Identification and Characterization of a Novel, Cold-Adapted d-Xylobiose- and d-Xylose-Releasing Endo-β-1,4-Xylanase from an Antarctic Soil Bacterium, Duganella sp. PAMC 27433

Endo-β-1,4-xylanase is a key enzyme in the degradation of β-1,4-d-xylan polysaccharides through hydrolysis. A glycoside hydrolase family 10 (GH10) endo-β-1,4-xylanase (XylR) from Duganella sp. PAMC 27433, an Antarctic soil bacterium, was identified and functionally characterized. The XylR gene (1122-bp) encoded an acidic protein containing a single catalytic GH10 domain that was 86% identical to that of an uncultured bacterium BLR13 endo-β-1,4-xylanase (ACN58881). The recombinant enzyme (rXylR: 42.0 kDa) showed the highest beechwood xylan-degrading activity at pH 5.5 and 40 °C, and displayed 12% of its maximum activity even at 4 °C. rXylR was not only almost completely inhibited by 5 mM N-bromosuccinimide or metal ions (each 1 mM) including Hg2+, Ca2+, or Cu2+ but also significantly suppressed by 1 mM Ni2+, Zn2+, or Fe2+. However, its enzyme activity was upregulated (>1.4-fold) in the presence of 0.5% Triton X-100 or Tween 80. The specific activities of rXylR toward beechwood xylan, birchwood xylan, oat spelts xylan, and p-nitrophenyl-β-d-cellobioside were 274.7, 103.2, 35.6, and 365.1 U/mg, respectively. Enzymatic hydrolysis of birchwood xylan and d-xylooligosaccharides yielded d-xylose and d-xylobiose as the end products. The results of the present study suggest that rXylR is a novel cold-adapted d-xylobiose- and d-xylose-releasing endo-β-1,4-xylanase.

Biocatalysts including xylanolytic GH enzymes are an attractive option for sustainable and clean manufacturing technologies owing to their numerous advantages; they are biodegradable, non-toxic, have low energy consumption, and outstanding catalytic properties. Endo-β-1,4-xylanases (EC 3.2.1.8) are the key enzymes responsible for the deconstruction of β-1,4-D-xylosic materials [3] and thus, have drawn much attention as promising

Isolation and Identification of a Xylanolytic Microorganism
To isolate xylanolytic Antarctic microorganisms, a soil sample was collected from King George Island, Antarctica (62 • 13 50.8 S; 58 • 46 30.9 W). For cultivation of bacterial isolates, 0.5 g of the soil sample was serially diluted up to 10 −2 using sterile 0.85% NaCl, after which 100 µL aliquots of the diluted sample suspensions were spread on 0.01× nutrient agar medium and incubated at 20 • C for 14 d. Of the heterotrophic bacteria grown on the solid medium, strain PAMC 27433 was purely isolated by culturing it on Reasoner's 2A (R2A) agar medium three or more times. Extraction of genomic DNA from the isolate was performed by using the Mini Tissue DNA kit (Cosmo Genetech Co. Ltd., Seoul, Korea) in accordance with the manufacturer's instructions. The 16S rRNA gene of strain PAMC 27433 was amplified with two universal primers (27F and 1492R), after which the resulting PCR (polymerase chain reaction) products were purified using LaboPass PCR purification kit (Cosmo Genetech Co. Ltd., Seoul, Korea) and sequenced with the same primers used for amplification. The sequence of its 16S rRNA gene was compared with those of type strains available in the EzBioCloud database (ChunLab Inc., Seoul, Korea) to find closely related species.

