J. Biochem. 2014;155(2):115–122

doi:10.1093/jb/mvt104

Solution structure of the chitin-binding domain 1 (ChBD1) of a hyperthermophilic chitinase from Pyrococcus furiosus Received August 21, 2013; accepted November 9, 2013; published online November 21, 2013

Shouhei Mine1, Tsutomu Nakamura1, Takaaki Sato1, Takahisa Ikegami2,* and Koichi Uegaki1,y 1 National Institute of Advanced Industrial Science and Technology (AIST), 1-8-31 Midorigaoka, Ikeda, Osaka 563-8577 and 2Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan

y Koichi Uegaki, National Institute of Advanced Industrial Science and Technology (AIST), 1-8-31 Midorigaoka, Ikeda, Osaka 563-8577, Japan. Tel: þ81-72-751-9269, Fax: þ81-72-751-8370, email: [email protected]

A chitinase, from Pyrococcus furiosus, is a hyperthermophilic glycosidase that effectively hydrolyses both a and b crystalline chitin. This chitinase has unique structural features; it contains two catalytic domains (AD1 and AD2) and two chitin-binding domains (ChBD1 and ChBD2). We have determined the structure of ChBD1, which significantly enhances the activity of the catalytic domains, by nuclear magnetic resonance spectroscopy. The overall structure of ChBD1 had a compact and globular architecture consisting of three anti-parallel b-strands, similar to those of other proteins classified into carbohydratebinding module (CBM) family 5. A mutagenesis experiment suggested three solvent-exposed aromatic residues (Tyr112, Trp113 and Tyr123) as the chitin-binding sites. The involvement of Tyr123 or the corresponding aromatic residues in other CBMs, has been demonstrated for the first time. This result indicates that the binding mode may be different from those of other chitin-binding domains in CBM family 5. In addition, the binding affinities of ChBD1 and ChBD2 were quite different, suggesting that the two ChBDs each play a different role in efficiently increasing the activities of AD1 and AD2. Keywords: Chitinase/chitin-binding domain/carbohydrate-binding module/CBM family 5. Abbreviations: CBD, cellulose-binding domain; CBM, carbohydrate-binding module; ChBD, chitin-binding domain; HSQC, heteronuclear single-quantum correlation; NMR, nuclear magnetic resonance; NOESY, nuclear Overhauser effect spectroscopy; Pf, Pyrococcus furiosus; TOCSY, total correlation spectroscopy; ORF, open reading frame.

Introduction Chitin is an insoluble crystalline of a b-1,4-linked polymer of N-acetyl-D-glucosamine, and is a major

ß The Authors 2013. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved

115

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

*Takahisa Ikegami, Institute for Protein Research, Osaka University, 3-2 Yamadaoka, Suita, Osaka 565-0871, Japan. Tel: þ81-6-6879-4334, Fax: þ81-6-6879-8600, email: tiik@ protein.osaka-u.ac.jp

constituent of the shells of crustaceans such as crabs and shrimps, the exoskeletons of insects and the cell walls of many fungi. Crystalline chitins, consisting of chitin chains connected in bundle morphology, are classified into a and b forms. The a crystalline chitin is the most abundant form in nature. Chitinase, an enzyme that hydrolyses the b-(1,4) bond of chitin, is expressed in diverse organisms ranging from archaea to mammals. In nature, two putative chitinase genes (PF1233 and PF1234) were annotated in the Pyrococcus furiosus genome database. However, we did not detect chitinase activity in the culture medium of P. furiosus; we found that the stop codon of PF1234 was located in the middle of the putative catalytic domain. When the nucleotide (1,006th adenine from the ORF start position) was deleted from PF1234 and the resultant frame shift taken into account, the two genes were successfully combined into a longer gene (its product is abbreviated as Pf-ChiA) (1, 2) (Fig. 1A). This Pf-ChiA demonstrated the maximal enzyme activity for not only the b-form of chitin, but also the a-form at 490 C (2). Pf-ChiA is a large protein of 1,075 amino acids; it is comprised of two catalytic domains, AD1 and AD2, and two chitin-binding domains, ChBD1 and ChBD2 (Fig. 1B). Intermolecular interactions between the ChBDs and chitin are expected to force the catalytic domains to be tethered in close proximity to the surface of chitin, resulting in efficient hydrolysis of the crystalline chitin. To further understand the enzymatic mechanism of Pf-ChiA, structural analysis of the functional domains of Pf-ChiA was required. The carbohydrate-binding domains (modules; CBMs) are defined as functional domains having carbohydrate-binding activity and at present 32,130 modules have been classified into 67 families mainly on the basis of the amino acid sequences (CAZY database, http://www.cazy.org/ Carbohydrate-BindingModules.html). We previously reported the tertiary structure of the ChBD2 domain, constituted of 100 amino acids of Pf-ChiA and the detailed mechanism of its functions (9). ChBD2 possesses three Trp residues (Trp274, Trp308 and Trp326) longitudinally aligned on the surface of the protein that interact with crystalline chitin. Based on amino acid sequence homology, ChBD2 is classified into CBM family 2, whereas ChBD1 (the core structure consisting of about 50 amino acids) is classified into CBM family 5. These two CBM families show no amino acid sequence homology to each other. Of the CBM family 5 members, the structures of at least four have been determined, i.e. chitinase 60 (ChBDChi60) from Moritella marina (10), chitinase C (ChBDChiC) from Streptomyces griseus HUT6037 (3),

