The gh43kk gene was obtained from a nucleotide sequence of a metagenomic subclone Biof1_09 (Genbank JQ581599 [1]. The gene was amplified by polymerase chain reaction (PCR) using specific primers containing BamHI and EcoRI restriction site (5’-GGGCGCGGATCCATGTACGCCAT GAAAAA CATTTTTAAATATG-3’) and (5’-GGGCCGCCGGAATTC TGTAGAAGTGCCACCTTTCAAC -3’). The amplification protocol was according to the Taq DNA polymerase manual (Thermo Scienctific, USA), starting with an initial denaturation at 95°C for 3 min, 30 cycles of denaturation at 95°C for 30 s, annealing at 42 °C for 1 min 30 s and extension at 72°C for 1 min 30 s, final extension at 72°C for 10 min. The PCR product was purified by QIAquick^® PCR Purification Kit (Qiagen, Germany) and was digested with FastDigest BamHI and EcoRI (Thermo Scienctific, USA). Afterward, the gh43kk fragment was inserted into pET28a(+) expression vector (Novagen, USA) by T4 DNA ligase (New England Biolabs, USA). The recombinant expression vector (pET/gh43kk) was transformed into E. coli BL21 by heat-shock technique. The transformants were selected by colony PCR and then also confirmed by sequencing (Macrogen, Korea).

The molecular weight and isoelectric point of the translated nucleotides of the gene gh43kk were calculated by Expasy software (https://web.expasy.org/compute.pi/). The patterns of amino acid residue conservation and divergence in a protein family were analyzed by searching in the Conserved Domain Database (CDD, https://www.ncbi.nlm.nih.gov/Structure/cdd /cdd.shtml).

A three-dimensional model of Gh43kk was generated by RaptorX (http://raptorx.uchicago.edu/StructurePrediction/ myjobs/62894490_578596/) with a score of 32. The P-value for the relative global quality was 0.0365. The GDT (global distance test) and uGDT (un-normalized GDT) values of the model were 30 and 33, respectively.

The expression of the recombinant protein was done under low temperature as follows: OD₆₀₀ of overnight bacterial culture was adjusted to 0.2 and the incubation was continued at 37°C until the OD₆₀₀ reached 0.4. Isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to a final concentration of 0.1 mM, and the cell suspension was further incubated for 20 h at 16°C. Lysates were prepared from 100 ml E. coli culture. Bacterial cells were harvested by centrifugation at the speed of 7,500 rpm, 4 °C for 10 min. Each gram of the resulting pellet was lysed in 5 ml of 50 mM sodium acetate pH 5, 300 mM NaCl, 20 mM imidazole and 90% (v/v) B-PER™ Bacterial Protein Extraction Reagent (Thermo Scientific, USA). Lysozyme (final concentration 1 mg/ml (w/v) and cOmplete™ Protease Inhibitor Cocktail (Sigma-Aldrich, USA) was further added to the lysate, then incubated at room temperature for 30 min prior to sonication. Sonication (Vibra-Cell™, Sonics, was done at 35% amplitude (pulse duration of 5 s on and 10 s off) on ice. The lysate was centrifuged at 18,000 g for 30 min at 4°C and the supernatant was sterile filtered through 0.22 µm membrane filter. Protino^® Ni-NTA Agarose (Macherey Nagel, Germany) was used for purification according to the manufacturer's instruction. Briefly, 5 ml of HisPur™ Ni-NTA Resin (Thermo Fisher Scientific, USA) was loaded into an empty gravity flow column with filter frits (Protino^® column 14 ml, Macherey-Nagel, Germany). The agarose was equilibrated with lysis buffer. The clear lysate was then added into the column, and incubated with the Ni-NTA agarose with gentle shaking on an orbital shaker for 60 min at 4°C. After letting the lysate flow through the column, the agarose was washed twice with 50 mM sodium acetate pH 5, 300 mM NaCl, and 30 mM imidazole. The recombinant protein was eluted with 50 mM sodium acetate pH 5, 300 mM NaCl, and 300 mM imidazole. The eluate was then brought through the Vivaspin^® 6 (cut off 10 kDa; GE Healthcare) to change the buffer to 20 mM Sodium acetate buffer pH 5.4. The purified protein was checked by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) using 13.5% acrylamide-separating gel. Western blot analysis using anti-his tag monoclonal antibody was done to confirm the recombinant protein. Concentration and yields of the purified protein were calculated by Bradford protein assay. Purity and recovery of the recombinant proteins were estimated from protein band intensity in SDS-PAGE quantified by ImageJ software (National Institutes of Health, USA).

