Introduction

Xylan is one of the major hemicellulosic polysaccharides in plant cell wall. It exists as a branched heteropolysaccharide containing a β-(1,4)-D-xylose backbone with substituting groups comprising different pentose and hexose sugars and sugar acids (Beg et al. 2001). Xylans can be broadly classified as homoxylans, arabinoxylans, glucuronoxylans and arabinoglucuronoxylans according to their main sugar unit (Polizeli et al. 2005). Biodegradation of xylan requires a group of xylanolytic enzymes with different specificities working co-operatively. Among them, endoxylanase (EC 3.2.1.8) is the most important enzyme that cleaves the internal β-1,4 bonds in the xylan main chain into xylooligosaccharides of different lengths (Biely 1985).

Microbial xylanases represent a major group of industrial enzymes applicable to degrade plant derived xylan with potent applications in saccharification of lignocellulosic substrates to xylose and to a promising prebiotic xylooligosaccharides (XOS) (Dodd and Cann 2009). In addition, they can be used in animal feed (Abelilla and Stein 2019), pulp and paper (Walia et al. 2017), textile (Singh et al. 2020) and food industries (Adiguzel et al. 2019). XOS are oligosaccharides of xylose residues that are linked by β-(1,4)-glycosidic bonds. The degree of polymerization (DP) of XOS can vary from 2 to 6 xylose units and their chemical side chains and substitution pattern can vary depending on xylan sources and the production process (Moure et al. 2006; Poletto et al. 2020; Samanta et al. 2015).

A number of xylanases from various bacteria, yeast and fungi have been isolated and characterized (Alokika and Singh 2019; Ding et al. 2018; Liu et al. 2018). Filamentous fungi are widely used as sources of xylanases due to their higher enzyme production levels in comparison with bacteria and yeasts (Ahmed et al. 2009). However, xylanases are produced and secreted together with other hydrolytic enzymes, making it difficult to obtain an endoxylanase with no other enzymes contributing to unwanted side activities. Heterologous expression of fungal endoxylanases with varying biochemical properties and specificities from a number of fungal origins, for example, in genera Aspergillus, Trichoderma, and Penicillium have been reported in various and heterologous hosts (Jun et al. 2009; Asano et al. 2005; Sriprang et al. 2006; Zhuo et al. 2018). Searching for new xylanases with high specificity for XOS production at moderate temperature in a suitable range to inhibit microbial growth while allowing high solubility of substrates is of great interest with an advantage to save the energy cost in production process (Choudhary et al. 2016).

In this study, we further explore the application of the recombinant endoxylanase originated from a thermoresistant fungus A. niger BCC14405 (Asano et al. 2005) and previously expressed in Pichia pastoris (Ruanglek et al. 2007). The recombinant xylanase XylB was further characterized for its biochemical properties with the focus on its specificities on production of XOS. The enzyme showed high endoxylanase activity with high specificity towards production of xylobiose to xylohexaose (X2–X6) preferred for prebiotic application (Aachary and Prapulla 2011; Samanta et al. 2015) while showed no side activities on cellulose. The enzyme is considered potent for further application in production of xylooligosaccharides and saccharification of hemicellulose fraction in lignocellulosic biomass.

Materials and methods

Materials, strains, plasmids and chemicals

pPICZαA (Invitrogen, Carlsbad, CA) was used as the expression vector. Escherichia coli DH5α (Invitrogen, Carlsbad, CA) was used as a host strain for DNA cloning. P. pastoris KM71 (Invitrogen, Carlsbad, CA) was used for recombinant protein expression. Beechwood xylan and carboxymethyl cellulose (CMC) were purchased from Sigma-Aldrich (St. Louis, MO). Wheat arabinoxylan and xylooligosaccharides (X2-X6) were purchased from Megazyme (Wicklow, Ireland). Avicel PH101 was purchased from Fluka (Milwaukee, WI) and used for preparing phosphoric acid swollen cellulose (PASC) (Zhang et al. 2006). All chemicals and reagents were analytical grade and obtained from major chemical suppliers (Sigma, Merck and Fluka).

