Cloning, purification, and characterization of a thermostable β-glucosidase from Thermotoga thermarum DSM 5069

Abstract

A 56-kDa β-glucosidase (TthBgl) derived from Thermotoga thermarum DSM 5069 was expressed and purified from Escherichia coli BL21 (DE3). The purified enzyme showed hydrolytic activity towards only p-nitrophenyl-β-D-glucopyranoside among the synthetic glycosides tested. The pH maximum was 5.0, and under the conditions tested, maximal activity was at 85 ºC, and pH stability occurred from 5.0 to 6.0. After being incubated at 80 ºC for 120 min, TthBgl retained 80% of its original activity. The β-glucosidase had no apparent requirement for metal ions or other co-factors, but its activity was significantly inhibited by 0.1% SDS and 1mM Cu2+, in which only 3% and 10% residual activity was maintained, respectively. The Vmax of TthBgl was 8.79 U mg-1 for p-nitrophenyl-β-D-glucopyranoside, while the Km was 2.41 mM. The Enzyme activity was gradually inhibited by the addition of glucose, but remained approximately 50% of its original value in 500 mM glucose. 789.25 mg/L glucose was released from cellobiose by the incubation of 0.2 U/mL TthBgl for 9 h at 75 ºC. According to a phylogenetic analysis, TthBgl belongs to the glycosyl hydrolase family 3 (GH3).

Full Article

Cloning, Purification, and Characterization of a Thermostable β-Glucosidase from Thermotoga thermarum DSM 5069

Lingfeng Long,^a,b.c Hao Shi,^a,b Xun Li,^a,b Yu Zhang,^a,b Jinguang Hu,^c and Fei Wang ^a,b,*

A 56-kDa β-glucosidase (TthBgl) derived from Thermotoga thermarum DSM 5069 was expressed and purified from Escherichia coli BL21 (DE3). The purified enzyme showed hydrolytic activity towards only p-nitrophenyl-β-D-glucopyranoside among the synthetic glycosides tested. The pH maximum was 5.0, and under the conditions tested, maximal activity was at 85 ºC, and pH stability occurred from 5.0 to 6.0. After being incubated at 80 ºC for 120 min, TthBgl retained 80% of its original activity. The β-glucosidase had no apparent requirement for metal ions or other co-factors, but its activity was significantly inhibited by 0.1% SDS and 1mM Cu²⁺, in which only 3% and 10% residual activity was maintained, respectively. The V_max of TthBgl was 8.79 U mg^-1 for p-nitrophenyl-β-D-glucopyranoside, while the K_m was 2.41 mM. The Enzyme activity was gradually inhibited by the addition of glucose, but remained approximately 50% of its original value in 500 mM glucose. 789.25 mg/L glucose was released from cellobiose by the incubation of 0.2 U/mL TthBgl for 9 h at 75 ºC. According to a phylogenetic analysis, TthBgl belongs to the glycosyl hydrolase family 3 (GH3).

Keywords: GH3 β-glucosidase; Thermotoga thermarum DSM 5069; Thermostability

Contact information: a: College of Chemical Engineering, Nanjing Forestry University, Nanjing, Jiangsu 210037, China; b: Jiangsu Key Lab of Biomass-based Green Fuels and Chemicals, Nanjing, Jiangsu 21003,7 China; c: Department of Wood Science, University of British Columbia, Vancouver, BC, Canada V6T 1Z4; *Corresponding author: hgwf@njfu.edu.cn

INTRODUCTION

Cellulose, a polymer of glucose units joined together by β-1,4-glycosidic bonds, is the most abundant and renewable lignocellulosic biomass resource on earth (Denman et al. 1996). The degradation of cellulose requires the synergistic action of several types of enzymes, including endoglucanase (EC 3.2.1.4), cellobiohydrolase (EC 3.2.1.91), and β-glucosidase (EC 3.2.1.21) (Yan et al. 2012). After the actions of endoglucanase and cellobiohydrolase, β-glucosidase (Bgl) removes glucose from either non-reducing cellobiose or cellotriose (Wu et al. 2013). Because of its contribution to the hydrolysis of cellulose, Bgl has attracted significant attention for use in a wide range of industrial processes, such as improving the flavor of wine and clarifying juice in beverage and food production (Ketudat Cairns and Esen 2010; González-Pombo et al. 2011).