Molecular Cloning of the Endo-β-1,4-Xylanase (XylR) Gene
PCR amplification of the gene coding for mature recombinant XylR proteins was performed employing the two gene-specific oligonucleotides XylR-F (5 -CATATGGCGGAAGA CACGCCGGAACC-3 ) and XylR-R (5 -AAGCTTTCACGCCCGCGGCGG-3 ). The designed upstream (XylR-F) and downstream (XylR-R) primers contained an NdeI restriction site and a HindIII restriction site, respectively. The genomic DNA of Duganella sp. PAMC 27433, which was extracted using a NucleoSpin gDNA Clean-up (Macherey-Nagel), was used as a template and the reaction was carried out using a PCR thermal cycler (TaKaRa), as previously described [21]. Briefly, the PCR mixture (50 µL) contained 2.5 units of FastStart Taq DNA polymerase (Roche), 2.5 mM of each dNTP, 100 pmol of each primer, 20 ng of template DNA, and a PCR buffer. The initial template denaturation was conducted for 4 min at 95 • C, followed by 35 cycles of 30 s at 95 • C, 30 s at 60.5 • C, and 1 min 10 s at 72 • C. After separating the amplified PCR products by electrophoresis on a 1.2% agarose gel, the desired gene products were isolated using a NucleoSpin Gel and PCR Clean-up (Macherey-Nagel, Düren, Germany). The purified gene products (1065-bp) were then inserted into a pGEM-T easy vector (Promega, Madison, WI, USA). The pGEM-T easy/xylR vectors, which were transformed into Escherichia coli DH5α competent cells, were isolated from the recombinant cells and subsequently cleaved with restriction endonucleases NdeI and HindIII to yield xylR fragments with the corresponding cohesive ends. After purifying the xylR fragments, they were cloned into a pET-28a(+) vector (Novagen, Darmstadt, Germany) with the same cohesive ends and then the constructed pET-28a(+)/xylR vectors were introduced into E. coli BL21 competent cells.

Overproduction and Purification of Recombinant Proteins
Overproduction of the N-terminal (His) 6 -tagged XylR (rXylR) was performed by culturing the recombinant E. coli BL21 cells harboring pET-28a(+)/xylR in a 5-L baffled flask, which included 1 L of Luria-Bertani broth (Difco) plus 25 mg/L of kanamycin, in a rotary shaker (150 rpm) for 10 h at 30 • C. The induction of the XylR gene was performed by adding 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) after the absorbance of the culture at 600 nm reached 0.4-0.5. Following cultivation, the rXylR-expressing cells were harvested from the culture broth by centrifugation (7000× g) for 15 min at 4 • C and then stored at −20 • C for 3 h. To purify rXylR, the frozen cells were completely resuspended in a binding buffer [22] and disrupted by sonication. In this study, rXylR proteins were only produced in an active form and thus, the soluble fraction exhibiting high endo-β-1,4xylanase activity toward beechwood xylan was selectively recovered by centrifugation (12,000× g) for 20 min at 4 • C. The active rXylR proteins were purified by affinity column chromatography employing a HisTrap HP (GE Healthcare, Uppsala, Sweden) (5 mL) column connected to fast protein liquid chromatography system (Amersham Pharmacia Biotech, Uppsala, Sweden). Elution of the recombinant enzymes from the column was done using a linear gradient of 20-500 mM imidazole at a flow rate of 2 mL/min. The active fractions containing rXylR were selectively combined and then desalted with a HiPrep 26/10 desalting column (GE Healthcare, Uppsala, Sweden) employing 50 mM sodium phosphate buffer (pH 6.0) as the mobile phase. The desalted fractions including purified rXylR proteins were collected, pooled, and used for further analysis.

Analysis of Proteins
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis of the denatured rXylR proteins in a 12.0% gel was conducted to determine its relative molecular mass. After completion of the electrophoresis, visualization of the proteins separated by SDS-PAGE was performed by staining the gel with Coomassie Brilliant Blue R-250. Quantitative analysis of the protein concentrations was performed by Bradford assay employing bovine serum albumin as a standard [21].

Enzyme Assays
Endo-β-1,4-xylanase activity was assayed by quantitatively measuring the amount of D-xylose and its reducing oligosaccharides produced from biocatalytic degradation of beechwood xylan employing 3,5-dinitrosalicylic acid (DNS) reagent and D-xylose as a standard. The standard assay mixture (0.5 mL) contained 1.0% beechwood xylan or 5 mM pNP-sugar derivatives together with an appropriately diluted enzyme preparation (0.05 mL) in 50 mM sodium phosphate buffer (pH 5.5). The biocatalytic reaction was initiated by adding 0.05 mL of diluted enzyme solution to the reaction mixture and achieved at 40 • C for 10 min. After completion of the enzyme reaction, 0.75 mL of the DNS reagent was added to the reaction mixture, followed by boiling for 5 min to develop the red-brown color. One unit (U) of endo-β-1,4-xylanase activity toward β-1,4-D-xylans or pNP-sugar derivatives was defined as the amount of rXylR required to produce 1 µmol of reducing sugar or pNP, respectively, per min under standard assay conditions.

