The catalytic mechanism of dye-decolorizing peroxidase DyP may require the swinging movement of an aspartic acid residue Toru Yoshida 1 , Hideaki Tsuge 2 , Hiroki Konno 1 , Toru Hisabori 1 and Yasushi Sugano 1 1 R1-7 Chemical Resources Laboratory, Tokyo Institute of Technology, Japan 2 Department of Bioresources and Environmental Sciences, Kyoto Sangyo University, Japan Introduction Heme peroxidase, one of the best-studied and most ubiquitous enzymes, oxidizes a variety of substrates, including phenolic and azo compounds. To date, many heme peroxidases have been isolated from vari- ous organisms, and the details of these can be found in the high-quality PeroxiBase (http://peroxibase. toulouse.inra.fr/index.php) database [1]. Heme perox- idases are classified into six families: animal peroxid- ases, nonanimal peroxidases, catalases, diheme cytochrome c peroxidases, dye-decolorizing peroxidase (DyP)-type peroxidases, and haloperoxidases. The cat- alytic cycle of peroxidases has been well studied [2], and proceeds as follows: Resting state þ H 2 O 2 ! compound I þ H 2 O Compound I þ AH 2 ! compound II þ AH Compound II þ AH 2 ! resting state þ AHþH 2 O Combined with the above: 2AH 2 þ H 2 O 2 ! 2H 2 O þ 2AH AH 2 and AH• mean a substrate and a radical pro- duct, respectively. In essence, enzymes that catalyze the above reaction are defined as peroxidases, even though details of the catalytic cycles or the nature of the Keywords catalytic mechanism; DyP; DyP-type peroxidase; heme protein; swing mechanism Correspondence Y. Sugano, R1-7 Chemical Resources Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku Yokohama 226-8503, Japan Fax: +81 45 924 5268 Tel: +81 45 924 5235 E-mail: ysugano@res.titech.ac.jp (Received 23 February 2011, revised 1 April 2011, accepted 4 May 2011) doi:10.1111/j.1742-4658.2011.08161.x The dye-decolorizing peroxidase (DyP)-type peroxidase family is a unique heme peroxidase family. The primary and tertiary structures of this family are obviously different from those of other heme peroxidases. However, the details of the structure–function relationships of this family remain poorly understood. We show four high-resolution structures of DyP ( EC 1.11.1.19), which is representative of this family: the native DyP (1.40 A ˚ ), the D171N mutant DyP (1.42 A ˚ ), the native DyP complexed with cyanide (1.45 A ˚ ), and the D171N mutant DyP associated with cyanide (1.40 A ˚ ). These structures contain four amino acids forming the binding pocket for hydrogen peroxide, and they are remarkably conserved in this family. Moreover, these structures show that OD2 of Asp171 accepts a proton from hydrogen peroxide in compound I formation, and that OD2 can swing to the appropriate position in response to the ligand for heme iron. On the basis of these results, we propose a swing mechanism in com- pound I formation. When DyP reacts with hydrogen peroxide, OD2 swings towards an optimal position to accept the proton from hydrogen peroxide bound to the heme iron. Abbreviations DyP, dye-decolorizing peroxidase; PDB, Protein Data Bank. FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS 2387 intermediates obtained may differ from those of stan- dard ubiquitous heme peroxidases. Compound I repre- sents an intermediate with an Fe 4+ oxoferryl center [3] and a porphyrin cationic radical [4,5], whereas com- pound II is obtained when one electron has been removed from compound I. The resulting AH• (an AH radical) is further converted to various products via non- enzymatic reactions (e.g. radical coupling steps). A great deal of information is available on the first step of the cycle, i.e. the generation of compound I using H 2 O 2 . In the most popular model, a distal histi- dine is essential, serving as an acid–base catalyst, and an equally essential arginine supports compound I for- mation [6]. In other words, most heme peroxidases have both a distal histidine and an essential arginine [7]. A simplified scheme of compound I formation, according to many studies [8–10], is shown in Fig.1. The reaction starts with deprotonation of hydrogen peroxide by histidine (an acid–base catalytic residue). Although the proposed peroxidase–H 2 O 2 complex existing just before the formation