G Model

ARTICLE IN PRESS

BIOTEC 7090 1–10

Journal of Biotechnology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec

Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases

1

2

Q1

3 4

Yuanyuan Ma a,b,c , Xuewei Liu a,d , Yanchen Yin a,d , Chao Zou a,d , Wanchao Wang a,d , Shaolan Zou a,b,c , Jiefang Hong a , Minhua Zhang a,b,c,∗ a

R&D Center for Petrochemical Technology, Tianjin University, Tianjin 300072, China Collaborative Innovation Center of Chemical Science and Engineering, Tianjin 300072, China c Key Laboratory for Green Chemical Technology of Ministry of Education, R&D Center for Petrochemical Technology, Tianjin University, Tianjin 300072, China d Department of Biochemical Engineering, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China

5

b

6 7 8 9 10

11 26

a r t i c l e

i n f o

a b s t r a c t

12 13 14 15 16 17

Article history: Received 29 October 2014 Received in revised form 26 March 2015 Accepted 18 April 2015 Available online xxx

18

25

Keywords: ␤-Glucosidase Saccharomycopsis fibuligera Trichoderma reesei Pichia pastoris Response surface method Biochemical property

27

1. Introduction

19 20 21 22 23 24

The ␤-glucosidases from Saccharomycopsis fibuligera (SfBGL1) and Trichoderma reesei (TrBGL1) were cloned and expressed in Pichia pastoris. Methanol concentration and pH significantly affected the production. The combined effects of the two factors were optimized by using the response surface method, resulting in a 137% and 84% increase in rTrBGL1 and rSfBGL1 yield compared to single-factor experiment. Structure and biochemical properties of the two enzyme were investigated and compared. They belong to glycosyl hydrolase family 3 and exhibit significant hydrolysis activity and low-level transglycosylation activity. The two enzymes show similar substrate affinity and ion-tolerance, and both of them can be activated by Cr6+ , Mn2+ and Fe2+ . The rSfBGL1 has greater catalytic speed, higher specific activity and acid-tolerance than rTrBGL1, but rTrBGL1 is more thermostable and has higher optimal temperature than rSfBGL1. This study provides a useful and quick optimal method for recombinant enzyme production and makes a valuable comparison of biochemical properties, which opens important avenues of exploration for relationship between structure and function and further practical applications. © 2015 Published by Elsevier B.V.

Q2 28 29 30 31 32 33 34 35 36 37 38 39

Cellulose is the most abundant organic polymer on Earth, which can be hydrolyzed into glucose for further utilization, but the high cost of cellulase presents a significant barrier to commercialization (Merino and Cherry, 2007). ␤-Glucosidase is a component of cellulase system. As a rate-limiting regulator, it plays an important role in degradation of cello-oligosaccharides, which influences the liberation of glucose from cellulose (Barnett et al., 1991; Chen et al., 2008). Nevertheless, they are poorly produced by native celluaseproducing strains. Besides cellulose degradation, ␤-glucosidases have been widely used in agricultural, medical and diversified industrial fields (Bhatia et al., 2002; Husain, 2010). ␤-Glucosidases belong to a large family of glycoside hydrolase (GH), and many

∗ Corresponding author at: R&D Center for Petrochemical Technology, Tianjin 300072, China. Tel.: +86 22 87401536; fax: +86 22 27406119. E-mail addresses: [email protected] (Y. Ma), lxw [email protected] (X. Liu), [email protected] (Y. Yin), [email protected] (C. Zou), [email protected] (W. Wang), [email protected] (S. Zou), [email protected] (J. Hong), [email protected] (M. Zhang).

of them have dual activities of hydrolysis and transglycosylation, which catalyze synthesis of poly- or oligo-saccharide (Bohlin et al., 2013; Ketudat Cairns and Esen, 2010). Hence, it is very necessary to investigate the specific functional property of several key members of this family for rational utilization. Trichoderma reesei has been employed in commercial cellulase preparations. It can secret high amounts of cellulolytic enzymes, but the ␤-glucosidase yield is low compared with that from other filamentous fungi (Barnett et al., 1991; Zhang et al., 2010). The ␤glucosidases I (TrBGL1) encoded by bgl1 gene is an extracellular, cellulose-inducible ␤-glucosidase (Mach et al., 1995). The enzyme has been tried to be expressed in filamentous fungi and Saccharomyces cerevisiae to improve expression levels (Cummings and Fowler, 1996; Zhang et al., 2010). The ␤-glucosidase gene bgl1 from Saccharomycopsis fibuligera (Sfbgl1) has drawn increasing attention as a potential donor of the ␤-glucosidase gene for the improvement of production (Shen et al., 2008; van Rooyen et al., 2005; Zhang et al., 2012), and the expression level of Sfbgl1 gene in S. cerevisiae is higher than that of other fungal ␤-glucosidase genes (van Rooyen et al., 2005). Although Sfbgl1 and Trbgl1 have been successfully heterologously expressed, the expression levels still leave room for

http://dx.doi.org/10.1016/j.jbiotec.2015.04.016 0168-1656/© 2015 Published by Elsevier B.V.

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

2

86

improvement. Various ␤-glucosidases have been tried to be used as supplement in cellulase mixture to enhance the saccharification of cellulose, however the low productivity and unknown biochemical property of TrBGL1or SfBGL1 limit the application for the purpose. Therefore, productivity improvement is a consideration of the most importance. An appropriate expression system and an effective optimal method are necessary for efficient protein production. Pichia pastoris possesses many advantages such as posttranslational modification, easy manipulation, and higher protein expression levels. Therefore, it is worth to try to express bioactive ␤-glucosidases in P. pastoris host and optimize the conditions for high productivity. Presently, the biochemical properties of TrBGL1 and SfBGL1 are little reported, and the potential differences of structure and biochemical properties between the two ␤-glucosidases are not elucidated. Post-translational modifications by host usually make their biochemical characteristics different from the native enzyme. However, functional properties are important for their usage as gene resources, enzyme modification and industrial application. Consequently, in the present study, the Trbgl1 (GenBank accession no. TRU09580) and Sfbgl1 (HQ891006) genes were cloned and over-expressed in P. pastoris, and the production conditions were quickly optimized by the response surface methods (RSM). The structural models and biochemical properties of purified proteins were revealed and compared for the first time.

