Molecular characterization of second tomato α1,3/4-fucosidase (α-Fuc’ase Sl-2), a member of glycosyl hydrolase family 29 active toward the core α1,3-fucosyl residue in plant N-glycans

Molecular characterization of second tomato α1,3/4-fucosidase (α-Fuc’ase Sl-2), a member of... Abstract In a previous study, we molecular-characterized a tomato (Solanum lycopersicum) α1, 3/4-fucosidase (α-Fuc’ase Sl-1) encoded in a tomato gene (Solyc03g006980), indicating that α-Fuc’ase Sl-1 is involved in the turnover of Lea epitope-containing N-glycans. In this study, we have characterized another tomato gene (Solyc11g069010) encoding α1, 3/4-fucosidase (α-Fuc’ase Sl-2), which is also active toward the complex type N-glycans containing Lea epitope(s). The baculovirus-insect cell expression system was used to express that α-Fuc’ase Sl-2 with anti-FLAG tag, and the expression product (rFuc’ase Sl-2), was found as a 65 kDa protein using SDS-PAGE and has an optimum pH of around 5.0. Similarly to rFuc’ase Sl-1, rFuc’ase Sl-2 hydrolyzed the non-reducing terminal α1, 3-fucose residue on LNFP III and α1, 4-fucose residues of Lea epitopes on plant complex type N-glycans, but not the core α1, 3-fucose residue on Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc or Fucα1-3GlcNAc. However, we found that both α-Fuc’ases Sl-1 and Sl-2 were specifically active toward α1, 3-fucose residue on GlcNAcβ1-4(Fucα1-3)GlcNAc, indicating that the non-substituted β-GlcNAc linked to the proximal GlcNAc residue of the core tri-saccharide moiety of plant specific N-glycans must be a pre-requisite for α-Fuc’ase activity. A 3 D modelled structure of the catalytic sites of α-Fuc’ase Sl-2 suggested that Asp192 and Glu236 may be important for binding to the α1, 3/4 fucose residue. α-fucosidase, α1-3 fucose, N-glycan metabolism, plant N-glycan, Solanum lycopersicum Almost all the secreted-type proteins produced in eukaryotic cells are N-glycosylated, suggesting that the N-glycosylation of proteins and the subsequent modification of the glycan moiety are ubiquitous and pivotal biological processes in eukaryotes, especially the multicellular organisms (1, 2). Plant N-glycans are usually classified into four subgroups: high-mannose type, hybrid type, biantennary-complex type and pauci-mannose type. The final structure of these N-glycans, however, varies by species due to the action of a unique set of processing enzymes acting on N-glycans in the secretory pathway of plant cells (3, 4). Plant complex-type N-glycans carry a distinctive attribute with α1, 3-linked fucose residue to the proximal GlcNAc residue and β1, 2-linked xylose residue to β-mannose residue (4–6). Some secreted glycoproteins contain large complex-type N-glycans bearing Lewis a (Lea) epitope (Galβ1-3(Fucα1-4)GlcNAc) at their non-reducing end (7, 8). In addition, we have found that the rice or water plant (Egeria densa) contain various complex type free N-glycans (FNGs), GN2-type FNGs (GN1M3FX, GN2M3FX, Gal1Fuc1GN1M3FX, and Gal2Fuc2GN2M3FX) and GN1-type FNGs (GN1M3X-GN1, Gal1Fuc1GN1M3X-GN1, Gal2Fuc2GN2M3X-GN1) (5, 9, 10). The occurrence of these defucosylated/secreted type FNGs suggests that α-fucosidase must be involved in the degradation pathway of Lea-containing complex type N-glycans working in the secretory pathway or extracellular space. Some plant glycosidases have been found in the cell wall or extracellular space (11–14), although it is well known that many plant glycosidases involved in the turnover of glycoconjugates are localized in the acidic vacuole or protein body. Moreover, the Lea epitope is not found on vacuolar proteins, but is enriched at the plasma membrane and extracellular glycoproteins (4). These observations indicate that α-fucosidase acting on α1, 4-fucosyl linkage may reside in the cell membrane or apoplast to modify the Lea structure. However, the complete scheme of the degradation pathway of plant-specific N-glycans has not been elucidated because α1, 3-fucosidase acting on the plant specific core structures, such as M3FX or MFX, has not been identified so far (6). As for the plant α-fucosidase responsible for plant complex-type N-glycans, it has been reported that almond, Arabidopsis (AtFuc1) and rice α-fucosidases that belong to GH29 are active toward the Lea epitope on plant N-glycans and LNFP III, but not toward the core α1-3 fucosyl residue on Man3-1Xyl1Fuc1GlcNAc2 (15–17). Based on the genetic information of the rice α-Fuc’ase gene (17), two putative tomato (Solanum lycopersicum) α-Fuc’ase genes (Solyc03g06980 and Solyc11g069010) were found in the genome database and we have succeeded to express and characterize one α-Fuc’ase encoded by Solyc03g06980 (α-Fuc’ase Sl-1) (18). α-Fuc’ase Sl-1 was substantially hydrolyzed the non-reducing terminal α1, 3-fucose residue on LNFP III and α1, 4-fucose residues of Lea epitopes on plant complex-type N-glycans, but not by the α1, 3-fucose residue on the plant core penta-oligosaccharide unit (Manβ1-4[Xylβ1-2]GlcNAcβ1-4[Fucα1-3]GlcNAc) or Fucα1-3GlcNAc (GN1F). Molecular 3 D modelling of α-Fuc’ase Sl-1 construction and structure/sequence interpretation based on comparison with a homologous α-fucosidase from Bifidobacterium longum subsp. infantis (Blon_2336) (19) indicated that an unsubstituted β-GlcNAc residue in the GlcNAcβ1-4(Fucα1-3)GlcNAc (GN2F) structure might be pre-requisite to fix the α1, 3-fucosylated substrate at the sugar-binding pocket commonly found in GH29-B (18, 19). However, in our previous study, we could not prepare the substrate, GN2F and thus have not proved our hypothesis to date. As described above, we have characterized the α-Fuc’ase encoded by the Solyc03g06980 (α-Fuc’ase Sl-1) gene, but the molecular characterization of the remaining putative α-Fuc’ase genes (Solyc11g069010) has not been performed. In this study, therefore, we have expressed the gene using the insect cell (Sf9) system and confirmed that the gene product (α-Fuc’ase Sl-2) is also a GH29 family α1, 3/4-Fuc’ase. α-Fuc’ase Sl-2 was active toward the Lea-epitope in the plant complex-type N-glycans and LNFP III but not the M3FX or MFX, indicating that the substrate specificity of the recombinant α-Fuc’ase Sl-2 (rFuc’ase Sl-2) is similar to that of α-Fuc’ase Sl-1. Interestingly, however, we have found that rFuc’ase Sl-1 and rFuc’ase Sl-2 were active toward GlcNAcβ1-4(Fucα1-3)GlcNAc-PA (GN2F) but not MF and GN1F, indicating that these two tomato α-Fuc’ases require the non-reducing terminal β-linkage GlcNAc residue of GN2F for activity and must be involved in the complete degradation of the plant specific N-glycans. Molecular 3D modelling of α-Fuc’ase Sl-2 based on comparisons with the homologous enzyme of Bifidobacterium longum subsp. infantis (Blon_2336) α-fucosidase (19) provided an insight into the catalytic mechanism. Materials and Methods Materials Tomato seeds (KG172) were kindly gifted by the Research Institute, Kagome Co. Ltd. (Tochigi, Japan). Reagents were of analytical or electrophoresis grade and were purchased from Wako Pure Chemical Industries Ltd. (Osaka, Japan), Nacalai Tesque, Inc. (Kyoto, Japan) and Sigma (St. Louis, MO, USA). Genetic manipulation enzymes with other reagents were purchased from Takara Bio Inc. (Kyoto, Japan), Applied Biosystems™ (Austin, TX, USA) and Promega (Madison, WI, USA). Anti-Flag® M2 affinity gel was from SIGMA (Saint Louis, USA). PA-labelled Lacto-N-fucopentaose III (LNFP III) and Lacto-N-fucopentaose I (LNFP I) was purchased from Takara Bio (Kyoto, Japan). Authentic PA-sugar chains (Gal2Fuc2GN2M3FX, M3FX, MFX, MF and GNF) were prepared from glycoproteins produced by rice cultured cells (20), and momordin-a (21). Spodoptera frugiperda (Sf9) cells were purchased from Invitrogen (Carlsbad, CA, USA). MF, GN2F and GNF were also prepared from MFX using tomato β-xylosidase (22), β-mannosidase (Helix pomatia, Sigma) and β-N-acetylhexosaminidase (Jack bean, Sigma). Their structures were confirmed by ESI-MS and MS/MS analyses as shown in Supplementary Fig. S1. Molecular cloning and recombinant expression The coding sequence (1554 bp) of Solyc11g069010 (locus: Solyc11g069010; allele: SOLYC11G069010 in Sol Genomics Network: www.solgenomics.net/search/locus), described here as α-Fuc’ase Sl-2, was amplified through PCR and cloned into pGEM-T easy vector, and thereafter sub-cloned into pFastBac™-1 using the Bac-to-Bac Baculovirus Expression System (Invitrogen). In every cloning step, the specific primers were designed (Supplementary Table SI) and the amplified PCR products were subjected to 1% (w/v) agarose gel electrophoresis to view proper DNA banding on the gel (data not shown). The correct orientation of the gene of interest and all of the constructs were verified by sequencing. The constructed recombinant pFastBac™-1 vector served as a donor plasmid, and was used to transform into the E. coli DH10Bac with the bacmid DNA. The donor plasmid (recombinant pFastBac™-1) transposed the α-Fuc’ase Sl-2 gene into the bacmid DNA with the help of a helper plasmid available in the DH10Bac E.coli, and eventually constructed the baculovirus expression vector (recombinant bacmid). The selection of the recombinant bacmid was carried out with kanamycin, gentamycin and tetracycline resistance. Bacmid vectors for insert sizes were also checked by restriction digestion with EcoRI and NotI following 1% (w/v) agarose gel electrophoresis. The bacmid expression construct was finally transfected into Sf9 insect cells for expression of the α-Fuc’ase Sl-2 gene. The Sf9 (cell line of Spodoptera frugiperda) cells were transfected with this recombinant bacmid DNA for the expression of α-Fuc’ase Sl-2 gene using Cellfectin® II reagent (Invitrogen) according to the manufacturers protocol. At 72 h post-infection, supernatants were collected for the recombinant baculovirus as P1 virus stock, and the recombinant virus stock was amplified for two more successive generations as P2 and P3 stock. A time-course study was performed to select the highest expression level using P3 viral stock. Insect cell culture and transfection The insect cell line Spodoptera frugiperda Sf9 was cultured in TNM-FH insect medium (Invitrogen) supplemented with 10% fetal bovine serum (FBS) at 27°C. The Sf9 cells were cotransfected with the baculovirus as instructed by the manufacturer. Bacmid DNA is a modified wild baculovirus DNA, which contains a lethal deletion and cannot develop into a viable virus by itself. Recombination between the flanking regions of the polyhedron gene from the donor vector and modified wild-type baculovirus DNA therefore results in 100% recombinant baculovirus DNA. To express the protein, 8 × 105 Sf9 cells were seeded in a 100 mm2 culture petridish and cotransfected with P3 virus stock. Purification, SDS-PAGE and Western blotting Infected Sf9 cells were harvested by scraping in 25 mM phosphate buffer saline (PBS) (pH 6.5), supplemented with 0.1% (v/v) Triton X-100 and protease inhibitor cocktail (Roche). Cells were homogenized by sonication, and crude lysate was applied to the ANTI-FLAG® M2 Affinity Gel (SIGMA), and purified proteins according to the manufacturers’ indications. The protein concentration was quantified spectrophotometrically at 280 nm using bovine serum albumin (BSA) as a standard. Proteins were separated by gel electrophoresis on 12% (w/v) SDS-PAGE gel under reducing conditions (with 5% 2-mercaptoethanol) according to the method described by Laemmli (23). Gels were stained with a Silver staining kit (Silver Stain Kit II, Wako Co., Japan). Precision Plus Protein™ Standards (Bio-Rad Laboratories, Inc., Hercules, CA) were used as molecular weight markers. Following gel electrophoresis, the proteins on the gel were transferred onto a polyvinylidene fluoride (PVDF) membrane (Amersham Hybond®-P; GE Healthcare UK Ltd., Buckinghamshire, UK). The membrane was blocked with Tris buffered saline (TBS) containing 0.05% (v/v) Tween® 20 and 1% BSA. The membrane was probed with the Anti-FLAG M2 monoclonal antibody (SIGMA) at room temperature for 2 h, and was further incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (1:10,000 dilution) (Santa Cruz Biotechnology, Inc., Dallas, TX) for 30 min. The detection of target proteins was performed using ECL SELECT™ chemiluminescence reagent Amersham Biosciences), according to the manufacturer’s protocol. Enzyme activity assay Assays of rFuc’ase Sl-2 activities were basically performed according to the methods described in our previous paper (18). Briefly, rFuc’ase Sl-2 activities during the purification step were assayed at 37°C by incubating with PA-labelled