Construction of recombinant plasmid

The xylB gene was amplified from pET28/xylB plasmid (Sriprang et al. 2006) using primers XylBF/EcoRI (5′-GCGAATTCTCGACCCCGAGCTCGACC-3′) and XylBR/XbaI (5′-GCTCT AGAGCCTGAACAGTAATGGAGGA-3′). Restriction sites for cloning are underlined. The gene was cloned into pPICZαA expression vector and transformed into E. coli DH5α. The transformed clones were selected on LB agar containing 25 µg mL−1 Zeocin. The plasmid harboring the inserted gene was extracted from the selected clone and then confirmed by sequencing using AOX forward and AOX reverse primers. The recombinant plasmid pPICZαA-xylB was linearized with PmeI and transformed into P. pastoris KM71 using electroporation and then selected on YPD agar with 100 µg mL−1 Zeocin. The transformed clone was confirmed by colony PCR using both AOX primers and gene-specific primers.

Protein expression

Expression of the recombinant enzyme from P. pastoris KM71 was performed according to the method described in the supplier’s protocol (MAN0000035). P. pastoris KM71 containing xylB gene was cultured overnight in 5 mL YPD broth at 30 °C and used as a starter. For protein expression, 0.2% (v/v) starter was inoculated into 500 mL BMGY medium in a 2-L flask and then cultured at 30 °C with 200 rpm shaking for 16 h (OD600 = 5–8). The cells were collected by centrifugation at 5000×g for 5 min. The cell pellet was then resuspended in 100 mL BMMY medium and transferred to a 1-L flask. The culture was induced in BMMY medium at 30 °C with 200 rpm shaking. Methanol was added to 0.3% final concentration at 24 h intervals until 72 h. The crude enzyme was harvested by centrifugation at 5000×g at 4 °C for 10 min. The crude enzyme preparation was stored at 4 °C for further experiments. The protein concentration was determined by Bradford assay using the Bio-Rad Protein assay solution (Bio-Rad, Hercules, CA).

Enzyme activity assay

The enzyme activities were determined by measuring the released reducing sugar based on dinitrosalicylic acid (DNS) assay (Miller 1959) using xylose as a standard. The standard reaction contained crude enzyme XylB at an appropriate concentration and 1% (w/v) of the substrate selected from beechwood xylan dissolved in 100 mM sodium phosphate buffer, pH 6.0. The reactions were incubated at 45 °C for 10 min or otherwise specified. After incubation, the reactions were stopped by addition of twofold volumes of DNS and boiled for 10 min. The released reducing sugar was determined by measuring the absorbance at 540 nm. The absorbance data were used to calculate the enzyme activities in which 1 unit of enzyme was defined as the amount of enzyme that produces 1 µ mole of product from substrate per 1 min under the specified condition.

Biochemical characterization

In order to determine the effect of temperature on enzyme activity, the XylB was incubated with 1% (w/v) beechwood xylan dissolved in 100 mM sodium phosphate buffer (pH 6.0) for 10 min. The temperatures of reaction were varied from 30 to 80 °C to determine the optimal temperature. To investigate the optimal pH, XylB was incubated with 1% (w/v) beechwood xylan dissolved in 100 mM buffers with varying pH range, including sodium acetate buffer (pH 4.0–5.5), sodium phosphate buffer (pH 6.0–8.0) and Tris–HCl buffer (pH 8.0–9.0) at 45 °C for 10 min.

To examine thermostability, XylB was pre-incubated in 100 mM sodium phosphate buffer (pH 6.0) at 40–60 °C for 0, 15, 30, 45 and 60 min. The incubated XylB was then added to the reaction containing 1% (w/v) of beechwood xylan dissolved in 100 mM sodium phosphate buffer (pH 6.0) for 10 min. For substrate specificity, various substrates including beechwood xylan, Avicel: PH101 microcrystalline cellulose, carboxymethyl cellulose (CMC), wheat arabinoxylan, phosphoric acid swollen cellulose (PASC) were used to examine the substrate specificity of XylB. Enzyme kinetics of XylB was studied using 0–10 mg mL−1 xylan as the substrate based on Michaelis–Menten equation with 0–120 min of incubation.

