Production and purification of xylanases and lignins

Four xylanases classified to CAZy family GH11 were studied (Table 1). TrXyn1 and TrXyn2 were endogenous proteins of T. reesei and were homologously produced. On the contrary, Xyl40, which has been isolated from a metagenomic library and is putatively of prokaryotic origin, was produced heterologously in a prokaryotic host E. coli and a eukaryotic host T. reesei. The purified T. reesei single-domain xylanases TrXyn1 and TrXyn2, and the two-domain Xyl40 produced in E. coli, all contained a single major band according to the SDS-PAGE analysis (Additional file 1: Fig. S1). A truncated version Xyl40-CD was produced heterologously in T. reesei with the CBM removed. Furthermore, to prevent overglycosylation and consequent misfolding of the putatively prokaryotic protein in the eukaryotic production host, two out of three glycosylation sites were removed from the catalytic domain.

Table 1 Characteristics of GH11 xylanases used in the present study

The purification of the single-domain xylanase Xyl40-CD from T. reesei culture supernatant resulted in two protein pools. The first pool corresponded to one 23.2 kDa size protein and the second pool corresponded to a mix of a 23.2 kDa and an 18.9 kDa size protein. These accounted for 30% and 70% of the total protein in the second pool, based on the SDS-PAGE band intensities, respectively. Deglycosylation of the second pool of Xyl40-CD with endoH resulted in a disappearance of the upper band, while a single band corresponding to molecular weight of 18.9 kDa remained (Additional file 1: Fig. S2). This indicated that the two protein bands in SDS-PAGE represented glycosylated and non-glycosylated forms of the Xyl40-CD. The glycosylated form of Xyl40-CD in the first pool was used in the following experiments. In addition, the deglycosylated Xyl40-CD was used in studying the role of glycosylation in lignin-derived inhibition.

All the xylanases are classified to family GH11 and had a similar fold in the catalytic domain; however, they differed in several aspects summarised in Table 1. The studied xylanases had a catalytic domain of approximately 20 kDa, but Xyl40, carrying a CBM60 domain, had substantially higher molar mass of 35.9 kDa (Table 1). The CBM60 of bacterium Cellvibrio japonicus has been reported to target the binding on xylans, galactans and cellulose through a ligand binding cleft [21]. All xylanases had a low amount of potential salt bridges (3–5). The two forms of Xyl40 had two to three potential disulphide bridges, whereas the TrXyn1 and TrXyn2 did not have any (Table 1). According to Tenkanen et al. [19], the T. reesei xylanases TrXyn1 and TrXyn2 had only minor amount of structural glycans, < 1% of weight. However, TrXyn1 had no potential N-glycosylation sites, whereas TrXyn2 had three potential glycosylation sites (Fig. 6) and has been reported to be glycosylated [22]. Xyl40 was produced in the prokaryotic host in which glycosylation does not occur. As described above, the production of Xyl40-CD resulted in N-glycosylation of a fraction of the enzyme and the glycosylated fraction was used in the experiments. The deglycosylation of the recombinant Xyl40-CD resulted in a 4.3 kDa reduction in size on the SDS-PAGE. The specific activities on birch wood xylan of the xylanases varied between 3200 and 23,000 nkat/mg protein at pH 5, TrXyn2 having the highest specific activity (Table 1). Previously, TrXyn2 has been observed to have a higher specific activity than TrXyn1 and Xyl40 on birch wood xylan [19, 20]. The genetically engineered Xyl40-CD produced in T. reesei had four times higher specific activity on birch wood xylan than the Xyl40 produced in E. coli. The presence of glycans can increase or decrease xylanase activity and thermal stability [23, 24]. The higher activity detected with Xyl40-CD may arise not only from glycosylation but also from the different production strains. Proteins may fold incorrectly depending on the production strain, especially correct disulphide bond formation in the periplasm of E. coli may not always be successful, resulting in loss of enzyme function [25].

Lignin was isolated from steam pretreated spruce by extensive enzymatic hydrolysis followed by a protease treatment to remove the bound enzymes. The lignin content of the enzymatic hydrolysis residues (EnzHR) was 85.6% with a residual carbohydrate content of 13.9% and nitrogen content of 0.4% after protease treatment (Additional file 1: Table S1). These are within previously reported values of enzymatically isolated lignins [8, 10, 26]. Protease activity was measured from 1% (w/V) EnzHR lignin suspension to confirm that the lignin preparation was not contaminated by proteases used in the isolation procedure. The protease activity of the EnzHR lignin was below the quantitation limit indicating that over 99% of the initially added protease activity had been removed.