Effects of pH, Temperature, and Chemicals on the rXylR Activity
The optimum pH of rXylR was evaluated at pH values ranging from 3.5 to 10.5 at 40 • C for 10 min employing the following buffer systems (50 mM): sodium citrate (pH 3.5-5.5), sodium phosphate (pH 5.5-7.5), Tris-HCl (pH 7.5-9.0), and glycine-NaOH (pH 9.0-10.5). The pH stability of the enzyme was investigated by determining its remaining endo-β-1,4-xylanase activity after pre-incubation of 1 h at 4 • C in the aforementioned pH buffers. In this case, the enzyme reaction was initiated by adding 1.0% beechwood xylan to the reaction mixture. The effect of temperature on the endo-β-1,4-xylanase activity of rXylR was examined by reacting the enzyme with beechwood xylan at 4,8,18,30,35,40,45,50, and 55 • C for 10 min in 50 mM sodium phosphate buffer (pH 5.5). To assess the thermostability of rXylR, it was first pre-incubated at the corresponding temperature for 15, 30, and 60 min, respectively, in 50 mM sodium phosphate buffer (pH 5.5), then the substrate was added to the reaction mixture to initiate the biocatalytic reaction. The residual endo-β-1,4-xylanase activity of the enzyme was assayed for 10 min at 40 • C. The effects of divalent cations (each 1 mM) and chemical substances (each 5 mM or 0.5%) on the endo-β-1,4-xylanase activity of rXylR was determined after pre-incubation of rXylR at 4 • C for 10 min in the standard assay mixture that contained the chemical of interest.

Analysis of the Degradation Products
Biocatalytic degradation of birchwood xylan (2 mg/0.2 mL) or D-xylooligosaccharides (X 2 -X 6 , each 1 mg/0.2 mL) was done by reacting the β-1,4-D-xylosic substrates with rXylR (1 µg) in 50 mM sodium phosphate buffer (pH 5.5) for 2 h at 35 • C, during which time rXylR was more thermostable at the temperature than at 40 • C. The biocatalytic reaction was stopped by boiling the reaction mixtures at 100 • C for 5 min and the resulting degradation products were subsequently analyzed by liquid chromatography/tandem mass spectrometry (LC-MS/MS) using D-xylose (X 1 ) and D-xylooligosaccharides (X 2 -X 6 ) as standards. High performance liquid chromatography (HPLC) analysis was achieved using a Finnigan Surveyor Modular HPLC system (Thermo Electron Co., Waltham, MA, USA) equipped with an Asahipak NH2P-50 2D column (5 µm, 2.0 × 150 mm, Shodex) [23]. Moreover, LC-MS was carried out by employing a Finnigan LCQ Advantage MAX ion trap mass spectrometer (Thermo Electron Co., Waltham, MA, USA) equipped with an electrospray ionization (ESI) source [23].

Binding Assay
The substrate-binding ability of rXylR was evaluated using diverse insoluble polymers including Avicel PH-101, oat spelts xylan, ivory nut mannan, lignin, chitin from crab, chitosan, curdlan from A. faecalis, and wheat arabinoxylan. Before the binding assay, the insoluble materials were first washed four times with sterile distilled water to eliminate any residual water-soluble carbohydrates, after which they were rewashed with 50 mM sodium phosphate buffer (pH 5.5). Binding experiments of rXylR to candidate polymers with a specific microstructure were performed using the assay mixtures, which included an appropriately diluted enzyme preparation (5.0 U/mL) together with an equal volume of insoluble polymer in a 1.5 mL Eppendorf tube, on ice for 2 h. The assay mixtures were vigorously stirred every 5 min during the reaction period. The supernatants containing rXylR proteins unbound to the evaluated polymers were carefully retrieved by centrifugation (12,000× g) and applied directly to the quantitative assay of protein concentration and endo-β-1,4-xylanase activity.

