Characterization of a GH Family 43 β-Xylosidase Having a Novel Carbohydrate-binding Module from Paenibacillus xylaniclasticus Strain TW1

Abstract

Paenibacillus xylaniclasticus strain TW1, a gram-positive facultative anaerobic bacterium, was isolated as a xylanolytic microorganism from the wastes of a pineapple processing factory. A gene encoding one of its xylanolytic enzymes, a β-xylosidase, was cloned and sequenced. Sequence analysis revealed that this β-xylosidase, named PxXyl43A, was composed of a glycoside hydrolase (GH) family 43 subfamily 12 catalytic module and an unknown function module (UM). The full-length PxXyl43A (PxXyl43A) was heterologously expressed in Escherichia coli and purified. Recombinant PxXyl43A exhibited hydrolysis activity against both p-nitrophenyl-β-D-xylopyranoside (pNPX) and p-nitrophenyl-α-L-arabinofuranoside at specific activities of 250 and 310 mU/mg, respectively. The optimal reaction pH and temperature for pNPX hydrolysis were 7.1 and 54 ˚C, respectively. At pH 7.0 and 54 ˚C, the K_m and k_cat for pNPX were 1.2 mM and 2.8 ± 0.15 s^-1, respectively. It was also discovered that the recombinant unknown function module of PxXyl43A (PxXyl43A-UM) could bind to insoluble xylans like birchwood xylan and oat spelt xylan, whereas it did not bind to cellulosic substrates such as ball-milled cellulose, carboxymethyl cellulose or lichenan. The PxXyl43A-UM's binding constant value K_a for oat spelt xylan was 2.0 × 10^-5 M^-1. These results suggest that PxXyl43A possesses a novel carbohydrate-binding module, named as CBM91, specific for xylan-containing polysaccharides.

Abbreviations

BMC, ball-milled cellulose; BSA, bovine serum albumin; BWX, birch wood xylan; CBM, carbohydrate-binding module; CMC, carboxymethyl cellulose; GH, glycoside hydrolase; OSX, oat spelt xylan; pNP, p-nitrophenol; pNPAf, pNP-α-L-arabinofranoside; pNPG, pNP-β-D-glucopyranoside; pNPGal, pNP-β-D-galactopyranoside; pNPX, pNP-β-D-xylopyranoside; UM, unknown function module; X2, xylobiose; X3, xylotriose; X4, xylotetraose; X5, xylopentaose; X6, xylohexaose.

INTRODUCTION

Lignocellulose is one of the most abundant polysaccharide biomass as a carbon source for sustainable energy production. For biofuels production, the rigid lignocellulose, which consists of cellulose, hemicellulose and lignin, is required to be degraded into fermentable monosaccharides. Lignocellulose degradation has been accomplished by variety of microbial enzymes, such as, cellulases, hemicellulases, β-glucosidases and β-xylosidases. Among them, β-xylosidase is essential for complete digestion of hemicellulose, since it cleaves the β-glycoside bonds of xylan unit.¹⁾ It is known that some of the β-xylosidases, which belong to glycoside hydrolase (GH) family 43, possess carbohydrate-binding module (CBM), that is non-catalytic sugar binding modules to facilitate the enzyme catalytic activity. So far, xylan-binding CBMs appended to GH family 43 β-xylosidases were identified and classified into CBM families 6, 13, and 35.²⁾³⁾⁴⁾

Paenibacillus xylaniclasticus strain TW1, a gram-positive facultative anaerobic bacterium, was isolated as a xylanolytic bacterium from sludge in an anaerobic digester fed with pineapple wastes in Thailand.⁵⁾ The partial genome sequencing of the TW1 strain was performed to show that this strain would possess many xylanolytic enzymes. The TW1 strain has been shown to produce many xylanolytic enzymes and extracellular multienzyme complexes when incubated with 0.5% of birch wood xylan (BWX) and secretes cellulose-binding proteins when incubated with 0.5% of corn hull.⁶⁾ Thus, the TW1 strain is a promising bacterium for degradation and saccharification of lignocellulosic biomass. The TW1 strain showed especially high xylanolytic activity, since it could grow on oat spelt xylan (OSX) as the sole carbon source.⁶⁾ However, the various xylanolytic enzymes of the TW1 strain have yet to be identified and characterized. In this study, we focused on one of the putative β-xylosidase genes from the TW1 strain, named PxXyl43A, because this gene contains an unknown function module at the C-terminus instead of a typical CBM. We investigated the xylanolytic activities of PxXyl43A as well as the polysaccharide-binding activity of the unknown function module using purified recombinant proteins. Here, we report a novel CBM derived from P. xylaniclasticus.