S. Mine et al.

A

B

Fig. 1 Construction of Pf-ChiA. (A) Two ORFs annotated as chitinase genes in the genome database. The primarily active chitinase has been divided into two non-active fragments by the stop codon of the PF1234 gene. (B) Schematic structure of Pf-ChiA. Arrow indicates the position of the stop codon in PF1234. (C) The amino acid sequences of some CBMs that are classified into CBM family 5 or 12. The exposed aromatic residues, which are involved in ligand binding, are boxed and written in bold. ChBDChiC represents chitinase C from Streptomyces griseus HUT6037 (3, 4). ChBDChiB represents chitinase B from Serratia marcescens (5). ChBDEGZ represents endoglucanase Z from Erwinia chrysanthemi (6). Six representative structures of CBMs that belong to CBM family 5 or 12 were aligned using GASH (7), a robust method for aligning protein structures. As GASH deals with a pairwise alignment alone, all the pairwise combinations were input to the tool and the alignment results were further modified manually, so that they were consistent in the form of a multiple alignment. The sequence similarities, estimated by ClustalW2 (8) independently of any conformation, between ChBD1 and each of ChBDChi60, CBDEGZ, ChBDChiC, ChBDChiB, and ChBDChiA1 were 27, 37, 26, 35 and 20%, respectively.

endoglucanase Z (CBDEGZ) from Erwinia chrysanthemi (6) and chitinase B (ChBDChiB) from Serratia marcescens (5). In all the four cases, two or three aromatic residues were exposed to solvent, which indicates that these residues likely interact with carbohydrate substrates. However, the difficulty in determining the positions of these exposed ChBD1 aromatic residues from the amino acid sequence alignment alone has prevented the prediction of which residues bind to crystalline chitin (Fig. 1C). Here, we determined the structure of ChBD1 by nuclear magnetic resonance (NMR) spectroscopy to examine its structural characterization and classification.

Materials and Methods Protein expression and purification The complementary DNA (cDNA) encoding ChBD1 (Ser65Thr135) was obtained from the P. furiosus gene library (PF1234) by polymerase chain reaction and was cloned into the pET32 Ek/LIC vector (Novagen), which includes a thioredoxin-(His)6 tag followed by a PreScission protease (GE Healthcare) cleavage site. Escherichia coli strain BL21(DE3) was transformed with the expression vector. Protein expression and purification was achieved using the same procedures as described previously (1). Briefly, the expressed fusion protein was collected and purified by a Ni2þ-chelating column. Following the removal of thioredoxin by PreScission protease, ChBD1 was purified by anion exchange chromatography and gel-filtration chromatography. The resultant ChBD1 had two additional residues (H2N-Gly-Pro-) derived from the PreScission protease recognition sequence at its amino terminus.

