Cloning and High-Level Expression of α-Galactosidase cDNA from Penicillium purpurogenum

Abstract

The cDNA coding for Penicillium purpurogenum α-galactosidase (αGal) was cloned and sequenced. The deduced amino acid sequence of the α-Gal cDNA showed that the mature enzyme consisted of 419 amino acid residues with a molecular mass of 46,334 Da. The derived amino acid sequence of the enzyme showed similarity to eukaryotic αGals from plants, animals, yeasts, and filamentous fungi. The highest similarity observed (57% identity) was to Trichoderma reesei AGLI. The cDNA was expressed in Saccharomyces cerevisiae under the control of the yeast GAL10 promoter. Almost all of the enzyme produced was secreted into the culture medium, and the expression level reached was approximately 0.2 g/liter. The recombinant enzyme purified to homogeneity was highly glycosylated, showed slightly higher specific activity, and exhibited properties almost identical to those of the native enzyme from P. purpurogenum in terms of the N-terminal amino acid sequence, thermoactivity, pH profile, and mode of action on galacto-oligosaccharides.

α-Galactosidase (αGal) (EC 3.2.1.22) is of particular interest in view of its biotechnological applications. αGal from coffee beans demonstrates a relatively broad substrate specificity, cleaving a variety of terminal α-galactosyl residues, including blood group B antigens on the erythrocyte surface. Treatment of type B erythrocytes with coffee bean αGal results in specific removal of the terminal α-galactosyl residues, thus generating serological type O erythrocytes (8). Cyamopsis tetragonoloba (guar) αGal effectively liberates the α-galactosyl residue of galactomannan. Removal of a quantitative proportion of galactose moieties from guar gum by αGal improves the gelling properties of the polysaccharide and makes them comparable to those of locust bean gum (18). In the sugar beet industry, αGal has been used to increase the sucrose yield by eliminating raffinose, which prevents normal crystallization of beet sugar (28). Raffinose and stachyose in beans are known to cause flatulence. αGal has the potential to alleviate these symptoms, for instance, in the treatment of soybean milk (16).

αGals are also known to occur widely in microorganisms, plants, and animals, and some of them have been purified and characterized (5). Dey et al. showed that αGals are classified into two groups based on their substrate specificity. One group is specific for low-M_r α-galactosides such as pNPGal (p-nitrophenyl-α-d-galactopyranoside), melibiose, and the raffinose family of oligosaccharides. The other group of αGals acts on galactomannans and also hydrolyzes low-M_r substrates to various extents (6).

We have studied the substrate specificity of αGals by using galactomanno-oligosaccharides such as Gal³Man₃ (6³-mono-α-d-galactopyranosyl-β-1,4-mannotriose) and Gal³Man₄ (6³-mono-α-d-galactopyranosyl-β-1,4-mannotetraose). The structures of these galactomanno-oligosaccharides are shown in Fig. 1. Mortierella vinacea αGal I (11) and yeast αGals (29) are specific for the Gal³Man₃ having an α-galactosyl residue (designated the terminal α-galactosyl residue) attached to the O-6 position of the nonreducing end mannose of β-1,4-mannotriose. On the other hand, Aspergillus niger 5-16 αGal (12) and Penicillium purpurogenum αGal (25) show a preference for the Gal³Man₄ having an α-galactosyl residue (designated the stubbed α-galactosyl residue) attached to the O-6 position of the third mannose from the reducing end of β-1,4-mannotetraose. The M. vinacea αGal II (26) acts on both substrates to almost equal extents. The difference in specificity may be ascribed to the tertiary structures of these enzymes.

FIG. 1.

Structures of galactomanno-oligosaccharides.

Genes encoding αGals have been cloned from various sources, including humans (3), plants (20, 32), yeasts (27), filamentous fungi (4, 17, 24, 26), and bacteria (1, 2, 15). αGals from eukaryotes show a considerable degree of similarity and are grouped into family 27 (10).

Here we describe the cloning of P. purpurogenum αGal cDNA, its expression in Saccharomyces cerevisiae, and the purification and characterization of the recombinant enzyme.