Azurine cross-linked cellulose (AZCL-HE-cellulose) (Megazyme, Ireland) or beechwood xylan (Megazyme, Ireland) was mixed with molten sterile agar (prepared with 50 mM sodium acetate buffer pH 5) to a final concentration of 1% (w/v). Various amounts of the purified protein, as well as the commercial cellulose mixture, Celluclast™ (Novozymes, Denmark) and Cellubrix^® (Novozymes, Denmark) as positive control and protein elution buffer (0.1 M sodium acetate buffer pH 5) containing imidazole as negative control were dropped into agar wells. After incubation at 37°C for 2 days, the plate containing beechwood xylan was stained with 0.33% Gram’s iodine and clear zones around the agar wells were observed. When the plate containing AZCL-HE-cellulose was used, smooth blue zones around the agar well indicated the activity of cellulose-degrading enzymes.

Enzyme activity was determined by incubating substrates with the purified enzyme in 100 mM sodium acetate buffer pH 5 at 50°C for 30 min, which will be used as a standard condition for the rest of the enzyme activity experiment. Reducing sugars were assayed by the DNS method with glucose as a standard. The concentration of pulverized substrates (avicel, beechwood xylan, cellobiose, cellulose, carboxymethyl cellulose) was 1% (w/v), while the size of filter paper was 0.5×1.5 cm. Negative controls were done by incubating the substrates without the enzyme (using elution buffer instead). The values of colorimetric detection (in reducing sugars) of controls were subtracted from the results of the actual reactions. All the experiments were done in three replicates. One unit of enzyme activity was defined as the amount of enzyme that releases 1 µmol of glucose equivalents per minute. Specific activity was exhibited in the unit of activity enzyme in one mg of protein. Protein concentration was determined by using the method as described by Bradford [12] with BSA as a standard.

The kinetic studies were determined using beechwood xylan concentrations from 0.5 to 5 mM, respectively. Reducing sugars were measured after the reactions were incubated at 50°C for 30 min. The Michaelis constant (K_m) and maximum velocity (V_max) were determined by means of Lineweaver–Burk plots [13]

TLC was performed as modified by Zhou and coworkers [14]. One percent (w/v) of substrates were incubated with Gh43kk protein under optimal conditions. Subsequently, aliquots of 2 μl each were spotted on thin layer chromatography (TLC) silica gel 60 plates (Merck, USA), along with standards of glucose, cellobiose, xylose, and arabinose, respectively. Separation was carried out by TLC with a mobile phase (ethanol/1-butanol/water at a ratio of 3:5:2) at room temperature. Sugars were detected by spraying the plates with ethanol and sulfuric acid (ratio 95%:5% v/v), followed by heating at 100°C for 10 min. The experiment was done in triplicates. Relative rates of flow (R_f) were calculated as R_f = X/ Y, by which X = migration distance of the sample; and Y = migration distance of the solvent.