Thermal stability of the GH11 xylanases

Thermal stability of the enzymes was compared by measuring residual activities of TrXyn1, TrXyn2, Xyl40-CD, and Xyl40 after incubation at 40–70 °C in buffer with 0.1 mg/ml BSA for 2 or 24 h. The metagenomic xylanases Xyl40 and Xyl40-CD had significantly higher thermal stability than the T. reesei xylanases (Fig. 1). The metagenomic xylanases retained 80–92% of activity after 24 h of incubation at 40–60 °C and only significantly lost their activity after incubation at 70 °C, having 9–15% residual activity. TrXyn1 and TrXyn2 were both stable at 40 °C, but after 24 h at 50 °C, TrXyn2 retained only 39% of the original activity, whereas the activity of TrXyn1 was totally lost. No residual activity was detected for TrXyn1 or TrXyn2 at incubation for 2 h at 60 °C or above. The measured thermal stabilities of these xylanases were in accordance to previously reported values for TrXyn1, TrXyn2 and Xyl40 [19, 20].

Fig. 1
figure 1

Thermal stability of xylanases. Residual activity of xylanases after a 2 h and b 24 h incubation at temperatures 40 °C, 50 °C, 60 °C and 70 °C. Activity measurements were performed at 50 °C with the standard xylanase assay before and after incubation

The effect of lignin on the xylanases

Inhibitory effect of lignin on xylan hydrolysis

The inhibitory effect of lignin on TrXyn1, TrXyn2, Xyl40-CD and Xyl40 was first compared by carrying out hydrolysis assays with 1% birch wood xylan as substrate in the presence or absence of 1% lignin-rich enzymatic hydrolysis residue (EnzHR) at 40 °C. This temperature was selected based on the measured temperature stabilities of these enzymes (Fig. 1). The enzymes were dosed on protein weight basis (1 µg protein/g dry xylan). TrXyn2 and Xyl40-CD had up to seven times higher specific activity on birch wood xylan than TrXyn1 and Xyl40, which was also seen in the hydrolysis progress curves (Fig. 2, left column). The addition of EnzHR lignin led to a 12–27% decrease in hydrolysis yields compared to the control without lignin for all xylanases (Fig. 2, left column). The most thermostable Xyl40-CD was least affected by lignin.

Fig. 2
figure 2

Effect of lignin on xylan hydrolysis at 40 °C (left column) and 50 °C (right column). Hydrolysis of xylan (1% w/V) by xylanases a, b TrXyn1, c, d TrXyn2, e, f Xyl40 and g, h Xyl40-CD in 50 mM Na-citrate buffer, pH 5. (■) Xylan hydrolysed as such, () xylan supplemented with 1% w/V enzymatic hydrolysis residue (EnzHR) lignin from steam pretreated spruce and (▲) xylan supplemented with a buffer soluble fraction of the same lignin. The hydrolysis was followed for 48 h at 40 °C and for 3 h or 24 h at 50 °C

The effect of temperature on the hydrolysis yields was further studied at 50 °C at selected time points (Fig. 2, right column). At this temperature, the hydrolysis yield of the most thermostable xylanase, Xyl40-CD, was least affected by lignin having only a 19% decrease in yield after 3 h (Fig. 2h). Instead, the hydrolysis yield of the most thermolabile xylanase TrXyn1 was significantly affected by lignin, having a 49% reduction in yield already after 3 h of hydrolysis at 50 °C (Fig. 2b). However, also the yield of the thermostable Xyl40 decreased by 66% after the addition of EnzHR lignin after 3 h (Fig. 2f). TrXyn1 and Xyl40 had a lower hydrolysis rate than TrXyn2 and Xyl40-CD, and therefore, the hydrolysis with these enzymes was carried out to up to 24 h. The hydrolysis yield of TrXyn1 did not increase after 3 h in the presence of EnzHR lignin, confirming the detrimental effect of the lignin on the activity of this enzyme (Fig. 2b). Instead, the hydrolysis reaction with Xyl40 proceeded, and after 24 h the reduction in hydrolysis yield by lignin was only 38% lower compared to the control, indicating that the lignin slowed the xylan hydrolysis but did not completely inhibit the enzyme (Fig. 2f).