MATERIALS AND METHODS

Strain, plasmid and media. P. xylaniclasticus strain TW1 was isolated previously from the wastes of a pineapple processing factory in Thailand.⁵⁾ The Escherichia coli strains αINV (Invitrogen, Thermo Fisher Scientific, Inc., Waltham, MA, USA) and ME9806 (iVEC3) (National BioResource Project (NBRP), Mishima, Japan) were used as cloning hosts while E. coli JM109 and BL21(DE3)(TOYOBO CO., LTD., Japan) were used as protein expression hosts.⁷⁾ The pCR2.1 plasmid (Invitrogen) was used for cloning, and the pQE30 plasmid (QIAGEN Benelux B.V., Venlo, Netherland) and pET16b plasmid (Novagen Inc., Madison, WI, USA) were used for expression of recombinant His-tagged proteins. Transformed E. coli was cultivated in LB liguid medium supplemented with ampicillin (50 µg/mL). The recombinant proteins of the full length of PxXyl43A (PxXyl43A) and the unknown function module at the C-terminus of PxXyl43A (PxXyl43A-UM) were expressed using the plasmids pET16b-PxXyl43A and pQE30-PxXyl43A-UM, respectively.

Cloning of the full length of PxXyl43A and the unknown function module at the C-terminal. A DNA fragment (1,585 bp) including the region encoding PxXyl43A (1,560 bp) was obtained from genomic DNA of P. xylaniclasticus strain TW1 by PCR using primers #1 and #2 (Table 1). A DNA fragment of PxXyl43A-UM (684 bp) was amplified from genomic DNA of P. xylaniclasticus strain TW1 using primers #7 and #8. The linear pET16b for iVEC cloning was amplified using primers #5 and #6. The DNA fragment of PxXyl43A-UM was inserted into pCR2.1. After digestion with HindIII and BamHI, the DNA fragment was ligated into the corresponding site of pQE30. The DNA fragment including the region encoding PxXyl43A was amplified again using primers #3 and #4 and inserted into pET16b by in vivo cloning using E. coli strain ME9806 (iVEC3).⁷⁾ Each of the inserted DNA fragments was sequenced to confirm the absence of mutations.

Table 1. List of primers.

1	AACAGTATAGCAGTTAAATACGAAA
2	TTAGGACAATCCTCATTAAGCTCCG
3	CGGCCATATCGAAGGTCGTCATATGAACAGTATAGCAGTTAAATA
4	GCCCCAAGGGGTTATGCTAGTTAGGACAATCCTCATTAAGCTCCG
5	CATATGACGACCTTCGATATGGC
6	TAACTAGCATAACCCCTTGGGG
7	GGATCCATGAATGTCGAAATTCTCGCAGGA
8	AAGCTTTCAAGACAGCGGTGTATAGTCCAG

Expression and purification of recombinant proteins of PxXyl43A. Transformed E. coli cells having pET16b-PxXyl43A or pQE30-PxXyl43A-UM were grown overnight at 37 ˚C in LB liguid medium supplemented with ampicillin (50 µg/mL). When the optical density of the cells at 600 nm reached 0.4, the lac operon trigger isopropyl-1-thio-β-D-galactopyranoside was added to a final concentration of 0.5 mM. After several hours of incubation, cells were collected by centrifugation at 12,000 rpm and disrupted with sonication in 50 mM sodium phosphate buffer (pH 7.0). The recombinant proteins were purified from the cell-free extracts with the Profinia system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) according to the manufacturer's instructions. The purified proteins were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and the protein concentration was determined using a Protein Assay kit (Bio-Rad Laboratories) with bovine serum albumin (BSA) as the standard.

Substrate specificity assays of PxXyl43A. The xylosidase activity of PxXyl43A was measured by incubating the protein in the presence of 0.25 mM p-nitrophenyl-β-D-xylopyranoside (pNPX) in 50 mM sodium phosphate buffer (pH 7.0) at 54 ˚C for 10 min unless otherwise stated. The amount of p-nitrophenol released from pNPX was measured by the absorbance at 420 nm. All enzyme assays were repeated at least three times to confirm reproducibility. pNP-β-D-glucopyranoside (pNPG), pNP-α-L-arabinofranoside (pNPAf), and pNP-β-D-galactopyranoside (pNPGal) were used to determine the substrate specificity of PxXyl43A. One unit (U) of enzyme activity for pNP-glycoside was defined as the enzyme amount required to liberate 1 µmol of p-nitrophenol per min. OSX was used as insoluble xylan at 1.5% (w/v). After enzymatic reactions, reducing sugars from OSX were quantified using the 3,5-dinitrosalicylic acid method.⁸⁾

Determination of optimal pH and temperature and thermostability of PxXyl43A. To determine the optimal pH for β-xylosidase activity, the recombinant PxXyl43A solution was incubated for 5 min at different pHs (pH 3.0 to pH 10) using Britton-Robinson's buffer at 54 ˚C in the presence of 0.25 mM pNPX as a substrate. For optimal temperature determination, PxXyl43A was incubated at different temperatures (15 to 75 ˚C) for 5 min in 50 mM potassium phosphate buffer (pH 7.0) in the presence of 0.25 mM pNPX as a substrate. For thermostability assay, PxXyl43A were incubated at different temperatures (40 to 57 ˚C) for 2h in 50 mM potassium phosphate buffer (pH 7.0), and then the residual β-xylosidase activity were determined. All enzyme assays were repeated at least three times to confirm reproducibility.

