Characterization and high-efficiency secreted expression in Bacillus subtilis of a thermo-alkaline β-mannanase from an alkaliphilic Bacillus clausii strain S10

Background β-Mannanase catalyzes the cleavage of β-1,4-linked internal linkages of mannan backbone randomly to produce new chain ends. Alkaline and thermostable β-mannanases provide obvious advantages for many applications in biobleaching of pulp and paper, detergent industry, oil grilling operation and enzymatic production of mannooligosaccharides. However, only a few of them are commercially exploited as wild or recombinant enzymes, and none heterologous and secretory expression of alkaline β-mannanase in Bacillus subtilis expression system was reported. Alkaliphilic Bacillus clausii S10 showed high β-mannanase activity at alkaline condition. In this study, this β-mannanase was cloned, purified and characterized. The high-level secretory expression in B. subtilis was also studied. Results A thermo-alkaline β-mannanase (BcManA) gene encoding a 317-amino acid protein from alkaliphilic Bacillus clausii strain was cloned and expressed in Escherichia coli. The purified mature BcManA exhibited maximum activity at pH 9.5 and 75 °C with good stability at pH 7.0–11.5 and below 80 °C. BcManA demonstrated high cleavage capability on polysaccharides containing β-1,4-mannosidic linkages, such as konjac glucomannan, locust bean gum, guar gum and sesbania gum. The highest specific activity of 2366.2 U mg−1 was observed on konjac glucomannan with the Km and kcat value of 0.62 g l−1 and 1238.9 s−1, respectively. The hydrolysis products were mainly oligosaccharides with a higher degree of polymerization than biose. BcManA also cleaved manno-oligosaccharides with polymerization degree more than 3 without transglycosylation. Furthermore, six signal peptides and two strong promoters were used for efficiently secreted expression optimization in B. subtilis WB600 and the highest extracellular activity of 2374 U ml−1 with secretory rate of 98.5% was obtained using SPlipA and P43 after 72 h cultivation in 2 × SR medium. By medium optimization using cheap nitrogen and carbon source of peanut meal and glucose, the extracellular activity reached 6041 U ml−1 after 72 h cultivation with 6% inoculum size by shake flask fermentation. Conclusions The thermo-alkaline β-mannanase BcManA showed good thermal and pH stability and high catalytic efficiency towards konjac glucomannan and locust bean gum, which distinguished from other reported β-mannanases and was a promising thermo-alkaline β-mannanase for potential industrial application. The extracellular BcManA yield of 6041 U ml−1, which was to date the highest reported yield by flask shake, was obtained in B. subtilis with constitutive expression vector. This is the first report for secretory expression of alkaline β-mannanase in B. subtilis protein expression system, which would significantly cut down the production cost of this enzyme. Also this research would be helpful for secretory expression of other β-mannanases in B. subtilis. Electronic supplementary material The online version of this article (10.1186/s12934-018-0973-0) contains supplementary material, which is available to authorized users.

Background β-Mannanase, as one of the mannan-degrading enzymes, plays an important role in the hydrolysis of hemicellulose and catalyzes the cleavage of β-1,4-linked internal linkages of mannan backbone randomly to produce new chain ends [1,2]. β-mannanases have many potential applications in industrial processes such as biobleaching of pulp and paper, detergent industry, food and feed industry, coffee processing industry, oil and gas industry, pharmaceutical applications, and second generation biofuel [1]. They are also involved in breaking down plant tissues in the cell walls [3] as well as digesting and absorbing nutrients in lower animals by degrading mannan polymers [4]. Owing to the importance and increasing application demand, many β-mannanases from different biological sources have been studied by purification, gene cloning, enzymatic characterization, structure and catalytic mechanisms, and engineering of improved features [1,5]. Besides a few β-mannanases from plants to lower animals [3,4], most of reported β-mannanases were from microorganisms including fungi and bacteria. Among bacteria, β-mannanases were mostly produced by Gram-positive strains like Bacillus species [6][7][8][9][10][11][12][13][14] and Clostridia species [15]. Only a few Gram negative strains such as Pseudomonas [16], Vibrio [17], Thermotoga [18], Rhodothermus [19], Klebsiella [20] and Enterococcus [21] have been reported to produce β-mannanase. Besides these, some β-mannanases from actinomycetes such as Streptomyces [22], Cellulomonas [23] and Thermobifida [24] have also been studied. For fungi, the most common mannolytic group belongs to the genus Aspergillus [25][26][27][28][29], while some species from genus Penicillium [30], Trichoderma [31] and Neosartorya [32] have also been investigated. Particularly, β-mannanases from extremophilic microorganisms have drawn special attention due to their unusual properties like thermo-, acid-, or alkalitolerance which may be desirable for specific industrial processes [33]. Thermo-alkaline β-mannanases were usually used in biobleaching of pulp and paper, detergent industry, oil grilling operation and enzymatic production of mannooligosaccharides because of their high activity and good stability at high pH value and temperature [1]. Although a number of alkaline and thermostable mannanases were reported in the past decades, only a few are commercially exploited as wild or recombinant enzymes [34][35][36]. Therefore, there is still a pressing need to seek for new enzymes that are efficient and stable at alkaline or high temperature conditions to promote the development of the industrial application of β-mannanases.
In addition to enzyme properties, high yield is another important factor in the industrial application of enzymes. However, the yield of β-mannanases produced from native microbes, especially from extremophilic microbes, is usually low and hard to be improved, which led to the increasing attention on the heterologous production of β-mannanases in recent years [33]. Many β-mannanase genes originated from different microbes have been cloned and heterologously expressed in the frequently used protein expression systems such as Escherichia coli [37][38][39], Pichia pastoris [33,[40][41][42] and Bacillus subtilis [5,43,44] in the past decade. Compared to E. coli, P. pastoris and B. subtilis can secreted the enzyme directly into the culture medium, which making the downstream separation and purification much easier. However, most of the reported β-mannanases expressed in these two expression systems were acidic or neutral enzymes and only alkaline β-mannanase gene from alkaliphilic Bacillus sp. N16-5 was found heterologously expressed in P. pastoris with the highest yield of 1114 U ml −1 and 6336 U ml −1 by shake flask and fed-batch fermentation, respectively [33,45]. No heterologous and secreted expression of alkaline β-mannanase in B. subtilis was reported in the literature. In addition, B. subtilis possesses its own advantages in generally producing proteins with higher secretion concentration and well-characterized secretion pathways compared to P. pastoris [5]. Even when comparable secretion capacity occurs, B. subtilis still shows higher growth rate than P. pastoris, which makes it a priority in industrial application. On the other hand, bacteria producing β-mannanase, especially alkaline β-mannanases, are mostly confined to Gram-positives including various Bacillus species. Therefore, B. subtilis would be preferred expression system for highlevel secreted production of β-mannanases including alkaline β-mannanases originated from Bacillus strains.
In previous research, we isolated an alkaliphilic strain identified as Bacillus clausii strain S10 from soda lake in Inner Mongolia, China, which showed high alkaline β-mannanase activity. In this study, we described the gene cloning, heterologous expression, purification and characterization of this alkaline β-mannanase BcManA. Further a recombinant B. subtilis WB600 strain containing pMA5 vector carrying the BcManA gene was constructed. Different signal peptides and strong promoters were also used for the evaluation of high-level secreted production, along with the medium and inoculum amount optimization. Finally, the highest yield of β-mannanase reported to date by shake flask was obtained.