plant complex type N-glycans containing Lea determinants (Gal2Fuc2GN2M3FX) or LNFP III in 0.1 M sodium citrate buffer (pH 5.0). A reaction mixture of 50 μl containing about 10 pmol of substrate (Gal2Fuc2GN2M3FX or LNFP III), and 240 ng of rFuc’ase Sl-2 were used in the assay. The substrates and enzymatic products were analysed by SF-HPLC using a Shodex Asahipak NH2P-50 column (4.6 × 250 mm). The column was equilibrated with 80% (v/v) acetonitrile/water. The PA-sugar chains were eluted by increasing the water content of the water-acetonitrile mixture from 26 to 50% linearly at a flow rate of 0.7 ml/min, and were detected with a Jasco FP-920 Intelligent Fluorescence detector (Jasco, Inc., Easton, MD) (excitation at 310 nm, emission at 380 nm). The optimum pH of rFuc’ase Sl-2 was investigated using Gal2Fuc2GN2M3FX as a substrate at different pH levels ranging from 3.0–8.0. The buffers used were 0.1 M glycine–HCl (pH 3.0 and 3.5), 0.1 M sodium citrate (pH 4.0–5.0), 0.1 M MES (pH 6.0–7.0) and 0.1 M Tris–HCl (pH 8.0). The effects of metal ions (FeCl2, MnCl2, CoCl2, CaCl2, ZnCl2, CuCl2 and MgCl2) and EDTA on rFuc’ase Sl-2 activity were also investigated using Gal2Fuc2GN2M3FX as a substrate. The substrate specificity of rFuc’ase Sl-2 was analysed using authentic PA-sugar chains (10–20 pmol), LNFP I, LNFP III and PA-labelled N-glycans, Gal2Fuc2GN2M3FX, M3FX, MFX, MF, GN2F and GN1F. A reaction mixture of 50 μl for each substrate was incubated with rFuc’ase Sl-2 (240 ng) in 0.1 M sodium citrate buffer (pH 5.0) at 37°C for 4 h or overnight. When LNFP I, LNFP III, Gal2Fuc2GN2M3FX, M3FX and MFX were used as substrates, the products were analysed by SF-HPLC. When GNF and GN2F were used as substrates, the reaction mixture was analysed by RP-HPLC using a Cosmosil 5C18-AR column (4.6 × 250 mm). The column was equilibrated with 0.05% (v/v) trifluoroacetic acid (TFA)/water. The PA-sugar chains were eluted by increasing the acetonitrile content from 0 to 7% linearly at a flow rate of 1.2 ml/min. Sequence analysis, phylogenetic tree construction and homology modelling The signal peptide sequence of α-Fuc’ase Sl-2 was predicted using SignalP4.1 online tools (http://www.cbs.dtu.dk/services/SignalP/). The theoretical molecular mass and an isoelectric point (pI) were determined with the ExPASy proteomics server (http://web.expasy.org/cgi-bin/compute_pi/pi_tool). To identify the potential N-glycosylation sites, the online software NetGlyc 1.0 (http://www.cbs.dtu.dk/services/NetNGlyc/) was used. The conserved domains (CD) in the deduced amino acid sequence of α-Fuc’ase Sl-2 were identified by the CD-search tool (CDD V3.0-44354 PSSMs) available at the NCBI website (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). The phylogenetic tree, employing distance/neighbour-joining method, was constructed using the amino acid sequences of α-Fuc’ase Sl-2 with 10 characterized or deduced α-fucosidases from different sources (Blon_2336, Bifidobacterium longum; BT_2192, Bacteroides thetaiotaomicron; Ss_Fuc_142, Streptomyces sp.; BbAfcB, Bifidobacterium bifidum; α-Fuc’ase Os, Oryza sativa; Solyc-, S. lycopersicum; At_Fuc1, A. thaliana; CPF_2130, Clostridium perfringens; TfFuc1, Tannerella forsythia), which were mainly found in the CAZy database. We used the SWISS-MODEL web server (http://swissmodel.expasy.org/) and ModWeb server (https://modbase.compbio.ucsf.edu/modweb/) for homology modelling of α-Fuc'ase-Sl-2 according to the method described by Hossain et al. (24). The amino acid sequence of the second tomato α-fucosidase (α-Fuc'ase-Sl-2) composed of 518 residues (GenBank accession no: XM_004251058.2; GI: 723745655) was used to identify the template structures in the template library of the SWISS-MODEL server, and homology modelling was carried out in project mode. The online ModWeb Comparative Modelling Server version SVN.r1597 was also used for further modelling to compare with the models obtained from the SWISS-MODEL server. The qualities of the modelled structures were assessed using DFire (25), QMEAN (26), PROCHECK (27), WHAT_IF (28) methods and the ModEval Model evaluation server (29). The 3D-model structure of α-Fuc-Sl-2 was aligned with other crystal structures to ascertain the active site residues responsible for catalytic activity. COFACTOR software, a structure-based method for the biological function annotation of protein molecules, was used to identify functional insights, including ligand-binding sites and gene-ontology terms (30, 31). COACH, another software for binding site prediction, was used to compare the results obtained from COFACTOR (32). The RosettaDock server was used to dock the bona fide inhibitor of α-fucosidase, deoxyfuconojirimycin (33). UCSF Chimera and SWISS-PdbViewer were used to view the models and prepare images. Results and Discussion Construction of the α-Fuc’ase Sl-2 gene expression vector We previously cloned, expressed, and functionally-characterized an α-fucosidase (α-Fuc’ase Sl-1) from tomato (18). In this study, we cloned and expressed another α-fucosidase gene, Solyc11g069010, which is supposed to have α1, 3/4- fucosidase activity towards Lea epitopes and core α1, 3-fucose residues on complex type N-glycans. For this, an α-Fuc’ase Sl-2 gene expression vector (baculovirus expression vector) was constructed. Insert sequence and size of the expression vector (bacmid) were confirmed through sequencing and restriction digestion methods, respectively (data not shown). The nucleotide sequence of the α-Fuc’ase Sl-2 gene is shown in Fig. 1. Fig. 1 View largeDownload slide Deduced amino acid sequence from the coding sequence of Solyc11g069010, α-Fuc’ase Sl-2. The primers used to amplify and sequencing of the coding sequence (2073 bp) are indicated with underlined letters. α-Fuc’ase Sl-2 consists of 518 amino acid residues, and the N-terminal 26 amino acid residues that encode a predicted signal peptide are underlined. The signal peptide cleavage site is shown by a vertical arrow. The predicted N-glycosylation sites (N-X-S/T) are represented in an underlined bold font. Fig. 1 View largeDownload slide Deduced amino acid sequence from the coding sequence of Solyc11g069010, α-Fuc’ase Sl-2. The primers used to amplify and sequencing of the coding sequence (2073 bp) are indicated with underlined letters. α-Fuc’ase Sl-2 consists of 518 amino acid residues, and the N-terminal 26 amino acid residues that encode a predicted signal peptide are underlined. The signal peptide cleavage site is shown by a vertical arrow. The predicted N-glycosylation sites (N-X-S/T) are represented in an underlined bold font. Production of rFuc’ase Sl-2 by Sf9 cells and its purification Sf9 cells were cotransfected with the baculovirus expression vector containing the α-Fuc’ase Sl-2 sequence, which was inserted downstream of the polyhedron promoter. Cells were grown as described in the Materials and Methods. We carried out a time-course study to optimize the expression level of the gene, and found that 72 h post-infection was the proper expression time. As the α-Fuc’ase Sl-2 expression construct contained a FLAG-tag in its C-terminal, we successfully purified the enzyme by using ANTI-FLAG® M2 Affinity Gel. The purified protein (rFuc’ase Sl-2) from cell extracts showed a single protein band with a molecular weight of about 65 kDa when analysed with SDS-PAGE (Fig. 2A). By using western blot analysis, the expected protein band was detected from the purified protein (Fig. 2B, L1) and crude enzyme (Fig. 2B, L2) using ANTI-FLAG® M2 monoclonal antibody. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s) that did not bind to the affinity gel tightly but was recognized by the antibody. No band was detected from infected culture (Fig. 2B, L3) medium or from unprocessed cells (Fig. 2B, L4), suggesting rFuc’ase Sl-2 is not secreted out of culture broth or is not a secretory protein. Fig. 2 View largeDownload slide SDS-PAGE and western blot analysis of rFuc’ase Sl-2. (A) The purified rFuc’ase Sl-2 from the culture medium was subjected to 12% polyacrylamide gel electrophoresis under reducing conditions. The gel was stained with Silver stain. M stands for marker proteins of Precision Plus Protein™ Standards. (B) Purified rFuc’ase Sl-2 from the culture medium was applied to the 12% polyacrylamide gel electrophoresis under reducing conditions. The protein on the gel was transferred into the polyvinylidene fluoride (PVDF) membrane for western blot analysis. The membrane was probed with primary antibody (anti-FLAG monoclonal antibody) and was then incubated with secondary antibody (goat anti-mouse IgG-HRP) as described in the text. A ∼60 kDa protein band of rFuc’ase Sl-2 was detected by Image Quant LAS 500. M, molecular markers; L1, purified rFuc’ase Sl-2; L2, crude lysate of infected Sf9 cells; L3, uninfected Sf9 cell lysate; L4, culture medium of infected Sf9 cells. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s). Fig. 2 View largeDownload slide SDS-PAGE and western blot analysis of rFuc’ase Sl-2. (A) The purified rFuc’ase Sl-2 from the culture medium was subjected to 12% polyacrylamide gel electrophoresis under reducing conditions. The gel was stained with Silver stain. M stands for marker proteins of Precision Plus Protein™ Standards. (B) Purified rFuc’ase Sl-2 from the culture medium was applied to the 12% polyacrylamide gel electrophoresis under reducing conditions. The protein on the gel was transferred into the polyvinylidene fluoride (PVDF) membrane for western blot analysis. The membrane was probed with primary antibody (anti-FLAG monoclonal antibody) and was then incubated with secondary antibody (goat anti-mouse IgG-HRP) as described in the text. A ∼60 kDa protein band of rFuc’ase Sl-2 was detected by Image Quant LAS 500. M, molecular markers; L1, purified rFuc’ase Sl-2; L2, crude lysate of infected Sf9 cells; L3, uninfected Sf9 cell lysate; L4, culture medium of infected Sf9 cells. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s). Sequence interpretation and phylogenetic study α-Fuc’ase Sl-2 gene encodes a 518-amino-acid-long polypeptide chain, of which, the first 26 N-terminal amino acids form the predicted signal peptide (Fig. 1). A putative cleavage site resides in between 26th and 27th residues. The theoretical molecular mass and pI were 59.47 kDa and 5.69, respectively. The molecular mass (65 kDa) determined by SDS-PAGE was slightly bigger than that of the theoretical mass, probably due to the addition of N-glycan(s). In addition, the polypeptide chain contains six potential N-glycosylation sites (N-X-S/T) at residue numbers 83, 249, 322, 366, 440 and 511. A conserved domain database (CDD) search found two conserved domains for α-Fuc’ase Sl-2, namely, the COG3669 domain (carbohydrate transport and metabolism, 43-479) and cl07893 domain (alpha amylase catalytic domain, 77-337), which belong to GH family 29. A neighbour-joining phylogenetic tree was prepared (Fig. 3). The close members of α-Fuc’ase Sl-2 in the clades are mainly characterized as active toward α-1, 3/1, 4-fucosidic linkage, suggesting that α-Fuc’ase Sl-2 might have a similar substrate specificity and play a similar role in the turnover of N-glycoproteins in plant cells. The sequence homology was analysed by the NCBI protein BLAST tool, and found a higher homology between α-Fuc’ase Sl-2 and AtFuc1 (69%) than between α-Fuc’ase Sl-2 and α-Fuc’ase Sl-1 (51%). Fig. 3 View largeDownload slide Phylogenetic tree of GH29 family α-Fuc’ases. The phylogenetic tree was drawn using the amino acid sequence of α-Fuc’ase Sl-2 with nine other characterized or deduced GH 29 family α-fucosidases from different sources (plant/bacteria/fungi etc.) found in the CAZy database. The names of the organisms are described in the text. Fig. 3 View largeDownload slide Phylogenetic tree of GH29 family α-Fuc’ases. The phylogenetic tree was drawn using the amino acid sequence of α-Fuc’ase Sl-2 with nine other characterized or deduced GH 29 family α-fucosidases from different sources (plant/bacteria/fungi etc.) found in the CAZy database. The names of the organisms are described in the text. Functional properties The activity of rFuc’ase Sl-2 against Gal2Fuc2GN2M3FX was found optimal at a pH of around 5.0 (Supplementary Fig. S2), suggesting that the enzyme resides and functions in acidic organelle such as the vacuole or cell wall. The HPLC profiles of the rFuc’ase Sl-2 activity assay against the Gal2Fuc2GN2M3FX substrate at different pH levels (pH 4.0, 5.0 and 7.0) are shown in Fig. 4A and B. The optimum pH of rFuc’ase Sl-2 was found to be similar