Kinetic analysis of β-xylosidase activity of PxXyl43A for pNPX. To determine the reaction kinetics of PxXyl43A for pNPX, the reaction was performed at 54 ˚C in 50 mM potassium phosphate buffer (pH 7.0) for 1.5 min in the presence of different concentrations of pNPX (1.8 to 9.0 mM).

Assessment of the inhibitory effect of xylose on β-xylosidase activity of PxXyl43A. The inhibitory effect of excess xylose against the β-xylosidase activity of PxXyl43A was assessed using 0.25 mM pNPX as a substrate under the presence of different concentrations of xylose (10 to 200 mM) at 54 ˚C for 5 min.

Analysis of hydrolysis product of xylooligosaccharides. Xylooligosaccharides with different molecular weights, xylobiose (X2) (Wako), xylotriose (X3) (Wako), xylotetraose (X4) (Megazyme), xylopentaose (X5) (Megazyme) and xylohexaose (X6) (Megazyme) (10 µg of each), were incubated with 0.1 U of purified PxXyl43A at 54 ˚C for 1 h in 10 µL of 50 mM potassium phosphate buffer (pH 7.0). The reaction products were analyzed by thin-layer chromatography (TLC) on a DC-Fertigplatten SIL G-25 plate (Macherey-Nadgel) developed with a 1-butanol/acetic acid/water (1:2:1, v/v/v) solvent mixture. The products were visualized by soaking the plate in methanol-sulfuric acid solution (5% sulfuric acid in methanol) and baking at 130 ˚C for several minutes.

Macroarray assay for PxXyl43A-UM binding to polysaccharides. A macroarray assay of the recombinant protein of the unknown function module (PxXyl43A-UM) was performed using polysaccharides, including BWX, OSX, ball-milled cellulose (BMC), carboxymethyl cellulose (CMC) and lichenan as the substrates, according to previously reported procedures.⁹⁾¹⁰⁾ Aliquots (1 μL of 0.2-10 μg/µL) of each polysaccharide were applied to an untreated nitrocellulose sheet (MS-Millipore membrane filters) (Millipore Co., Burlington, MA, USA). The nitrocellulose sheet was left at room temperature for 30 min to dry. After blocking with 5% (w/v) skim milk in 5 × phosphate buffer (PBS) for 1 h, the sheet was incubated with PxXyl43A-UM (12.5 µg/mL) in 5 × PBS for 1 h. After extensively washing with 5 × PBS, the nitrocellulose sheet was incubated at room temperature with a 1,000-fold dilution of anti-His horseradish peroxidase (HRP)-conjugated antibody (Cat#1014981) (QIAGEN) in 5 × PBS for 1 h. The sheet was washed with distilled water and incubated with freshly prepared HRP substrate (0.05% 3,3′-diaminobenzidine tetrahydrochloride and 0.005% H₂O₂) to detect PxXyl43A-UM bound to each polysaccharide. The reaction was stopped by washing the nitrocellulose sheet with distilled water.

Binding assay of PxXyl43A-UM for polysaccharides. OSX, BWX, BMC, and lichenan were used as substrates, after washed three times with 50 mM potassium phosphate buffer (pH 7.0) to remove soluble saccharides. The protein solutions (0.4 mg/mL) of PxXyl43A-UM or BSA as a control were incubated with 5 mg of each substrate for 10 min at room temperature in 50 mM potassium phosphate buffer (pH 7.0). After centrifugation (15,000 rpm, 5 min, 4 ˚C), the supernatant was collected as the unbound proteins, and the precipitate rinsed with 50 mM potassium phosphate buffer (pH 7.0) as the bound proteins. The unbound and bound proteins were denatured with SDS sample buffer containing 2% SDS and 1% β-mercaptoethanol and then analyzed by SDS-PAGE.

Determination of the binding affinity parameters of PxXyl43A-UM for OSX. Different concentrations (1.6 µM-16 µM) of PxXyl43A-UM were added to 20 mL of 3% OSX and incubated for 5 min on ice with gentle mixing. The solutions were centrifuged (15,000 rpm, 2 min, 25 ˚C), and the free protein concentration in the supernatant was measured by Lowry's method.¹¹⁾ The bound protein concentration was determined from the difference between the total protein concentration and the protein concentration in the supernatant. Absorbance parameters were estimated using the following equation based on the Langmuir adsorption isotherm: 1/[PC]=(1/K_a[PC]_max)(1/[PC])+(1/[PC]_max), where [PC], where [PC], Ka (M^-1) and [PC]_max are the concentration of the protein complex with carbohydrate, the equilibrium absorption constant and the maximum amount of protein bound to polysaccharide, respectively.¹²⁾

RESULTS AND DISCUSSION

DNA sequence analysis of PxXyl43A.