116

Proteins uniformly labelled with 15N were obtained by growing the E. coli cells in an M9 minimal medium containing 0.05% (w/v) 15 NH4Cl as the sole nitrogen source. Uniformly 15N- and 13 C-labelled proteins were obtained in an M9 medium containing 0.05% (w/v) 15NH4Cl and 0.1% (w/v) [13C6]-glucose as the sole sources for nitrogen and carbon, respectively. Fractional labelling for stereospecific assignment of the germinal methyl groups of valine and leucine residues was achieved by growing the bacteria first in nonlabelled M9 medium and then by adding 10% [13C6]-glucose as the labelled carbon source at the time of induction (11). Approximately 10 mg of purified, uniformly 15N-labelled ChBD1 was obtained from 1 l of culture. The protein was dissolved at an approximate concentration of 1.0 mM in a 20 mM 2-morpholinoethanesulfonic acid buffer (pH 6.0) containing 10% D2O for the NMR experiments.

Resonance assignments and structure determination NMR spectra were acquired with a Bruker AV400M, DRX500 or DRX600 spectrometer equipped with a pulsed-field gradient probe or with a DRX800 spectrometer equipped with a cryogenic probe at 30 C. The sequence-specific backbone resonance assignment was obtained from three-dimensional (3D) CBCA(CO)NH, HNCACB, HNCO and HN(CA)CO spectra. Aliphatic side-chain assignment was achieved using mainly 3D C(CO)NH, H(CCO)NH, HBHA(CO)NH and HCCH-total correlation spectroscopy(HCCHTOCSY) spectra. The methyl groups of valine and leucine residues were stereospecifically assigned from two-dimensional (2D) highresolution and constant time 1H-13C-heteronuclear single-quantum correlation (HSQC) spectra using the 10% 13C-labelled sample (11). Tyrosine, phenylalanine and tryptophan side chains were assigned from 2D nuclear Overhauser effect spectroscopy (NOESY), TOCSY, (Hb)Cb(CgCd)Hd and (Hb)Cb(CgCdC")H" spectra (12). All data were processed with the NMRPipe program (13) and analysed with the Sparky program (14).

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

C

Structural studies of chitin-binding domain of a hyperthermophilic chitinase The structures of ChBD1 were calculated by CYANA 2.1 (15) using nuclear Overhauser effect (NOE)-derived distance restraints obtained from 3D 15N-edited and 13C-edited NOESY; dihedral angle restraints predicted on the basis of deviations in the backbone chemical shifts compared with those of random coils using TALOS (16). Hydrogen bonds were predicted from structures that were calculated without any hydrogen bond restraints using a tool implemented in Chimera for identifying possible hydrogen bonds. It was further checked whether the NOE patterns observed around the candidates were typical of the secondary structures. In the final calculation, 16 hydrogen bond restraints were applied as 2.8—3.3 A˚ for N-O pairs and 1.8—2.3 A˚ for H-O pairs. In addition, 49 pairs of  and c dihedral angle restraints in the main chain and 11 1 dihedral angle restraints in aromatic residues, which were estimated from 2D HN(Cg) and HN(COCg) experiments (17), were applied. From 100 target structures, 30 that had the minimum target functions were selected and analysed by PROCHECK-NMR software (18). They showed no distance restraint violation of 40.5 A˚ and no torsion angle restraint violation of 45 .

Mutants were constructed with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA), using the ChBD1 expression vector as a template. The purification procedures for the mutant proteins were the same as those used for the structure determination of wild-type ChBD1. Point mutants were named in the conventional manner. For example, mutants in which Tyr112, Trp113 or Tyr123 were replaced with Ala were designated as ChBD1Y112A, ChBD1W113A or ChBD1Y123A, respectively.

Binding assays of wild-type and mutant ChBD1 for chitin Chitin-binding was analysed at 25 C in a 20 mM phosphate buffer, pH 7.0, at a protein concentration of 40 mM. Chitin (Seikagaku Co.)

Data deposition The atomic coordinates of the NMR structures with the constraints used have been deposited in the Protein Data Bank with accession code 2rts. The resonance assignment has been deposited in the BioMagResBank with accession code 11530.

Results and Discussion Structural determination of ChBD1

A 2D 1H-15N heteronuclear single-quantum correlation (HSQC) spectrum of ChBD1 is shown in Fig. 2, consistent with a 73 amino acid protein containing two extra residues (Gly and Pro) from the PreScission protease cleavage site. Complete backbone assignments for all residues were obtained, except for the N-terminal Gly and the subsequent Pro residues. The structure of ChBD1 was determined according to the procedure described in the ‘Materials and Methods’ section. Table I shows the structural and restraint statistics of the 30 structures that had the lowest target functions. As shown in Fig. 3A, the

Fig. 2 2D 1H-15N HSQC spectrum of ChBD1. The letters with and without ‘sc’ indicate the assignments to the side-chain amino and backbone amide groups, respectively.

