Introduction

Xylan is one of the major structural polysaccharides in plant cells and the second most abundant polysaccharide after cellulose in nature. Given its structural complexity, enzymatic hydrolysis of xylan needs cooperative action of a variety of enzymes (Moreira and Filho 2016). Among these xylanolytic enzymes, endo-xylanase (EC 3.2.1.8), which is normally termed xylanase, is particularly important because it catalyzes cleavage of the internal β-1,4-glycosidic linkages of xylan backbone, thereby generating xylooligosaccharides (XOS) with low degree of polymerization (DP) (Moreira and Filho 2016). On the basis of amino acid sequence and three-dimensional structure, majority of endo-xylanases are categorized as glycosyl hydrolase family 10 and 11 (GH10 and GH11) (http://www.cazy.org/). To date, a large number of endo-xylanases have been characterized and applied in various fields of industry, such as pulp biobleaching (Walia et al. 2017), waste paper deinking (Dhiman et al. 2014), animal feed production (Harris and Ramalingam 2010), bread making (Butt et al. 2008), biofuel production (Bhalla et al. 2015), and prebiotic production (Jain et al. 2015).

Generally, an enzyme performance is highly sensitive to reaction temperature. Biotechnological application of xylanase (similar to many other enzymes) is normally hindered by its temperature-dependent catalytic property. Therefore, in recent years, the use of extremophilic xylanases to satisfy the industrial demands for xylanase that can catalyze under harsh conditions has been attracting interest. Currently, a large number of thermophilic xylanases have been exploited (Kumar et al. 2018). By contrast, less attention is given to psychrophilic or cold-active xylanases, despite their considerable potential applications in many industrial processes. Cold-active xylanase can be especially used in industrial processes where undesirable chemical reactions occur at high temperature (Santiago et al. 2016), and they are particularly suitable in food and feed industry applications (Cavicchioli et al. 2002, 2011). An example of cold-active xylanase application is its usage for improving dough stability and flexibility and for increasing bread volume and crumb structure (Collins et al. 2006; Butt et al. 2008; Dornez et al. 2011). In addition, cold-active xylanase possesses the general advantage of a cold-active enzyme for application, such as energy-saving characteristic and inherently broad substrate specificity relative to its thermophilic counterparts (Santiago et al. 2016).

To date, the number of cold-active xylanases is largely limited (Santiago et al. 2016). No more than 20 cold-active xylanolytic enzymes have been heterogeneously expressed and characterized (Santiago et al. 2016). These cold-active xylanases were generally obtained from cultured psychrophilic microorganisms or metagenomes (Vester et al. 2015). Mining of enzyme from unexplored genome databases (such as GenBank) can be an alternative method to characterize novel enzymes (Lauro et al. 2010; Gong et al. 2013). Therefore, we provided special attention to sequenced genomes of marine microorganisms that are considered reservoirs of novel and extremophilic enzymes (Littlechild 2015). Luteimonas abyssi, XH031T, a novel species of Luteimonas, was isolated from deep-sea sediment of the South Pacific Gyre (genome accession number in GenBank: NZ_KQ759763) (Fan et al. 2014). According to genome annotation, a xylanase (GenBank accession numbers: WP_082672697.1, termed as laXynA) is presumably produced by the strain. In the present study, the amino acid sequence and structural features of putative xylanase were analyzed. The biochemical properties, kinetic parameters, and cleavage patterns were characterized. Our results indicated that laXynA is a typical cold-active endo-xylanase belonging to the GH10 family. Additionally, extreme halophilicity is a prominent characteristic of laXynA.

