Role of Acinetobacter calcoaceticus 3,4-dihydrocoumarin hydrolase in oxidative stress defence against peroxoacids Kohsuke Honda, Michihiko Kataoka, Eiji Sakuradani and Sakayu Shimizu Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Japan The physiological role of a bifunctional enzyme, 3,4-dihydrocoumarin hydrolase (DCH), which is capable of both hydrolysis of ester bonds and organic acid-assisted bromination of organic compounds, was investigated. Purified DCH from Acinetobacter calcoaceticus F46 cata- lysed dose- and time-dependent degradation of peracetic acid. The gene (dch) was cloned from the chromosomal DNA of the bacterium. The dch ORF was 831 bp long, corresponding to a protein of 272 amino acid residues, and the deduced amino acid sequence showed high similarity to those of bacterial serine esterases and perhydrolases. The dch gene was disrupted by homologous recombination on the A. calcoaceticus genome. The dch disruptant strain was more sensitive to growth inhibition by peracetic acid than the parent strain. On the other hand, the recombinant Escheri- chia coli cells expressing dch were more resistant to peracetic acid. A putative catalase gene was found immediately downstream of dch, and Northern blot hybridization ana- lysis revealed that they are transcribed as part of a polycis- tronic mRNA. These results suggested that in vivo DCH detoxifies peroxoacids in conjunction with the catalase, i.e. peroxoacids are first hydrolysed to the corresponding acids and hydrogen peroxide by DCH, and then the resulting hydrogen peroxide is degraded by the catalase. Keywords: Acinetobacter calcoaceticus; catalase; 3,4-dihydro- coumarin hydrolase; perhydrolase; peroxoacid. In a previous study, we found and isolated a novel lactonohydrolase, 3,4-dihydrocoumarin hydrolase (DCH), from Acinetobacter calcoaceticus F46 [1]. The amino acid sequences of the N-terminal and internal peptide of DCH exhibited high similarity to those of several serine-hydro- lases and perhydrolases (nonhaem haloperoxidases). Fur- thermore, the enzyme also showed both hydrolysis activity toward aromatic lactones, such as 3,4-dihydrocoumarin (Scheme 1), and organic acid-assisted bromination activity toward monochlorodimedon (2-chloro-5,5-dimethyl-1,3- cyclohexanedione). Perhydrolases were originally identified as the enzymes catalysing halogenation reactions of various organic com- pounds, because these enzymes catalyse the halogenation of various organic compounds in the presence of hydrogen peroxide, halide ions, and organic acids. Various perhydro- lases have been isolated, mainly from Pseudomonas and Streptomyces species [2–8]. Several of these species are known to produce halogenated metabolites. Perhydrolases had been, as a result of this, thought to be involved in the synthesis of halogenated metabolites in vivo, and had been called Ônonhaem haloperoxidasesÕ, to distinguish them from the ÔtrueÕ haloperoxidases, which are haem- or vanadium- dependent enzymes [9]. However, there is now evidence that perhydrolases do not participate in the biosynthesis of halogenated compounds. A perhydrolase-deficient mutant of Pseudomonas fluorsecens yielded a chlorinated metabo- lite, pyrrolnitrin, like the parent strain [4], and the Strep- tomyces enzyme, which chlorinates tetracycline, is not related to the perhydrolase [10]. X-ray crystallographic analysis of the perhydrolase of Streptomyces aureofaciens revealed an a/b-hydrolase fold, with a Ser-Asp-His catalytic triad [11] that is conserved in the serine-hydrolase family. Interestingly, this catalytic triad is also highly conserved in the amino acid sequences of perhydrolase [12,13]. Based on recent results, a reaction mechanism has been proposed for the halogenation reaction catalysed by a perhydrolase with peroxoacid and hypohalous acid as reaction intermediates. The enzyme catalyses the peroxidation of the organic acid to O O H 2 O OH COOH 3 ,4-Dihydrocoumarin 3-(2-Hydroxyphenyl)propionic aci d Scheme 1. Correspondence to M. Kataoka, Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606–8502, Japan. Fax/Tel.: +81 75 7536462, E-mail: kataoka@kais.kyoto-u.ac.jp Abbreviations: DCH, 3,4-dihydrocoumarin hydrolase; dch,gene encoding 3,4-dihydrocoumarin hydrolase; cat, gene encoding putative catalase; IPTG, isopropyl thio-b- D -galactoside; LB, Luria–Bertani. Enzymes: 3,4-dihydrocoumarin hydrolase (EC 3.1.1.35); catalase (EC1.11.1.6); chloroperoxidase (EC1.11.1.10). Note: The nucleotide sequences reported here have been submitted to DDBJ/EMBL/GenBank databases under accession number AB092339. Note: a web site is available at http://www.hakko.kais.kyoto-u.ac.jp/ lab-e/index-e.html (Received 30 September 2002, revised 22 November 2002, accepted 28 November 2002) Eur. J. Biochem. 270, 486–494 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03407.x peroxoacid in the presence of H 2 O 2 through a reaction mechanism that closely resembles transesterification. Sub- sequently, the halide ion is oxidized by the peroxoacid to hypohalous acid, and then nonenzymatic halogenation of organic compounds such as monochlorodimedon by the resulting hypohalous acid takes place [1,14–16] (Fig. 1). Actually, DCH also requires organic acids, such as formic acid, acetic acid, propionic acid and n-butyric acid, to catalyse the halogenation reaction [1]. In addition, it has been shown that several lipases catalyse the synthesis of peroxoacids [17]. These results suggest that the halogenation reactions catalysed by perhydrolase do not reflect the physiological roles of these enzymes. However, it seems improbable that perhydrolase in vivo catalyses the formation of peroxoacids, which are very toxic for cells. So, we predicted that these enzymes catalyse the reverse