Cohnella amylopullulanases: Biochemical characterization of two recombinant thermophilic enzymes

Some industries require newer, more efficient recombinant enzymes to accelerate their ongoing biochemical reactions in harsh environments with less replenishment. Thus, the search for native enzymes from extremophiles that are suitable for use under industrial conditions is a permanent challenge for R & D departments. Here and toward such discoveries, two sequences homologous to amylopullulanases (EC 3.2.1.41, GH57) from an endogenous Cohnella sp., [Coh00831 (KP335161; 1998 bp) and Coh01133 (KP335160: 3678 bp)] were identified. The genes were heterologously expressed in E. coli to both determine their type and further characterize their properties. The isolated DNA was PCR amplified with gene specific primers and cloned in pET28a, and the recombinant proteins were expressed in E. coli BL21 (DE3). The temperatures and pH optima of purified recombinants Coh 01133 and Coh 00831 enzymes were 70°C and 8, and 60°C and 6, respectively. These enzymes are stable more than 90% in 60°C and 50°C for 90 min respectively. The major reactions released sugars which could be fractionated by HPLC analysis, from soluble starch were mainly maltose (G2), maltotriose (G3) and maltotetraose (G4). The enzymes hydrolyzed pullulan to maltotriose (G3) only. Enzyme activities for both proteins were improved in the availability of Mn2+, Ba2+, Ca2+, and Mg2+ and reduced in the presence of Fe2+, Li2+, Na2+, Triton X100 and urea. Moreover, Co2+, K+, and Cu2+ had a negative effect only on Coh 01133 enzyme.


Introduction
Costs associated with food and beverage enzymes are expected to reach $1.7 billion US dollars by 2018 [1]. Among the diverse relevant enzymes, starch hydrolyzing enzymes (namely αamylases, α-amylase-pullulanases, and amylopullulanases) have found their way into starch and baking industries as catalysts [2]. These enzymes fall into two families of glycosyl hydrolases, GH13 and GH57, and both having four conserved regions that classify them within the family of α-amylases [3,4]. Pullulanases are being used in industrial applications, such as onestep liquefaction-saccharification towards the production of sugar syrups and as an anti-staling agent in the baking industry, and are mainly produced by thermophilic bacteria and archaea [2,3]. This enzyme not only has application in food and beverage industries, it is being used within ethanol, detergents, dishwashing, laundry detergents, textile, and pulp and paper industries as well. Starch enzymatic alteration into sugar syrups is done at high temperature, in liquefaction, and saccharification steps follows at 60˚C. Furthermore starch bioprocessing at higher temperature improves starch solubility, restricts microbial contamination, reduces its viscosity, decreases reaction times and more economical [5]. Therefore it seems necessity to discover suitable thermoactive and thermostable enzyme which improves saccharification rate and the other process yield [6]. Amylopullulanases (EC 3.2.1.41) are classified into two groups: Type I that cleaves α-(1!6) linkages in pullulan and branched oligosaccharides such as amylopectin, producing maltotriose and unbranched oligosaccharides, and Type II pullulanase or amylopullulanase that has both amylase and pullulanase activity [7,8]. The latter cleaves starch glycosidic bonds at both α-(1!6) and α-(1!4) linkages, releasing a remaining polysaccharide. In contrast, α-amylase cleaves only α-(1!4) linkages in varieties of substrates such as starch, glycogen, and cyclodextrins [9]. According to the number of active sites within the enzyme amylopullulanase can be divided into two subgroups [10]. Most thermophilic amylopullulanases possess one active site [11].
Here, two amylopullulanases from an endogenous Cohnella sp. A01 toward their identification and biochemical characterization were heterologously expressed in E. coli in the search for industrial enzymes. The enzyme features were compared with some previously known orthologs, and phylogenetic analysis were carried out to define their relatedness to either GH13 or GH57.

Bioinformatic analysis
Amylopullulanase and α-amylase protein sequences of GH13 and GH57 were obtained from CaZy (http://www.cazy.org; Table A in S1 File). Multiple sequence alignment was performed with the sequences using ClustalW. A phylogenetic tree was established using neighbor-joining with 1000× bootstrap value in MEGA5 [13]. Relative molecular weights and isoelectric points were predicted using ProParam at ExPASy (http://web.expasy.org/protparam/) [10].