87

2. Materials and methods

62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

88 89

90 91 92 93 94 95 96

97 98

2.1. Bioinformatics analysis and construction of expression vectors Bioinformatics analysis of TrBGL1 and SfBGL1 was carried out as described in the Supplementary Material. TrBGL1 and SfBGL1 genes were cloned into pPICZ␣C and pPICZ␣A vectors, resulting in pPICZ␣C-TrBGL1 and pPICZ␣A-SfBGL1 plasmids, respectively (Fig. S1). TrBGL1 and SfBGL1 genes were then integrated into P. pastoris Gs115 genomic DNA by electroporation, and positive transformants were verified by PCR (Supplementary Material). 2.2. Selection of high-level expression colonies and expression optimization

113

In order to determine the productivity of the selected colonies, induction experiments with pH 6.0 of buffered glycerol/methanolcomplex (BMGY/BMMY) media were carried out as shown by Supplementary Material. The effect of initial pH 4–7 of media on the production of target proteins was also investigated. BMMY containing 0.5% and 1.0% methanol were used for the production of rSfBGL1 (recombinant SfBGL1) and rTrBGL1 (recombinant TrBGL1), respectively. The pH 5.0 of BMGY/BMMY was used to investigate the effect of methanol concentration on production, and methanol was added to a final concentration of 0.25%, 0.5%, 1%, 1.5% and 2%, respectively. The culture was taken daily to determine cell growth and expression levels of rBGLs (recombinant ␤-glucosidases). The culture supernatants were collected for SDS-PAGE analysis and activity assay of ␤-glucosidase. The activity was expressed as units per milliliter culture media.

114

2.3. Enzyme assay

99 100 101 102 103 104 105 106 107 108 109 110 111 112

115 116 117 118 119 120

The ␤-glucosidase activity was determined by measuring the hydrolysis of p-nitrophenyl-beta-d-glucopyranoside (pNPG, Sigma). The activity assay was conducted in a reaction mixture (1 ml) consisting of 100 ␮L of an appropriately diluted supernatant sample or purified rBGLs, 50 mM sodium acetate buffer (pH 5.0) and 5 mM pNPG. The reaction was conducted at 40 ◦ C for 10 min,

and then terminated by adding 1 ml of 1 M Na2 CO3 . After 5 min, the absorbance of the resulting mixture was measured at 405 nm, and the concentration of released pNP (p-nitrophenol) was calculated. One unit (U) of enzyme activity was defined as the amount of enzyme that catalyzed the formation of 1 ␮mol PNP per minute under the conditions of the assay. Relative expression levels of recombinant enzyme in culture supernatants were evaluated by the units of ␤-glucosidase activity per milliliter of culture (u/ml). All enzyme assays were performed in triplicate. 2.4. Optimization of ˇ-glucosidase production by response surface method

123 124 125 126 127 128 129

131

(1)

where Y is the predicted response, ˇ1 and ˇ2 are linear coefficients, Q3 ˇ4 and ˇ5 are quadratic coefficients, ˇ3 is interaction coefficients and ˇ0 is the constant coefficient. Purification and deglycosylation of ␤-glucosidases The rBGLs with C-terminal hexahistidine (His 6) tag were purified using Ni-NTA His·Bind resins system (Merk, 70666) according to the manufacturer’s instructions with slight modification as previously described (Wang et al., 2014; Liu et al., 2014). Purified proteins were exchanged by ultrafiltration into 50 mM NaAc buffer (pH 4.8) and then quantified by Bradford’s assay for functional investigation. N-linked and O-linked glycosylation sites were predicted, and the N-linked glycosylation of recombinant proteins was confirmed by digestion of purified protein with PNGase F (NEB, P0704S) as described in Supplementary Material. Deglycosylation of protein was assessed by a shift in electrophoretic mobility on a 10% SDS-PAGE gel. 2.5. Measurement of kinetic parameters The Michaelis–Menten constant (Km ) and the maximal reaction velocities (Vmax ) were determined at 40 ◦ C using pNPG (0.04–0.18 mM) as substrate by Lineweaver–Burk plot. The reaction was performed as described in Section 2.3. The turnover number (kcat ) is the number of moles of substrate converted to product each second per mole of enzyme, and described as following formula: kcat =

122

130

Response surface method was used to investigate the combined effect of pH and temperature on the production. A two-factors, fivelevels central composite design (CCD) was used. The cultures were incubated at 28 ◦ C with shaking at 250 rpm for 12 days. The culture supernatants from day 10 were used to measure the activity of rBGLs. Enzyme activity value per ml of supernatant liquid (u/ml) was used as the response value. Design ExpertTM version 7.0.3 (StatEase, Inc., Minneapolis, USA) was used for experimental design and data analysis (Dong et al., 2012). The experimental results are usually fitted via the response surface regression procedure based on the following second-order polynomial equation: Y = ˇ0 + ˇ1 x1 + ˇ2 x2 + ˇ3 x1 x2 + ˇ4 x12 + ˇ5 x22

121

Vmax Etotal

The specific activity, a quantity that is used to monitor enzyme purification, is defined as units per microgram of protein. Effect of metal ions and chaotropic agents on activity The residual activity in presence of various metal ions and reagents was determined after pre-incubating purified rBGLs in 50 mM sodium acetate buffer (pH 5.0) containing different concentration of salts and detergents at 20 ◦ C for 1 h. A reaction mixture with pNPG but without ␤-glucosidase was used as a negative control. The absorbance value of control with each chemical was subtracted from that of reaction containing that chemical, resulting in a final activity value. The relative activity was defined as the

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

132 133 134 135 136 137 138 139 140 141 142

143

144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159

160

161 162 163 164 165 166

167

168 169 170 171 172 173 174 175 176 177 178

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

3

Fig. 1. Comparison of the overall structures of SfBGL1 and TrBGL1. Both SfBGL1 and TrBGL1 have 3 domains (D1, D2 and D3). The alpha helices, ␤-strands and coils are shown in magenta, blue and gray, respectively. The predicted highly conserved active sites are indicated by black arrows.