Materials and methods

Bacterial strains, gene, and chemicals

Escherichia coli DH5α (Invitrogen) and BL21 Rosetta (DE3) (Novagen) were used as host cells for gene cloning and heterologous expression, respectively. DNA fragments encoding full open reading frame (ORF) of laXynA on the pUC57-simple vector (pUC57-laXynA, restriction sites EcoR1 and Not1 were added to the flanking 5′- and 3′-ends of the ORF, respectively) were synthesized by GENEWIZ (Suzhou, China). Xylose (X1) was purchased from Merck Life Science (Shanghai). Xylobiose (X2), xylotriose (X3), xylotetraose (X4), xylopentaose (X5), xylohexaose (X6), and beechwood xylan were purchased from Megazyme (Ireland).

Sequence and structural analyses

The amino acid sequence of laXynA was compared with deposited sequences in the protein database (NCBI) using the BLASTp algorithm on the BLAST server (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Signal peptide prediction was conducted using SignalP 4.1 (http://www.cbs.dtu.dk/services/SignalP/). A three-dimensional model of laXynA was generated by homology modeling using SWISS-MODEL (https://swissmodel.expasy.org/interactive). The credibility of the structural model was evaluated by MolProbity (http://molprobity.biochem.duke.edu/) (Chen et al. 2010). The secondary structure of laXynA was assigned using the DSSP program (http://swift.cmbi.ru.nl/gv/dssp/) (Kabsch and Sander 1983). Structure-based sequence alignment was performed on the T-Coffee server (http://tcoffee.crg.cat/apps/tcoffee/do:regular) and ESPript3.0 (http://espript.ibcp.fr/ESPript/cgi-bin/ESPript.cgi). Structural figures were created using PyMOL (Schrödinger).

Expression and purification of laXynA

pUC57-laXynA plasmids (amplified in E. coli DH5α) were double-digested with EcoR1 and Not1. Gel purified DNA fragments encoding laXynA from the digestion were inserted into pET-28a expression vector. The recombinant plasmid, named pET-28a-laXynA, was verified by DNA sequencing and transformed into E. coli BL21 Rosetta (DE3) competent cells. A 5 ml overnight culture of transformed E. coli BL21 Rosetta (DE3) in Luria–Bertani medium was inoculated into 500 ml ZYM 5052 autoinduction medium in a 2-l flask containing antibiotic kanamycin and chloramphenicol with the final concentration of 100 and 34 µg/ml, respectively. The culture grew at 37 °C until the optical density at 600 nm approximately reached 1. Subsequently, the culture temperature was changed to 20 °C and maintained for 16 h. The cells were harvested by spinning the tubes at 4 °C. The cell pellet resuspended in lysis buffer (500 mM NaCl, 20 mM imidazole, 20 mM Na2HPO4, pH 7.4) was broken by a high-pressure homogenizer. The supernatant of the cell lysate was filtered over a 0.45-µm filter and applied to a 5-ml HisTrap column (GE Healthcare). After equilibrating the column with the lysis buffer on an Äkta prime purifier (GE Healthcare), the recombinant protein was eluted with an elution buffer (500 mM NaCl, 500 mM imidazole, 20 mM Na2HPO4, pH 7.4) in a gradient elution mode. Elution fractions were collected and pooled as indicated by sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS–PAGE). A 10 kDa cut-off concentrator (Millipore) was used to exchange the buffer of the eluted protein for a phosphate buffer (10 mM Na2HPO4, pH 7.4) and concentrate the yielded enzyme properly. The final isolated enzyme was stored in 1-ml aliquots at − 80 °C. The protein concentration was determined by a Bradford protein assay kit (Beyotime, China).

Enzymatic assay and kinetic parameter determination

Xylanase activity was quantified by using the 3,5-dinitrosalicylicacid (DNS) method as described by Bailey et al. (1992). A standard reaction mixture consisted of 0.5 ml of diluted enzyme (approximate 1:20), 1 ml of buffer, and 0.5 ml of beechwood xylan (10 mg/ml). After incubation in a water bath for 15 min, the reaction was stopped with 2 ml of DNS reagent. Afterward, the mixture was heated at 100 °C for 5 min and cooled to room temperature. The absorption of the reaction mixture at 520 nm was measured on a UV/Vis spectrophotometer. Xylose standard was used to plot the calibration curve. One unit (U) of xylanase activity was defined as the amount of enzyme that can release 1 µmol of reducing sugars from beechwood xylan equivalent to xylose per minute. The specific activity was presented as units per milligram of protein. All activity assays were conducted three times.