Cloning and expression of amylopullulanases
A single clone of Cohnella sp. A01 was cultured in NB at 60˚C for 72 h. Genomic DNA was isolated using a DNA extraction kit. Coh 01133 gene (3678 bp) and Coh 00831 gene (1998 bp) were PCR amplified using two gene specific primer pairs containing properly engineered restriction sites that were used for following restrictions and ligations as below: The amplicons were digested with BamH I and Nhe I in Coh 00831, and Nde I and Sal I in Coh 01133. The restricted fragments were cloned in pET28a using 1 U T4 DNA ligase and transformed in BL21 (DE3) [14]. The T 7 promoter was induced with 1 mM IPTG for 4 h at 22˚C.

Purification of recombinant amylopullulanases
Coh 01133 amylopullulanase was purified with Ni-NTA Sepharose affinity chromatography at 4˚C according to the Invitrogen protocol. Coh 00831 amylopullulanase was purified via anion exchange chromatography using DEAE-Sepharose column (2 × 20 cm). A single step ion exchange chromatography method has been developed for purification of Coh 00831 amylopullulanase. The crude extract was dissolved in phosphate buffer (10 mM) having a pH between 6 and 8 and with an increase of pH 0.5. The samples were added to an anion exchange matrix using DEAE-Sepharose equilibrated with 10 mM phosphate buffer at the corresponding test pH. The amount protein and amylopullulanase activity of the supernatant were determined. The buffer ionic strength effect on adsorption was also specified by varying concentrations of phosphate (10, 20, 30 and 40 mM) at pH 8. NaCl solution optimum ionic strength was determined by eluting the adsorbed protein using a 0.1 to 1.5 M solution.
After finding the optimum parameters, DEAE-Sepharose column equilibrated with 20 mM phosphate buffer, pH = 8.0 [15,16]. The proteins were eluted with 0-0.5 M NaCl gradient, and eluents were monitored at 280 nm and further tracked via SDS-PAGE. Both proteins were dialyzed against a 50 mM sodium acetate buffer (pH = 5.5).

Recombinant amylopullulanases enzyme assay
Amylopullulanase activity assay was performed using spectrophotometric method based on Millers method [17] while pullulan and starch were used as enzyme substrate. Production of reducing sugar in the reaction mixture shows the enzyme activity by addition of 3,5-dinitrosalicylic acid (DNS). The reaction mixture (0.5 ml) containing 1% (w/v) substrate solution, 50 mM sodium acetate buffer (pH = 5.5), and an enzyme source was incubated at 60˚C for 30 min. One unit of pullulanase activity is defined as the amount of enzyme which produces 1 μmol of reducing sugar (with glucose as the standard) per minute under assay conditions. pH profiles were performed in 50 mM mixed buffer containing sodium phosphate, sodium acetate, and glycine. For pH stability, enzymes were incubated for 90 min at pH = 3-12 (5mM mixed buffer) and the relevant activities were monitored at 540 nm; data was presented using Prism5.
Furthermore, in the temperature profile, relative enzyme activities were determined in temperatures ranging 20-100˚C. Meanwhile, enzymes were incubated at 20-100˚C for 90 min at their optimum pHs. The heat-treated enzymes were used for assay. Following the addition of DNS, enzyme activities were determined at 540 nm to perform temperature stabilities. The relevant graphs were prepared with Prism5.
The products from soluble starch and pullulan hydrolyzed by the amylopullulanases were examined with HPLC (Agilent 1260 A) using an Aminex HPX-87H column (Bio-Rad, USA) connected to a RID detector. The products were eluted with a mixture of acetonitrile: doubledeionized HPLC-grade water (60: 40 [V/V]) as solvent at a flow rate of 1 ml per min and column temperature of 20˚C.
Kinetic parameters for enzyme catalysis for both enzymes using pullulan and starch as substrates were obtained by Prism5.