117

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

Construction of mutants for the binding assay

was added to a final concentration of 10 mg/ml, and the suspension was incubated at 25 C for 20 h. Each mixture was centrifuged at 25 C for 10 min at 23,000g to separate the supernatant from the substrates with bound protein and the absorbance at 280 nm (A280) of the supernatant was measured to quantify the amount of protein unbound to the insoluble ligands. The amount of bound protein was calculated from the difference between the initial protein concentration and the free protein concentration after binding. The amount of bound wild-type ChBD1 was 2.67 mmol/(g chitin), corresponding to 33% of the total protein.

S. Mine et al. Table I. Experimental restraints and structural statistics.

665 260 185 31 173 16 109 all 883 197 30 0

Percent 79.5 17.7 2.7 0.0

0.212 0.713

The ensemble of 30 structures with the minimum target functions were selected from 100 independent calculations. They exhibited no violations from any distance restraints by more than 0.5 A˚ or from any angle restraints by more than 5 . Among 20 methyl groups of four valines and six leucines, 14 methyl groups were successfully assigned stereospecifically. The methyl groups of Val 10, Leu 15 and Leu 16 failed to be assigned owing to their peak overlaps. a The PROCHECK-NMR program was used to evaluate the quality of structures. The Ramachandran plot itself (one of the output pages of PROCHECK-NMR program) has been supplied as Supplementary Data. b The precision of the atomic coordinates is defined as the average root mean square deviation (RMSD) from the mean structure obtained by averaging the 30 structures.

atomic coordinates have been well defined throughout the protein molecule, with the exception of the N-terminal region (Ser65-Tyr86). This region was confirmed as being flexible in solution by negative or low values of steady state {1H}-15N heteronuclear NOE experiments (Fig. 3C). The root mean square deviations from the averaged structure were calculated as 0.21 A˚ for the backbone and 0.71 A˚ for all heavy atoms of the welldefined region (Pro87-Thr135). ChBD1 had a compact and globular conformation, comprised of three anti-parallel b-strands (Fig. 3B), b1 (Ile100-Tyr103), b2 (Lys106-Ala110) and b3 (Trp126-Glu131). There was no region characteristic of an a-helix. The backbone conformation was very similar to that of known tertiary structures of the proteins of CBM family 5, such as ChBDChi60, ChBDChiB, ChBDChiC and CBDEGZ. However, the aromatic residues responsible for ligand binding were not conserved in the same way as observed in the latter three CBMs, as described below. Chitin-binding site of ChBD1

A unique structural feature of ChBD1 is that three aromatic residues, Tyr112, Trp113 and Tyr123, are exposed and aligned linearly on one side of the molecular surface. These three residues are likely to be involved in chitin binding because aromatic rings arranged on the flat face of proteins are proposed to be stacked on every other pyranose ring of polysaccharides through hydrophobic interactions (21). To confirm the involvement of these aromatic residues 118

Structural comparison with other CBMs

CBM families have been classified into seven groups on the basis of the conservation of protein folds (22). Although members of CBM family 5 have similar folding in their backbones, the positions of the aromatic side-chains that directly interact with substrates are different. To predict ligand specificity more precisely, CBM family 5 has been further classified into three groups based on a comparison of the ligand-binding sites (3). The first group, represented by CBDEGZ, has three exposed aromatic residues, including Trp18, Trp43 and Tyr44, which are aligned linearly and play major roles in the binding to cellulose (Fig. 5). A set of three aromatic residues with these characteristics has been observed in multiple cellulose-binding domains for which the structure has been determined (21, 23—25). Furthermore, all CBDEGZ-group members are likely to contain a relatively flexible loop in the N-terminal region with an aromatic residue, such as Trp18, exposed to solvent. The second group, represented by ChBDChiB and ChBDChiC, has two exposed aromatic residues (Trp479 and Tyr481 in ChBDChiB, Trp59 and Trp60 in ChBDChiC) that interact directly with substrates (Fig. 5). These WW motifs (Trp—Trp, Tyr—Trp or Trp—Tyr) are conserved widely in CBM family 5 (3, 22, 26, 27), suggesting that the second group constitutes the majority of CBMs in this CBM family. However, the CBMs in this group lack a region that corresponds to the loop ranging from Trp13 to Gln22 of CBDEGZ, which contains the exposed aromatic ring of Trp18. When the structure of ChBDChiC was determined, extensive analysis of the sequential alignment of CBM family 5 predicted the existence of a third group of members that could be separated from these two groups (Fig. 1B) [this hypothetical group was described as the fourth group (3)]. However, the classification of this group was ambiguous at that time because none of the structures belonging to this group had been determined. However, our structure of ChBD1, which is classified in this third group, has revealed that it has three exposed aromatic residues, Tyr112, Trp113 and Tyr123. Although the former two residues constitute the WW motif, which is also seen in the first and second groups, the third aromatic residue corresponding to Tyr123, which is