MATERIALS AND METHODS

Strains, plasmids, media, and cultivation conditions.

P. purpurogenum no. 618 was isolated from soil and maintained on a medium containing 2.0% (wt/vol) agar, 4.0% (wt/vol) malt extract, 0.2% (wt/vol) NH₄NO₃, 0.1% (wt/vol) KH₂PO₄, and 0.05% (wt/vol) MgSO₄ · 7H₂O. Escherichia coli INVaF′ and plasmid pCRII (Invitrogen) were used for TA cloning of amplified DNA fragments and for preparation of single-stranded plasmid DNA. S. cerevisiae WS3-2A (MATα leu2 ura3 ade8 cys3) and plasmid YEp51 were kindly provided by Y. Jigami (National Institute of Bioscience and Human-Technology, Japan) and used for expression of the αGal cDNA.

Luria-Bertani medium supplemented with ampicillin (100 μg/ml) was used for cultivation of the E. coli transformants. Recombinant strains of S. cerevisiae were cultivated at 30°C in YPD medium (1% [wt/vol] yeast extract, 2% [wt/vol] polypeptone, 2% [wt/vol] glucose). To express the αGal cDNA, galactose was added to the medium instead of glucose.

Amino acid sequencing of αGal purified from P. purpurogenum.

αGal was purified to homogeneity from the culture filtrate of P. purpurogenum as previously reported (25). The purified enzyme was treated with trypsin or V8 protease. The resulting peptides were isolated by reverse-phase high-performance liquid chromatography, and their N-terminal amino acid sequences were determined by a protein sequencer (G1005A; Hewlett-Packard Co.).

Cloning and sequencing analysis of αGal cDNA.

Restriction endonucleases and other enzymes were purchased from Takara Shuzo Co. and used in accordance with the manufacturer’s instructions. Total RNA was prepared from mycelia by the phenol-chloroform method (30), and poly(A)⁺ RNA was purified with an oligo(dT)-cellulose column. A DNA fragment encoding a portion of the P. purpurogenum αGal gene was amplified by the reverse transcription (RT)-PCR method with a set of P1 [5′-GCI(T/C)TIGGITGGAA(T/C)(A/T)(G/C)ITGGAA-3′] (I = inosine) and P2 [5′-(T/C)TTCAT(A/T/G)AT(A/T/G/C)GCCCA-3′] primers designed from the N-pep and V-pep sequences in Fig. 2, respectively.

FIG. 2.

Full-length cDNA encoding P. purpurogenum αGal. Four peptide sequences, N-pep, T1-pep, T2-pep, and V-pep, which were obtained from purified P. purpurogenum αGal are underlined. The designed oligonucleotide primers use for RT-PCR, P1 and P2, were based on N-pep and V-pep, respectively. The nucleotide sequences corresponding to the P1 and P2 primers are shown with arrows indicating the 5′-to-3′ direction. Primers P3 to P6 were designed based on the nucleotide sequences of DNA fragment amplified by RT-PCR and used for 5′ and 3′ RACE. These primers are shown with arrows indicating the 5′-to-3′ direction. The termination codon is indicated by an asterisk. Putative N-glycosylation sites are in shaded boxes.

To determine the nucleotide sequence of the full-length cDNA coding for P. purpurogenum αGal, the 5′ and 3′ RACE (rapid amplification of cDNA ends; Marathon cDNA Amplification Kit [Clontech]) technique was used. The 5′ RACE product was amplified with primer P3 (5′-ACCCCAAGATGGGACGTCGGC-3′) (nucleotides [nt] 775 to 795 in Fig. 2) and Marathon adapter primer AP1 (5′-CCATCCTAATACGACTCACTATAGGGC-3′) and then subjected to nested PCR using primer P4 (5′-GATCTTGAACAGCGACCAAGGC-3′) (nt 715 to 736 in Fig. 2) and adapter primer AP2 (5′-ACTCACTATAGGGCTCGAGCGGC-3′). To obtain the 3′ RACE product, the primary PCR using primers P5 (5′-ATGGTACCGCTCAGCAGGTCC-3′) (nt 386 to 406 in Fig. 2) and AP1 was followed by a nested PCR using primers P6 (5′-GCGCCGGATATGAGACGTGTGCTGG-3′) (nt 439 to 464 in Fig. 2) and AP2. The 5′ and 3′ RACE products were cloned into the pCRII vector, and sequence analysis of both strands of the cloned genes was performed by using the 373 DNA sequencer (Applied Biosystems Inc.).