180

ratio of ␤-glucosidase activity with added reagents to that without them.

181

3. Results

179

182 183

184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213

214 215

216 217 218

3.1. Comparative sequence and structure analysis of TrBGL1 and SfBGL1 Bioinformation analysis shows that TrBGL1 and SfBGL1 share 42% identity in their amino acid sequences, and both belong to glycosyl hydrolase family 3 (GHF3). The sequence conservation of GHF3 enzymes is relatively low, and only a few crystal structures of GHF3 enzymes have been reported (Pozzo et al., 2010; Yoshida et al., 2010). This study thus simulated their three-dimensioned structure, and the spatial conformation of three specific domains (D1: the GHF3 N-terminal domain, D2: C-terminal domain and D3: fibronectin type III domain of unknown function) and two linkers were for the first time presented (Fig. 1 and Fig. S2). Their D1 both formed a triose phosphate isomerase (TIM) barrel and contained conserved SDW and KHF motif. The Asp of SDW motif might function as the catalytic nucleophile residue, and His in KHF motif had also been implicated as the potential H+ donor. Several highly conserved amino acids such as D, R, K, H, R, M, Y, D, and W in this domain were putative active sites. D2 formed an ␣/␤ sandwich and had a putative active site, the acid/base amino acid residue E (E452 for SfBGL1, and E457 for TrBGL1). These active sites in D1 and D2 can form hydrogen bonds and Vander Waals (vdW) with carbohydrate substrates, and the interaction between ␤-glucosidase and substrates is thus mediated. Obvious structure difference was observed in D3. The D3 of SfBGL1 had a visible redundant structure compared with that of TrBGL1, and the redundant sequence (aa686-753) was also predicted to form two alpha helices (Fig. S2). No putative active site was shown in D3, but their D3 both can form FnIII fold that is only presented in the structure of TnBgl3B in publicly available GHF3 structures. The location of D3 is distant from the catalytic site implying that it can have no direct effect on the hydrolysis of smaller substrates (Pozzo et al., 2010). 3.2. Optimization of ˇ-glucosidase production in P. pastoris by single-factor experiments Random selected Mut+ transformants of P. pastoris for each construct were used to examine the expression levels of target proteins. The ␤-glucosidase activity was found in culture supernatant of all

recombinant strains but not in that of control strain Gs115, and strain TrBGL1-3 and SfBGL1-8 showed the highest yield among transformants containing Trbgl1 and Sfbgl1 genes, respectively (Fig. S3). Consequently, they were selected to optimize ␤-glucosidase production. Three different media formulae were tried, and BMMY gave the highest yield. The activity of recombinant enzyme was hardly detected in MMH and BMMH media, and the growth of strains cultured in the two media was slow compared with that in BMMY (Fig. S4). When strains were cultured in BMMY, the effect of culture time on production was investigated. The rBGL yield showed a dramatic increase from day 0 to10, and decreased from day 10 to 12 post-induction. The cell number of both strains increased from day 0 to day 10, and slightly decreased from day 10 to 12. An increase in cell number during the induction may help to improve production. Interestingly, they were cultured under the same conditions, but the expression levels of rSfBGL1 were much higher than that of rTrBGL1 (Fig. S5). Concentration of methanol in media also affected the protein yield. The optimum methanol concentration for rTrBGL1 production was 1.0%, followed by 1.5% and 0.5%, and rTrBGL1 was hardly secreted into the media with 2% methanol (Fig. 2a and e). The optimal growth of strain TrBGL1-3 was also observed when methanol concentration was 1.5% and 1%, and cell growth was slightly inhibited by 2% methanol (Fig. 2c). The optimum methanol concentration for rSfBGL1 production was about 0.5%, followed by 0.25% and 1%, and methanol concentration above 1.5% caused the inhibition of rSfBGL1 expression (Fig. 2b and e). Correspondingly, the growth of strain SfBGL1-8 showed similar trend with the enzyme production (Fig. 2d). The results show that excess methanol restrains cell proliferation resulting in a low protein yield, and the optimal methanol concentration is 0.5%/1.0% for growth and the production of rSfBGL1/rTrBGL1. Effect of medium pH on rBGL secretion was detected. The relative optimal protein expression levels of SfBGL1-8 were obtained in the range of pH 4.5–6.5, and the ␤-glucosidase activity in culture supernatants was the highest at pH 4.5, and then decreased with an increase of pH (Fig. 3a and e). Interestingly, a slight lower molecular weight was observed in rSfBGL1 from pH 5.5 culture than that from other pH value (Fig. 3e). Different pH culture condition can affect the glycosylation pattern of foreign protein expressed in yeast cells (Nakamura et al., 2006), possibly resulting in a difference in molecular weight among SfBGL1. The highest expression level of rTrBGL1 was obtained at pH 5.5, and the production level at pH 4.5 and 5.0 were slightly inferior to that at pH 5.5, whereas production was

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263

G Model BIOTEC 7090 1–10 4

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

Fig. 2. The ␤-glucosidase production and growth curve of recombinant strains grown on media containing different methanol concentrations. Panel (a) and (b) indicate the enzyme activity of secreted rTrBGL1 and rSfBGL1, respectively; panel (c) and (d) show the growth tendency of strain TrBGL1-3 and SfBGL1-8, respectively; SDS-PAGE analysis of rBGLs in culture supernatants (panel e).

