2015 Volume 62 Issue 4 Pages 153-157
Chitin oligosaccharide deacetylase (COD) is an enzyme that generates β-N-acetyl-D-glucosaminyl-(1,4)-D-glucosamine from N,N′-diacetylchitobiose. COD has been found only in species of Vibrio bacteria. However, a homology search in the sequence databases revealed that Shewanella woodyi ATCC51908 also encodes a protein with sequence similarity to COD. Analysis of the deduced amino acid sequence of the S. woodyi COD (Sw-COD) confirmed that the protein contains the same motifs as CODs from Vibrio bacteria. The COD-encoding gene, which includes a signal sequence, was cloned from the chromosomal DNA of S. woodyi, the expression plasmid containing this gene was constructed, and then the plasmid was introduced into Escherichia coli HMS174(DE3) cells. The recombinant Sw-COD (Sw-rCOD) was produced in culture medium with the aid of the signal peptide and purified from culture supernatant. The properties of Sw-rCOD (substrate specificity, optimal pH, etc.) were similar to those of the CODs from Vibrio bacteria.
COD, chitin oligosaccharide deacetylase; GlcNAc-GlcN, β-N-acetyl-D-glucosaminyl-(1,4)-D-glucosamine; CE-4, carbohydrate esterase family 4; PDD, polysaccharide deacetylase domain; CBD, carbohydrate-binding domain.
Chitin oligosaccharide deacetylase (COD, EC 3.5.1.105) is an enzyme that hydrolyzes the acetamide bond of the second N-acetyl-D-glucosamine (GlcNAc) residue from the non-reducing end of chitin oligosaccharides. Historically, CODs have been identified and characterized from the following species of Vibrio bacteria: Vibrio alginolyticus H-8,1) Vibrio cholerae EI Tor N16961,2) Vibrio parahaemolyticus KN1699,3) Vibrio sp. SN184,4) and Vibrio harveyi ATCC BAA-1116.5) Each of these enzymes show highest activity against the N,N′-diacetylchitobiose [(GlcNAc)2] in chitin oligosaccharides. Previously, we reported that β-N-acetyl-D-glucosaminyl-(1,4)-D-glucosamine (GlcNAc-GlcN), which is generated from (GlcNAc)2 by COD, functions in several COD-producing Vibrio bacteria as both an inducer of chitinase production and a chemoattractant.6) 7) These findings indicate that COD is an important enzyme for chitin decomposition by these bacteria. Until now, COD has been reported only in members of the genus Vibrio. However, we postulated that CODs should be present in other genera of bacteria that utilize chitin as a nutrient source. We therefore searched the genomes of newly sequenced bacteria for predicted proteins with homology to COD enzymes. In the present paper, we report about COD derived from a non-Vibrio bacterial species.
We utilized the blast-p program (DNA Data Bank of Japan, DDBJ BLAST version 2.2.24) to identify candidate novel COD homologs. Our search revealed that Shewanella woodyi ATCC51908 harbors the gene of protein (GenBank Protein ID ACA84860) of which primary structure is highly homologous to that deduced from nucleotide sequences of the COD gene of V. harveyi ATCC BAA-1116 (GenBank Gen ID 5554180). A global homology search by GENETYX MAC program (Ver. 16.0.9) demonstrated approximately 66% identity between the amino acid sequence of the putative COD of S. woodyi and those of the above-mentioned Vibrio CODs. Motif analysis using the Pfam database (version 27.0, http://pfam.xfam.org/) confirmed that the S. woodyi protein, like the Vibrio homologs, contains one polysaccharide deacetylase motif (PDM, amino acid no. 23-127) and two family 12 carbohydrate-binding modules (CBMs; amino acid no. 329-372 and 380-405). Recently, the three-dimensional structures of the CODs of V. cholerae