Expression of P. purpurogenum αGal cDNA in yeast.

To construct expression vector YEp-PGA the 5′ and the 3′ RACE products in pCRII were digested with ClaI and HindIII. They were then ligated at these sites. The full-length cDNA was amplified with SalP (5′-GCGGTCGACATGTTAAGTAGTGTAACTGTAGC-3′) and M13 primer M4 (5′-GTTTTCCCAGTCACGAC-3′). SalP included a SalI cleavage site just before the initiation codon of the αGal gene. The cDNA was then digested with SalI and BamHI and ligated with YEp51 between the SalI and BamHI sites. The plasmid was transferred into S. cerevisiae WS3-2A by electroporation using 0.2-cm-diameter cuvettes at 7.5 kV/cm, 200 Ω, and 25 μF with a Gene Pulser (Bio-Rad Laboratories). An SD minus Leu plate (0.67% [wt/vol] yeast nitrogen base; 2% [wt/vol] glucose; 20-μg/ml [each] His, adenosine, and uracil) was used for selection of the yeast transformants. S. cerevisiae WS3-2A carrying YEp-PGA was first grown in 20 ml of YPD medium at 30°C for 24 h. The cells were harvested by centrifugation and cultivated in 100 ml of YPGal medium (1% [wt/vol] yeast extract, 2% polypeptone, 2% [wt/vol] galactose) at 30°C with shaking to express the P. purpurogenum αGal gene.

Enzyme assay and measurement of protein concentration.

αGal activity was assayed by measuring the amount of p-nitrophenol released from p-nitrophenyl-α-d-galactopyranoside (21). One unit of activity was defined as the amount of enzyme releasing 1 μmol of p-nitrophenol from pNPGal per min at pH 4.0 and 40°C.

The distribution of protein in the purification process was determined by measuring the A₂₈₀ and assuming that the absorbance at a concentration of 1 mg of protein/ml is 1.0. The protein contents of the enzyme preparations were measured with a Bio-Rad DC Protein Assay Kit with bovine serum albumin as the standard.

Purification of recombinant αGal.

The culture supernatant (100 ml) was harvested by centrifugation after 9 days of growth. The supernatant was concentrated with Centriprep 10 (Amicon) and dialyzed against 20 mM sodium acetate buffer, pH 5.4, and put on a DEAE-Sepharose Fast Flow column (1.3 by 21 cm; Pharmacia) equilibrated with the same buffer. Proteins were eluted with a linear gradient of 0 to 0.2 M NaCl. The active fractions were collected, concentrated, and dialyzed against 10 mM sodium acetate buffer, pH 3.5, and put on a Mono-S HR 5/5 column (Pharmacia) equilibrated with the same buffer. Proteins were eluted with a linear gradient of 0 to 0.2 M NaCl. Active fractions were collected, concentrated, and put on a HiPrep 16/60 Sephacryl S-200 HR column (Pharmacia) equilibrated with 10 mM sodium acetate buffer, pH 3.5, containing 0.15 M NaCl. The column was washed with the same buffer until protein could no longer be detected in the eluent.

Electrophoretic analysis.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out in 10% polyacrylamide gel as described by Laemmli (14). The proteins in the gel were visualized by staining with Coomassie brilliant blue R-250. The molecular mass was estimated with markers (10 kDa Protein Ladder; Gibco BRL).

Preparation of galactomanno-oligosaccharides.

A galactomanno-oligosaccharide having an α-1,6-galactosyl stub on β-1,4-mannotetraose, Gal³Man₄, was prepared from a hydrolysate of copra galactomannan by using Streptomyces β-mannanase (11). In addition, a galactomanno-oligosaccharide with a terminal galactose at the nonreducing end of β-1,4-mannotriose, Gal³Man₃, was prepared from Gal³Man₄ by cutting off the nonreducing mannosyl residue end of the saccharide with Aspergillus niger β-mannosidase (13).

