Arch Microbiol (2002) 178 : 71–74 DOI 10.1007/s00203-002-0428-0

O R I G I N A L PA P E R

Claudia Herles · Annett Braune · Michael Blaut

Purification and characterization of an NADH oxidase from Eubacterium ramulus

Received: 14 December 2001 / Revised: 20 March 2002 / Accepted: 30 March 2002 / Published online: 7 May 2002 © Springer-Verlag 2002

Abstract An NADH oxidase from the strictly anaerobic Eubacterium ramulus was purified to homogeneity. The enzyme is composed of two types of subunits with molecular masses of 40 and 30 kDa. The molecular mass of the native enzyme is 450 kDa according to gel filtration and PAGE analysis. Six to eight mol of FAD were found per mol of native enzyme. The NADH-specific enzyme was inhibited by N-bromosuccinimide and sulfhydryl reagents such as N-ethylmaleimide, CuCl2 or ZnCl2. The physiological function of the purified enzyme is unclear, but the demonstration of NADH-dependent O2-consumption suggests that it plays a role in the scavenging of oxygen.

lish anaerobiosis. During the consumption of oxygen, the growth of C. butyricum ceases but resumes without noticeable cell damage after the oxygen in the medium is completely reduced. Furthermore, NADH oxidases might be involved in the regulation of the NAD/NADH ratio in anaerobic bacteria (Niimura et al. 2000). Here we describe the isolation and characterization of an NADH oxidase from Eubacterium ramulus.

Materials and methods Growth conditions and preparation of cell extract

Keywords NADH oxidase · Eubacterium ramulus

Introduction Although oxygen is toxic for obligate anaerobes, several enzymes catalyzing NADH-dependent reduction of oxygen have been found in anaerobes (Liu and Scopes 1993; Higuchi et al. 1994). NADH oxidases produce either H2O2 or H2O (Nishiyama et al. 2001); however, superoxide may also be formed (Maeda et al. 1992). Oxygen-consuming NADH-dependent flavin oxidoreductases have also been purified from Eubacterium sp. strain VPI 12708 (Franklund et al. 1993) and Eubacterium lentum (Feighner and Hylemon 1980). The physiological function of such enzymes in anaerobic bacteria is not yet clear. Under air, cell extracts of Clostridium butyricum consume oxygen when NADH or NADPH is added (Kawasaki et al. 1998). In this organism, the oxidation of NADH is considered as a defense mechanism to reduce traces of oxygen and reestab-

C. Herles · A. Braune · M. Blaut (✉) Gastrointestinale Mikrobiologie, Deutsches Institut für Ernährungsforschung (DIfE), Arthur-Scheunert-Allee 114–116, 14558 Bergholz-Rehbrücke, Germany e-mail: [email protected], Tel.: +49-33200-88470, Fax: +49-33200-88407

Eubacterium ramulus was cultured anaerobically in ST medium (Kamlage et al. 1999) containing per liter: 9 g tryptically digested peptone from meat, 1 g proteose peptone, 3 g meat extract, 4 g yeast extract, 6 g glucose, 3 g NaCl, 2 g Na2HPO4, 0.5 ml Tween 80, 0.25 g cystine, 0.25 g cysteine, 0.1 g MgSO4·7H2O, 5 mg FeSO4·7H2O and 3.4 mg MnSO4·2H2O at pH 7.0 and 37 °C. Growth was monitored by measuring the optical density (OD) at 600 nm. Cells were harvested under air or under N2/CO2 (80/20 v/v) at the end of the exponential growth phase. For the preparation of cell extracts, cells were suspended in 50 mM potassium phosphate buffer, pH 6.8, at a ratio of 1 g of cells (wet mass) per 2 ml buffer. After addition of DNase I, the suspension was passed twice through a chilled French pressure cell (Aminco, Silver Springs, MD., USA) at 130 MPa. The cytoplasmic fraction was obtained by centrifuging the cell extract at 40,000×g (30 min, 4 °C) and the resulting supernatant subsequently at 110,000×g for 1 h at 4 °C. Purification of NADH oxidase All purification steps were carried out under air. The soluble cell extract of E. ramulus was loaded onto a DEAE-Sephacel column (2.5×15 cm, Amersham Pharmacia Biotech, Freiburg Germany) equilibrated with buffer A (50 mM potassium phosphate, pH 6.8). The column was washed with 100 ml of buffer A at a flow rate of 1 ml min–1. Bound proteins were eluted with a 100-ml linear gradient of 0 to 0.25 M KCl in buffer A followed by 50 ml 0.25 M KCl in buffer A and a 250-ml linear gradient of 0.25 M to 1 M KCl in buffer A. Five-ml fractions were collected. Pooled active DEAE fractions were concentrated using a Centriprep 10 cartridge (Amicon, Witten, Germany). To avoid overloading of the column, only portions of the concentrate were subjected repeatedly to gel filtration on a Superdex 200 column (Amersham Pharmacia Biotech).

