Purification and Characterization of a Thermostable β-Mannanase from Bacillus subtilis BE-91: Potential Application in Inflammatory Diseases

β-mannanase has shown compelling biological functions because of its regulatory roles in metabolism, inflammation, and oxidation. This study separated and purified the β-mannanase from Bacillus subtilis BE-91, which is a powerful hemicellulose-degrading bacterium using a “two-step” method comprising ultrafiltration and gel chromatography. The purified β-mannanase (about 28.2 kDa) showed high specific activity (79, 859.2 IU/mg). The optimum temperature and pH were 65°C and 6.0, respectively. Moreover, the enzyme was highly stable at temperatures up to 70°C and pH 4.5–7.0. The β-mannanase activity was significantly enhanced in the presence of Mn2+, Cu2+, Zn2+, Ca2+, Mg2+, and Al3+ and strongly inhibited by Ba2+ and Pb2+. K m and V max values for locust bean gum were 7.14 mg/mL and 107.5 μmol/min/mL versus 1.749 mg/mL and 33.45 µmol/min/mL for Konjac glucomannan, respectively. Therefore, β-mannanase purified by this work shows stability at high temperatures and in weakly acidic or neutral environments. Based on such data, the β-mannanase will have potential applications as a dietary supplement in treatment of inflammatory processes.


Introduction
Mannan consists of a series of complex polysaccharides, which are found in the cell wall of marine algae [1]. The backbone is comprised of -1,4-linked mannose residues. Konjac glucomannan is a randomly arranged polymer of -1,4-linked glucose and mannose residues at ratio of 1.0 : 1.6. Both the backbones of mannan and Konjac are modified by -1,6-linked galactosyl residues to form galactomannan and galactoglucomannan, respectively [2].
Although multiple -mannanase-producing bacteria have been reported [15,16], they are far from the diverse industry needs. Currently, acidic and alkaline -mannanase has been proposed to meet the industrial demands [17]. However, the requirements of high energy in production and the environmental impact limit their development. Neutral and weakly acidic -mannanase with lower energy for production has attracted considerable interest over the past few years; however, it has rarely been characterized. It is clarified that -mannanase with high activity in short fermentation time confers lower costs during the production procedures. Therefore, the exploitation of strains producing high -mannanases activity is valuable and profitable. In current study, we isolated and preserved a powerful hemicellulose-degrading bacterium (BE-91). Then we explored the efficient purification process and characterized the enzymatic properties of its -mannanase.

Classification of Strain BE-91.
The 16S rDNA of strain BE-91 was PCR amplified from genomic DNA using the Bacterial Identification PCR Kit (TaKaRa, Japan). The obtained 16S rDNA was sequenced by the ABI 3730XL 96capillary DNA analyzer. The primers were as follows: P1 5 -AGAGTTTGATCMTGGCTCAG-3 and P2 5 -TACGGY-TACCTTGTTACGACTT-3 . The resulting sequence aligned closely with the related standard sequences of other bacteria retrieved from GenBank. Neighbor-joining clusters were constructed by Mega 6.0 [19].

Enzymatic
Assays. -mannanase activity was estimated by initiating the reaction at 65 ∘ C for 10 min in 0.05 mol/L citric acid/0.1 mol/L Na 2 HPO 4 buffer (pH 6.0) with 0.2% (w/v) Konjac glucomannan as substrate. The amounts of reducing sugar in the reaction were quantified based on a standard curve generated with mannose using the 3,5-dinitrosalicylic acid (DNS) method. One unit (IU) of -mannanase activity was defined as the amount of protein producing 1 mol/L of reducing sugar per minute (e.g., mannose) under standard conditions [20].