reaction, i.e. hydrolytic degradation of peroxoacids, in vivo.Inorder to confirm this assumption, we carried out functional analysis of a bifunctional hydrolase-perhydrolase, with DCH as a model enzyme. Here we describe the cloning of the DCH gene (dch)fromA. calcoaceticus F46, a gene disruption experiment and the physiological role of the enzyme. Materials and methods Chemicals and enzymes A stock solution of peracetic acid of 32% (w/v) concentra- tion in equilibrium with < 6% hydrogen peroxide and 40–45% acetic acid (Aldrich Chemical Co.) was used. Restriction enzymes, alkaline phosphatase (calf intestine), T4 DNA ligase and Ex Taq DNA polymerase were from Takara Shuzo Co. (Kyoto, Japan). All other reagents were commercially available and of analytical grade. Strains, plasmids and growth media A. calcoaceticus F46 was previously isolated from a soil sample [1]. Escherichia coli JM109 was used for gene cloning and expression. Microorganisms were cultured on Luria– Bertani (LB) medium [18] at 28 °CforA. calcoaceticus,and 37 °CforE. coli. In selective media, ampicillin and kana- mycin were used at 100 lgÆmL )1 and 50 lgÆmL )1 , respect- ively. pT7Blue (Novagen) was used for subcloning of the PCR products. pBluescript II SK(+) (Toyobo, Osaka, Japan) was used as a cloning vector. pKK223-3 (Amersham Bioscience) was used as expression vector, and pKT231 [19] was used for construction of the dch disruptant. Enzyme assay, SDS/PAGE and protein determination Acinetobacter strains and the recombinant E. coli strains were cultured in 3 mL LB medium. For cultivation of the recombinant E. coli strains, isopropyl thio-b- D -galactoside (IPTG) was added to a final concentration of 2 m M .After cultivation for 12 h, cells were harvested by centrifugation, resuspended in 1 mL 20 m M potassium phosphate buffer pH 7.0, and disrupted by sonication. The supernatant obtained on centrifugation (14 000 g,4°C, 20 min) was used as the enzyme solution. 3,4-Dihydrocoumarin-hydrolysing activity and mono- chlorodimedon-brominating activity were determined as described previously [1]. Catalase activity was determined according to the method of Hildebraunt and Roots [20]. Peracetic acid-hydrolysing activity was measured as des- cribed below. SDS/PAGE was performed in a 12.5% polyacrylamide slab gel using the Tris/glycine buffer system described by King and Laemmli [21]. Protein concentrations were determined as described by Bradford [22] using BSA as standard. Determination of peracetic acid-hydrolysing activity of purified DCH DCH was purified from A. calcoaceticus F46 as described previously [1]. The assay mixture for peracetic acid-hydro- lysing activity comprised 100 m M sodium citrate buffer pH 5.5, 0.01% (w/v; 1.31 m M ) peracetic acid and enzyme in a final volume of 500 lL. Aftera 10 min incubation at 30 °C, the reaction was stopped by adding 500 lL1 M HCl. The amount of peracetic acid remaining in the reaction mixture was determined as monochlorodimedon-brominating ability [16]. One-hundred lL of the reaction mixture was added to a detection mixture comprising 100 m M KBr and 50 l M monochlorodimedon in a final volume of 2.5 mL. After 10 min at room temperature, the absorbance at 290 nm was measured with a Shimadzu UV-240 spectrophotometer (Kyoto, Japan). Blanks without the enzyme and without peracetic acid were included if necessary. A molar extinction coefficient of 19 900Æ M )1 Æcm )1 for monochlorodimedon was used for calculation of enzyme activity. One unit of enzyme was defined as the amount of enzyme catalysing the consumption of 1 lmol peracetic acid per min at 30 °C. Enzyme Ser OOC-R Enzyme Ser OH R-COOHR-COOOH H 2 O 2 H + , Br - HBrO H 2 O H 2 O Cl HH 3 C H 3 C O O Cl Br H 3 C H 3 C O O Monochlorodimedon Monochloromonobromodimedon ii iii iv i Fig. 1. Predicted reaction mechanism of bromination reactions with bacterial nonhaem haloperoxidases. (i) Nucleophilic attack of an active serine residue on the carboxyl carbon atom of the organic acid and formation of an acyl-enzyme intermediate; (ii) hydrolytic cleavage of the acyl-enzyme on nucleophilic attack by hydrogen peroxide; (iii) nonenzymatic formation of hypohalous acid from the peroxoacid and a halide ion; (iv) nonenzymatic halogenation of organic com- pounds such as monochlorodimedon by the hypohalous acid. Ó FEBS 2003 Role of 3,4-dihydrocoumarin hydrolase of Acinetobacter (Eur. J. Biochem. 270) 487 Oxidative stress sensitivity assays To test the susceptibility of A. calcoaceticus strains and the recombinant E. coli strains to oxidative stress agents, disk inhibition assays were performed. Cells grown overnight in 2 mL LB medium were mixed with 10 mL LB medium containing 1% agar at 42 °C and then poured onto an LB agar plate (135 · 95 mm). For the recombinant E. coli strains, IPTG was added to a final concentration of 2 m M . Sterilized filter disks containing 10 lL peracetic acid, H 2 O 2 , or acetic acid at an appropriate concentration, were placed on the agar plates. The plates were incubated overnight at 28 °CforA. calcoaceticus strains or at 37 °CforE. coli strains, and then the halos of growth inhibition were measured. DNA isolation Total DNA of A. calcoaceticus waspreparedasfollows. A. calcoaceticus was cultivated overnight in 500 mL LB medium. Cells were collected by centrifugation and resus- pended in 12 mL buffer comprising 25% sucrose, 50 m M Tris/HCl pH 8.0, 50 m M EDTA. Lysozyme (Seikagaku Co., Tokyo, Japan) and proteinase K (Wako Pure Chemi- cals, Osaka, Japan) were added to the mixture at final concentrations of 2 mgÆmL )1 and 1.25 mgÆmL )1 , respect- ively. After incubation at 37 °C for 30 min, an equal volume of 2% SDS was added for cell lysis. DNA was purified from the lysate by phenol/chloroform (1 : 