Identification a Novel Raw-Starch-Degrading-α-Amylase from a Tropical Marine Bacterium

Problem statement: Bacteria from the surface of the tropical marine hard coral Acropora sp. were screened for producing raw-starch-degrading-α-amylase. Approach: Based on its 16s rDNA sequence, a bacterium that produced the highest amylolitic activity was identified as Bacillus amyloliquifaciens ABBD. The bacterial isolate secreted a α-amylase extracellularly and then the enzyme was partially purified by ammonium sulfate precipitation followed by anion exchange chromatography. Results: Electrophoresis results both SDS-PAGE and native PAGE suggested that the enzyme was a heterodimeric protein (97 kDa) consisting of 45 and 55 kDa subunits. The α-amylase had an optimum pH of 7.0 and temperature of 60°C. More than 80% activity of the enzyme was retained under high salt conditions (up to 20% NaCl). The enzyme remained stable at 50°C for 1 h. Starch hydrolysis by the enzyme at 70°C yielded oligosaccharides (G2-G4) and at room temperature yielded glucose/maltose (G1 and G2). Conclusion: The B. amyloliquifaciens ABBD α-amylase was capable of degrading various raw starch granules from corn, rice, cassava and sago at room temperature.

According to Henrissat (1991), α-amylases belong to the Glycosyl Hydrolase (GH) family 13 (http://afmb.cnrs-mrs.fr/CAZY/fam/GH13.html). The structural features of the family include the core domain (domain A) consisting of a (β/α) 8 barrel, the domain B protruding from domains A third β-strand and the Cterminal domain (C) with a Greek key motif. Owing to the incapability to penetrate crystalline structures, only few amylolytic enzymes are capable of attacking raw starch granules. The latter enzymes possess the starch binding domain predominantly at their C-termini (for review see (Machovic and Janecek, 2006). α-Amylases degrading raw starch granules have been reported to occur in a wide range of organisms (Demirkan et al., 2005;Matsubara et al., 2004;Iefuji et al., 1996;Bandani, 2009a,2009b), however, few reports on α-amylases from marine sources are known (Vidilaseris et al., 2009). Here we report a hard coral associated bacterium producing rawstarch-degrading-α-amylase, Bacillus amyloliquifaciens ABBD. The enzyme properties such as quaternary structure, biochemical features and its action on raw starch granules as well as its hydrolysis pattern are described.

MATERIALS AND METHODS
Bacterial isolate and culture condition: Of three isolates obtained from the surface of the hard coral Acropora sp. collected from Bandengan Water, Jepara, North Java, Indonesia by scuba diving at depth of approximately 5 m, the bacterial strain ABBD displaying highest starch degrading activity was chosen for this study (Fig. 1). The isolate was routinely maintained on Marine Broth (MB) medium containing 0.25% bacto peptone, 0.05% yeast extract and 0.1% soluble starch (Merck) in sea water. The culture was grown in a shaker incubator (New Brunswick Scientific model C25KC) at 37°C and 150 rpm for 18 h.
Identification of 16S rDNA: Bacterium grown to the stationary phase was harvested by centrifugation at 5000×g for 10 min at 4°C. Genomic DNA was isolated and purified using the Wizard Genomic Purification Kit (Promega). Degenerated forward-and reverse primers for amplification of the 16S DNA were 5'-AGAGTTTGATC (A/C) TGGCTCAG-3' (BactF1) and 5'-GGTTAC(G/C)TTTGTTACGACTT-3' (UniB1), respectively. PCR amplification was performed in a final concentration of 25 µL, containing 5 ng chromosomal DNA, 0.8 µm of each primer, 0.4 mm dNTP, 0.025 U Taq DNA polymerase, 50 mm KCl, 1.5 mm MgCl 2 and 10 mm Tris-HCl, pH 8.3. The following profile was applied for the PCR: DNAdenaturation at 94°C for 1 min, primer annealing at 50°C for 1 min and elongation at 72°C for 2 min. The initial and final cycle included denaturation at 94°C for 4 min and elongation at 72°C for 7 min to ensure full extension of the product. Amplified PCR products were analyzed in a 1% agarose gel and nucleotide sequences determined by the method of dye-end terminator using BactF1 and UniB1 as forward and reverse primer, respectively, performed by Macrogen Inc., (Seoul, Korea). The 16S rDNA sequence of the isolate was aligned with those in the GenBank database. Multiple alignments of sequences and calculation of the level of sequence similarity were carried out using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The neighborjoining phylogenetic analysis was performed by applying the MegAlign DNAstar program.