72 Protein was eluted with 50 mM potassium phosphate buffer, pH 6.8, containing 50 mM KCl at a flow rate of 0.5 ml min–1. Fractions of 0.4 ml were collected. Active fractions were combined and desalted with 5 mM potassium phosphate buffer, pH 6.8, using a Centricon 10 cartridge (Amicon). Protein was applied to a hydroxyapatite column CHT I (BioRad, Munich, Germany) previously equilibrated with 5 mM potassium phosphate buffer, pH 6.8. The column was washed with 15 ml of the same buffer and protein was eluted with a 60-ml linear gradient of 5 to 500 mM potassium phosphate, pH 6.8, at a flow rate of 2 ml min–1 and collected in 0.5-ml fractions. Pooled fractions were loaded onto a Mono Q column (Amersham Pharmacia Biotech) equilibrated with 50 mM potassium phosphate, pH 6.8. NADH oxidase was eluted with a linear gradient of 0 to 1 M KCl in buffer A at a flow rate of 1 ml min–1. The eluent was collected in 0.5-ml fractions.

Enzyme and protein assays NADH oxidase activity was measured spectrophotometrically under oxic conditions at 35 °C by monitoring the oxidation of NADH at 340 nm (ε340=6.3 mM–1cm–1) with a Cary 1 UV-visible double beam spectrophotometer (Varian, Darmstadt, Germany). The reaction mixture contained the following in a final volume of 1 ml: 50 mM potassium phosphate buffer pH 6.8, 150 µM NADH, 150 µM FAD and an appropriate amount of enzyme. The reaction was initiated by addition of the enzyme preparation. One unit of activity was defined as the amount of enzyme catalyzing the oxidation of 1 µmol NADH per min. Oxygen consumption was measured in the presence of 300 µM NADH and 15 µM FAD in air-saturated potassium phosphate buffer (50 mM, pH 6.8) using an oxygen electrode (Rank Bros., Cambridge, UK). In order to identify the product of NADH oxidation, 60 µg catalase (Amersham Pharmacia Biotech) was added to the reaction mixture. Inhibitory substances (1 mM each) were preincubated with the purified enzyme for 10 min at 4 °C. Alternative electron acceptors (150 µM) were tested in the absence of oxygen. The protein concentration was determined by the method of Bradford (1976) using the BioRad dye reagent (BioRad) and bovine serum albumin as the standard.

Electrophoresis conditions Native and SDS-PAGE were done according to the method of Laemmli (1970). Phosphorylase b (94 kDa), albumin (67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), trypsin inhibitor (20.1 kDa), α-lactalbumin (14.4 kDa) (Amersham Pharmacia Biotech) were used as standards in SDS-PAGE. Proteins were stained with Coomassie brilliant blue 250 (0.2% in methanol/acetic acid/water, 45:45:10, v/v/v). Native gels were also stained for NADH oxidase activity by immersion in standard NADH oxidase assay mixture containing in addition 500 µg of nitro blue tetrazolium ml–1 followed by incubation at 37 °C for 15 min. Bands with NADH oxidase activity appeared purple on a yellow background. Gels were analyzed with a ChemiDoc System using the Quantity One Quantification Software (BioRad).

Determination of molecular mass The molecular mass of native NADH oxidase was estimated by gel filtration and native PAGE. For PAGE, precast gels with a linear gradient of 4 to 20% acrylamide (BioRad) were used. Gel filtration was carried out on a Superdex 200 column (Amersham Pharmacia Biotech) at a flow rate of 0.5 ml min–1 with an eluent buffer consisting of 50 mM potassium phosphate and 50 mM KCl (pH 6.8). Standards (Amersham Pharmacia Biotech) included thyroglobin (669 kDa), ferritin (440 kDa), catalase (232 kDa), lactate dehydrogenase (140 kDa), and albumin (67 kDa).