Strains, plasmids, and materials
Alkaliphilic strain B. clausii S10 which was used to clone the mannanase gene was isolated from a soda lake in Inner Mongolia of China. B. clausii S10 was preserved in the China General Microbiological Culture Collection Center (CGMCC 1.15293). The plasmids and bacteria strains used in this study for gene cloning and expression are listed in Table 1. Mannooligosaccharides (mannobiose, mannotriose, mannotetraose, mannopentose, and mannohexose) were purchased from Megazyme (Wicklow, Ireland). All the enzymes for DNA manipulations were purchased from NEB Inc. (Ipswich, MA, USA). Isopropyl-β-dthiogalactopyranoside (IPTG), ampicillin, kanamycin and imidazole were from Amresco Inc. (Solon, OH, USA). All other chemicals were of reagent grade.

Gene cloning of BcManA
The genomic DNA of B. clausii S10 was extracted using Bacterial DNA Kit (OMEGA Bio-tek, Norcross, USA), and the restriction enzyme Sau3A was employed to obtain randomly digested chromosomal fragments. The 4.0-to 8.0-kb fragments were recovered and purified by a Gel Extraction Kit (OMEGA Bio-tek) and ligated into pUC118 vector previously digested with BamHI and treated with calf intestinal alkaline phosphatase. The ligation product was then transformed into E. coli DH5α cells by electroporation transformation and plated onto Luria-Bertani (LB) agar plates containing 60.0 μg ml −1 of ampicillin and incubated at 37 °C overnight. Grown colonies were picked with sterilized toothpicks, plated in 150.0 μl LB medium containing ampicillin (60.0 μg ml −1 ) in sterilized 96-well microplates, and incubated overnight at 37 °C with shaking at 900 rpm. Fifty microliters of sterilized 60.0% glycerol was then added to each well, and plates were stored at − 80 °C to create the genomic DNA library.
Colonies in the microplate were spotted onto LBampicillin (60.0 μg ml −1 ) agar medium (pH 8.0) containing 0.5% locust bean gum by a sterile 96-pin replicator. After cultivating for 20 h at 37 °C, 1.0% sterile Congo red solution was softly poured onto the LB agar medium over the colonies. The colony which has transparent zone after 15 min soakage was considered to be the positive The mature BcManA encoding gene was expressed as a recombinant fusion protein containing an N-terminal His-tag in E. coli and was purified to near homogeneity by a two-step process including heat treatment and nickelnitrilotriacetic acid (Ni-NTA) affinity chromatography. Cells of about 150.0 ml culture of E. coli BL21 (DE3) harboring pET28a-ManA were harvested by centrifugation at 8000×g at 4 °C for 15 min, washed with binding buffer (20 mM Tris-HCl buffer containing 500 mM NaCl and 5 mM imidazole, pH 7.9) and then suspended in 20.0 ml of the same buffer. The suspended cells were disrupted by sonication and the supernatant was obtained by centrifugation at 16,000×g for 10 min at 4 °C. The supernatant was incubated at 60 °C for 20 min and then centrifuged at 16,000×g for 15 min at 4 °C. The second supernatant was loaded onto a pre-equilibrated Ni-NTA column filled with Ni-NTA His·Bind Resin (Novagen, CA, USA) followed by washing with 10.0 ml of binding buffer and subsequently with 15.0 ml of washing buffer (20 mM Tris-HCl buffer containing 500 mM NaCl and 60 mM imidazole, pH 7.9). Finally, the protein was eluted with 3.0 ml of elution buffer (20 mM Tris-HCl buffer containing 500 mM NaCl and 1 M imidazole, pH 7.9).
The obtained protein solution was desalted using a desalting column (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) with 20 mM Tris-HCl buffer (pH 7.5). The purity of the proteins was assessed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The protein concentration was determined by Quick Start Bradford protein assay (Bio-Rad, Hercules, CA, USA) with bovine serum albumin (0.125-1.0 mg ml −1 ) as standard.