of compound I (the boxed state in Fig. 1) has not been experimentally demonstrated, it is believed that H 2 O 2 must interact with heme iron prior to compound I formation [8]. Deprotonation of H 2 O 2 is extremely improbable, because the pK a of this molecule is 11.6 [11]. However, H 2 O 2 bound to heme iron (Fe 3+ ) is estimated to have apKa of 3.2–4.0 [8]. If this is indeed the case, deproto- nation of H 2 O 2 appears to be reasonable, and a classi- cal reaction path (termed the Poulos–Kraut mechanism) has been proposed [6]. DyP ( EC 1.11.1.19) from the fungus Bjerkandera ad- usta Dec 1 (formerly called Thanatephorus cucumeris Dec 1) is a type of heme peroxidase, but also mediates hydrolysis of anthraquinone rings, indicating that the enzyme is bifunctional [12,13]. DyP from B. adusta Dec 1 is a member of the DyP-type peroxidase family [14], which is further subdivided into subfamilies A, B, C, and D, according to PeroxiBase. Research on DyP-type peroxidase family enzymes commenced only 15 years ago, much later than work on the more com- mon peroxidases. However, in recent times, several major studies of DyP-type peroxidases have appeared [15–17]. The DyP-type peroxidase AnaPX from the cyanobacterium Anabaena sp. PCC 7120 (type D) shares some characteristics with DyP from B. adusta Dec 1 [18]. However, two other types of DyP-type per- oxidase, YcdB (type A) [19] and YfeX (type B) from Escherichia coli, have been reported to cooperatively capture iron [20]. Surprisingly, the functions of YcdB and YfeX thus appear to have little to do with peroxi- dase activity. These studies suggest that DyP-type per- oxidases represent a novel form of heme enzyme, expressing activities that are not confined to peroxidase action. Moreover, such enzymes are widely distributed from bacteria to metazoa, indicating that the DyP-type peroxidase family is not exceptionally small, but is rather a sizeable grouping of proteins sharing type- unique characteristics. Thus, a key cluster of proteins has evolved to play various roles in a variety of organ- isms [14]. It is thus important to understand the vari- ous characteristics of DyPs. In the present study, we focused on elucidation of the catalytic mechanism, employing tertiary structural analysis. Three X-ray crystal structures of members of the DyP-type peroxidase family, bound to heme or proto- porphyrin IX, have been deposited in the Protein Data Bank (PDB). YcdB (PDB ID 2WX7, 2.3 A ˚ resolution) is a type A enzyme; TyrA (PDB ID 2iiz, 2.3 A ˚ resolu- tion) is a type B enzyme [21]; and DyP (PDB ID 2D3Q, 2.96 A ˚ resolution) is a type D enzyme [12]. Surprisingly, the levels of primary structural identity between DyP (2D3Q) and the other enzymes are very low (< 5%), although the overall tertiary structures are very similar [14]. One of the most interesting char- acteristics of DyP-type peroxidases is that the catalytic residue is not histidine but rather aspartic acid (Asp171 in DyP) [12]. However, the details of struc- ture–function relationships in the DyP family remain poorly understood. Here, we obtained four structures of a DyP (type D) enzyme at 1.40–1.45 A ˚ resolution: native DyP (native); the D171N mutant DyP (D171N), native DyP complexed with cyanide (CN) (native-CN), and the D171N mutant DyP associated with CN (D171N-CN), and identified the precise positions of Asp171 and associated residues. These structures show that OD2 accepts a proton from H 2 O 2 bound to the heme iron. Nevertheless, in the native structure, OD2 is too far away to accept the proton from the H 2 O 2 . On the other hand, the four structures obviously show that OD2 is able to swing to the appropriate position Fe 3+ Fe 3+ Fe 4+ + H O Fe 3+ : : : OH O OH H 2 O Compound IResting state N NH NH+ NH N NH N NH – O Fig. 1. Schematic diagram of the most popular mechanism advanced to explain compound I formation by peroxidases. Struc- tures of the resting state, two deduced intermediates and com- pound I are shown. The black bars enclosing iron atoms represent porphyrin rings. The acid–base catalytic residue is indicated as the imidazole group of histidine. In compound I, the plus sign and the black dot indicate when the porphyrin ring is a cationic radical. The box