O1 str. NHCC-010F (PDB ID: 4NY2)8) and V. parahaemolyticus KN1699 (PDB ID: 3WX7)9) were determined. These structures confirmed that these proteins consist of one polysaccharide deacetylase domain (PDD) and two carbohydrate-binding domains (CBDs). Our investigation9) of the domain function of strain KN1699 COD confirmed that the PDD alone can function as a catalyst, and that the CBDs possess strong chitin-binding activity. According to the motif analysis, it is speculated that the S. woodyi protein also consists of these domains. Two amino acid residues (His and Asp) (inverted triangles a in Fig. 1) that serve as general acid-base catalysts and three amino acid residues (His-His-Asp triad) (inverted triangles b in Fig. 1) that coordinate the active site zinc ion are conserved in PDD region of Vibrio CODs. Moreover, each of the two CBDs contain six aromatic amino acids (inverted triangles c in Fig. 1) that are predicted to be involved in chitin binding.8) 9) These amino acids exist at corresponding positions in the primary structures of each of the bacterial COD homologs, including the S. woodyi protein (Fig. 1). Thus, the predicted protein from S. woodyi harbors all of the structural elements associated with COD activity in the Vibrio COD enzymes. These facts allow us to think that this protein of S. woodyi is obviously COD. This bacterium scarcely produced COD (data not shown). We therefore investigated the functional properties of S. woodyi COD in comparison to those of the Vibrio homologs using recombinant S. woodyi COD isolated following overproduction in Escherichia coli host cells.
Alignment of amino acid sequences of a putative COD from Shewanella woodyi ATCC51908 and CODs from several Vibrio strains.
These sequences correspond to the putative or known CODs from the following bacterial strains: 1, Shewanella woodyi ATCC51908; 2, Vibrio alginolyticus H-8 (GenBank Protein ID BAB21759); 3, Vibrio cholerae EI Tor N16961 (GenBank Protein ID AAF94439); 4, Vibrio parahaemolyticus KN1699 (GenBank Protein ID BAG70715); 5, Vibrio sp. SN184 (GenBank Protein ID BAG82921); and 6, Vibrio harveyi ATCC BAA-1116. Amino acids highlighted in black and gray are, respectively, identical and highly conserved (> 50%) among these proteins. Boxed N-terminal sequences represent signal peptides. Double-headed arrows indicate the extent of the carbohydrate-binding modules. Lower-case letters (a, b, c) with inverted triangles indicate structural elements as described in the text.
S. woodyi ATCC5190810) was obtained from the American Type Culture Collection (Manassas, USA). Bacteria belonging to the genus Shewanella were found in areas of seawater and freshwater.11) The strain we used was isolated from deep sea.10) Strain ATCC51908 harbors several chitinase genes, suggesting that this strain is bacterium that utilizes chitin as a nutrient source.
After the strain ATCC51908 cells were grown for 16 h at 28°C on an agar plate containing 1.87% (w/v) Marine broth 2216 (Becton, Dickinson and Company, Franklin Lakes, USA), the cells were harvested from the plate, and chromosomal DNA was isolated using an Isoplant II kit (Wako Pure Chemical Ind., Osaka, Japan). The target gene (GenBank Gene ID 6114798), along with flanking sequences (14 bp upstream, containing the predicted Shine-Dalgarno sequence), was amplified by PCR (Ex Taq DNA polymerase, 1 × Ex Taq buffer, dNTP mixture; Takara Bio Inc., Shiga, Japan). The PCR reaction employed 760 ng of the chromosomal DNA as template and 10 pmol of the following primers: 5′-GAATTCACAAGGAAATAACAATGAAATTAAC-3′ (forward primer) and 5′-CTCGAGTTAGTTTGCTAAGAAC-3′ (reverse primer). (underlined letters designate the EcoRI site of the forward primer and the XhoI site of the reverse primer; letters in italics designate added non-complimentary nucleotides; bold letters designate start codon, which was changed from GTG to ATG). The PCR consisted of 30 amplification cycles with the following conditions: denaturation at 94°C for 1 min; annealing at 55°C for 1 min; and elongation at 72°C for 1.5 min. All PCR experiments were conducted using a C1000 Thermal Cycler (Bio-Rad Laboratories, Hercules, USA). PCR products were isolated from agarose gel slices using a QIAquick gel extraction kit (Qiagen GmbH, Hilden, Germany) and ligated into pGEM-T Easy Vector (Promega Corporation, Madison, USA) by TA cloning to yield plasmid pGEM-SwCOD. E. coli DH5α (Takara Bio Inc.) was transformed with pGEM-SwCOD, and the plasmid was amplified by cultivating the transformed cells for 16 h with shaking (200 rpm) at 37°C in 10 mL of Luria-Bertani (LB) medium supplemented with 50 µg/mL ampicillin. After harvesting the E. coli cells by centrifugation (3,300 × G for 3 min at 4°C), plasmid was isolated from the cells using a Miniprep kit (Qiagen). The target gene was excised from the plasmid using EcoRI (Toyobo Co., Ltd., Osaka, Japan) and XhoI (Takara Bio Inc.), and ligated into EcoRI/XhoI-double digested pET-21(＋) vector (Merck KGaA, Darmstadt, Germany) to yield the plasmid (pET-SwCOD) for the production of the recombinant COD of S. woodyi ATCC51908 (Sw-rCOD). Nucleotide sequence analysis of the target gene in pET-SwCOD was performed using a Big Dye Terminator V3.1 Cycle Sequencing kit and an ABI 3130xl DNA sequencer (Applied Biosystems, Santa Clara, USA).
E. coli HMS174(DE3) cells were transformed with pET-SwCOD and one resulting transformant was used for the production of Sw-rCOD. The transformant cells were cultivated with shaking (160 rpm) at 30°C in 1 L of LB medium supplemented with 50 µg/mL ampicillin until the OD600 reached 0.45. Isopropyl-β-thiogalactopyranoside then was added into the culture broth at a final concentration of 0.5 mM and the culture was incubated for an additional 20 h under the same conditions. After removing the cells from 1 L of culture broth by centrifugation (3,000 × G for 10 min at 20°C), (NH4)2SO4 was added to the obtained culture supernatant to 80% saturation in order to precipitate the proteins. The resulting precipitate was collected by centrifugation (3,000 × G for 15 min at 4°C), dissolved in a small amount of 20 mM sodium phosphate buffer (pH 7.0), and then dialyzed against the same buffer to yield the crude enzyme solution. This solution was loaded onto a DEAE-Sepharose Fast Flow resin (GE Healthcare, Buckinghamshire, England) column (size: φ2.5 × 15 cm) pre-equilibrated with the same buffer. Proteins were eluted from the column with a linear gradient of 0 to 0.8 M NaCl in the same buffer (total volume: 400 mL). The eluate containing the protein that show COD activity was concentrated to 1 mL using an Amicon Ultra-15 Centrifugal Filter Device (Merck KGaA) and loaded onto a Bio-Gel P-100 Fine resin (Bio-Rad Laboratories) column (size: φ1.5 × 90 cm) pre-equilibrated with 20 mM sodium phosphate buffer (pH 7.0). The eluate containing the protein that show COD activity was re-equilibrated to 20 mM sodium phosphate buffer (pH 7.0) containing 0.7 M (NH4)2SO4 by ultrafiltration using a Stirred Cell Model 8400 equipped with Ultrafiltration Discs YM-10 (Merck KGaA). Target protein was further purified by column chromatography