264 265 266 267 268 269 270 271 272 273 274 275 276 277 278

inhibited at pH 6–7 (Fig. 3b and e). Optimal expression was obtained at a relative acidic environment, which was also more propitious to cell growth (Fig. 3c and d) implying that biomass accumulation contributes to enzyme production improvement. Accordingly, both the increased cell densities and the inhibited protease activity may result in a high-level secretion of exogenous proteins at acid conditions. Except for above factors, many factors influence foreign gene expression in P. pastoris, such as copy number of target genes, types of host strains, culture conditions, medium composition, cell densities and addition of protease inhibitors (eg, casamino acids, PMSF or commercial protease inhibitor cock tails) (Zhu et al., 2011; Liu et al., 2014; Wang et al., 2014). The current study revealed that methanol concentration and pH could affect production (Figs. 2 and 3), and the two factors were thus further optimized by RSM.

3.3. Optimal productions of rBGLs by RSM

279

We first tried to optimize production of rBGLs in P. pastoris by RSM. The empirical relationship between the response and the variables was expressed by the following fitting second-order polynomial equation (Eq. (2)/Eq. (3) for SfBGL1/TrBGL1 production): Y = 2.55 − 0.34x1 − 0.4x2 − 0.23x1 x2 − 0.37x12 + 0.022x22 Y=

1.46 + 0.27x1 − 0.098x2 − 0.088x1 x2 − 0.13x12

− 0.035x22

280 281 282 283

(2)

284

(3)

285

A three-dimensional response surface (Fig. 4) was generated to show the interrelationships between pH and methanol concentration and determine the optimal conditions for beta-glucosidase production. The sufficiency of the second-order model was checked by the analysis of variance (ANOVA) presented in Table S1. The

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

286 287 288 289 290

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

5

Fig. 3. The ␤-glucosidase production and growth curve of recombinant strains grown at different pH media. Panel (a) and (b) depict the enzyme activity of rSfBGL1 and rTrBGL1 secreted by recombinant strains, respectively; the growth tendency of strain TrBGL1-3 and SfBGL1-8 grown on media with various pH values are shown at panel (c) and (d), respectively; panel (e) displays SDS-PAGE analysis of recombinant enzyme in culture supernatants. 291 292 293 294 295 296 297 298 299 300 301 302 303 304

F-value of 14.65/13.87 (SfBGL1/TrBGL1) implies that the model was significant, as there was only a 0.26%/0.3% (SfBGL1/TrBGL1) probability that a large F-value could occur due to noise. The model coefficient of determination R2 was 0.9243/0.9204 (SfBGL1/TrBGL1), indicating that the fitting degree of equation was good, and there was a high degree of correlation between the predicted value and measured value. The P-value below 0.05 indicates that both model terms were significant. These statistical tests indicated that the two models were adequate for predicting the rBGL yield, and the non-significant value of lack of fit “Prob > F” more than 0.05 also showed that the quadratic model was valid for the present study. The two models predicted that the maximum protein yield was 1.90/3.43 U/ml for TrBGL1 (pH 4.5 and 1.03% methanol)/SfBGL1

(pH 2.8 and 0.734% methanol). The mean value of three verification experiments showed that final 1.97/3.20 U of activity per milliliter was present in the supernatant liquids of TrBGL13/SfBGL1-8 culture. The good agreement between the predicted and the experimental results verified the validity of the models. 3.4. Purification, deglycosylation and kinetic parameters of rBGLs The purified rTrBGL1/rTrSfBGL1 migrated in SDS-PAGE gel more slowly than expected and gave bands with an approximate molecular mass of 90/120 kDa (Fig. 5), which was higher than the predicted molecular weights of 79.6/98.5 kDa. It may due to glycosylation. The rTrBGL1/rSfBGL1 had 5/14 putative N/O-glycosylation sites, and the molecular weights of rBGLs were close to the predicted

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

305 306 307 308 309

310

311 312 313 314 315 316

G Model BIOTEC 7090 1–10 6

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

Fig. 4. Response surface plot for rBGL production showing the interactive effects of methanol concentration and pH.

328

values after deglycosylation (Fig. 5). Previous reports have already disclosed that slower migration of recombinant proteins produced by P. pastoris is due to glycosylation (Potvin et al., 2012; Wang et al., 2014). Purified enzymes were used for examination of biochemical characterization. The Km value of rTrBGL1/rSfBGL1 toward the pNPG was 0.209/0.212 mM, and maximum velocity (Vmax ) was 1.4 × 10−5 /3.31 × 10−5 mol/l min−1 . The turnover number (Kcat) of TrBGL1/SfBGL1 was 7.28/10.65 s−1 . These results indicated that they shared similar substrate affinity, but rSfBGL1 had greater catalytic speed and higher specific activity (6.34 ␮mol/␮g min) than rTrBGL1(4.7 ␮mol/␮g min, Table S2).

329

3.5. Optimal temperature and pH, and thermal stability

317 318 319 320 321 322 323 324 325 326 327

330 331 332 333 334 335 336

The optimum temperature for rTrBGL1/rSfBGL1 activity was 70/50 ◦ C (Fig. 6a), and the rTrBGL1 showed higher relative activity at high temperature (60–90 ◦ C) than rSfBGL1. The rTrBGL1 and rSfBGL1 exhibited an optimal activity at pH 5.0 and pH 4.0 (Fig. 6b), respectively. The acid conditions were propitious to catalysis, and their relative activity still remained approximately 70% at pH 2.0, indicating an acid-tolerant property.

The two enzymes were quite stable at 30 and 40 ◦ C. The rTrBGL1 also showed good stability at 50 ◦ C and retained 92.7% of initial activity after 1 h incubation at 50 ◦ C (Fig. 6c). The rTrBGL1 kept more original activity than rSfBGL1 at 50–80 ◦ C, especially at 60 and 70 ◦ C, showing a better thermal stability (Fig. 6c). 3.6. Susceptibility toward chemical reagents

Table 1 The effect of metal ion and chaotropic agents on rBGLs. Statistical differences between the treatments and the control were evaluated by ANOVA, and nonsignificant data (P > 0.001) were not shown (samples treated by KCl and Tris) or indicated by hyphen. Final concentration

Relative activity % SfBGL1

Control Metal ion NaCl AgNO3 K2 Cr2 O7 FeSO4 MnCl2

Fig. 5. SDS-PAGE analysis of deglycosylated ␤-glucosidases. Lanes P and D indicate that affinity-purified and deglycosylated rBGLs, respectively.