1, v/v) extraction, precipitation by isopropanol, RNase-treatment, and then re-precipitation with ethanol. The resultant precipitate, i.e. total DNA, was dissolved in TE buffer consisting of 10 m M Tris/HCl pH 8.0, 1 m M EDTA. Cloning of dch from the genome of A. calcoaceticus F46 To amplify the fragment of DNA encoding DCH from the total genomic DNA of A. calcoaceticus F46byPCR,two highly degenerate primers were designed based on the amino acid sequences of the N-terminal (VDIFYKDW) and internal peptide (EDDQVVPFE) of the purified DCH [1]. The primers used were a sense primer, GTIGA (C/T)AT(A/T/C)TT(C/T)TA(C/T)AA(A/G)GA(C/T)TGG, and an antisense primer, TC(A/G)AAIGGIACIAC(C/ T)TG(A/G)TC(A/G)TC(C/T)TC, where ÔIÕ denotes inosine. The reaction mixture comprised 1 lg total DNA, 250 pmol each primers, 8 nmol each dNTP and 1.5 U Ex Taq polymerase in a final volume of 50 lL. The reaction mixture was run on a thermal cycler (T Gradient; Biometra, Go ¨ ttingen, Germany) using a program of 1 min at 94 °C, 1minat55°C, and 1 min at 72 °C for 30 cycles. The PCR product was approximately 0.7 kb in length, and was used as a probe. Southern blotting and colony hybridization were performed essentially as described by Sambrook et al.[18] using an AlkPhos Direct system (Amersham Bioscience). The probe hybridized to a 5.7-kb XbaIfragmentoftotal DNA from A. calcoaceticus F46. The enriched gene library was prepared by separating XbaI-digested total DNA on an agarose gel, ligating the isolated DNA in the range of 5.7 kb into pBluescript SK II (+), and then transforming the ligation mixture into E. coli JM109. Colony hybridization led to the identification of a single recombinant clone harbouring the plasmid, which contained the 5.7 kb XbaI fragment, and the plasmid was named pADH10. A restriction map of pADH10 is shown in Fig. 2. Nucleotide sequencing Nucleotide sequencing was performed by the dideoxy chain termination method with a CEQ DTCS kit and a CEQ 2000XL DNA analysis system (Beckman Coulter). The sequencing reaction was performed as described in the instruction manual. Expression of dch in E. coli The dch ORF was amplified by PCR using pADH10 and the following two oligonucleotides as a template and as primers, respectively. The sense primer (CAGGAGA ATTCAAAATGGG) contained an EcoRI recognition site (in italics), and the antisense primer (GCCTAAGCTTAAA ACTTAAC) contained a HindIII recognition site (in italics). The conditions for PCR were the same as those described above. The PCR-amplified products were ligated into pT7Blue (designated pDCH1), and sequenced to verify that they had correctly encoded DCH. A plasmid, designated pDCH21, was constructed by ligation of the 860-bp fragment (derived from pDCH1 on EcoRI–HindIII diges- tion) with pKK223-3, and was then transformed into E. coli JM109. Northern hybridization A. calcoaceticus F46 was grown in LB medium to the early exponential growth phase. After cells had been harvested by centrifugation, total RNA was obtained by the acid/ guanidinium thiocyanate/phenol/chloroform method using ISOGEN (Nippon Gene Co., Tokyo, Japan). Total RNA was electrophoresed on a 1% agarose gel in 20 m M Mops buffer containing 1 m M EDTA and 2.2 M formaldehyde, and then transferred to a Hybond-N membrane (Amer- sham Bioscience) in 20 · NaCl/Cit (1 · NaCl/Cit is 0.15 M NaCl, 15 m M sodium citrate). Internal fragments of dch (derived from pDCH21 on EcoRI–HindIII digestion) and a dch cat unknown protein putative transposase 1 kb XbaI XbaI BglII EcoT22I Eco T22I 5.7 kb Fig. 2. Proposed gene organization and restriction map of the cloned region from A. calcoaceticus F46 bearing dch. Proteins encoded by ORFs other than that of dch are based on their similarity to other proteins in the protein sequence database (see text for details). Arrows indicate the direction and extent of sequence determination. 488 K. Honda et al. (Eur. J. Biochem. 270) Ó FEBS 2003 putative catalase gene (derived from pADH10 on EcoT22I digestion, Fig. 2) were used as specific probes. Hybridiza- tion was carried out using the AlkPhos Direct system according to the manufacturer’s protocol. rRNA was used as a standard for loaded total RNA and was visualized by ethidium bromide staining. Construction of a dch disruptant The construction of a gene replacement vector is illustrated in Fig. 3A. A 913-bp PCR product containing a C-terminal truncated DHase gene was generated by PCR with primers dch-I (GAGCTGAATTCTTTGGTCA) containing an EcoRI site (in italics), and dch-II (ATGTAAGATCTGA ACTGGACTTCG) containing a BglII site (in italics), cloned into pT7Blue, excised with EcoRI and BglII, and then replaced at the EcoRI and BglII sites of pADH10. The resulting plasmid, named pDHEB, contains the gene encoding DCH lacking the C-terminal region, in which Asp and His residues involved in the catalytic triad are contained. The kanamycin resistance gene (Km R ), including its own promoter, was amplified by PCR using pKT231 as a template. The sequences flanking the BamHI site were as follows: sense primer, GGATCCGGACCAGTTGGT GATTTT, and antisense primer, GGATCCTTAGAAAA ACTCATCGAGC (BamHI recognition are shown in italics). The amplified Km R was excised with BamHI and then inserted into the BglII site of pDHEB (designated pDHKm R ). The resulting dch disruption plasmid, pDHKm R , was linearized and then introduced into A. cal- coaceticus F46 by electroporation. Transformation method A. calcoaceticus F46 was transformed by electropolation using a Gene Pulser II (Bio-Rad Laboratories) under the following conditions: 0.1-cm cuvette, 400 W,25lF, and 16 kVÆcm )1 . Competent cells for electroporation were prepared by the method of Leahy [23]. After the pulse, 1.5 mL of 30 °C-prewarmed LB was added immediately, followed by