Purification of -Mannanase.
The bacterial -mannanase was purified using a two-step process involving ultrafiltration (Sartorius, Germany) and gel filtration. The fermentation liquid was fractionated orderly by 100 kDa, 50 kDa, and 5 kDa membrane thresholds. The solution filtered with 5 kDa < MW < 50 kDa was further purified on a Sephadex G-100 gel column (B1.6 cm × 100 cm, Pharmacia). The eluate was obtained at a rate of 0.5 mL/min and collected in 5 mL fractions.mannanase activity was determined by the DNS method, whereas the protein was quantified by the Coomassie brilliant blue staining against bovine serum albumin (BSA) standard [21].

The Determination of Apparent Molecular Weight.
The molecular mass of the -mannanase was determined by SDS-PAGE (Bio-Rad, USA), with 3% stacking gel and 12% separating gel [22]. The protein bands were stained with 0.01% Coomassie brilliant blue R-250 and destained with a water-methanol-acetic acid (9 : 9 : 2) solvent. Zymogram analysis was performed by the method of Chanhan [17]. The molecular weight of -mannanase was derived from the relative mobility of molecular weight markers resolved simultaneously.
2.6. The Effect of Temperature on the Activity and Stability of -Mannanase. The activity of -mannanase was assayed at a range of temperatures between 50 and 70 ∘ C in 0.05 mol/L citric acid-0.1 mol/L Na 2 HPO 4 buffer at pH 6.0. The thermostability was assessed by preincubating the enzyme, without a substrate, at different temperatures varying over 20-80 ∘ C for 30 min. The residual activity was promptly measured by the DNS method. The -mannanase activity was considered to be 100% when preincubated at 4 ∘ C.

The Effect of pH on the Activity and Stability of -
Mannanase. -mannanase activity was evaluated by incubating the purified enzyme at different pH conditions ranging from 4.0 to 8.0 in 0.05 mol/L citric acid-0.1 mol/L Na 2 HPO 4 buffer at 4 ∘ C. The stability at a particular pH was tested by preincubating the purified enzyme, without a substrate, for 30 min in various 0.05 mol/L citric acid-0.1 mol/L Na 2 HPO 4 buffers at pH 3.0-8.5 at 4 ∘ C. The residual -mannanase activity was immediately measured after treatment by the DNS procedure.

The Effect of Metal Ions on the Activity of -Mannanase.
In order to examine the effects of metal ions on the activity of -mannanase, the enzyme was incubated for 30 min at 4 ∘ C in the presence of various 1.0 mmol/L metal ions, AlCl 3 , BaCl 2 , and NH 4 Cl. The residualmannanase activity was measured at a specific condition and that of the treatment in the absence of additives as a control.

Statistical Analysis.
Each -mannanase activity experiment was performed in triplicate and expressed as mean ± SD (standard deviation). The statistical analyses were performed with SPSS 15.0 (SPSS Inc., Chicago IL, USA). One-way or two-way analysis of variance (ANOVA) was used to compare various treatment groups.

Screening of the High -Mannanase Activity Producing
Strain. Four bacteria were stochastically selected for the -mannanase activity assay. Figure 1 exhibited the halos produced on the screening plate. Table 1 summarized the -mannanase activity of the four bacteria (strain BE-23 without -mannanase activity was used as a negative control). Strain BE-91 fermented for 9 h exhibited the highest activity, up to 273.7 IU/mL. Wild-type B. subtilis MA139 yielded a maximum -mannanase activity of 170 IU/mL after 3 days of fermentation, and the maximum enzyme activity of B. subtilis TJ-102 was 205.3 IU/mL at 38 h [25,26]. Notably, BE-91 secreted -mannanase with higher activity in shorter time.

Classification of B.
subtilis BE-91. The 1,508 bp sequence of 16S rDNA of strain BE-91 was analyzed by a phylogenic tree (Figure 2). The homology between BE-91 16S rDNA (gi 260159552) and B. subtilis 16S rDNA (gi 530330588 and gi 341831474) was 99%. It was confirmed that the similarity of B. subtilis type strains about 16S rRNA gene sequence is higher than 98% [27,28]. We also obtained ≥98% similarity to 16S rRNA gene sequences of B. subtilis isolates.