1 mM 1 mM 1 mM 1 mM 1 mM 5 mM 1 mM 1 mM 1 mM 1 mM 1 mM 1 mM 1 mM 1 mM

CoCl2 CaCl2 MgSO4 NiSO4 BaCl2 (NH4 )2 SO4 CuSO4 ZnSO4 Chaotropic agents 10 mM EDTA 0.4 M Urea 2% (v/v) Ethanol 4% Ethanol

338 339 340 341

342

The catalytic activity of enzymes can vary widely depending on the mode of interaction of metal ions with enzymes (Medyantseva et al., 1998). The study of effects of metal ions on enzyme activity thus has both theoretical and practical significance. One mM of FeSO4 , K2 Cr2 O7 and MnSO4 significantly enhanced their activity above 50%, which implies that they may be involved in enzyme function. The Cr2 O7 2− was the most effective activator resulting in a 321/241% increase of rSfBGL1/rTrBGL1 activity (Table 1). AgNO3 manifested the strongest inhibitory activity leading to almost

Effectors

337

TrBGL1

100

100

91.4 ± 0.8 27.0 ± 0.6 321.3 ± 0.9 151.4 ± 0.9 218.8 ± 0.44 285.0 ± 15.6 78.3 ± 0.9 81.1 ± 1.0 75.6 ± 1.5 73.8 ± 0.2 71.4 ± 2.2 75.7 ± 0.4 69.6 ± 0.6 65.2 ± 0.4

– 78.9 ± 0.9 241.3 ± 0.6 167.4 ± 1.7 200.7 ± 1.9 128.9 ± 3.4 – – – – – – 89.6 ± 0.3 –

72.0 ± 0.2 63.2 ± 0.1 – 110.5 ± 0.3

84.3 ± 0.5 91.5 ± 1.5 117.3 ± 0.7 131.5 ± 0.5

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

343 344 345 346 347 348 349 350 351

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

7

Fig. 6. Biochemical properties of ␤-glucosidases. The optimal temperature (panel a), pH (panel b) and thermal stability (panel c) of rTrBGL1 and rSfBGL1 are shown, and error bars indicate standard deviations. The pNPG was used as substrate to determine the optimal temperature, pH and thermal stability, and the reactions were carried out as previously described. The pH 4–8 of 50 mM sodium acetate buffer was used to study the effect of pH on activity. The effect of temperature on ␤-glucosidase activity was assayed at temperature ranging from 30 to 90 ◦ C. Thermal stabilities of rBGLs were determined by assaying for residual enzyme activity after incubation at various temperatures (30–80 ◦ C) for 0–180 min without substrate. Panel (d) shows the effect of concentration of metal ions on ␤-glucosidase activity. The ion concentrations range from 0.01 mM to 5 mM, and Y-axis illustrates the relative activity value in presence of Mg2+ (left), Ca2+ (left), Co2+ (left), and Mn2+ (right). 352 353 354 355 356 357 358 359 360

83–84% activity loss. Except for the above three activators and AgNO3 , other reagents inhibited rSfBGL1 activity by 6–35%, but they slightly inhibited or enhanced that of rTrBGL1 (89–105% of original activities), indicating a much greater tolerance of rTrBGL1 for chemical reagents treatment (Table 1). The ␤-glucosidase TtBGL was almost completely inactivated by Zn2+ , Cu2+ , Ag+ , or Hg2+ (Pei et al., 2012), whereas rSfBGL1/rTrBGL1 nearly kept 70–90% of activity in presence of 1 mM Zn2+ or Cu2+ , indicating their robustness to the reagents.

The inhibitory or activating effect of Ca2+ , Mg2+ , Co2+ and Mn2+ correlates linearly with the concentration of the effectors (Medyantseva et al., 1998Medyantseva 1998). An approximate Q4 linear relationship between the degree of inhibition of rSfBGL1 activity and concentrations of ions was observed with addition of 0.1–1 mM Mg2+ and Ca2+ ions. A linear dependence of an increase in rTrBGL1 activity on the concentration of Co2+ was also observed in presence of 0.1–5 mM Co2+ , and no linear relationship was found in the presence of Mg2+ , Ca2+ and Mn2+ . Nevertheless, Mn2+ led to

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

361 362 363 364 365 366 367 368 369

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

8

Fig. 7. Transglycosylation activity. Five micromole of purified enzyme and 75 mg of anhydrous glucose were incubated in 200 ␮L acetate buffer (10 mM, pH 5.0) at 40 ◦ C. Aliquots taken after 0–24 h were loaded on 3K Nanosep ultrafiltration devices (Pall Nanosep Omega 3K, OD003C34) and centrifuged to remove rBGLs. Sugar profile was then detected using HPLC with an Aminex HPX-87H column (Bio-Rad) at 60 ◦ C. 4 mM H2 SO4 was used as the mobile phase at 0.5 ml/min. Individual sugar (glucose, cellobiose, cellotriose) concentrations were determined by peak area. Square and circle indicate the cellotriose and cellobiose, respectively. Solid and hollow symbols represent for the oligosaccharides produced by rSfBGL1 and rTrBGL1, respectively.