to those of α-fucosidases from almond (15, 16), Arabidopsis (AtFuc1) (16), rice (α-Fuc’ase Os) (17) and tomato (α-Fuc’ase Sl-1) (18). Fig. 4 View largeDownload slide Activities of rFuc’ase Sl-2 toward N-glycans, Gal2Fuc2GN2M3FX-PA. (A) A reaction mixture (50 μl) of rFuc’ase Sl-2 (240 ng) + substrate (10 pmol) + 0.1 M citrate buffer (pH 5.0) was incubated at 37°C for 1 h and was analysed by size-fractionation (SF)-HPLC. (B) Substrate with the same volume of reaction mixture and different pH levels (4.0, 5.0 and 6.0) was incubated at 37°C for 4 h and was analysed by SF-HPLC. Fig. 4 View largeDownload slide Activities of rFuc’ase Sl-2 toward N-glycans, Gal2Fuc2GN2M3FX-PA. (A) A reaction mixture (50 μl) of rFuc’ase Sl-2 (240 ng) + substrate (10 pmol) + 0.1 M citrate buffer (pH 5.0) was incubated at 37°C for 1 h and was analysed by size-fractionation (SF)-HPLC. (B) Substrate with the same volume of reaction mixture and different pH levels (4.0, 5.0 and 6.0) was incubated at 37°C for 4 h and was analysed by SF-HPLC. For the investigation on the effect of metal ions, various divalent metal cations (FeCl2, MnCl2, CoCl2, CaCl2, ZnCl2, CuCl2 and MgCl2) and EDTA were added at final concentrations of 5 mM to the reaction mixture. The rFuc’ase Sl-2 activity was not stimulated explicitly with the addition of metal ions or EDTA; Fe2+, and Cu2+ rather utterly inhibited the enzyme activity (Supplementary Table SII), suggesting that these metal ions might co-ordinate with some essential amino acids in the catalytic site. Substrate specificity of rFuc’ase Sl-2 The substrate specificity of rFuc’ase Sl-2 was investigated using several N-glycans or other oligosaccharides bearing α-fucosyl residue as substrates. The products and the remaining substrates of the enzymatic reaction were analysed by SF-HPLC or RP-HPLC, and the results are summarized in Table I. We confirmed that the rFuc’ase Sl-2 substantially hydrolyzed the α1, 4-Fuc residue on the Lea epitopes of plant complex type N-glycans (Fig. 4A) and the α1, 3-fucose residue on LNFP III. No activity, even at a high enzyme concentration, was noticed with α1, 2-fucosyllactose and p-nitrophenol-α-L-fucopyranoside (pNP-Fuc) as substrates. The substrate specificity of rFuc’ase Sl-2 was quite similar to those of the almond fucosidase I (15, 16), Arabidopsis fucosidase (AtFuc1) (16) and rFuc’ase Sl-1 (18). However, when we used GlcNAcβ1-4(Fucα1-3)GlcNAcPA (GN2F) and Fucα1-3GlcNAc-PA (GN1F) as substrates, we found that rFuc’ase Sl-2 hydrolyzed the α1, 3-fucosyl linkage on GN2F (Fig. 5A) but not on GN1F (Fig. 5B), indicating that the β1-4 linked GlcNAc residue can be a substitute for β1-4Gal residue linked to the β-GlcNAc residue in LNFP III for fixing the α1-3Fuc residue in the catalytic site. Furthermore, in addition to rFuc’ase Sl-2, we also confirmed that rFuc’ase Sl-1 (18) also hydrolyzed the α1, 3-fucosyl linkage on GN2F (data not shown). However, the reaction rates of these two tomato α-Fuc’ase specimens toward GN2F (GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) was much slower than those toward the Lea unit in the plant complex type N-glycan and LNFP III, indicating that the pyridylamino group attached to the proximal GlcNAc or the open-ring form of the GlcNAc might wield a negative influence on the enzyme activity. Table I. Substrate specificity of α-Fuc’ase Sl-2     ND: Any products were not detected under the analytical condition described in the text. Table I. Substrate specificity of α-Fuc’ase Sl-2     ND: Any products were not detected under the analytical condition described in the text. Fig. 5 View largeDownload slide Activities of rFuc’ase Sl-2 toward GN2F and GN1F. A reaction mixture (50 μl) consisting of rFuc’ase Sl-2 (240 ng), substrates (10 pmol) and 0.1 M citrate buffer (pH 5.0) was incubated at 37°C overnight, and the products were analysed by reversed-phase (RP)-HPLC using a Cosmosil 5C18-AR column. (A) HPLC profile of GN2F incubated with rFuc’ase Sl-2. (B) HPLC profile of GN1F incubated with rFuc’ase Sl-2. GN2, GlcNAcβ1-4GlcNAc-PA; GN, GlcNAc-PA. Some peaks marked with asterisks were not N-glycans, but rather contaminative fluorescence substances. Fig. 5 View largeDownload slide Activities of rFuc’ase Sl-2 toward GN2F and GN1F. A reaction mixture (50 μl) consisting of rFuc’ase Sl-2 (240 ng), substrates (10 pmol) and 0.1 M citrate buffer (pH 5.0) was incubated at 37°C overnight, and the products were analysed by reversed-phase (RP)-HPLC using a Cosmosil 5C18-AR column. (A) HPLC profile of GN2F incubated with rFuc’ase Sl-2. (B) HPLC profile of GN1F incubated with rFuc’ase Sl-2. GN2, GlcNAcβ1-4GlcNAc-PA; GN, GlcNAc-PA. Some peaks marked with asterisks were not N-glycans, but rather contaminative fluorescence substances. Resolved 3 D-modelled structure and mechanism of action The overall 3D modelled structure of α-Fuc’ase Sl-2 is shown in Fig. 6A, which consists of 443 amino acids (37–479) in its structural part. The modelled structure contains two domains: an N-terminal GH29 α-fucosidase domain containing a (β/α)8 TIM barrel, and a C-terminal carbohydrate binding module (CBM) 32 domain with a β-sandwich structure. The CD database search suggested that the α-Fuc’ase Sl-2 consists of three characteristic domains: (1) the α-amylase catalytic domain family, (2) the F5/8 type C domain (discoidin domain family) and (3) the COG3669 domain. Two domains, the α-amylase catalytic domain family (amino acids: 77–337) and the COG3669 domain (amino acids: 43-479), overlap each other on the amino acid sequence, and these two domains could not be shown together with different colours in Fig. 6A. The modelled structure contains two domains, as described in the ‘Sequence interpretation and phylogenetic study’ section. The structural features of α-Fuc’ase Sl-2 were compared upon superimposition on the template (pdb id: 3uesA) structure, and showed close resemblance between the two structures (data not shown). The predicted ligand binding sites of α-Fuc'ase-Sl-2, from the superimposition, were found to be Phe54, His56, Trp67, His105, His106, Tyr151, Asp192, Ala194, Glu236, Asp280 and Trp287. The ligand binding sites of α-Fuc'ase-Sl-2 were further assessed with a potential α-fucosidase inhibitor, deoxyfuconojirimycin (DFU), using the RosettaDock server and the residues were identified as Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288, which is shown in a generated space filled view in Fig. 6B. Finally, the modelled structure of α-Fuc'ase-Sl-2 was aligned with the template to identify the predicted catalytic nucleophile and catalytic acid/base residue. From the alignment view (Fig. 6C), the residues of Asp192 and Glu236 in the α-Fuc’ase Sl-2 modelled structure were the catalytic proton donor and catalytic nucleophile/base, respectively, when compared with the location of corresponding residues on the template (Fig. 6C). Multiple sequence alignment of the amino acid sequences of α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Bifidobacterium longum α-fucosidase (pdb: 3uesA), revealed that the catalytically important residues were also conserved among them as shown in Fig. 7, which indicates a similar mode of action to their native substrates. Fig. 6 View largeDownload slide 3 D modelled structure of α-Fuc’ase Sl-2 resolved by the SWISS-MODEL web server. (A) Ribbon diagram of the overall 3D modelled structure of tomato α-Fuc’ase-Sl-2. The catalytic domain is shown by various colours, from the blue at the N terminus to yellowish green at the C terminus. The strands and helices that form the (β/α)8 barrel are labelled in red. The C-terminal carbohydrate-binding (β-sandwich, F5/8 type C) domain is shown in red, with 8-strands labelled in blue letters. The N and C termini of modelled structure are also labelled. (B) Space filled view (hydrophobic surface) of the modelled structure (α-Fuc’ase-Sl-2) with docked DFU (Deoxyfucojirinomycin). Active site residues (Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288) interacted with ligand (DFU) are coloured black. The DFU located in the active site cleft is indicated as an interactive colour stick. Chimera was used to prepare the image. (C) Structural alignments of the active sites of α-Fuc’ase-Sl-2 (modelled structure) and 3ues (Bifidobacterium longum subsp. infantis) in complex with DFU (Deoxyfuconojirimycin). All active site residues and the DFU from both the 3ues- and α-Fuc’ase-Sl-2-DFU complexes are drawn in stick format. The residues of modelled α-Fuc’ase-Sl-2 are shown in green colour. The two DFUs resided in the centre position of both 3ues and α-Fuc’ase-Sl-2 are hydrogen bonded (cyan colour) with active site residues of the two structures. The strands and helices that form the (β/α)8 TIM barrel are labelled in green (α-Fuc’ase-S2) and red (3ues). Only catalytically important residue numbers are shown in the figure; other residues are not shown for the sake of clarity. Chimera was used to prepare the image. Fig. 6 View largeDownload slide 3 D modelled structure of α-Fuc’ase Sl-2 resolved by the SWISS-MODEL web server. (A) Ribbon diagram of the overall 3D modelled structure of tomato α-Fuc’ase-Sl-2. The catalytic domain is shown by various colours, from the blue at the N terminus to yellowish green at the C terminus. The strands and helices that form the (β/α)8 barrel are labelled in red. The C-terminal carbohydrate-binding (β-sandwich, F5/8 type C) domain is shown in red, with 8-strands labelled in blue letters. The N and C termini of modelled structure are also labelled. (B) Space filled view (hydrophobic surface) of the modelled structure (α-Fuc’ase-Sl-2) with docked DFU (Deoxyfucojirinomycin). Active site residues (Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288) interacted with ligand (DFU) are coloured black. The DFU located in the active site cleft is indicated as an interactive colour stick. Chimera was used to prepare the image. (C) Structural alignments of the active sites of α-Fuc’ase-Sl-2 (modelled structure) and 3ues (Bifidobacterium longum subsp. infantis) in complex with DFU (Deoxyfuconojirimycin). All active site residues and the DFU from both the 3ues- and α-Fuc’ase-Sl-2-DFU complexes are drawn in stick format. The residues of modelled α-Fuc’ase-Sl-2 are shown in green colour. The two DFUs resided in the centre position of both 3ues and α-Fuc’ase-Sl-2 are hydrogen bonded (cyan colour) with active site residues of the two structures. The strands and helices that form the (β/α)8 TIM barrel are labelled in green (α-Fuc’ase-S2) and red (3ues). Only catalytically important residue numbers are shown in the figure; other residues are not shown for the sake of clarity. Chimera was used to prepare the image. Fig. 7 View largeDownload slide Sequence alignment of α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 with its homolog Blon_Fuc. Sequence alignments among α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Blon_Fuc from B. longum subsp. infantis were performed using the ClustalW program. Predicted ligand binding amino acids at the active site of α-Fuc’ase-Sl-2, α-Fuc’ase-Sl-1 and Blon_Fuc described in the text are conserved, and are indicated as boxes. Fig. 7 View largeDownload slide Sequence alignment of α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 with its homolog Blon_Fuc. Sequence alignments among α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Blon_Fuc from B. longum subsp. infantis were performed using the ClustalW program. Predicted ligand binding amino acids at the active site of α-Fuc’ase-Sl-2, α-Fuc’ase-Sl-1 and Blon_Fuc described in the text are conserved, and are indicated as boxes. Sakurama et al. reported that two microbial α-fucosidases, B. thetaiotaomicron enzyme (BT_2192) (34) and B. bifidum enzyme (BbAfcB) (35) belong to GH29-B, which is active toward the α1, 3/4-fucosyl linkages in Lewis x and Lea epitopes, but not toward α1, 2-fucosyl linkages nor pNP-Fuc (35). Based on the substrate specificity, the tomato α-Fuc’ase Sl-2, as well as α-Fuc’ase Sl-1 (18) and rice α-Fuc’ase (OsFuc1) (17), seemed to belong to GH29-B but not GH29-A, such as in L. caseii (AlfA, AlfB, and AlfC) (36). Both rFuc’ase Sl-1 and rFuc’ase Sl-2 could not hydrolyze MFX, MF and GN1F, but were active against α1, 3/4-fucosyl linkages in LFNP III and the Lea epitopes, indicating a critical role of the non-reducing terminal β-Gal residue of these substrates. In the case of BT_2192, it has also been found that a β-Gal-binding site consisting of Trp230, Glu254 and Asp277 plays a critical role in hydrolytic activity toward the Lea epitopes, and