Sequence analysis showed that the open reading frame of the PxXyl43A gene consists of 1,560 nucleotides encoding a protein of 520 amino acids with a predicted molecular mass of 57 kDa (Fig. 1A). The DNA sequence coding PxXyl43A was registered with DDBJ with the accession number LC648486. A BLAST homology analysis indicated that the first half of the DNA sequence of PxXyl43A (1-881 bp) was homologous to that of the GH family 43 enzymes, with 72% similarity to GH43_XYL1 from Paenibacillus lautus and 69% similarity to the enzyme from Paenibacillus sp. JDR-2. Phylogenetic analysis with eleven GH43 enzymes, whose three-dimensional structures are registered in the PDB, indicated that PxXyl43A belongs to GH43 subfamily 12 (Fig. 1B). However, there was no homologous sequence match to the latter half of the sequence of the PxXyl43A gene (882-1,560 bp). The signal peptide analysis using SignalP 5.0 predicted that PxXyl43A has no signal peptide, suggesting either that this protein is an intracellular enzyme or that this bacterium has another expression system independent of signal peptides. Thus, PxXyl43A is expected to be an intercellular enzyme belonging to GH43 subfamily 12, defined as being composed of a GH43 β-xylosidase catalytic module at the N-terminus and an unknown-function module, named UM, at the C-terminus (Fig. 1A). According to the database, some of the known GH43 enzymes have a catalytic module at the N-terminus and an additional unknown-function module, similar to PxXyl43A-UM, at the C-terminus.^{13)14)15)16)17)} However, the amino acid sequences of these unknown-function modules did not show similarity to that of the function-characterized modules, classified as CBM. Therefore, we decided to characterize the catalytic activity of the full-length PxXyl43A as well as the carbohydrate-binding activity of the unknown-function module (UM) derived from PxXyl43A.

Fig. 1. The domain structure of PxXyl43A and the phylogenetic analysis with GH43 enzymes.

　A: Schematic diagram of PxXyl43A and its recombinants, PxGH43 and PxGH43-UM. The numbers represent the number of amino acids. PxGH43 family proteins consist of a GH43 catalytic domain (GH43) and an unknown function module (UM), which are connecting by a linker peptide between base pairs 287 and 321. B: Phylogenetic tree of GH43 subfamily 11 to 14. Each alignment was obtained using NCBI. The phylogenetic tree was built using the neighbor-joining method in Mega X. The following enzymes were used in this analysis: GH family 43 protein from Cladosporium acalyphae SL-16 (AQX83203), xylosidase/arabinofuranosidase from Aspergillus niger DSM 26641 (ALN49267), β-xylosidase/α-L-arabinofuranosidase from Phanerochaete chrysosporium BKM-F-1767 (AFW16059), two β-1,4-xylosidases (BAF39209 and BAF39984) from Bifidobacterium adolescentis ATCC 15703, unnamed protein product from Aspergillus oryzae RIB40 (BAE55534), Pc16g08140 from Penicillium rubens Wisconsin 54-1255 (CAP93484), β-xylosidase from Geobacillus thermoleovorans IT-08 (ABC75004), XynB from Paenibacillus sp. JDR-2 (ABV90487), GH family 43 protein from Paenibacillus lautus E7593-69 (AYB47526), GH family 43 protein from Paenibacillus crassostreae LPB0068 (AOZ93067).

Cloning, Expression and purification of PxXyl43A.

The amplified DNA encoding the full-length protein (PxXyl43A) and the unknown-function module (PxXyl43A-UM) were 1,585 and 684 bp, respectively (Fig. 1A). The molecular sizes of the recombinant proteins of PxXyl43A and PxXyl43A-UM were estimated at 57 and 25 kDa, respectively by SDS-PAGE, corresponding to the predicted molecular sizes (data not shown). The concentration of purified recombinant proteins of PxXyl43A and PxXyl43A-UM were 11 and 1.1 mg/mL, respectively.

Assay for β-xylosidase activity of PxXyl43A.

