2015 Volume 62 Issue 1 Pages 1-6
We characterized a β-L-arabinopyranosidase AbpBL (BLLJ_1823) belonging to the glycoside hydrolase family 27 (GH27) from Bifidobacterium longum subsp. longum JCM1217. The recombinant AbpBL expressed in Escherichia coli hydrolyzed pNP-β-L-arabinopyranoside but not pNP-α-D-galactopyranoside. The enzyme also liberated L-arabinose from the β-L-arabinopyranosyl side chain of larch wood arabinogalactan. However, we could not detect any β-L-arabinopyranosidase activity or remarkable transcriptional induction in cultured cells of B. longum subsp. longum. Mutagenesis experiments revealed that I56D and I56A mutants both exhibited β-L-arabinopyranosidase and α-D-galactopyranosidase activities. AbpBL Ile-56 residue is a critical residue for the specificity of β-L-arabinopyranosidase.
AG-II, type-II arabinogalactan; β-Arap, β-L-arabinopyranose; α-Araf, α-L-arabinofuranose; Arap-Araf-Gal3, β-Arap-α-Araf-β-1,6-galactotriose; pNP, p-nitrophenyl; GH, glycoside hydrolase family; LWAG, larch wood arabinogalactan; HPAEC-PAD, high-performance anion-exchange chromatography with pulsed amperometric detection; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry.
The type II arabinogalactan (AG-II) chains in larch wood arabinogalactan (LWAG) comprise a β-1,3-linked galactan backbone and β-1,6-linked galactan side chains with α-L-arabinofuranose (α-Araf) and β-L-arabinopyranose (β-Arap) substitutions.1) The terminal β-Arap-1,3-α-Araf structures were also found in the arabinogalactan protein from wheat flour2) and gum arabic.3) β-L-Arabinopyranosidase (EC 3.2.1.88) was first purified from seeds of plant Cajanus indicus.4) 5) Ichinose et al. cloned and characterized a β-L-arabinopyranosidase from Streptomyces avermitilis, and revealed the three-dimensional structure.6) The enzyme belongs to the glycoside hydrolase family 27 (GH27), which is mainly comprised of α-D-galactopyranosidase and α-N-acetylgalactosaminidase. Because α-D-galactopyranoside and β-Arap have same pyranose structure, α-D-galactopyranosidase can potentially possess β-L-arabinopyranosidase activity. They revealed that Glu-99 is a critical residue that modulated β-L-arabinopyranosidase and α-D-galactopyranosidase activities in S. avermitilis. The bifunctional enzymes harboring Cys as the critical residues were characterized from Fusarium oxysporum.7) However, these enzymes showed partial α-D-galactopyranosidase activity in addition to a β-L-arabinopyranosidase activity. In 2012, a homologous enzyme exhibiting specificity toward the β-Arap residues was characterized from Geobacillus stearothermophilus.8) The enzyme conserved Ile as the critical residue.
Recently, we characterized BLLJ_1840 as an exo-β-1,3-galactanase (EC 3.2.1.145) in Bifidobacterium longum subsp. longum JCM1217.9) The enzyme hydrolyzed the β-1,3-galactan backbone, bypassing the β-1,6-galactan side chains, and released β-1,6-galactooligosaccharides and their derivatives containing terminal β-Arap and α-Araf residues. This strain encoded a β-L-arabinopyranosidase candidate (BLLJ_1823), which conserved Ile as the critical residue. Therefore, we predicted that B. longum subsp. longum strains would utilize a β-L-arabinopyranosidase for the assimilation of β-L-arabinopyranosyl oligosaccharides. In this study, we cloned the abpBL (BLLJ_1823) gene from B. longum subsp. longum JCM1217 and characterized the recombinant protein as a β-L-arabinopyranosidase.