Several studies report that hydrolysis residue lignins inhibit enzymes by non-productive binding and inactivating enzymes [6, 7, 10, 11]. Previously, thermostable cellulases have been seen to exhibit lower binding to lignin and increased lignin tolerance in hydrolysis conditions compared to thermolabile cellulases [13]. As such, inhibition of the thermostable xylanase variants Xyl40-CD and Xyl40 even at 40 °C was surprising, suggesting that thermostability is not the only measure for predicting lignin-derived inactivation. Furthermore, the Xyl40 was inhibited by lignin more than Xyl40-CD at the elevated temperature. The major differences between the Xyl40 enzymes were that Xyl40-CD was glycosylated and contained no CBM, whereas Xyl40 was not glycosylated, but had a CBM60 connected to the catalytic domain (Table 1).

Effect of glycosylation on xylan hydrolysis

The role of N-glycosylation on lignin tolerance was assessed using the enzymatically deglycosylated form of Xyl40-CD. Removal of N-glycans from Xyl40-CD had only a minor effect on the hydrolysis yields on pure xylan but had a remarkable impact on lignin tolerance (Fig. 3). In the presence of EnzHR lignin, the glycosylated form of Xyl40-CD had a 19% lower hydrolysis degree at 24 h when compared to the control without lignin, whereas the non-glycosylated form had a 38% lower hydrolysis degree at 24 h (Fig. 3). This indicates that besides thermal stability, structural glucans improved preservation of the xylanase activity in the presence of lignin.

Fig. 3
figure 3

Effect of N-glycosylation of xylanase Xyl40-CD on lignin tolerance. Hydrolysis yield of xylan (1% w/V) (square) and hydrolysis yield of xylan supplemented with 1% EnzHR lignin from steam pretreated spruce (circle). Hydrolysis performed with a glycosylated (filled symbol) and deglycosylated (open symbol) form of Xyl40-CD in 50 mM Na-citrate buffer, pH 5 at 40 °C

Effect of soluble lignin-derived components on xylan hydrolysis

The role of soluble lignin-derived compounds in the inhibition of xylanases was first studied using a buffer soluble fraction extracted from the EnzHR lignin. The buffer soluble lignin did not inhibit the xylanases confirming that the observed inhibition was due to the solid lignin. Instead, 7–8% increase in the hydrolysis yields was detected for TrXyn2 and Xyl40-CD in the presence of buffer soluble lignin (Fig. 2). These enzymes had higher specific activities than TrXyn1 and Xyl40 (Table 1), both of which were not as clearly affected by the buffer soluble lignin. To further explore the activating effect of the buffer soluble fraction on the xylanases, effects of individual phenolic compounds on the hydrolysis was tested. The compounds were selected based on their presence on lignocellulose pretreatment liquor [27] or in the buffer soluble fraction of isolated lignins [11]. Phenols, protocatechuic acid, syringic acid, ferulic acid, vanillic acid, p-coumaric acid, acetovanillone, vanillin, syringaldehyde and homovanillyl alcohol, were tested at concentrations of 10, 100 and 1000 µg/ml. No activating effects were seen, and only syringaldehyde showed a clear inhibitory effect on the hydrolysis for all the tested xylanases, the effect being most prominent with TrXyn1 with a 50% reduction in the hydrolysis yield (Additional file 1: Table S2). Syringaldehyde was not detected in the buffer soluble fraction of EnzHR lignin isolated from pretreated spruce [11], further confirming that the observed inhibition with EnzHR lignin was due to the solid lignin. Previously, soluble lignin-derived compounds have been seen to be inhibitory to cellulases and xylanases [12, 28]. However, in some cases, low concentrations of monophenols, such as vanillic acid, acetovanillone, protocatechuic acid and ferulic acid, have been reported to increase xylanase activity [16, 29].

