Characterization of the bga1-encoded glycoside hydrolase family 35 b-galactosidase of Hypocrea jecorina with galacto-b- D-galactanase activity Christian Gamauf 1 , Martina Marchetti 2 , Jarno Kallio 3 , Terhi Puranen 3 , Jari Vehmaanpera ¨ 3 , Gu ¨ nter Allmaier 2 , Christian P. Kubicek 1 and Bernhard Seiboth 1 1 Research Area Gene Technology and Applied Biochemistry, Institute of Chemical Engineering, Vienna University of Technology, Austria 2 Research Area Instrumental Analytical Chemistry, Institute of Chemical Technology and Analytics, Vienna University of Technology, Austria 3 Roal Oy, Rajama ¨ ki, Finland The enzyme b-galactosidase (EC 3.2.1.23) catalyses the hydrolysis of terminal nonreducing b-d-galactose residues in b-d-galactosides as, for example, lactose (1,4-O-b-d-galactopyranosyl-d-glucose) and structur- ally related compounds. It is found in plants and animals, as well as in a wide variety of microorganisms including yeasts, fungi, bacteria and Archaea. Accord- ing to the Carbohydrate Active Enzymes database (http://www.cazy.org/) [1], b-galactosidases are members of four different glycoside hydrolase (GH) families (GH1, GH2, GH35 and GH42), indicating their structural diversity. In biotechnology, they are mainly applied for the hydrolysis of lactose to d-glucose and d-galactose in various products of the dairy industry [2,3]. This results in improved quality of the end product (softer texture of ice cream, faster ripening of cheese, etc.) and also lessens the problem of lactose intolerance, which is prevalent in more than half of the world’s population [4]. In addition, b-gal- actosidases catalyse transgalactosylation reactions of various b-d-galactosides including lactose. Recently, these galacto-oligosaccharides have attracted consider- able interest because of a proposed beneficial effects on human health [5,6]. The filamentous fungus Hypocrea jecorina (ana- morph: Trichoderma reesei) is a potent producer of Keywords b-galactosidase; galactanase; Hypocrea jecorina; substrate specificity; transglycosylation Correspondence C. Gamauf; Research Area Gene Technology and Applied Biochemistry, Institute of Chemical Engineering, Vienna University of Technology, Getreidemarkt 9 ⁄ 166-5, A-1060 Vienna, Austria Fax: +43 158 801 17299 Tel: +43 158 801 17265 E-mail: gamauf@mail.zserv.tuwien.ac.at Website: http://www.vt.tuwien.ac.at/ (Received 16 January 2007, accepted 22 January 2007) doi:10.1111/j.1742-4658.2007.05714.x The extracellular bga1-encoded b-galactosidase of Hypocrea jecorina (Trichoderma reesei) was overexpressed under the pyruvat kinase (pki1) promoter region and purified to apparent homogeneity. The monomeric enzyme is a glycoprotein with a molecular mass of 118.8 ± 0.5 kDa (MALDI-MS) and an isoelectric point of 6.6. Bga1 is active with several disaccharides, e.g. lactose, lactulose and galactobiose, as well as with aryl- and alkyl-b-d-galactosides. Based on the catalytic efficiencies, lactitol and lactobionic acid are the poorest substrates and o-nitrophenyl-b-d-galacto- side and lactulose are the best. The pH optimum for the hydrolysis of gal- actosides is  5.0, and the optimum temperature was found to be 60 °C. Bga1 is also capable of releasing d-galactose from b-galactans and is thus actually a galacto- b- d-galactanase. b-Galactosidase is inhibited by its reac- tion product d-galactose and the enzyme also shows a significant trans- ferase activity which results in the formation of galacto-oligosaccharides. Abbreviations ACN, acetonitrile; GH, glycoside hydrolase; LIF, laser-induced fluorescence; oNPG, o-nitrophenyl-b- D-galactopyranoside; pNPG, p-nitrophenyl- b- D-galactopyranoside. FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS 1691 cellulolytic and hemicellulolytic enzymes and due to its strong promoters and excellent secretion capacities it is also of interest for the expression of heterologous proteins [7]. One of the carbon sources used for cellulase production is lactose, which requires an extracellular b-galactosidase activity for being assimilated by the fungus. The gene encoding this enzyme, bga1, has recently been described [8]. Like most other fungal b-galactosidases, the encoded protein belongs to GH35. The ability of filamentous fungi to assimilate and grow on lactose is enigmatic, as lactose is unlikely to occur in their natural environment. This argument is also substantiated by the finding that bga1 is induced by d-galactose and l-arabinose, thus pointing to a role for the enzyme in the degradation of plant poly- and oligosaccharides which are present in the natural hab- itat of a rhizosphere-competent and potentially endo- phytic fungus like Trichoderma [9]. Consequently, our aim was to purify Bga1 and study its substrate profile. We reasoned that this information may provide us with a hint towards the role of this enzyme in the physiology of the fungus, as well as to its potential use in biotechnology. Results Purification