370 371

372

a significant activity increase (200.7/285.0% for rTrBGL1/rSfBGL1) (Fig. 6d). 3.7. Transglycosylation properties

385

Transglycosylation activity was examined using glucose as substrate. The peaks of the cellobiose at 7.34 min and cellotriose at 6.73 min as shown by HPLC data were visible but not observed in control sample, indicating their transglycosylation activity. The production of cellobiose and cellotriose catalyzed by rSfBGL1 increased more quickly than that by rTrBGL1 from 3 h to 48 h, and the rSfBGL1/rTrBGL1 finally acted catalytically to produce 19.15/0.58 mg/ml of cello-oligosaccharides, suggesting that rSfBGL1 has higher transglycosylation activity (Fig. 7). Although rSfBGL1 synthesized 3.33 mg/ml oligosaccharide at 6 h, but 12.6% (approximately 126 mg/ml) oligocellulose was obtained by TcBGL at 6 h (Pal et al., 2010), indicating that rSfBGL1 and rTrBGL1 exist lower transglycosylation activity than transglycosylating enzymes.

386

4. Discussion

373 374 375 376 377 378 379 380 381 382 383 384

387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402

In this paper, rSfBGL1 and rTrBGL1 were successfully expressed in P. pastoris and several factors impacting production were investigated. BMMY medium is propitious to growth and ␤-glucosidase expression. BMMY contains peptone and yeast extracts compared to BMMH and MMH, and thus providing adequate nutrients for growth. Peptone possibly acts as alternative and competing substrates for one or more problem proteases, and can also repress protease induction caused by nitrogen limitation (Macauley-Patrick et al., 2005). The highest expression levels are usually obtained in BMMY compared with other two media (Toonkool et al., 2006), but BMMH is sometimes reported to be a better expression medium (Al-Rashed et al., 2010; Slámová et al., 2012). So the medium must be optimized for the production of each protein. Methanol is used as a carbon source and also as an inducer. The 0.5%/1.0% methanol was optimal for the production of rSfBGL1/rTrBGL1. Depending on the sorts of heterologous proteins, the 0.25–3.0% (v/v) of methanol

is usually optimal for various protein production in P. pastoris (Gao and Shi, 2013). Excess methanol inhibits cell growth and protein production (Macauley-Patrick et al., 2005), and the low levels of methanol may not be enough to initiate transcription (Cereghino Q5 and Cregg 2000). Accurate regulation of the methanol concentration is thus necessary to maintain induction of genes and prevent the accumulation of methanol to a toxic level. Besides, glycerol and sorbitol was fed as carbon source in mixtures with methanol in the induction phase to enhance cell densities and avoid the toxicity of excess methanol to cells, resulting in a productivity improvement of heterologous proteins (Potvin et al., 2012). P. pastoris is capable of growing across a relatively broad pH range (3.0–7.0), which allows considerable freedom in adjusting the pH to one that is not optimal for a problem protease (MacauleyPatrick et al., 2005). The optimal pH for rSfBGL1/rTrBGL1 production is 4.5/5.5 indicating the existence of neutral proteases in culture. Both methanol concentration and pH of medium showed significant effects on the rBGL yield. Therefore, the two factors were further simultaneously optimized to improve production. RSM is considered as an efficient method for non-linear optimization, however, to date, there are no reports on optimization of rBGL production in P. pastoris by using it. We first used RSM to optimize extracellular rBGL production. Considerable improvements were obviously observed, and the production of rTrBGL1/rSfBGL1 was increased by 137/84% compared with the maximum yields obtained in single-factor experiments, proving that RSM can quickly define the optimum conditions for extracellular production of enzyme. Activity determination under different conditions can introduce some inaccuracy in the values of parameters for systemic comparison of their properties. We first reported on a comparison of biochemical properties between rTrBGL1 and rSfBGL1 under same conditions, which provides useful information for using them as a resource of gene expression and further improvement of enzyme properties. TrBGL1 showed better thermo-tolerance and SfBGL1 had better acid-tolerance. The optimum temperature and pH of native TrBGL1 were 70 ◦ C/pH 4.6 (Chen et al., 1992), which is consistent with that of rTrBGL1 in this study. The optimal temperature of several ␤-glucosidases is also reported to be about 70 ◦ C (Harnpicharnchai et al., 2009; Pei et al., 2012). The rTrBGL1 also showed better thermostability than rSfBGL1, and it retained approximately 90% of activity after 3 h of incubation at 30–50 ◦ C and 58–77% of that after 60 ◦ C for 1–3 h. The thermal stability is higher than that of some thermostable ␤-glucosidases (Pei et al., 2012) but is less than and nearly that of TnBgl3B (Pozzo et al., 2010) and BCC2871 BGLI (Harnpicharnchai et al., 2009), respectively. The difference in thermal stability of rSfBGL1 and rTrBGL1 may be due to different spatial structure, especially the difference in FnIII of D3. FnIII of TnBgl3B has been speculated to contribute to thermostability by acting as a clamp that reduces the amplitude of relative movements between D1 and D2, but its function has not yet been unambiguously demonstrated (Pozzo et al., 2010). FnIII has also been found in a number of bacterial extracellular GHs, but is little reported from GHF3. Thus the motif is worth to further study. Metal ions and chaotropic agent can affect activity of enzyme. The study reveals that FeSO4 , K2 Cr2 O7 and MnSO4 are activator, and this is the first report that K2 Cr2 O7 can activate enzymes. In previous studies, strong activation by Mn2+ was observed in several ␤-glucosidases, such as CcBglA, NkBgl1 and TrBgl2 (Jeng et al., 2011), but the real role of Mn2+ has not been clarified. Fe2+ usually inhibits activity of ␤-glucosidases. Thus, Fe2+ , Mn2+ or Cr2O72− coordinated ␤-glucosidase structure should be solved in order to clarify the real role of the ions in enzyme function. Both rTrBGL1 and rSfBGL1 were not sensitive to the given chaotropic agent but they were slightly sensitive to EDTA. Interestingly, ethanol did not inhibit but promote the activity, and 4% ethanol enhanced