using a Phenyl Sepharose High Performance resin (GE Healthcare) column (size: φ1.5 × 8 cm) pre-equilibrated with 20 mM sodium phosphate buffer (pH 7.0) containing 0.7 M (NH4)2SO4. Proteins were eluted from the column with a linear gradient of 0.7 to 0 M (NH4)2SO4 in 20 mM sodium phosphate buffer (pH 7.0) (total volume: 400 mL). The eluate containing the protein that show COD activity was dialyzed against 20 mM sodium phosphate buffer (pH 7.0) and stored at 4°C. The purified protein ran as a single band with a molecular mass of approximately 48.6 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Fig. 2). Sw-rCOD was purified 7.2-fold with 17.2% recovery from crude enzyme solution. The specific activity of the purified enzyme against (GlcNAc)2 was 7.8 unit (U)/mg of protein. Assays of COD activity were conducted according to the colorimetric method reported by Dishe and Borenfreund12) using (GlcNAc)2 as a substrate. The reaction was performed at 37°C in 1 mL of 20 mM sodium phosphate buffer (pH 7.0) containing enzyme and 1.0 mM substrate (standard assay condition). One unit (U) of COD activity was defined as the amount of enzyme required to produce 1 µmol of D-glucosamine (GlcN) residues per minute under these assay conditions. Protein concentrations in enzyme solutions were determined by Lowry’s method using bovine serum albumin (Thermo Fisher Scientific, Waltham, USA) as a standard.13) The N-terminal amino acid sequence of the recombinant protein (obtained using a Procis 492 cLC protein sequencer; Life Technologies, Carlsbad, USA) was determined as SDKTIYLTFDDGPMNATPALIDTL, indicating that a signal peptide consisting of 22 amino acid residues (MKLTTLTCALSLALASISYAQA) was cleaved from the N-terminus when the expressed protein crossed the E. coli cytoplasmic membrane (Fig. 1).
SDS-PAGE analysis of purified Sw-rCOD.
Proteins in the gel were stained with Coomassie Brilliant Blue R250 (Tokyo Chemical Ind., Tokyo, Japan). Lanes 1 and 2 contain molecular mass standards (SIMASIMA Unstained Low Range Ladder, Cosmo Bio Co., Ltd., Tokyo, Japan) and purified Sw-rCOD, respectively.
We investigated the optimal reaction conditions of Sw-rCOD using (GlcNAc)2 as a substrate. This enzyme showed the highest activity around pH 7 when assayed at 37°C, and at 37°C when assayed at pH 7.0 (Fig. 3). These data indicated that the optimal reaction pH and temperature of Sw-rCOD are similar to those of the Vibrio CODs. Next, we investigated the substrate specificity of Sw-rCOD under the standard assay conditions using chitin oligosaccharides of various polymerization degrees. As shown in Table 1, Sw-rCOD showed activity not only for (GlcNAc)2 but also some still lower activity for N,N′,N′′-triacetylchitotriose (GlcNAc)3. Although very low, this enzyme exhibited activity against N,N′,N′′,N′′′-tetraacetylchitotetraose (GlcNAc)4 too. In this assay condition, the activity of Sw-rCOD against GlcNAc and N,N′,N′′,N′′′,N′′′′-pentaacetylchitopentaose (GlcNAc)5 was not detected. Such a specificity has also been observed in the CODs from V. parahaemolyticus KN1699,3) Vibrio sp. SN184,4) and V. harveyi ATCC BAA-1116.5) Treatment of the compound produced from (GlcNAc)2 by Sw-rCOD with β-N-acetylhexosaminidase gave GlcNAc and GlcN (Supplemental Fig. 1; See J. Appl. Glycosci. Web site), indicating that this COD hydrolyzes acetamide bond of reducing end GlcNAc residue of (GlcNAc)2.4) 14) These facts indicate that there are no remarkable differences in the properties of Sw-rCOD compared to those of the Vibrio CODs.
Optimal reaction conditions for Sw-rCOD.