indicates the region that is the focus of the present study. Catalytic residue Asp171 swings in view T. Yoshida et al. 2388 FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS in response to ligand for heme iron. On the basis of these analyses, we propose a new mechanism for the formation of compound I by DyP. When H 2 O 2 binds to the heme iron of DyP, OD2 of the catalytic residue Asp171 swings to a position that is optimal for interac- tion with the H 2 O 2 . Results Approach pathway and binding site of H 2 O 2 In all heme peroxidases, H 2 O 2 is believed to bind to the heme-distal side. This means that an H 2 O 2 approach pathway from the enzyme molecular surface to the heme-distal region must exist. To date, no details of any such pathway have been reported for DyP-type peroxid- ases. In the present study, we obtained a native structure at 1.40-A ˚ resolution (PDB ID 3AFV). This high-resolu- tion structure showed much more precise details of enzyme conformation and molecular surface arrange- ment than were previously available [12]. Heme was not exposed to the molecular surface. However, a large cav- ity was evident towards the heme-distal side (Fig. 2A), and formed a small cylindrical pocket about 3 A ˚ in both diameter and height. H 2 O 2 appeared to fit into this pocket. In contrast, bulky substrates, such as anthraqui- none, did not appear to bind into this pocket. Further- more, this region was surrounded by side chains of four amino acids: Asp171 (the catalytic residue), Arg329, Leu354, and Phe356. Overall, the results suggest that H 2 O 2 may reach the heme-distal side by passing through the cavity, and then bind and react (with heme) within the pocket (Fig. 2B). Conservation of residues forming the binding pocket for H 2 O 2 Primary sequence identities among different types of DyP-type peroxidases are, at most, 15%. Indeed, this value falls to < 5% when types A and D are compared. Nevertheless, YcdB, TyrA, and DyP, of types A, B, and D, respectively, have very similar b-barrel folds and a-helical structural regions. As described above, we confirmed that the side chains of Asp171, Arg329, Leu354 and Phe356 form a binding pocket for H 2 O 2 . On the basis of X-ray crystal struc- tures and multiple sequence alignments among all DyP-type peroxidase family members (types A, B, C, and D), we confirmed conservation of the four residues that form binding pockets for H 2 O 2 . In YcdB (type A), the relevant residues are Asp200, Arg312, Leu331, and Phe333. In TyrA (type B), the residues are Asp151, Arg242, Leu255, and Phe257. Although BtDyP (type B) does not bind heme or protoporphy- rin IX (PDB ID 2gvk), the relevant residues in this protein seem to be Asp157, Arg245, Thr260, and Phe262. No X-ray crystal structure of a type C enzyme has been obtained. However, multiple sequence align- ment of all DyP-type peroxidases in PeroxiBase showed that the four residues discussed above were remarkably conserved as compared with other residues (Fig. S1). These results show that the b-barrel fold, the a-helical structural regions and the binding pocket for H 2 O 2 are conserved in members of the DyP-type peroxidase family. D171N structure We previously reported that the enzyme activity of the D171N was 1 ⁄ 3000th that of native [12]. We obtained a D171N structure at 1.42-A ˚ resolution (PDB ID 3MM1) (Fig. 3), and compared this with the native structure at 1.40 A ˚ resolution. Both structures were superimposed on the heme plane and several rmsd val- ues were calculated (Table 1). The overall structure of the two enzymes was very similar. The structures at the heme-distal side were also similar. In both struc- tures, two water molecules were positioned in the H 2 O 2 D171 R329 L354 L354 F356 75° AB R329 D171 F356 Heme plane H 2 O 2 3 Å Large cavity 3 Å 3 Å Fig. 2. Approach pathway and binding site of H 2 O 2 in DyP. (A) The cutaway view indicates the molecular surface of the entire structure. The black square indicates the heme plane. The broken arrow in the large cavity shows the pathway taken by H 2 O 2 when it approaches the heme-distal side. (B) Close-up views of an end of the cavity [circled in (A)]. This region of the cavity forms a binding pocket for H 2 O 2 . The broken