of b-galactosidase Bga1 In order to facilitate the purification of the bga1-enco- ded b-galactosidase, we fused its ORF to the promoter region of the pyruvate kinase-encoding pki1 gene and cultivated the resulting recombinant strain on d-glucose. Under these conditions, b-galactosidase formation parallels growth, and the culture superna- tant was thus harvested when two-thirds of the carbon source had been consumed. The resulting filtrate had a specific activity of 83.5 nkatÆmg )1 (with o-nitrophenyl- b-d-galactopyranoside; oNPG) and analysis by SDS ⁄ PAGE showed a protein band in the expected molecular mass range (110 kDa; Fig. 1), which already accounted for a significant part of the total secreted protein. Concentration by ultrafiltration and subse- quent purification by gel filtration and cation-exchange chromatography (Table 1) yielded a single protein band at  110 kDa, proving a homogenously purified protein. The specific activity with oNPG was 828.2 nkatÆmg )1 , indicating a roughly tenfold enrich- ment over the original activity. Physicochemical properties and stability of Bga1 The size of denatured H. jecorina Bga1, measured using SDS ⁄ PAGE (see above), is  110 kDa. Because virtually the same molecular mass was determined by gel-permeation chromatography of the native enzyme, we conclude that the enzyme is a monomer. The iso- electric point of the unfolded protein was determined as 6.6, which is in good agreement with the value calculated from the amino acid sequence (6.35) using the protparam tool (http://www.expasy.org/tools/ protparam.html), indicating the absence of any post- translational modification altering the charge of the protein (e.g. phosphate, sulfate). Bga1 is stable over a broad range of pH values, which extends far into the alkaline pH range and inac- tivation occurs rapidly only below pH 3.0 (data not shown). Incubation of the enzyme at different tempera- tures up to 60 °C for 1 h also led to recovery of almost all of the original activity, but temperatures over 65 °C led to rapid denaturation (data not shown). Bga1 is a glycoprotein MALDI-TOF-MS in the linear mode was used to determine the exact molecular mass, and a value of 118 789 ± 485 Da was obtained based on single-, dou- 1 kD 200 150 120 85 100 70 60 50 40 23 4 Fig. 1. SDS ⁄ PAGE of the purified Bga1. Lanes: 1, molecular size marker; 2, culture supernatant; 3, combined factions after gel filtra- tion; 4, purified protein after ion-exchange chromatography. Each lane contained 10 lg protein. Table 1. Purification of the H. jecorina b-galactosidase. Step Total protein (mg) Total activity (nkat) Specific activity (nkatÆmg )1 ) Yield (%) Enrichment (fold) Culture filtrate 24.45 2041.2 83.5 100.0 1 Gel filtration 1.29 693.0 535.8 34.0 6.4 Ion-exchange chromatography 0.32 261.0 828.2 12.8 9.9 b-Galactosidase of Hypocrea jecorina C. Gamauf et al. 1692 FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS ble- and triple-charged molecular ions (Fig. 2A). This is higher than the theoretical value calculated from the amino acid sequence of the mature protein (109 301 Da). The asymmetric peak of the singly charged molecule indicates a heterogeneity which was also observed in the double- and triple-charged mole- cules at the high mass side of the peaks (indicated by asterisks in Fig. 2A). Capillary gel electrophoresis- on-the-chip confirmed that the isolated protein exhibits heterogeneities which are again reflected as an asym- metric peak shape (Fig. 2B, tailing is marked with an asterisk). Because Bga1 is an extracellular protein, glycosyla- tion was expected, and this assumption was supported by glycoprotein staining (data not shown). Taking the average size of Trichoderma N-glycosylation structures of 1500 Da into account [10], the difference between the theoretical and the determined molecular mass of 9500 Da (see above) would predict the presence of six N-glycosylation antennae. Analysis of the Bga1 amino acid sequence using the NetNGlyc tool (http:// www.cbs.dtu.dk/services/NetNGlyc/) revealed the pres- ence of 11 consensus sites for N-linked glycosylation at positions 287, 402, 434, 536, 544, 627, 709, 782, 810, 836 and 930. Because Bga1 is orthologous to a Penicil- lium sp. b-galactosidase, for which the 3D structure and the attached N-glycans have been determined [11], we compared the positions in the two proteins. This analysis (Fig. 3) showed that of the seven positions identified in Penicillium sp., four are conserved in Bga1. All these positions are located at the surface of the jelly roll domain of Bga1 and thus most probably are glycosylated in vivo. To further substantiate these findings, in gel digestion of the highly purified protein was carried out. Seventeen peptides resulting from Bga1, evenly distributed over the protein, could be detected by MALDI-RTOF-MS (Fig. 3). This sums up to a sequence coverage of 12.4% for the mature pro- tein (without the predicted signal