Effect of metal ions and chemical surfactants

To investigate the effect of metal ions and chemical surfactants, various ions and surfactants were added to the enzyme reactions containing 1% (w/v) beechwood xylan dissolved in 100 mM sodium phosphate buffer (pH 6.0) with the final concentration of 1 mM for metal ions or 0.1% (v/v) for surfactant and chemicals at 45 °C for 10 min. The metal ions and surfactants used in this study included Na+ (NaCl), K+ (KCl), Zn2+ (ZnCl2), Co2+ (CoCl2 · 6H2O), Mn2+ (MnCl2), Li+ (LiCl), Mg2+ (MgCl2·6H2O), Cu2+ (CuCl2·2H2O), Ni2+ (NiCl2 · 6H2O), Ca2+ (CaCl2·2H2O), Tween 20 (% v/v), Triton X-100 (% v/v), β-mercaptoethanol (% v/v), urea (% w/v) and sodium dodecyl sulphate (SDS) (% w/v). Differences in means among the treatments were identified using independent-samples T-test (p < 0.05) using SPSS 16.0.

Cleavage pattern analysis

The product profile of XylB on hydrolysis of xylooligosaccharides (X2-X6) was studied. The reaction mixture containing 1 mg mL−1 of xylooligosaccharides dissolved in 100 mM sodium phosphate buffer (pH 6.0) and 100 μg of XylB was incubated in Thermomixer C (Eppendorf, Hamburg, Germany) at 45 °C with shaking at 1000 rpm for 24 h. The sugar products released into the supernatant were analyzed using Dionex3000 high performance liquid chromatography (Thermo Fisher Scientific, Waltham, MA) equipped with Aminex HPX-87H column and refractive index detector (Shodex, Kyoto, Japan) using 5 mM sulfuric acid as a mobile phase at 65 °C with a flow rate 0.5 mL min−1.

Production of xylooligosaccharides (XOS) from beechwood xylan

Enzymatic hydrolysis was studied in 1 mL reaction containing 1% (w/v) of beechwood xylan dissolved in 100 mM sodium phosphate (pH 6.0) with an enzyme dosage of 2.5, 5 and 10 mg g−1 substrate. The reaction was incubated at 45 °C with shaking at 200 rpm for 24 h. Profiles of xylooligosaccharides were analyzed using HPLC with the condition described above. The experiments were carried out in triplicates.

Results

Heterologous expression of xylB in P. pastoris KM71

The xylB gene (546 bp) encoding for xylanase (GH11) was previously isolated from A. niger BCC14405. The mature gene was expressed as an intracellular soluble protein with the approximate molecular weight of 21 kDa in E. coli BL21 (Sriprang et al. 2006). However, the recombinant XylB expressed in E. coli required further downstream processes including cell disruption and purification step. Therefore, in this study, the xylB gene was cloned and expressed as a secreted protein in P. pastoris KM71 with a molecular weight of approximately 25 kDa after induction at 24–72 h (Fig. 1). The enzyme was expressed as a single major band on SDS-PAGE with > 95% homogeneity. No activity of xylanase was detected in P. pastoris KM71 strain carrying an empty vector (pPICZαA without XylB gene) used as the negative control. The molecular weight of the protein obtained is larger than the prediction (21 kDa). This could be due to post-translational glycosylation (Thak et al. 2020). The yield of enzyme production obtained in shake flask scale was 116 mg protein L−1 of culture medium. Due to its remarkably high homogeneity, XylB was further characterized with no additional purification step.

Fig. 1
figure 1

Expression of recombinant XylB in P. pastoris KM 71. Protein expression was induced with 3% (v/v) methanol at 30 °C, 200 rpm for 24–72 h. Lane M represents unstained protein marker (Thermo Scientific). Lane 1–3 represent recombinant XylB after induced 24, 48 and 72 h, respectively

Full size image

Biochemical properties

The enzyme activity of XylB was tested against beechwood xylan under various conditions in order to investigate the effect of pH and temperature. For the optimal pH, enzyme activity was determined under various pH ranging from pH 4.0 to 9.0. XylB exhibited the highest xylanase activity in 100 mM sodium phosphate pH 6.0. Remarkably, more than 80% remaining activity was observed at pH 5.0–6.0 (Fig. 2a). For the optimal temperature of XylB, enzyme activity was determined using beechwood xylan as a substrate by comparing the activity at various temperatures ranging from 30 to 80 °C. According to the results, XylB exhibited the highest activity at 45 °C with more than 60% of its maximal activity at a temperature between 40 and 50 °C (Fig. 2b). For thermostability, more than 80% and 40% remaining activity were observed after the enzyme was incubated without substrate for 15–60 min at 40 °C and 50 °C, respectively. However, the activity substantially decreased after incubation at 60 °C for 15 min and was completely lost after incubation for 45 min (Fig. 2c).