Xylanase adsorption and inactivation by lignin

The mechanism of the inhibition by the EnzHR lignin at 40 °C was studied using adsorption experiments. The xylanases were incubated with the EnzHR lignin for up to 24 h. After incubation, the supernatant (free enzyme) and solid fraction (lignin-bound enzyme) were separated by centrifugation. The lignin-bound fraction was washed with buffer containing BSA to remove loosely bound xylanases form the lignin surface (wash buffer). Residual xylanase activity was measured from all fractions and the free and solid fraction were run on SDS-PAGE. Control reactions without lignin were carried out for each of the enzyme to monitor their stability in the test conditions. After incubation with the EnzHR lignin, enzyme activity was lost from the solution indicating that all the xylanases studied were adsorbed to lignin and/or become inactive in the solution (Fig. 4). In case of the thermostable Xyl40-CD and Xyl40, a relatively high proportion of xylanase activity was observed in the solid fraction, comprising 22–49% of the initial xylanase activity added to the reaction, whereas in case of the less thermostable enzymes, TrXyn1 and TrXyn2, all the activity was lost within the 24-h incubation (Fig. 4).

Fig. 4
figure 4

Xylanase activity after incubation with enzymatic hydrolysis residue (EnzHR) lignin. Distribution of xylanase activities for xylanases a TrXyn1, b TrXyn2,c Xyl40 and d Xyl40-CD is presented as a percentage of the initially measured activity using the standard assay conditions (50 °C). Xylanase was incubated in 50 mM Na-citrate buffer, pH 5 at 40 °C in the presence of EnzHR lignin isolated from steam pretreated spruce (Lignin) or without lignin (Control). After incubation for 3 or 24 h the solution was centrifuged and the supernatant (Free enzyme) was separated from the solid lignin containing lignin-bound enzymes. The loosely bound enzymes were washed from the lignin with buffer (loosely bound enzyme) and after centrifugation the pellet containing solid lignin was resuspended into buffer for the activity assay (lignin-bound enzyme)

SDS-PAGE analysis of the free enzyme and lignin-bound enzyme revealed that, indeed, the xylanases TrXyn1, TrXyn2 and Xyl40-CD adsorbed to EnzHR lignin, as the xylanase band was clearly visible in the lignin-bound fraction (57–90% of band intensity) and only a very faint band could be detected in the solution (1–8% band intensity) (Fig. 5). The intensities of the protein bands in the controls without lignin and lignin-bound fractions were very similar for TrXyn1, TrXyn2 and Xyl40-CD, indicating that most of the enzyme could be extracted from lignin using the SDS-PAGE sample buffer. In our earlier work with cellulases, inactivation of the enzymes by lignin adsorption was associated also by impaired release of the enzymes from lignin in the SDS-PAGE analysis, which was interpreted to be due to denaturation of the enzymes on lignin surface [13, 26]. The results suggest that for the xylanases studied here, such irreversible binding was not needed for the enzyme inactivation/inhibition. Unexpectedly, for Xyl40, there was a clear difference in band intensities between 3-h and 24-h samples in the SDS-PAGE (Fig. 5c), while the activity remained constant between 3 and 24 h in the activity assay (Fig. 4c). One explanation for this difference may be that the enzyme produced in E. coli was only partially folded properly, i.e. all the protein observed on the SDS-PAGE was not composed of active enzyme. Possibly the unfolded protein was sensitive to precipitation in the test conditions and therefore was not detected in the SDS-PAGE at 24 h. However, more research would be needed to confirm these speculations.

Fig. 5
figure 5

Xylanase distribution after incubation with enzymatic hydrolysis residue (EnzHR) lignin. Distribution of xylanases a TrXyn1, b TrXyn2, c Xyl40 and d Xyl40-CD to the supernatant (free enzyme) and solid (lignin-bound enzyme) fractions after incubating 10 µg/ml xylanase with 10 mg/ml EnzHR lignin from isolated from steam pretreated spruce. Enzyme incubated without lignin was used as a control sample. These are same samples as are presented in Fig. 4. Xylanase band is indicated with a triangle and band intensities compared to control as presented as % of control sample at 3 h (% data not shown for Xyl40). A band for BSA, which was added after incubation with lignin, is visible in each sample above the 50 kDa standard. Each xylanase was run on their own gel, but for presentation purposes for b, bands were cut to remove unnecessary lanes and indicated as a gap in the image