the GN1F structure, lacking a Gal residue, cannot be a substrate for BT_2192 (34, 35). As described above, in this study, we confirmed that both α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 can hydrolyze the α1-3 fucosyl linkage in GN2F, indicating that the β1-4 GlcNAc residue can be a substitute for β1-4Gal residue linked to the β-GlcNAc residue in LNFP III for fixing the α1-3 Fuc residues in the catalytic site. Recently, Kato et al. found that one of the Arabidopsis α-Fuc’ases, AtFuc1, hydrolyzed the α1-3 fucosyl linkage in GNF2 and the reaction rate of AtFuc1 toward the GN2F was much faster than that toward GN2F-PA (36). These results suggest that GH29-B α-Fuc’ases must require the β1-4 linkage Gal or GlcNAc residue that is linked to the GlcNAc carrying the α1-3/4 Fuc residue, and supported the hypothetical hydrolytic mechanism based on the 3 D modelled structures of the two tomato α-Fuc’ases. Furthermore, the substrate specificity of these two tomato α-Fuc’ase specimens, α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2, indicates that the complete degradation of plant specific N-glycans requires defucosylation prior to β-GlcNAc’ase acting toward the GlcNAcβ1-4GlcNAc(-Asn) unit. Supplementary Data Supplementary Data are available at JB Online. Acknowledgements The authors are grateful to the Department of Instrumental Analysis, Advanced Science Research Center, Okayama University, for ESI-MS analysis. The authors would like to thank Editage (www.editage.jp) for English language editing. Funding This work was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Basic Research C [no. 24580494 to M.M. and no. 17K0819709 to YK)] and Fostering Joint International Research (no. 15K0784 to M.M.) and the Research Grants of Mizutani Foundation for Glycoscience (to Y.K.). Conflict of Interest None declared. References 1 Lerouge P., Cabanes-Macheteau M., Rayon C., Fischette-Lainé A.-C., Gomord V., Faye L. ( 1998) N-glycoprotein biosynthesis in plants: recent developments and future trends. Plant Mol. Biol . 38, 31– 48 Google Scholar CrossRef Search ADS PubMed  2 Kobata A. ( 2007) Glycoprotein glycan structures in Comprehensive Glycoscience–from Chemistry to Systems Biology , Vol. 1. ( Kamerling H., ed.) pp. 39– 72, Elsevier, Oxford Google Scholar CrossRef Search ADS   3 Schoberer J., Strasser R. ( 2011) Sub-compatmental organization of Golgi-resident N-glycan processing enzymes in plants. Mol. Plant.  4, 220– 228 Google Scholar CrossRef Search ADS PubMed  4 Strasser R. ( 2016) Plant protein glycosylation. Glycobiology  26, 926– 939 Google Scholar CrossRef Search ADS PubMed  5 Maeda M., Kimura Y. ( 2014) Structural features of free N-glycans occurring in plants and functional features of de-N-glycosylation enzymes, ENGase and PNGase. Y. Front. Plant Sci  5, 429 6 Ishimizu T. ( 2015) Plant N-glycans and their degrading enzymes in Glycoscience: Biology and Medicine . ( Endo T., Seeberger P.H., Hart G.W., Wong C.-H., Taniguchi N., eds.) pp. 1111– 1117, Springer, Tokyo Google Scholar CrossRef Search ADS   7 Fitchette-Laine A.-C., Gomord V., Cabanes M., Michalski J.-C., Macary M., Foucher B., Cavelier B., Hawes C., Lerouge P., Faye L. ( 1997) N-glycans harboring the Lewis a epitope are expressed at the surface of plant cells. Plant J.  12, 1411– 1417 Google Scholar CrossRef Search ADS PubMed  8 Melo S., Nimtz M., Conradt H.S., Fevereiro P.S., Costa J. ( 1997) Identification of the human Lewisa carbohydrate motif in a secretory peroxidase from a plant cell suspension culture (Vaccinium myrtillus L.). FEBS Lett . 415, 186– 191 Google Scholar CrossRef Search ADS PubMed  9 Maeda M., Kimura M., Kimura Y. ( 2010) Intracellular and extracellular free N-glycans produced by plant cells: occurrence of unusual plant complex-type free N-glycans in extracellular spaces. J. Biochem . 148, 681– 692 Google Scholar CrossRef Search ADS PubMed  10 Maeda M., Ebara N., Tani M., Vavricka J.C., Kimura Y. ( 2017) Occurrence of plant complex type free N-glycans carrying a single GlcNAc residue at the reducing end in a fresh-water plant, Egeria densa: a possible mechanism for biosynthesis of free N-glycans in plants. Glycoconj. J.  34, 229– 240 Google Scholar CrossRef Search ADS PubMed  11 Augur C., Benhamou N., Darvill A., Albersheim P. ( 1993) Purification, characterization and cell wall localization of an alpha-fucosidase that inactivsties a xyloglucan oligosaccharin. Plant J.  3, 415– 426 Google Scholar CrossRef Search ADS PubMed  12 Gossarau R. ( 1991) Catalytic histochemistry of acid and newtral hydrolases in plant seedlings. Histochem. J . 23, 483– 489 Google Scholar CrossRef Search ADS PubMed  13 Iglesias N., Abelenda J.A., Rodino M., Sampedro J., Revilla G., Zarra I. ( 2006) Apoplastic glycosidases active against xyloglucan oligosaccharides of Arabidopsis thaliana. Plant Cell Physiol . 47, 55– 63 Google Scholar CrossRef Search ADS PubMed  14 Günl M., Neumetzler L., Kraemer F., de Souza A., Schultink A., Pena M., York W.S., Pauly M. ( 2011) AXY8 encodes an α-fucosidase, underscoring the importance of apoplastic metabolism on the fine structure of Arabidopsis cell wall polysaccharides. Plant Cell  23, 4025– 4040 Google Scholar CrossRef Search ADS PubMed  15 Ogata-Arakawa M., Muramatsu T., Kobata A. ( 1977) α-L-fucosidases from almond emulsin: characterization of the two enzymes with different specificities. Arch. Biochem. Biophys . 181, 353– 358 Google Scholar CrossRef Search ADS PubMed  16 Zeleny R., Leonard R., Dorfner G., Dalik T., Kolarich D., Altmann F. ( 2006) Molecular cloning and characterization of a plant α1,3/4-fucosidase based on sequence tags from almond fucosidase I. Phytochemistry  67, 641– 648 Google Scholar CrossRef Search ADS PubMed  17 Rahman M.Z., Fujishige M., Maeda M., Kimura Y. ( 2016) Rice α-fucosidase active against plant complex type N-glycans containing Lewis a epitope: purification and characterizeation. Biosci. Biotechnol. Biochem . 80, 291– 294 Google Scholar CrossRef Search ADS PubMed  18 Rahman M.Z., Maeda M., Itano S., Hossain M.A., Ishimizu T., Kimura Y. ( 2017) Molecular characterization of tomato α1,3/4-fucosidase, a member of glycosyl hydrolase family 29, involved in degradation of plant complex type N-glycans. J. Biochem . 161, 421– 432 Google Scholar PubMed  19 Sakurama H., Fushinobu S., Hidaka M., Yoshida E., Honda Y., Ashida H., Kumagi H., Yamamoto K., Katayama T. ( 2012) 1,3-1,4-α-L-fucosynthase that specifically introduces Lewis a/x antigens into type-1/2 chains. J. Biol. Chem . 287, 16709– 16719 Google Scholar CrossRef Search ADS PubMed  20 Maeda M., Kimura Y. ( 2006) Glycoform analysis of N-glycans linked to glycoproteins expressed in rice culture cells: predominant occurrence of complex type N-glycans. Biosci. Biotechnol. Biochem . 70, 1356– 1363 Google Scholar CrossRef Search ADS PubMed  21 Kimura Y., Minami Y., Tokuda T., Nakajima S., Takagi S., Funatsu G. ( 1991) Primary structures of N-linked oligosaccharides of momordin-a, ribosome-inactivating protein from Momordica charantia seeds. Agric. Biol. Chem . 55, 2031– 2036 Google Scholar CrossRef Search ADS PubMed  22 Yokouchi D., Ono N., Nakamura K., Maeda M., Kimura Y. ( 2013) Purification and characterization of β-xylosidase that is active for plant complex type N-glycans from tomato (Solanum lycopersicum): removal of core α1-3 mannosyl residue is prerequisite for hydrolysis of β1-2 xylosyl residue. Glycoconj. J.  30, 463– 472 Google Scholar CrossRef Search ADS PubMed  23 Laemmli U.K., Favre M. ( 1973) Maturation of the head of bacteriophage T4. J. Mol. Biol.  80, 575– 599 Google Scholar CrossRef Search ADS PubMed  24 Hossain M.A., Roslan H.A., Karim M.R., Kimura Y. ( 2016) Molecular phylogeny, 3D-structural insights, docking and mechanisms of action of plant beta-galactosidase. Int. J. Bioinform. Res. Appl . 12, 149– 179 Google Scholar CrossRef Search ADS   25 Pieper U., Webb B.M., Barkan D.T., Schneidman-Duhovny D., Schlessinger A., Braberg H., Yang Z., Meng E.C., Pettersen E.F., Huang C.C., Datta R.S., Sampathkumar P., Madhusudhan M.S., Sjolander K., Ferrin T.E., Burley S.K., Sali A. ( 2011) MODBASE, a database of annotated comparative protein structure models and associated resources. Nucleic Acids Res . 39, D465– D474 Google Scholar CrossRef Search ADS PubMed  26 Zhou H., Zhou Y. ( 2009) Distance-scaled, finite ideal-gas reference state improves structure-derived potentials of mean force for structure selection and stability prediction. Protein Sci . 11, 2714– 2726 Google Scholar CrossRef Search ADS   27 Benkert P., Biasini M., Schwede T. ( 2011) Toward the estimation of the absolute quality of individual protein structure models. Bioinformatics  27, 343– 350 Google Scholar CrossRef Search ADS PubMed  28 Laskowski R.A., MacArthur M.W., Moss D.S., Thornton J.M. ( 1993) PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr.  26, 283– 291 Google Scholar CrossRef Search ADS   29 Hooft R.W., Vriend G., Sander C., Abola E.E. ( 1996) Errors in protein structures. Nature  381, 272 Google Scholar CrossRef Search ADS PubMed  30 Melo F., Sanchez R., Sali A. ( 2002) Statistical potentials for fold assessment. Protein Sci . 11, 430– 444 Google Scholar CrossRef Search ADS PubMed  31 Roy A., Yang J., Zhang Y. ( 2012) COFACTOR: an accurate comparative algorithm for structure-based protein function annotation. Nucleic Acids Res . 40, W471– W477 Google Scholar CrossRef Search ADS PubMed  32 Yang J., Roy A., Zhang Y. ( 2013) BioLiP: a semi-manually curated database for biologically relevant ligand-protein interactions. Nucleic Acids Res . 41, D1096– D1103. Google Scholar CrossRef Search ADS PubMed  33 Lyskov S., Gray J.J. ( 2008) The RosettaDock server for local protein-protein docking. Nucleic Acids Res . 36, W233– W238 Google Scholar CrossRef Search ADS PubMed  34 Sakurama H., Tsutsumi E., Ashida H., Katayama T., Yamamoto K., Kumagai H. ( 2012) Differences in the substrate specificities and active-site structures of two α-L-fucosidases (glycoside hydrolase family 29) from Bacteroids thetaiotaomicron. Biosci. Biotechnol. Biochem . 76, 1022– 1024 Google Scholar CrossRef Search ADS PubMed  35 Ashida H., Miyake A., Kiyohara M., Wada J., Yoshida E., Kumagai H., Katayam T., Yamamoto K. ( 2009) Two distinct α-L-fucosidases from Bifidobacterium bifidum are essential for the utilization of fucosylated milk oligosaccharides and glycoconjugates. Glycobiology  19, 1010– 1017 Google Scholar CrossRef Search ADS PubMed  36 Kato S., Hayashi M., Kitagawa M., Maeda M., Kimura Y., Igarashi K., Kajiura H., Kasahara M., Ishimizu T. ( 2018) Degradation pathway of plant complex-type N-glycans: identification and characterization of a key α1,3-fucosidase from glycoside hydrolase family 29. Biochem. J . 475, 305– 317 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations Fuc L-fucose Gal D-galactose GlcNAc N-acetyl-D-glucosamine Glc D-glucose GN1M3FX GlcNAcβ1-2Manα1-6(Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) GN2M3FX GlcNAcβ1-2Manα1-6(GlcNAcβ1-2Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA GN2F GlcNAcβ1-4(Fucα1-3)GlcNAc-PA GN1F Fucα1-3GlcNAc-PA GN2F Fucα1-3GlcNAcβ1-4GlcNAc-PA Lacto-N-fucopentaose I Fucα1-2Galβ1-3GlcNAcβ1-3Galβ1-4Glc-PA Lacto-N-fucopentaose III Galβ1-4(Fucα1-3)GlcNAcβ1-3Galβ1-4Glc-PA (Lea)1GN1M3FX Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-6(GlcNAcβ1-2Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) (Lea)2M3FX Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-6(Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-3)(Xylβ1-2) Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA Lea Lewis a Man D-mannose MF Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA MFX Xylβ1-2Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA M3FX Manα1-6(Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucβ1-3)GlcNAc-PA PA pyridylamino PTC plant complex type RP-HPLC reversed-phase HPLC rFuc’ase Sl recombinant α-fucosidase from Solanum lycopersicum SF-HPLC size-fractionation HPLC Xyl D-xylose α-Fuc’ase Sl α-fucosidase from Solanum lycopersicum 3uesA α1, 3/4-fucosidase from Bifidobacterium longum © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Biochemistry Oxford University Press

Molecular characterization of second tomato α1,3/4-fucosidase (α-Fuc’ase Sl-2), a member of glycosyl hydrolase family 29 active toward the core α1,3-fucosyl residue in plant N-glycans

Loading next page...
 
/lp/ou_press/molecular-characterization-of-second-tomato-1-3-4-fucosidase-fuc-ase-eg4WEqLED0
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
ISSN
0021-924X
eISSN
1756-2651
D.O.I.