The optimal temperature for enzyme activity was performed at various temperatures ranging from 10 to 90°C. To determine temperature stability of the enzyme, the purified enzyme was incubated at different temperatures for 30 min prior to the enzyme activity assay. Determination of optimum pH was done in 100 mM sodium acetate buffer (pH 3-6), 100 mM phosphate buffer (pH 6-8) and 100 mM tris-HCl buffer (pH 8-10). For the test of pH stability, the protein was pre-incubated in buffers with different pH for 1 h at 4°C, then the activity was assayed under the same condition as mentioned in Enzyme activity assay section. The effects of metal ions compounds (Ba²⁺ Ca²⁺ Co²⁺Cu²⁺, Fe²⁺, Li⁺, Mg²⁺, Mn²⁺, Ni²⁺, Sr²⁺ and Zn²⁺, salts (KCl, NaCl), chelating agent (Ethylenediaminetetraacetic acid; EDTA), and 1% (w/v) detergent (sodium dodecyl sulfate; SDS) were determined under standard condition. The concentrations of metal ions, salts and EDTA were 5 mM. Control was buffer without addition of any tested chemicals or metal ions. Beechwood xylan was used as substrate for all experiments.

The data was analyzed by using the two-sample t-test (Satterthwaite’s method), by which a p-value of ≤ 0.05 is considered significant.

The gene gh43kk contains putative nucleotides of 330 bp, which encode a polypeptide of 110 amino acid residues with a theoretical molecular mass of 12.81 kDa and an isoelectric point of 4.5. The protein Gh43kk shared conserved regions within the amino acid sequences of the glycosyl hydrolase family 43. Among those are amino acid residues which were reported to be found on the active site of the enzyme, namely aspartic acid (Asp43), glutamine (Gln93), glycine (Gly107), and Threonine (Thr108) (Graciano 2012) The multiple amino acid sequence alignment in Fig. (1) shows that these key residues are well conserved in Gh43kk protein.

As shown in the three-dimensional model (Fig. 2), the N-terminal region of Gh43kk displays a β-meander motif containing 4 consecutive antiparallel β-strands linked together by hairpin loops. The C-terminal region shows a v-shaped helix-break-helix fold [15]. The conserved catalytic aspartate (Asp43) is located in the hairpin loops linking the first pair of the antiparallel β-sheets.

Fig. (1). Multiple amino acid sequence alignment of GH43KK and other similar enzymes revealing the conserved regions within the amino acid sequences of glycosyl hydrolase family 43. Significant identity or similarity of amino acid residues is marked with an asterisk (similar), colon (highly identical), or dot (moderately identical). Amino acid residues, which were reported to be found on the active site, are indicated with red arrows. GH43KK was aligned with beta-xylosidase/alpha-L-arabinofuranosidase from following microorganisms: Eubacterium rectale M104/1 (CBK93970.1), Caulobacter crescentus NA1000 (ACF39706.1), Paenibacillus lactis 154 (EHB68557.1)

Fig. (2). Overall structure of the putative Gh43kk. The N-terminal, yellow colored antiparallel β-strands were linked together by hairpin loops. The catalytic amino acid, aspartate (Asp43) residue, is located at the hairpin loop of the peripheral β-sheet (red circle). The helical C-terminal region is shown in pink.

The gene gh43kk was successfully cloned, and protein expression was optimized. The recombinant Gh43kk protein was effectively overexpressed at 37°C by using IPTG as an inducer, however, a large amount of the expressed protein remained as an inclusion body (Fig. S1), which obstructs the purification. The recombinant protein was soluble when the temperature was shifted to 16°C, therefore, the recombinant protein Gh43kk was successfully expressed and purified at that temperature (Fig. 3).

Fig. (3) shows the purification step with the purified protein band of about 23 kDa. The purified recombinant protein was more than 95% pure, with around 75% recovery. Approximately 10 milligrams of recombinant protein was obtained from 1 liter of bacterial expression culture (1-1.5 g wet weight). The recombinant Gh43kk was confirmed by Western blot analysis and a protein band of about 23 kDa was visualized (Fig. 3B).

Fig. (3). SDS-PAGE showing purification step of the recombinant protein (A) I = uninduced protein; T = total cell lysate after induction; P = pellet; S = supernatant; Ft = flow-through; W = washing; E = elution; C = concentrated protein by Vivaspin® column; M = protein standard maker.

Activities of the purified Gh43kk protein were detected both on agar and in liquid suspension using various substrates. Smooth blue-colored zone or clear zones were observed on agar plates containing AZCL-Cellulose and xylan as substrates, respectively (Fig. 4). The same result was observed by spectrophotometric experiments, by which the enzyme showed the highest hydrolytic activity towards the beechwood xylan with a specific activity of 286.37 ± 7.37 mU·mg^-1 protein followed by CMC with 129.78 ± 7.91 mU·mg^-1 protein, respectively (Table 1). The K_m for Gh43kk on beechwood xylan was 0.153 µM, while the Turnover number (K_cat) was 0.0213, and Kcat/Km of the enzyme was 0.143.

Fig. (4). Agar well diffusion test of the purified recombinant protein Gh43kk on AZCL-Cellulose (i) and beechwood xylan (ii) agar plates. B = 0.1 M sodium acetate buffer pH 5, C = mixture of diluted Celluclast™ and Cellubrix®, E = 0.1 M sodium acetate buffer pH 5 with 300 mM imidazole, 2 = 17.27 mg purified protein and 3 = 25.91 mg purified protein.

Thin Layer Chromatography (TLC) analysis of hydrolysate of different substrates treated by the recombinant protein Gh43kk revealed several patterns of sugar products (Fig. 5). Both glucose and cellobiose were detected when avicel, cellulose and filter paper were hydrolyzed with Gh43KK. Only glucose was observed when CMC and rice straw were degraded with the recombinant enzyme. Gh43KK also exhibits its xylanase activity since xylan could be hydrolyzed as well, yielding xylose as a product.

The purified recombinant Gh43kk could work well (with an activity of more than 80%) in pH ranging from 4 to 8, with an optimum pH of 4 (Fig. 6A). However, the enzyme activity of about 60% was still observed at pH 3 and 9, suggesting its capability in working at a very wide range of pH (Fig. 6A). The enzyme maintained its activity of more than 60% when pre-incubated overnight in buffer with pH ranged from 4 to 9, where pH 5 to 7 was shown to have the least effect on its activity (Fig. 6B). The recombinant protein Gh43kk showed increased activity when the temperature rose toward 50°C, which was also the optimum temperature for the enzyme (Fig. 6C). Its activity decreased rapidly when the temperature exceeded 50°C (Fig. 6C). The recombinant protein maintained more than 80% of its activity after pre-incubation at the temperature under 50°C (Fig. 6D).

The highest activity (with an activity of about 120%) of the recombinant enzyme was observed when manganese ions were present. Slightly higher activity of the enzyme was also observed with the presence of magnesium ion, however it did not significantly enhance the activity of the enzyme (p-value of 0.21). Moderate inhibition (activity of about 76-79%) of the enzyme Gh43kk was observed when copper, barium, potassium and sodium ions were present. Metal ions of zinc, calcium and strontium, and sodium dodecyl sulfate (SDS) were shown to strongly inhibit the activity of the recombinant enzyme. Gh43kk protein almost lost its activity when incubated with Ethylenediaminetetraacetic acid (EDTA) (with an activity of about 4.68%) (Table 2).

Fig. (5). (A) Thin Layer Chromatography of hydrolysis result after treatment of different substrates with the recombinant protein Gh43kk under optimal condition (B). Rf value of each substrate observed by TLC in this experiment.

Fig. (6). Effect of temperature and pH on activity and stability of the recombinant protein Gh43kk. (A) Effect of pH, (B) pH stability, (C) Effect of temperature, (D) Temperature stability. Blue lines indicate experiments conducted with sodium acetate buffer. Orange lines indicate experiment conducted with Tris buffer. Beechwood xylan was used as substrate.