The β-xylosidase activity of the recombinant PxXyl43A toward pNPX as a soluble substrate was measured. As a result, PxXyl43A showed hydrolytic activity for pNPX, with a specific activity of 250 ± 37 mU/mg. The kinetic parameters K_m and k_cat of the β-xylosidase activity of PxXyl43A for pNPX were 1.2 mM and 2.8 ± 0.15 s^-1, respectively. As the k_cat values for pNPX of other GH43 β-xylosidases were reported to be 1.3‒57 s^-1, PxXyl43A has a relatively slow turnover rate for pNPX compared to other GH43 β-xylosidases.^{18)19)20)21)22)} The K_m data also suggest that PxXyl43A has relatively high affinity for pNPX, as the K_m values of other GH43 β-xylosidases for pNPX were 0.38‒17 mM.^18)19)20)23) The β-xylosidase activity of PxXyl43A toward insoluble xylan OSX was also tested because the TW1 strain possesses high xylanolytic activity toward OSX.⁶⁾ It has been reported that some GH43 enzymes have β-xylosidase activity toward insoluble xylan.^24)25)26) However, it was found that PxXyl43A did not show β-xylosidase activity toward OSX (data not shown). These results suggest that the TW1 strain possesses other β-xylosidases responsible for the degradation of insoluble xylan.

Substrate specificity of PxXyl43A.

To assess the substrate specificity of PxXyl43A for sugars other than xylose, the hydrolytic activities toward pNPAf, pNPG, and pNPGal were also measured. Some of the GH43 enzymes from the rumen metagenome and from Bacillus sp. HJ14 exhibit activity for both pNPX and pNPAf while some GH43 enzymes possess hydrolytic activity toward only one of the two.^21)22)24)27) As the exception, the GH43 enzyme from Geobacillus stearothermophilus T-6 hydrolyzes a wide variety of pNP substrates: namely, pNPX, pNPAf, pNPG, pNPGal, pNP-α-L-rhamnopyranoside, pNP-β-D-fucopyranoside, and pNP-β-D-mannopyranoside.¹⁹⁾ We found that PxXyl43A showed hydrolytic activity for pNPAf, but not for pNPG or pNPGal. The specific activity for pNPAf was determined as 310 ± 29 mU/mg, a value comparable to that of the activity for pNPX (250 ± 37 mU/mg). Thus, PxXyl43A is a GH43 enzyme that acts on both pNPX and pNPAf. However, PxXyl43A did not show hydrolytic activity for pNPGal and pNPG, similar to most other reported GH43 enzymes.

Optimal pH and temperature and thermostability of PxXyl43A.

The optimal reaction condition of PxXyl43A was determined using pNPX as a substrate. The data showed that the optimal pH of PxXyl43A was 7.0 and that over 80% of its maximum activity was retained at pH 5.0 to 8.0 (Fig. 2A). The optimal pHs of other GH43 β-xylosidases have been reported to be pH 5.5 to pH 7.5, so the optimal pH of PxXyl43A falls within the usual range.^{19)20)21)22)28)29)} The optimal temperature of PxXyl43A was 54 ˚C, and over 50% of its maximum activity was retained at temperatures from 40 to 60 ˚C (Fig. 2B). Similarly, the optimal reaction temperatures reported for pNPX of GH43 β-xylosidases from other mesophilic microbes are 50 to 65 ˚C.^19)20)29)30) As to the thermostability, the activity of PxXyl43A was stably maintained at 40 to 52 ˚C, but decreased drastically at temperatures higher than 52 ˚C (Fig. 2C). Considering that the optimal growth temperature of P. xylaniclasticus TW1 is around 40 ˚C, it seems reasonable that PxXyl43A exhibits its enzymatic activity at this temperature.⁵⁾ Thus, the data indicated that PxXyl43A possesses thermal resistance up to 52 ˚C.

Fig. 2. Characterization of β-xylosidase activity of PxGH43.

　The optimal pH (A), optimal temperature (B), thermostability (C) and effect of xylose concentration (D) were determined for the β-xylosidase activity of PxGH43 in the presence of 0.25 mM pNPX as a substrate. Values are the means of triplicate experiments ± the standard deviations.

Stability of β-xylosidase activity of PxXyl43A in the presence of high xylose.

It has been reported that most GH43 enzymes show lower β-xylosidase activity in the presence of high concentrations of xylose.²³⁾³¹⁾ For example, a GH43 subfamily 12 enzyme from Lactobacillus brevis, XynB2, lost 66% of its β-xylosidase activity in the presence of 100 mM xylose.²³⁾ By contrast, PxXyl43A retained 85% of its β-xylosidase activity in the presence of 100 mM of xylose and 74% of its activity at 200 mM xylose (Fig. 2D). These data indicate that PxXyl43A possesses high tolerance for product inhibition by D-xylose. This characteristic of PxXyl43A would be advantageous for lignocellulosic biomass saccharification in the presence of xylose at high concentrations.

Analysis of hydrolysis products.

To study the mode of action of PxXyl43A on different-length xylooligosaccharides, reactions using X2, X3, X4, X5, and X6 as substrates were analyzed by TLC (Fig. 3). When using X2 as a substrate, xylose was released as a product of PxXyl43A after 2 h reaction and increased every hour. When X3 was used as a substrate, xylose was released slowly and detected after 4 h reaction. On the other hand, when X4, X5, and X6 were used as substrates, no reaction products were detected. These results suggest that PxXyl43A is an exo-acting enzyme which releases xylose from small xylooligosaccharides up to DP 3.