peptide). The detec- ted peptides represent amino acid sequences not bearing potential N-glycosylation sites. This failure in detecting glycopeptides despite step-wise elution to overcome suppressing effects during the MALDI pro- cess might be explained by the much higher ionization efficiency for nonglycosylated peptides which were preferentially eluted in the solution containing 50 and 75% (v ⁄ v) acetonitrile (ACN) where also most of the N-glycosylated peptides were expected. None of these tryptic peptides contained a consensus for potential N-glycosylation, and thus none of the sites mentioned above could be falsified. Optimal temperature and pH for catalysis The optimal temperature and pH for the reaction of Bga1 with oNPG as substrate was determined (Fig. 4): maximal activity was found at pH 5.0 and 60 °C, and a substantial decrease was noted at alkaline pH and at temperatures > 60 °C. The enzyme retained more activity at acidic pH. Substrate specificity for b-galactosides and kinetic properties of Bga1 The substrate specificity of the purified H. jecorina Bga1 was determined towards oNPG, methyl-b- d-galactoside and various disaccharides, and the kin- etic constants were calculated (Table 2). The catalytic efficiencies (k cat ⁄ K m ) indicate that oNPG and lactu- lose (4-O-b-d-galactopyranosyl-d-fructose) are the best substrates for Bga1. This is the result of both a lower Michaelis constant K m , as well as a higher V max for these substrates. In addition, Bga1 also showed significant activity with galactobiose (4-O-b- d-galactopyranosyl-d-galactose), while its activity with lactitol (4-O-b-d-galactopyranosyl-d-glucitol), lactobionic acid (4-O-b-d-galactopyranosyl-d-gluconic acid) and methyl-b-d-galactoside was poor. The effect of hydrolysis products (each at 10 mm final concentration) on the hydrolysis of oNPG by Bga1 under standard assay conditions was studied. 0 50 100 %Int 50 000 100 000 m /z [M] + [M] 3+ [M] 2+ * * * 100 200 300 [FU] 20 30 40 50 60 70 [s] 150 000 A B Fig. 2. Establishment of the exact molecular mass by MS and pro- tein purity by capillary gel electrophoresis-on-the-chip. (A) Determin- ation of the exact molecular mass of H. jecorina Bga1 by positive ion MALDI-TOF-MS in the linear mode. The singly, doubly and triply charged ions are indicated by respective symbols. Asterisks mark the asymmetric peak flank, indicating a heterogenic glycosylation pattern. (B) Capillary gel electrophoresis-on-the-chip electrophero- gram of 0.45 lg Bga1 detected by LIF (FU, fluorescence units). The asterisk again marks the asymmetric peak flank, indicating a heterogenic glycosylation pattern. C. Gamauf et al. b-Galactosidase of Hypocrea jecorina FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS 1693 Although the activity was only slightly reduced in the presence of 10 mmd-glucose (data not shown), d-galactose had a significant impact on oNPG hydro- lysis (Fig. 5). The inhibition was competitive with an inhibition constant K i of 1 mm. When the reaction products were monitored using HPLC, formation of transglycosylation products became evident (Fig. 6). Peaks with different retention times than the substrates and hydrolysis products of the reaction were detected with all disaccharides tested, and the concentration of the putative transglycosyla- tion products was inversely correlated with that of d-galactose. The degree of transglycosylation was dependent on the concentration of the substrate, as also found in other studies [12,13]. Activity of Bga1 on polymeric substrates The ability of Bga1 to hydrolyse b-galactosidic bonds in polymeric substrates was tested with two b-1,4-galactans (from lupin and potato) and one b-1,3- ⁄ b-1,6-arabinogalactan (from larch wood). Fig. 3. Alignment of the b-galactosidases from H. jecorina and Penicillium sp. Shaded residues are conserved in both enzymes. The solid lines indicate the tryptic peptides detected by MALDI peptide mass finger- printing. Predicted (H. jecorina; CAD70669) and confirmed (Penicillium sp.; CAF32457) N-glycosylation sites are marked by darker shading and sites conserved in both sequences by arrows. Diamonds indicate amino acid residues involved in substrate binding and catalysis [11]. b-Galactosidase of Hypocrea jecorina C. Gamauf et al. 1694 FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS HPLC analysis of the reaction products confirmed that the enzyme released d-galactose in a time- and concentration-dependent manner. The reaction fol- lowed a Michaelis–Menten kinetic, and the constants for the polymeric substrates are given in Table 3. Turnover numbers (k cat ) on all three polymeric sub- strates were in the same range as on galactobiose (5.68 s )1 ) and significantly higher than on most other galactosides tested (Table 2). The k cat ⁄ K m ratios are less favourable, but the lower values can be explained taking into account the fraction of d-galactose units in the polymers that are actually