Zymogram analysis
Isolated recombinant enzymes were separated on a 10% native-PAGE containing 1% (w/v) pullulan. The gel was washed with 2.5% Triton X100 and incubated for 1 h with the proper buffer providing optimum pH for each enzyme, 50 mM either sodium acetate (pH = 6.0) or sodium phosphate (pH = 8.0), at 4˚C. The gels were incubated at 60˚C for 1 h; then Lugol's reagent was added to the gel, and enzyme activity was photographed.  (Table 1) [3,18,19]. The deduced amino acid sequences of the amylopullulanases were compared with the sequences of the related enzymes available through GenBank (Table A in S1 File). Cohnella sp. A01 00831 amino acid sequence had 58% identity with the reported amylopullulanase from D. turgidum, while the highest percentage identity with Cohnella sp. A01 01133 amino acid sequence was noted with amylopullulanase from M. pulmonis (47%) ( Table B in S1 File). Phylogenetic analysis of Coh 01133 amino acid sequence and Coh 00831 amino acid sequence within the bacterial tree indicated that these two proteins belong to a single clade grouped with GH57 (Fig 1). No significant similarity found between Coh 00831 and Coh 01133 amino acid sequences. Coh 01133 and Coh 00831 enzymes were predicted to have relative molecular weights of~127 and~70 kDa with isolectric points of 6 and 5.37, respectively.

Expression and purification of the recombinant pullulanase
Coh 01133 and Coh 00831 genes were PCR amplified (Fig 2A and 2B, lanes 2), cloned in pET28a, expressed in BL21 (DE3) in the presence of 1 mM IPTG at 22˚C, and purified (Fig 2 and Figure A in S1 File.). Although both recombinant enzymes were capable of hydrolyzing both starch and pullulan (Table C in S1 File), they were more efficient with starch: k cat /K m = 0.20 × 10 4 and 16 × 10 4 in contrast to pullulan 0.14 × 10 4 and 0.07 × 10 4 for Coh 00831 enzyme and Coh 01133 enzyme, respectively. Enzyme activities were demonstrated by zymogram analysis in the presence of pullulan (Fig 2A and 2B, lanes 7). Coh 01133 enzyme was purified 7.09-fold, with a yield of 53% via affinity column chromatography using Ni-NTA Sepharose, and had a specific activity of 52.79 U/mg ( Table 2). The enzyme had a higher relative activity at a pH range of 5-9 with optimum activity at pH = 8.0 (Fig 3A, "left panel"). The enzyme kept 92% of its activity at pHs 7.0, 8.0 and 9.0 for a duration of 90 min (Fig 3B, "left panel"). It was active at temperatures higher than 45˚C up to 90˚C with the optimum temperature being 70˚C (Fig 3C, "left panel"). The analysis of Coh 01133 enzyme temperature stability demonstrated that the enzyme keeps 90% of its activity at temperatures between 20-60˚C for 90 min with decline at temperatures above 60˚C (Fig 3D, "left panel").
Coh 00831 enzyme was purified 23.04-fold, with a yield of 24% via anion exchange chromatography using DEAE-Sepharose, and had a specific activity of 118.84 U/mg ( Table 2). Recombinant Coh 00831 enzyme had greater activity at pHs ranging 4-9 with optimum activity noted at pH = 6.0 (Fig 3A, right panel). The enzyme was relatively stable at pHs 5.0, 6.0, and 7.0 ( Fig  3B, right panel). The enzyme was active at temperatures ranging 40-70˚C with the highest activity at 60˚C (Fig 3C, right panel). The enzyme is stable more than 92% in 50˚C for 90 min and kept 85% of its activity at 60˚C, once incubated in a range of temperatures (20-100˚C) for 90 min (Fig 3D, right panel).
The enzymes' kinetic parameters in the presence of two substrates, pullulan and starch, were calculated with Prism5. In Coh 01133 enzyme, the K m for pullulan was determined to be 0.28 mg/ml, and that for starch was 0.279 mg/ml. From the Lineweaver-Burk plot, the V max values for pullulanase were calculated to be 5.099 μmol/min for pullulan and 10.45 μmol/min for starch. In Coh 00831 enzyme, the K m for pullulan was 0.21 mg/ml and for starch was 0.1898 mg/ml, and the V max values were calculated to be 4.375 μmol/min for pullulan and 5.647 μmol/min for starch. The kinetic parameters (K m , V max , k cat and k cat /K m ) of the enzymes have been summarized in Table C in S1 File. The k cat value of the Coh 01133 enzyme toward pullulan substrate was smaller compared with Coh 00831 enzyme. In contrast, Coh 00831 enzyme had a lower k cat value for soluble starch compared with those of Coh 01133 enzyme (Table C in S1 File).