Fig. 3. Thin-layer chromatography of hydrolysis products of PxGH43 released from xyloorigosaccharides.

　Reactions were carried out for 1-4 h using 1 mg/mL of each oligosaccharide substrate: X2, X3, X4, X5 or X6. Xylose (X1) and X2 were applied as standards.

Binding specificity of PxXyl43A-UM for insoluble polysaccharides.

To investigate the binding specificity and affinity of PxXyl43A-UM for insoluble polysaccharides, a binding test and macroarray assay were performed (Figs. 4A and 4B). For the binding test, BWX, OSX, BMC or lichenan were used as insoluble polysaccharides (Fig. 4A). In cases of OSX and BWX as insoluble xylan, almost all the PxXyl43A-UM protein was detected in the bound fraction, whereas BSA was detected mostly in the unbound fraction (Fig. 4A). These results suggest that PxXyl43A-UM possesses binding activity to insoluble xylan. On the other hand, PxXyl43A-UM did not bind to cellulosic substrates, such as BMC and lichenan (Fig. 4A). The macroarray assay confirmed that PxXyl43A-UM bound to BWX and OSX but not to BMC, CMC or lichenan (Fig. 4B). Among these polysaccharides, PxXyl43A-UM showed the highest affinity for BWX. These results suggest that PxXyl43A-UM specifically binds to insoluble xylan.

Fig. 4. Binding assay of PxGH43-UM to insoluble polysaccharides.

　A, The binding of PxGH43-UM to insoluble polysaccharides, BMS, OSX, BWX and lichenan, were analyzed by SDS-PAGE using BSA as a control. M, molecular weight marker. B, Macroarray assay of PxGH43-UM binding to each insoluble polysaccharide (0.2-10 μg): BWX, OSX, BMC, CMC and lichenan.

Binding affinity of PxXyl43A-UM for OSX.

Different concentrations of PxXyl43A-UM were incubated with OSX, and the bound protein concentration was determined from the difference between the total protein concentration and the protein concentration in the supernatant. The binding constant value (K_a) of PxXyl43A-UM for OSX was determined. As a result, the K_a value was 2.0×10⁵ M^-1 and [PC]_max was 10 µmol/g for 0.5 g/L OSX. This K_a value is comparable to that of the already reported xylan-specific CBMs, belonging to CBM families 6, 13, and 35.²⁾³⁾⁴⁾ The amino acid sequence of PxXyl43A-UM showed no remarkable similarity to those of these already characterized CBMs. There is a report that GH43 β-xylosidase from Thermobifida fusca also has a C-terminus additional module, which is 30% identical to PxXyl43A-UM. However, in this case, the C-terminus module seemed not to be required for xylan-binding activity of the enzyme; indeed, no xylan-specific binding was shown by this module.¹⁴⁾ Our present study shows that PxXyl43A-UM is a novel CBM that specifically binds to xylan, the first representative of CBM91. However, the physiological significance of PxXyl43A-UM's binding ability for insoluble xylan is not clear, because PxXyl43A itself showed no β-xylosidase activity toward insoluble xylan. In addition, PxXyl43A seems to be an intracellular enzyme, since it does not possess a sequence for a typical signal peptide. There are some GH43 β-xylosidases, CoXyl43A from a compost metagenome and XacGH43_1 from Xanthomonas citri, that do not have C-terminal module.³²⁾³³⁾ Hence, it is likely that the xylan-binding ability may not be involved in GH43 β-xylosidase activity. Nevertheless, the xylan-binding activity as well as the β-xylosidase activity of PxXyl43A could be utilized as a tool for efficient saccharification of xylan-containing biomass by combination with other degradative enzymes for insoluble xylan.

CONFLICTS OF INTEREST

The authors declare that they have no conflict of interest.

ACKNOWLEDGMENTS

We thank Rina Shirasaki for CBM binding assay.