Cross-linking of protein subunits In order to determine the number of different subunits, a cross-linking experiment was carried out with glutardialdehyde according to Ishiura et al. (1990). Purified NADH oxidase was treated with glutardialdehyde (final concentration 0.05%) at 37 °C for 15 min. After incubation, 1 M glycine (pH 8.0) was added to the reaction mixture (final concentration 0.3 M) to stop the reaction. The protein solution was concentrated with Centricon 10 cartridges (Amicon) and subjected to SDS-PAGE.

Spectroscopy and identification of the flavin component The UV-visible absorption spectrum of purified NADH oxidase was recorded on a Varian Cary 1 UV-visible double beam spectrophotometer. The flavin component was identified by thin-layer chromatography as described by Baron and Hylemon (1995) using riboflavin, FMN, and FAD as standards. The amount of flavin was quantified spectrophotometrically at 450 nm using an extinction coefficient of 11.3 mM–1cm–1.

N-terminal amino-acid sequence determination For N-terminal amino-acid sequence analysis, 3 µg of purified enzyme was subjected to SDS-PAGE, blotted onto a polyvinyl diflouride membrane and stained with Coomassie brilliant blue. The amino-acid sequence analysis was determined at the WITA GmbH (Teltow, Germany) using a Procise Sequencer.

Chemicals FAD, FMN, and riboflavin were from Fluka (Deisenhofen, Germany). Bovine serum albumin and nitro blue tetrazolium were from Sigma (Deisenhofen, Germany). Glutardialdehyde was from Roth (Karlsruhe, Germany). All other chemicals were of high purity from commercial sources.

Results and discussion Enzyme purification NADH oxidase activity was recovered from the cytoplasmic fraction of cell extracts of E. ramulus. The enzyme was purified with a four-step protocol (Fig. 1, Table 1). NADH oxidase was purified to homogeneity with an overall 98-fold purification and a recovery of 17%. Gel filtration with the Superdex 200 column was the most effective purification step, resulting in an 18-fold purification and 76% recovery. Further purification steps including hydroxyapatite chromatography on CHT I and chromatography on Mono Q led to an enzyme preparation that was >95% homogeneous as judged by SDS-PAGE. In native PAGE and gel filtration, the enzyme displayed an apparent molecular mass of 450 kDa and 455 kDa, respectively.

Enzyme composition The enzyme consists of two polypeptides of 40 (α) and 30 (β) kDa (Fig. 1), which was confirmed by protein cross-

73

Fig. 1 SDS-PAGE of NADH oxidase from Eubacterium ramulus at each purification step. Lane 1 Cell extract (15 µg), lane 2 pooled DEAE Sephacel fractions (14 µg), lane 3 pooled Superdex 200 fractions (10 µg), lane 4 pooled fractions of hydroxyapatite CHT I (10 µg), lane 5 NADH oxidase after Mono Q (3 µg), lane M molecular mass standards with 94, 67, 43, 30, 20.1 and 14.4 kDa. Gel was stained with Coomassie-blue R250

linking experiments. Cross-linking of the purified NADH oxidase with glutardialdehyde resulted in a loss of the 40and 30-kDa protein bands and the appearance of a protein band (>175 kDa) between the stacking and resolving gels in SDS-PAGE after protein denaturation. When this crosslinked protein preparation was subjected to native gradient PAGE, a protein band exhibiting the same molecular mass (450 kDa) as the native purified enzyme was detectable. It may therefore be concluded that the NADH oxidase isolated from E. ramulus consists of two different subunits. The subunit composition was examined by quantifying the intensity of the protein bands stained with Coomassie brilliant blue in a SDS-polyacrylamide gel. A ratio of 1.9± 0.2 α-subunit to one β-subunit was found. Assuming a similar staining behavior of the two polypeptides, this analysis suggests an α8β4- rather than an α6β6-arrangement. The N-terminal amino-acid sequences of the α- and β-subunits were AAMEQSLIEQYKRVYVFAQQVPRMGVML and MKIIVEIAQVPDTAGGVVFNPDIRTLYA, respectively. The first 14 amino acids of the β-subunit showed 64% identity to the β-subunit of putative electron transfer flavoproteins of Clostridium acetobutylicum and Clostridium perfringens (Boynton et al. 1996; Shimizu et al. 2002).