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

Number of experimental restraints NOE distance restraints Total Intra-residual Sequential (j i — j j ¼ 1) Medium-range (j i - j j  4) Long-range (j i - j j44) Hydrogen bonds Dihedral angle restraints Structure statistics Ramachandran plota Residues in most favourable region Residues in additionally allowed region Residues in generously allowed region Residues in disallowed region RMSD of atomic coordinatesb Residues 87—135 Backbone (A˚) Heavy atoms (A˚)

(Tyr112, Trp113 and Tyr123) in the binding interactions, each one was replaced with alanine through site-directed mutagenesis. Figure 4 summarizes the effect of each mutant on binding to chitin powder. The binding affinities of the three mutants decreased to 30—40% of that of wild-type ChBD1. Although the degree of inactivation was not as large as observed for other CBMs, the result shows that the three residues of ChBD1 are at least likely to be involved in the binding activity to chitin. Each of the three residues may act additively to the substrate. However, it has been reported that in two ChBDs of CBM family 5, ChBDChiC and ChBDChiB, only two aromatic residues are exposed and involved in chitin binding (3—5). Therefore, ChBD1 has a structure possessing three exposed aromatic residues for chitin binding, probably in the same way as ChBDChi60.

Structural studies of chitin-binding domain of a hyperthermophilic chitinase

250 Relative bound protein (%)

located in a relatively flexible loop region near the C-terminus, was not observed in the other two groups. The same kinds of sequences are also observed for some other CBMs, such as those of endoglucanase 5A from Bacillus agaradhaerens and endo-1,4-bxylanase from Xylanimicrobium pachnodae. They have aromatic residues at the position corresponding to Tyr123 of ChBD1. Recently, the crystal structure of a chitinase (Chi60 from M. marina, MmChi60) was reported (10). ChBDChi60 has three aromatic residues that are exposed to solvent. Among them, Trp533 and Trp534 correspond to the WW motif like the other ChBDs and CBDs in CBM family 5. The remaining Trp546 matched the third aromatic residue, Tyr123, of our ChBD1 as shown in the result of the conformational alignment (Fig. 1C). The contribution of Trp546 to the chitin-binding activity has not been assayed biochemically, but our mutation and structure alignment results strongly suggest that ChBDChi60 belong to the third group of CBM family 5 in the same way as our ChBD1 of Pf-ChiA (or to the fourth group when chitinase A1 from Bacillus circulans WL-12 is also taken into account) and that its Trp546 is involved in the chitin-binding together with Trp533 and Trp534.

200 150 100 50 0

wild

Y112A

W113A

Y123A

ChBD2

Fig. 4 Assays for binding of the three mutants of ChBD1 and wildtype ChBD2 to insoluble chitin substrate. The binding affinity for each mutant is shown as a value (%) relative to that of wild-type ChBD1.

All members of these three groups have similar backbone conformations, but have distinct characteristics in the most important parts: those involved in substrate recognition. These differences are interesting in terms of the possibility of different interaction mechanisms. The structure of the chitin-binding domain of chitinase A1 from B. circulans WL-12 (ChBDChiA1) should 119

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

Fig. 3 Solution structures of ChBD1. (A) Stereo view of backbone superposition of the final 30 calculated structures of ChBD1. The figure was drawn with Chimera (19). (B) Stereo view of ribbon drawing of the representative structure of ChBD1, which had the lowest target function value. The molecular orientation is the same as in (A) and the three b-strands are depicted in yellow. Three aromatic residues involved in chitin binding are labelled. The figure was drawn with PyMol (20). (C) Plot of {1H}-15N NOE against the amino acid residue number for ChBD1. The data are represented by the peak intensity ratio INOE/Iref, where Iref and INOE were measured in the absence or presence of 1H saturation, respectively.