arrow shows the approach pathway of H 2 O 2 . The pocket is delineated by double-headed arrows. Four residues forming the pocket are shown in stick format. T. Yoshida et al. Catalytic residue Asp171 swings in view FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS 2389 binding pocket of H 2 O 2 , and the positional relation- ships of the four residues forming the binding pocket were very similar. OD1 of Asp171 (Asn171) formed hydrogen bonds with the amide nitrogen of Gly172 and NH1 of Arg329, but did not form a hydrogen bond with a water molecule in the binding pocket. OD2 (ND2) did not form a hydrogen bond with the peptide chain, but formed a bond with a water mole- cule in the binding pocket. No polar atoms that could form hydrogen bonds with OD2 (ND2) were noted in the peptide chain within 5.0 A ˚ of OD2 (ND2). Excep- tionally, the positions of Asp171 and Asn171 seemed to differ between the two structures. In fact, the rmsd for Asp171(Asn171) was notably larger than that for other residues. It is of particular interest that the rmsd for CA of Asp171(Asn171) was very small, but the rmsd for OD2 (ND2) was clearly large. This probably results in alternation of proton acceptor and donor between the native and the D171N. In the native pro- tein, OD2 is the proton acceptor and W1227 the pro- ton donor. In contrast, in the D171N mutant, ND2 is the proton donor and W1200 the proton acceptor. These results strongly support the previous suggestion that Asp171 functions as a catalytic residue [12]. Coordination of CN CN coordination was examined by spectroscopy and X-ray crystallography. Because the speed of reaction between peroxidase and H 2 O 2 is very high, the binding mode has not been experimentally demonstrated. On the other hand, CN binds stably to the heme iron. Actually, complexes of peroxidase with CN have been believed to mimic peroxidase–H 2 O 2 complexes [22]. Although the binding mode of the heme iron differs between H 2 O 2 and CN, the position of the carbon atom of CN bound to the heme iron mimics the posi- tion of the proximal oxygen of H 2 O 2 bound to the heme iron during compound I formation. Therefore, the carbon atom position of the CN should provide basic information for understanding the interaction between OD2 of Asp171 and the proximal oxygen of H 2 O 2 bound to the heme iron in compound I formation. The addition of CN to native led to a shift in the Soret band from 406 nm to 421 nm, and created an additional absorption maximum at 535 nm with a shoulder at 565 nm (Fig. 4). This change was similar to that observed in horseradish peroxidase and Arthro- myces ramosus peroxidase upon binding of CN [2,23]. The data suggest that CN bound to the heme iron of native, and that the electron state of the iron then changed from high-spin to low-spin. We obtained a Native Native-CN D171N D171N-CN D171 D171 N171 N171 G172 G172 G172 G172 R329 R329 R329 R329 L354 L354 L354 L354 F356 F356 F356 F356 CN CN W1200 W1227 Fig. 3. Structures at the heme-distal side of the native, D171N, native-CN and D171N-CN enzymes. Heme molecules are shown as white sticks. Blue spheres represent water molecules and W means oxygen of water. Broken lines between two atoms indicate that the distance between these atoms is < 3.4 A ˚ . The 2F o ) F c electron density map at 1r is shown in pink for water molecules and cyanide ions. Brown circles represent the OD1 atom of Asp171 or Asn171, and black circles represent the OD2 atom of Asp171 or the ND2 atom of Asn171. Table 1. Rmsd values between two structures. Native and D171N Native and native-CN Native and D171N-CN All CA 0.08 0.38 0.57 Heme plane a 0.02 0.04 0.04 Asp171 (Asn171) All atoms 0.29 0.53 0.46 Main chain 0.20 0.40 0.34 Side chain 0.35 0.64 0.55 CG, OD1, OD2 (ND2) atoms 0.39 0.70 0.59 OD2 (ND2) atom 0.61 1.03 0.78 Arg329 All atoms 0.08 0.24 0.36 Main chain 0.07 0.24 0.34 Side chain 0.09 0.25 0.36 Leu354 All atoms 0.04 0.29 0.22 Main chain 0.04 0.31 0.25 Side chain 0.04 0.28 0.18 Phe356 All atoms 0.16 0.22 0.27 Main chain 0.05 0.26 0.30 Side chain 0.20 0.18 0.24 a The heme plane represents the 24 atoms of porphyrin. Catalytic residue Asp171 swings in view T. Yoshida et al. 2390 FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS native-CN enzyme structure at 1.45 A ˚ resolution (PDB ID 3MM2) and one of D171N-CN at 1.40 A ˚ resolu- tion (PDB ID 3MM3) (Fig. 3). In both structures, CN bound almost vertically to the heme plane, and the nitrogen atom of CN formed a hydrogen bond with an adjacent water molecule. When the native-CN struc- ture was compared with that of the native, by superim- position in the heme plane, the rmsd for Asp171, especially OD2, was apparently large (Table 1). When D171N-CN was compared with the native by superim- position on the heme plane, the rmsd for Asn171, especially ND2, was again rather large. Interestingly, OD2 of Asp171 and ND2 of Asn171 formed hydrogen bonds with different molecules. In native CN, OD2 formed a bond with the water molecule adjacent to CN. In contrast, ND2 formed a hydrogen bond with CN of D171N-CN. In the four structures obtained in the present study, OD1 could not always form a hydrogen bond with a molecule in the binding pocket of H 2 O 2 , but always participated in formation of two hydrogen bonds with the peptide chain. In contrast, OD2 (ND2) could always form a hydrogen bond with a molecule in the binding pocket of H 2 O 2 , but could not always engage in hydrogen bonding with the peptide chain. These results strongly suggest that OD2, rather than OD1, accepts a proton from H 2 O 2 in the binding pocket. Discussion Swinging of Asp171 It is important to note that OD2 of Asp171 in native- CN and ND2 of Asn171 in D171N-CN formed hydro- gen bonds with different molecules. As a result, the positions of OD2 and ND2 were very different. This appears to have been induced by variation in the pro- tonation state of OD2 and ND2. In native-CN, OD2 is a proton acceptor at pH 6.0. The crystallization con- dition was also at pH 6.0. Therefore, OD2 cannot form a hydrogen bond with CN, which is a proton acceptor, but can form such a bond with a water mole- cule adjacent to CN, which serves as a proton donor. Thus, OD2 is shifted in position, in a direction away from the heme plane, as compared with the native. On the other hand, in D171N-CN, ND2 is a proton donor at pH 6.0. Therefore, ND2 can form a hydrogen bond with CN. As a result, ND2 is shifted in position in a direction towards the heme plane, as compared with the native. This suggests that OD2 can change position in response to ligand status in the binding pocket. This flexibility of OD2 seems to be associated with the fact that OD2 does not possess a polar atom that can form a hydrogen bond. Moreover, such flexibility was strongly supported by superimposition of the struc- tures of the native, D171N, native-CN and D171N- CN enzymes in the heme plane (Fig. 5). Because of differences in the chosen hydrogen bond partners, OD2 (ND2) seems to swing around OD1, by over 37°. These results suggest that OD2 can swing in response to ligand status. The catalytic residue Asp171 swings to form the compound I intermediate At which position does OD2 accept a proton from H 2 O 2 ? Because an X-ray crystal structure at 1.4 A ˚ reso- lution cannot show hydrogen atoms, we discuss this issue from the viewpoint of the distance between OD2 and the proximal oxygen. We replace the carbon atom position of CN with a position of the proximal oxygen Native Native-CN 5 Fig. 4. UV–visible spectra of native and native-CN enzymes. Spec- tra from 450 nm to 700 nm are shown at five-fold magnification. Solution conditions were 4 l M DyP in 25 mM citrate buffer (pH 5.5) containing 0.5 M NaCl, with or without 100 mM KCN. CN 2.05 A 37.0 OD1 OD2 (ND2) D171 (N171) D171 (N171) OD1 OD2 (ND2) 60 Fig. 5. Comparison of Asp171 (Asn171) locations between native (green), D171N (blue), native-CN (yellow) and D171N-CN (red) enzymes. These four structures are superimposed on the heme planes. Gray broken lines show the shortest distances between OD2 (ND2) and the carbon atom of CN for each structure. The dis- tance between the CN carbon atom and the iron atom of heme is 2.05 A ˚ . On the right, the relevant four residues are rotated by 60° to assist in an understanding of differences in residue positions. T. Yoshida et al. Catalytic residue Asp171 swings in view FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS 2391 of H 2 O 2 as