Activity assay of BcManA
The enzymatic activity of BcManA was determined by measuring the release of reducing sugars by the 3,5-dinitrosalicylic acid (DNS) method with mannose as the standard. The reaction mixture comprised 190.0 μl of 50 mM glycine-NaOH (pH 9.5) containing 0.5% substrate (w/v), and 10.0 μl of appropriately diluted enzyme solution. The reaction was incubated at 70 °C for 10 min and stopped by adding 200.0 μl DNS reagent, followed by heating in a boiling water bath for 5 min. The absorbance at 540 nm was measured with a Spectra Max 190 Microplate Reader (Molecular Devices, Sunnyvale, CA, USA). One unit of enzyme activity is defined as the amount of enzyme required to release 1 μmol of mannose-reducing sugar equivalents per min. The activity measurements were repeated three times.
Influence of metal ions and some reagents on enzyme activity was determined by incubating the enzyme in 1 mM of metal ions (Hg 2+ , Mg 2+ , Na + , Mn 2+ , Zn 2+ , Cu 2+ , Pb 2+ , Ca 2+ , Ag + , Fe 3+ , K + , Rb + , Cs + , Fe 2+ , Ni 2+ and Co 2+ ) and EDTA by measuring the relative activity under the standard assay condition. The effect of some protein denatures such as SDS, on enzyme stability was determined by the residual activity after incubation them of different concentration with the enzyme of 0.1 mg ml −1 at 30 °C for 30 min.

Substrate specificity and kinetic parameters
The substrate specificity of purified recombinant BcManA was assayed by incubating the enzyme solution with 0.5% (w/v) substrates, including konjac glucomannan, locust bean gum, guar gum, sesbania gum, carboxymethyl cellulose, xylan, glucan, pectin, chitin, soluble starch and α-1,6-linked yeast mannan under standard reaction conditions (pH 9.5 and 75 °C). To detect the exoglycosidase activity, the p-nitrophenyl glycoside substrate (5 mM) was incubated with 1 μM enzyme for 30 min. The release of p-nitrophenol was monitored by measuring the absorbance at 405 nm. The hydrolysis kinetic parameters were determined at 75 °C in 50 mM glycine-NaOH buffer (pH 9.5) after a reaction time of 10 min using 0.1 μM enzyme and substrate of konjac glucomannan (0.3-5.0 mg ml −1 ), locust bean gum (0.5-6.0 mg ml −1 ), guar gum (1.0-8.0 mg ml −1 ) and sesbania gum (1.0-8.0 mg ml −1 ). The k cat and K m values were calculated by GraphPad Prism 5.0 software (http://www.graph pad. com/prism /) using non-linear regression. All data are expressed as the means of triplicate measurements.

Catalytic products analysis by thin-layer chromatography (TLC)
The time course of hydrolysis was performed using 10 µM enzyme and 2% (w/v) substrate at 65 °C. For konjac glucomannan and locust bean gum, 10 µl aliquots were taken out at 0 and 90 min and boiled for 2 min to stop enzymatic hydrolysis. Whereas for mannooligosaccharides, aliquots were taken out at 0, 10, 30, 60 and 90 min and stop the enzyme activity by boiling. The catalytic products were then analyzed by thin-layer chromatography (TLC) using Silica gel 60 F 254 (Merck, Darmstadt, Germany). The expansion solvent used was n-butanol/H 2 O/acetic acid (2:1:1, by vol). Oligosaccharides were detected by spraying the dried plates with a solution of acetone/diphenylamine/phenylamine/phosphoric acid (100 ml:2 g:2 ml:10 ml) and heating at 110 °C for 5 min.