accessible for the b-galactosidase (those that are on the nonreducing ends of the chains, e.g. 36.8% in arabinogalactan from larch wood) [14]. The enzyme showed highest activity against the d-galactose-richest galactan from lupin, which is consistent with an activity only against b-galactosidic bounds. This was also supported by the finding that no other monosaccharides were detectable during HPLC analysis of the enzymatic assays. Bga1 was also unable to hydrolyse p-nitro- phenyl-b-l-arabinofuranoside, which is the most typical l-arabinose-linkage found in arabinogalactans [15]. Discussion In this study, we characterized a b-galactosidase from H. jecorina belonging to GH family 35. Although b-galactosidases have been purified and characterized from a variety of sources [3,5,16], their GH family affi- liation is unknown in most cases, which makes com- parison of the obtained results difficult. To the best of our knowledge, this is the first report of the enzymo- logical properties of a b-galactosidase identified as a member of GH35. H. jecorina Bga1 hydrolysed all b-galactosides tes- ted, but with clearly different preference: based on the ratio of k cat ⁄ K m , the aromatic artificial galactoside oNPG was the most preferred substrate, which may indicate a beneficial role of hydrophobicity in one of the steps of the hydrolysis. Among naturally occur- ring galactosides, galactobiose was the best substrate. Substitution of the O-4-linked hexose by a corres- ponding polyol (lactitol) or a corresponding sugar acid (lactobionic acid) severely impaired the catalytic efficacy. This was in higher proportions due to a decrease in k cat , suggesting that such substitutions may interfere with proton assistance of the acid⁄ base residue or the nucleophilic attack on C-1 of d-galac- tose [17]. Interestingly, lactose was also a comparably poor substrate for Bga1, which coincides with the low rate of its assimilation by H. jecorina, and supports the assumption that lactose is not the natural sub- strate for Bga1. 0 20 40 60 80 100 23456789 pH enzyme activity [arbitary units] 0 20 40 60 80 100 10 20 30 40 50 60 70 80 Temperature [°C] en z ym ea c t ivity [a rb ita r y u nits] A B Fig. 4. pH and temperature optimum of the H. jecorina Bga1. The activity was assayed at the indicated pH (A) and temperatures (B) as described in the Experimental procedures. Table 2. Kinetic parameters of H. jecorina b-galactosidase with various b-galactosides. Substrate K m (mM) V max (nkatÆmg )1 ) k cat (s )1 ) k cat ⁄ K m (LÆg )1 Æs )1 ) o-Nitrophenyl-b- D-galactopyranoside 0.36 ± 0.01 144.61 ± 0.15 17.31 ± 0.02 159.41 ± 5.41 Galactobiose 9.06 ± 2.48 47.46 ± 6.26 5.68 ± 0.69 1.83 ± 0.51 Lactobionic Acid 15.31 ± 2.34 0.83 ± 0.04 0.10 ± 0.004 0.018 ± 0.003 Lactitol 19.77 ± 2.00 1.04 ± 0.05 0.13 ± 0.01 0.018 ± 0.002 Lactose 8.79 ± 1.00 5.78 ± 0.16 0.69 ± 0.02 0.23 ± 0.02 Lactulose 0.56 ± 0.06 12.74 ± 0.34 1.52 ± 0.04 7.98 ± 0.78 Methyl-b- D-galactopyranoside 2.85 ± 0.39 0.71 ± 0.03 0.09 ± 0.003 0.15 ± 0.02 C. Gamauf et al. b-Galactosidase of Hypocrea jecorina FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS 1695 The low affinity for lactose also explains the relat- ively poor growth of H. jecorina on this carbon source: Seiboth et al. [8], using p-nitrophenyl-b-d-galacto- pyranoside (pNPG) as a substrate, reported extra- cellular Bga1 activity during growth on lactose to be 86 lmolÆ(minÆg mycelial dry weight) )1 . Taking the ratio of K m and V max for lactose : pNPG, as deter- mined in this study, in consideration, an actual lactose hydrolysing activity of 0.4 lmolÆ(minÆg mycelial dry weight) )1 for lactose can be calculated. Thus, assuming that H. jecorina grows on an initial lactose concentra- tion of 10 or 20 gÆL )1 (80–85% substrate saturation), this is equivalent to a hydrolysis rate of  20 mg lac- toseÆ(hÆg mycelial dry weight) )1 or 0.5 gÆ(dayÆg mycelial dry weight) )1 . Despite being less efficient on lactose than on other b-galactosides, Bga1 may still offer some advantages for lactose hydrolysis from a biotechnological per- spective: although its K m for lactose is 8.8 mm, this is significantly lower than the corresponding values of other fungal b-galactosidases used commercially (e.g. Talaromyces thermophilus,18mm; Aspergillus oryzae, 36–180 mm; A. niger, 54–99 mm; Kluyveromyces fragil- is, 15–52 mm; K. lactis,35mm) [18,19]. In addition, the ratio of K i to K m calculated for d-galactose and lactose, which can be interpreted as a specificity con- stant that determines preferential binding of the sub- strate vs. that of the monosaccharide end products, is 0.11, which compares favourably with the K i,Gal ⁄ K m,Lac ratio reported, e.g. for A. oryzae and A. niger (0.01 and 0.006, respectively) [19]. In view of these advan- tages, the ability of Bga1 to form transglycosylation products, albeit a known property of GH35 b-galacto- sidases [5,12,20,21], is also noteworthy. b-Linked oligosaccharides derived from d-galactose and other hexoses are of considerable interest, partially as poten- tial prebiotics for the food industry, as well as in phar- macology and medicine [22,23]. The main reason for studying the enzymological properties of Bga1 in more detail was our interest in the physiological role of this protein in H. jecorina. 