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468

G Model BIOTEC 7090 1–10

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487

488

489 490 491 492 493 494 495 496 497 498 499 500 501 502 503

Q6 504

10.5/31.5% activity of rSfBGL1/rTrBGL1 (Table 1), which may be due to that the change in polarity of the medium caused by alcohols could stabilize enzyme conformation (Karnaouri et al., 2013; Mateo and Di Stefano, 1997). Previous study has reported that short chain alcohols can activate several ␤-glucosidase from fungi (Karnaouri et al., 2013; Harnpicharnchai et al., 2009). Some glycosidases have the ability to catalyze transglycosylation reactions, and enzymatic glycosylation is particularly useful for the modification of complex biologically active substances. The transglycosylation property for ␤-glucosidases from family GHF1 has been studied recently. This is the first report of the transglycosylation properties of rSfBGL1/rTrBGL1 from GHF3. For a ␤-glucosidases, hydrolysis and transglycosylation are two divergent routes. High substrate concentrations act to shift the flux from the hydrolytic to the transglycosylation pathway, and transglycosylation can account for a moderate part of the slowdown in hydrolysis (Bohlin et al., 2013). The rSfBGL1 and rTrBGL1 show relative low transglycosylation activity, demonstrating that they behave as hydrolytic enzyme. 5. Conclusions We have successfully cloned and expressed two ␤-glucosidase genes TrBGL1 and SfBGL1 in P. pastrois. Single-factor experiments show that methanol concentration and pH are the key factors affecting protein expression, and a final significant improvement in yield (137/84% of increase in rTrBGL1/rSfBGL1 production) is quickly obtained using RSM. Functional and structural characterization of the two rBGLs provides a useful comparison between them and gives key information for using them as gene resources and rational design of hybrid enzymes. Metal ions Cr6+ , Mn2+ and Fe2+ result in 51–221% increase of activities as activators. Cr6+ is the most effective activator, and it is the first time to report that it can significantly activate enzyme, which gives new angle of view about improvement of enzyme activity. The current study thus provides a useful production strategy and valuable insights to biochemical function of ␤-glucosidases. Uncited reference

du Plessis et al. (2010).Acknowledgements We gratefully acknowledge Ms Huina Dong, Dr. Kun Zhang for 506 advice on experiments. We also thank Shicheng Yang, Xiangchen 507 Guan and Jingjing Li who helped to carry out cloning and enzyme 508 Q7 assay. This work was supported by National Natural Science Foun509 dation of China (NSFC-30900033). 510 505

511

512 513 514

515

516 517 518 519 520 521 522 523 524 525 526 527

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jbiotec.2015.04. 016 References Al-Rashed, S.A.A., Bakar, F.D.A., Said, M., Hassan, O., Rabu, A., Illias, R.M., Murad, A.M.A., 2010. Expression and characterization of the recombinant Trichoderma virens endochitinase Cht2. Afr. J. Microbiol. Res. 4 (16), 1758–1767. Barnett, C.C., Berka, R.M., Fowler, T., 1991. Cloning and amplification of the gene encoding an xetracellular ␤-glucosidase from Trichoderma reesei: evidence for improved rates of saccharification of cellulosic substrates. Nat. Biotechnol. 9, 562–567. Bhatia, Y., Mishra, S., Bisaria, V.S., 2002. Microbial ␤-glucosidases: cloning, properties, and applications. Crit. Rev. Biotechnol. 22, 375–407. Bohlin, C., Praestgaard, E., Baumann, M., Borch, K., Praestgaard, J., Monrad, R., Westh, P., 2013. A comparative study of hydrolysis and transglycosylation activities of fungal ␤-glucosidases. Appl. Microbiol. Biotechnol. 97, 159–169.