Effect of some metal ions and chemical materials
Among the metal ions and detergents tested, Mn 2+ , Ba 2+ , Ca 2+ , and Mg 2+ improved the activity of both enzymes, while Fe 2+ , Li 2+ , Na 2+ , Triton X100, and urea caused sharp reductions in enzyme activity. More specifically, Co 2+ , Cu 2+ , and K + in Coh 01133 amylopullulanase reduced enzyme activity greatly. EDTA reduced the activities of both enzymes, specifically at 5 mM. In contrast, iodoacetamide had no major effect on enzyme activity (Table 4).

Discussion
Industrial enzymes have shown great promise for many aspects of human life for the last half century or so; they can improve catalytic reaction rates, considering environmental issues as opposed to chemical biocatalysts. Starch byproducts resulting from enzymatic hydrolysis have  many applications in the food and pharmaceutical industries. Moreover, the use of starch hydrolyzing enzymes such as α-amylases and amylopullulanases in detergent industries compared with other enzymes has improved the efficiency of commercial enzyme complexes. It is apparent that the industry is booming in its different fields, and accordingly looking into other more efficient enzymes with greater stability. Such enzymes are generally obtained from extermophilic microorganisms like archaea, bacteria, and fungi. Among bacteria, some species of Geobacillus and Bacillus are known to produce thermophilic amylopullulanase [20][21][22].
Research on the thermophilic amylopullulanases is attractive not only for understanding the enzyme stability mechanisms, but also for finding better enzymes with more efficient usage in industrial process [23]. Here and in the quest for such enzymes, a native thermophilic bacterium has been isolated, namely Cohnella sp. A01 (accession number JN208862.1), from shrimp ponds in southern parts of Iran. Following genome sequencing (data not yet published), two amylopullulanase genes, Coh 00831 gene and Coh 01133 gene, were PCR amplified and used for heterologous protein expression and further biochemical characterizations. The enzymes demonstrated themselves to be thermophilic as expected from the niche from which the bacterium was isolated. The K m values for the two recombinant enzymes were the lowest for pullulan and starch when compared with previously reported homologues from other bacteria ( Table C in S1 File), indicating a greater affinity toward the substrates. Here, the findings are discussed for each gene separately, and the results are compared with other bacterial homologs.
Coh 00831 amylopullulanase: Phylogenetic analysis of this protein with bacterial members of families GH13 and GH57 demonstrated that Coh 00831 enzyme was well-grouped with GH57 and was closer to the amylopullulanases of this family (Fig 1). Almost all mesophilic amylopullulanases belong to GH13 family and the thermostable counterparts come under GH57 or GH13 family [24]. Coh 00831 enzyme with 666 aa is among the common bacterial amylopullulanases, considering its relative molecular weight (70 kDa). The recombinant enzyme had an optimum temperature of 60˚C, higher than a few previously reported bacterial amylopullulanases (Table C in S1 File). The enzyme is stable more than 92% in 50˚C for 90 min. Most of thermostable amylopullulanase were active at acidic or neutral pH while Coh 00831 amylopullulanase was active at pHs ranging 4-9, reaching an optimum at pH = 6.0. K m values were illustrative of no differences between enzyme affinity toward either starch or pullulan (Table C in S1 File). However, k cat was smaller for pullulan, indicating the enzyme's preference toward starch as substrate. The purified Coh 00831 enzyme hydrolyzed pullulan to generate G3 and hydrolyzed soluble starch to produced G2, G3, G4 and G6, which are nearly the same as those of other amylopullulanases [25][26][27]. Among the cations tested, Mg 2+ , Ca 2+ , and Mn 2+ improved enzyme activity by more than 50%. The cationic inhibitors were Na + , K + , Ni 2+ , Fe 2+ , Zn 2+ , Cu 2+ , Co 2+ , and Sr 2+ (Table 4). In general, the used detergents reduced enzyme activity. Enzyme inactivates by 6 M urea. EDTA had very little effect on the enzyme activity. Some common chelating agents used in industrial cleaning compounds include EDTA, phosphates and sodium citrate. Iodoacetamide did not inactivate the enzyme, indicating no cysteine residue within the enzyme active site (Table 4).