incubation at 30 °C for 4 h and then plating. E. coli was transformed by the method of Hanahan [24]. Results Peracetic acid hydrolysis catalysed by DCH Titration with sodium thiosulfate is commonly used to measure peracetic acid concentration. For exact determia- tion, the concentration of peracetic acid needs to be increased and too high a concentration of peroxoacid inhibits enzyme activity. Therefore we developed another method for the determination of small amounts of peracetic acid (see Materials and methods). As described in the Introduction, monochlorodimedon is converted to mono- bromomonochlorodimedon in the presence of peracetic acid and bromine. Based on this principle, the amount of peracetic acid could be determined by measuring spectro- photometrically the decrease in monochlorodimedon. The standard assay with a peracetic acid solution of known concentration, obtained by dilution of 32% peracetic acid, revealed that the concentration could be determined exactly up to 0.1% (13.1 m M ) (data not shown). Under the standard assay conditions, the purified DCH catalysed the degradation of peracetic acid in a dose- and time-dependent manner (Fig. 4). In contrast, the cell-free extractofthedchD Acinetobacter strain did not show peracetic acid-degradation activity (Table 1). These results confirmed that DCH catalysed the degradation of peracetic acid. Lineweaver–Burk treatment of the data yielded an apparent K m for peracetic acid of 0.390 m M at pH 5.0. This value was lower than those for other substrates, such as 3,4-dihydrocoumarin (0.806 m M ,pH7.0)andDL-b-acetyl- thioisobutyrate methyl ester (25.9 m M ,pH7.0)[1].Onthe other hand, the V max for peracetic acid was 12 600 UÆmg )1 , which was much higher than those for other substrates. Nucleotide and deduced amino acid sequences The nucleotide sequence of the 5.7 kb XbaIregionof pADH10 was determined (Fig. 2). An ORF, which exten- ded from position 231 to 1061 and encoded 276 amino acid residues, was confirmed to be dch by the N-terminal and internal amino acid sequences of the purified DCH [1]. The Fig. 3. Gene disruption of dch in A. calco- aceticu s F46. (A) Construction of gene dis- ruption vector pDHKm R . ÔDÕ and ÔHÕ within dch of pADH10 represent Asp and His resi- dues that are involved in the catalytic triad. The arrows represent the primers used for PCR to amplify the catalytic triad-defective dch. (B) Genomic Southern analysis of XbaI- digested total DNAs (10 lg each) from the dchD strain (lane 1) and wild-type strain (lane 2), with the thermostable alkaline phos- phatase-labeled dch fragment as a probe. Ó FEBS 2003 Role of 3,4-dihydrocoumarin hydrolase of Acinetobacter (Eur. J. Biochem. 270) 489 deduced amino acid sequence of DCH exhibited high similarity to those of the esterases of Pseudomonas putida MR-2068 (65.1% identity in overall amino acid sequence) [25], P. fluorescens SIK WI (43.3%) [12,26], and P. fluores- cens DSM 50106 (22.9%) [27], and the nonhaem haloper- oxidases of P. putida (76.7%) [28], P. pyrrocinia (67.0%) [29], S. lividans (65.1%) [7], Rhodococcus erythropolis (64.6%) [8], and S. aureofaciens (46.7%) [30,31] (Fig. 5). The active amino acid residues of serine-hydrolases, the so-called catalytic triad, consisting of the consensus motif, Gly-X-Ser-X-Gly, and Asp and His residues, were highly conserved among these proteins. Downstream of dch,an additional three ORFs were identified (Fig. 2). The second ORF extending from position 1139 to 2659, encoded a polypeptide of 506 amino acids with a calculated M r of 55 909 Da. The deduced amino acid sequence of this protein exhibited significant similarity to those of the catalases of P. fluorescens (84% identity in overall amino acid sequence) [32], Haemophilus influenzae (84%) [33], etc. The third ORF (from position 2728 to 3288) encoded a deduced protein of 186 amino acids with a calculated M r of 20 922 Da, which showed no significant similarity to any other proteins in the protein sequence database. The last ORF (from position 4390 to 5262) encoded a protein of 290 amino acids (M r ¼ 33 168 Da) and exhibited high similar- ity to the transposases of Methylomonas aminofaciens (98% identity) [34], E. coli (98%) [35], etc. Expression of dch in E. coli A cell-free extract of E. coli JM109 transformed with pDCH21 exhibited specific 3,4-dihydrocoumarin-hydro- lysing activity of 116 UÆmg )1 and monochlorodimedon- brominating activity of 14.3 UÆmg )1 ,i.e. 30-fold increases in comparison with those of A. calcoaceticus F46 (See Table 1). E. coli JM109 and E. coli JM109 bearing pKK223-3 did not have enzyme activity. These results confirm that DCH catalyses both the hydrolysis and halogenation reactions. SDS/PAGE of a cell-free extract of E. coli bearing pDCH21 gave a thick band corresponding to DCH and revealed that the level of DCH produced by the transformant was very high (Fig. 6). Gene disruption of dch in A. calcoaceticus F46 To determine the physiological role of DCH in A. calco- aceticus F46, the chromosomal dch gene was destroyed by homologous recombination with disruption vector pDHKm R as described in Materials and methods. The transformant was selected for the kanamycin resistance phenotype. The dch gene disruption was confirmed by Southern analysis (Fig. 3B). The chromosomal DNAs of both the wild-type strain and the dchD strain were digested with XbaI. The signals of the wild-type strain and dchD strain corresponded to sizes of 5.7 kb and 6.3 kb, respect- ively. This difference is expected to result from insertion of a single copy of the Km R gene. 3,4-Dihydrocoumarin-hydrolysing activity, monochloro- dimedon-brominating activity, and peracetic acid-degrading activity were compared between the wild-type strain and Fig. 4. Dose- and time-dependent decomposition of peracetic acid catalysedbyDCH.