0 20 40 60 80 100 0246810 Galactose [mM] enzyme activity [arbitary units] 0.25 mM oNPG 0.75 mM oN PG A 0 0.02 0.04 0.06 0.08 0.10 0.12 -2 -4 2 4 6 8 10 12 0.25 mM oNPG 0.75 m M oNPG Galactose [mM] K i 1/ Var b t a ry uni [i t s ] B 0.25 mM oNPG 0.75 m M oNPG Fig. 5. Inhibition of oNPG hydrolysis by D-galactose. (A) Addition of D-galactose to the basic b-galactosidase assay (see Experimental procedures) results in a significant reduction of the enzyme activity. The reduction can be partially overcome by increasing the substrate concentration, indicating a competitive inhibition mechanism. (B) Dixon diagram for determination of the inhibition constant K i . 6 6 7 7 4 4 8 8 9 9 10 10 11 11 12 12 13 13 14 14 retention time [min] retention time [min] detector units detector units 0 0 2 2 4 4 6 6 1 1 2 2 3 3 A B Fig. 6. Formation of transglycosylation products during lactose hydrolysis by H. jecorina b-galactosidase. Chromatogram of a partial hydrolysis of 10 m M (A) and 100 mM (B) lactose by Bga1. Peaks: 1, transglycosylation product; 2, lactose; 3, glucose; 4, galactose. b-Galactosidase of Hypocrea jecorina C. Gamauf et al. 1696 FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS Species of Trichoderma are known to be able to colon- ize and grow in the rhizosphere of plants [9,24]. In bacteria, rhizosphere competence is related to the abil- ity to utilize arabinogalactan from the root mucilage [25], and it is likely that similar mechanisms may have evolved in rhizosphere competent fungi too. In any case, Bga1 is able to act on polymeric b-1,3- and b-1,4-galactans. A role for Bga1 in arabinogalactan degradation may also explain why bga1 expression and that of the Leloir pathway genes gal1 (encoding galac- tokinase) and gal7 (encoding galactose-1-phosphate uridylyltransferase) is induced by both d-galactose and l-arabinose [26,27]. The ability to attack galactose poly- mers may also lend to speculate about the structure of the Bga1 protein, but further studies are necessary to find out which domains could be involved in recogni- tion and⁄ or binding to the polysaccharides. Also, it will be intriguing to learn whether other enzymes act synergistically with Bga1. A prerequisite for such stud- ies, however, is a more detailed knowledge of the structure of the commercially available b-galactans which can be used as a model for such studies. Experimental procedures Substrates Unless indicated otherwise, all substrates and peptides and proteins for MS calibration were purchased from Sigma (St Louis, MO) and at least of analytical grade. Arabino- galactan was purchased from Fluka (Buchs, Switzerland) and galactan from lupin and potato was from Megazyme (Bray, Ireland). Strain and culture conditions To purify the extracellular b-galactosidase, H. jecorina strain PKI-BGA13, a recombinant of QM9414 (ATCC 26921), which carries multiple copies of a pki1:bga1 cassette allowing overexpression of Bga1 during growth on d-glucose, was used. Construction of the expression vector and of the bga1- overexpressing strain was performed as described previously [8]. The fungus was cultivated in a Braun Biostat ED bio- reactor (working volume 10 L) for 30 h using the following medium (gÆL )1 ): d-glucose 40, bacto peptone 4, yeast extract 1, (NH 4 ) 2 SO 4 2.8, KH 2 PO 4 4, MgSO 4 Æ7H 2 O 0.6, CaCl 2 0.6, FeSO 4 Æ7H 2 O 0.005, MnSO 4 ÆH 2 O 0.0016, ZnSO 4 Æ7H 2 O 0.0014, CoCl 2 0.002. Struktol (1 mLÆL )1 ) was added at the beginning of the cultivation as an antifoam agent. The fermenter was inoculated with 200 mL of a shake flask preculture grown for 65 h in the same medium. The temperature was kept constant at 28 °C and the pH between 4.8 and 5.2 by addition of 12.5% (w ⁄ v) NH 4 OH or 17.5% (w ⁄ v) H 3 PO 4 , respectively. Aeration was 1 vvm, and Impeller speed was set to 200 r.p.m., and after 12 h linearly increased to 750 r.p.m. at a rate of 30 r.p.m.