Isolation and
Purification of -Mannanase. 2,000 mL of fermentation liquor was purified by ultrafiltration and chromatography. Specific activity, recovery, and multiple purifications at each step were summarized in Table 2. The recovery of -mannanase in B. subtilis BE-91 exceeded 66.0%; multiple purifications achieved 32.9-fold pure -mannanase activity, and the specific activity of the purified enzyme reached 79,859.2 IU/mg. The purified -mannanase was shown to be homogeneous judged by SDS-PAGE analysis (Figure 3). Compared with the previous separation and purification methods [29,30], the two-step method has the advantages of high efficiency, high yield, and easy operation.

Optimal pH and Stability of -Mannanase.
The optimal pH and the stability of BE-91 -mannanase were measured at various pHs. The optimum enzyme activity was obtained at pH 6.0 (Figure 6), and more than 80% maximal activity was retained at pH 4.5-7.0 ( Figure 7). Interestingly, the optimal pH of BE-91 -mannanase was the same as that of B. subtilis MA139 (pH 6.0), an enzyme that can potentially be used  as a feed additive for monogastric animals [25]. At pH < 4.0, the -mannanase activity was negligible, retaining <80% of its maximal value obtained after incubation at pH > 7.5, 4 ∘ C for 30 min. A relatively broad zone of optimum activity was observed. Therefore, BE-91 -mannanase can be considered a weakly acidic and neutral enzyme, thereby rendering suitability for animal feed industry [38].

The Effect of Metal Ions on -Mannanase Stability.
The effect of a variety of metal ions on -mannanase activity was measured ( Table 3). The highest induction was achieved with Mn 2+ , which showed 168% baseline activity, followed by Al 3+ , Ca 2+ , Cu 2+ , Zn 2+ , Mg 2+ , and NH 4 + , respectively. K + and Fe 3+ had no obvious effects on -mannanase activity in these conditions. Ba 2+ and Pb 2+ greatly inhibited the enzyme activity to a final rate of 83% and 74%, respectively. This suggests that BE-91 -mannanase should not be contaminated by Ba 2+ and Pb 2+ .

Kinetic
Parameters. The purified enzyme hydrolyzed Konjac glucomannan but only slightly hydrolyzed ivory nut mannan, guar galactomannan, and 1,4-beta-D-mannan. Wheat arabinoxylan, beechwood xylan, and CMC were barely hydrolyzed, as shown in Table 4. This -mannanase exhibited the highest activity with Konjac glucomannan, enriched in glucose units. This finding suggests thatmannanase of BE-91 preferentially hydrolyzes the -1,4linkage of the glucosylated mannan backbone.  and max values of this -mannanase estimated by the Lineweaver-Burk plot were 7.14 mg/mL and 107.5 mol/ min/mL, respectively, for locust bean gum, versus 1.749 mg/ mL and 33.45 mol/min/mL for Konjac glucomannan, respectively. These results displayed higher affinity ofmannanase towards natural Konjac glucomannan ( max / , 19.1 mol/min/mg) than the locust bean gum ( max / , 15.0 mol/min/mg), similar to the values obtained for Penicillium pinophilum C1 and Penicillium freii F63, hence constituting it as an adequate candidate in food industry for the production of oligosaccharides [17,18,39].

Conclusion
B. subtilis bacteria are abundant, moderately stable, and mostly nonpathogenic microorganisms. Our results indicated that B. subtilis BE-91 could be considered a prominent candidate for the production of extracellular -mannanase. In addition, this study developed an advanced purification approach, "two-step method," with high efficiency, high yield, and easy operation. Furthermore, the -mannanase purified from BE-91 was extremely stable at relatively high temperatures and various weak acidic or neutral pHs. Finally, the enzyme showed a higher affinity towards natural Konjac glucomannan, a major functional food material. Therefore, thismannanase, purified and characterized from B. subtilis BE-91 for the first time, is suitable for inflammatory diseases.