9

Chen, H., Hayn, M., Esterbauer, H., 1992. Purification and characterization of two extracellular ␤-glucosidases from Trichoderma reesei. Biochim. Biophys. Acta 1121, 54–60. Chen, M., Zhao, J., Xia, L., 2008. Enzymatic hydrolysis of maize straw polysaccharides for the production of reducing sugars. Carbohydr. Polym. 71, 411–415. Cummings, C., Fowler, T., 1996. Secretion of Trichoderma reesei ␤-glucosidase by Saccharomyces cerevisiae. Curr. Genet. 29, 227–233. Dong, H., Zhao, X., Ma, Y., Zhang, M., 2012. Optimization of a synthetic medium for ethanol production by xylose-fermenting Zymomonas mobilis using response surface methodology. Chin. Sci. Bull. 57, 3782–3789. du Plessis, L., Rose, S.H., van Zyl, W.H., 2010. Exploring improved endoglucanase expression in Saccharomyces cerevisiae strains. Appl. Microbiol. Biotechnol. 86, 1503–1511. Gao, M., Shi, Z., 2013. Process control and optimization for heterologous protein production by methylotrophic Pichia pastoris. Chin. J. Chem. Eng. 21, 216–226. Harnpicharnchai, P., Champreda, V., Sornlake, W., Eurwilaichitr, L., 2009. A thermotolerant ␤-glucosidase isolated from an endophytic fungi Periconia sp., with a possible use for biomass conversion to sugars. Protein Expr. Purif. 67, 61–69. Husain, Q., 2010. Beta galactosidases and their potential applications: a review. Crit. Rev. Biotechnol. 30, 41–62. Jeng, W.Y., Wang, N.C., Lin, M.H., Lin, C.T., Liaw, Y.C., Chang, W.J., Liu, C.I., Liang, P.H., Wang, A.H.J., 2011. Structural and functional analysis of three ␤-glucosidases from bacterium Clostridium cellulovorans, fungus Trichoderma reesei and termite Neotermes koshunensis. J. Struct. Biol. 173, 46–56. Karnaouri, A., Topakas, E., Paschos, T., Taouki, I., Christakopoulos, P., 2013. Cloning, expression and characterization of an ethanol tolerant GH3 beta-glucosidase from Myceliophthora thermophila. Pers. J. 1, e46, http://dx.doi.org/10.7717/peerj. 46 Ketudat Cairns, J.R., Esen, A., 2010. ␤-Glucosidases. Cell. Mol. Life Sci. 67, 3389–3405. Liu, X., Liu, C., Ma, Y., Hong, J., Zhang, M., 2014. Heterologous expression and functional characterization of a novel cellulose-disruptive protein LeEXP2 from Lycopersicum esculentum. J. Biotechnol. 186, 148–155. Macauley-Patrick, S., Fazenda, M.L., McNeil, B., Harvey, L.M., 2005. Heterologous protein production using the Pichia pastoris expression system. Yeast 22, 249–270. Mach, R.L., Seiboth, B., Myasnikov, A., Gonzalez, R., Strauss, J., Harkki, A.M., Kubicek, C.P., 1995. The bgl1 gene of Trichoderma reesei QM 9414 encodes an extracellular, cellulose-inducible beta-glucosidase involved in cellulase induction by sophorose. Mol. Microbiol. 16, 687–697. Mateo, J.J., Di Stefano, R., 1997. Description of the ␤-glucosidase activity of wine yeasts. Food Microbiol. 14, 583–591. Medyantseva, E.P., Vertlib, M.G., Budnikov, G.K., 1998. Metal ions as enzyme effectors. Russ. Chem. Rev. 67, 225–232. Merino, S.T., Cherry, J., 2007. Progress and challenges in enzyme development for biomass utilization. Adv. Biochem. Eng. Biotechnol. 108, 95–120. Nakamura, S., Ban, M., Kato, A., 2006. Preparation of bioactive and surface functional oligomannosyl neoglycoprotein using extracellular pH-sensitive glycosylation of mutant lysozyme having N-linked signal sequence in yeast. Bioconjug. Chem. 17, 1170–1177. Pal, S., Banik, S.P., Ghorai, S., Chowdhury, S., Khowala, S., 2010. Purification and characterization of a thermostable intra-cellular ␤-glucosidase with transglycosylation properties from filamentous fungus Termitomyces clypeatus. Bioresour. Technol. 101, 2412–2420. Pei, J., Pang, Q., Zhao, L., Fan, S., Shi, H., 2012. Thermoanaerobacterium thermosaccharolyticum beta-glucosidase: a glucose-tolerant enzyme with high specific activity for cellobiose. Biotechnol. Biofuels 5, 31, http://dx.doi.org/10.1186/ 1754-6834-5-31 Potvin, G., Ahmad, A., Zhang, Z., 2012. Bioprocess engineering aspects of heterologous protein production in Pichia pastoris. A review. Biochem. Eng. J. 64, 91–105. Pozzo, T., Pasten, J.L., Karlsson, E.N., Logan, D.T., 2010. Structural and functional analyses of beta-glucosidase 3B from Thermotoga neapolitana: a thermostable three-domain representative of glycoside hydrolase 3. J. Mol. Biol. 397, 724–739. Shen, Y., Zhang, Y., Ma, T., Bao, X., Du, F., Zhuang, G., Qu, Y., 2008. Simultaneous saccharification and fermentation of acid-pretreated corncobs with a recombinant Saccharomyces cerevisiae expressing ␤-glucosidase. Bioresour. Technol. 99, 5099–5103. Slámová, K., Bojarová, P., Gerstorferová, D., Fliedrová, B., Hofmeisterová, J., Fiala, M., Pompach, P., Kˇren, V., 2012. Sequencing, cloning and high-yield expression of a fungal ␤-N-acetylhexosaminidase in Pichia pastoris. Protein Expr. Purif. 82, 212–217. Toonkool, P., Metheenukul, P., Sujiwattanarat, P., Paiboon, P., Tongtubtim, N., Ketudat-Cairns, M., Ketudat-Cairns, J., Svasti, J., 2006. Expression and purification of dalcochinase, a beta-glucosidase from Dalbergia cochinchinensis Pierre, in yeast and bacterial hosts. Protein Expr. Purif. 48, 195–204. van Rooyen, R., Hahn-Hagerdal, B., La Grange, D.C., Van Zyl, W.H., 2005. Construction of cellobiose-growing and fermenting Saccharomyces cerevisiae strains. J. Biotechnol. 120, 284–295. Wang, W., Liu, C., Ma, Y., Liu, X., Zhang, K., Zhang, M., 2014. Improved production of two expansin-like proteins in Pichia pastoris and investigation of their functional properties. Biochem. Eng. J. 84, 16–27.

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613

G Model BIOTEC 7090 1–10 10 614 615 616 617 618 619 620

ARTICLE IN PRESS Y. Ma et al. / Journal of Biotechnology xxx (2015) xxx–xxx

Yoshida, E., Hidaka, M., Fushinobu, S., Koyanagi, T., Minami, H., Tamaki, H., Kitaoka, M., Katayama, T., Kumagai, H., 2010. Role of a PA14 domain in determining substrate specificity of a glycoside hydrolase family 3 beta-glucosidase from Kluyveromyces marxianus. Biochem. J. 431, 39–49. Zhang, J., Zhong, Y., Zhao, X., Wang, T., 2010. Development of the cellulolytic fungus Trichoderma reesei strain with enhanced beta-glucosidase and filter paper activity using strong artificial cellobiohydrolase 1 promoter. Bioresour. Technol. 101, 9815–9818.

Zhang, L., Guo, Z., Hong, J., Ding, Z., Gao, Z., He, Z., Shi, G., 2012. Expressing ␤glucosidase from Saccharomycopsis fibuligera in industrial ethanol producing yeast and evaluation of the expressing sufficiency. Ann. Microbiol. 62, 539–544. Zhu, T., You, L., Gong, F., Xie, M., Xue, Y., Li, Y., Ma, Y., 2011. Combinatorial strategy of sorbitol feeding and low-temperature induction leads to high-level production of alkaline ␤-mannanase in Pichia pastoris. Enzyme Microb. Technol. 49, 407–412.

Please cite this article in press as: Ma, Y., et al., Expression optimization and biochemical properties of two glycosyl hydrolase family 3 beta-glucosidases. J. Biotechnol. (2015), http://dx.doi.org/10.1016/j.jbiotec.2015.04.016

621 622 623 624 625 626 627