Substrate specificity.

The action of αGal on oligosaccharides and locust bean gum was monitored by determining the release of d-galactose by using d-galactose dehydrogenase (23). The reaction mixture in 0.5× McIlvaine buffer, pH 4.5, containing the 0.1% (wt/vol) substrate was incubated at 37°C for 24 h. The reaction was terminated by boiling for 5 min. The reaction mixture (40 μl) was added to 100 μl of 1 M Tris-HCl (pH 8.6), 10 μl of 10 mM NAD⁺, and water to make a final volume of 180 μl. The A₃₄₀ was measured as a blank, and 5 μl of d-galactose dehydrogenase was added to start the reaction. The solution was incubated at 37°C for 30 min, and the A₃₄₀ was measured.

Hydrolyses of galacto-oligosaccharides such as melibiose, raffinose, Gal³Man₃, and Gal³Man₄ by the purified native and recombinant αGals were done at pH 4.0 and 30°C. The sugar sample after the enzyme reaction was analyzed by thin-layer chromatography (TLC; Silica gel 60; Merck) for characterization of the hydrolysis products. The reaction products were developed with 1-propanol–nitromethane–water (5:2:3, vol/vol). The sugars on the plate were detected by heating at 140°C for 5 min after spraying with sulfuric acid.

Nucleotide sequence accession number.

The αGal cDNA sequence is available in the DDBJ, EMBL, and GenBank databases under accession no. AB008367.

RESULTS

Cloning and characterization of the P. purpurogenum αGal cDNA.

The gene encoding P. purpurogenum αGal was cloned by PCR with designed primers based on partial amino acid sequences of the purified protein. The nucleotide sequence and deduced amino acid sequence of the 5′ and 3′ RACE products are shown in Fig. 2. Examination of the sequence revealed the presence of one open reading frame of 1,371 bp. The nucleotide sequences of the overlap region of these fragments (between P6 and P4) were identical, and the amino acid sequences of the purified enzyme identified by Edman degradation (N-pep, T1-pep, T2-pep, and V-pep) were found in the sequence. The coding sequence consisted of 19 amino acids of signal sequence and 420 amino acids of mature αGal with a molecular mass of 46.3 kDa. Nine putative N-glycosylation sites were found in the sequence, and this is coincident with the reactivity with concanavalin A (25).

A comparison of the amino acid sequences of Trichoderma reesei (17), S. carlsbergensis (27), M. vinacea (24, 26), coffee bean (32), and human (3) αGals with that of P. purpurogenum αGal is depicted in Fig. 3. P. purpurogenum αGal showed a considerable degree of homology with these enzymes (35 to 57%). However, bacterial αGals, such as those of Escherichia coli (2, 15) and Streptococcus mutans (1), showed relatively little (less than 20%) homology to P. purpurogenum αGal.

FIG. 3.

Sequence homology of αGals from different sources. The amino acid sequences of P. purpurogenum αGal (P.p.), T. reesei AGLI (T.r.), S. carlsbergensis αGal (Yeast), M. vinacea αGalI (M.v.I), M. vinacea αGalII (M.v.II), coffee bean αGal (Coffee), and human αGalA (Human) were aligned for optimal sequence similarity by using the program GENETYX (Software Development, Tokyo, Japan). Hyphens indicated gaps, and the yeast and human αGal sequences were truncated at the C terminus as indicated by asterisks. Identical amino acid residues, five of seven or more at the same position, are shaded, and cysteine residues located at the insertion sequences are dotted.

P. purpurogenum αGal and T. reesei AGLI showed the highest similarity among the αGals (the sequence identity was 57%). In addition; a unique 34-amino-acid insertion from residues 150 to 183 of P. purpurogenum αGal was also observed in the sequence of T. reesei AGLI. These two enzymes had nine Cys residues at identical positions, including two Cys residues in the insertion and C-terminal regions. Thus, it is likely that these enzymes are in similar tertiary structures.