In control samples incubated without lignin, TrXyn1 and TrXyn2 retained 80 and 63% of activity, respectively, of the initial xylanase activity for 24 h (Fig. 4a and b). For TrXyn1, the residual activity measured in buffer was in line with the activity measured in buffer with BSA in the thermal stability assay (Fig. 1). Instead, TrXyn2 retained 80% of activity at 40 °C in the thermal stability assay carried out with BSA in the buffer (Fig. 1). There was also a significant difference between Xyl40-CD and Xyl40 in the control reactions: the Xyl40-CD retained most of its activity, 79–87%, over the 24-h incubation at 40 °C, while the activity of Xyl40 declined to 39% within the first three hours. In the thermostability assessment, where the enzymes were incubated at similar conditions, but with 0.1 mg/ml BSA included in the buffer, the Xyl40 was clearly more stable at 40 °C (Fig. 1). The presence of BSA had thus a key role in preservation of the activity of TrXyn2 and Xyl40 in the absence of a substrate. BSA can be added to dilute enzyme mixtures, either to prevent enzyme adsorption onto the tube walls or to stabilise the enzyme. In the adsorption experiment, BSA could not be added to the reaction as BSA is known to bind to lignin reducing the adsorption of other enzymes to lignin surface [30]. In our experiments low protein binding tubes were used instead.

Comparison of structural features of the GH11 xylanases

The two GH11 xylanases from T. reesei and the two variants of the metagenomic xylanase Xyl40 were computationally characterised focusing on structural features possibly affecting thermal stability, adsorption to lignin and activity in the presence of lignin. The analysis was done on levels of amino acid sequence and tertiary structures. The metagenomic xylanase Xyl40-CD shared 43% sequence identity with TrXyn1 and 53% with TrXyn2. GH11 xylanases have a highly conserved structure comprising two β-sheets packed against each other forming a shape described as a “right hand” [31]. Structure-based sequence alignment of the catalytic domains of TrXyn1, TrXyn2 and Xyl40-CD showed that the conserved areas in the protein sequences (Fig. 6) were located in the β-strands in the palm region, at the catalytic site as well as the α-helix on the outer surface and C-terminal end in the finger region.

Fig. 6
figure 6

Structure-based sequence alignment of Xyl40-CD, TrXyn1 and TrXyn2 performed with the ESPrit web server. Secondary structural elements of the Xyl40-CD are indicated above the alignment (β-strand arrows and α-helices). Strictly identical amino acid residues are marked in white letters on a black background. Regions of conserved, highly similar residues are framed in thin-lined boxes with bold letters. Possible cysteine bridges are highlighted in purple. Possible salt bridges are highlighted in orange for Xyl40-CD, green for TrXyn1 and blue for TrXyn2; predicted N-glycosylation sites are marked as triangles in orange for Xyl40-CD (empty triangles correspond to removed glycosylation sites of Xyl40-CD) and blue for TrXyn2

Disulphide bridges and salt bridges

There was a clear difference detected between the mesophilic T. reesei xylanases and the thermophilic Xyl40-CD in the amount of cysteine (Cys) residues. The catalytic domain of Xyl40-CD contains two pairs of Cys residues and the CBM60 of Xyl40 a third pair, all pairs within 2.03 Å distance and as such, likely sites for disulphide bridges. On the contrary, TrXyn1 and TrXyn2 sequences did not contain any Cys residues. The Cys-pairs in the 3D model of Xyl40-CD were located in the N-terminal region and thumb region (Fig. 7). A similar disulphide bridge in the N-terminal region has been found in the catalytic domain of family GH11 xylanase EvXyn11. The Xyl40 xylanase originates from an uncultured bacterium and is identical to the PDB template sequence of EnXyn11A used for modelling Xyl40-CD [32]. The 15 min half-life of EvXyn11 has been reported to be improved from 64 °C to 89 °C after seven mutations in the N-terminal region [32], thus highlighting the importance of the N-terminal region for improved thermal stability of GH11 xylanases.

Fig. 7.
figure 7

3D structural model of Xyl40-CD. Predicted N-glycosylation site Asn24 in magenta, two potential disulphide bond sites (Cys2-Cys28 and Cys126-Cys132) in stick form in yellow, five potential salt bridges in stick form with positive sites in blue and negative in red. Visualised with ChimeraX