Identification of a Xylanolytic Antarctic Isolate
A Gram-negative bacterium, strain PAMC 27433, was identified as Duganella sp. based on its 16S rRNA gene sequence (GenBank accession number: MW820638) that was 98.2% similar to that of Duganella pernnla FT109W T . Strain PAMC 27433 was deposited in the Polar and Alpine Microbial Collection under code no. Duganella sp. PAMC 27433.

Molecular Characterization of the GH10 Endo-β-1,4-Xylanase Gene
The XylR gene (GenBank accession number: MW735678) of the xylanolytic Antarctic bacterium Duganella sp. PAMC 27433 was discovered to contain a 1122-bp open reading frame that codes for an acidic protein (pI: 6.84) of 373 amino acids with a deduced molecular mass of 41,350 Da. When analyzed by the SignalP 5.0 server (http://www.cbs.dtu.dk/ services/SignalP/), the premature XylR was predicted to be an extracellular protein with a signal sequence that might be post-translationally processed between Ala23 and Ala24 in the N-terminus region ( Figure 1). On the other hand, the (His) 6 -tagged recombinant XylR lacking a signal peptide evaluated in this study was predicted to be a polypeptide with a calculated pI of 6.79 and a deduced molecular mass of 41,234 Da. Protein BLAST and Pfam analyses of the primary structure of XylR indicated that the enzyme might be a non-modular carbohydrolase made up of a single catalytic GH10 domain (from Leu31 to Asp364).
Sequence alignment between the primary structure of XylR and those of its structural homologs available in the NCBI database indicated that the catalytic GH10 domain of the enzyme exhibited the highest sequence identity (86%) with that of uncultured bacterium BLR13 endo-β-1,4-xylanase A precursor (GenBank accession number: ACN58881), which had not yet been structurally and functionally characterized. Moreover, the catalytic GH10 domain of XylR shared 84, 82, and 79% sequence identities with that of Rugamonas sp. SG757 endo-β-1,4-xylanase (WP_176653060), Duganella levis endo-β-1,4-xylanase (WP_161054863), and Duganella rivi endo-β-1,4-xylanase (WP_154357092), respectively, which were recorded in the NCBI database only. To the best of our knowledge, no report concerning the functional characteristics of microbial GH10 endo-β-1,4-xylanases available in the NCBI database, which share sequence identity of >50% with XylR, has been published to date.

Purification and SDS-PAGE Analysis of rXylR Proteins
It has been demonstrated that when overexpressed in E. coli BL21, some bioactive proteins are frequently produced as insoluble inclusion bodies that are purified through an on-column protein refolding method [25]. A tri-modular GH10 endo-β-1,4-xylanase with a fibronectin type 3 domain and a carbohydrate-binding module 2 from Cellulosimicrobium sp. strain HY-13 [26] and a bi-modular GH10 endo-β-1,4-xylanase with a ricin-type β-trefoil lectin domain-like domain from Luteimicrobium xylanilyticum HY-24 [22] are examples of enzymes that were formed as inactive protein aggregates in the recombinant cells. However, similar to other non-modular GH10 endo-β-1,4-xylanases [17,18,21], rXylR comprised of only a single catalytic GH10 domain was overproduced in an active form in E. coli BL21. Thus, the enzyme could be simply purified to electrophoretic homogeneity by basic affinity chromatography using a HisTrap HP column.

Purification and SDS-PAGE Analysis of rXylR Proteins
It has been demonstrated that when overexpressed in E. coli BL21, some bioactive proteins are frequently produced as insoluble inclusion bodies that are purified through an on-column protein refolding method [25]. A tri-modular GH10 endo-β-1,4-xylanase with a fibronectin type 3 domain and a carbohydrate-binding module 2 from Cellulosimicrobium sp. strain HY-13 [26] and a bi-modular GH10 endo-β-1,4-xylanase with a ricintype β-trefoil lectin domain-like domain from Luteimicrobium xylanilyticum HY-24 [22] are examples of enzymes that were formed as inactive protein aggregates in the recombinant cells. However, similar to other non-modular GH10 endo-β-1,4-xylanases [17,18,21], rXylR comprised of only a single catalytic GH10 domain was overproduced in an active form in E. coli BL21. Thus, the enzyme could be simply purified to electrophoretic homogeneity by basic affinity chromatography using a HisTrap HP column.