It is interesting that there are even numbers of Cys residues in the insertion region; for example, residues 25 to 34 of human αGal contain two Cys residues, residues 147 to 180 of P. purpurogenum (146 to 180 of T. reesei) contain two Cys residues, and yeast αGal residues 198 to 219 and M. vinacea αGal I residues 196 to 215 contain four Cys residues. These Cys residues might have a role in maintaining the stability of these enzymes by forming an S-S bridge(s) in the molecule.

Expression and purification of recombinant αGal in S. cerevisiae.

P. purpurogenum αGal cDNA was expressed in S. cerevisiae under the control of the yeast GAL10 promoter. S. cerevisiae cells carrying YEp-PGA were cultured in YPGal medium, and αGal production was monitored. αGal was secreted into the medium, and the activity reached about 63 U/ml of medium at 216 h (equivalent to 0.21 g/liter of medium). Little αGal activity was detected in the periplasmic space or intracellular fractions throughout the culture period. No background activity was detected when the host cells carrying the expression vector YEp51 were cultured under the same conditions (data not shown). Recombinant αGal was purified to homogeneity by using three chromatographic steps (Table 1). Starting from the 100-ml culture medium, 6.75 mg of the purified αGal was obtained with 33% recovery.

TABLE 1.

Purification of recombinant αGal

Step	Enzyme activity (U)	Amt of protein (mg)	Sp act (U/mg)	Yield (%)	Purification (fold)
Dialysis	6,150	1,677	3.67	100	1
DEAE-Sepharose FF	3,690	66.6	55.4	60.0	15.1
Mono-S HR	2,838	10.02	284	46.3	77.4
Sephacryl S-200 HR	2,016	6.75	299	32.8	81.5

Characterization of recombinant αGal.

Purified recombinant αGal showed a single but broad protein band with the characteristics of a glycoprotein on SDS-PAGE, and its molecular mass was estimated to be in the range of 70 to 100 kDa (Fig. 4). The apparent molecular mass of the recombinant enzyme was 10 to 30 kDa larger than that of the native enzyme (25); however, no difference was found by SDS-PAGE between the recombinant and native enzymes after treatment with endoglycosidase H (data not shown). This suggests that the differences in the molecular masses existed in the carbohydrate moieties. The specific activity of the purified enzyme was 300 U/mg, which is slightly higher than that of the native enzyme (245 U/mg), and this may be due to the difference between the carbohydrate moieties or the purity of the enzymes.

FIG. 4.

Homogeneity and molecular mass determination of recombinant αGal by SDS-PAGE. The enzyme (10 μg) was electrophoresed on 10.0% (wt/vol) polyacrylamide gel and stained with Coomassie brilliant blue R-250. Lanes: 1, molecular size markers; 2, recombinant αGal; 3, recombinant αGal digested with 2 mU of endo-β-N-acetylglucosaminidase H at pH 5.3 and incubated 37°C for 16 h.

The N-terminal amino acid sequence of purified αGal determined by Edman degradation was found to be identical to that of the native enzyme (data not shown). This result indicates that the produced recombinant enzyme is properly processed to yield the mature form in yeast cells.

Some properties of the recombinant enzymes are summarized in Table 2. The recombinant enzyme was most active at pH 4.5 and 55°C, and it was stable from pH 4.0 to 6.0 and up to 40°C. The effects of pH and temperature on the activity and stability of the recombinant enzyme were identical to those on the native enzyme.

TABLE 2.

Comparison of some properties of recombinant and native αGals

αGal	Optimum pH	Optimum temp (°C)	pH range for stability	Temp range (°C) for stability	Molecular mass (kDa)
					Undigested	After diges-tion with endoglyco-sidase H
Recombinant	4.5	55	4.0–6.0	<40	70–100	52–60
Native	4.5	55	4.0–6.0	<40	67	55

The substrate specificity of the recombinant enzyme is shown in Fig. 5 and 6. The best substrate for the enzyme was Gal³Man₄, which was prepared from the β-mannanase digest of copra galactomannan, followed by raffinose. However, melibiose, stachyose, and Gal³Man₃ were not effectively hydrolyzed compared with Gal³Man₄. The enzyme also hardly liberated galactosyl residues from the polymer substrate galactomannan (data not shown).

