Fig. 4. Inset, Replot of slopes (ƒ) and intercepts () versus H2O2 concentration.

I. de la Mata et al. / Enzyme and Microbial Technology 27 (2000) 234 –239

Our results show that the enzyme treated with hydrogen peroxide was unable to dimerize. Among all the amino acid residues that are vulnerable to hydrogen peroxide oxidation, only tryptophan residues and a cysteine residue were oxidized in the standard reaction conditions, indicating their possible involvement in the dimerization process. The dimeric aggregation state was mainly connected with the stability of the native enzyme [21]. Trp243, which is conserved in all D-amino acid oxidase sequences so far [22], could be located at the interface domain between both subunits of the protein, according to the three-dimensional structure of the pig kidney enzyme, that should share common features with the yeast one [23]. This tryptophan residue could be essential for dimerization and its structured fluorescence emission (Fig. 3B) be quenched in the monomer due to the contact with the polar surrounding medium. Other tryptophan residues that are well conserved in the D-amino acid oxidase sequence alignment [22] are Trp55 and Trp70 and could be located at the FAD-binding domain [23]. Also, the absence of FAD in the apoprotein monomer can contribute to quench their structured fluorescence emission. As stated in Section 3, one cysteine residue per monomer was oxidized in the presence of H2O2 either in apo- or holoenzyme, thus the involvement of a cysteine residue in the dimerization process cannot be ruled out. Cysteine 259 belongs to a region showing a high degree of conservation in D-amino acid oxidase sequences as does tryptophan 243. This cysteine residue corresponds to cysteine 263 in pig kidney enzyme and, according to the three-dimensional structure of this enzyme, would be located in the ␣I3 helix on the interface domain [23]. Previous studies on the kinetic mechanism of D-amino acid oxidase from R. gracilis [19] pointed to the possibility that the catalytic reaction would involve a ternary mechanism, with hydrogen peroxide and piruvic acid being the first and second products, respectively, released from the enzyme surface. The results of product inhibition studies presented here (hydrogen peroxide behaves as a noncompetitive inhibitor and pyruvic acid as a competitive inhibitor) support the formation of a ternary complex and the proposed order of product release. To conclude, our results provide evidence about the double harmful effect of hydrogen peroxide on the reaction catalyzed by D-amino acid oxidase. It causes a change in the protein due the oxidation of tryptophan and cysteine residues that prevents dimerization and a kinetic effect as a non-competitive inhibitor with Ki values (Kii and Kis) in the same order of magnitude as the Km (1 mM) for the D-alanine as substrate.

References [1] Savidge JA. Conversion of cephalosporins. In: Vandamme EJ, editor. Biotechnology of industrial antibiotics. New York: Dekker, 1984. p. 205–15.

239

[2] Isogai T, Ono H, Ishitani Y, Kojo H, Ueda Y, Kohsaka M. Structure and expression of cDNA for D-amino acid oxidase active against cephalosporin C from Fusarium solani. J Biochem 1990;108:1063–9. [3] Szwajcer Dey E, Flygare S, Mosbach K. Stabilisation of D-amino acid oxidase from yeast Trigonopsis variabilis used for production of glutaryl-7-aminocephalosporanic acid from cephalosporin C. Appl Biochem Biotechnol 1991;24:239 –50. [4] Pilone MS, Pollegioni L, Buto S. Stability and kinetic properties of immobilized Rhodotorula gracilis D-amino acid oxidase. Biotechnol Appl Biochem 1992;16:252– 62. [5] Caldwell P, Luck DC, Weissbach H, Brot N. Oxidation of the methionine residues of Escherichia coli ribosomal protein L12 decreases the protein’s biological activity. Proc Natl Acad Sci USA 1978;75: 5349 –52. [6] Means GE, Feeney RE. Reducing and oxidizing reagents. In: Chemical modification of proteins. San Francisco: Holden–Day Inc., 1971. p. 162–5. [7] Ramo´n F, de la Mata I, Iannacone S, Castillo´n MP, Acebal C. Chemical modification of histidyl residues in D-amino acid oxidase from Rhodotorula gracilis. J Biochem 1995;118:911– 6. [8] Casalin P, Pollegioni L, Curti B, Pilone MS. A study on apoenzyme from Rhodotorula gracilis D-amino acid oxidase. Eur J Biochem 1991;197:513–7. [9] Massey V, Curti B. A new method of preparation of D-amino acid oxidase apoprotein and a conformational change after its combination with flavin adenine dinucleotide. J Biol Chem 1966;214:3417–23. [10] Ellman GL. Tissue sulfhydryl groups. Arch Biochem Biophys 1959; 82:70 –2. [11] Ghisla S, Massey V, Lhoste JM, Mayhew SG. Fluorescence and optical characteristics of reduced flavines and flavoproteins. Biochemistry 1974;13:589 –97. [12] Stinson RA, Holbrook JJ. Equilibrium binding of nicotinamide nucleotides to lactate dehydrogenases. Biochem J 1973;131:719 –28. [13] Newmann NP. Oxidation with hydrogen peroxide. Methods Enzymol 1972;25:393– 400. [14] Pilone MS, Pollegioni L, Casalin P, Curti B, Ronchi S. Properties of D-amino acid oxidase from Rhodotorula gracilis. Eur J Biochem 1989;180:199 –204. [15] Curti B, Ronchi S, Pilone Simonetta M. D-, and L-amino acid oxidases. In: Mu¨ller F, editor. Chemistry and biochemistry of flavoenzymes. Boca Raton: CRC Press, 1992. p. 69 –94. [16] Fonda ML, Anderson BM. Inactivation by maleimides. J Biol Chem 1969;244:666 –74. [17] Pollegioni L, Campaner S, Raibekas AA, Pilone MS. Identification of a reactive cysteine in the flavin-binding domain of Rhodotorula gracilis D-amino acid oxidase. Arch Biochem Biophys 1997;343:1–5. [18] Hachimori Y, Horinishi H, Kurihara K, Shibata K. Sattes of amino acid residues in proteins. Different reactivities with H2O2 of tryptophan residues in lysozyme, proteinases and zymogens. Biochim Biophys Acta 1964;93:346 –52. [19] Pollegioni L, Langkav B, Tischer W, Ghisla S, Pilone MS. Kinetic mechanism of D-amino acid oxidase from Rhodotorula gracilis and Trigonopsis variabilis. J Biol Chem 1993;268:13850 –7. [20] Cleland WW. Determining the chemical mechanisms of enzymecatalyzed reactions by kinetics studies. Adv Enzymol Relat Areas Mol Biol 1977;45:273–387. [21] Pollegioni L, Ceciliani F, Curti B, Ronchi S, Pilone MS. Studies on the structural and functional aspects of Rhodotorula gracilis D-amino acid oxidase by limited trypsinolysis. Biochem J 1995;310:577– 83. [22] Faotto L, Pollegioni L, Ceciliani F, Ronchi S, Pilone MS. The primary structure of D-amino acid oxidase. Biotechnol Lett 1995;17: 193– 8. [23] Mattevi A, Vanoni MA, Todone F, et al. Crystal structure of D-amino acid oxidase: a case of active site mirror-image convergent evolution with flavocytochrome b2. Proc Natl Acad Sci USA 1996;93:7496 – 501.