Biochemical Characterization of Recombinant Enzymes
To date, only four endo-β-1,4-xylanases from Antarctic microorganisms have been genetically and biochemically characterized (Table 1). Of the four, two, thermolabile GH10 endo-β-1,4-xylanases from the mold Cladosporium sp. [13] and the yeast C. adeliae [28], were reported to be most active at 50 °C and pH 6.0 for beechwood xylan and at 45-50 °C in pH range 5.0-5.5 for birchwood xylan. Additionally, a GH10 endo-β-1,4-xylanase from Psychrobacter sp. strain 2-17 [14] and a GH8 endo-β-1,4-xylanase from Pseudoalteromonas haloplanktis [12] exhibited the maximum biocatalytic activity for birchwood xylan at 35 °C and pH 8.0 and at 35 °C in pH range 5.3-8.0, respectively. However, in this study, the highest endo-β-1,4-xylanase activity of rXylR for beechwood xylan was observed at 40 °C and pH 5.5 (Figure 4a,b), which suggested that the pH and temperature optima of the enzyme were distinguished from those of the aforementioned endo-β-1,4-xylanases from Antarctic microorganisms. It is interesting to note that rXylR was capable of degrading beechwood xylan even at 4 °C, although at that temperature, the enzyme showed approximately 12% of its maximum endo-β-1,4-xylanase activity. Furthermore, at 8 °C, the relative activity of rXylR to decompose the substrate was evaluated to be approximately 20% of its maximum biocatalytic activity, implying that it is a cold-adapted biocatalyst. When the enzyme reaction was performed at pH values below 5.0 or above 7.0, and at temperatures exceeding 40 °C, the endo-β-1,4-xylanase activity of rXylR was considerably reduced. However, it should be noted that rXylR was relatively stable in a broad range of pH values (3.5-10.5) similar to a cold-active GH10 endo-β-1,4-xylanase from the marine bacterium E. rosea sp. nov. JL3085 T [27]. Actually, the enzyme retained over 80% of residual endo-β-1,4-xylanase activity at those pH values even when exposed to a temperature of 4 °C for 1 h without beechwood xylan as a substrate (Figure 4c). The half-life of rXylR at 40 °C was determined to be approximately 30 min (Figure 4d), indicating that it was more stable at the temperature compared to thermolabile GH10 endo-β-1,4-xylanases