(A) Dose-dependent decomposition of peracetic acid. Purified DCH (6.67, 13.3 and 26.7 ng) was added to 500 lLof the standard reaction mixture. The enzyme reactions were carried out for 5 min at 30 °C. (B) Time-dependent decomposition of peracetic acid. The enzyme reactions were carried out at 30 °C, for 5, 10, 15, 20 or 30 min, with 3.33 ng purified DCH. After the reactions had been terminated by the addition of 500 lL1 M HCl, 100 lLofthereaction mixture was transferred to the detection mixture to give a final volume of 2500 lL. The rates of peracetic acid-degradation were determined by monitoring the decrease in A 290 , and calculated with a molar absorption coefficient of 19 900 M )1 Æcm )1 for monochlorodimedon. Table 1. Comparison of the enzyme activities of the A. calcoaceticus dchD strain and parent strain F46. Preparation of cell-free extracts and determination of each enzyme activity were performed as described in the text. Enzyme activity (UÆmg )1 ) F46 dchD 3,4-Dihydrocoumarin-hydrolysing activity 4.00 0 Monochlorodimedon-brominating activity 0.443 0 Peracetic acid-hydrolysing activity 27.3 0 Catalase activity 105 118 490 K. Honda et al. (Eur. J. Biochem. 270) Ó FEBS 2003 dchD strain (Table 1). A cell-free extract of the dchD strain did not show any enzyme activity. On the other hand, catalase activity of the cell-free extract of the dchD strain was almost same as that of the wild-type (Table 1), and it was assumed that the expression level of the putative catalase gene (cat), which was encoded immediately downstream of dch, was not affected by dch disruption. Northern analysis To determine the transcription unit of dch,totalRNA extracted from early exponential growth phase cells of A. calcoaceticus F46 was probed with specific internal DNA fragments derived from either dch or the cat gene located downstream of dch. Transcripts of  0.8 and  2.6 kb that hybridized to the dch probe were observed, suggesting that dch was transcribed as both a monocistronic mRNA and part of a polycistronic mRNA (Fig. 7). In addition, the presence of mRNA components of  1.5 and  2.6 kb that hybridized to cat indicated that cat was also transcribed as both a single cistron and part of polycistronic mRNA. These finding that dch and cat were transcribed as parts of the same polycistronic mRNA indicated that both of them were regulated by the same promoter. DCH GYVTTKDG VDIFYKDWGPRDAPV 23 EST-P SYVTTKDG VQIFYKDWGPRDAPV 23 EST-F S TFVAKDG TQIYFKDWGS GKP 21 EST-F1 MAVQWL I AAGVLVGAS VVFWGL SAWMTRR I EAAVP GNGR FVEVDGER F HYYE EGK - - GP P 58 BPO-EST MSYVTTKDG VQIFYKDWGPRDAPV 24 CPO-P PYVTTKDN VEIFYKDWGPKDAQ P 23 CPO-L GTVTTSDG TNIFYKDWGPRDGLP 23 HPO-R PFVTASDG TEIFYKDWGS GRP 21 BPO-A1 PICTTRDG VEIFYKDWGQ GRP 21 BPO-A2 P F I TVGQEN S T S I DLYYEDHGT - - GQP 25 DCH I F F HHGWPL S SDDWDAQML F FLKEG FRVVAHDRRGHGR STQVW- DGHDMDHYADDVAAVV 82 EST-P I HFHHGWPL SADDWDAQML F FLAHGYRVVAHDRRGHGR S S QVW- DGHDMDHYADDVAAVV 82 EST-F VL F SHGWLLDADMWEYQMEYL S SRGYR T I AFDRRGFGR S DQPW- TGNDYDT FADD I AQL I 80 EST-F1 LVMI HGLMGS SRNL TYAL SRQLR EHF RV I T LDR PG SGYS TRHKGTAADL PAQARQVAA F I 118 BPO-EST I H F H H G W P L S A D D W D A Q M L F F L G Q G F R V F A H D R R GHGR S SQV S - DGHDMDHYADDVAAVV 83 CPO-P I VFHHGWPL SGDDWDAQML F FVQKGYRV I AHDRRGHGR SAQV S - DGHDMDHYAADAFAVV 82 CPO-L VVF HHGWPL SADDWDNQML F F L SHGYRV I AHDRRGHGR SDQ P S - TGHDMDTYAADVAAL T 82 HPO-R I MFHHGWPL S SDDWDSQL L FLVQRGYRV I AHDRRGHGR SAQVG - HGHDMDHYAADAAAVV 80 BPO-A1 VVF I HGWPLNGDAWQDQLKAVVDAGYRG I AHDRRGHGH S T PVW- DGYD FDT FADDLNDL L 80 BPO-A2 VVL I HGF PL SGH SWERQ SAAL LDAGYRV I TYDRRGFGQ S SQP T - TGYDYDT FAADLNTVL 84 DCH EYLGVQGAVHVGH S TGGGEVAYYVARY - - PND PVAKAVL I SAV P PLMVKT ESN PDG - L PK 139 EST-P AHLG I QGAVHVGHS TGGGEVVRYMARH - - PADKVAKAVL I AAVP PLMVQT PDN PGG - LPK 139 EST-F EHLDLKEVTLVGFS MGGGDVARY I ARHG - - SARVAGLVL LGAVT PL FGQKPDYPQG - VP L 137 EST-F1 NQLGLDKPLVLGHS LGGAISLALALDH PEA-VSGLVLVAPLT HPQPRLPL 167 BPO-EST E H L G T QGAVHVGHS TGGGEVVRYMARY - - PNDKVAKGVL I AAVP PLMVQT PGNPGG - LPK 140 CPO-P EALDLRNAVH I GHS TGGGEVARYVANDGQ PAGRVAKAVLVSAV P PLMLKT E SNP EG - LP I 141 CPO-L EALDLRGAVH I GHS TGGGEVARYVARAE - - PGRVAKAVLVS AVP PVMVKSDTNP DG - L P L 139 HPO-R AHLGLRDVVHVGH S TGGGEVARYVARHG - - AGRVAKAVL I GAVP PLMVQTE SN P EG - L PV 137 BPO-A1 TDLDLRDVTLVAHS MGGGELARYVGRHG - - TGR LR S AVL L SA I P PVMI KSDKN PDG - V PD 137 BPO-A2 ET LDLQDAVLVGF S MGTGEVARYVS SYG - - TAR I AKVA F LAS LE P F LLKTDDNPDGAAP Q 142 DCH EVFDDLQNQL FKNR SQ FYHDVP AGP FYG FNR P - GAKVS EPVVLNWWR 191 EST-P SVFDG FQAQVAS NRAQFYRDV PAGP FYGYNR P - GVDA S EGIIGNWWR 191 EST-F DVFARFKTELLKDRAQF I SDFNA- PFYGINK- -GQVVS QGVQTQT LQ 187 EST-F1 VFWS LAVRPAWLRRFVANTLTVPMGL-LTRRSVVKGVF APDAA P EDF ATRGGGL 220 BPO-EST S V F D D F QVQVATNRAQ FYRDVP SG P FYGYNR P - GAKS S E G V I G N W W R 192 CPO-P EVFDG FRKALADNRAQF FLDVP TGP FYG FNRA - GATVH QGVIRNWWR 193 CPO-L EVFDE FRAALAANRAQ FY I DVP SGP FYG FNRE - GATV S QGLIDHWWL 191 HPO-R EVFDGFREAVVTNRSQFYLDLASGPFYGFNRP -GADI S QGVIQNWWR 189 BPO-A1 EVFDALKNGVL TER SQFWKDTAEG F F - SANR P - GNKVT QGNKDAFWY 188 BPO-A2 E F FDG I VAAVKADRYA FYTGF FND - FYNLDENLGTR I S EEAVRNSWN 194 DCH QGMMGGAKAHYDG I VAF SQTD F T E AL - - - KK I EV PVL I LHGEDD QVVP F E I S GKKSAE LV 242 EST-P QGMI GSAKAHYDG I VAF SQTDF TEDL - - - KG I TQP VLVMHGDDD Q I VPYENSGLLSAKLL 242 EST-F I AL LA S LKATVDCVTAFAE TDF R PDM- - - AK I DVP T LV I HGDGD Q I VP FE TTGKVAAE L I 238 EST-F1 LGMRPDN FYAA S SE I ALVNDCL PGMVKRY PQLAL P I GL I YGAQD KVLDF RRHGQALADKV 280 BPO-EST Q G M I G S AKAHYDGVVAF SQTDF TEDL - - - KK I QQPVLVMHGDDD QI VPYENSGPLSAKLL 243 CPO-P QGMEGS AKAHYDG I KAF S E TDQT EDL - - - KS I TV P T LVL HGEDD Q I VP I ADAALKS I KLL 244 CPO-L