Materials. Gum arabic and p-nitrophenyl (pNP) substrates were both obtained from Sigma-Aldrich (St. Louis, USA). LWAG was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). β-Arap-α-Araf-β-1,6-galactotriose (Arap-Araf-Gal3), β-1,6-Gal2, and β-1,6-Gal3 were prepared from LWAG as previously described,9) and β-1,2-Araf2 was prepared from extensin.10) α-L-Arabinofuranosidase (EC 3.2.1.55) and β-galactosidase (EC 3.2.1.23) from Aspergillus niger were purchased from Megazyme (Wicklow, Ireland). All other chemicals were obtained from Wako Pure Chemical Industries Ltd. (Osaka, Japan).
Expression and purification of recombinant AbpBL. The translational start codon (Met) for the other B. longum homologues, BL0177 and BLD1538, was predicted to be 15 amino acids upstream of the Met for BLLJ_1823 (GenBank ID: BAJ67488). Because the N-terminal 15 amino acids were conserved in almost all of the bifidobacteria as shown in JSTAGE Supplementary Material Fig. S1, we predicted that the ATG start codon of AbpBL would be 45 nucleotides upstream of the start codon for BLLJ_1823. The abpBL gene corresponding to the nucleotide sequence in the region 2191789‒2193198 (1,410 nucleotides) of B. longum subsp. longum JCM1217 encodes 469 amino acids. This nucleotide sequence has been deposited in the third party annotation section of DDBJ/GenBank/EBI Data Bank with accession number BR001234. The forward (5′-AAGAAGGAGATATACCATGACGGATCTTCCGCAGA-3′) and reverse (5′-GTGGTGCTCGAGTGCGGCCGCGCGTCGGTCCAGGGCCAC-3′) primers were designed from the nucleotides 2193180‒2193198, and 2191792‒2191809, respectively. The underlined texts represent the nucleotides complementary to the template. The amplicon was subsequently cloned into the pET-23d vector (Novagen Inc., Madison, USA) using the GeneArt (Life Technologies, Regensburg, Germany). The resulting pET23d‒abpBL plasmid was sequenced on an ABI 3100 DNA sequencer using a BigDye Terminator 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, USA). The plasmid was transformed into E. coli BL21 (λDE3) cells, and subsequently grown at 20°C by using the Overnight Express Autoinduction System (Novagen). The cell cultures were subsequently centrifuged, and the pellets were then resuspended in the BugBuster Protein Extraction Reagent (Novagen). The His-tagged AbpBL protein was purified with a column containing the TALON Metal Affinity Resin (Clontech Laboratories Inc., Palo Alto, USA). The eluted protein was desalted and concentrated by using a 10-kDa ultrafiltration membrane (Millipore Co., Billerica, USA).
Enzyme assays. The hydrolytic activity of the AbpBL enzyme was assayed using pNP-β-L-arabinopyranoside (pNP-β-Arap) as a substrate. The 200 μL reaction mixture contained 1 mM of the substrate, 7.9 mU/mL of AbpBL, and 50 mM sodium acetate buffer (pH 5.5). After incubating the reaction mixture at 40°C, the enzymatic reactions were terminated by the addition of 300 μL of 100 mM Na2CO3. The absorbance of the released pNP was measured at 400 nm. One unit of enzyme activity was defined as the amount of enzyme required to produce 1 μmol of pNP per minute.
The hydrolytic activity toward Arap-Araf-Gal3 was analyzed as follows: Arap-Araf-Gal3 was incubated with 7.9 mU/mL of AbpBL in 40 μL of 50 mM sodium acetate buffer (pH 5.5). After incubating the reaction mixture at 40°C for 16 h, the reaction was stopped by boiling this mixture for 3 min. This reaction mixture was used for the enzymatic digestibility analysis with A. niger α-L-arabinofuranosidase and β-galactosidase. The reaction mixtures were assessed using high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). The oligosaccharides were analyzed with a CarboPac PA-1 column (φ 4 × 250 mm; Dionex Corp., Sunnyvale, USA). The column was eluted at a flow rate of 1.0 mL/min using the following gradient: 0‒5 min, 100% eluent A (0.1 M NaOH); 5‒30 min, 0‒100% eluent B (0.5 M sodium acetate and 0.1 M NaOH); and 30‒35 min, 100% eluent B. The oligosaccharides were collected and examined using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS; Bruker Daltonics, Leipzig, Germany).