REFERENCES

• 1)  L.R. Moreira and E.X. Filho: Insights into the mechanism of enzymatic hydrolysis of xylan. Appl. Microbiol. Biotechnol., 100, 5205-5214 (2016).
• 2)  K. Sakka, M. Nakanishi, M. Sogabe, T. Arai, H. Ohara, A. Tanaka, T. Kimura, and K. Ohmiya: Isothermal titration calorimetric studies on the binding of a family 6 carbohydrate-binding module of Clostridium thermocellum XynA with xlylooligosaccharides. Biosci. Biotechnol. Biochem., 67, 406-409 (2003).
• 3)  A.B. Boraston, P. Tomme, E.A. Amandoron, and D.G. Kilburn: A novel mechanism of xylan binding by a lectin-like module from Streptomyces lividans xylanase 10A. Biochem. J., 350, 933-941 (2000).
• 4)  A. Dhillon, K. Sharma, V. Rajulapati, and A. Goyal: The multi-ligand binding first family 35 Carbohydrate Binding Module (CBM35) of Clostridium thermocellum targets rhamnogalacturonan I. Arch. Biochem. Biophys., 654, 194-208 (2018).
• 5)  C. Tachaapaikoon, S. Tanasupawat, P. Pason, S. Sornyotha, R. Waeonukul, K.L. Kyu, and K. Ratanakhanokchai: Paenibacillus xylaniclasticus sp. nov., a xylanolytic-cellulolytic bacterium isolated from sludge in an anaerobic digester. J. Microbiol., 50, 394-400 (2012).
• 6)  C. Tachaapaikoon, K.L. Kyu, P. Pason, and K. Ratanakhanockchai: A novel multienzyme complex from a newly isolated facultative anaerobic bacterium, Paenibacillus sp. TW1. Act. Biol. Hung., 63, 288-300 (2012).
• 7)  S. Nozaki and H. Niki: Exonuclease III (XthA) enforces in vivo DNA cloning of Escherichia coli to create cohesive ends. J. Bacteriol., 201, e00660-18 (2019).
• 8)  G.L. Miller: Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem., 31, 426-428 (1959).
• 9)  L. McCartney, H.J. Gilbert, D.N. Bolam, A.B. Boraston, and J.P. Knox: Glycoside hydrolase carbohydrate-binding modules as molecular probes for the analysis of plant cell wall polymers. Anal. Biochem., 326, 49-54 (2004).
• 10)  Y. Araki, S. Karita, T. Tsuchiya, M. Kondo, and M. Goto: Family 17 and 28 carbohydrate-binding modules discriminated different cell-wall sites in sweet potato roots. Biosci. Biotechnol. Biochem., 74, 802-805 (2010).
• 11)  O.H. Lowry, N.J. Rosebrough, A.L. Farr, and R.J. Randall: Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193, 265-275 (1951).
• 12)  N.R. Gilkes, E. Jervis, B. Henrissat, B. Tekant, R.C. Miller, R.A.J Warren, and D.G. Kilburn: The adsorption of a bacterial cellulase and its two isolated domains to crystalline cellulose. J. Biol. Chem., 267, 6743-6749 (1992).
• 13)  A. Rohman, N. van Oosterwijk, N.N.T. Puspaningsih, and B.W. Dijkstra: Structural basis of product inhibition by arabinose and xylose of the thermostable GH43 β-1,4-xylosidase from Geobacillus thermoleovorans IT-08. PLoS One, 13, e0196358 (2018).
• 14)  S. Moraïs, O. Salama-Alber, Y. Barak, Y. Hadar, D.B. Wilson, R. Lamed, Y. Shoham, and A.E. Bayer: Functional association of catalytic and ancillary modules dictates enzymatic activity in glycoside hydrolase family 43 β-xylosidase. J. Biol. Chem., 287, 9213-9221 (2012).
• 15)  S. Yoshida, C.W. Hespen, R.L. Beverly, R.I. Mackie, and I.K. Cann: Domain analysis of a modular α-L-arabinofuranosidase with a unique carbohydrate binding strategy from the fiber-degrading bacterium Fibrobacter succinogenes S85. J. Bacteriol., 192, 5424-5436 (2010).
• 16)  C. Brüx, A. Ben-David, D. Shallom-Shezifi, M. Leon, K. Niefind, G. Shoham, Y. Shoham, and D. Schomburg: The structure of an inverting GH43 β-xylosidase from Geobacillus stearothermophilus with its substrate reveals the role of the three catalytic residues. J. Mol. Biol., 359, 97-109 (2006).
• 17)  L.M. Zanphorlin, M.A.B. de Morais, J.A. Diogo, M.N. Domingues, F.H.M. de Souza, R. Ruller, and M.T. Murakami: Structure-guided design combined with evolutionary diversity led to the discovery of the xylose-releasing exo-xylanase activity in the glycoside hydrolase family 43. Biotechnol. Bioeng., 116, 734-744 (2019).
• 18)  D.B. Jordan, X.L. Li, C.A. Dunlap, T.R. Whiteheada, and M.A. Cotta: β-D-xylosidase from Selenomonas ruminantium of glycoside hydrolase family 43. Appl. Biochem. Biotechnol., 137, 93-104 (2007).