QGMMGAANAHYEC I AAF S E TDF TDDL - - - KR I DV PVLVAHGTDD QVVPYADAA PKSAEL L 242 HPO-R QGMTGS AQAHYEG I KAF SE TDF TDDL - - - RA I DVP T L I MHGDDD QI VP I ANSAETAVTLV 240 BPO-A1 MAMAQT I EGGVRCVDAFGYTDF TEDL - - - KKFD I PT LVVHGDDD QVVP I DATGRKSAQ I I 239 BPO-A2 TAASGG F FAAAAAP T TW- YTD FRAD I PRIDVPALILHGTGD RT LP I ENTARV FHKAL 244 DCH KNGKL I SYPG F PH G- -MPTTEAETINKDLLAF I RS 275 EST-P PNGT LKTYQGYP H G - - MP TTHADV I NADL LAF I R S 275 EST-F KGAE LKVYKDAPH G - - FAVTHAQQLNEDL LA F LKR 271 EST-F1 PGLKLQVVEGRGH M- - L P I TATARVVEAVLHVAKRVR PVE TATVLHP P FALANK 332 BPO-EST P N G T L K T Y K G F P H G MP T THADV I NADLVA F I R S 276 CPO-P QNGT LKTY PGYSH G - - MLTVNADVLNADL LAFVQA 277 CPO-L ANATLKSYEGL PH G- -MLS THPEVLNPDLLAFVKS 275 HPO-R KNARLKVYPGL SH G - - MCT VNADTVNADL L S F I E S 273 BPO-A1 PNAE LKVYEGS SH G I AMVPGDKEKFNRDL L E FLNK 274 BPO-A2 PSAEYVEVEGAPH G- - LLWTHAEEVNTALLAFLAK 277 Fig. 5. Alignment of DCH of A. calc oaceticu s F46 with several serine-hydrolases and perhydrolases. The deduced amino acid sequences of DCH, the esterases of P. putida MR-2068 (EST-P) [25], P. fluorescence SIK WI (EST-F) [12,26], and P. fluorescence DSM 50106 (EST-F1) [27], and the perhydrolase of P. putida (BPO-EST) [28], P. pyrrocinia (CPO-P) [29], S. lividans (CPO-L) [7], R. erythropolis (HPO-R) [8], and S. aureofaciens (BPO-A1, BPO-A2) [30,31] are aligned. Identical amino acid residues are enclosed in boxes. Ser, Asn and His residues which are thought to be involved in the catalytic triad are denoted by black boxes. Ó FEBS 2003 Role of 3,4-dihydrocoumarin hydrolase of Acinetobacter (Eur. J. Biochem. 270) 491 Sensitivity to peroxides The tolerance of each strain to peroxides, i.e. peracetic acid and hydrogen peroxide, was measured by means of disk inhibition assays (Table 2). The dchD strain was more sensitive than the wild-type strain to peracetic acid. On the other hand, E. coli expressing DCH showed greater toler- ance to peracetic acid than E. coli bearing pKK223-3, but the difference in the diameters of growth inhibition zones of transformed E. coli strains was not appreciable in contrast with those of Acinetobacter strains. For this reason, it was assumed that E. coli JM109 itself possessed a certain tolerance to peroxoacids. The sensitivities to hydrogen peroxide and acetic acid were not affected in either Acinetobacter strain (dchD or wild-type) or E. coli strains (pDCH21 and pKK223-3 transformants). Discussion In the previous study, we found that the amino acid sequences of the N-terminal and internal peptide of DCH exhibited significant similarity to those of several hydrolases and perhydrolase, and that DCH was a bifunctional enzyme capable of both hydrolysis of ester bonds and halogenation [1]. The results of nucleotide sequencing analysis of dch were also in good agreement with this observation. The active site amino acid residues of serine-hydrolases, the so-called catalytic triad, consisting of the consensus motif Gly-X- Ser-X-Gly and Asp and His residues, were highly conserved in the deduced amino acid sequences of dch and homolog- ous proteins, and these enzymes were suggested to belong to the serine-ÔhydrolaseÕ family. On the other hand, the recombinant E. coli expressing dch showed not only lactone-hydrolysing activity but also monochlorodimedon- brominating activity. This fact indicates that both the hydrolysis of ester compounds and the halogenation of organic compounds must be catalysed by the same enzyme. In fact, several bifunctional enzymes, other than DCH, that catalyse both hydrolysis and hologenating reactions have already been reported [26,27]. However, the physiological roles of these enzymes were not known. To elucidate the functions of these bifunctional enzymes, we directed our 1234 97.4 66.3 42.4 30.0 20.1 kDa Fig. 6. SDS/PAGE of the purified DCH and cell-free extracts of recombinant E. co li strains. Lane 1, purified DCH, 10 lg; lanes 2 and 3, cell-free extracts of E. coli JM109 bearing pDCH21 and pKK223-3, respectively; lane 4, molecular mass standards (Daiichi Chemicals, Tokyo, Japan): phosphorylase b (97.4 kDa), BSA (66.3 kDa), aldo- lase (42.4 kDa), carbonic anhydrase (30.0 kDa), and trypsin inhibitor (20.1 kDa). The gel was stained for protein with Coomassie brilliant blue R-250 and destained in ethanol/acetic acid/water (3 : 1 : 6, v/v/v). Fig. 7. Northern analysis of total RNA from early log phase A. calco- aceticus F46 cells. Thrity lg total RNA was loaded on each lane. The internal gene fragment of either dch (A) or cat (B) was used as a probe. The approximate sizes of the transcripts are indicated. 492 K. Honda et al. (Eur. J. Biochem. 270) Ó FEBS 2003 attention to the ability of dch to catalyse the peroxidation of organic acids, which could be an intermediary reaction of halogenation, and it is assumed that the role of DCH in vivo is the degradation of peroxoacids. Peroxoacids are powerful antimicrobial agents like hydrogen peroxide, and are used for sterilization in several industrial settings. Anderson and Miller reported that cells of a plant-colonizing bacterium, P. putida,weremore sensitive to killing by peracetic acid when they lacked a major catalase activity [36]. However, it was assumed that the catalase catalysed the degradation of hydrogen per- oxide, which was part of the peracetic acid-equilibrium mixture and did not act directly on peracetic acid. On the other hand, Picard et al. revealed that a perhydrolase, CPO-T, isolated from S. aureofaciens, could degrade per- acetic acid [16] and here we demonstrate that purified DCH catalysed the decomposition of peracetic acid, in a time- and dose-dependent manner. Both of these results indicate that DCH and CPO-T act on peracetic acid itself. Although the products