The hydrolysis rate was analyzed using 1.0% LWAG with 7.9 mU/mL of AbpBL in 40 μL of 50 mM sodium acetate buffer (pH 5.5). After incubating the reaction mixture at 40°C for 36 h, the releasing L-arabinose was quantified by HPAEC-PAD. The total L-arabinose content in LWAG was measured by the sugar composition analysis as described previously.10)
The substrate specificity toward the pNP-substrates was analyzed as follows: One mM of substrates were incubated at 40°C for 16 h with 7.9 mU/mL of AbpBL in 40 μL of 50 mM sodium acetate buffer (pH 5.5). The reaction products were spotted on a silica gel 60 aluminum plate (Merck, Darmstadt, Germany) using a 7:1:2 (v/v/v) n-propanol/EtOH/water solvent mixture. The sugars were visualized by spraying orcinol-sulfate reagent on the plate.11)
The pH dependence of the enzyme activity was determined using pNP-β-Arap as the substrate between pH 3.0 and 8.0 using the following buffers: 50 mM sodium acetate (pH 3.0‒6.0), 50 mM MES (pH 5.5‒7.0), and 50 mM HEPES (pH 7.0‒8.0). The effect of temperature on enzyme activity was determined using 50 mM sodium acetate buffer (pH 5.5) at 25‒45°C.
Kinetic analysis. The kinetic parameters for AbpBL were determined using 0.5‒8.0 mM pNP-β-Arap. The 200 μL reaction mixture containing 50 mM sodium acetate buffer (pH 5.5), and 7.9 mU/mL of AbpBL was incubated at 40°C for 10 min. The samples were analyzed as described above.
Site-directed mutagenesis. The KOD-Plus-Mutagenesis Kit (Toyobo Co., Ltd., Osaka, Japan) was used to introduce amino acid substitutions into AbpBL by using the primers shown in JSTAGE Supplementary Material Table S1. The primers I56_reverse and I56D (or I56A)_forward were used to design the constructs for the I56D (or I56A) mutants. These mutant enzymes were expressed and purified by using the same procedure as that for the wild-type enzyme.
Assays of bacterial enzyme activities and quantitative real-time PCR. B. longum subsp. longum JCM1217 was cultured at 37°C under anaerobic conditions using the AnaeroPack system (Mitsubishi Gas Chemical, Tokyo, Japan) on a peptone-yeast extract-Fildes (PYF) medium12) containing 1.0% L-arabinose, glucose, galactose, β-1,2-Araf2, β-1,6-Gal2, β-1,6-Gal3, or LWAG. The cell cultures were centrifuged at 17,000 × G for 20 min, and the resultant pellets were washed with 50 mM sodium acetate buffer (pH 6.0). Afterwards, the pellets were sonicated with a Branson Sonifier 250 (Danbury, USA). The cell lysates were centrifuged at 17,000 × G for 10 min, and the cell pellets were resuspended with 50 mM sodium acetate buffer (pH 6.0). The lysate supernatants and the sonicated cell pellets were incubated with pNP-β-Arap at 40°C for 16 h, and then analyzed by TLC and HPAEC-PAD as described above. For the quantitative real-time PCR, total RNA extraction, reverse transcription, and quantitative real-time PCR were performed as described previously.9) The primers for the analysis of rpoB and abpBL are shown in JSTAGE Supplementary Material Table S1.
Sequence analysis of AbpBL.