The Michaelis–Menten constant (Km), maximum rate of reaction (Vmax), and turnover number (kcat) for the isolated laXynA were determined using beechwood xylan as the substrate at concentrations varying from 0.5 to 6 mg/ml in McIlvaine buffer (pH 7) for 15 min. The enzyme kinetic constant values were calculated by constructing double-reciprocal plots.

Effects of temperature and pH on the activity and stability of laXynA

The optimal pH of laXynA was determined at 30 °C in buffers at pH 3–11. pH stability was monitored by assessing the residual activity at pH 7 after incubating the enzyme in buffers with various pH at 4 °C for 2 h. The pH gradient buffers used for enzymatic assays included McIlvaine buffer for pH 3–8 and glycine-NaOH buffer (50 mM) for pH 9–11. The optimal temperature was estimated over the range of 0–90 °C in the McIlvaine buffer of pH 7. The temperature stability of laXynA was evaluated by incubating the enzyme for 2 h at 10–90 °C, cooling on ice, and measuring the residual activity at 30 °C.

Effects of metal ions and NaCl on activity

The effects of different metal ions on the activity of laXynA were investigated by adding 5 mM (final concentration) FeSO4, MnSO4, CaCl2, CoCl2, or MgSO4 to the standard reaction mixture. The effect of salt concentration on xylanase activity was evaluated with 0–4 M NaCl in the reaction mixture at pH 7 and 40 °C.

Analysis of hydrolytic product

Hydrolysis products of beechwood xylan and XOS with backbone length of 2–6 (X2–X6) by laXynA were investigated by thin layer chromatography (TLC). A reaction mixture consisting of 12.5 µl of appropriately diluted enzyme, 12.5 µl of substrate, and 25 µl of McIlvaine buffer (pH 7) was incubated at 40 °C for 1 h in a thermal cycler and the reaction was stopped by heating at 95 °C for 10 min. An appropriate volume of the sample was spotted onto a Silica Gel plate (Merck, Germany). For each of the substrates, a reaction with heat-inactivated laXynA (95 °C, 10 min) was used as negative control. The TLC plate was developed with n-butanol/islpropanol/acetic acid/water (7:5:2:4, v/v). Spots were visualized by spraying the plate with a solution of 3% (w/v) urea in n-butanol/ethanol/water/phosphoric acid (80:8:5:7, v/v) and heated at 115 °C for 20 min in an oven.

Results

Recombinant laXynA production and purification

According to the genome annotation of L. abyssi, XH031T, the ORF of laXynA is 1098 bp in length, and it encodes a putative xylanase with 365 residues. The theoretical molecular weight and isoelectric point are 39590.12 Da and 4.53, respectively. Recombinant laXynA was produced in E. coli and purified using His tag affinity chromatography. SDS–PAGE analysis of the fractions eluted by imidazole indicated that the recombinant laXynA displayed a band of approximately 45 kDa, which is larger than the calculated molecular mass of 39590.12 Da because of the additional 34 amino acids in the N-terminus of laXynA derived from pET-28a vector (Fig. 1). The relatively pure fractions were pooled and the protein was applied in the enzymatic assays. The protein concentration of the final isolated recombinant laXynA was about 1 mg/ml.

Fig. 1
figure 1

SDS–PAGE analysis of the isolated laXynA. Lane 1, protein marker; lane 2, final isolated recombinant laXynA after Ni-affinity chromatography

Full size image

Biochemical characterization of recombinant laXynA

With beechwood xylan as the substrate, laXynA showed the highest activity at 30–40 °C, peaking at 30 °C (Fig. 2a). The enzyme retained more than 64, 55, and 29% of the maximum activity at 10, 5, and 0 °C, respectively. The effect of pH on the activity of laXynA was investigated at a pH range of 3–11. As shown in Fig. 2b, laXynA is a neutral xylanase (optimal pH of 6.5). However, the enzyme was active across a broad pH range as indicated by the retention of around 40% activity between pH 6 and 9.

Fig. 2
figure 2

Effects of pH and temperature on laXynA activity and stability. a Effect of temperature on the activity of laXynA. b Effect of pH on the activity of laXynA. c Temperature stability of laXynA. The 100% activity was obtained from the enzyme under normal storage temperature (4 °C). d pH stability of laXynA. The 100% specific activity is approximately 30 U/mg. Data are means ± standard deviations from three repeats. The figure was prepared using GraphPad Prism

Full size image

LaXynA displayed low thermostability as demonstrated by a loss of more than 75% activity after 2 h incubation at 40 °C and almost complete loss of activity after 2 h incubation at 90 °C (Fig. 2c). LaXynA was stable at pH ranging from 6 to 11. It retained more than 90% of the maximum activity after 2 h incubation at this pH range at 4 °C (Fig. 2d).

Among the tested metal ions, Mg2+, Ca2+, and Fe2+ show no evident effect on the activity of laXynA at a concentration of 5 mM; Mn2+ showed negative effect of 0.25-fold whereas Co2+ enhanced the activity slightly (Fig. 3a).

Fig. 3
figure 3

Effects of metal ions and salt on laXynA activity. a Effects of metal ions. b Effects of NaCl concentration. The 100% specific activity is approximately 30 U/mg. The figure was prepared using GraphPad Prism

Full size image

NaCl (0.5–3 M) exerted positive effect on the activity of laXynA (Fig. 3b). When 0.5 M NaCl was added into the reaction mixture, laXynA activity increased by about fourfold relative to the reaction condition with no NaCl added. The degree of improvement decreased with the increased NaCl concentration (Fig. 3b). Nevertheless, the positive effect was present until the NaCl concentration reached 4 M.

Kinetic study on laXynA

A linear relationship between xylanase activity and reaction time was observed from 0 to 15 min at a beechwood xylan concentration of 10 mg/ml (data not shown). Therefore, a reaction time of 15 min was adopted for kinetic assays. Kinetic values were determined at three different temperatures. The maximum velocity of the isolated laXynA at 10 °C was 9 U/mg. This value significantly increased to 18 and 49 U/mg when the reaction temperature was changed to 25 and 40 °C, respectively (Table 1). The Km values for laXynA determined at 10 and 40 °C were 3.2 mg/ml, and the value decreased to 1.6 mg/ml at 25 °C. The kcat values increased with the increase in tested temperatures. The kcat/Km values at 25 and 40 °C were 4.1 and 5.6 times that at 10 °C, respectively (Table 1).

Table 1 Kinetic values of laXynA toward beechwood xylan
Full size table

Hydrolytic property of laXynA

Hydrolytic property of laXynA was investigated with beechwood xylan and XOS (X2–X6) as the substrates. The main hydrolytic products of beechwood xylan by laXynA were X2, X4, and a few of X1 after 1 h reaction at 40 °C (Fig. 4). No xylosidase activity was observed for laXynA. X3 was hydrolyzed to X2 and X1 (Fig. 4). The majority of X4 was hydrolyzed to X2. The hydrolytic products of X5 were dominated with X2, X1, X3, and X4. A similar product profile for X5 was observed for X6 under the same reaction condition (Fig. 4).

Fig. 4
figure 4

Hydrolytic products of beechwood xylan and XOS by TLC. Lanes 1 and 2, samples of beechwood xylan incubated with heat-inactivated and untreated laXynA, respectively; lane 3, sample of xylose incubated with untreated laXynA; lanes 4 and 5, samples of xylobiose (X2) incubated with heat-inactivated and untreated laXynA, respectively; lanes 6 and 7, samples of xylotriose (X3) incubated with heat-inactivated and untreated laXynA, respectively; lanes 8 and 9, samples of xylotetraose (X4) incubated with heat-inactivated and untreated laXynA, respectively; lanes 10 and 11, samples of xylopentaose (X5) incubated with heat-inactivated and untreated laXynA, respectively; lanes 12 and 13, samples of xylohexaose (X6) incubated with heat-inactivated and untreated laXynA, respectively. Migration positions of XOS (X1–X6) are indicated by height of XOS abbreviations

Full size image

Sequence analysis and structural modeling of laXynA

Signal peptide analysis revealed the presence of a signal peptide with cleavage site between Gly24 and Asp25 in laXynA. The cleavage will produce a mature protein of 341 amino acids. BLASTp search against GenBank protein database (mature laXynA as the query sequence) demonstrated that laXynA belongs to the GH10 family and only consists of a catalytic domain. The search also showed that laXynA shared the highest identity (68%) with endo-xylanase from Xanthomonas sp. (XynB).

A structural model of laXynA containing residues 52–307 was built by the SWISS-MODEL server using crystal structure of XynB from Xanthomonas species (Protein Data Bank entry: 4PN2) (Santos et al. 2014) as the template (Fig. 5). The model showed a small MolProbity core of 1.25 (99th percentile), which indicated good quality of the overall structure. The predicted structure of laXynA exhibited a featured (β/α)8-barrel fold of GH10 xylanases (Fig. 5). The secondary structure of laXynA was assigned by DSSP on the basis of the three-dimensional model. A structure-based alignment of catalytic domains indicated that laXynA was not well-conserved among the characterized cold-active GH10 xylanases (Fig. 6). LaXynA shared 20–26% sequence identity with other cold-active GH10 xylanases. The most notable difference between laXynA and other cold-active GH10 xylanases was located in the loop regions, particularly the loops connecting β1 and β2, and β7 and α7 (Fig. 6).

Fig. 5
figure 5

Overall structural model (left) and detail of substrate-binding cleft (right) of laXynA. The carbon atoms of catalytic residues in the overall structure are shown in green. The approximate positions of putative subsites are indicated by large black numbers. The carbon atoms of the amino acids constituting each subsites are shown in different colors, with yellow, green, magenta, and brown for − 2, − 1, + 1, and + 2 subsites, respectively. The carbon atoms of 310-helix where Phe251 is located is shown in cyan. The figure was prepared using PyMOL

Full size image
Fig. 6
figure 6

Structure-based alignment of the catalytic domain of laXynA with other characterized cold-active GH10 xylanases. The accession numbers of the xylanase sequences are as follows: WP_082672697.1, the xylanase in this study; ADN44261.1, XynGR40 from goat rumen contents; Q21DH6, Xyn10C from Saccharophagus degradans 2–40; AGC01501.1, XynAGN16 from Arthrobacter sp. GN16; WP_013072455.1, XynA from Zunongwangia profunda; WP_012025843.1, Xyn10A from Flavobacterium johnsoniae; AFE82288.1, XynAHJ2 from Bacillus sp. HJ2; AEB69780.1, XynA from Sorangium cellulosum So9733-1; AGA16736.1, Xyn10A from Bacillus sp. SN5; ACN76857.1, XynA from Glaciecola mesophila KMM 241; AAY98787.1, Xyn10 from Flavobacterium sp. Conserved and identical amino acids are highlighted in yellow and red, respectively. Acid/base catalyst and catalytic nucleophile are indicated with red asterisks (*); the putative residues involving substrate-binding are indicated with green asterisks (*). The secondary structural elements are displayed above the corresponding sequences. Gaps are indicated by dashes. The figure was generated by ESPript3.0

Full size image

The putative catalytic acid–base Glu164 and nucleophile Glu280 were situated at the end of β-strand 4 and 7, respectively (Figs. 5, 6), consistent with the active-site topology of GH10 xylanase (Pollet et al. 2010). Amino acids, either participating in the catalytic reaction or substrate binding in the glycon region of substrate-binding cleft (subsites − 2 and − 1), were highly conserved among cold-active GH10 xylanases (Fig. 6). In the aglycon region, two aromatic amino acids, namely, Tyr216 and Phe251, can be assumed to be the binding sites for + 1 and + 2 xylose residues, respectively. Tyr216 was highly conserved among GH10 xylanases, and Phe251 situated at the bottom of active cleft was only found in laXynA (Figs. 5, 6). A 310-helix, where Tyr216 was located, was also distinct for laXynA (Figs. 5, 6).

Discussion

Xylanase, in particular endo-xylanase, plays the most important role in xylan degradation in nature. Xylanolytic enzymes have also shown considerable potential in applications, such as animal feed, food processing, and textile (Dhiman et al. 2008; Cavicchioli et al. 2011). Exploring novel xylanases and understanding their enzymatic properties are critical for their efficient and effective usage. L. abyssi XH031T is a recently isolated marine bacterium from deep-sea sediment of the South Pacific (Fan et al. 2014). The organism extensively produces cold-active enzymes, including polysaccharide hydrolyases, such as amylase, cellulose, and chitinase (Zhang et al. 2015). In the present study, laXynA, a predicted xylanase encoded in the genome of L. abyssi XH031T, was characterized. Amino acid sequence alignment and three-dimensional structure modeling revealed that laXynA belongs to the GH10 family. It showed the highest identity of 68% with endo-xylanase from Xanthomonas sp. but relatively low identity with other characterized cold-active GH10 xylanases. The optimum temperature of laXynA was approximately 30 °C, and the xylanase retained important percentage of optimal activity (more than 55%) at 5 °C. Meanwhile, laXynA was found to be thermolabile. These properties of laXynA are similar to those of cold-active xylanases characterized from various microbes (Santiago et al. 2016). These results indicated that laXynA is a new member of cold-active xylanase (Table 2).

Table 2 Molecular and biochemical characteristics of cold-active xylanases
Full size table

An excellent feature of laXynA is its extremely halophilicity. The activity of the enzyme was considerably enhanced by NaCl at 0.5–3 M (Fig. 3b). The highest activity of laXynA was observed in the presence of about 0.5 M NaCl, which is close to the salt concentration of seawater (about 0.6 M NaCl); this result indicated that laXynA likely functions in seawater. To date, a few halophilic xylanases have been reported. Among them, Xyn10C from S. degradans 2–40 (cold-active) (Ko et al. 2016), XynA from Z. profunda (cold-active) (Liu et al. 2014), and XynFCB from Thermoanaerobacterium saccharolyticum NTOU1 (thermostable) (Hung et al. 2011a) display their highest activity at 2–3 M NaCl; several halophilic xylanases, including XynA from G. mesophila KMM 241 (cold-active) (Guo et al. 2009), Xyn10A from Bacillus sp. SN5 (cold-active) (Bai et al. 2012), and XynA from T. saccharolyticum NTOU1 (thermostable) (Hung et al. 2011b), just like observation on laXynA, display the highest activity at about 0.5 M NaCl. For all of these halophilic xylanases, the degrees of activity enhancement by NaCl were 1.2–1.9-fold (Liu et al. 2014). By contrast, the activity enhancement by NaCl for laXynA is the most significant (approximately fourfold). Notably, laXynA and all the other halophilic xylanases mentioned in this study are produced by microorganisms inhabiting marine environments or soda lakes, which indicated that the halophilic property was derived from environment-driven adaption. General molecular mechanism of protein adaptation to high salinity based on bioinformatic analysis may be attributed to the excess surface-exposed acid over basic amino acids (DasSarma and DasSarma 2015). LaXynA is a good halophilic protein model to verify the validity of the proposed mechanism. More negatively charged (52, Glu + Asp) residues than the positively charged (23, Arg + Lys) ones were present on laXynA, and all of the acidic amino acids were located on solvent-accessible surface. Our modeled structure revealed that the overall surface of laXynA was largely covered with negative electrostatic potential (Fig. 7).

Fig. 7
figure 7

Surface electrostatic of laXynA. a Surface presentation of active-site cleft side of laXynA. b 90° rotated view relative to (a) c 90° rotated view relative to (b) d 90° rotated view relative to (c) The surface electrostatic is colored from blue (positive potential) to red (negative potential). The figure was prepared using PyMOL

Full size image

In principle, cold-active enzymes exhibit higher Km value than those of their thermostable counterparts (Georlette et al. 2004). The Km values of thermostable xylanases are normally in the range of 0.1–5 mg/ml (Basit et al. 2018). Nevertheless, most of characterized cold-active xylanases show relatively low Km values (< 5 mg/ml) (Table 2). LaXynA, as a cold-active xylanase, showed relatively low Km value of 1.6 mg/ml determined at 25 °C toward beechwood xylan. Under the similar experimental condition, low Km value was also found for cold-active xylanases, such as XynGR40 from goat rumen contents (1.8 mg/ml) (Wang et al. 2011), XynA from G. mesophila KMM 241 (1.22 mg/ml) (Guo et al. 2009), and Xyn10 from Flavobacterium sp. (1.8 mg/ml) (Lee et al. 2006a) (Table 2). Low Km value (< 1 mg/ml) was also observed for cold-active xylanase; Xyn10A from Bacillus sp. SN5 showed a Km value of 0.5 mg/ml (Bai et al. 2012). Unusually, laXynA showed a twofold higher Km value determined at 10 °C (3.2 mg/ml) and 40 °C (3.2 mg/ml) than that obtained at 25 °C (1.6 mg/ml). This result suggested that the affinity between the enzyme and xylan (or several XOS with particular lengths) reached the optimum at around 25 °C. Temperature-induced structural changes may account for this particular substrate-binding behavior. It has been shown that a temperature-dependent structural modification on substrate-binding cleft of xylanase 10B from Thermotoga petrophila changed the XOS binding at the aglycone subsites (Santos et al. 2010). Several similar studies on cold-active xylanases also showed the varied Km values at different measuring. However, the of Km values changed in different tendencies. XynAGN16L (1.46 and 2.21 mg/ml towards beechwood xylan at 10 and 30 °C, respectively) (Zhou et al. 2015a) and XynA from G. mesophila KMM 241 (0.78 and 1.22 mg/ml toward beechwood xylan at 4 and 30 °C, respectively) (Guo et al. 2009) showed an increased Km value at higher temperature; by contrast, the Km values of xynGR40 (2.2 and 1.8 mg/ml toward beechwood xylan at 10 and 30 °C, respectively) (Wang et al. 2011) and Xyn10 from F. johnsoniae (10.6 and 8.4 mg/ml towards birchwood xylan at 10 and 30 °C, respectively) (Chen et al. 2013) decreased as a result of rising temperature.