of the enzyme reaction were not determined in this study, DCH was thought to convert peracetic acid to acetic acid and hydrogen peroxide because of the existing evidence that DCH and other perhydrolases are able to catalyse the reverse reaction, i.e. the formation of a peracetic acid from acetic acid and hydrogen peroxide [1,12,16,28] and because DCH could catalyse the halogenation reaction in the presence of not only acetic acid but also other organic acids, such as formic acid, propionic acid, and n-butyric acid [1]: peroxoacids corresponding to these organic acids might also serve as substrates of the enzyme. DCH was originally isolated as a lactonohydrolase, which was specific for aromatic lactones, such as 3,4-dihydrocoumarin, 2-coumaranone and homogentisic acid lactone, and considered to participate in the degrada- tive metabolic pathway of polycyclic aromatic compounds [1]. However, judging from the K m and V max values for peracetic acid and other substrates, peracetic acid is likely to be a natural substrate for DCH in vivo.AdchD mutant derived from A. calcoaceticus F46 was more sensitive to growth inhibition by peracetic acid than the parent strain. On the other hand, E. coli expressing dch showed increased resistance to peracetic acid. These results also suggested that DCH detoxifies peroxoacids and plays a role in the oxidative stress defence system in vivo. Superoxide dismu- tase, catalase, peroxidase, etc. are already known to be antioxidant enzymes, and have been well investigated. However, enzymes responsible for defence against peroxo- acids have not yet been found. DCH detoxifies peracetic acid in a unique manner, i.e. ÔhydrolyticÕ degradation, and might be a new example of such antioxidant enzymes. In addition to these observations, a putative catalase gene was found immediately downstream of dch. Although the expression of these genes was constitutive and not induced on the addition of either peracetic acid or hydrogen peroxide to the culture medium (data not shown), Northern analysis revealed that they were tran- scribed as both monocistronic mRNAs and parts of the same polycistronic mRNA. This indicates that they are regulated by the same promoter, and DCH and the catalase might play a cooperative roles in the oxidative defence system, i.e. at first, peroxoacids are hydrolysed to the corresponding organic acids and hydrogen peroxide by DCH, and then the resulting hydrogen peroxide is degraded by the catalase. In prokaryotes, however, genes abutting each other tend to be transcribed as a polycis- tronic mRNA. To confirm this novel enzymatic antioxi- dative stress system, further investigations on DCH and other perhydrolases are required. Recently, with the great advances in genetic analysis, a lot of perhydrolase-like genes have been found in various bacteria. The novel oxidative stress defence system presented in this study might be more widespread in nature than we thought. Acknowledgements This work was supported in part by Grants-in-Aid for Scientific Research nos. 01729 (to K. H.) and 14360054 (to M. K.) from the Japan Society for the Promotion of Science. We also thank T. Ishige, A.Tani,Y.SakaiandN.Katofortheirhelpinthegenedisruptionof Acinetobacter. References 1. Kataoka, M., Honda, K. & Shimizu, S. (2000) 3,4-Dihydro- coumarin hydrolase with haloperoxidase activity from Acine- tobacter calcoaceticus F46. Eur. J. Biochem. 267, 3–10. 2. Wiesner, W., van Pe ´ e, K H. & Lingens, F. (1988) Purification and characterization of a novel bacterial non-heme chloroperoxidase from Pseudomonas pyrrocinia. J. Biol. Chem. 263, 13725–13732. 3. Itoh, N., Morinaga, N. & Kouzai, T. (1994) Purification and characterization of a novel metal-containing nonheme bromo- peroxidase from Pseudomonas putida. Biochim. Biophys. Acta 998, 63–68. Table 2. Oxidative stress susceptibility of each strain. Cells were grown overnight in 2 mL LB medium, mixed with 10 mL LB medium containing 1% agar and then poured onto an LB agar plate. A sterilized filter disk containing 10 lL peracetic acid, H 2 O 2 , or acetic acid at the concentrations indicated, was immediately placed onto the agar plate. The plates were incubated overnight at 28 °CforAcinetobacter strains, or at 37 °CforE. coli strains, and the diameters of growth halos were measured. Diameter of growth inhibition halo (mm) Peracetic acid H 2 O 2 Acetic acid Strain 0.6% 0.06% 10% 3% 10% 1% A. calcoaceticus F46 4.4 0 1.0 0.5 4.0 0 A. calcoaceticus dchD 7.8 1.4 1.0 0.5 4.5 0 E. coli JM109 (pDCH21) 0 0 4.3 0 6.2 0 E. coli JM109 (pKK223-3) 2.0 0 3.0 0 6.0 0 Ó FEBS 2003 Role of 3,4-dihydrocoumarin hydrolase of Acinetobacter (Eur. J. Biochem. 270) 493 4. Kirner, S., Krauss, S., Sury, G., Lam, S.T., Ligon, J.M. & van Pe ´ e, K H. (1996) The non-haem chloroperoxidase from Pseudomonas fluorescens and its relationship to pyrrolnitrin biosynthesis. Microbiology 142, 2129–2135. 5. Weng,M.,Pfeifer,O.,Krauss,S.,Lingens,F.&vanPe ´ e, K H. (1991) Purification, characterization and comparison of two non- heme bromoperoxidases from Streptomyces aureofaciens ATCC 10762. J. Gen. Microbiol. 137, 2539–2546. 6. Bantleon, R., Altenbuchner, J. & van Pe ´ e, K H. (1994) Chloro- peroxidase from Streptomyces lividans: isolation and character- ization of the enzyme and the corresponding gene. J. Bacteriol. 176, 2339–2347. 7. De Schrijver, A., Nagy, I., Schoofs, G., Proost, P., Vanderleyden, J., van Pe ´ e, K H. & De Mot, R. (1997) Thiocarbamate herbicide- inducible nonheme haloperoxidase of Rhodococcus erythropolis NI86/21. Appl. Environ. Microbiol. 63, 1911–1916. 8. Burd, W., Yourkevich, O., Voskoboev, A.J. & van Pe ´ e, K H. (1995) Purification and properties of a non-haem chloroperoxi- dase from Serratia marcescens. FEMS Microbiol. Lett. 129, 255– 260. 