10.1093/jb/mvy029
Publisher site
See Article on Publisher Site

Abstract

Abstract In a previous study, we molecular-characterized a tomato (Solanum lycopersicum) α1, 3/4-fucosidase (α-Fuc’ase Sl-1) encoded in a tomato gene (Solyc03g006980), indicating that α-Fuc’ase Sl-1 is involved in the turnover of Lea epitope-containing N-glycans. In this study, we have characterized another tomato gene (Solyc11g069010) encoding α1, 3/4-fucosidase (α-Fuc’ase Sl-2), which is also active toward the complex type N-glycans containing Lea epitope(s). The baculovirus-insect cell expression system was used to express that α-Fuc’ase Sl-2 with anti-FLAG tag, and the expression product (rFuc’ase Sl-2), was found as a 65 kDa protein using SDS-PAGE and has an optimum pH of around 5.0. Similarly to rFuc’ase Sl-1, rFuc’ase Sl-2 hydrolyzed the non-reducing terminal α1, 3-fucose residue on LNFP III and α1, 4-fucose residues of Lea epitopes on plant complex type N-glycans, but not the core α1, 3-fucose residue on Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc or Fucα1-3GlcNAc. However, we found that both α-Fuc’ases Sl-1 and Sl-2 were specifically active toward α1, 3-fucose residue on GlcNAcβ1-4(Fucα1-3)GlcNAc, indicating that the non-substituted β-GlcNAc linked to the proximal GlcNAc residue of the core tri-saccharide moiety of plant specific N-glycans must be a pre-requisite for α-Fuc’ase activity. A 3 D modelled structure of the catalytic sites of α-Fuc’ase Sl-2 suggested that Asp192 and Glu236 may be important for binding to the α1, 3/4 fucose residue. α-fucosidase, α1-3 fucose, N-glycan metabolism, plant N-glycan, Solanum lycopersicum Almost all the secreted-type proteins produced in eukaryotic cells are N-glycosylated, suggesting that the N-glycosylation of proteins and the subsequent modification of the glycan moiety are ubiquitous and pivotal biological processes in eukaryotes, especially the multicellular organisms (1, 2). Plant N-glycans are usually classified into four subgroups: high-mannose type, hybrid type, biantennary-complex type and pauci-mannose type. The final structure of these N-glycans, however, varies by species due to the action of a unique set of processing enzymes acting on N-glycans in the secretory pathway of plant cells (3, 4). Plant complex-type N-glycans carry a distinctive attribute with α1, 3-linked fucose residue to the proximal GlcNAc residue and β1, 2-linked xylose residue to β-mannose residue (4–6). Some secreted glycoproteins contain large complex-type N-glycans bearing Lewis a (Lea) epitope (Galβ1-3(Fucα1-4)GlcNAc) at their non-reducing end (7, 8). In addition, we have found that the rice or water plant (Egeria densa) contain various complex type free N-glycans (FNGs), GN2-type FNGs (GN1M3FX, GN2M3FX, Gal1Fuc1GN1M3FX, and Gal2Fuc2GN2M3FX) and GN1-type FNGs (GN1M3X-GN1, Gal1Fuc1GN1M3X-GN1, Gal2Fuc2GN2M3X-GN1) (5, 9, 10). The occurrence of these defucosylated/secreted type FNGs suggests that α-fucosidase must be involved in the degradation pathway of Lea-containing complex type N-glycans working in the secretory pathway or extracellular space. Some plant glycosidases have been found in the cell wall or extracellular space (11–14), although it is well known that many plant glycosidases involved in the turnover of glycoconjugates are localized in the acidic vacuole or protein body. Moreover, the Lea epitope is not found on vacuolar proteins, but is enriched at the plasma membrane and extracellular glycoproteins (4). These observations indicate that α-fucosidase acting on α1, 4-fucosyl linkage may reside in the cell membrane or apoplast to modify the Lea structure. However, the complete scheme of the degradation pathway of plant-specific N-glycans has not been elucidated because α1, 3-fucosidase acting on the plant specific core structures, such as M3FX or MFX, has not been identified so far (6). As for the plant α-fucosidase responsible for plant complex-type N-glycans, it has been reported that almond, Arabidopsis (AtFuc1) and rice α-fucosidases that belong to GH29 are active toward the Lea epitope on plant N-glycans and LNFP III, but not toward the core α1-3 fucosyl residue on Man3-1Xyl1Fuc1GlcNAc2 (15–17). Based on the genetic information of the rice α-Fuc’ase gene (17), two putative tomato (Solanum lycopersicum) α-Fuc’ase genes (Solyc03g06980 and Solyc11g069010) were found in the genome database and we have succeeded to express and characterize one α-Fuc’ase encoded by Solyc03g06980 (α-Fuc’ase Sl-1) (18). α-Fuc’ase Sl-1 was substantially hydrolyzed the non-reducing terminal α1, 3-fucose residue on LNFP III and α1, 4-fucose residues of Lea epitopes on plant complex-type N-glycans, but not by the α1, 3-fucose residue on the plant core penta-oligosaccharide unit (Manβ1-4[Xylβ1-2]GlcNAcβ1-4[Fucα1-3]GlcNAc) or Fucα1-3GlcNAc (GN1F). Molecular 3 D modelling of α-Fuc’ase Sl-1 construction and structure/sequence interpretation based on comparison with a homologous α-fucosidase from Bifidobacterium longum subsp. infantis (Blon_2336) (19) indicated that an unsubstituted β-GlcNAc residue in the GlcNAcβ1-4(Fucα1-3)GlcNAc (GN2F) structure might be pre-requisite to fix the α1, 3-fucosylated substrate at the sugar-binding pocket commonly found in GH29-B (18, 19). However, in our previous study, we could not prepare the substrate, GN2F and thus have not proved our hypothesis to date. As described above, we have characterized the α-Fuc’ase encoded by the Solyc03g06980 (α-Fuc’ase Sl-1) gene, but the molecular characterization of the remaining putative α-Fuc’ase genes (Solyc11g069010) has not been performed. In this study, therefore, we have expressed the gene using the insect cell (Sf9) system and confirmed that the gene product (α-Fuc’ase Sl-2) is also a GH29 family α1, 3/4-Fuc’ase. α-Fuc’ase Sl-2 was active toward the Lea-epitope in the plant complex-type N-glycans and LNFP III but not the M3FX or MFX, indicating that the substrate specificity of the recombinant α-Fuc’ase Sl-2 (rFuc’ase Sl-2) is similar to that of α-Fuc’ase Sl-1. Interestingly, however, we have found that rFuc’ase Sl-1 and rFuc’ase Sl-2 were active toward GlcNAcβ1-4(Fucα1-3)GlcNAc-PA (GN2F) but not MF and GN1F, indicating that these two tomato α-Fuc’ases require the non-reducing terminal β-linkage GlcNAc residue of GN2F for activity and must be involved in the complete degradation of the plant specific N-glycans. Molecular 3D modelling of α-Fuc’ase Sl-2 based on comparisons with the homologous enzyme of Bifidobacterium longum subsp. infantis (Blon_2336) α-fucosidase (19) provided an insight into the catalytic mechanism. Materials and Methods Materials Tomato seeds (KG172) were kindly gifted by the Research Institute, Kagome Co. Ltd. (Tochigi, Japan). Reagents were of analytical or electrophoresis grade and were purchased from Wako Pure Chemical Industries Ltd. (Osaka, Japan), Nacalai Tesque, Inc. (Kyoto, Japan) and Sigma (St. Louis, MO, USA). Genetic manipulation enzymes with other reagents were purchased from Takara Bio Inc. (Kyoto, Japan), Applied Biosystems™ (Austin, TX, USA) and Promega (Madison, WI, USA). Anti-Flag® M2 affinity gel was from SIGMA (Saint Louis, USA). PA-labelled Lacto-N-fucopentaose III (LNFP III) and Lacto-N-fucopentaose I (LNFP I) was purchased from Takara Bio (Kyoto, Japan). Authentic PA-sugar chains (Gal2Fuc2GN2M3FX, M3FX, MFX, MF and GNF) were prepared from glycoproteins produced by rice cultured cells (20), and momordin-a (21). Spodoptera frugiperda (Sf9) cells were purchased from Invitrogen (Carlsbad, CA, USA). MF, GN2F and GNF were also prepared from MFX using tomato β-xylosidase (22), β-mannosidase (Helix pomatia, Sigma) and β-N-acetylhexosaminidase (Jack bean, Sigma). Their structures were confirmed by ESI-MS and MS/MS analyses as shown in Supplementary Fig. S1. Molecular cloning and recombinant expression The coding sequence (1554 bp) of Solyc11g069010 (locus: Solyc11g069010; allele: SOLYC11G069010 in Sol Genomics Network: www.solgenomics.net/search/locus), described here as α-Fuc’ase Sl-2, was amplified through PCR and cloned into pGEM-T easy vector, and thereafter sub-cloned into pFastBac™-1 using the Bac-to-Bac Baculovirus Expression System (Invitrogen). In every cloning step, the specific primers were designed (Supplementary Table SI) and the amplified PCR products were subjected to 1% (w/v) agarose gel electrophoresis to view proper DNA banding on the gel (data not shown). The correct orientation of the gene of interest and all of the constructs were verified by sequencing. The constructed recombinant pFastBac™-1 vector served as a donor plasmid, and was used to transform into the E. coli DH10Bac with the bacmid DNA. The donor plasmid (recombinant pFastBac™-1) transposed the α-Fuc’ase Sl-2 gene into the bacmid DNA with the help of a helper plasmid available in the DH10Bac E.coli, and eventually constructed the baculovirus expression vector (recombinant bacmid). The selection of the recombinant bacmid was carried out with kanamycin, gentamycin and tetracycline resistance. Bacmid vectors for insert sizes were also checked by restriction digestion with EcoRI and NotI following 1% (w/v) agarose gel electrophoresis. The bacmid expression construct was finally transfected into Sf9 insect cells for expression of the α-Fuc’ase Sl-2 gene. The Sf9 (cell line of Spodoptera frugiperda) cells were transfected with this recombinant bacmid DNA for the expression of α-Fuc’ase Sl-2 gene using Cellfectin® II reagent (Invitrogen) according to the manufacturers protocol. At 72 h post-infection, supernatants were collected for the recombinant baculovirus as P1 virus stock, and the recombinant virus stock was amplified for two more successive generations as P2 and P3 stock. A time-course study was performed to select the highest expression level using P3 viral stock. Insect cell culture and transfection The insect cell line Spodoptera frugiperda Sf9 was cultured in TNM-FH insect medium (Invitrogen) supplemented with 10% fetal bovine serum (FBS) at 27°C. The Sf9 cells were cotransfected with the baculovirus as instructed by the manufacturer. Bacmid DNA is a modified wild baculovirus DNA, which contains a lethal deletion and cannot develop into a viable virus by itself. Recombination between the flanking regions of the polyhedron gene from the donor vector and modified wild-type baculovirus DNA therefore results in 100% recombinant baculovirus DNA. To express the protein, 8 × 105 Sf9 cells were seeded in a 100 mm2 culture petridish and cotransfected with P3 virus stock. Purification, SDS-PAGE and Western blotting Infected Sf9 cells were harvested by scraping in 25 mM phosphate buffer saline (PBS) (pH 6.5), supplemented with 0.1% (v/v) Triton X-100 and protease inhibitor cocktail (Roche). Cells were homogenized by sonication, and crude lysate was applied to the ANTI-FLAG® M2 Affinity Gel (SIGMA), and purified proteins according to the manufacturers’ indications. The protein concentration was quantified spectrophotometrically at 280 nm using bovine serum albumin (BSA) as a standard. Proteins were separated by gel electrophoresis on 12% (w/v) SDS-PAGE gel under reducing conditions (with 5% 2-mercaptoethanol) according to the method described by Laemmli (23). Gels were stained with a Silver staining kit (Silver Stain Kit II, Wako Co., Japan). Precision Plus Protein™ Standards (Bio-Rad Laboratories, Inc., Hercules, CA) were used as molecular weight markers. Following gel electrophoresis, the proteins on the gel were transferred onto a polyvinylidene fluoride (PVDF) membrane (Amersham Hybond®-P; GE Healthcare UK Ltd., Buckinghamshire, UK). The membrane was blocked with Tris buffered saline (TBS) containing 0.05% (v/v) Tween® 20 and 1% BSA. The membrane was probed with the Anti-FLAG M2 monoclonal antibody (SIGMA) at room temperature for 2 h, and was further incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (1:10,000 dilution) (Santa Cruz Biotechnology, Inc., Dallas, TX) for 30 min. The detection of target proteins was performed using ECL SELECT™ chemiluminescence reagent Amersham Biosciences), according to the manufacturer’s protocol. Enzyme activity assay Assays of rFuc’ase Sl-2 activities were basically performed according to the methods described in our previous paper (18). Briefly, rFuc’ase Sl-2 activities during the purification step were assayed at 37°C by incubating with PA-labelled plant complex type N-glycans containing Lea determinants (Gal2Fuc2GN2M3FX) or LNFP III in 0.1 M sodium citrate buffer (pH 5.0). A reaction mixture of 50 μl containing about 10 pmol of substrate (Gal2Fuc2GN2M3FX or LNFP III), and 240 ng of rFuc’ase Sl-2 were used in the assay. The substrates and enzymatic products were analysed by SF-HPLC using a Shodex Asahipak NH2P-50 column (4.6 × 250 mm). The column was equilibrated with 80% (v/v) acetonitrile/water. The PA-sugar chains were eluted by increasing the water content of the water-acetonitrile mixture from 26 to 50% linearly at a flow rate of 0.7 ml/min, and were detected with a Jasco FP-920 Intelligent Fluorescence detector (Jasco, Inc., Easton, MD) (excitation at 310 nm, emission at 380 nm). The optimum pH of rFuc’ase Sl-2 was investigated using Gal2Fuc2GN2M3FX as a substrate at different pH levels ranging from 3.0–8.0. The buffers used were 0.1 M glycine–HCl (pH 3.0 and 3.5), 0.1 M sodium citrate (pH 4.0–5.0), 0.1 M MES (pH 6.0–7.0) and 0.1 M Tris–HCl (pH 8.0). The effects of metal ions (FeCl2, MnCl2, CoCl2, CaCl2, ZnCl2, CuCl2 and MgCl2) and EDTA on rFuc’ase Sl-2 activity were also investigated using Gal2Fuc2GN2M3FX as a substrate. The substrate specificity of rFuc’ase Sl-2 was analysed using authentic PA-sugar chains (10–20 pmol), LNFP I, LNFP III and PA-labelled N-glycans, Gal2Fuc2GN2M3FX, M3FX, MFX, MF, GN2F and GN1F. A reaction mixture of 50 μl for each substrate was incubated with rFuc’ase Sl-2 (240 ng) in 0.1 M sodium citrate buffer (pH 5.0) at 37°C for 4 h or overnight. When LNFP I, LNFP III, Gal2Fuc2GN2M3FX, M3FX and MFX were used as substrates, the products were analysed by SF-HPLC. When GNF and GN2F were used as substrates, the reaction mixture was analysed by RP-HPLC using a Cosmosil 5C18-AR column (4.6 × 250 mm). The column was equilibrated with 0.05% (v/v) trifluoroacetic acid (TFA)/water. The PA-sugar chains were eluted by increasing the acetonitrile content from 0 to 7% linearly at a flow rate of 1.2 ml/min. Sequence analysis, phylogenetic tree construction and homology modelling The signal peptide sequence of α-Fuc’ase Sl-2 was predicted using SignalP4.1 online tools (http://www.cbs.dtu.dk/services/SignalP/). The theoretical molecular mass and an isoelectric point (pI) were determined with the ExPASy proteomics server (http://web.expasy.org/cgi-bin/compute_pi/pi_tool). To identify the potential N-glycosylation sites, the online software NetGlyc 1.0 (http://www.cbs.dtu.dk/services/NetNGlyc/) was used. The conserved domains (CD) in the deduced amino acid sequence of α-Fuc’ase Sl-2 were identified by the CD-search tool (CDD V3.0-44354 PSSMs) available at the NCBI website (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). The phylogenetic tree, employing distance/neighbour-joining method, was constructed using the amino acid sequences of α-Fuc’ase Sl-2 with 10 characterized or deduced α-fucosidases from different sources (Blon_2336, Bifidobacterium longum; BT_2192, Bacteroides thetaiotaomicron; Ss_Fuc_142, Streptomyces sp.; BbAfcB, Bifidobacterium bifidum; α-Fuc’ase Os, Oryza sativa; Solyc-, S. lycopersicum; At_Fuc1, A. thaliana; CPF_2130, Clostridium perfringens; TfFuc1, Tannerella forsythia), which were mainly found in the CAZy database. We used the SWISS-MODEL web server (http://swissmodel.expasy.org/) and ModWeb server (https://modbase.compbio.ucsf.edu/modweb/) for homology modelling of α-Fuc'ase-Sl-2 according to the method described by Hossain et al. (24). The amino acid sequence of the second tomato α-fucosidase (α-Fuc'ase-Sl-2) composed of 518 residues (GenBank accession no: XM_004251058.2; GI: 723745655) was used to identify the template structures in the template library of the SWISS-MODEL server, and homology modelling was carried out in project mode. The online ModWeb Comparative Modelling Server version SVN.r1597 was also used for further modelling to compare with the models obtained from the SWISS-MODEL server. The qualities of the modelled structures were assessed using DFire (25), QMEAN (26), PROCHECK (27), WHAT_IF (28) methods and the ModEval Model evaluation server (29). The 3D-model structure of α-Fuc-Sl-2 was aligned with other crystal structures to ascertain the active site residues responsible for catalytic activity. COFACTOR software, a structure-based method for the biological function annotation of protein molecules, was used to identify functional insights, including ligand-binding sites and gene-ontology terms (30, 31). COACH, another software for binding site prediction, was used to compare the results obtained from COFACTOR (32). The RosettaDock server was used to dock the bona fide inhibitor of α-fucosidase, deoxyfuconojirimycin (33). UCSF Chimera and SWISS-PdbViewer were used to view the models and prepare images. Results and Discussion Construction of the α-Fuc’ase Sl-2 gene expression vector We previously cloned, expressed, and functionally-characterized an α-fucosidase (α-Fuc’ase Sl-1) from tomato (18). In this study, we cloned and expressed another α-fucosidase gene, Solyc11g069010, which is supposed to have α1, 3/4- fucosidase activity towards Lea epitopes and core α1, 3-fucose residues on complex type N-glycans. For this, an α-Fuc’ase Sl-2 gene expression vector (baculovirus expression vector) was constructed. Insert sequence and size of the expression vector (bacmid) were confirmed through sequencing and restriction digestion methods, respectively (data not shown). The nucleotide sequence of the α-Fuc’ase Sl-2 gene is shown in Fig. 1. Fig. 1 View largeDownload slide Deduced amino acid sequence from the coding sequence of Solyc11g069010, α-Fuc’ase Sl-2. The primers used to amplify and sequencing of the coding sequence (2073 bp) are indicated with underlined letters. α-Fuc’ase Sl-2 consists of 518 amino acid residues, and the N-terminal 26 amino acid residues that encode a predicted signal peptide are underlined. The signal peptide cleavage site is shown by a vertical arrow. The predicted N-glycosylation sites (N-X-S/T) are represented in an underlined bold font. Fig. 1 View largeDownload slide Deduced amino acid sequence from the coding sequence of Solyc11g069010, α-Fuc’ase Sl-2. The primers used to amplify and sequencing of the coding sequence (2073 bp) are indicated with underlined letters. α-Fuc’ase Sl-2 consists of 518 amino acid residues, and the N-terminal 26 amino acid residues that encode a predicted signal peptide are underlined. The signal peptide cleavage site is shown by a vertical arrow. The predicted N-glycosylation sites (N-X-S/T) are represented in an underlined bold font. Production of rFuc’ase Sl-2 by Sf9 cells and its purification Sf9 cells were cotransfected with the baculovirus expression vector containing the α-Fuc’ase Sl-2 sequence, which was inserted downstream of the polyhedron promoter. Cells were grown as described in the Materials and Methods. We carried out a time-course study to optimize the expression level of the gene, and found that 72 h post-infection was the proper expression time. As the α-Fuc’ase Sl-2 expression construct contained a FLAG-tag in its C-terminal, we successfully purified the enzyme by using ANTI-FLAG® M2 Affinity Gel. The purified protein (rFuc’ase Sl-2) from cell extracts showed a single protein band with a molecular weight of about 65 kDa when analysed with SDS-PAGE (Fig. 2A). By using western blot analysis, the expected protein band was detected from the purified protein (Fig. 2B, L1) and crude enzyme (Fig. 2B, L2) using ANTI-FLAG® M2 monoclonal antibody. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s) that did not bind to the affinity gel tightly but was recognized by the antibody. No band was detected from infected culture (Fig. 2B, L3) medium or from unprocessed cells (Fig. 2B, L4), suggesting rFuc’ase Sl-2 is not secreted out of culture broth or is not a secretory protein. Fig. 2 View largeDownload slide SDS-PAGE and western blot analysis of rFuc’ase Sl-2. (A) The purified rFuc’ase Sl-2 from the culture medium was subjected to 12% polyacrylamide gel electrophoresis under reducing conditions. The gel was stained with Silver stain. M stands for marker proteins of Precision Plus Protein™ Standards. (B) Purified rFuc’ase Sl-2 from the culture medium was applied to the 12% polyacrylamide gel electrophoresis under reducing conditions. The protein on the gel was transferred into the polyvinylidene fluoride (PVDF) membrane for western blot analysis. The membrane was probed with primary antibody (anti-FLAG monoclonal antibody) and was then incubated with secondary antibody (goat anti-mouse IgG-HRP) as described in the text. A ∼60 kDa protein band of rFuc’ase Sl-2 was detected by Image Quant LAS 500. M, molecular markers; L1, purified rFuc’ase Sl-2; L2, crude lysate of infected Sf9 cells; L3, uninfected Sf9 cell lysate; L4, culture medium of infected Sf9 cells. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s). Fig. 2 View largeDownload slide SDS-PAGE and western blot analysis of rFuc’ase Sl-2. (A) The purified rFuc’ase Sl-2 from the culture medium was subjected to 12% polyacrylamide gel electrophoresis under reducing conditions. The gel was stained with Silver stain. M stands for marker proteins of Precision Plus Protein™ Standards. (B) Purified rFuc’ase Sl-2 from the culture medium was applied to the 12% polyacrylamide gel electrophoresis under reducing conditions. The protein on the gel was transferred into the polyvinylidene fluoride (PVDF) membrane for western blot analysis. The membrane was probed with primary antibody (anti-FLAG monoclonal antibody) and was then incubated with secondary antibody (goat anti-mouse IgG-HRP) as described in the text. A ∼60 kDa protein band of rFuc’ase Sl-2 was detected by Image Quant LAS 500. M, molecular markers; L1, purified rFuc’ase Sl-2; L2, crude lysate of infected Sf9 cells; L3, uninfected Sf9 cell lysate; L4, culture medium of infected Sf9 cells. The lower band found in L2 might be a proteolytic product from rFuc’ase Sl-2 by the action of endogenous protease(s). Sequence interpretation and phylogenetic study α-Fuc’ase Sl-2 gene encodes a 518-amino-acid-long polypeptide chain, of which, the first 26 N-terminal amino acids form the predicted signal peptide (Fig. 1). A putative cleavage site resides in between 26th and 27th residues. The theoretical molecular mass and pI were 59.47 kDa and 5.69, respectively. The molecular mass (65 kDa) determined by SDS-PAGE was slightly bigger than that of the theoretical mass, probably due to the addition of N-glycan(s). In addition, the polypeptide chain contains six potential N-glycosylation sites (N-X-S/T) at residue numbers 83, 249, 322, 366, 440 and 511. A conserved domain database (CDD) search found two conserved domains for α-Fuc’ase Sl-2, namely, the COG3669 domain (carbohydrate transport and metabolism, 43-479) and cl07893 domain (alpha amylase catalytic domain, 77-337), which belong to GH family 29. A neighbour-joining phylogenetic tree was prepared (Fig. 3). The close members of α-Fuc’ase Sl-2 in the clades are mainly characterized as active toward α-1, 3/1, 4-fucosidic linkage, suggesting that α-Fuc’ase Sl-2 might have a similar substrate specificity and play a similar role in the turnover of N-glycoproteins in plant cells. The sequence homology was analysed by the NCBI protein BLAST tool, and found a higher homology between α-Fuc’ase Sl-2 and AtFuc1 (69%) than between α-Fuc’ase Sl-2 and α-Fuc’ase Sl-1 (51%). Fig. 3 View largeDownload slide Phylogenetic tree of GH29 family α-Fuc’ases. The phylogenetic tree was drawn using the amino acid sequence of α-Fuc’ase Sl-2 with nine other characterized or deduced GH 29 family α-fucosidases from different sources (plant/bacteria/fungi etc.) found in the CAZy database. The names of the organisms are described in the text. Fig. 3 View largeDownload slide Phylogenetic tree of GH29 family α-Fuc’ases. The phylogenetic tree was drawn using the amino acid sequence of α-Fuc’ase Sl-2 with nine other characterized or deduced GH 29 family α-fucosidases from different sources (plant/bacteria/fungi etc.) found in the CAZy database. The names of the organisms are described in the text. Functional properties The activity of rFuc’ase Sl-2 against Gal2Fuc2GN2M3FX was found optimal at a pH of around 5.0 (Supplementary Fig. S2), suggesting that the enzyme resides and functions in acidic organelle such as the vacuole or cell wall. The HPLC profiles of the rFuc’ase Sl-2 activity assay against the Gal2Fuc2GN2M3FX substrate