Biochemical Characterization of Recombinant Enzymes
To date, only four endo-β-1,4-xylanases from Antarctic microorganisms have been genetically and biochemically characterized (Table 1). Of the four, two, thermolabile GH10 endo-β-1,4-xylanases from the mold Cladosporium sp. [13] and the yeast C. adeliae [28], were reported to be most active at 50 • C and pH 6.0 for beechwood xylan and at 45-50 • C in pH range 5.0-5.5 for birchwood xylan. Additionally, a GH10 endo-β-1,4-xylanase from Psychrobacter sp. strain 2-17 [14] and a GH8 endo-β-1,4-xylanase from Pseudoalteromonas haloplanktis [12] exhibited the maximum biocatalytic activity for birchwood xylan at 35 • C and pH 8.0 and at 35 • C in pH range 5.3-8.0, respectively. However, in this study, the highest endo-β-1,4-xylanase activity of rXylR for beechwood xylan was observed at 40 • C and pH 5.5 (Figure 4a,b), which suggested that the pH and temperature optima of the enzyme were distinguished from those of the aforementioned endo-β-1,4-xylanases from Antarctic microorganisms. It is interesting to note that rXylR was capable of degrading beechwood xylan even at 4 • C, although at that temperature, the enzyme showed approximately 12% of its maximum endo-β-1,4-xylanase activity. Furthermore, at 8 • C, the relative activity of rXylR to decompose the substrate was evaluated to be approximately 20% of its maximum biocatalytic activity, implying that it is a cold-adapted biocatalyst. When the enzyme reaction was performed at pH values below 5.0 or above 7.0, and at temperatures exceeding 40 • C, the endo-β-1,4-xylanase activity of rXylR was considerably reduced. However, it should be noted that rXylR was relatively stable in a broad range of pH values (3.5-10.5) similar to a cold-active GH10 endo-β-1,4-xylanase from the marine bacterium E. rosea sp. nov. JL3085 T [27]. Actually, the enzyme retained over 80% of residual endo-β-1,4-xylanase activity at those pH values even when exposed to a temperature of 4 • C for 1 h without beechwood xylan as a substrate (Figure 4c). The half-life of rXylR at 40 • C was determined to be approximately 30 min (Figure 4d), indicating that it was more stable at the temperature compared to thermolabile GH10 endo-β-1,4-xylanases from Antarctic fungal species [13,28], which were almost completely inactivated in 30 min at 40 • C. The thermostability of rXylR was gradually downregulated in a temperature-dependent manner when subjected to temperatures of 25-40 • C for the same pre-incubation period, while the enzyme was drastically inactivated when pre-incubated for 10 min at 45 • C (Figure 4d). at 40 °C. The thermostability of rXylR was gradually downregulated in a temperature-dependent manner when subjected to temperatures of 25-40 °C for the same pre-incubation period, while the enzyme was drastically inactivated when pre-incubated for 10 min at 45 °C (Figure 4d).  As shown in Figure 5, rXylR did not show endo-β-1,4-xylanase activity for beechwood xylan in the presence of tryptophan residue-directed modifiers such as Hg 2+ (1 mM) and Nbromosuccinimide (5 mM). A similar observation was also made when some bacterial GH10 endo-β-1,4-xylanases were reacted with β-1,4-D-xylosic polysaccharides in the presence of the aforementioned compounds [5,16,22,29]. Taken together, these findings corresponded to the fact that the oxidizing substances Hg 2+ and N-bromosuccinimide modify the indole ring of highly conserved tryptophan residues in the active site of GH10 endo-β-1,4-xylanases, which play a critical role in the enzyme-substrate interaction [24]. Interestingly, rXylR was not only almost completely inhibited in the presence of 1 mM Ca 2+ or Cu 2+ but also noticeably suppressed by 1 mM Ni 2+ , Zn 2+ , or Fe 2+ , although it was partially inhibited by 1 mM Mg 2+ , Ba 2+ , or Co 2+ . The significant inhibitory effects of the indicated divalent cations on the rXylR activity were very comparable to the inhibitory or stimulatory effects of the same compounds on the biocatalytic activity of some cold-active GH10 endo-β-1,4-xylanases [14,16,18,27,30]. For example, the metal ions (each 1 mM), Ca 2+ and Cu 2+ , had no inhibitory effect on the activity of Psychrobacter sp. strain 2-17 GH10 endo-β-1,4xylanase [14] and Glaciecola mesophila KMM 241 GH10 endo-β-1,4-xylanase [30]. Moreover, two GH10 endo-β-1,4-xylanases from B. cellulosilyticus [18] and E. rosea sp. nov. JL3085 T [27] were only slightly sensitive to 1 mM Cu 2+ , Fe 2+ , Ni 2+ , or Zn 2+ . On the other hand, in the presence of 1 mM Cu 2+ or Fe 2+ , the biocatalytic activity of a GH10 endo-β-1,4-xylanase from F. johnsoniae was greatly enhanced by approximately 2.2-fold and 1.5-fold, respectively [16]. Compared to the slight suppression of Psychrobacter sp. strain 2-17 GH10 endo-β-1,4xylanase by the non-ionic detergent Triton X-100 (1.0%) [14], it is noteworthy that the rXylR activity could be upregulated by >1.45-fold in the presence of 0.5% Triton X-100 or Tween 80 ( Figure 5), which is similar to some GH10 functional homologs [16,23,26]. In addition, no significant inhibition or stimulation of rXylR was observed when the enzyme was pre-incubated in the presence of sulfhydryl reagents (each 5 mM) including iodoacetamide, sodium azide, and N-ethylmaleimide, while 1 mM ethylenediaminetetraacetic acid (EDTA) caused 27% reduction in its original endo-β-1,4-xylanase activity. The inhibitory effect of rXylR exerted by EDTA was close to that of Psychrobacter sp. strain 2-17 GH10 endo-β-1,4xylanase by the same compound [14]. Conversely, the xylanolytic activity of a cold-active GH10 endo-β-1,4-xylanase from G. mesophila KMM 241 was slightly promoted by the metal chelator, EDTA [30].

Substrate Specificity and Hydrolytic Properties of rXylR
The substrate specificity of the cold-adapted rXylR enzyme, which was determined under the standard assay conditions using different types of polysaccharides and pNPsugar derivatives, is listed in Table 2. Of the tested polymeric substrates, the enzyme could preferentially decompose D-xylose-based hemicelluloses with the following order: bee-

Substrate
V max (U mg − The results of HPLC analysis clearly showed that rXylR was capable of hydrolyzing birchwood xylan and D-xylooligosaccharides with a degree of polymerization in the range of 2-6, even though the susceptibility of D-xylobiose (X 2 ) molecules to the enzyme was insignificant (Table 4). Enzymatic degradation of the β-1,4-D-xylosic materials resulted in the production of D-xylose (X 1 ) in addition to X 2 as the dominant end product, regardless of the degree of polymerization of the substrates. Specifically, the degradation products of birchwood xylan produced by rXylR were identified as X 1 (37.1%) and X 2 (62.9%). D-Xylotriose (X 3 ) was also hydrolyzed by the enzyme to X 1 (19.3%) and X 2 (80.7%) without the formation of detectable amounts of D-xylooligosaccharides longer than X 3 , indicative of the lack of transglycosylation activity. Furthermore, the production of X 1 in the biocatalytic reactions of D-xylooligomers and birchwood xylan was gradually increased with an increase of the β-1,4-D-xylosic substrate chain length (Table 4). Based on the hydrolytic properties of rXylR, we propose that the enzyme is the first cold-adapted X 2 and X 1 -releasing endo-β-1,4-xylanase from Antarctic microorganisms, which can be employed as a promising candidate for the low-temperature preparation of simple sugars from β-1,4-D-xylosic materials via a one-step procedure. Considering that rXylR only produced X 1 and X 2 as final hydrolysis products from both D-xylooligomers (X 2 -X 6 ) and birchwood xylan, the biocatalytic ability of the enzyme was very comparable to that of the known GH10 enzymes for the D-xylose-based oligomeric and polymeric substrates. However, in contrast to rXylR, a cold-active GH10 endo-β-1,4-xylanase from G. mesophila KMM 241 could not hydrolyze X 2 and X 3 [30]. In addition, the breakdown of either Dxylooligomers (X 4 -X 6 ), β-1,4-D-xylans, or both, by cold-active GH10 endo-β-1,4-xylanases from Cladosporium sp. [13], B. cellulosilyticus [18], and G. mesophila KMM 241 [30] always accompanied the formation of either X 2 , X 3 , X 4 , or in combination, and not X 1 . Furthermore, cold-active GH10 endo-β-1,4-xylanases from F. johnsoniae [16] and the environmental DNA of goat rumen contents [15] were reported to produce mixtures of X 1 , X 2 , and longer D-xylooligosaccharides in the degradation processes of β-1,4-D-xylans.

Conclusions
The cold-adapted GH10 endo-β-1,4-xylanase (XylR) from an Antarctic soil bacterium, Duganella sp. PAMC 27433 is a novel X 2 -and X 1 -releasing biocatalyst displaying peculiar characteristics in its primary structure, sensitivity to metal ions, biocatalytic activity, substrate specificity, degradation patterns of β-1,4-D-xylosic substrates, and substratebinding capacities to insoluble polymers. Unlike known cold-active GH10 endo-β-1,4xylanases [16,18,30], rXylR does not have transglycosylation activity and lacks the ability to produce D-xylooligosaccharides longer than X 2 , and therefore can be exploited as a potential biocatalyst for the one-step production of fermentable sugars (X 2 and X 1 ) from various β-1,4-D-xylan polysaccharides at low temperatures. From an ecological context, the data presented in this study reflect the contribution and biological importance of xylanolytic Antarctic bacteria to the bioremediation of hemicellulosic materials in the cold environment.