Æh )1 . Purification of b-galactosidase The fermenter broth was withdrawn, the biomass separated by centrifugation (3500 g, 15 min, 4 °C) and aliquots of the culture supernatant were stored at )80 °C until use. The supernatant was then centrifuged at 15 000 g and 4 °C for 30 min to remove particulate matter and filtered through Amicon Ultra columns (Millipore, Bedford, MA) with a cut-off value of 5000 Da. The concentrated protein solution ( 250 l L) was diluted in 50 mm citrate buffer, pH 5.5 containing 150 mm NaCl, concentrated again (final volume  500 lL) and loaded onto a HR 16 ⁄ 50 column packed with Superose 12 prep grade (GE Bioscience, Chalfont, UK) and equilibrated with the same buffer. Fractions were collected and assayed for b-galactosidase activity, and those that contained > 20% of the peak fraction activity were pooled and concentrated as above. For further purification, the concentrate was diluted in 10 mm citrate buffer pH 5.5 (buffer A) to a final volume of 10 mL, and loaded onto a Mono S HR 5 ⁄ 5 column (GE Bioscience) previously equili- brated with 10 column volumes of buffer A. The column was then washed with further 10 column volumes of buf- fer A, and thereafter the bound proteins were eluted by applying a linear gradient of 0–0.6 mm NaCl in a total of 40 mL of buffer A. Fractions were assayed for b-galactosi- dase activity and those containing > 20% of the activity in the peak fraction were pooled. b-Galactosidase activity assay Unless stated otherwise, b-galactosidase was assayed by measuring the hydrolysis of oNPG in 50 mm acetate buffer pH 5. The reaction was started by the addition of oNPG Table 3. Kinetic parameters of H. jecorina b-galactosidase with various galactans Substrate Bound % Gal K m (gÆL )1 ) V max (nkatÆmg )1 ) k cat (s )1 ) k cat ⁄ K m (LÆg )1 Æs )1 ) Arabinogalactan b-1,3 ⁄ b-1,6 79 13.59 ± 1.87 36.50 ± 4.88 4.34 ± 0.58 0.32 ± 0.06 Galactan (Lupin) b-1,4 91 38.34 ± 6.87 57.80 ± 10.32 6.87 ± 1.23 0.18 ± 0.05 Galactan (Potato) b-1,4 87 25.70 ± 4.82 27.93 ± 5.20 3.32 ± 0.62 0.13 ± 0.03 C. Gamauf et al. b-Galactosidase of Hypocrea jecorina FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS 1697 (final concentration in the basic assay was 3 mm) to give a total reaction volume of 1 mL. The assay was incubated at 30 °C for 30–60 min and stopped by the addition of 3 mL of 1 m Na 2 CO 3 . Absorbance was measured at 405 nm (e oNP ¼ 4530 LÆmol )1 Æcm )1 ) against a blank sample. Activ- ities are given in nanokatals, one nkat being equivalent to the release of 1 nmol o-nitrophenol per second under the conditions given above. Specific activities are related to 1 mg of protein, determined by the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). To determine the activity of b-galactosidase on other substrates, the following procedure was used: 1 lg purified b-galactosidase was incubated with appropriate amounts of potential substrates in 50 mm acetate buffer pH 5.0 in a total volume of 1 mL for 2–24 h at 30 °C. The reaction was stopped by boiling the incubation mixture at 95 °C for 10 min, and, after cooling on ice and centrifugation in an Eppendorf centrifuge (5 min), the amount of liberated d-galactose was determined by HPLC using an Aminex HPX-87H column (Bio-Rad) with 10 mm H 2 SO 4 as the mobile phase at a flow rate of 0.5 mLÆmin )1 (35 °C). The concentration of d-galactose in the sample was calculated from a calibration curve and used to determine the nmol of d-galactose per second formed in the assay. The activity of b-galactosidase with lactulose could not be measured in this way, because d-galactose and d-fruc- tose displayed similar retention times in our HPLC analy- sis. Therefore, we used d-galactose dehydrogenase (Sigma) to quantify the produced d-galactose. To this end, the pH of the incubation mixture (processed as above) was adjusted to pH 8.6 by addition of 1 m NaOH, then 10 lL10mm NAD + and 0.1 m phosphate buffer pH 8.6 up to 1 mL were added. The reaction was started by addition of 60 mU of d-galactose dehydrogenase (incubation: 1 h at 30 °C). The amount of NADH (e NADH ¼ 6300 LÆmol )1 Æcm )1 ) pro- duced was determined by measuring the absorbance at 340 nm against a blank (incubation mixture without sub- strate). In these assays one nkat was defined as 1 nmol d-galactose formed per second under the conditions given and again related to the protein concentration. To determine K m and V max , the activity of Bga1 with the given substrates was assayed at at least five different sub- strate concentrations in an appropriate range. Each meas- urement was performed in triplicate and the Enzyme Kinetics Module in sigma plot 2001 (Systat Software Inc., Point Richmond, CA) was used to calculate the K m and V max values. The errors represent the standard deviation from the measurement and regression. The molecular mass of Bga1 used for the calculation of k cat was 118 789 Da. Temperature and pH optimum and enzyme stability To determine the optimal temperature and pH for the assay described above, different temperatures (15–75 °C) and 0.1 m McIlvaine buffer (citric acid ⁄ Na 2 HPO 4 pH 3–8) were used. The stability was investigated by incubating the enzyme for 1 h at the given pH or temperature and then assaying the activity as described above. Throughout these experiments, the enzyme concentration was 1 lgÆmL )1 . Biochemical analytical methods Standard methods, as described previously [28] were used for SDS ⁄ PAGE, isoelectric focusing and Coomassie Brilli- ant Blue staining. Glycoprotein staining of SDS ⁄ PAGE gels was performed with the Pro-Q Emerald 300 Glycoprotein Gel and Blot Staining Kit (Molecular Probes, Eugene, OR). Capillary