The kcat of laXynA increased with the increase in tested temperatures (10, 25, and 40 °C) (Table 1). This phenomenon was also observed for XynA from G. mesophila KMM 241 (Guo et al. 2009), XynGR40 (Wang et al. 2011), and XynAGN16L (Zhou et al. 2015a). LaXynA displayed an intermediate kcat value of 35.6 s−1 at 40 °C. Actually, the kcat values of cold-active xylanases toward xylan determined at similar temperature covered a considerably broad range (6.84–1247 s−1). The kcat value of laXynA is much higher than that of the cold-active xylanases, such as Xyn10 from S. cellulosum So9733-1 (6.84 s−1 toward beechwood xylan at 30 °C) (Wang et al. 2012) and F. johnsoniae. (10.7 s−1 toward beechwood xylan at 35 °C) (Chen et al. 2013). The kcat values are higher but similar to that of laXynA for XynAGN16 (49.2 s−1 toward beechwood xylan at 45 °C) (Zhou et al. 2015b) and XynA from Z. profunda (47.3 s−1 toward beechwood xylan at 30 °C) (Liu et al. 2014). Several cold-active xylanases, mainly belonging to the GH8 family, such as XynA from G. mesophila KMM 241 (609 s−1 toward beechwood xylan at 35 °C) (Guo et al. 2009) and xylanases from Pseudoalteromonas haloplanktis (1247 s−1 toward birchwood xylan at 25 °C) (Collins et al. 2002), show much larger kcat values than that of laXynA.

In principle, to maintain activity at low temperature, an increase in kcat is necessary for cold-active enzymes (Georlette et al. 2004; Santiago et al. 2016). Therefore, among the characterized cold-active xylanases, the xylanase from P. haloplanktis most satisfactorily satisfies the evolutionary principle of a cold-adapted enzyme, as demonstrated by its considerably large Km and kcat values. LaXynA and most of other cold-active xylanases exhibit a relatively small Km and kcat values, which implied that their adaption to cold environment is not sufficient or in a unusual evolutionary road.

The hydrolytic product of xylan produced by most of GH10 endo-xylanases is a mixture of XOS (low DP less than five) with a small percentage of xylose (Linares-Pasten et al. 2018). LaXynA degraded beechwood xylan to X1, X2, and X4 after 1 h incubation at 40 °C, which suggested its endo-xylanase property (Fig. 4). Structural data combined with kinetic activity on XOS have demonstrated that generally, 4–7 subsites are present in the active-site cleft of GH10 endo-xylanases (Pollet et al. 2010), and the highly conserved − 2, − 1, and + 1 subsites play a crucial role in glycosidic bond cleavage (Ducros et al. 2000). In accordance with this canonical property of GH10 endo-xylanase, subsites − 2 to + 1 in the active cleft of laXynA can execute the catalytic cleavage of X3 (Fig. 4). The dominant hydrolytic product of X4 by laXynA was X2, which demonstrated that + 2 to − 2 binding and cleavage was preferred by laXynA (Fig. 4). This preferred cleavage mode can be explained by the presence of Phe251 in the + 2 subsite of laXynA. This aromatic residue likely formed hydrophobic stacking interaction with + 2 xylose residue, which is consistent with the dominant stacking interactions in the aglycon region of GH10 endo-xylanases (Zolotnitsky et al. 2004). In contrast, cold-active endo-xylanases such as XynA (GH10) and XynB (GH8) from G. mesophila KMM 241 were unable to hydrolyze X3 and X4, thereby suggesting the requirement of at least four and five subsites for effective cleavage (Guo et al. 2009, 2013). At least four subsites are also required for cleavage for cold-active Xyn8 (GH8) from an environmental genomic DNA library (Lee et al. 2006b).

Conclusions

We expressed and characterized a new member of cold-active endo-xylanases (laXynA) from marine microorganism. Several characteristics of laXynA are attractive for industry applications. LaXynA degraded xylan to small XOS with a small proportion of xylose, which suggested that it can be used for prebiotic XOS production. LaXynA exhibited excellent activity in high-salinity environment, which implied a potential usage in biotechnological processes of sea food and saline food. However, similar to most of other cold-active xylanases, the catalytic efficiency of laXynA is not as high as that of their mesophilic counterparts (Georlette et al. 2004). Therefore, to search for highly active cold-active xylanases or modify the current cold-active xylanases by protein engineering may be significant. This study reported the over-expression and biochemical characterization of a cold-active xylanase from a Luteimonas species for the first time. This study can also help in elucidating hemicellulose utilization in deep-sea sediment of the South Pacific.