described in Results. The distances between OD2 (ND2) and the proximal oxygen are shown in Fig. 5. In the native and D171N-CN structures, the dis- tances are 4.06 A ˚ and 3.46 A ˚ , respectively. We think that this indicates a significant difference. That is because this difference is caused by the difference between the positions of hydrogen bond partners of OD2 in the native and ND2 in D171N-CN. In the native, the distance, 4.06 A ˚ , is too long to permit reac- tion with H 2 O 2 . This shows that this position of OD2 in the native is not appropriate for involvement in the reac- tion. In contrast, in D171N-CN, the distance, 3.46 A ˚ ,is less than in the native. The position of OD2 in D171N- CN is appropriate for involvement in the reaction. Our argument is illustrated in Fig. 6A. Thus, OD2 never accepts a proton when in the position occupied by OD2 in the native, but does so when in the position occupied by OD2 in D171N-CN (Fig. 6A). On the basis of the dual conclusions that OD2 accepts a proton when in the position occupied by OD2 in D171N-CN, and that OD2 can swing from the position occupied in the native to the position seen in D171N-CN, we propose a swing mechanism for the formation of compound I by DyP (Fig. 6B). To accept a proton, OD2 of Asp171 swings towards the position occupied by OD2 in D171N-CN. After compound I formation, OD2 of Asp171 returns to the position characteristic of the native. The side chain of the cata- lytic residue of DyP is not located just above the heme iron, unlike in other heme peroxidases. Moreover, the side chain of DyP is arranged in parallel rather than vertically to the heme plane. This novel location and arrangement seems to produce the swing mechanism. However, this swing mechanism may be needed in type C and D DyP-type peroxidases. Nevertheless, we believe that this structural study paves the way to understanding the structure–function relationships of the DyP-type peroxidase family. Experimental procedures Crystallization of native and D171N proteins Native and D171N proteins were purified with a modifica- tion of published procedures [12,24,25]. Purified samples were deglycosylated by endoglycosidase H (Roche Diagnos- tics, Tokyo, Japan). Samples were loaded onto Superdex 75 columns (GE Healthcare Japan, Tokyo, Japan), and the de- glycosylated fractions were concentrated to 20 mgÆmL )1 by ultrafiltration. Crystallization was achieved with the hang- Fig. 6. The swinging mechanism of Asp171. (A) Interpretation of distances between Asp171 and a virtual proximal oxygen of H 2 O 2 in the native and D171N-CN. The native and D171N-CN structures are superimposed on the heme planes. Note that Asn171 of D171N-CN is shown as Asp171. To avoid misunderstanding, only the heme of the native is shown in white. The position of the CN carbon mimics the position of the proximal oxygen of H 2 O 2 . Double-headed arrows show whether the distance between Asp171 and the virtual proximal oxy- gen is appropriate to permit reaction. (B) Schematic diagram of the proposed mechanism of compound I formation by DyP. Structures of the resting state, two deduced intermediates, and compound I are shown. The two black bars enclosing an iron atom indicate the porphyrin ring. Broken lines and arrows indicate hydrogen bonds and the swinging direction of the OD2 atom of Asp171, respectively. In compound I, a plus sign and a black dot indicate when the porphyrin ring is a cationic radical. Catalytic residue Asp171 swings in view T. Yoshida et al. 2392 FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS ing drop vapor diffusion method. Drops containing 1 lLof a20mgÆmL )1 protein solution (0.1 m Mes at pH 6.0; 0.5 m NaCl) and 1 lL of mother solution [0.1 m Mes at pH 6.0; 48% (w ⁄ v) poly(ethylene glycol) 8000] were equilibrated against 500 lL of reservoir solution [0.1 m Mes at pH 6.0; 0.25 m NaCl; 28% (w ⁄ v) poly(ethylene glycol) 8000] at 278 K. Hexagonal crystals appeared after 2–3 weeks. Prior to data collection, crystals were soaked briefly in a cryopro- tective solution containing 0.1 m Mes at pH 6.0, 0.25 m NaCl, 30% (w ⁄ v) poly(ethylene glycol) 8000, and 25% (v ⁄ v) glycerol, and flash frozen in liquid nitrogen. CN-complexed crystals of native and D171N proteins CN-complexed crystals were prepared by soaking native and D171N proteins in a cryoprotective solution containing 0.1 m Mes at pH 6.0, 0.25 m NaCl, 30% (w ⁄ v) poly(ethylene glycol) 8000, 25% (w ⁄ v) glycerol, and 120 mm KCN. For both crystal types, binding of CN appeared to be complete within a few seconds, as assessed by visual monitoring of the crystal color change from brown to red. The crystals were flash frozen in liquid nitrogen. X-ray data collection and structural refinement Data collection from native, the D171N, native-CN and D171N-CN was performed with a wavelength of 1.0 A ˚ on a beamline PF-AR NE3A or PF-AR NW12A instrument at the Photon Factory (Tsukuba, Japan). Subsequent proce- dures, including processing, scaling, and refinement, were identical for all crystals. Datasets were processed and scaled with the hkl2000 program [26]. Structures were solved with the molecular replacement software of the ccp4 program suite [27] (molrep), employing a 2.96-A ˚ -resolution structure of DyP (PDB ID 2D3Q) as the starting point. Iterative refinement and model-building were subsequently per- formed with refmac5 [28] and coot [29]. Data collection and refinement statistics are summarized in Table S1. For native DyP, three datasets were collected, refined, and superimposed on the heme plane. The rmsd values for Asp171 were very small in all three datasets (Fig. S2; Table S2). Two datasets were collected and refined for the D171N, and superimposed on the heme plane. The rmsd values of Asn171 were rather large in both datasets. This was because the hydrogen bond between the protonated ND2 of Asn171 and a water molecule in the binding pocket was weak. UV–visible spectrophotometry (solution studies) All spectra were obtained with a Shimadzu UV-2400 PC spectrophotometer (Shimadzu Co., Kyoto, Japan) at 30 °C, with a spectral bandwidth of 1.0 nm, employing cuvettes of light path 1 cm. Solution conditions were 4 lm DyP in 25 mm citrate buffer (pH 5.5) containing 0.5 m NaCl, with or without 100 mm KCN. Acknowledgements This research was undertaken with the assistance of the Photon Factory in KEK (proposal numbers 2010G011 and 2008G063) and was partly supported by a Grant-in-Aid for Scientific Research (no. 22570136) from the Japan Society for the Promo- tion of Sciences. References 1 Koua D, Cerutti L, Falquet L, Sigrist CJ, Theiler G, Hulo N & Dunand C (2009) PeroxiBase: a database with new tools for peroxidase family classification. Nucleic Acids Res 37, D261–266. 2 Dunford HB (1999) Heme Peroxidases. Wiley, New York. 3 Lang G, Spartalin K & Yonetani T (1976) Mo ¨ ssbauer spectroscopic study of compound ES of cytochrome c peroxidase. Biochim Biophys Acta 451, 250–258. 4 Browlett WR & Stillman MJ (1981) Evidence for heme pi cation radical species in compound I of horseradish peroxidase and catalase. Biochim Biophys Acta 660, 1–7. 5 Kaneko Y, Tamura M & Yamazaki I (1980) Formation of porphyrin pi cation radical in zinc-substituted horse- radish peroxidase. Biochemistry 19, 5795–5799. 6 Poulos TL & Kraut J (1980) The stereochemistry of peroxidase catalysis. J Biol Chem 255, 8199–8205. 7 Welinder KG (1992) Superfamily of plant, fungal and bacterial peroxidases. Curr Opin Struct Biol 2, 388–393. 8 Jones P & Dunford HB (2005) The mechanism of com- pound I formation revisited. J Inorg Biochem 99, 2292– 2298. 9 Raven EL (2003) Understanding functional diversity and substrate specificity in haem peroxidases: what can we learn from ascorbate peroxidase? Nat Prod Rep 20, 367–381. 10 Derat E, Shaik S, Rovira C, Vidosich P & Alfonso-Pri- eto M (2007) The effect of a water molecule on the mechanism of formation of compound 0 in horseradish peroxidase. J Am Chem Soc 129, 6346–6347. 11 Uri N & Evans MG (1949) The dissociation constant of hydrogen peroxide and the electron affinity of the HO 2 radical. Trans Faraday Soc 45, 224–230. 12 Sugano Y, Muramatsu R, Ichiyanagi A, Sato T & Shoda M (2007) DyP, a unique dye-decolorizing peroxi- dase, represents a novel heme peroxidase family: Asp171 replaces the distal histidine of classical peroxid- ases. J Biol Chem 282, 36652–36658. T. Yoshida et al. Catalytic residue Asp171 swings in view FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS 2393 13 Sugano Y, Matsushima Y, Tsuchiya K, Aoki H, Hirai M & Shoda M (2009) Degradation pathway of an anthraquinone dye catalyzed by a unique peroxidase DyP from Thanatephorus cucumeris Dec 1. Biodegradation 20, 433–440. 