Construction of plasmids for secreted expression in B. subtilis
For constructing the control plasmid pMA5-manA, the mature BcManA-encoding gene without original signal peptide was obtained by PCR using the primer pair of F1 (5′-CGC GGA TCC CAA AGC GGC TTT CAC GTA AAAG-3′, where the underline indicates the BamHI site) and R1 (5′-CTA GCT AGC TTA ATC ACG TTT GAG CCC ATT TTC -3′, where the underline indicates the NheI site). The PCR program was the same as the construction of plasmid pET28a-manA. The PCR product was purified using a Gel Extraction Kit (OMEGA Bio-tek) and then digested with BamHI and NheI for ligation to the pMA5 which were cleaved with the same enzymes. Then the ligation product was transformed into E. coli DH5α cells and plated onto LB agar plates containing 50.0 μg ml −1 of ampicillin and incubated at 37 °C overnight. Besides the original peptide of BcManA, signal peptides including SP amyL , SP lipA , SP nprB , SP nprE , SP lipB and SP amyE from B. subtilis 168 were also used for secreted expression. On the other hand, other two constitutive promoters including P43 and P lapS , and the inducible promoter of P grac were also used for BcManA expression. The fragment lengths of P43 and P lapS were 326 bp and 308 bp, respectively. P grac containing the promoter, lacO sequences, and the regulator protein lacI was 1548 bp and originated from the plasmid pHT43 plasmid.
The expression plasmids were constructed by PCR using the modified Gibson assembly method [46] base on the plasmid pMA5-manA. Primers used to amplify the signal peptide and promoter fragments for expression plasmid construction were listed in Additional file 1: Tables S1, S2. The I-5 ™ 2 × High-Fidelity Master Mix (Tsingke Biotech Co., Ltd, China) was used for PCR amplification. The primers for construction of plasmids with different signal peptides directly included the signal peptide sequences and the PCR protocols were as follows: denaturation at 98 °C for 2 min, followed by 30 cycles of denaturation at 98 °C for 20 s, annealing at 60 °C for 20 s and extension at 72 °C for 2 min, and a final extension at 72 °C for 5 min. The plasmid pMA5-manA was used as the template. The PCR product was purified by a Cycle-Pure Kit (OMEGA Bio-tek) and then digested by DpnI for 6 h. Then 2 µl of this the product and 0.5 µl of Taq DNA ligase were added into 7.5 µl of assembly master mixture [46], and this mixture was incubated at 50 °C for 1 h. Then the product was directly transformed into competent E. coli DH5α, and positive colony samples were selected to be further validated by sequencing (Tsingke Biotech Co., Ltd, China).
For the construction of plasmids with different promoters, the PCR amplification program was divided into two parts: the amplification of promoter sequence and the pMA5-manA sequence. The primer pairs of F1 and R1, F2 and R2 in Additional file 1: Table S2 were used for the promoter and pMA5-manA sequences amplification, respectively. The PCR protocol for promoter P43 and P lapS was as follows: denaturation at 98 °C for 2 min, followed by 30 cycles of denaturation at 98 °C for 20 s, annealing at 56 °C for 20 s and extension at 72 °C for 10 s, and a final extension at 72 °C for 5 min. For the second part of amplification, the PCR protocol was the same as that for construction of plasmids with different signal peptides. All the PCR products were purified by a Cycle-Pure Kit (OMEGA Bio-tek) and then digested by DpnI for 6 h. Then each 1 µl of these corresponding products and 0.5 µl of Taq DNA ligase were added into 7.5 µl of assembly master mixture followed incubation at 50 °C for 1 h. Similarly, these reaction products were directly transformed into E. coli DH5α for further validation by sequencing. All the confirmed recombinant pMA5-manA derivative plasmids were than transformed into B. subtilis WB600 cells by electroporation transformation as previously [47] for secreted expression in B. subtilis.

High-level secreted expression of BcManA in B. subtilis by shake flask
For gene expression in flask, 0.5 ml of seed culture of B. subtilis strains harboring appropriate expression plasmids were inoculated in 50 ml of 2 × SR medium [5] containing 60.0 μg ml −1 of kanamycin and 10 μg ml −1 of chloramphenicol in 500 ml flasks. The expression cultivation was performed at 37 °C with 230 rpm. For the determination of cell growth and β-mannanase production curves, the Horikoshi-I medium (polypeptone 5.0 g l −1 , yeast extract 5.0 g l −1 , glucose 10.0 g l −1 , K 2 HO 4 1.0 g l −1 , MgSO 4 ·7H 2 O 0.2 g l −1 , 10% Na 2 CO 3 solution 100 ml l −1 ) and modified Horikoshi-I medium (glucose was replaced by 10.0 g l −1 of locust bean gum) were used for B. clausii S10, while 2 × SR medium was used for B. subtilis WB600 and the recombinant B. subtilis WB600 containing BcManA gene. The OD 600 value was used to indicate cell biomass. The β-mannanase activity in supernatant and whole cell lysate was assayed under standard condition after cultivation. Furthermore, some cheaper nitrogen source such as soybean powder, peanut meal, peptone, NH 4 Cl and carbon source such as glucose, glycerol, soluble starch, konjac flour, corn starch and dextrin were also used to replace tryptone or yeast extract in 2 × SR medium, respectively. Similarly, 50 ml of different medium in 500 ml flasks was used for the medium component optimization and enzyme yield determination.
In the CAZy database, the members of glycoside hydrolase family 5 have been classified in 56 subfamilies due to the divergent amino acid sequence up to now (http:// www.cazy.org/GH5.html). Based on the sequence comparisons, BcManA belongs to the glycosyl hydrolase family 5, subfamily 8 (GH5_8). As previous study, eight conserved residues of all family 5 enzymes were demonstrated to be responsible for the catalytic activity [10]. The multiple sequence alignment showed that these essential active residues were also conserved in BcManA as R79, H115, N153, D154, H219, Y221, D249 and W278 (Fig. 1). Among these residues, the aspartate residues D154 and D249 are suggested to be the proton donor and the nucleophile, which was consistent with D158 and D254 in BSP165MAN belonging to GH5_8, respectively [48]. Frequently, β-mannanases carry extra non-catalytic modules which generally are the carbohydrate binding modules (CBMs) besides the catalytic module [1]. The CBM are thought to enhance enzyme activity toward cellulose conjugated mannan. However, no obvious CBM modules were found in BcManA or the five high homologous mannanases by the sequence analysis. Moreover, as an extracellular enzyme, the BcManA protein also includes a typical amino-terminal signal sequence with 28 amino acids (Fig. 1), which revealed that the BcManA mature protein contained 289 amino acids and had a calculated molecular mass of 34.3 kDa and an isoelectric point of 4.6.