Table 1 Purification of NADH oxidase of Eubacterium ramulus

This is the first report on an NADH oxidase from an anaerobic bacterium that is composed of two different subunits. All NADH oxidases isolated so far from anaerobic bacteria possess two or more but identical subunits with molecular masses between 100 and 450 kDa. In Eubacterium sp. strain VPI 12708, a 210-kDa NADH:flavin oxidoreductase composed of three identical subunits of 72 kDa has been identified (Franklund et al. 1993). The NADH oxidase from Clostridium thermohydrosulfuricum, a 450-kDa protein, is made up of six identical 71-kDa subunits (Maeda et al. 1992). The solution containing the purified NADH oxidase of E. ramulus was slightly yellowish-green. The enzyme showed a typical flavin spectrum with peaks at 375 and 455 nm. The flavin component was released after boiling in absolute ethanol for 15 min in the dark and could be identified as FAD by thin-layer chromatography. The NADH oxidase contained 6–8 mol of FAD per mol native enzyme. Based on the proposed α8β4 structure (see above), it is reasonable to assume that each α-subunit (40 kDa) contains one FAD moiety. Catalytic properties of purified NADH oxidase The temperature optimum of NADH oxidation by the enzyme was 37 °C, the pH optimum was 7.0. The enzyme was specific for NADH, exhibiting only 5% of the activity with NADPH. The purified enzyme catalyzed the reduction of O2 in air-saturated buffer at a rate of 1 U mg–1. The addition of catalase to the reaction mixture led to the release of oxygen indicating that the enzyme catalyzed the formation of H2O2. Enzyme activity followed Michaelis-Menten kinetics. The Km of NADH was 38±2 µM and Vmax was 14.0± 1.0 µmol min–1 (mg protein)–1. Compared to the purified NADH oxidases from Eubacterium sp. VPI 12708 (48 U mg–1) and Clostridium thermohydrosulfuricum (28 U mg–1) (Franklund et al. 1993; Maeda et al. 1992), the enzyme of E. ramulus (15 U mg–1) is less active. It also shows a lower affinity for NADH than the NADH oxidase from C. thermohydrosulfuricum with a Km of 19±3 µM (Maeda et al. 1992). The NADH oxidase activity was sensitive to sulfhydryl reagents. The catalytic activity was reduced to 27% with CuCl2 and 25% with N-ethylmaleimide, whereas ZnCl2 inhibited NADH oxidation only slightly (78%). The inhibition of NADH oxidase by CuCl2 and N-ethylmaleimide

Step

Total protein (mg)

Total activity (U)

Specific activity (U/mg)

Purification (-fold)

Yield (%)

Cell extract DEAE-Sephacel Superdex 200 Hydroxyapatite CHT I Mono Q

600 170 8.4 2.4 1.1

90 73 68 31 15

0.15 0.43 8 13 15

1 3 55 84 98

100 81 76 34 17

74

indicates that cysteine is involved in catalysis as previously reported for several NADH oxidases (Maeda et al. 1992; Niimura et al. 1995). N-bromosuccinimide, a reagent known to selectively oxidize tryptophan residues, inhibited NADH oxidase by 90%. Inhibition by this reagent was also reported for the NADH oxidase of Eubacterium sp. strain VPI 12708 (Franklund et al. 1993). Metal-ionchelating agents such as EDTA and o-phenanthroline had no effect on enzyme activity. To obtain insight into the physiological function of this enzyme, several commercially available electron carriers, such as FAD, FMN, potassium hexacyanoferrate (II), methylene blue, and benzyl viologen, were tested under anoxic conditions. Under these conditions, none of the electron carriers except FAD led to an increase in NADH oxidation (170% compared to oxic conditions). This is in contrast to findings by Baron and Hylemon (1995), who showed that the activity of an NADH oxidase from an Eubacterium strain increases in the presence of K4Fe(CN)6 or methylene blue. In summary, the NADH oxidase purified from E. ramulus represents another enzyme from a strictly anaerobic bacterium that catalyzes the reduction of oxygen by NADH. The physiological function of this flavoprotein remains to be elucidated, although the ability of whole cells of E. ramulus (data not shown) and of the purified enzyme to consume O2 may be taken as an indication that the enzyme plays a role in the protection against oxygen. Acknowledgements We are grateful to Andreas Brune, Konstanz, for his support in carrying out the measurements with the oxygen electrode.