AbpBL lacks the putative signal peptide and C-terminal transmembrane domains, according to the SignalP 4.1 (http://www.cbs.dtu.dk/services/SignalP/) and InterPro servers (http://www.ebi.ac.uk/interpro/), which suggests that the enzyme localizes to the intracellular compartments. AbpBL (amino acids 14‒336) exhibits sequence homologies with other characterized GH27 β-L-arabinopyranosidases from G. stearothermophilus (50% identity) and S. avermitilis (24% identity), and with the bifunctional enzymes from F. oxysporum (24% identities) that show β-L-arabinopyranosidase/α-D-galactopyranosidase activities. The genes homologous to AbpBL are conserved in bifidobacteria (e.g., in B. longum strains, B. adolescentis, B. dentium, and B. asteroides). AbpBL forms a gene cluster with a GH127 β-L-arabinofuranosidase candidate (BLLJ_1826), a Lac-I type transcriptional regulator (BLLJ_1824), an AraC-type transcriptional regulator (BLLJ_1825), and two ABC transporter permeases (BLLJ_1820‒1821). The gene cluster is also conserved in almost all of the B. longum strains. Ile-56 is the critical residue that modulates the β-L-arabinopyranosidase and α-D-galactopyranosidase activities, and this residue is conserved in the β-L-arabinopyranosidase from G. stearothermophilus. As shown in JSTAGE Supplementary Material Fig. S1, the catalytic residues of the GH27 enzymes are also conserved in AbpBL (Asp186 for a nucleophile and Asp274 for a general acid/base).
Preparation and characterization of the recombinant AbpBL. The recombinant AbpBL protein was expressed at 20˚C as a soluble protein. The purified recombinant AbpBL protein migrated as a single band with an apparent molecular mass of 55 kDa on SDS-PAGE and displayed an m/z of 53,670 for MALDI-TOF MS (JSTAGE Supplementary Material Fig. S2). These molecular masses were in agreement with the calculated molecular mass of 53,753 Da. Furthermore, its homodimeric form (m/z 107,175) was estimated by MALDI-TOF MS analysis (JSTAGE Supplementary Material Fig. S2).
Synthetic pNP substrates were used to identify the substrate specificities of AbpBL. The enzyme released L-arabinose from pNP-β-Arap, but not from pNP-α-D-galactopyranoside (pNP-α-Gal), pNP-β-D-galactopyranoside (pNP-β-Gal), pNP-α-L-arabinofuranoside (pNP-α-Araf), or pNP-α-L-arabinopyranoside (pNP-α-Arap) (Fig. 1). The optimal temperature and pH for pNP-β-Arap were 40°C and 5.5, respectively (JSTAGE Supplementary Material Fig. S3). The Km and kcat values for pNP-β-Arap were calculated as 3.20 ± 0.088 mM and 9.02 ± 0.057 s−1, respectively.
TLC analysis of the AbpBL reactions with the pNP-substrates.
The pNP-substrates were incubated in the absence (lane a) or presence (lane b) of the recombinant enzyme at 40°C for 16 h. pNP-α-Araf (lane 1), pNP-α-Arap (lane 2), pNP-β-Arap (lane 3), pNP-α-Gal (lane 4), and pNP-β-D-Gal (lane 5) were used as substrates.
Natural polysaccharides and an oligosaccharide were used to identify the substrates of recombinant AbpBL. As shown in Fig. 2, the enzyme released L-arabinose from LWAG and gum arabic. The specific activity for LWAG was 22-fold higher than that for gum arabic (Table 1). AbpBL removed 14.7% of L-arabinose from total L-arabinose in LWAG. Next, a LWAG side chain, Arap-Araf-Gal3, was used as a substrate. The enzyme released L-arabinose and the peak corresponding to Arap-Araf-Gal3 (retention time: 13.0 min; m/z 767.77; calc. m/z 768.25) shifted to 13.6 min on HPAEC-PAD (Fig. 3a and 3b). MALDI-TOF MS analysis revealed a molecular ion peak at m/z 636.22 for Araf-Gal3 (calc. m/z 636.21). Furthermore, Araf-Gal3 was degraded to β-1,6-Gal3 (retention time: 12.1 min; m/z 505.26; calc. m/z 504.17) by A. niger α-L-arabinofuranosidase (Fig. 3c), but not degraded by A. niger β-galactosidase (data not shown). This suggested that a single α-L-arabinofuranoside was substituted on the non-reducing terminal of β-1,6-Gal3. In addition, the released β-1,6-Gal3 was degraded to galactose by A. niger β-galactosidase (Fig. 3d). Our data also supports the structure of Arap-Araf-Gal3, which was predicted as β-Arap-1,3-α-Araf-1,3-β-Gal-1,6-β-Gal-1,6-Gal.13) These results showed that AbpBL is a β-L-arabinopyranosidase with strict substrate specificity for β-Arap substitutions from natural polysaccharides.