Expression in E. coli and purification of recombinant BcManA
Using the shake-flask cultivation in LB medium, the total activity of the mature BcManA reached 343 U ml −1 with extracellular activity of 5 U ml −1 after 5 h induction using locus bean gum as substrate. The recombinant enzyme was purified by the above twostep process with a recovery activity of 40.9% and a final purity of about 97.5%. SDS-PAGE analysis showed that the purified recombinant mature BcManA has a molecular mass about 34 kDa (Fig. 2) which was consistent with the calculated molecular weight of 34.3 kDa. The reported Bacillus β-mannanases showed the majority of molecular mass from 30 to 70 kDa ( Table 2). The smallest and largest β-mannanases reported to date were found to be 22 kDa and 130 kDa from B. halodurans PPKS-2 and Bacillus sp. JAMB-750, respectively [49,50]. Moreover, after heating at 60 °C for 20 min, almost no BcManA activity loss was found, while most  of the E. coli proteins were denatured, which indicated that an appropriate heat treatment is efficient for the purification process. The specific activity of the purified BcManA was 2087.1 and 2366.2 U mg −1 with locust bean gum and konjac glucomannan as substrate, respectively. This activity was lower than the homologous β-mannanases of BSP165MAN from Bacillus sp. N16-5 (5065.0 U mg −1 ) [10], MAN5A from B. circulans CGMCC1554 (4839.0 U mg −1 ) [51] and BacK1MAN from B. circulans K-1 (3140.0 U mg −1 ) [52], but higher than BSP602MAN from Bacillus sp. strain JAMB-602 (287.0 U mg −1 ) [9]. In addition, although there are some β-mannanases which specific activity was higher than BcManA, the specific activity of BcManA was still higher than most other reported alkaline and Bacillus β-mannanases (Table 2).

Biochemical characterization of recombinant BcManA
The effect of pH and temperature on the recombinant BcManA was studied using locust bean gum as substrate. The recombinant BcManA had a pH optimum at 9.5 which was similar to β-mannanases from Bacillus sp. N16-5 [10] and Bacillus sp. strain JAMB-602 [9] but lower than those extreme alkaline enzymes from Streptomyces sp. CS428 (pH 12.5) [22], B. subtilis subsp. inaquosorum CSB31 (pH 12.5) [53], B. halodurans PPKS-2 (pH 11.0) [49], Bacillus sp. JAMB-750 (pH 10.0) [50] and B. cereus N1 (pH 10.0) [54]. By setting the activity at pH 9.5 as 100.0%, BcManA showed relative high activity at pH 7.0-10.5 and more than 50.0% of full activity was retained at that pH range. However, the relative activity declined rapidly at pH below 6.5 or above 10.5. There was only 32.5% and 34.1% of relative activity at pH 6.0 and 11.0 (Fig. 3a). BcManA was stable over an alkaline pH range from 6.0 to 11.5 (Fig. 3b), in which more than 55.0% of the original enzyme activity was retained after 6 h at 30 °C. However, the residual activity declined certainly when the pH values were lower than 5.5 or higher than 11.5, and only 20.5% and 25.8% of original activity was retained after pretreatment at pH 5.0 and 12.0, respectively (Fig. 3b). These indicated that BcManA was a typical moderate alkaline β-mannanase. An extremely alkaline mannanase (MnB31) from B. subtilis subsp. inaquosorum CSB31 was found to have the highest optimal pH of 12.5 and to be stable at pH range from 5.8 to 12.5, which was the most alkaline β-mannanase reported to date [53]. In addition, almost all the reported Bacillus β-mannanases displayed optimal pH at the range from 6.0 to 12.5 (Table 2), which indicated that Bacillus strains were the main sources of neutral and alkaline β-mannanases. Only the β-mannanases from Bacillus sp. MSJ-5 [55] and B. subtilis NM-39 [56] showed certain acidic enzymatic characteristics. Relative to neutral and alkaline β-mannanases, acidic β-mannanases mostly come from fungus, such as those from Penicillium oxalicum GZ-2 [30], Neosartorya fischeri P1 [32], Aspergillus niger LW-1 [29], Gloeophyllum trabeum [57], Phialophora sp. p13 [58], and Bispora sp. MEY-1 [59] showing optimal pH of 4.0, 4.0, 3.5, 2.5, 1.5, 1.5, respectively. Generally, most of reported β-mannanases showed optimal temperature of 40-75 °C [60]. There were only few β-mannanases that had optimal temperature above 75 °C, such as Man5 from Thermotoga maritime [18], MAN-P from A. niger [61], rMan5P1 from N. fischeri [32], Man5XZ3 from A. nidulans XZ3 [25], and ManBK01 from A. niger BK01 [27]. Man5 had optimal temperature of 90 °C which was the highest reported to date, whereas the other four fungus β-mannanases showed the highest activity at 80 °C. The optimal temperature of BcManA was 75 °C and more than 40.0% of the activity was maintained in the range of 50-85 °C (Fig. 3c). Although this optimal reaction temperature was lower than those five thermostable β-mannanases with pH optimum of 4.0-7.0, it was still the highest of all reported Bacillus and alkaline β-mannanases including the four highly homologous enzymes ( Table 2). Previous research indicated that most β-mannanases were stable at temperatures equal or less than 60 °C and the enzyme activity decreased rapidly at higher temperatures [60]. The BcManA showed good thermostability and almost no activity loss was found after 9 h incubation at 50 °C and 60 °C (Fig. 3d). Actually, no obvious activity loss was detected even after incubating BcManA at 50 °C for 52 h (data not shown). For incubation at 60 °C, the obvious activity loss appeared after incubation for more than 36 h, and about 68.1% of the original activity retained after incubation for 48 h. There was even about 80.0% and more than 50.0% of the original activity retained after incubation at 70 °C for 1 h and 3 h, respectively. The residual activity decreased rapidly when incubation at 80 °C and only about 50.0% and 32.0% activity retained after 0.5 and 1 h incubation, respectively. Although the three fungi β-mannanases including rMan5P1, Man5XZ3 and ManBK01 had higher optimal temperature, they showed less thermostability than BcManA. Only 50.0% activity was retained after incubating rMan5P1 at 70 °C for 10 min [32]. Meanwhile, about 70.0% activity of Man5XZ3 retained after 180 min incubation at 60 °C [25], whereas only 50.0% activity of ManBK01 retained after 15 min incubation at 80 °C [27]. Only MAN-P showing 61.0% residual activity after incubation at 80 °C for 120 min [61] and Man5 with half-life of 38 min at 90 °C [18] displayed better thermostability than BcManA. Furthermore, by comparison with other β-mannanases, the thermostability of BcManA was also better than all reported alkaline β-mannanases and most of Bacillus β-mannanases ( Table 2) except for that from B. subtilis TBS2 which showed about 60% activity retained at 80 °C [60]. β-mannanases have many applications in the industrial processes, such as paper industry, food industry, oil drilling, detergent industry, hydrolysis of coffee extract, textile and cellulosic fiber processing industry [1]. Among these applications, the activity and stability of β-mannanase at alkaline and thermal condition was crucial especially for enzymatic breaking of hydraulic fracturing fluids in oil drilling industry, biobleaching of pulp and paper in the paper industry, and enzymatic scouring and desizing processes in the textile industry [10,34]. Overall consideration of pH and temperature performance, BcManA distinguished from other reported β-mannanases and was a promising thermoalkaline β-mannanase for potential industrial application.
The effect of metal ions and chemicals on the activity of BcManA was evaluated under standard condition (Table 3). Hg 2+ , Pb 2+ , Ag + , Zn 2+ , Fe 2+ , and Fe 3+ strongly inhibited the activity of BcManA, whereas Cu 2+ , Mn 2+ , Ni 2+ and Co 2+ partially inhibited BcManA. Heavy metal irons are known to have strong affinities for sulfhydryl (-SH) groups and normally act as an irreversible inhibitor for enzyme activity. Similarly, heavy metal irons such as Ag 2+ and Hg 2+ also strongly inhibited activity of other highly homologous β-mannanases except for BacK-1MAN which activity was not affected by Ag 2+ [52]. Furthermore, Pb 2+ showed strong inhibition to BcManA and BSP602MAN but no affection on BSP165MAN [9,10]. Also Fe 2+ and Fe 3+ which strongly inhibited BcManA just showed unobvious affection on BSP165MAN and BacK1MAN [10,52]. The other tested metal ions and  (Table 3). Differently, only 27.0% original activity of MAN5A was retained after incubation with 1 mM EDTA [51]. However, BcManA was sensitive to guanidine hydrochloride and only 20.5% activity retained following incubation in 1.0 M guanidine hydrochloride (Table 3).