S. Mine et al.

ChBD1

A

Y112 Y123

D

W60

E

Y44

C

W18

W546

W533

W479

F

ChBDChi60 W534

ChBDChiB Y481

ChBDChiA1 Q679

W687 Fig. 5 Comparison of ChBD1 and five representative CBM family 5 and 12 CBMs: CBDEGZ (Protein Data Bank (PDB) code 1A1W), ChBDChiB (PDB code 1E15), ChBDChiC (PDB code 2D49), ChBDChi60 (PDB code 4HMC) and ChBDChiA1 (PDB code 1ED7), whose sequences are described in Fig. 1C. The root mean square deviation (RMSD) values for the backbone heavy atoms in the secondary structure regions were calculated using MatchMaker tool in Chimera. The RMSD values between ChBD1 and each of ChBDChi60, ChBDChiB, ChBDChiC, ChBDChiA1 and CBDEGZ were 0.897, 0.868, 0.834, 0.864 and 0.928 A˚, respectively. As the structures of ChBD1, ChBDChiC and ChBDChiA1 were determined by NMR, the protons of the aromatic residues are also displayed, whereas those of CBDEGZ, ChBDChiB and ChBDChi60 were determined by Xray crystallography and therefore, no coordinates of protons have been reported in PDB. Some important chitin- and cellulose-binding residues are displayed as stick models. b-Strands are shown in yellow. Graphics were drawn with Chimera.

also be discussed (Fig. 5). Even though ChBDChiA1 has a backbone conformation very similar to that of the above-mentioned CBMs in CBM family 5, it has been classified into CBM family 12. Interestingly, ChBDChiA1 has no exposed aromatic residues as observed in CBM family 5 members. Instead, Trp687 and Gln679 are reported to be likely involved in chitinbinding (28). Although Gln679 exists at a similar position to that of the exposed aromatic residues of CBM family 5 members, Trp687 is located on a different surface and is not fully exposed. Considering that ChBDChiA1 recognizes only insoluble or crystalline chitin substrates, but not soluble chitinous substrates, the conformational variation of binding sites may be related to how strongly and which type of chitin the corresponding ChBDs recognize. Structural comparison with ChBD2

Pf-ChiA contains two chitin-binding domains, ChBD1 and ChBD2 (Fig. 1B). Although both domains have three exposed aromatic residues on the same plane of each protein surface, the binding affinities of these domains may be different because their amino acid sequences and tertiary structures are quite different. Thus, we investigated their binding interactions with 120

chitin. As shown in Fig. 4, the binding affinity of ChBD2 is twice as strong as that of ChBD1. This difference suggests that the functions of the two domains in the context of intact Pf-ChiA are different, even if these CBMs have the same number of aromatic residues for chitin binding. Pf-ChiA possesses two catalytic domains, AD1 and AD2, and the hydrolytic activity of AD2 towards chitin was observed to be twice as high as that of AD1 (2). In addition, ChBD1 and ChBD2 enhanced the hydrolytic efficiency of AD1 and AD2, respectively (2). These results suggest that any concerted action of AD1 and AD2 involving ChBD1 and ChBD2 achieve synergistic hydrolysis of chitin, as described below, as a mechanism of chitin degradation of Pf-ChiA. As the initial step, ChBD2 binds strongly onto the chitin surface, resulting in a secure footing for the ensuing degradation steps by AD2. Then, the strong interaction may facilitate the subsequent hydrolysis reaction by AD2. Next, the binding of ChBD1 may allow AD1 to access the ends of chitin chains, which effectively releases the chitin residues from the chain ends through hydrolysis. Our structural analysis has revealed that ChBD1 of Pf-ChiA possesses a similar backbone conformation to

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

W59

W43

W113

ChBDChiC

CBDEGZ

B

Structural studies of chitin-binding domain of a hyperthermophilic chitinase

that of other CBM family 5 members, but a different binding interface and that CBM family 5 can be classified into at least three groups based on the respective binding interfaces of each member. We have reported the tertiary structures of ChBD1, ChBD2 (9) and AD2 (29, 30) of Pf-ChiA, whose function as a full chitinase against a form of crystalline chitin was recovered by a reconstruction of the gene. Further structural analysis and protein engineering of Pf-ChiA would facilitate the development of biomass exploitation technology.