HPAEC-PAD analysis of AbpBL reactions with polysaccharides.
LWAG (A) and gum arabic (B) were incubated with (b) or without (a) AbpBL at 40°C for 16 h.
Substrate specificity of AbpBL toward polysaccharides.a
aThe enzyme was incubated in the 40 μL reaction mixture containing 1.0% substrates and 50 mM sodium acetate buffer (pH 5.5). The reactions were stopped by the addition of 10 μL of 500 mM sodium hydroxide, and the releasing L-arabinose was quantified by HPAEC-PAD. One unit of enzyme activity toward natural substrates was defined as the amount of enzyme required to produce 1 μmol of the liberated L-arabinose per minute. bThe enzyme (0.79 mU/mL) was incubated at 40°C for 30 min. cThe enzyme (1.6 mU/mL) was incubated at 40°C for 180 min. dRelative activity was expressed as the percentage of the activity toward LWAG.
HPAEC-PAD analysis of the degradative products of Arap-Araf-Gal3 with various enzymes.
Arap-Araf-Gal3 was incubated with (b) or without (a) AbpBL at 40°C for 16 h. The reaction product (Araf-Gal3) was additionally incubated with Aspergillus α-L-arabinofuranosidase (c). The reaction product (β-1,6-Gal3) was incubated with Aspergillus β-galactosidase (d).
Mutagenesis studies of AbpBL.
As shown in JSTAGE Supplementary Material Fig. S1, the critical amino acid for modulating the β-L-arabinopyranosidase and/or α-D-galactopyranosidase activities is a conserved Ile or Ala (Ile-56 in the case of AbpBL) in bifidobacteria. Furthermore, an Asp residue is highly conserved among the characterized α-D-galactopyranosidases. Therefore, we selected Ala and Asp substitutions for the site-directed mutagenesis of AbpBL Ile-56. The I56D and I56A mutant enzymes were recovered in the soluble fractions using BugBuster. The I56D and I56A mutants both exhibited β-L-arabinopyranosidase and α-D-galactopyranosidase activities (Fig. 4). The specific activities for pNP-α-Gal were 0.119 U/mg for I56D mutant and 0.343 U/mg for I56A (Table 2). On the other hand, the β-L-arabinopyranosidase activities decreased to 1.4% for the I56D mutant and to 22% for the I56A mutant, relative to the wild-type enzyme. The ratios of the β-L-arabinopyranosidase and α-D-galactopyranosidase activities were 54:46 for the I56D mutant and 86:14 for the I56A mutant.
TLC analysis of the AbpBL mutants with the pNP-substrates.
The wild type AbpBL (lane b), the I56D mutant (lane c), and the I56A mutant (lane d) were incubated with pNP-β-Arap (lane 2) and pNP-α-Gal (lane 4) at 40°C for 16 h. Lane a, control without enzyme; lane 1, L-arabinose standard; lane 3, galactose standard.
The specific activities of the AbpBL mutants.
aSubstrate cleavage was not detected.
Bifidobacterial enzymatic activity and gene expression profiles of B. longum subsp. longum.