Analysis of hydrolysis product by TLC
The hydrolysis products from LBG and KGM by BcManA were analyzed by TLC. A mixture of mannose oligosaccharides consisting of mannose (M1), mannobiose (M2), mannotriose (M3), mannotetraose (M4), mannopentose (M5), and mannohexose (M6) were used as standard markers. As shown in Fig. 4a, the products hydrolysed from LBG to KGM were mainly oligosaccharides with various sizes after 90 in cumulative action, which indicated that BcManA was an endo-acting β-mannanase.
In addition, small amount of mannose was also found in the hydrolysed products, which was similar to the β-mannanases from Bacillus sp. N16-5 [ The catalytic activity of BcManA on the manno-oligosaccharides was also assayed. BcManA showed no activity on M2 (Fig. 4b) and very low activity on M3 (Fig. 4c). For M4-M6, the activity increased along with the degree of polymerization (DP) increase and the efficient hydrolytic activity was found on M5 and M6 (Fig. 4c-f ). The products mainly were manno-oligosaccharides with DP of ≥ 2 and only very few mannose was detected. These were similar to the β-mannanases from B. subtilis WL-3 [43], B. subtilis WL-7 [73] and Bacillus sp. AM-001 [74], but different from those from B. subtilis WY34 [67], Bacillus sp. JAMB-602 [9], Enterococcus casseliflavus FL2121 [21], and Pseudomonas sp. strain PT-5 [16] which only hydrolysed manno-oligosaccharides with DP ≥ 4. In addition, no products with higher DP than the substrates were found, which indicated that BcManA had no transglycosylation activity. Whereas, some β-mannanases Fig. 4 Thin-layer chromatography (TLC) analysis of hydrolysis products from mannopolysaccharides and manno-oligosaccharides. a Hydrolysis products from locust bean gum (L) and konjac glucomannan (K); b-f time course of manno-oligosaccharide degradation. b mannobiose; c mannotriose; d mannotetraose; e mannopentose; f mannohexose. M1-M6, the standard marker, mixture of manno-oligosaccharides. The numbers on the graphs indicate the reaction time such as that from B. subtilis WL-3 showed transglycosylation activity when manno-oligosaccharides with DP ≥ 3 as substrates [43]. Manno-oligosaccharides are generally supplemented in animal feed to improve digestibility and maintain gastrointestinal status, and also used as food additives for humans to promote selective growth of beneficial intestinal microflora [53,67]. As previous reports, the production of manno-oligosaccharides by acid hydrolysis of mannan is neither environmentally friendly nor cost-effective, whereas the use of alkalithermostable endo mannanase will further improve the yield of manno-oligosaccharides due to high solubility of mannan in alkaline media and reduction in viscosity of the reaction mixture [22]. So BcManA could be a good candidate for manno-oligosaccharides production from cheap mannan such as konjac power and locust bean gum under alkaline and high temperature condition efficiently.