Supplementary Data Supplementary Data are available at JB Online.

9.

10.

11.

Acknowledgements

Funding Grant-in-Aid for Scientific Research (No. 22570128) to K.U. from the JapanSociety for the Promotion of Sciences.

12.

13.

Conflict of interest None declared.

References 1. Nakamura, T., Ishikawa, K., Hagihara, Y., Oku, T., Nakagawa, A., Inoue, T., Ataka, M., and Uegaki, K. (2005) Crystallization and preliminary X-ray diffraction analysis of a chitin-binding domain of hyperthermophilic chitinase from Pyrococcus furiosus. Acta Crystallogr. Sect. F Struct. Biol. Cryst. Commun. 61, 476—478 2. Oku, T. and Ishikawa, K. (2006) Analysis of the hyperthermophilic chitinase from Pyrococcus furiosus: activity toward crystalline chitin. Biosci. Biotechnol. Biochem. 70, 1696—1701 3. Akagi, K., Watanabe, J., Hara, M., Kezuka, Y., Chikaishi, E., Yamaguchi, T., Akutsu, H., Nonaka, T., Watanabe, T., and Ikegami, T. (2006) Identification of the substrate interaction region of the chitin-binding domain of Streptomyces griseus chitinase C. J. Biochem. 139, 483—493 4. Kezuka, Y., Ohishi, M., Itoh, Y., Watanabe, J., Mitsutomi, M., Watanabe, T., and Nonaka, T. (2006) Structural studies of a two-domain chitinase from Streptomyces griseus HUT6037. J. Mol. Biol. 358, 472—484 5. van Aalten, D.M., Synstad, B., Brurberg, M.B., Hough, E., Riise, B.W., Eijsink, V.G., and Wierenga, R.K. (2000) Structure of a two-domain chitotriosidase from Serratia marcescens at 1.9 A˚ resolution. Proc. Natl Acad. Sci. USA 97, 5842—5847 6. Brun, E., Moriaud, F., Gans, P., Blackledge, M.J., Barras, F., and Marion, D. (1997) Solution structure of the cellulose-binding domain of the endoglucanase Z secreted by Erwinia chrysanthemi. Biochemistry 36, 16074—16086 7. Standley, D.M., Toh, H., and Nakamura, H. (2005) GASH: An improved algorithm for maximizing the number of equivalent residues between two protein structures. BMC Bioinformatics 6, 221 8. Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, H., Valentin, F.,

14. 15.

16.

17.

18.

19.

20. 21.

22.

23.

121

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

The authors thank Ms C. Kageyama (AIST) for technical assistance with gene manipulation and protein purification.

Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J., and Higgins, D.G. (2007) Clustal W and clustal X version 2.0. Bioinformatics 23, 2947—2948 Nakamura, T., Mine, S., Hagihara, Y., Ishikawa, K., Ikegami, T., and Uegaki, K. (2008) Tertiary structure and carbohydrate recognition by the chitin-binding domain of a hyperthermophilic chitinase from Pyrococcus furiosus. J. Mol. Biol. 381, 670—680 Malecki, P.H., Raczynska, J.E., Vorgias, C.E., and Rypniewski, W. (2013) Structure of a complete four-domain chitinase from Moritella marina, a marine psychrophilic bacterium. Acta Crystallogr. D 69, 821—829 Neri, D., Szyperski, T., Otting, G., Senn, H., and Wuthrich, K. (1989) Stereospecific nuclear magnetic resonance assignments of the methyl groups of valine and leucine in the DNA-binding domain of the 434 repressor by biosynthetically directed fractional 13C labeling. Biochemistry 28, 7510—7516 Yamazaki, T., Formankay, J.D., and Kay, L.E. (1993) Two-dimensional NMR experiments for correlating 13Cb and 1H"/d chemical shifts of aromatic residues in 13Clabeled proteins via scalar couplings. J. Am. Chem. Soc. 115, 11054—11055 Delaglio, F., Grzesiek, S., Vuister, G.W., Zhu, G., Pfeifer, J., and Bax, A. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277—293 Goddard, T.D. and Kneller, D.G. (2008) SPARKY 3. University of California, San Francisco Guntert, P., Mumenthaler, C., and Wuthrich, K. (1997) Torsion angle dynamics for NMR structure calculation with the new program DYANA. J. Mol. Biol. 273, 283—298 Shen, Y., Delaglio, F., Cornilescu, G., and Bax, A. (2009) TALOS plus: a hybrid method for predicting protein backbone torsion angles from NMR chemical shifts. J. Biomol. NMR 44, 213—223 Hu, J.S., Grzesiek, S., and Bax, A. (1997) Two-dimensional NMR methods for determining 1 angles of aromatic residues in proteins from three-bond JC’Cg and JNCg couplings. J. Am. Chem. Soc. 119, 1803—1804 Laskowski, R.A., Rullmann, J.A.C., MacArthur, M.W., Kaptein, R., and Thornton, J.M. (1996) AQUA and PROCHECK-NMR: Programs for checking the quality of protein structures solved by NMR. J. Biomol. NMR 8, 477—486 Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., and Ferrin, T.E. (2004) UCSF chimera - A visualization system for exploratory research and analysis. J. Comput. Chem 25, 1605—1612 DeLano, W.L. (2002) The PyMol Molecular Graphics System. DeLano Scientific, Palo Alto, CA, USA Tormo, J., Lamed, R., Chirino, A.J., Morag, E., Bayer, E.A., Shoham, Y., and Steitz, T.A. (1996) Crystal structure of a bacterial family-III cellulose-binding domain: a general mechanism for attachment to cellulose. EMBO J. 15, 5739—5751 Boraston, A.B., Bolam, D.N., Gilbert, H.J., and Davies, G.J. (2004) Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochem. J. 382, 769—781 Kraulis, J., Clore, G.M., Nilges, M., Jones, T.A., Pettersson, G., Knowles, J., and Gronenborn, A.M. (1989) Determination of the three-dimensional solution structure of the C-terminal domain of cellobiohydrolase I from Trichoderma reesei. A study using nuclear

S. Mine et al.

122

27. Uni, F., Lee, S., Yatsunami, R., Fukui, T., and Nakamura, S. (2012) Mutational analysis of a CBM family 5 chitin-binding domain of an alkaline chitinase from Bacillus sp. J813. Biosci. Biotechnol. Biochem. 76, 530—535 28. Hara, M., Sugimoto, H., Uemura, M., Akagi, K., Suzuki, K., Ikegami, T., and Watanabe, T. (2013) Involvement of Gln679, in addition to Trp687, in chitin-binding activity of the chitin-binding domain of chitinase A1 from Bacillus circulans WL-12. J. Biochem. 154, 185—193 29. Nakamura, T., Mine, S., Hagihara, Y., Ishikawa, K., and Uegaki, K. (2007) Structure of the catalytic domain of the hyperthermophilic chitinase from Pyrococcus furiosus. Acta Crystallogr. F 63, 7—11 30. Tsuji, H., Nishimura, S., Inui, T., Kado, Y., Ishikawa, K., Nakamura, T., and Uegaki, K. (2010) Kinetic and crystallographic analyses of the catalytic domain of chitinase from Pyrococcus furiosus - the role of conserved residues in the active site. FEBS J. 277, 2683—2695

Downloaded from http://jb.oxfordjournals.org/ at Michigan State University on February 27, 2015

magnetic resonance and hybrid distance geometry-dynamical simulated annealing. Biochemistry 28, 7241—7257 24. Mattinen, M.L., Kontteli, M., Kerovuo, J., Linder, M., Annila, A., Lindeberg, G., Reinikainen, T., and Drakenberg, T. (1997) Three-dimensional structures of three engineered cellulose-binding domains of cellobiohydrolase I from Trichoderma reesei. Protein Sci. 6, 294—303 25. Xu, G.Y., Ong, E., Gilkes, N.R., Kilburn, D.G., Muhandiram, D.R., Harris-Brandts, M., Carver, J.P., Kay, L.E., and Harvey, T.S. (1995) Solution structure of a cellulose-binding domain from Cellulomonas fimi by nuclear magnetic resonance spectroscopy. Biochemistry 34, 6993—7009 26. Itoh, Y., Watanabe, J., Fukada, H., Mizuno, R., Kezuka, Y., Nonaka, T., and Watanabe, T. (2006) Importance of Trp59 and Trp60 in chitin-binding, hydrolytic, and antifungal activities of Streptomyces griseus chitinase C. Appl. Microbiol. Biotechnol. 72, 1176—1184