gel electrophoresis-on-the-chip Chip-based separation of proteins was performed using a prototype instrument (2100 bioanalyser) of Agilent Tech- nologies (Waldbronn, Germany), which has been described in detail elsewhere [29]. Briefly, this instrument uses lab-on- a-chip technology to separate high molecular mass proteins electrophoretically in a linear polymer solution and is cou- pled to a laser-induced fluorescence (LIF) detector using a fluorescence dye. Determination of the molecular mass The native molecular mass of the b-galactosidase was deter- mined by gel-permeation chromatography using a 16 ⁄ 70 column filled with Bio-Gel A-1.5 m (Bio-Rad), equilibrated with 0.1 m acetate buffer pH 5 containing 0.5 m NaCl at a flow rate of 0.5 mLÆmin )1 . Bio-Rad Gel Filtration Standard proteins were used to calibrate the column. Mass spectrometric characterization Sample preparation for MALDI-TOF-MS of the native protein was carried out on a stainless steel target, applying the dried droplet preparation technique [30] using 2,4,6- trihydroxyacetophenone (20 mgÆmL )1 in methanol) as the matrix. Positive-ion mass spectra were recorded on a vacuum MALDI-TOF ⁄ curved field reflector TOF instru- ment (TOF 2 , Shimadzu Biotech, Manchester, UK) equipped with a nitrogen laser (k ¼ 337 nm) in the linear mode by accumulating 200–500 single unselected laser shots. External calibration was performed using an aqueous solution of b-galactosidase from Escherichia coli and the mass spectra were treated with the company-supplied smoothing algorithm. For in-gel digestion the respective protein band of an appropriate SDS ⁄ PAGE was excised manually with a stain- less steel scalpel and digested with trypsin (bovine pancreas, modified; sequencing grade, Roche, Mannheim, Germany) b-Galactosidase of Hypocrea jecorina C. Gamauf et al. 1698 FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS [31]. Extracted tryptic peptides were desalted and step-wise purified utilizing ZipTip Ò technology [32] (C 18 reversed phase, standard bed, Millipore, Bedford, MA) by loading the extracted peptide mixture onto C 18 -ZipTips which were activated by ACN ⁄ ultra pure water (1 : 1, v ⁄ v) and further equilibrated with water. After binding of the sample the tips were washed two times with 10 lL water for salt removal. Step-wise elution was performed consecutively using solutions consisting of 2, 10, 50 and 75% (v ⁄ v) ACN in water. The different fractions (3 lL each) were analysed by MALDI-curved field reflectron (RTOF)-MS. Sample preparation was again carried out on a stainless steel target, applying the thin-layer preparation technique with sinapic acid (6 mgÆmL )1 in water) as well as the dried-droplet tech- nique for 2,5-dihydroxybenzoic acid (10 mgÆmL )1 in water). Positive-ion mass spectra were recorded on the same instru- ment as mentioned above. External calibration was per- formed using an aqueous solution of standard peptides (Bradykinin fragment 1–7, human angiotensin II, somato- statin and ACTH fragment 18–39). Acknowledgements We thank Verena Seidl for her help with the isoelectric focusing analysis, Roland Mu ¨ ller for his help with CGE-on-the-chip experiment and Agilent Technologies for the loan of the instrument. Sanna Hiljanen-Berg and Sirpa Okko are thanked for skilful technical assist- ance. Work in the laboratory of CG, CPK and BS was supported by the Austrian Science Foundation FWF (P16143). References 1 Coutinho PM & Henrissat B (1999) Carbohydrate- active enzymes: an integrated database approach. In Recent Advances in Carbohydrate Bioengineering (Gilbert HJ, Davies G, Henrissat B & Svensson B, eds), pp. 3–12. Royal Society of Chemistry, Cambridge. 2 Holsinger VH & Kligerman AE (1991) Application of lactase in dairy foods and other foods containing lac- tose. Food Technol 45, 92–95. 3 Gekas V & Lo ´ pez-Leiva M (1985) Hydrolysis of lactose: a literature review. Process Biochem 20, 2–12. 4 Rings EH, Grand RJ & Buller HA (1994) Lactose intol- erance and lactase deficiency in children. Curr Opin Pediatr 6, 562–567. 5 Nakayama T & Amachi T (1999) b-Galactosidase, enzy- mology. In Encyclopedia of Bioprocess Technology: Fer- mentation, Biocatalysis and Bioseparation (Flickinger MC & Drew SW, eds), pp. 1291–1305. Wiley, New York, NY. 6 Boon MA, Janssen AE & van’t Riet K (2000) Effect of temperature and enzyme origin on the enzymatic syn- thesis of oligosaccharides. Enzyme Microb Technol 26, 271–281. 7 Penttila ¨ M, Limon C & Nevalainen H (2004) Molecular biology of Trichoderma and biotechnological applica- tions. In Handbook of Fungal Biotechnology (Arora DK, ed.), pp. 413–427. Marcel Dekker, New York, NY. 8 Seiboth B, Hartl L, Salovuori N, Lanthaler K, Robson GD, Vehmaanpera ¨ J, Penttila ¨ ME & Kubicek CP (2005) Role of the bga1-encoded extracellular b-galac- tosidase of Hypocrea jecorina in cellulase induction by lactose. Appl Environ Microbiol 71, 851–857. 