at different pH levels (pH 4.0, 5.0 and 7.0) are shown in Fig. 4A and B. The optimum pH of rFuc’ase Sl-2 was found to be similar to those of α-fucosidases from almond (15, 16), Arabidopsis (AtFuc1) (16), rice (α-Fuc’ase Os) (17) and tomato (α-Fuc’ase Sl-1) (18). Fig. 4 View largeDownload slide Activities of rFuc’ase Sl-2 toward N-glycans, Gal2Fuc2GN2M3FX-PA. (A) A reaction mixture (50 μl) of rFuc’ase Sl-2 (240 ng) + substrate (10 pmol) + 0.1 M citrate buffer (pH 5.0) was incubated at 37°C for 1 h and was analysed by size-fractionation (SF)-HPLC. (B) Substrate with the same volume of reaction mixture and different pH levels (4.0, 5.0 and 6.0) was incubated at 37°C for 4 h and was analysed by SF-HPLC. Fig. 4 View largeDownload slide Activities of rFuc’ase Sl-2 toward N-glycans, Gal2Fuc2GN2M3FX-PA. (A) A reaction mixture (50 μl) of rFuc’ase Sl-2 (240 ng) + substrate (10 pmol) + 0.1 M citrate buffer (pH 5.0) was incubated at 37°C for 1 h and was analysed by size-fractionation (SF)-HPLC. (B) Substrate with the same volume of reaction mixture and different pH levels (4.0, 5.0 and 6.0) was incubated at 37°C for 4 h and was analysed by SF-HPLC. For the investigation on the effect of metal ions, various divalent metal cations (FeCl2, MnCl2, CoCl2, CaCl2, ZnCl2, CuCl2 and MgCl2) and EDTA were added at final concentrations of 5 mM to the reaction mixture. The rFuc’ase Sl-2 activity was not stimulated explicitly with the addition of metal ions or EDTA; Fe2+, and Cu2+ rather utterly inhibited the enzyme activity (Supplementary Table SII), suggesting that these metal ions might co-ordinate with some essential amino acids in the catalytic site. Substrate specificity of rFuc’ase Sl-2 The substrate specificity of rFuc’ase Sl-2 was investigated using several N-glycans or other oligosaccharides bearing α-fucosyl residue as substrates. The products and the remaining substrates of the enzymatic reaction were analysed by SF-HPLC or RP-HPLC, and the results are summarized in Table I. We confirmed that the rFuc’ase Sl-2 substantially hydrolyzed the α1, 4-Fuc residue on the Lea epitopes of plant complex type N-glycans (Fig. 4A) and the α1, 3-fucose residue on LNFP III. No activity, even at a high enzyme concentration, was noticed with α1, 2-fucosyllactose and p-nitrophenol-α-L-fucopyranoside (pNP-Fuc) as substrates. The substrate specificity of rFuc’ase Sl-2 was quite similar to those of the almond fucosidase I (15, 16), Arabidopsis fucosidase (AtFuc1) (16) and rFuc’ase Sl-1 (18). However, when we used GlcNAcβ1-4(Fucα1-3)GlcNAcPA (GN2F) and Fucα1-3GlcNAc-PA (GN1F) as substrates, we found that rFuc’ase Sl-2 hydrolyzed the α1, 3-fucosyl linkage on GN2F (Fig. 5A) but not on GN1F (Fig. 5B), indicating that the β1-4 linked GlcNAc residue can be a substitute for β1-4Gal residue linked to the β-GlcNAc residue in LNFP III for fixing the α1-3Fuc residue in the catalytic site. Furthermore, in addition to rFuc’ase Sl-2, we also confirmed that rFuc’ase Sl-1 (18) also hydrolyzed the α1, 3-fucosyl linkage on GN2F (data not shown). However, the reaction rates of these two tomato α-Fuc’ase specimens toward GN2F (GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) was much slower than those toward the Lea unit in the plant complex type N-glycan and LNFP III, indicating that the pyridylamino group attached to the proximal GlcNAc or the open-ring form of the GlcNAc might wield a negative influence on the enzyme activity. Table I. Substrate specificity of α-Fuc’ase Sl-2     ND: Any products were not detected under the analytical condition described in the text. Table I. Substrate specificity of α-Fuc’ase Sl-2     ND: Any products were not detected under the analytical condition described in the text. Fig. 5 View largeDownload slide Activities of rFuc’ase Sl-2 toward GN2F and GN1F. A reaction mixture (50 μl) consisting of rFuc’ase Sl-2 (240 ng), substrates (10 pmol) and 0.1 M citrate buffer (pH 5.0) was incubated at 37°C overnight, and the products were analysed by reversed-phase (RP)-HPLC using a Cosmosil 5C18-AR column. (A) HPLC profile of GN2F incubated with rFuc’ase Sl-2. (B) HPLC profile of GN1F incubated with rFuc’ase Sl-2. GN2, GlcNAcβ1-4GlcNAc-PA; GN, GlcNAc-PA. Some peaks marked with asterisks were not N-glycans, but rather contaminative fluorescence substances. Fig. 5 View largeDownload slide Activities of rFuc’ase Sl-2 toward GN2F and GN1F. A reaction mixture (50 μl) consisting of rFuc’ase Sl-2 (240 ng), substrates (10 pmol) and 0.1 M citrate buffer (pH 5.0) was incubated at 37°C overnight, and the products were analysed by reversed-phase (RP)-HPLC using a Cosmosil 5C18-AR column. (A) HPLC profile of GN2F incubated with rFuc’ase Sl-2. (B) HPLC profile of GN1F incubated with rFuc’ase Sl-2. GN2, GlcNAcβ1-4GlcNAc-PA; GN, GlcNAc-PA. Some peaks marked with asterisks were not N-glycans, but rather contaminative fluorescence substances. Resolved 3 D-modelled structure and mechanism of action The overall 3D modelled structure of α-Fuc’ase Sl-2 is shown in Fig. 6A, which consists of 443 amino acids (37–479) in its structural part. The modelled structure contains two domains: an N-terminal GH29 α-fucosidase domain containing a (β/α)8 TIM barrel, and a C-terminal carbohydrate binding module (CBM) 32 domain with a β-sandwich structure. The CD database search suggested that the α-Fuc’ase Sl-2 consists of three characteristic domains: (1) the α-amylase catalytic domain family, (2) the F5/8 type C domain (discoidin domain family) and (3) the COG3669 domain. Two domains, the α-amylase catalytic domain family (amino acids: 77–337) and the COG3669 domain (amino acids: 43-479), overlap each other on the amino acid sequence, and these two domains could not be shown together with different colours in Fig. 6A. The modelled structure contains two domains, as described in the ‘Sequence interpretation and phylogenetic study’ section. The structural features of α-Fuc’ase Sl-2 were compared upon superimposition on the template (pdb id: 3uesA) structure, and showed close resemblance between the two structures (data not shown). The predicted ligand binding sites of α-Fuc'ase-Sl-2, from the superimposition, were found to be Phe54, His56, Trp67, His105, His106, Tyr151, Asp192, Ala194, Glu236, Asp280 and Trp287. The ligand binding sites of α-Fuc'ase-Sl-2 were further assessed with a potential α-fucosidase inhibitor, deoxyfuconojirimycin (DFU), using the RosettaDock server and the residues were identified as Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288, which is shown in a generated space filled view in Fig. 6B. Finally, the modelled structure of α-Fuc'ase-Sl-2 was aligned with the template to identify the predicted catalytic nucleophile and catalytic acid/base residue. From the alignment view (Fig. 6C), the residues of Asp192 and Glu236 in the α-Fuc’ase Sl-2 modelled structure were the catalytic proton donor and catalytic nucleophile/base, respectively, when compared with the location of corresponding residues on the template (Fig. 6C). Multiple sequence alignment of the amino acid sequences of α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Bifidobacterium longum α-fucosidase (pdb: 3uesA), revealed that the catalytically important residues were also conserved among them as shown in Fig. 7, which indicates a similar mode of action to their native substrates. Fig. 6 View largeDownload slide 3 D modelled structure of α-Fuc’ase Sl-2 resolved by the SWISS-MODEL web server. (A) Ribbon diagram of the overall 3D modelled structure of tomato α-Fuc’ase-Sl-2. The catalytic domain is shown by various colours, from the blue at the N terminus to yellowish green at the C terminus. The strands and helices that form the (β/α)8 barrel are labelled in red. The C-terminal carbohydrate-binding (β-sandwich, F5/8 type C) domain is shown in red, with 8-strands labelled in blue letters. The N and C termini of modelled structure are also labelled. (B) Space filled view (hydrophobic surface) of the modelled structure (α-Fuc’ase-Sl-2) with docked DFU (Deoxyfucojirinomycin). Active site residues (Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288) interacted with ligand (DFU) are coloured black. The DFU located in the active site cleft is indicated as an interactive colour stick. Chimera was used to prepare the image. (C) Structural alignments of the active sites of α-Fuc’ase-Sl-2 (modelled structure) and 3ues (Bifidobacterium longum subsp. infantis) in complex with DFU (Deoxyfuconojirimycin). All active site residues and the DFU from both the 3ues- and α-Fuc’ase-Sl-2-DFU complexes are drawn in stick format. The residues of modelled α-Fuc’ase-Sl-2 are shown in green colour. The two DFUs resided in the centre position of both 3ues and α-Fuc’ase-Sl-2 are hydrogen bonded (cyan colour) with active site residues of the two structures. The strands and helices that form the (β/α)8 TIM barrel are labelled in green (α-Fuc’ase-S2) and red (3ues). Only catalytically important residue numbers are shown in the figure; other residues are not shown for the sake of clarity. Chimera was used to prepare the image. Fig. 6 View largeDownload slide 3 D modelled structure of α-Fuc’ase Sl-2 resolved by the SWISS-MODEL web server. (A) Ribbon diagram of the overall 3D modelled structure of tomato α-Fuc’ase-Sl-2. The catalytic domain is shown by various colours, from the blue at the N terminus to yellowish green at the C terminus. The strands and helices that form the (β/α)8 barrel are labelled in red. The C-terminal carbohydrate-binding (β-sandwich, F5/8 type C) domain is shown in red, with 8-strands labelled in blue letters. The N and C termini of modelled structure are also labelled. (B) Space filled view (hydrophobic surface) of the modelled structure (α-Fuc’ase-Sl-2) with docked DFU (Deoxyfucojirinomycin). Active site residues (Trp67, His105, His106, Asp192, Glu236, Trp287 and Phe288) interacted with ligand (DFU) are coloured black. The DFU located in the active site cleft is indicated as an interactive colour stick. Chimera was used to prepare the image. (C) Structural alignments of the active sites of α-Fuc’ase-Sl-2 (modelled structure) and 3ues (Bifidobacterium longum subsp. infantis) in complex with DFU (Deoxyfuconojirimycin). All active site residues and the DFU from both the 3ues- and α-Fuc’ase-Sl-2-DFU complexes are drawn in stick format. The residues of modelled α-Fuc’ase-Sl-2 are shown in green colour. The two DFUs resided in the centre position of both 3ues and α-Fuc’ase-Sl-2 are hydrogen bonded (cyan colour) with active site residues of the two structures. The strands and helices that form the (β/α)8 TIM barrel are labelled in green (α-Fuc’ase-S2) and red (3ues). Only catalytically important residue numbers are shown in the figure; other residues are not shown for the sake of clarity. Chimera was used to prepare the image. Fig. 7 View largeDownload slide Sequence alignment of α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 with its homolog Blon_Fuc. Sequence alignments among α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Blon_Fuc from B. longum subsp. infantis were performed using the ClustalW program. Predicted ligand binding amino acids at the active site of α-Fuc’ase-Sl-2, α-Fuc’ase-Sl-1 and Blon_Fuc described in the text are conserved, and are indicated as boxes. Fig. 7 View largeDownload slide Sequence alignment of α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 with its homolog Blon_Fuc. Sequence alignments among α-Fuc’ase Sl-2, α-Fuc’ase Sl-1 and Blon_Fuc from B. longum subsp. infantis were performed using the ClustalW program. Predicted ligand binding amino acids at the active site of α-Fuc’ase-Sl-2, α-Fuc’ase-Sl-1 and Blon_Fuc described in the text are conserved, and are indicated as boxes. Sakurama et al. reported that two microbial α-fucosidases, B. thetaiotaomicron enzyme (BT_2192) (34) and B. bifidum enzyme (BbAfcB) (35) belong to GH29-B, which is active toward the α1, 3/4-fucosyl linkages in Lewis x and Lea epitopes, but not toward α1, 2-fucosyl linkages nor pNP-Fuc (35). Based on the substrate specificity, the tomato α-Fuc’ase Sl-2, as well as α-Fuc’ase Sl-1 (18) and rice α-Fuc’ase (OsFuc1) (17), seemed to belong to GH29-B but not GH29-A, such as in L. caseii (AlfA, AlfB, and AlfC) (36). Both rFuc’ase Sl-1 and rFuc’ase Sl-2 could not hydrolyze MFX, MF and GN1F, but were active against α1, 3/4-fucosyl linkages in LFNP III and the Lea epitopes, indicating a critical role of