FIG. 5.

Action of native (a) and recombinant (b) αGals on galactomanno-oligosaccharides. The reaction mixture was composed of 80 μl of 1% (wt/vol) substrate, 80 μl of McIlvaine buffer (pH 4.5), and 40 μl (0.4 U) of enzyme solution. The reaction was done at 30°C, and 20 μl of the reaction mixture was withdrawn at each time indicated. Three microliters of the mixture was used for TLC. Gal, authentic galactose; M, authentic mannose to mannopentaose, from top to bottom.

FIG. 6.

Actions of native (a) and recombinant (b) αGals on galacto-oligosaccharides. The reaction mixture was composed of 40 μl of 1% (wt/vol) substrate, 40 μl of McIlvaine buffer (pH 4.5), and 20 μl (0.2 U) of enzyme solution. The reaction was performed at 30°C, and 20 μl of the reaction mixture was withdrawn at each time indicated. Three microliters of the mixture was used for TLC. Gal, authentic galactose; Mel, melibiose; Raf, raffinose; Sta, stachyose.

DISCUSSION

αGals are classified into two groups based on substrate specificity (5). Some enzymes are specific for low-M_r substrates, and others are able to efficiently hydrolyze polymer substrates. We have found that there are three kinds of αGals which act on lowM_r substrates (26). The first group acts only on the terminal α-galactosyl residue of a substrate such as Gal³Man₃, the second one is specific only for the stubbed α-galactosyl residue of a substrate such as Gal³Man₄, and the third shows a preference for both residues. Recombinant and native P. purpurogenum αGals could hardly act on the polymer substrate and showed a preference for the stubbed galactosyl residue among the low-M_r substrates. T. reesei αGal was reported to show a synergistic action on galactomannan with β-mannanase and to effectively liberate galactose residues (31), suggesting that T. reesei αGal has a high ability to liberate the stubbed galactosyl residue from galactomanno-oligosaccharides having a stubbed α-galactosyl residue with a high yield (13). Until now, only M. vinacea αGalI and yeast αGals have shown specificity for the terminal α-galactosyl residue and have rarely been seen to act on the stubbed α-galactosyl residue of galactomanno-oligosaccharides. Other αGals show a preference for the stubbed α-galactosyl residue of the galactomanno-oligosaccharides.

Three genes encoding αGals from T. reesei were isolated by expression cloning, and some properties of the enzymes produced by yeast were analyzed (17). Based on its substrate specificity, AGLI might correspond to the 50-kDa αGal which was purified from T. reesei RUT C-30 and previously characterized (31). The physicochemical properties and substrate specificity of T. reesei AGLI can resemble those of P. purpurogenum αGal. This might also be due to the similarity of the primary structures of the two enzymes. Although αGal from T. reesei RUT C-30 was nonglycosylated (31), P. purpurogenum αGal was highly glycosylated. P. purpurogenum secreted over 10 times more αGal into the culture medium than did T. reesei (25, 31). In this study, S. cerevisiae secreted P. purpurogenum αGal into the culture medium at about 200 mg/liter. This value is about 10 times higher than that of the native αGal produced by P. purpurogenum (25). In order to increase recombinant αGal production, optimization of the recombinant yeast culture conditions has also been studied (7, 22, 33). The level of P. purpurogenum αGal expressed in yeast is comparable to those of coffee αGal expressed in yeast and insect cells (32).

Many αGals have been purified and characterized, and genes encoding αGals have been isolated from several sources. αGals from T. reesei and humans have been crystallized, and X-ray diffraction studies are in progress (9, 19). However, only a few αGals were studied based on structure-function relationships. Thus, the structures of the active sites and catalytically important amino acid residues still remain largely unknown, and little is known about the structure-function relationship of αGal. The experimental data obtained and the expression system used in this study will be useful in studying the structure-function relationships of αGals.