We examined the transcript levels of abpBL using previously prepared B. longum subsp. longum cDNAs grown in carbohydrate sources containing L-arabinose, glucose, galactose, β-1,2-Araf2, β-1,6-Gal2, β-1,6-Gal3, or LWAG. AbpBL showed similar levels of gene transcription (0.6‒2.4 fold) for several carbohydrate sources when compared with glucose (Fig. 5). In addition, we could not detect β-L-arabinopyranosidase activity in the cell lysates of B. longum subsp. longum JCM1217 grown in PYF medium containing L-arabinose, glucose, galactose, or LWAG (data not shown).
The transcript levels of abpBL in B. longum subsp. longum grown on several carbohydrate sources.
The fold changes were determined from the measured mRNA values of B. longum subsp. longum grown on glucose. Data indicate mean ± SEM (n ＝ 3).
We predicted that AbpBL to be a β-Arap specific β-L-arabinopyranosidase capable of degrading AG-II in B. longum subsp. longum. However, we could not detect the corresponding β-L-arabinopyranosidase activities in the bifidobacterial cell lysates. In addition, transcriptional analysis showed that abpBL was not remarkably induced by LWAG and other carbohydrate sources used in this study. In the case of BLLJ_1840 (exo-β-1,3-galactanase), the gene transcription increased 109-fold on LWAG as the carbohydrate source.9) These data suggest that AbpBL is a pseudogene without transcriptional control of its gene expression in B. longum subsp. longum. On the other hand, we confirmed the β-L-arabinopyranosidase activity in B. adolescentis JCM1275, which encoded an AbpBL homologous gene BAD_1525 (unpublished data). Therefore, β-L-arabinopyranosidase may be presence as the functional enzyme in the some species of bifidobacteria.
As the critical residue for modulating the enzymatic activity, Asp residue is conserved in GH27 α-galactosidases (JSTAGE Supplementary Material Fig. S1). Ichinose et al. reported that Asp substitution of the enzyme from S. avermitilis retained the β-L-arabinopyranosidase activity.6) We also confirmed that the AbpBL I56D mutant exhibited α-D-galactopyranosidase activity in addition to β-L-arabinopyranosidase activity. Therefore, almost all members of GH27 α-galactosidases may contain β-L-arabinopyranosidase activity in addition to α-D-galactopyranosidase activity. α-Galactosidases belonging into GH27 and GH36 share common domains of catalytic nucleophile and general acid/base.14) The degradative activity toward pNP-β-Arap was confirmed in an archaeal GH36 α-galactosidase,14) but not in an bacterial enzyme.15) These data suggested that GH27 and GH36 α-galactosidases potentially have β-L-arabinopyranosidase activity.
L-Arabinose exists as either terminal Araf or β-Arap-1,3-Araf structure in LWAG.1) From previous studies, the terminal β-Arap contained within LWAG was predicted to account for 26 and 27% in Western larch and Mongolian larch, respectively, of total L-arabinose.1) 16) In this study, AbpBL released 14.7% of L-arabinose from LWAG, which suggested that the enzyme releases roughly 56% of β-Arap from LWAG. We enzymatically determined that the β-Arap-1,3-Araf was substituted on a terminal position of galactose on β-1,6-Gal3. However, the oligosaccharide is a minor compartment in releasing sugars from LWAG by B. longum subsp. longum exo-β-1,3-galactanase. Ponder and Richards predicted that β-Arap-1,3-Araf unit bound to the non-reducing end of β-1,3-galactan main chain and a middle position of β-1,6-galactan side chain in LWAG.1) β-Arap-1,3-Araf unit may attach to LWAG with variety of structure.
The Ile residue in the β-L-arabinopyranosidase from G. stearothermophilus was predicted to cause a construct steric hindrance to the O-6 position of the galactose residue.8) However, this enzyme exhibited very weak level of the degradative activity for pNP-α-Gal and pNP-α-Araf. AbpBL has strict substrate specificity for β-Arap. AbpBL can be useful for the structural analysis of plant polysaccharides and glycoproteins.
This work was supported in part by the Takeda Science Foundation and by the JSPS KAKENHI Grant-in-Aid for Scientific Research (C), Grant Number 24580144.