• 19)  D. Shallom, M. Leon, T. Bravman, A. Ben-David, G. Zaide, V. Belakhov, G. Shoham, D. Schomburg, T. Baasov, and Y. Shoham: Biochemical characterization and identification of the catalytic residues of a family 43 β-D-xylosidase from Geobacillus stearothermophilus T-6. Biochemistry. 44, 387-397 (2005).
• 20)  K. Wagschal, C. Heng, C.C. Lee, and D.W.S. Wong: Biochemical characterization of a novel dual-function arabinofuranosidase/xylosidase isolated from a compost starter mixture. Appl. Microbiol. Biotechnol., 81, 855-863 (2009).
• 21)  J. Zhou, L. Bao, L. Chang, Y. Zhou, and H. Lu: Biochemical and kinetic characterization of GH43 β-D-xylosidase/ α-L-arabinofuranosidase and GH30 α-L-arabinofuranosidase/ β-D-xylosidase from rumen metagenome. J. Ind. Microbiol. Biotechnol., 39, 143-152 (2012).
• 22)  R. Zhang, N. Li, Y. Liu, X. Han, T. Tu, J. Shen, S. Xu, Q. Wu, J. Zhou, and Z. Huang: Biochemical and structural properties of a low-temperature-active glycoside hydrolase family 43 β-xylosidase: Activity and instability at high neutral salt concentrations. Food Chem., 301, 125266 (2019).
• 23)  H. Michlmayr, J. Hell, C. Lorenz, S. Böhmdorfer, T. Rosenau, and W. Kneifel: Arabinoxylan oligosaccharide hydrolysis by family 43 and 51 glycosidases from Lactobacillus brevis DSM 20054. Appl. Environ. Microbiol., 79, 6747-6754 (2013).
• 24)  T. Teeravivattanakit, S. Baramee, P. Phitsuwan, R. Waeonukul, P. Pason, and C. Tachaapaikoon: Novel trifunctional xylanolytic enzyme Axy43A from Paenibacillus curdlanolyticus strain B-6 exhibiting endo-xylanase, β-D-xylosidase, and arabinoxylan arabinofuranohydrolase activities. Appl. Environ. Microbiol., 82, 6942-6951 (2016).
• 25)  X.Yang, P. Shi, H. Huang, H. Luo, Y. Wang, W. Zhang, and B. Yao: Two xylose-tolerant GH43 bifunctional β-xylosidase/ α-arabinosidases and one GH11 xylanase from Humicola insolens and their synergy in the degradation of xylan. Food Chem., 148, 381-387 (2014).
• 26)  P. Shi, X. Chen, K. Meng, H. Huang, Y. Bai, H. Luo, P. Yang, and B. Yao: Distinct actions by Paenibacillus sp. strain E18 α-L-arabinofuranosidases and xylanase in xylan degradation. Appl. Environ. Microbiol., 79, 1990-1995 (2013).
• 27)  H. Ichinose, M. Yoshida, Z. Fujimoto, and S. Kaneko: Characterization of a modular enzyme of exo-1,5-α-L-arabinofuranosidase and arabinan binding module from Streptomyces avermitilis NBRC14893. Appl. Microbiol. Biotechnol., 80, 399-408 (2008).
• 28)  B.R. de Camargo, N.J. Claassens, B.F. Quirino, E.F. Noronha, and S.W.M. Kengen: Heterologous expression and characterization of a putative glycoside hydrolase family 43 arabinofuranosidase from Clostridium thermocellum B8. Enzyme Microb. Technol., 109, 74-83 (2018).
• 29)  Y.A. Kim and K.H. Yoon: Characterization of a Paenibacillus woosongensis β-xylosidase/ α-arabinofuranosidase produced by recombinant Escherichia coli. J. Microbiol. Biotechnol., 20, 1711-1716 (2010).
• 30)  A.H. Viborg, K.I.Sørensen, O. Gilad, D.B.Steen-Jensen, A. Dilokpimol, S. Jacobsen, and B. Svensson: Biochemical and kinetic characterisation of a novel xylooligosaccharide-upregulated GH43 β-D-xylosidase/α-L-arabinofuranosidase (BXA43) from the probiotic Bifidobacterium animalis subsp. lactis BB-12. AMB Express, 3, 56 (2013).
• 31)  A. Rohman, B.W. Dijkstra, and N.N.T. Puspaningsih: β-xylosidases: Structural diversity, catalytic mechanism, and inhibition by monosaccharides. Int. J. Mol. Sci., 20, 5524 (2019).
• 32)  T. Matsuzawa, S. Kaneko, N. Kishine, Z. Fujimoto, and K. Yaoi: Crystal structure of metagenomic β-xylosidase/ α-L-arabinofuranosidase activated by calcium. J. Biochem., 162, 173-181 (2017).
• 33)  M.A.B. Morais, J. Coines, M.N. Domingues, R.A.S. Pirolla, C.C.C. Tonoli, C.R. Santos, J.B.L. Correa, F.C. Gozzo, C. Rovira, and M.T. Murakami: Two distinct catalytic pathways for GH43 xylanolytic enzymes unveiled by X-ray and QM/MM simulations. Nat. Commun., 12, 367 (2021).