Glycoside hydrolases are classified into families based on their amino acid similarities. Of the 99 families classified, Bgls belong to families GH1, GH3, GH5, GH9, GH30, and GH116 (http://www.cazy.org/fam/acc_GH.html). In general, Bgls from thermophiles or hyperthermophiles are much more stable than those from mesophiles because their special structures allow them to maintain enzymatic activity at high temperatures. In addition to increasing the hydrolysis reaction rate, high temperature can also reduce the risk of contamination (dos Santos et al. 2011; Shi et al. 2014). Therefore, thermostable Bgls deriving from hyperthermophiles have great potential in industrial processes. For example, in the industrial production of paper and pulp, it is commonly accepted that a preliminary hemicellulose extraction stage produces a certain amount of solubilized glucan that is useless for pulping and papermaking processes. The high temperature waste containing hemicellulose and cellulose can be further processed into various end-products, such as bioethanol for energy purposes. Thus, enzymatic hydrolysis at high temperature not only can save energy for cooling the samples down, but also can decrease the substrate viscosity, a factor that inhibits the hydrolysis of biomass at high-solids loadings (Loaiza et al. 2015).

GH3 β-glucosidases have been purified and characterized from many organisms, including bacteria, eukaryotes, and archaea (http://www.cazy.org/fam/acc_GH.html; Roy et al. 2005; Kudou et al. 2014). Thermotoga thermarum 5069 is a hyperthermophilic bacterium that grows at 80 ºC; it was isolated from continental solfataric springs at Lac Abbe in Djibouti, Somalia, Africa (Windberger et al. 1989). In this study, a thermostable β-glucosidase from Thermotoga thermarum 5069 was cloned, expressed, and characterized.

EXPERIMENTAL

Materials and Methods

Bacterial strains, plasmid, and growth conditions

Escherichia coli Top10 and E. coli BL21 (DE3) were used as cloning and expression hosts, respectively. Both were grown at 37 ºC in Luria-Bertani medium (LB) containing ampicillin (100 μg/mL). The plasmid, pET-20b (Novagen, Darmstadt, Germany), was used as the cloning and expression vector.

Plasmid construction

A DNA fragment of about 1500 bp (GenBank Protein No. WP_013931664) from Thermotoga thermarum DSM 5069 was amplified by PCR using the following primers: TthBgl-1, 5’-GGAATTCCATATGACGCTTTCGGAGGTTGTTG-3’, and TthBgl-2, 5’-CCGCTCGAGCCTTAACACCTCCACCGGCAGT-3’. Restriction sites for NdeI and XhoI are underlined. The 1.5-kb PCR product was digested with NdeI and XhoI (Takara, Dalian, China) and ligated into the NdeI- and XhoI- linearized expression vector, pET-20b, to produce the recombinant plasmid pET-20b-TthBgl.

Expression and purification of TthBgl

The recombinant plasmid pET-20b-TthBgl was transformed into E. coli BL21 (DE3). The TthBgl gene was expressed in overnight cultures containing pET-20b-TthBgl in 200 mL of LB medium supplemented with ampicillin (100 μg/mL) and 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG). The cells were harvested by centrifugation (10,000 × g for 10 min at 4 ºC), washed twice, and re-suspended in 5 mL of 5 mM imidazole, 0.5 mM NaCl, and 20 mM Tris-HCl Buffer (pH 7.9). The cell extracts after sonication were treated at 60 ºC for 30 min, cooled in an ice bath, and centrifuged (10,000 × g for 30 min at 4 ºC). The obtained supernatant was loaded onto a Ni-NTA affinity column (Novagen, Darmstadt, Germany). A gradient of 20 mM to 200 mM imidazole was used for eluting the protein, and 1-mL fractions were collected for activity assays and SDS-polyacrylamide gel electrophoresis (SDS-PAGE) analysis.