Signal peptides and promoters determination for efficient secreted expression of BcManA in B. subtilis
Compared to E. coli which was extensively used for protein expression, B. subtilis displayed significant advantages in efficient secretion capacity of functional extracellular proteins directly into culture medium, which made downstream separation and purification processes much more simple [5,44]. In this study, we used the B. subtilis WB600, a protease-deficient strain, for secreted expression of BcManA. The multicopy plasmid pMA5 which containing the constitutive promoter P hpaII was used as the expression vector. Firstly, a suitable and efficient signal peptide is necessary for protein secretory expression in B. subtilis [5]. Five different signal peptides (SP lipA , SP amyE , SP lipB , SP aprE , SP amyL ) from the general secretory (Sec) pathway in B. sutilis and the original signal peptide of BcManA (SP ori ) were selected for BcManA secretory expression. After 48 h cultivation in 2 × SR medium by flask shaking with 1% (v/v) inoculation of overnight culture, significant protein bands were found on SDS-PAGE from the culture supernatant (Fig. 5). All the protein banks showed molecular weight about 34 kDa which was in accordance with the calculated molecular weight of the mature BcManA. This indicated that all of the used signal peptides secreted BcManA into the medium successfully. Among these six signal peptides, different level of extracellular activity was detected and the highest activity of 763 U ml −1 was obtained with secretory rate of 95.8% by B. subtilis WB600-3, which indicated that SP lipA was the most efficient signal peptide for BcManA expression in B. subtilis WB600 in this study (Table 5). This was similar to the β-mannanase from B. licheniformis DSM13 which also showed the highest extracellular activity of 533 U ml −1 after 72 h cultivation by SP lipA which however, was lower than BcManA [5]. Unexpectedly, B. subtilis WB600-1 containing the expression vector without signal peptide also showed high extracellular activity of 571 U ml −1 after 48 h cultivation with the secretory rate of 59.6%. We detected the β-galactosidase activity in the culture supernatant and the whole cell lysate of B. subtilis WB600-1, and about 60% of total β-galactosidase activity (the activity in the whole cell lysate was set to 100%) was found in culture supernatant (data not shown). This result indicated that the high extracellular β-mannanase activity in B. subtilis WB600-1 culture supernatant might be due to cell lysis after 48 h cultivation.
Besides the signal peptide, the promoter was another factor for efficient protein expression. So using strong promoters was one of the most efficient methods to increase the proteins production [5,44]. The original promoter P hpaII in pMA5 was a relatively strong promoter. In order to further increase the expression of BcManA, two other strong promote P43 and P lapS were selected. Promoter P43 was a well characterized constitutive strong promoter and widely used in expression of many proteins in B. subtilis [5,44]. Whereas, promoter P lapS was an artificial hybrid promoter containing P luxS and the − 10 region from B. licheniformis and the − 35 region of P Apr from B. subtilis, which was confirmed experimentally to be stronger than P43 for β-Gal secretory expression in B. subtilis [75]. The strains B. subtilis WB600-8 and B. subtilis WB600-9 containing these two strong promoters and SP lipA were also cultured in 2 × SR medium by flask shaking at 37 °C for 48 h, and the extracellular activity was assayed ( Table 5). The highest enzyme activity of 908 U ml −1 was produced by B. subtilis WB600-8 (P43) which was higher than that produced by B. subtilis WB600-3 (P hpaII ). Similar result was also found in the expression of B. licheniformis DSM13 β-mannanase and the activity of 1274 U ml −1 and 711 U ml −1 was produced by P43 and P hpaII after 72 h cultivation, respectively [5]. The activity of 222 U ml −1 was determined from B. subtilis WB600-9 (P lapS ) which was only about 24.4% of that from B. subtilis WB600-8 (P43). However, for the expression of β-Gal, the enzyme production driven by P lapS was more than 10 times of that by P43 [75]. This indicated that the strongest promoter P lapS might be not suitable for BcManA expression in B. subtilis WB600. So it shows that promoter with suitable strength is important for protein expression. Further, we determined the cell growth and BcManA production curve of B. subtilis WB600, B. subtilis WB600-8 and the original strain B. clausii S10 in different medium (Additional file 1: Fig. S2). No mannanase activity was detected in the B. subtilis WB600 and B. clausii S10 culture supernatant during the entire measuring period. However, small amount of activity was found when B. clausii S10 was cultured in the modified Horikoshi-I medium containing locust bean gum, which indicated that the mannanase production in B. clausii S10 was inducible. The BcManA yield reached to plateau after 72 h cultivation for both of B. subtilis WB600-8 and B. clausii S10. And the extracellular activity from B. subtilis WB600-8 reached 2374 U ml −1 with secretory rate of 98.5% at 72 h, which was much higher than that of native BcManA from B. clausii S10 (75 U ml −1 ). Also this enzyme activity was already higher than that produced by other reported B. subtilis strains by shake flask (Table 6) [5,43,44,76].