Coh 01133 amylopullulanase: Phylogenetic analysis of this protein, similar to Coh 00831 enzyme, grouped the protein sequence with GH57, next to other amylopullulanases (Fig 1) similar to mentioned above. Coh 01133 enzyme with 1226 aa and a relative mass of 125 kDa is among the large amylopullulanases of GH57, similar to proteins from Thermococus Sp. (Table C in S1 File). The enzyme's optimum temperature was 70˚C; accordingly, it can be classified as a thermophilic amylopullulanase. However, compared to other previously characterized amylopullulanases, this enzyme falls within the mid-range. It is suggested that thorough comparative studies with other known GH57 members (Table C in S1 File) be conducted toward engineering this enzyme to improve the optimal temperature. Given that most amylolitic industrial enzymes need to be active at temperatures higher than 50˚C, Coh 01133 amylopullulanase can be considered a suitable industrial enzyme. The enzyme is stable more than 90% in 60˚C for 90 min. Earlier studies were indicative of a pH stability of 5.5-6 for most pullulanase in GH57 [28][29][30][31]. In this study, the enzyme demonstrated an optimum pH of 8.0 with wide pH stability (5.0-9.0) that makes it suitable for use in the detergent industry. K m values for Coh 01133 enzyme (~0.28) shown in Table C in S1 File. However, k cat and k cat /K m values were greater for starch than pullulan, indicating that starch is a better substrate for the enzyme (Table C in S1 File). The action of the enzyme on pullulan and soluble starch formed G3 and G2, G3, and G4 respectively. Similar results have been observed for the other amylopullulanases [25][26][27]. Cations such as Mn 2+ , Ba 2+ Ca 2+ , and Mg 2+ improved the Coh 01133 amylopullulanase activity; for Ca 2+ and Mg 2+ up to 2-fold activity was noted. The improvement of amylopullulanase activity in the presence of Ca 2+ has been reported elsewhere; for instance, Micrococcus sp. pullulan hydrolyzing enzyme was activated after CaCl 2 addition [28,[32][33][34]. Additionally, Ca 2+ and Mg 2+ improved the temperature stability of the enzyme [35,36]. Some other cations, Li + , Na + , K + , Co 2+ , and Cu 2+ , severely reduced enzyme activity. Similar reductions were reported earlier, for example amylopullulanases activity of T. hydrothermalis did not affect by Na + and Mg 2+ while Mn 2+ increase its activity. Metal cations inhibitory effect such as Ni 2+ and Cu 2+ has been seen for almost all amylopullulanases [6,18,28,31,36,37].

Conclusion
Two amylopullulanse genes from an endogenous bacterium belonging to GH57, namely Coh 00831 (~70 kDa) and Coh 01133, (~127 kDa), were PCR amplified and heterologously expressed in E. coli. Recombinant enzymes were assayed against pullulan and starch and demonstrated greater affinity toward starch. Thus, both enzymes have the potential to be used for liquefaction in the starch industry. The effects of metal divalent cations were studied, and Ca 2+ , Mg 2+ , Mn 2+ , and Ba 2+ were demonstrated to improve enzyme activity by more than 50%.
Supporting information S1 File. Table A in S1 File: Bacterial α-amylases and amylopullulanases used for multiple sequence alignment and phylogenetic tree. Table B in S1 File: The % identity of proteins from this study with other Amylopullulanase. Table C in S1 File: Molecular weight, pH optimum, temperature optimum and Kinetic parameters of some amylopullulanases from GH13 and GH57. Figure A in S1 File: Chromatogram of Ni-NTA sepharose column (c1), chromatogram of ion exchange chromatography (c2) (PPTX)