14 Sugano Y (2009) DyP-type peroxidases comprise a novel heme peroxidase family. Cell Mol Life Sci 66, 1387–1403. 15 Ogola HJ, Hashimoto N, Miyabe S, Ashida H, Ishikawa T, Shibata H & Sawa Y (2010) Enhancement of hydrogen peroxide stability of a novel Anabaena sp. DyP-type peroxidase by site-directed mutagenesis of methionine residues. Appl Microbiol Biotechnol 87, 1727–1736. 16 Hofrichter M, Ullrich R, Pevyna MJ, Liers C & Lundell T (2010) New and classic families of secreted fungal heme peroxidases. Appl Microbiol Biotechnol 87, 871–897. 17 Liers C, Bobeth C, Pecyna M, Ullich R & Hofrichter M (2010) DyP-like peroxidases of the jelly fungus Auricularia auricular-judae oxidize nonphenolic lignin model compounds and high-redox potential dyes. Appl Microbiol Biotechnol 85, 1869–1879. 18 Ogola HJ, Kamiike T, Hashimoto N, Ashida H, Ishika- wa T, Shibata H & Sawa Y (2009) Molecular character- ization of a novel peroxidase from the cyanobacterium Anabaena sp. strain PCC 7120. Appl Environ Microbiol 75, 7509–7518. 19 Sturm A, Schierhorn A, Lindenstrauss U, Lilie H & Bru ¨ ser T (2006) YcdB from Escherichia coli reveals a novel class of Tat-dependently translocated hemopro- teins. J Biol Chem 281, 13972–13978. 20 Le ´ toffe ´ S, Heuck G, Delepelaire P, Lange N & Wandersman C (2009) Bacteria capture iron from heme by keeping tetrapyrrol skeleton intact. Proc Natl Sci Acad USA 106, 11719–11724. 21 Zubieta C, Joseph R, Krishna SS, McMullan D, Kapoor M, Axelrod HL, Miller MD, Abdubek P, Acosta C, Astakhova T et al. (2007) Identification and structural characterization of heme binding in a novel dye-decolorizing peroxidase, TyrA. Proteins 69, 234– 243. 22 Edwards SL & Poulos TL (1990) Ligand binding and structural perturbations in cytochrome c peroxidase. J Biol Chem 265, 2588–2595. 23 Fukuyama K, Kunishima N, Amada F, Kubota T & Matsubara H (1995) Crystal structures of cyanide- and triiodide-bound forms of Arthromyces ramosus peroxi- dase at different pH values. Perturbations of active site residues and their implication in enzyme catalysis. J Biol Chem 270, 21884–21892. 24 Sugano Y, Nakano R, Sasaki K & Shoda M (2000) Efficient heterologous expression in Aspergillus oryzae of a unique dye-decolorizing peroxidase, DyP, of Geo- trichum candidum Dec 1. Appl Environ Microbiol 66, 1754–1758. 25 Saijo S, Sato T, Tanaka N, Ichiyanagi A, Sugano Y & Shoda M (2005) Precipitation diagram and optimization of crystallization conditions at low ionic strength for deglycosylated dye-decolorizing peroxidase from a basidiomycete. Acta Crystallogr F 61, 729–732. 26 Otwinowski Z & Minor W (1997) Processing of x-ray diffraction data collected in oscillation mode. Methods Enzymol 276, 307–326. 27 Collaborative Computational Project Number 4 (1994) The CCP4 suite: programs for protein crystallography. Acta Crystallogr D 50, 760–763. 28 Vagin A & Teplyakov A (2000) An approach to multi- copy search in molecular replacement. Acta Crystallogr D 56, 1622–1624. 29 Emsley P & Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D 60, 2126–2132. Supporting information The following supplementary material is available: Fig. S1. Structure-based sequence alignments of YcdB, BtDyP, TyrA, and DyP. Fig. S2. Comparison of the locations of Asp171 (Asn171) side chains among structures obtained with different datasets. Table S1. Data collection and refinement statistics. Table S2. Rmsd values between two structures. This supplementary material can be found in the online version of this article. Please note: As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Catalytic residue Asp171 swings in view T. Yoshida et al. 2394 FEBS Journal 278 (2011) 2387–2394 ª 2011 The Authors Journal compilation ª 2011 FEBS . The catalytic mechanism of dye-decolorizing peroxidase DyP may require the swinging movement of an aspartic acid residue Toru Yoshida 1 ,. are defined as peroxidases, even though details of the catalytic cycles or the nature of the Keywords catalytic mechanism; DyP; DyP- type peroxidase; heme