Fig. 2
figure 2

Effects of pH and temperature on XylB activity. a Effect of pH was analyzed in the reaction containing 1% (w/v) of beechwood xylan in 100 mM sodium acetate buffer (pH 4.0–5.5), 100 mM sodium phosphate buffer (pH 6.0–8.0) and 100 mM Tris–HCl buffer (pH 8.0–9.0) at 45 °C. b Effect of temperature was investigated in the reaction containing 1% (w/v) of beechwood xylan in100 mM sodium phosphate buffer pH 6.0 at 30–80 °C. c Thermostability was investigated by incubated the enzyme in 100 mM sodium phosphate buffer pH 6.0 at 40, 50 and 60 °C for 15–60 min

Full size image

The effect of metal ions (Ca2+, Co2+, Cu2+, K+, Li+ Mg2+, Mn2+, Na+, Ni2+ and Zn2+), surfactants and chemicals (β-mercaptoethanol, SDS, urea, Triton X-100 and Tween 20) on XylB activity were evaluated. According to the results, Cu2+, Mg2+, Ni2+, Na+, Ca2+, Zn2+, K+, Li+ enhanced XylB activity with 8%-28.4% increase, whereas XylB activity drastically decrease with 47% loss of activity in the reaction contain Mn2+ (Table 1). Moreover, Mn2+, β-mercaptoethanol, urea and SDS led to 50% -90% reduction in enzyme activity in the respective conditions. Triton X-100 and Tween 20 did not influence the activity of the enzyme.

Table 1 The effects of metal ions and surfactants on the activity of recombinant XylB
Full size table

Substrate specificity on different polysaccharides

In order to characterize the specificity of XylB, the enzyme activity was tested against various hemicellulosic (beechwood xylan and wheat arabinoxylan) and cellulosic substrates (carboxymethyl cellulose (CMC), phosphoric acid swollen cellulose (PASC) and Avicel). The highest specific activity was detected against beechwood xylan (3852 U mg−1 protein) with a weak activity against wheat arabinoxylan (169 U mg−1 protein) (Table 2). In addition, XylB did not degrade any type of cellulose polysaccharides used in this experiment at a detectable level. The kinetic study of XylB was then determined using beechwood xylan and arabinoxylan as the substrates.

Table 2 Substrate specificity and kinetic parameters of recombinant XylB
Full size table

Product profile from xylooligosaccharide and beechwood xylan hydrolysis

In order to investigate the cleavage pattern of recombinant XylB, the hydrolysis products obtained from xylooligosaccharides (X2-X6) hydrolysis were analyzed. Based on the results, X3-X6 were mainly hydrolyzed into xylose (X1), xylobiose (X2) and xylotriose (X3) by XylB (Fig. 3). However, XylB was unable to cleave xylobiose under the experimental conditions. The hydrolysis of xylan from beechwood was then performed. According to the results, xylooligosaccharides released from beechwood xylan hydrolysis comprised xylobiose (X2), xylotriose (X3) and xylohexaose (X6) with the highest XOS yield of 708 mg g−1 substrate (equivalent to 70% conversion yield) using an enzyme dosage of 10 mg g−1 substrate (Fig. 4).

Fig. 3
figure 3

Product profile obtained from hydrolysis of xylooligosaccharides. The reactions contained 1 mg mL−1 of xylooligosaccharides (X2-X6) dissolved in 100 mM sodium phosphate buffer pH 6.0 and 100 µg mg−1 substrate of XylB. The reactions were incubated at 45 °C for 24 h. A Standard sugars (X1-X6); B Xylobiose; C Xylotriose; D Xylotetraose; E Xylopentaose and F Xylohexaose

Full size image
Fig. 4
figure 4

Product profile obtained from hydrolysis of beechwood xylan. The reactions contained 10 mg mL−1 of beechwood xylan dissolved in 100 mM sodium phosphate buffer pH 6.0 and 2.5–10 mg g−1 substrate of XylB. The reactions were incubated at 45 °C for 24 h

Full size image

Discussion

In our previous study, the gene encoding for the endoxylanase XylB was isolated from A. niger BCC14405 and successfully heterologously expressed as an intracellular protein in E. coli BL21 (DE3) (pLysS) expression system (Asano et al. 2005; Sriprang et al. 2006). According to the previous results, the wild type and recombinant xylanases exhibited the highest specificity towards xylan from birchwood with the activity of 5870 and 7663 U mg−1, respectively. This demonstrated the potential application of XylB for several biotechnological industries. However, the XylB was expressed as an intracellular protein that required additional cell disruption and protein purification steps. The production of recombinant enzyme in a secreted form is one of the important properties for economically feasible production in an industrial scale. Several reports demonstrated that P. pastoris has been successfully used as an expression host for production of extracellular recombinant proteins (Bunterngsook et al. 2017; Ding et al. 2018; Li et al. 2018; Zhuo et al. 2018). Expression of this enzyme in P. pastoris and optimization of the fermentation conditions for enzyme production was previously reported with demonstration of its application as an additive in animal feed (Ruanglek et al. 2007). However, the recombinant enzyme has not been characterized in detail for its biochemical properties and its application on other biotechnological purposes has not been studied.

In this work, the xylB gene was expressed in P. pastoris KM71 system by fusing with the alpha factor secretion signal under the control of methanol inducible promoter. The recombinant XylB was highly expressed as a secreted protein with a production yield of 116 mg protein L−1 of culture medium. Based on its biochemical characteristics, XylB showed high specificity against beechwood xylan (3852 U mg−1) and arabinoxylan (168 U mg−1) without any detectable cellulolytic activity on celluloses similar to the purified xylanase from A. niger BCC14405 reported previously (Asano et al. 2005). The structure of Endo-1,4-β-xylanase GH11 called jelly-roll scaffold contains two parts at the active binding site that make the enzyme specific to xylose-based polysaccharides. The first part is the thumb-loop of GH11, which makes the cleft narrow in subsites and is designed for binding xylose-based polysaccharides (Sulzenbacher et al. 1999) but not for glucose-based substrates (Paës et al. 2012). The second part is the highly conserved Val or Leu residue at position 35 of GH11, which binds only to xylan and prevents cellulose-binding because of the lack of hydrogen bonds with the O6 of glucose residue (Sidhu et al. 1999; Sulzenbacher et al. 1999). According to amino acid sequence alignment of XylB compared with the data in SWISS-MODEL (https://swissmodel.expasy.org/), the results suggested that XylB show similarities to Xylanase 11C (Endo-1,4-β-xylanase) from Talaromyces cellulolyticus (PDB ID: 3wp3) (Kataoka et al. 2014) at 67% identity and 78% homology. The structure prediction revealed the presence of both conserved parts in XylB, consistent with its specificity on xylan polysaccharides. This specificity towards xylan is a unique feature of endo-xylanase belonging to GH11 (Paës et al. 2012). For example, an endoxylanase from Trichoderma sp. exhibited xylanase activity towards xylan from beechwood, wheat arabinoxylan and rye arabinoxylan, without hydrolytic activity on CMC and Avicel (Fu et al. 2019). A xylanase purified from Myceliophthora heterothallica F.2.1.4 was active only on beechwood xylan (de Oliveira Simões et al. 2019). However, XylB obtained in this work revealed relatively low optimal temperature (45 °C) compared to endoxylanase purified from A. niger BCC14405 crude enzyme (55 °C) (Asano et al. 2005) and recombinant XylB produced in E. coli expression system (50 °C) (Sriprang et al. 2006). This could be due to differences in post-translational modifications of protein expressed in different microbial systems. The differences in optimal working conditions and specific activity of XylB to the enzyme expressed in P. pastoris (Ruanglek et al. 2007) could come from the use of different sources of xylan substrates (birchwood and beechwood xylan) and components in enzyme assay reactions. The lower optimal temperature of recombinant XylB in our study is desirable for biotechnological application of the enzyme for XOS production. The optimal temperature range of 45–50 °C was reported to be desirable as it inhibits growth of contaminating microbes while allows high solubility of the hemicellulosic substrates (Choudhary et al. 2016). This can also lead to lower processing temperature compared to the use of thermophilic enzymes (Damásio et al. 2011; Qiao et al. 2014; Sriyapai et al. 2011). The optimal temperature of XylB was relatively low compared to most of the reported GH11 endo-xylanases which work optimally at 40–80 °C, while our enzyme showed significantly higher specific activity compared to most previously reported enzymes originated from various microbial sources (Table 3). Our previous work on XylB expression in the E. coli system (Sriprang et al. 2006) did not report the yield of enzyme production and showed only the specific activity 7663 U mg−1 protein against birchwood xylan. The data thus cannot be directly compared to the use of different substrates. The specific activity against arabinoxylan of XylB was 168 U mg−1 which was higher than commercial xylanases such as endo-1,4-β-Xylanase M4 (Product code: E-XYAN4; Megazyme) from A. niger (80 U mg−1) and endo-1,4-β-Xylanase M3 (Product code: E-XYTR3; Megazyme) from Trichoderma longibrachiatum (100 U mg−1).