The presence of salt bridges between negatively or positively charged amino acids in the xylanase structures was assessed using an ESBRI web server tool. The predicted number of salt bridges in xylanases TrXyn1, TrXyn2 and catalytic domain of the Xyl40 varied between 3 and 5 (Table 1). Interestingly, the location of the predicted salt bridges was very different in the three xylanases, except for the conserved salt bridge [33] buried in the enzyme core (Fig. 6). In Xyl40-CD, there were two distinct salt bridges, one between the α-helix and palm region and the other close to the C-terminal in the finger region (Fig. 7). The α-helix and N-terminal regions have been suggested to be the initiation sites for thermal unfolding for GH11 xylanases based on molecular dynamic and mutagenesis studies [34]. Stabilisation of these regions with salt bridges can potentially increase the thermal stability of the enzyme. Although the stabilising effect of salt bridges has not been as clearly demonstrated as in the case of disulphide bridges, a higher number of salt bridges are often found in more thermostable enzymes than in their mesophilic counterparts [33].

Surface hydrophobicity and charge

The surface hydrophobicity of the xylanases was analysed with Rosetta protein modelling software. Here, each enzyme is given a hydrophobic patch (Hpatch) score, which is an indicator of larger uniform hydrophobic patches on the protein surface. The Hpatch scores for the xylanases varied between 0.64 and 3.04, with the highest individual score of 2.24 going to the CBM60 of Xyl40 (Table 1). For comparison, Hpatch scores for carbohydrate active enzymes have been previously reported to vary between 0.8 and 2.8 for xylanases, 7.0 and 16.0 for endo- and exocellulases, and 17.9 and 45.9 for β-glucosidases [9, 11]. The analysed xylanases had low surface hydrophobicity compared to cellulases and β-glucosidases. Previously, Sammond et al. [9] have observed a correlation between high Hpatch score and protein binding to lignin [9].

All the xylanases had a pI of over 5, indicating that they carried a net positive charge under the hydrolysis pH 5 (Table 1). The xylanases Xyl40, Xyl40-CD and TrXyn1, with pIs ranging from 8.4 to 9, carried a higher positive charge than TrXyn1, which had a pI of 5.5. The pI is an indication of the net surface charge, but the local charge distribution on the enzyme surface may vary more significantly. The local surface charges of enzymes was modelled using the Protein-Sol web tool [35]. The Protein-sol patches algorithm was developed to predict protein solubility in buffer but has been expanded to calculate hydrophobic patches and electric charge distribution on the protein surface at pH 6.3. Since the catalytic domains of Xyl40 and Xyl40-CD were nearly identical in terms of amino acid sequences, only Xyl40-CD was used in the analysis and compared to TrXyn1 and TrXyn2. As expected, the enzymes TrXyn2 and Xyl40 had more positively charged areas on the protein surface compared to TrXyn1. The active sites of all the three xylanases were negatively charged, but TrXyn1 had a significantly larger negatively charged area surrounding the active site cleft (Fig. 8a). Interestingly, all xylanases had a positively charged area in the back of the palm region, on the opposing side of the catalytic cleft (Fig. 8b).

Fig. 8
figure 8

Surface charge distribution on xylanases TrXyn1, TrXyn2 and Xyl40-CD calculated with Protein-Sol web tool. Charge at pH 6.3 a from the front of the active site palm region and b turned 180° from the back of the palm region. Cartoon image on Xyl40-CD to represent protein orientation. Negatively charged areas are shown in red and positively charged in blue

Amino acid composition and N-glycosylation sites

Comparison of the amino acid composition (Table 2) revealed that the Xyl40-CD had features similar to other known thermostable GH11 xylanases [33]. The catalytic domain of Xyl40-CD had higher threonine (Thr) to serine (Ser) ratio (1.31) than TrXyn1 (0.78) or TrXyn2 (0.73). High Thr/Ser ration is proposed to improve β-sheet formation leading to more rigid fold [33]. The Xyl40-CD also had relatively high arginine and tryptophan content and low valine content, all of which have been found to be characteristic for thermostable xylanases [33]. The catalytic domain of Xyl40 had three predicted N-glycosylation sites of which two were removed from the engineered Xyl40-CD produced in T. reesei. The remaining N-glycosylation site Asn24 was located in the finger region of Xyl40-CD (Figs. 6 and 7). Xyl40 was produced in E. coli and as such was not glycosylated. TrXyn1 had no predicted N-glycosylation sites and TrXyn2 had three potential N-glycosylation sites (Fig. 6).

Table 2 Amino acid composition and Thr/Ser ratio of xylanases TrXyn1, TrXyn2 and Xyl40-CD

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (