Introduction

Lectins are carbohydrate-binding proteins of nonimmune origin, which agglutinate cells and/or precipitate glycoconjugates by interacting with sugar chains [1]. The wide distribution of lectins in all biological species, from viruses and bacteria to animals and plants, suggests that lectins have important physiological functions. Many biological functions of these lectins have been elucidated, and they have been used in biological and biomedical applications. Many lectins from mushrooms have also been isolated, and several potential uses have been discovered, including mitogenic, antiproliferative, antitumor, antiviral, and immunestimulating functions [2].

Goldstein and Winter [3] reported that mushroom lectins have at least 6 families on the basis of their sequence homology and characteristics: fungal immunomodulatory protein, galectin family, mushroom lectin (CXL) family, ricin family, β-propeller-fold lectins and Jacalin-related lectin. However, the presence of more lectin families is assumed based on the sequences until now. One is the family monocot mannose-binding lectin and other is Lyophyllum descades lectin (LDL) [4]-type lectin. The sequence with the motif of monocot mannose-binding lectin was reported for several fungi including Aspergillus oryzae [5], and xylose-binding lectin from Ascomycota [6] was also deduced to be this type from the sequences of peptide fragments. On the other hand, LDL showed no sequence homology to lectins reported so far.

Kruger et al. [7] isolated and characterized M. oreades agglutinin (MOA) from fairy ring mushrooms. The mushroom is edible and popularly found in Europe and US. MOA strongly precipitated blood type B substance and presented a high affinity for Galα1-3Gal. Yagi et al. [8] reported that the diversity of mushroom lectin in a single species might be correlated with geographical location. Our preliminary finding for the extract from Japanese M. oreades indicated that the hemagglutination activity was not due to MOA but to a different lectin.

In this study, we describe the isolation, characterization, and heterologous expression of a new lectin MOL as GNA-like lectin.

Materials and methods

Materials

The fruit bodies of M. oreades were collected in June in Jigenji Park in Kagoshima, Japan. They were stored at −80 °C until use. CNBr-activated Sepharose 4B was purchased from Amersham Biosciences (Uppsala, Sweden). Trypsin, chymotrypsin and Staphylococcus aureus V8 protease were purchased from Wako Chemical Industries (Osaka, Japan). Sugars and glycoproteins were purchased from Sigma (St. Louis, Mo). All other reagents were commercially available. Asialofetuin-Sepharose 4B was prepared by coupling asialofetuin to CNBr-activated Sepharose 4B.

Isolation of the lectin from fruiting bodies of M. oreades (MOL)

All steps of purification were done at 4 °C. The fruiting bodies of M. oreades (100 g) were homogenized with 10 volumes of phosphate buffered saline (PBS), pH 7.5. After the extraction overnight, the extract was filtered and centrifuged at 10,000 g for 20 min. The clear supernatant was concentrated against ammonium sulfate. The precipitate was collected by centrifugation at 15,000 g for 20 min and then dissolved in PBS. The clear, crude extract obtained was loaded onto an asialofetuin-Sepharose 4B column (1.5 × 12 cm) equilibrated with PBS. Unbound protein was washed with PBS, and the adsorbed lectin was then eluted with 0.2 M glycine-HCl containing 0.5 M NaCl, pH 2.5. Eluted fractions were neutralized with 2 M Tris–HCl, pH 8.0, immediately. Hemagglutination activity was measured throughout all purification with trypsinized rabbit erythrocytes.

Molecular mass determination of MOL

The molecular mass of MOL was measured by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOFMS, Bruker Daltonics, Bremen, Germany). Tricine SDS-PAGE was performed in 4 % stacking gel and 16.5 % resolving gel as described by Schagger and Jagow [9]. The gel was stained with Coomassie Brilliant Blue R-250. The molecular mass of MOL was measured by gel filtration on a TSKgel G2000SW (7.5 × 300 mm) column.

Amino acid sequencing of MOL

MOL (4.0 nmol each) was digested with trypsin (enzyme: substrate = 1:50, mol/mol) or chymotrypsin (1:50) in 100 μL of 50 mM sodium bicarbonate (pH 8.0) at 37 °C for 24 h. MOL (4.0 nmol each) was digested with S. aureus V8 protease (enzyme: substrate = 1:50, mol/mol) in 100 μl of 50 mM phosphate buffer (pH 7.8) at 37 °C for 24 h. Each digest was separated by reverse-phase HPLC using a COSMOSIL 5 C18 MS-II column (4.6 × 250 mm; Nacalai, Kyoto, Japan) with a linear gradient of acetonitrile (0–48 % in 60 min) in 0.1 % (v/v) trifluoroacetic acid. The N-terminal amino acid sequences of MOL and its proteolytic peptide fragments were analyzed with an Applied Biosystems Procise 492 protein sequencer [10].

Genomic DNA sequence analysis

Genomic DNA of M. oreades was isolated using a DNeasy plant mini kit (Qiagen, Venlo, Netherlands) following the manufacturer’s protocol. Templates (70 ng/μL) were amplified using two sets of degenerate primers designed from partial amino acid sequence of MOL MQTDGNCV (residues 63–70) and AVWHTATN (78–85). PCR was performed with TAKARA LA Taq (Takara Bio, Kyoto, Japan) for 25 cycles of repeating denaturation at 94 °C for 30 s, annealing at 52 °C for 30 s, and extension at 72 °C for 1 min to obtain the gene-specific sequence of the genomic DNA encoding MOL.

cDNA sequencing of MOL

Total RNA was purified from 100 mg of M. oreades using an RNeasy plant mini kit (Qiagen) according to the manufacturer’s protocol. cDNA was synthesized using ReverTra Ace -α-® (Toyobo, Osaka, Japan) according to the supplier’s protocols. Two μg of total RNA was used as a template, and transcription was carried out at 42 °C for 20 min. Synthesized cDNA was used as a template for PCR. 3′-RACE was performed using a 3′-full RACE Core set (Takara Bio) following the manufacturer’s protocols. The template cDNA was synthesized using 2.5 μM of an oligo-d(T) primer containing an adapter sequence. The primer MOL-F1 (ATGCAGACAGATGGTAATTGCGTCGGTTAC) was synthesized on the basis of the partial sequence of the genomic DNA. The reverse primer corresponded to the adapter sequence of the transcript primer. 3′-RACE was performed with 5 ng of template cDNA, 5 μM of primer MOL-F1, and 3 μM of adapter primer. The procedure consisted of 25 cycles of 94 °C for 30 s, 55 °C for 30 s, and 72 °C 2 min. 5′-RACE was performed using a 5′-full RACE Core set (Takara Bio) following the manufacturer’s protocols. cDNA was synthesized with 200 pmol/ml of the 5′-terminal phosphorylated transcript primer. Digestion of RNA generated a single strand cDNA, which was then ligated to form a concatemer or circular template. 5′-RACE was performed using TAKARA LA Taq containing 1 μl of the template with 5 μM of primers S1 (GGACTGGGATTTGAACGGAGCGAATTCGAC) and A1 (CAACGGTCTCAGCAATTTAGTGGACCACGG). The amplification consisted of 25 cycles of 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 1 min followed by a denaturation step at 94 °C for 3 min using the product. Nested PCR was done under the same conditions with the internal primer S2 (GGAACGTTTCGTTATGACCAGTTGTGTAC) and A2 (CCGCCTTTGGCATAGACAACCAAGTTTCCG). The resulting product was analyzed as described above.

DNA sequence analysis

The PCR products were subcloned into the plasmid pGEM-T (Promega, Madison, WI) after purification by agarose gel electrophoresis. DNA sequencing was performed by cycle sequencing on an ABI DNA Sequencer (Model 310; Applied BioSystems, Foster City, CA) using the T7promoter and SP6 promoter primers and BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied BioSystems). Sequence data were analyzed using GENETIX-MAC (ver. 15). Homologous sequences were searched using the BLAST and FASTA programs. Sequence alignments were made using the CLUSTAL W program [11].

Expression of the recombinant MOL (rMOL)

A full-length coding sequence PCR product was digested with NdeI and BamHI and cloned into an isopropyl-1-thio-α-D-galactopyranoside-inducible pET3a expression vector (Novagen, Darmstadt, Germany). rMOL was expressed in a BL21 (DE3) strain of E. coli. Induced bacteria were collected by centrifugation and then resuspended in PBS. After the cell suspension was sonicated, the insoluble fraction was removed by centrifugation (10,000 × g, 20 min). rMOL was purified by affinity chromatography with a column of asialofetuin-Sepharose 4B in a similar manner for MOL.

Protein determination

Protein amounts were determined by the method of Lowry et al. [12] with bovine serum albumin as the standard.

Hemagglutination and inhibition assay

Hemagglutination assays were performed in 96-well microtiter plates, in a final volume of 70 μl containing 50 μl of serially diluted lectin solution and 20 μl of a 4 % (v/v) suspension of trypsinized rabbit erythrocytes. After incubation for 1 h at room temperature, the agglutination titer was recorded for the maximum dilution giving positive agglutination. Carbohydrate-binding specificity was determined by the inhibition of agglutination of trypsinized rabbit erythrocytes. Stock solutions of the inhibitors (25 μl) were diluted twofold in 96-well microtiter plates, followed by the addition of 25 μl of lectin solutions with a hemagglutination titer of 8. After 1 h of incubation, 20 μl of 4 % trypsinized rabbit erythrocyte suspension was added to the mixture, and agglutination was measured after standing for 1 h at room temperature. The results were expressed as the minimum concentration of inhibitor required for the complete inhibition of hemagglutination.

Glycan array

Sugar-binding profiling of MOL was determined as described by Tateno et al. [13].

Glycoproteins and glycoside-polyacrylamide used for glycan-array are shown in Table 1.

Table 1 Glycans used for glycan array
Full size table

Thermal stability

Thermal stability profile was obtained by incubating aliquots of MOL at 25–100 °C in a water bath for 30 min. After the thermal treatment, the lectin solutions were chilled on ice, and the remaining activity was determined. Activity at each temperature was expressed as the relative activity compared with the control at 4 °C.

Results and discussion

Purification of MOL

Table 2 shows the summary of purification of MOL from 100 g of fruiting bodies by asialofetuin-Sepharose 4B. Its yield was 75 %, and 16.2 mg of MOL was obtained.

Table 2 Purification of MOL
Full size table

Amino acid sequence of MOL

The peptide sequences after enzymatic digestion and Edman sequencing were assigned to the amino acid sequence deduced from the DNA (Fig. 1). cDNA sequencing of MOL was done by PCR and 5′- and 3′-RACE. The cDNA comprised 561 nucleotides with an open reading frame of 360 nucleotides encoding 120 amino acids (GenBankTM accession number AB 649027). The results agreed with the peptide sequence, but 3 residues (107–109) were not obtained from the peptide sequences. Two putative N-glycosylation sites were found in the sequence, but MOL was not glycosylated.

Expression and characterization of MOL

The MOL open reading frame was cloned into a pET-3a expression vector. The protein was produced in E. coli BL21 (DE3) and purified as described above. rMOL had an electrophoretic mobility in SDS-PAGE identical to that of native protein at 13 kDa (Fig. 2) and eluted as a dimer at the same elution volume as MOL from a TSKgel G2000SW column.

Fig. 1
figure 1

Nucleotide sequence and deduced amino sequence of MOL. The numbers on the left and right margins indicate the order of amino acid residues and nucleotides, respectively. Amino acid residues of mannose-binding motif are shaded in gray. The arrows represent amino acid residues determined by Edman degradation. Dashed line is chymotryptic peptides, continuous line is tryptic peptides, long dashed dotted line is S. aureus V8-digested peptides, and dotted line is putative site of a digestion. Initiation and stop codons are indicated in italics and by an asterisk, respectively

Full size image
Fig. 2
figure 2

Tricine SDS-PAGE analysis of MOL and recombinant MOL. Lane M, protein molecular weight marker; Lane 1, precipitation of bacterial lysate; Lane 2, supernatant of bacterial lysate; Lane 3, flow-through fraction; Lane 4, purified recombinant MOL by affinity chromatography on asialofetuin-Sepharose 4B column; Lane 5, purified MOL by affinity chromatography on asialofetuin-Sepharose 4B column

Full size image

The molecular mass of MOL is calculated to be 13,103 from 120 amino acids. The MOL and rMOL were also subjected to MALDI-TOF mass spectrometry. MOL showed a molecular mass of 13,012.8 ± 3.0 Da whereas rMOL had a slightly lower mass of 12,968.5 ± 3.0 Da. The N-terminus of MOL was blocked but chymotryptic peptides of MOL showed SYVHPYGSTLPENG. The N-terminal sequence was also determined to be SYVHPYGSTL from rMOL. The initial methionine residue was removed.

Hemagglutination activity of MOL

MOL agglutinated native and trypsinized rabbit erythrocytes at a minimum concentration of 0.7 μg/ml. However, MOL did not agglutinate any human erythrocytes, irrespective of trypsin treatment. MOL retained full hemagglutinating activity after the heat treatment at 60 °C for 30 min. Treatment at 70 °C for 30 min inactivated MOL completely.

Sugars and glycoproteins specificity of MOL

The inhibition of the hemagglutinating activity of MOL and rMOL by various saccharides and glycoproteins is summarized in Table 3. For MOL, mannose was weakly inhibitory, other mono- and oligosaccharides were not inhibitory, and asialofetuin was the most potent inhibitor. Quail ovomucoid and bovine thyroglobulin with trimannosyl core and high mannose-N-glycans respectively weakly inhibited hemagglutination, but quail albumin and bovine pancreatic RNase B with only high mannose-N-glycans did not. The specific activity and hemagglutination inhibition of rMOL were almost the same as the native lectin.

Table 3 Inhibition by sugars and glycoproteins of hemagglutinating activity of MOL and rMOL Minimum concentration required for the complete inhibition of titer 4 hemagglutinating activity
Full size table

The result of glycan array analysis of rMOL is shown in Fig. 3. MOL did not react to invertase and ovalbumin, proteins with high mannose N-glycans, and yeast mannans, but preferred asialo- and agalacto-derivatives of α1-acid glycoprotein, fetuin, and transferin. Tateno et al. [13] showed the sugar binding-pattern of concanavalin A, a typical mannose-binding lectin with the same method. The pattern of Con A is quite different from that in Fig. 3. Apparently, sugar-binding specificity of MOL is different from those of typical mannose- binding lectins, and close to that of TxcL1 [14].

Fig. 3
figure 3

Glycan-array of MOL. Cy3-labeled MOL was applied on the array and binding was detected by the scanner. Scan image of MOL was analyzed with the Array Pro analyzer ver. 4. 5

Full size image

Sequence homology

BLAST searches showed the highest similarity of the MOL sequence to Galanthus nivalis agglutinin (GNA)-like protein from liverwort Marchantia polymorpha [15], mannose-binding lectin from Zea mays [16], mannose specific lectin from Allium ampeloprasum [17], antifungal protein from Gastrodia elata (NCBI accession number AAG53445), mannose-binding lectin from Picea abies (NCBI accession number AAZ30388), and D-mannose binding lectin from Signiliparus rugosus (NCBI accession number ZP_07963966). All of these proteins were bulb-type mannose-specific lectins that have the mannose-binding motif QXDXNXVXY (Fig. 4) [18]. The sequence of the first bulb-type mannose-specific lectin was reported from G. nivalis [19], and subsequently, many lectins have been isolated from monocots and liverworts. And the genes were also found in bacteria Signiliparus, conifers, and some fungi [5, 18]. Furthermore, one lectin of Ascomycota is deduced to be this family of lectin from the partial sequences [6]. However, the Ascomycota lectin showed completely different sugar-binding specificity, showing the preference to xylose and inulin. MOL also recognized complex N-glycans rather than high mannose N-glycans, though the two mannose-binding motifs were found. The sugar binding specificity of MOL is closer to those of GNAmaize [16] and RVL [20], which recognized not only high mannose N-glycans but also complex N-glycans. MOL showed sequence homology to mannose-binding lectins from Z. may and M. polymorpha, with sequence identities of 38 %. MOL included two tandemly repeated homologous domains (residues 64–84 and 96–116) with 62 % internal homology.

Fig. 4
figure 4

Comparison of sequences of MOL with some bulb-type mannose-specific lectins. Sequence alignment of bulb-type mannose-specific lectins from P. abies, Z. may, M. polymorpha, A. ampeloprasum, G. elata, S. rugosus, R. vivipara, and MOL, performed using Clustal W. Identical residues are indicated by asterisks, and similar residues are indicated by dashes or colons. The amino acids that constitute the mannose-binding motifs are shaded in gray

Full size image

Conclusion

This is the first report on Basidiomycota mannose-binding GNA-like lectin with the mannose-binding motif found in many monocots. The finding shows that the bulb-type lectin is common in plant and fungi as similarly known for the ricin type lectins and the Jacalin-related lectins. However, the sugar binding specificities of MOL were close to that of GNAmaize and RVL, and was different from that of Ascomycota lectin. These GNA-like lectins are found not to be highly mannose-specific. Japanese M. oreades may contain a lectin differing from those of European and US M. oreades [21, 22].