Table 3 Comparison of optimal conditions of several endoxylanase GH11
Full size table

According to the effect of metal ions and chemical additives on XylB activity, most of the metal ions including Cu2+, Mg2+, Ni2+, Na+, Ca2+, Zn2+, K+, Li+ enhanced XylB activity while β-mercaptoethanol and SDS led to reduction in enzyme activity under the respective conditions. An inhibitory effect of Mn2+ on xylanase activity was reported (Cheng et al. 2012; Mamo et al. 2006). The reduction in enzyme activity by SDS could be due to its property as an anionic surfactant which could destabilize existing interactions in the protein structure leading to enzyme denaturation (da Silva et al. 2017). The effect of β-mercaptoethanol could be due to protein conformation shifting, resulted from changes in the overall charge and structure of the protein, which can lead to decreased enzyme activity (Ahmed et al. 1996).

Regarding its cleavage pattern on xylooligosaccharide (X2–X6), the recombinant XylB showed a capability to act on xylotriose to xylohexaose and released xylobiose (X2) as the major product. Furthermore, xylobiose was not hydrolyzed by the XylB, suggesting that the XylB acted on oligosaccharides containing at least three xylose units. This was supported by the high saccharification efficiency (70% conversion) of xylan from beechwood in which the XOS (X2, X3 and X6) were released. The remaining X6 in the hydrolysis reaction using xylan as the substrate could represent incomplete digestion of the substrate under the experimental conditions. The unlabeled peaks were the background of sodium phosphate buffer (Fig. 3b–f). Its cleavage patterns were similar to some xylanases in glycosyl hydrolase family 11 that hydrolyzed hemicellulose substrates into XOS with different chain lengths. For example, XynST11 from Streptomyces sp. B6 could not hydrolyze xylobiose and the hydrolysis product of xylotriose was xylobiose (Liu et al. 2020). In the same way, the major products from hydrolysis of xylotetraose and xylopentaose were xylotetraose and xylotriose. Some xylanases from other studies were reported to be incapable of xylotriose hydrolysis (Chang et al. 2017; Valenzuela et al. 2014; Chen et al. 2009). In general, the major XOS products from enzymatic hydrolysis are xylobiose, xylotriose, and xylotetraose (Linares-Pasten et al. 2018). The product profile of XylB agrees well with some GH11 xylanases previously reported. The endoxylanase isolated from bagasse metagenome and its variants were recently reported to hydrolyze beechwood xylan into X3–X6 (Boonyapakron et al. 2020). Berrin and coworkers reported the predominant production of xylobiose and xylotriose as end products from wheat soluble arabinoxylan hydrolysis by PgXynA and PfXynC GH11 xylanases from Penicillium griseofulvum (Berrin et al. 2007). In addition, a GH11 xylanase (XynST11) isolated from Streptomyces sp. B6 produced much less xylose and a higher amount of XOS with a DP from 2 to 4 than the GH10 family enzyme (XynST10) (Liu et al. 2020). Overall, these previous works demonstrate the advantages of GH11 xylanases for XOS production and great potential of XylB for conversion of the plant-derived xylan to oligosaccharides.

Conclusion

In summary, recombinant XylB isolated from A. niger BCC14405 fungal strain was highly expressed in a secreted form in the yeast expression system. XylB is a cellulase-free xylanase that exhibits catalytic capability to hydrolyze xylan from beechwood and produce XOS. The properties of XylB demonstrate its potential applications for XOS production and lignocellulosic biomass saccharification.