Enzyme assay

Standard β-glucosidase assays (200 μL) contained 20 mM p-nitrophenyl-β-D-glucopyranoside (10 μL) and diluted enzyme (10 μL) in 50 mM imidazole-potassium buffer. The reaction was incubated at 85 °C for 10 min and terminated by the addition of 0.6 mL of 1 M Na₂CO₃. The amount of p-nitrophenol released was determined by measuring the absorbance at 405 nm. One unit of enzyme activity was defined as the amount of enzyme releasing 1 μmol p-nitrophenyl-β-D-glucopyranoside (pNPG) per minute. The purified protein concentration was determined by the Bradford method (1976) using bovine serum albumin (BSA) as a standard.

Characterization of TthBgl

The optimal pH of TthBgl was analyzed in 50 mM imidazole-potassium buffer covering a pH range of 4.0 to 6.5 in increments of 0.5 at 80 ºC. The pH stability was examined by pre-incubating TthBgl for 1 h. The effect of temperature on enzymatic activity was determined by incubation at temperatures ranging from 70 ºC to 95 ºC for 10 min. TthBgl was pre-incubated at 75 ºC, 80 ºC, 85 ºC, and 90 ºC for 30 min, 60 min, 90 min, and 120 min, respectively. Thermostability assays were conducted by measuring residual TthBgl activity. The results were expressed as percentages of the original activity.

The substrate specificity of the enzyme was tested using the chromogenic p-nitrophenyl (pNP)-glycosides pNP-α-D-glucopyranoside, pNP-β-D-glucopyranoside, pNP-α-D-galactopyranoside, pNP-β-D-galactopyranoside, pNP-β-D-mannopyranoside, pNP-β-D-xylopyranoside, and pNP-α-L-arabinofuranoside (Sigma, St. Louis, USA) as well as cellobiose (Aladdin, Shanghai, China). Enzyme activity was examined at 85 ºC and pH 5.0 for 10 min with pNPG from 0.2 mM to 1 mM to calculate kinetic parameters (V_max and K_m) and turnover number (k_cat) according to the Lineweaver-Burk method. The influence of various glucose concentrations (100, 200, 300, 400, 500, 600, 700, 800, 900, and 1000 mM) on β-glucosidase activity was assayed. The K_i value of glucose was defined as amount of glucose required for inhibiting 50% of the β-glucosidase activity and was given as the averages of three separate determinations.

The effects of metal ions and chemical reagents on TthBgl activity were also determined. Cu²⁺, Co²⁺, Al³⁺, Ba²⁺, Mn²⁺, Ni²⁺, Ca²⁺, Mg²⁺, and Zn²⁺ at final concentrations of 1 mM were mixed with the enzyme for 2 h, and TthBgl activity was measured. Chemical reagents, including Tween 80, Tris, SDS, and EDTA, were also mixed with the enzyme for 2 h at final concentrations of 0.05%, 0.05%, 0.1%, and 0.5 mM, respectively.

Hydrolysis of cellobiose and HPLC analysis of cellobiose product

The enzyme (0.2 U) was condensed using a SpeedVac concentrator (Thermo, Waltham, USA) and combined with 0.02 g of cellobiose in 50 mM imidazole-potassium buffer (pH 5.0) in a 2-mL bioreactor at 75 ºC. The enzymatic reaction was carried out for 1, 3, 5, and 9 h and stopped by cooling the reaction mixture to -20 ºC. Samples were centrifuged at 13,000 × g for 15 min to remove the protein. Then, 0.4 mL of supernatant was mixed with 1.2 mL ethanol and lyophilized, and the pellets were re-suspended in 400 μL of water. Separation and quantification of glucose were performed using a Prevail carbohydrate column (4.6 × 250 mm, 5 μm) with high-performance liquid chromatography (HPLC) (model 1260, Agilent, Santa Clara, USA) with an evaporative light scattering (ELSD) detector (Alltech, Shanghai, China). Acetonitrile/water (75/25%, v/v) was used as the mobile phase with a flow rate of 0.8 mL/min.