the non-reducing terminal β-Gal residue of these substrates. In the case of BT_2192, it has also been found that a β-Gal-binding site consisting of Trp230, Glu254 and Asp277 plays a critical role in hydrolytic activity toward the Lea epitopes, and the GN1F structure, lacking a Gal residue, cannot be a substrate for BT_2192 (34, 35). As described above, in this study, we confirmed that both α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2 can hydrolyze the α1-3 fucosyl linkage in GN2F, indicating that the β1-4 GlcNAc residue can be a substitute for β1-4Gal residue linked to the β-GlcNAc residue in LNFP III for fixing the α1-3 Fuc residues in the catalytic site. Recently, Kato et al. found that one of the Arabidopsis α-Fuc’ases, AtFuc1, hydrolyzed the α1-3 fucosyl linkage in GNF2 and the reaction rate of AtFuc1 toward the GN2F was much faster than that toward GN2F-PA (36). These results suggest that GH29-B α-Fuc’ases must require the β1-4 linkage Gal or GlcNAc residue that is linked to the GlcNAc carrying the α1-3/4 Fuc residue, and supported the hypothetical hydrolytic mechanism based on the 3 D modelled structures of the two tomato α-Fuc’ases. Furthermore, the substrate specificity of these two tomato α-Fuc’ase specimens, α-Fuc’ase Sl-1 and α-Fuc’ase Sl-2, indicates that the complete degradation of plant specific N-glycans requires defucosylation prior to β-GlcNAc’ase acting toward the GlcNAcβ1-4GlcNAc(-Asn) unit. Supplementary Data Supplementary Data are available at JB Online. Acknowledgements The authors are grateful to the Department of Instrumental Analysis, Advanced Science Research Center, Okayama University, for ESI-MS analysis. The authors would like to thank Editage (www.editage.jp) for English language editing. Funding This work was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Basic Research C [no. 24580494 to M.M. and no. 17K0819709 to YK)] and Fostering Joint International Research (no. 15K0784 to M.M.) and the Research Grants of Mizutani Foundation for Glycoscience (to Y.K.). Conflict of Interest None declared. References 1 Lerouge P., Cabanes-Macheteau M., Rayon C., Fischette-Lainé A.-C., Gomord V., Faye L. ( 1998) N-glycoprotein biosynthesis in plants: recent developments and future trends. Plant Mol. Biol . 38, 31– 48 Google Scholar CrossRef Search ADS PubMed  2 Kobata A. ( 2007) Glycoprotein glycan structures in Comprehensive Glycoscience–from Chemistry to Systems Biology , Vol. 1. ( Kamerling H., ed.) pp. 39– 72, Elsevier, Oxford Google Scholar CrossRef Search ADS   3 Schoberer J., Strasser R. ( 2011) Sub-compatmental organization of Golgi-resident N-glycan processing enzymes in plants. Mol. Plant.  4, 220– 228 Google Scholar CrossRef Search ADS PubMed  4 Strasser R. ( 2016) Plant protein glycosylation. Glycobiology  26, 926– 939 Google Scholar CrossRef Search ADS PubMed  5 Maeda M., Kimura Y. ( 2014) Structural features of free N-glycans occurring in plants and functional features of de-N-glycosylation enzymes, ENGase and PNGase. Y. Front. Plant Sci  5, 429 6 Ishimizu T. ( 2015) Plant N-glycans and their degrading enzymes in Glycoscience: Biology and Medicine . ( Endo T., Seeberger P.H., Hart G.W., Wong C.-H., Taniguchi N., eds.) pp. 1111– 1117, Springer, Tokyo Google Scholar CrossRef Search ADS   7 Fitchette-Laine A.-C., Gomord V., Cabanes M., Michalski J.-C., Macary M., Foucher B., Cavelier B., Hawes C., Lerouge P., Faye L. ( 1997) N-glycans harboring the Lewis a epitope are expressed at the surface of plant cells. Plant J.  12, 1411– 1417 Google Scholar CrossRef Search ADS PubMed  8 Melo S., Nimtz M., Conradt H.S., Fevereiro P.S., Costa J. ( 1997) Identification of the human Lewisa carbohydrate motif in a secretory peroxidase from a plant cell suspension culture (Vaccinium myrtillus L.). FEBS Lett . 415, 186– 191 Google Scholar CrossRef Search ADS PubMed  9 Maeda M., Kimura M., Kimura Y. ( 2010) Intracellular and extracellular free N-glycans produced by plant cells: occurrence of unusual plant complex-type free N-glycans in extracellular spaces. J. Biochem . 148, 681– 692 Google Scholar CrossRef Search ADS PubMed  10 Maeda M., Ebara N., Tani M., Vavricka J.C., Kimura Y. ( 2017) Occurrence of plant complex type free N-glycans carrying a single GlcNAc residue at the reducing end in a fresh-water plant, Egeria densa: a possible mechanism for biosynthesis of free N-glycans in plants. Glycoconj. J.  34, 229– 240 Google Scholar CrossRef Search ADS PubMed  11 Augur C., Benhamou N., Darvill A., Albersheim P. ( 1993) Purification, characterization and cell wall localization of an alpha-fucosidase that inactivsties a xyloglucan oligosaccharin. Plant J.  3, 415– 426 Google Scholar CrossRef Search ADS PubMed  12 Gossarau R. ( 1991) Catalytic histochemistry of acid and newtral hydrolases in plant seedlings. Histochem. J . 23, 483– 489 Google Scholar CrossRef Search ADS PubMed  13 Iglesias N., Abelenda J.A., Rodino M., Sampedro J., Revilla G., Zarra I. ( 2006) Apoplastic glycosidases active against xyloglucan oligosaccharides of Arabidopsis thaliana. Plant Cell Physiol . 47, 55– 63 Google Scholar CrossRef Search ADS PubMed  14 Günl M., Neumetzler L., Kraemer F., de Souza A., Schultink A., Pena M., York W.S., Pauly M. ( 2011) AXY8 encodes an α-fucosidase, underscoring the importance of apoplastic metabolism on the fine structure of Arabidopsis cell wall polysaccharides. Plant Cell  23, 4025– 4040 Google Scholar CrossRef Search ADS PubMed  15 Ogata-Arakawa M., Muramatsu T., Kobata A. ( 1977) α-L-fucosidases from almond emulsin: characterization of the two enzymes with different specificities. Arch. Biochem. Biophys . 181, 353– 358 Google Scholar CrossRef Search ADS PubMed  16 Zeleny R., Leonard R., Dorfner G., Dalik T., Kolarich D., Altmann F. ( 2006) Molecular cloning and characterization of a plant α1,3/4-fucosidase based on sequence tags from almond fucosidase I. Phytochemistry  67, 641– 648 Google Scholar CrossRef Search ADS PubMed  17 Rahman M.Z., Fujishige M., Maeda M., Kimura Y. ( 2016) Rice α-fucosidase active against plant complex type N-glycans containing Lewis a epitope: purification and characterizeation. Biosci. Biotechnol. Biochem . 80, 291– 294 Google Scholar CrossRef Search ADS PubMed  18 Rahman M.Z., Maeda M., Itano S., Hossain M.A., Ishimizu T., Kimura Y. ( 2017) Molecular characterization of tomato α1,3/4-fucosidase, a member of glycosyl hydrolase family 29, involved in degradation of plant complex type N-glycans. J. Biochem . 161, 421– 432 Google Scholar PubMed  19 Sakurama H., Fushinobu S., Hidaka M., Yoshida E., Honda Y., Ashida H., Kumagi H., Yamamoto K., Katayama T. ( 2012) 1,3-1,4-α-L-fucosynthase that specifically introduces Lewis a/x antigens into type-1/2 chains. J. Biol. Chem . 287, 16709– 16719 Google Scholar CrossRef Search ADS PubMed  20 Maeda M., Kimura Y. ( 2006) Glycoform analysis of N-glycans linked to glycoproteins expressed in rice culture cells: predominant occurrence of complex type N-glycans. Biosci. Biotechnol. Biochem . 70, 1356– 1363 Google Scholar CrossRef Search ADS PubMed  21 Kimura Y., Minami Y., Tokuda T., Nakajima S., Takagi S., Funatsu G. ( 1991) Primary structures of N-linked oligosaccharides of momordin-a, ribosome-inactivating protein from Momordica charantia seeds. Agric. Biol. Chem . 55, 2031– 2036 Google Scholar CrossRef Search ADS PubMed  22 Yokouchi D., Ono N., Nakamura K., Maeda M., Kimura Y. ( 2013) Purification and characterization of β-xylosidase that is active for plant complex type N-glycans from tomato (Solanum lycopersicum): removal of core α1-3 mannosyl residue is prerequisite for hydrolysis of β1-2 xylosyl residue. Glycoconj. J.  30, 463– 472 Google Scholar CrossRef Search ADS PubMed  23 Laemmli U.K., Favre M. ( 1973) Maturation of the head of bacteriophage T4. J. Mol. Biol.  80, 575– 599 Google Scholar CrossRef Search ADS PubMed  24 Hossain M.A., Roslan H.A., Karim M.R., Kimura Y. ( 2016) Molecular phylogeny, 3D-structural insights, docking and mechanisms of action of plant beta-galactosidase. Int. J. Bioinform. Res. Appl . 12, 149– 179 Google Scholar CrossRef Search ADS   25 Pieper U., Webb B.M., Barkan D.T., Schneidman-Duhovny D., Schlessinger A., Braberg H., Yang Z., Meng E.C., Pettersen E.F., Huang C.C., Datta R.S., Sampathkumar P., Madhusudhan M.S., Sjolander K., Ferrin T.E., Burley S.K., Sali A. ( 2011) MODBASE, a database of annotated comparative protein structure models and associated resources. Nucleic Acids Res . 39, D465– D474 Google Scholar CrossRef Search ADS PubMed  26 Zhou H., Zhou Y. ( 2009) Distance-scaled, finite ideal-gas reference state improves structure-derived potentials of mean force for structure selection and stability prediction. Protein Sci . 11, 2714– 2726 Google Scholar CrossRef Search ADS   27 Benkert P., Biasini M., Schwede T. ( 2011) Toward the estimation of the absolute quality of individual protein structure models. Bioinformatics  27, 343– 350 Google Scholar CrossRef Search ADS PubMed  28 Laskowski R.A., MacArthur M.W., Moss D.S., Thornton J.M. ( 1993) PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr.  26, 283– 291 Google Scholar CrossRef Search ADS   29 Hooft R.W., Vriend G., Sander C., Abola E.E. ( 1996) Errors in protein structures. Nature  381, 272 Google Scholar CrossRef Search ADS PubMed  30 Melo F., Sanchez R., Sali A. ( 2002) Statistical potentials for fold assessment. Protein Sci . 11, 430– 444 Google Scholar CrossRef Search ADS PubMed  31 Roy A., Yang J., Zhang Y. ( 2012) COFACTOR: an accurate comparative algorithm for structure-based protein function annotation. Nucleic Acids Res . 40, W471– W477 Google Scholar CrossRef Search ADS PubMed  32 Yang J., Roy A., Zhang Y. ( 2013) BioLiP: a semi-manually curated database for biologically relevant ligand-protein interactions. Nucleic Acids Res . 41, D1096– D1103. Google Scholar CrossRef Search ADS PubMed  33 Lyskov S., Gray J.J. ( 2008) The RosettaDock server for local protein-protein docking. Nucleic Acids Res . 36, W233– W238 Google Scholar CrossRef Search ADS PubMed  34 Sakurama H., Tsutsumi E., Ashida H., Katayama T., Yamamoto K., Kumagai H. ( 2012) Differences in the substrate specificities and active-site structures of two α-L-fucosidases (glycoside hydrolase family 29) from Bacteroids thetaiotaomicron. Biosci. Biotechnol. Biochem . 76, 1022– 1024 Google Scholar CrossRef Search ADS PubMed  35 Ashida H., Miyake A., Kiyohara M., Wada J., Yoshida E., Kumagai H., Katayam T., Yamamoto K. ( 2009) Two distinct α-L-fucosidases from Bifidobacterium bifidum are essential for the utilization of fucosylated milk oligosaccharides and glycoconjugates. Glycobiology  19, 1010– 1017 Google Scholar CrossRef Search ADS PubMed  36 Kato S., Hayashi M., Kitagawa M., Maeda M., Kimura Y., Igarashi K., Kajiura H., Kasahara M., Ishimizu T. ( 2018) Degradation pathway of plant complex-type N-glycans: identification and characterization of a key α1,3-fucosidase from glycoside hydrolase family 29. Biochem. J . 475, 305– 317 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations Fuc L-fucose Gal D-galactose GlcNAc N-acetyl-D-glucosamine Glc D-glucose GN1M3FX GlcNAcβ1-2Manα1-6(Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) GN2M3FX GlcNAcβ1-2Manα1-6(GlcNAcβ1-2Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA GN2F GlcNAcβ1-4(Fucα1-3)GlcNAc-PA GN1F Fucα1-3GlcNAc-PA GN2F Fucα1-3GlcNAcβ1-4GlcNAc-PA Lacto-N-fucopentaose I Fucα1-2Galβ1-3GlcNAcβ1-3Galβ1-4Glc-PA Lacto-N-fucopentaose III Galβ1-4(Fucα1-3)GlcNAcβ1-3Galβ1-4Glc-PA (Lea)1GN1M3FX Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-6(GlcNAcβ1-2Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA) (Lea)2M3FX Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-6(Galβ1-3(Fucα1-4)GlcNAcβ1-2Manα1-3)(Xylβ1-2) Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA Lea Lewis a Man D-mannose MF Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA MFX Xylβ1-2Manβ1-4GlcNAcβ1-4(Fucα1-3)GlcNAc-PA M3FX Manα1-6(Manα1-3)(Xylβ1-2)Manβ1-4GlcNAcβ1-4(Fucβ1-3)GlcNAc-PA PA pyridylamino PTC plant complex type RP-HPLC reversed-phase HPLC rFuc’ase Sl recombinant α-fucosidase from Solanum lycopersicum SF-HPLC size-fractionation HPLC Xyl D-xylose α-Fuc’ase Sl α-fucosidase from Solanum lycopersicum 3uesA α1, 3/4-fucosidase from Bifidobacterium longum © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The Journal of BiochemistryOxford University Press

Published: Feb 10, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off