9 Harman GE, Howell CR, Viterbo A, Chet I & Lorito M (2004) Trichoderma species – opportunistic, avirulent plant symbionts. Nat Rev Microbiol 2, 43–56. 10 Salovuori I, Makarow M, Rauvala H, Knowles J & Ka ¨ a ¨ ria ¨ inen L (1987) Low molecular weight high-mannose type glycans in a secreted protein of the filamentous fun- gus Trichoderma reesei. Bio ⁄ Technology 5, 152–156. 11 Rojas AL, Nagem RA, Neustroev KN, Arand M, Adamska M, Eneyskaya EV, Kulminskaya AA, Garratt RC, Golubev AM & Polikarpov I (2004) Crystal struc- tures of b-galactosidase from Penicillium sp. and its complex with galactose. J Mol Biol 343, 1281–1292. 12 Hung MN & Lee BH (2002) Purification and characteri- zation of a recombinant b-galactosidase with transgalac- tosylation activity from Bifidobacterium infantis HL96. Appl Microbiol Biotechnol 58, 439–445. 13 Mayer J, Conrad J, Klaiber I, Lutz-Wahl S, Beifuss U & Fischer L (2004) Enzymatic production and complete nuclear magnetic resonance assignment of the sugar lac- tulose. J Agric Food Chem 52, 6983–6990. 14 Willfo ¨ r S, Sjo ¨ holm R, Laine C & Holmbom B (2002) Structural features of water-soluble arabinogalactans from Norway spruce and Scots pine heartwood. Wood Sci Technol 36, 101–110. 15 Voragen AGJ, Pilnik W, Thibault JF & Renard CMGC (1995) Pectins. Food Polysaccharides and Their Applica- tions (Stephen AM, ed.), pp. 287–339. Marcel Dekker, New York, NY. 16 Crueger A & Crueger W (1984) Carbohydrates. Biotech- nology (Kieslich K, ed.), pp. 421–457. Verlag Chemie, Weinheim. 17 Sinnott ML (1990) Catalytic mechanisms of enzymic glycosyl transfer. Chem Rev 90, 1171–1202. 18 Aehle W (2004) Enzymes in Industry. Wiley-VCH, Wein- heim. 19 Jurado E, Camacho F, Luzo ´ n G & Vicaria JM (2002) A new kinetic model proposed for enzymatic hydrolysis of lactose by a b-galactosidase from Kluyveromyces fra- gilis. Enzyme Microb Technol 31, 300–309. 20 Onishi N & Tanaka T (1995) Purification and properties of a novel thermostable galacto-oligosaccharide-produ- cing b-galactosidase from Sterigmatomyces elviae CBS8119. Appl Environ Microbiol 61, 4026–4030. C. Gamauf et al. b-Galactosidase of Hypocrea jecorina FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS 1699 21 Yanahira S, Suguri T, Yakabe T, Ikeuchi Y, Hanagata G & Deya E (1992) Formation of oligosaccharides from lactitol by Aspergillus oryzae b-d-galactosidase. Carbo- hydr Res 232, 151–159. 22 Gopal PK, Prasad J & Gill HS (2003) Effects of the consumption of Bifidobacterium lactis HN019 (DR10TM) and galacto-oligosaccharides on the micro- flora of the gastrointestinal tract in human subjects. Nutr Res 23, 1313–1328. 23 Charalampopoulos D, Wang R, Pandiella SS & Webb C (2002) Application of cereals and cereal components in functional foods: a review. Int J Food Microbiol 79, 131–141. 24 Benitez T, Rincon AM, Limon MC & Codon AC (2004) Biocontrol mechanisms of Trichoderma strains. Int Microbiol 7, 249–260. 25 Knee EM, Gong FC, Gao M, Teplitski M, Jones AR, Foxworthy A, Mort AJ & Bauer WD (2001) Root mucilage from pea and its utilization by rhizosphere bacteria as a sole carbon source. Mol Plant Microbe Interact 14, 775–784. 26 Seiboth B, Hartl L, Pail M, Fekete E, Karaffa L & Kubicek CP (2004) The galactokinase of Hypocrea jecorina is essential for cellulase induction by lactose but dispensable for growth on d-galactose. Mol Microbiol 51, 1015–1025. 27 Seiboth B, Hofmann G & Kubicek CP (2002) Lactose metabolism and cellulase production in Hypocrea jecor- ina: the gal7 gene, encoding galactose-1-phosphate uri- dylyltransferase, is essential for growth on galactose but not for cellulase induction. Mol Genet Genom 267, 124– 132. 28 Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA & Stuhl K (2006) Current Protocols in Molecular Biology. Greene ⁄ Wiley Inter- science, New York, NY. 29 Mueller O, Hahnenberger K, Dittmann M, Yee H, Dubrow R, Nagle R & Ilsley D (2000) A microfluidic system for high-speed reproducible DNA sizing and quantitation. Electrophoresis 21, 128–134. 30 Kussmann M, Nordhoff E, Rahbek-Nielsen H, Haebel S, Rossel-Larsen M, Jakobsen L, Gobom J, Mirgorods- kaya E, Kroll-Kristensen A, Palm L et al. (1997) Mat- rix-assisted laser desorption ⁄ ionization mass spectrometry sample preparation techniques designed for various peptide and protein analytes. J Mass Spec- trom 32, 593–601. 31 Shevchenko A, Wilm M, Vorm O & Mann M (1996) Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem 68, 850–858. 32 Pluskal MG (2000) Microscale sample preparation. Nat Biotechnol 18, 104–105. b-Galactosidase of Hypocrea jecorina C. Gamauf et al. 1700 FEBS Journal 274 (2007) 1691–1700 ª 2007 The Authors Journal compilation ª 2007 FEBS . Characterization of the bga1-encoded glycoside hydrolase family 35 b-galactosidase of Hypocrea jecorina with galacto-b- D-galactanase activity Christian. com- parison of the obtained results difficult. To the best of our knowledge, this is the first report of the enzymo- logical properties of a b-galactosidase