Bioinformatics analysis

Amino acid sequence similarities were examined by means of the system BLAST (http://blast.ncbi.nlm.nih.gov/). Sequence alignment of several Bgls was performed using Clustal X2 (Larkin et al. 2007). The neighbor-joining (NJ) and maximum-parsimony (MP) trees were created in Mega 6 software (megasoftware.net) in order to postulate phylogenetic relationships of various Bgls from different organisms.

RESULTS AND DISCUSSION

Expression and Purification

A putative gene encoding β-glucosidase from T. thermarum was cloned and expressed. After purification via heat treatment and Ni²⁺ column affinity chromatography, the enzyme showed a single band on SDS-PAGE, with a molecular mass of approximately 56 kDa (Fig. 1), which is consistent with the theoretical value of 56,264 kDa, which had been separately determined with the Compute pI/Mw tool (http://web.expasy.org/compute_pi/). The monomeric mass of β-glucosidase in this study was similar to the value of a 56 kDa β-glucosidase from Thermotoga lettingae TMO (Compute pI/Mw tool), which showed a 63.35% sequence similarity with the TthBgl (described below). Its mass was lower than the 95 kDa GH3 β-glucosidase from Thermotoga maritima MSB8 (Gabelsberger et al. 1993), the 81 kDa GH3 β-glucosidase from Thermotoga neapolitana (Compute pI/Mw tool), the 77.8 kDa GH3 β-glucosidase from Thermofilum pendens (Li et al. 2013), and the 83.5 kDa GH3 β-glucosidase from Talaromyces emersonii (Collins et al. 2007).

Fig. 1. SDS-PAGE analysis of purified TthBgl. Lane M: protein marker, Lane 1: culture supernatant of E. coli BL21 (DE3), Lane 2: culture supernatant treated at 60 ºC for 30 min, Lane 3: TthBgl purified by Ni²⁺ affinity column chromatography

Biochemical and Kinetic Parameters

The enzyme activity was examined over a pH range of 3.5 to 6.5 at 80 ºC. The maximum activity of TthBgl was recorded at pH 5.0 (Fig. 2a). TthBgl had a higher acid-tolerance than β-glucosidases from T. maritima MSB8 (pH 6.2) and Thermotoga petrophila RKU-1 (pH 7) (GenBank Protein No. ABQ46970) (Gabelsberger et al. 1993; Haq et al. 2012) but a similar optimal pH to β-glucosidase from Pholiota adiposa (Jagtap et al. 2013). The optimal temperature of TthBgl was 85 ºC (Fig. 2b), which is higher than the optimum for enzymes from Actinosynnema mirum (37 ºC), Azospirillum irakense (45 ºC), Flavobacterium johnsoniae (37 ºC), Martelella mediterranea (45 ºC), Myceliophthora thermophile (60 ºC), Aspergillus niger (70 ºC), and Thermoascus aurantiacusCBMAI-756 (75 ºC) (Rashid and Siddiqui 1997; Faure et al. 2001; Leite et al. 2008; Mao et al. 2010; Hong et al. 2012; Cui et al. 2013; Zhao et al. 2015). With good thermostability, the enzyme maintains its activity for a longer time, and the amount needed in the reaction is reduced (Pei et al. 2012; Shi et al. 2013). After pre-incubation at 80 ºC for 120 min, TthBgl retained 80% of the initial activity (Fig. 2c).

Fig. 2. Enzymatic properties of recombinant TthBgl. a: Optimal pH of TthBgl (80 ºC, pH 4.0 to 6.0 for 10 min). b: Optimal temperature (pH 5.0, 70 to 95 ºC for 10 min). c: Thermostability of TthBgl (pH 5.0, 75 to 90 ºC for 0, 30, 60, 90, and 120 min). d: pH stability (80 ºC, pH 4.0 to 6.5 for 1 h). The maximum activity was defined as 100% (a, b). The initial activity was defined as 100% (c, d). Data shown are from one typical experiment that was repeated three times, and the variation about the mean was below 5%.

β-glucosidase from Thermofilum pendens maintained 95% activity after incubation for 120 min at 90 ºC (Li et al. 2013), and the residual activity of His-tagged β-glucosidase from T. maritima is 75% after being incubated at 80 ºC for 60 min (Xue et al. 2009). The highly thermostable β-glucosidase from T. petrophila RKU-1 (ABQ46970) retains 80% activity after incubation for 120 min at 80 ºC (Haq et al. 2012). Therefore, compared with other thermostable Bgls in the GH3 family, TthBgl had a high thermostability at 80 ºC and pH stability at pH 5.0 to 6.0 with 90% remaining activity (Fig. 2d).

Kinetic parameters of TthBgl were obtained from Lineweaver-Burk plots. The results showed that its V_max and K_m were 8.79 U mg^-1 and 2.41 mM (R² = 0.99), respectively, and k_cat was 8.25 s^-1.

In terms of substrate specificity, β-glucosidases are divided into three groups: those that hydrolyze oligosaccharides only, those with a strong affinity for aryl-β-glucose, and those able to hydrolyze multiple substrates (Zhao et al. 2013). When the substrate specificity was assayed with different substrates, TthBgl presented a strong affinity to p-nitrophenyl-β-D-glucopyranoside, confirming that this enzyme is a β-glucosidase (data not shown). Thus, TthBgl from Thermotoga thermarum DSM 5069 may belong to the second group, those having a strong affinity for aryl-β-glucose.

Bgl activity is known to be affected by many factors. Glucose concentration during the degradation of cellobiose or cellotriose inhibits enzyme activity via a feedback loop (Liu et al. 2011). Most Bgls that hydrolyze cellobiose are extremely sensitive to their own product, glucose (Saha and Bothast 1996), and in this study, TthBgl from Thermotoga thermarum DSM 5069 was no exception (Fig. 3). The relative activity of TthBgl decreased continuously as more glucose was added to the reaction. At 100 mM glucose, the relative activity declined steeply to around 75%, showing that TthBgl was sensitive to glucose. However, TthBgl still remained 50% of the original value in 500 mM, which is higher than the thermostable GH3 β-glucosidase from Penicillium brasilianum (2.3 mM), Talaromyces emersonii(0.245 mM), and the β-glucosidase III from Aspergillus tubingensis CBS 643.92 (470 mM) (Krogh et al. 2010; Murray et al. 2004; Decker et al. 2001). The reaction was conducted with pNPG as the substrate (Table 1).

Fig. 3. Effects of glucose on TthBgl activity. The values are the mean of three separate experiments, and the standard deviations were below 5%.

Table 1. Characteristics of Recombinant β-glucosidase of T. thermarum

Because Bgls can be inhibited or activated by metal ions or chemical reagents, the influence of several cations and chemical reagents on TthBgl was also investigated (Table 2). The addition of cations and chemical reagents did not noticeably increase the activity of TthBgl. However, recombinant TthBgl was completely inhibited by 0.1% SDS and 1 mM Cu²⁺. In most cases, the cation Cu²⁺ is an inhibitor of glycoside hydrolases (Shi et al. 2014). The glycoside hydrolases, e.g., β-xylosidase and α-L-arabinofuranosidase, from Thermotoga thermarum 5069 are also somewhat inhibited by Cu²⁺ (Shi et al. 2013, Shi et al. 2014). The enzyme was moderately inhibited by Co²⁺, Al³⁺, Zn²⁺, and Tris, and its enzyme activity remained roughly constant after the addition of 0.5 mM EDTA, which confirmed that no metal ions were required in the reaction.

Table 2. Effects of Cations and Chemical Reagents on Purified TthBgl Activity

Note: The activity of the enzyme without pre-incubation was defined as 100%. The experiment was performed in triplicate.

Release of glucose from cellobiose was also tested. A hydrolytic reaction was carried out by adding purified enzyme sample into mixtures at 75 ºC. This was done because TthBgl exhibited high thermostability at this temperature. The quantities of glucose released from cellobiose were 118.27 mg/L, 299.23 mg/L, 454.82 mg/L, and 789.25 mg/L for enzymatic reactions lasting 1, 3, 5, and 9 h, respectively.