(A) Optimal reaction pH. An assay mixture (100 µL) containing 1.25 µg Sw-rCOD and 10 mM (GlcNAc)2 was incubated at 37°C for 20 min. In this experiment, the following buffers were used: pH 3－6, 20 mM sodium citrate buffer (◯); pH 6－8, 20 mM sodium phosphate buffer (□); pH 8－11, 20 mM sodium borate buffer (△). After 20 min, hydrolysis products (GlcNAc-GlcN) in the reaction mixture were separated on thin-layer chromatography plates [Silica Gel 60, 0.25 mm, Merck KGaA; Mobile phase solvent, n-buthanol/methanol/16% aqueous ammonia 5/4/3 (v/v/v)], and were visualized by soaking the plate in an aqueous solution containing 2.4% (w/v) phosphomolybdic acid, 5% (v/v) H2SO4, and 1.5% (v/v) H3PO4, followed by heating. Then, GlcNAc-GlcN on the plate was quantified using Gel DocTM XR Plus equipped with Image LabTM ver. 2.0 (Bio-Rad Laboratories). Enzyme activity was expressed as the percentage of the activity at pH 7.0, which was defined as 100%. (B) Optimal reaction temperature. The enzyme reaction was assayed using the standard assay mixture containing 12.5 µg of Sw-rCOD. The enzyme activity was evaluated by the colorimetric method. Enzyme activity was expressed as the percentage of the activity at 37°C, which was defined as 100%.
Substrate specificity of CODs.
Vp-rCOD, Vsp-rCOD, and Vh-rCOD correspond to the recombinant CODs of V. parahaemolyticus KN1699, Vibrio sp. SN184, and V. harveyi ATCC BAA-1116, respectively. n.d., no detectable activity. a The reactions were performed under the standard assay conditions using GlcNAc and chitin oligosaccharides, which were purchased from Seikagaku Biobusiness (Tokyo, Japan), as the substrates. The values of U/mg of protein (mean ± SEM) were determined by three independent experiments. b These data are derived from our previous papers.3) 4) 5)
To perform a kinetic study of the Sw-rCOD reaction, 8 mL of 20 mM sodium phosphate buffer (pH 7.0) containing 10.24 µg of Sw-rCOD and (GlcNAc)2 (1.0, 2.0, 3.0, 4.0, or 5.0 mM) was incubated at 37°C. Aliquots (1 mL/time point) were withdrawn from each reaction mixture every 5 min for 30 min, and were heated at 95°C for 30 min in a hot dry bath to stop the enzymatic reaction. Heat-denatured enzyme was used in the control reaction mixture. Enzyme activity was determined according to the above-mentioned colorimetric method. The values of Vmax and Km were obtained from double-reciprocal plots of the reaction curves. The molecular mass of Sw-rCOD (44.0 kDa) was calculated from deduced amino acid sequence. Based on the results from three independent experiments, the parameter values (mean ± SEM) were determined as follows: Vmax, 15.7 ± 0.46 µmol・min-1・mg-1 of protein; Km, 0.67 ± 0.07 mM; kcat, 11.5 s-1; and kcat / Km, 17.2 mM-1・s-1. Recently, we reported the kinetic parameters of recombinant V. parahaemolyticus KN1699 COD (Vp-rCOD).9) Comparison of the parameters indicate that the Vmax value of Sw-rCOD is 2.5-fold lower than that of Vp-rCOD; the Km value of Sw-rCOD is 2.8-fold higher than that of Vp-rCOD; the kcat value of Sw-rCOD is 2.6-fold lower than that of Vp-rCOD; and the kcat / Km value of Sw-rCOD is 7.1-fold lower than that of Vp-rCOD. These data indicate that the catalytic function of Sw-rCOD is inferior to that of Vp-rCOD. Amino acid sequence identity between Sw-rCOD and Vp-rCOD is 66.9%. We believe that the difference in catalytic efficiency between these CODs reflects the differences in their primary structures.
In conclusion, although there were differences in catalytic efficiency, S. woodyi COD and Vibrio CODs shared similar properties. In future work, we plan to identify and characterize CODs from bacteria of various genera.
We thank Saori Yamaoka, Kyoko Teramoto, and Tomoyo Kiuchi of the General Research Institute of the College of Bioresource Sciences, Nihon University, for their technical support. This research was supported by The Science Research Promotion Funds from the Promotion and Mutual Aid Corporation for Private Schools of Japan, and by the College of Bioresource Sciences, Nihon University.