References Baron SF, Hylemon PB (1995) Expression of the bile acid-inducible NADH:flavin oxidoreductase gene of Eubacterium sp. VPI 12708 in Escherichia coli. Biochim Biophys Acta 1249: 145–154 Boynton ZL, Bennet GN, Rudolph FB (1996) Cloning, sequencing and expression of clustered genes encoding β-hydroxybutyrylcoenzyme A (CoA) dehydrogenase, crotonase and butyrylCoA dehydrogenase from Clostridium acetobutylicum ATCC 824. J Bacteriol 178:3015–3024 Bradford MM (1976) A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254

Feighner SD, Hylemon PB (1980) Characterization of a corticosteroid 21-dehydroxylase from the intestinal anaerobic bacterium, Eubacterium lentum. J Lipid Res 21:585–593 Franklund CV, Baron SF, Hylemon PB (1993) Characterization of the baiH gene encoding a bile acid-inducible NADH:flavin oxidoreductase from Eubacterium sp. strain VPI 12708. J Bacteriol 175:3002–3012 Higuchi M, Shimada M, Matsumoto J, Yamamoto Y, Rhaman A, Kamio Y (1994) Molecular cloning and sequence analysis of the gene encoding the H2O2-forming NADH oxidase from Streptococcus mutans. Biosci Biotechnol Biochem 58:1603– 1607 Ishiura S, Tsukahara T, Sugita H (1990) Molecular and biochemical properties of the ATP-stimulated multicatalytic proteinase ingensin, from rat liver. Int J Biochem 22:1195–1201 Kamlage B, Hartmann L, Gruhl B, Blaut M (1999) Intestinal microorganisms do not supply associated gnotobiotic rats with conjugated linoleic acid. J Nutr 129:2212–2217 Kawasaki S, Nakagawa T, Nishiyama Y, Benno Y, Uchimura T, Komagata K, Kozaki M, Niimura Y (1998) Effect of oxygen on the growth of Clostridium butyricum (type species of the genus Clostridium), and the distribution of enzymes for oxygen and for active oxygen species in clostridia. J Ferment Bioengin 86:368–372 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 Liu XL, Scopes RK (1993) Cloning, sequencing and expression of the gene encoding NADH oxidase from the extreme anaerobic thermophile Thermoanaerobicum brockii. Biochim Biophys Acta 1174:187–190 Maeda K, Truscott K, Liu XL, Scopes K (1992) A thermostable NADH oxidase from anaerobic extreme thermophiles. Biochem J 284:551–555 Niimura Y, Poole LB, Massey V (1995) Amphibacillus xylanus NADH oxidase and Salmonella typhimurium alkyl-hydroperoxide reductase flavoprotein components show extremely high scavenging for both alkyl hydroperoxide and hydrogen peroxide in the presence of S. typhimurium alkyl-hydroperoxide reductase 22-kDa protein component. J Biol Chem 270:25645– 25650 Niimura Y, Nishiyama Y, Saito D, Tsuji H, Hidaka M, Miyaji T, Watanabe T, Massey V (2000) A hydrogen peroxide-forming NADH oxidase that functions as an alkyl hydroperoxide reductase in Amphibacillus xylanus. J Bacteriol 182:5046–5051 Nishiyama Y, Massey V, Takeda K, Kawasaki S, Sato J, Watanabe T, Niimura Y (2001) Hydrogen peroxide-forming NADH oxidase belonging to the peroxiredoxin oxidoreductase family: existence and physiological role in bacteria. J Bacteriol 182: 5046–5051 Shimizu T, Ohtani K, Hirakawa H, Ohshima K, Yamashita A, Shiba T, Ogasawara N, Hattori M, Kuhara S, Hayashi H (2002) Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc Natl Acad Sci USA 99:996–1001