Multiple Sequence Alignment and Phylogenetic Analysis of TthBgl

To further classify TthBgl, 22 candidate sequences were used to construct phylogenetic trees using the neighbor-joining (NJ) and maximum-parsimony (MP) methods (Fig. 4). Both methods exhibited similar topological structures (MP tree not shown). There were two well-supported clades in NJ trees, and each of them delegated a glycoside hydrolases family. Clade I was the GH3 β-glucosidase mainly from bacteria but also included archaea and fungi. Clade II was the GH1 β-glucosidase from bacteria and archaea. Although Bgls exist in families GH1, GH3, GH5, GH9, GH30, and GH116, only a few of them belonged to families GH5, GH9, GH30, and GH116. Hence, the phylogenetic tree did not present Bgls from these families mentioned above.

Fig. 4. Neighbor-joining (NJ) tree analysis of 22 β-glucosidase amino acid sequences from different organisms. Numbers on nodes correspond to percentage bootstrap values for 1000 replicates.

The amino acid sequence analysis indicated that TthBgl β-glucosidase belonged to the GH3 family. Through multiple sequence alignment using Clustal X2, TthBgl from Thermotoga thermarum possessed 65.07% sequence similarity with the β-glucosidase-related glycosidase from Thermotoga profunda (WP_041083663), 63.35% from Thermotoga lettingae TMO (WP_012003721), and 60.44% from Thermotoga caldifontis (WP_041076019). However, TthBgl β-glucosidase from Thermotoga thermarum shared only 13.72% sequence similarity with the Bgls from Thermotoga maritima MSB8 (WP_004082478) and Thermotoga petrophila RKU-1 (ABQ46916) and 13.32% similarity with the Bgl from Thermotoga neapolitana DSM 4359 (ACM22846). Therefore, the new GH3 TthBgl would be expected to exhibit some distinct properties. To further investigate the sequence similarity, the amino acid sequences of Clade I were aligned again. The exoglucanase from Hordeum vulgare (barley) was used as a structure-determined representative from the family. The equivalent amino acid residue in TthBgl is conserved, D254 (Fig. 5).

Fig. 5. Sequence alignment of the amino acid sequences of Thermotoga thermarum DSM 5069 with those of Thermotoga lettingae (T. lettingae), Thermotoga profunda (T. profunda) and Thermotoga caldifontis (T. caldifontis). The active sites are indicated as * on the top of the alignment.

The phylogenetic tree revealed that TthBgl is closely related to a β-glucosidase-related glycosidase from T. lettingae TMO. Unfortunately, the enzyme from T. lettingae TMO has not been characterized, and consequently, it was not possible to compare the enzymatic properties of these two enzymes. The tree also showed a distant relationship between β-glucosidases from T. thermarum DSM 5069 and T. maritima MSB8. For substrate specificity, TthBgl only hydrolyzed p-nitrophenyl-β-D-glucopyranoside, while β-glucosidases from T. maritima MSB8 and T. petrophila RKU-1 (ABQ46916) hydrolyze several types of pNP-glycosides (Gabelsberger et al. 1993; Cota et al. 2015). In terms of K_m, TthBgl had a value of 2.41 mM for pNPG, which was lower than that of T. petrophila RKU-1 (ABQ46970) (2.8 mM) but higher than that of T. maritima MSB8 (0.43 mM) and T. petrophila RKU-1 (ABQ46916) (0.38 ± 0.03 mM) (Xue et al. 2009; Haq et al. 2012; Cota et al. 2015).

CONCLUSIONS

1. A β-glucosidase gene from Thermotoga thermarum DSM 5069, TthBgl, was cloned and sequenced. The recombinant protein was expressed, purified, and characterized for enzymatic properties.
2. Amino acid sequence analysis of TthBgl protein placed the enzyme in family 3 of the glycoside hydrolases. The recombinant enzyme possessed high thermostability and substrate specificity.

ACKNOWLEDGMENTS

This work was financially supported by the State Forestry Administration (No. 2014-4-37), the National Natural Science Foundation of China (No. 31370572, 31270612), and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

REFERENCES CITED