Optimization for high-level secreted expression of BcManA in B. subtilis
The medium component was important for industrial enzyme production by microbes. In 2 × SR medium, yeast extract and tryptone which were expensive and generally used only in laboratory were the carbon and nitrogen source. In order to promote the industrial production performance of the recombinant B. subtilis WB600-8 for BcManA, different cheap carbon and nitrogen sources were selected for evaluation. Firstly, peptone, soybean powder, peanut meal and the inorganic nitrogen of NH 4 Cl with 3% (w/v) concentration was used to replace tryptone for BcManA expression. As shown in Fig. 6a, obvious extracellular activity was detected after 48 h cultivation and the highest activity (1050 U ml −1 ) was found by using peanut meal which was slightly higher than that by using tryptone (908 U ml −1 ). When the culture time extended to 72 h, a higher activity using peanut meal reached 2645 U ml −1 compared with 2374 U ml −1 using tryptone. Peanut meals are cheap and abundant byproducts of oil extraction from peanut seeds. So they can be used as cheap nitrogen source for β-mannanase production by recombinant B. subtilis.
Using peanut meal as nitrogen source, different carbon source such as glucose, glycine, corn starch, soluble starch, konjac flour and dextrin with 2% (w/v) concentration were used instead of yeast extract for BcManA expression. After 48 h cultivation, the highest activity of only 69 U ml −1 was determined using glucose as carbon source among these modified mediums, which was at very low level compared with that using yeast extract. These results indicated that yeast extract might be important for BcManA production in B. subtilis. On the other hand, we found that the cell concentration in the glucose medium was observably higher than that in the yeast extract medium, along with pH decrease. This phenomenon indicated that glucose in medium could improve B. subtilis cells growth to high biomass, however, which resulted in more acid by-products and low level of protein expression. Nevertheless, yeast extract should be partially substituted by cheap carbon source such as glucose to balance the protein expression and cell growth. Different proportions substitution of yeast extract by glucose was performed with the total concentration of 5% (w/v). As shown in Fig. 6b, the highest activity of 3804 U ml −1 was found after 72 h cultivation in the medium containing 3% yeast extract and 2% glucose, which was about 2.5 times as much as that only using 5% yeast extract. Further, the effect of peanut meal concentration on BcManA expression was also evaluated. As shown in Fig. 6c, the highest yield of 4599 U ml −1 was obtained after 72 h when 5% peanut meal was used in the modified medium. In addition, the inoculation amount is also an important fact for protein expression. The extracellular enzyme yield under different inoculation amount was detected using the optimally modified 2 × SR medium (5% peanut meal, 3% yeast extract, 2% glucose, 0.6% K 2 HPO 4 , pH 7.2). Finally, the highest activity was determined to be 6041 U ml −1 after 72 h cultivation when the inoculum size of 6% (v/v) was used (Fig. 6d), which was about 80 times that of native BcManA from B. clausii S10.
In the past decade, many strains such as Bacillus sp., P. pastoris, E. coli, Aspergillus sp. and some least frequently used strains have been applied for β-mannanases production (Table 6). Among these reports, B. subtilis and P. pastoris are the most commonly used host strains for β-mannanases expression, and most of the β-mannanases had neutral or acid optimal pH. Song et al. reported the secreted expression of β-mannanase from B. licheniformis DSM13 in B. subtilis 1A751 through protein synthesis and secretion optimization strategy, and they improved the final production to 2207 U ml −1 by shake flask which was considered to be the highest among the reported Bacillsu strains [5]. Our result showed a yield of 6041 U ml −1 which was about 2.7 times as much as that expressed in B. subtilis 1A751, which indicating the highest β-mannanase yield produced by Bacillus strains by shake flask. Compared to Bacillus, much more secreted expression of β-mannanase was performed in P. pastoris strains ( Table 6). The highest yield by shake flask was 1114 U ml −1 with productivity of only 7.7 U ml −1 h −1 after 144 h cultivation in P. pastoris GS115 using pPIC9K vector [55]. Besides these commonly used strains, some other host strains such as A. niger [77], A. sojae [26], Yarrowia lipolytica [78], Lactobacillus plantarum [79] and Trichoderma reesei [31] were also used for β-mannanases production. The extracellular activity of β-mannanase from A. aculeatus MRC11624 was found to be 16,596 nkat ml −1 (about 996 U ml −1 ) after 192 h cultivation in A. niger D15 [77]. Higher activity of 2460 U ml −1 with productivity of 14.6 U ml −1 h −1 was obtained for ThMan5 after 72 h cultivation in T. reesei QM9414A [31]. In the present study, combined with selection of suitable signal peptide for efficient secretory, strong promoters to enhance the transcription and translation, and optimization of medium for strain growth and protein expression, the extracellular activity of BcManA reached 6041 U ml −1 with productivity of 83.9 U ml −1 h −1 after 72 h cultivation by flask shake, which was higher than those of not only Table 6 Secreted production of recombinant β-mannanases in different strains by shake flask a The time is the induction time b The activity defined as the amount of enzyme required to convert one mole of substrate per second and reported in katal (nkat ml −1 ) Bacillus strains but also all other reported strains by flask shake fermentation. Also the modified 2 × SR medium was cheaper than the normally used 2 × SR medium and could be applied for other protein expression in B. subtilis. Moreover, a much higher production of BcManA would be obtained by cultivating this strain in fed-batch fermentors with process optimization.

Conclusions
The thermo-alkaline β-mannanase (BcManA) gene from B. clausii was cloned and heterologous expressed. The purified recombinant BcManA showed high catalytic activity on konjac glucomannan and locust bean gum under alkaline and high temperature conditions. Overall consideration of pH and temperature performance, BcManA distinguished from other reported β-mannanases and was a promising thermo-alkaline β-mannanase for potential industrial application. Combined with medium optimization and selection of suitable signal peptide and strong promoters, the extracellular BcManA yield of 6041 U ml −1 was obtained in B. subtilis WB600 with constitutive expression vector, which representing the highest yield reported to date by flask shake. This would significantly cut down the production cost of this enzyme and provide a guideline for secretory expression of other β-mannanases in B. subtilis. Fig. 6 Extracellular β-mannanase activity of the recombinant B. subtilis WB600-8 cultured in 50 ml of various modified 2 × SR mediums. a Different nitrogen sources with 3% concentration in 2 × SR medium; b different proportions substitution of yeast extract (YE) by glucose (G) in the modified 2 × SR medium with 3% peanut meal as nitrogen source; c different peanut concentration in the modified 2 × SR medium with 3% yeast extract and 2% glucose; d different doses of inoculation in the final modified medium containing 5% peanut meal, 3% yeast extract, 2% glucose and 0.6% K 2 HPO 4 . Error bars represent standard deviations