9. van Pe ´ e, K H. (1996) Biosynthesis of halogenated metabolites by bacteria. Annu. Rev. Microbiol. 50, 375–399. 10. Dairi, T., Nakano, T., Aisaka, K., Katsumata, R. & Hasegawa, M. (1995) Cloning and nucleotide sequence of the gene responsible for chlorination of tetracycline. Biosci. Biotechnol. Biochem. 59, 1099–1106. 11. Hofmann, B., To ¨ lzer, S., Pelletier, I., Altenbuchner, J., van Pe ´ e, K H. & Hecht, H.J. (1998) Structural investigation of the cofactor-free chloroperoxidases. J. Mol. Biol. 279, 889–900. 12. Pelletier, I. & Altenbuchner, J. (1995) A bacterial esterase is homologous with non-haem haloperoxidases and displays bro- minating activity. Microbiolgy 141, 459–468. 13. Pelletier, I., Altenbuchner, J. & Mattes, R. (1995) A catalytic triad is required by the non-heme haloperoxidases to perform halo- genation. Biochim. Biophys. Acta 1250, 149–157. 14. Hohaus, K., Altmannm, A., Burd, W., Fischer, I., Hammer, P.E., Hill, D.S., Ligon, J.M. & van Pe ´ e, K H. (1997) NADH-depend- ent halogenases are more likely to be involved in halometabolite biosynthesis than haloperoxidases. Angew Chem. Int. Ed. Engl. 36, 2012–2013. 15. Kirk, O. & Conrad, L.S. (1999) Metal-free haloperoxidases: Fact of artifact? Angew Chem. Int. Ed. Engl. 38, 977–979. 16. Picard, M., Gross, J., Lubbert, E., To ¨ lzer,S.,Krauss,S.,vanPe ´ e, K H. & Berkessel, A. (1997) Metal-free bacterial haloperoxidases as unusual hydrolases: Activation of H 2 O 2 by the formation of peracetic acid. Angew Chem. Int. Ed. Engl. 36, 1196–1199. 17. Bjo ¨ rkling, F., Frykman, H., Godtfredsen, S.E. & Kirk, O. (1992) Lipase catalyzed synthesis of peroxycarboxylic acids and lipase mediated oxidations. Tetrahedron 48, 4587–4592. 18. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 19. Bagdasarian, M., Lurz, R., Ru ¨ ckert, B.H., Franklin, F.C., Bagdasarian, M.M., Frey, J. & Timmis, K.N. (1981) Specific- purpose plasmid cloning vectors. II. Broad-host-range, high-copy- number, RSF1010-derived vectors, and a host-vector system for gene cloning in Pseudomonas. Gene 16, 237–247. 20. Hildebraunt, A.G. & Roots, I. (1975) Reduced nicotinamide adenine dinucleotide phosphate (NADPH)-dependent formation and breakdown of hydrogen peroxide during mixed function oxidation reactions in liver microsomes. Arch. Biochem. Biophys. 171, 385–397. 21. King, J. & Laemmli, U.K. (1971) Polypeptides of the tail fibres of bacteriophage T4. J. Mol. Biol. 62, 465–477. 22. Bradford, M.M. (1966) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 23. Leahy, J.G. (1994) Transformation of Acinetobacter calcoaceticus RAG-1byelectroporation.Can. J. Microbiol. 40, 233–236. 24. Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J. Biol. Chem. 166, 557–580. 25. Ozaki, E., Sakimae, A. & Numazawa, R. (1995) Nucleotide sequence of the gene for a thermostable esterase from Pseudo- monas putida MR-2068. Biosci. Biotechnol. Biochem. 59, 1204–1207. 26. Choi, K.D., Jeohn, G.H., Rhee, J.S. & Yoo, O.J. (1990) Cloning and nucleotide sequence of an esterase gene from Pseudomonas fluorescens and expression of the gene in Escherichia coli. Aglic. Biol. Chem. 54, 2039–2045. 27. Khalameyzer, V., Fischer, I., Bornscheuer, U.T. & Altenbuchner, J. (1999) Screening, nucleotide sequence, and biological characterization of an esterase from Pseudomonas fluorescens with high activity toward lactones. Appl. Environ. Microbiol. 65, 477–482. 28. Itoh, N., Kawanami, T., Liu, J Q., Dairi, T., Miyakoshi, M., Nitta,C.&Kimoto,Y.(2001)Cloningandbiochemicalchar- acterization of Co 2+ -activated bromoperoxidase-esterase (per- hydrolase) from Pseudomonas putida. Biochim. Biophys. Acta 1545, 53–66. 29. Wolffram, C., Lingens, F., Mutzel, R. & van Pe ´ e, K H. (1994) Chloroperoxidase-encoding gene from Pseudomonas pyrrocinica: sequence, expression in heterologous hosts, and purification of the enzyme. Gene 130, 131–135. 30. Pfeifer, O., Pelletier, I., Altenbuchner, J. & van Pe ´ e, K H. (1992) Molecular cloning and sequencing of a non-haem bromoper- oxidasegenefromStreptomyces aureofaciens ATCC 10762. J. Gen. Microbiol. 138, 1123–1131. 31. Pelletier, I., Pfeifer, O., Altenbuchner, J. & van Pe ´ e, K H. (1994) Cloning of a second non-haem bromoperoxidase gene from Streptomyces aureofaciens ATCC 10762: sequence analysis, expression in Streptomyces lividans and enzyme purification. Microbiology 140, 509–516. 32. Peters, M., Heinaru, A. & Nurk, A. (2001) Plasmid-encoded catalase KatA, the main catalase of Pseudomonas fluorescens strain Cb36. FEMS Microbiol. Lett. 200, 235–240. 33. Bishai, W.R., Smith, H.O. & Barcak, G.J. (1994) A peroxide/ ascorbate-inducible catalase from Haemophilus influenzae is homologous to the Escherichia coli katE gene product. J. Bacte- riol. 176, 2914–2921. 34. Sakai, Y., Mitsui, R., Katayama, Y., Yanase, H. & Kato, N. (1999) Organization of the genes involved in the ribulose mono- phosphate pathway in an obligate methylotrophic bacterium, Methylomonas aminofaciens 77a. FEMS Microbiol. Lett. 176, 125– 130. 35. Bofosian, G., Bilyeu, K. & O’Neil, J.P. (1993) Genome rear- rangements by residual IS10 elements in strains of Escherichia coli K-12 which had undergone Tn10. Gene 133, 12–22. 36. Anderson, A.J. & Miller, C.D. (2001) Catalase activity and the survival of Pseudomonas putida, a root colonizer, upon treatment with peracetic acid. Can. J. Microbiol. 47, 222–228. 494 K. Honda et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . Role of Acinetobacter calcoaceticus 3,4-dihydrocoumarin hydrolase in oxidative stress defence against peroxoacids Kohsuke Honda, Michihiko Kataoka, Eiji. SDS/PAGE and protein determination Acinetobacter strains and the recombinant E. coli strains were cultured in 3 mL LB medium. For cultivation of the recombinant