Enzyme preparation:
To the supernatant obtained from bacterial cultures, ammonium sulfate was gently added to reach 80% saturation, followed by centrifugation at 22.000×g for 30 min at 4°C. The precipitate was dissolved in a minimum volume of 25 mm Glycine-NaOH buffer, pH 9.0 and dialyzed against the same buffer. The supernatant was loaded onto a Resource-Q column (Pharmacia) equilibrated with 25 mm Glycine-NaOH buffer, pH 9.0. After washing and The strain cultured in MB agar containing 2% bacto agar at 37°C overnight, was stained by iodine solution containing 2% KI and 0.2% I 2 .
A clear zone on a blue background showed amylolytic activities removing unbound proteins, the column was eluted by applying a linear gradient from 0-0.5% NaCl in the same buffer at a flow rate of 6 mL min −1 . Active fractions were pooled and concentrated using a filter membrane (GE Healthcare) (30 kDa molecular weight cut off) and used for the characterization.
Enzyme activity assay: Amylolytic activities were measured by the method of Fuwa (1954): 50 µL of soluble starch in 25 mm Glycine-NaOH buffer, pH 9.0 was mixed with 50 µL of enzyme solution. After incubation at 60°C for 10 min, the enzymatic reaction was terminated by the addition of 50 µL 1 M HCl. To the reaction mixture, 50 µL iodine solution containing 2% KI and 0.2% I 2 and 800 µL 25 mm Glycine-NaOH buffer, pH 9.0 was added. Absorption was measured at 600 nm and the activity was calculated using soluble starch as the standard. Hydrolysis of 1 mg mL −1 soluble starch in 1 min was defined as 1 U mL −1 of enzyme activity. Specific activity is given as U mg −1 protein.
Protein determination: Protein concentration was determined by the method of Bradford (1976) using bovine serum albumin as a standard.

Electrophoresis:
Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) under non reducing conditions was carried out according to Laemmli (1970) using 10% acrylamide gels. Electrophoresis was run at 150 V for 60 min at room temperature. After electrophoresis, a part of the gel was silver stained. To identify amylase band (clear zone on a blue background), the other part of gel was incubated for 45 min at 70°C in 1% soluble starch containing 25 mm Glycine-NaOH buffer, pH 9.0 and then stained with iodine solution.
Non-denaturing gel electrophoresis (native-PAGE) was performed similar to the SDS-PAGE procedure but lacking sodium dodecyl sulfate. Zymogram staining was performed by incubating the gels for 45 min at 70°C in 1% soluble starch in 25 mm Glycine-NaOH buffer, pH 9.0 and then stained with iodine solution.
Effect of pH on enzyme activity: The optimal pH of the α-amylase was determined by standard assays at different pH values (4.0-10.0). The pH solution was adjusted using a JBScreen pH-2D buffer system (Jena Bioscience). The buffer contained succinic acid, sodium dihydrogen phosphate, glycine (2:7:7) and was adjusted with 1 m NaOH to reach the appropriate pH.
Effect of salinity on enzyme activity: The effect of NaCl on α-amylase activity was determined by the enzyme at different NaCl concentrations (0-20%) under standard assay conditions at pH 7.0 Effect of temperature on enzyme activity and stability: To determine the optimum temperature for the α-amylase activity, assays were performed at 30, 40, 50, 60, 70 and 80°C. For checking thermal stability, enzymes were heated for 15-60 min at 40, 50, 60, 70 and 80°C. Subsequently, samples were immediately cooled on ice and the residual activities were determined.

Determination of degradation products: Thin Layer Chromatography
(TLC) was employed for identification of oligosaccharides in starch hydrolysates (Hansen, 1975). Briefly, each sample containing starch hydrolysates was spotted on a silica gel plate and then developed in a chromatographic jar in a solvent system of butanol, ethanol and water (5:5:3) until the solvent reached the top of the plate. After drying at room temperature to get rid of the solvent, plates were sprayed with 50% concentrated H 2 SO 4 in methanol and subsequently dried in an oven at 110°C for 10 min to develop the black color of the spots. Sugars (G1-G7) were used as markers.
Degradation of raw starch granules: Amylolytic activity on raw starch granules from rice, corn, cassava and sago was checked by mixing 10% of the respective raw starch suspension in a volume ratio of 1:1 with the enzyme solution. Reaction mixtures were incubated at 37°C for 4 days, then centrifuged and the pellet was dried in an oven at 50°C. The pellets were photographed using a scanning electron microscopy (SEM Jeol JSM6063lA0).

RESULTS AND DISCUSSION
The bacterium characteristics: During the microbial screening process -of the three culturable isolates obtained from the surface of the hard coral Acropora sp.strain ABBD was found to exhibit the highest amylolytic activities (Fig. 1). Cells of strain ABBD positively reacted in the Gram-staining procedure and had a rodlike shape. Sequences of two PCR products obtained by applying primers BactF1 and UniB1 and chromosomal DNA of the strain ABBD as a template yielded the almost complete nucleotide sequence of the 16S rDNA of 1414 bases. The sequence has been deposited in the GenBank database under accession no FJ654647.
By performing a BLAST search within the GenBank database, the most similar sequences were identified. Subsequently, based on selected sequences of close neighbors, a phylogenetic tree was constructed (Fig. 2). At the close of the analysis B. amyloliquefaciens strains GH58 and GH28 (Meerak et al., 2008) were identified as being almost identical, i.e. with 99% identity. Thus, with respect to the phylogenetic tree of the 16S rDNA obtained by applying the clustal method (Higgins and Sharp, 1989), strain ABBD is considered to be a representative of the species B. amyloliquefaciens.   Lane 1, 2 and 3 represent SDS-PAGE of protein marker, crude enzyme and partially purified enzyme, respectively; whereas lane 4 indicates zymogram of partially purified enzyme Quaternary structure of the α-amylase: Since the B. amyloliquefaciens α-amylases are routinely extracellular enzymes, it was possible to partially purify the enzyme from the supernatant of the isolate; we applied a strong anion exchange column (Resource-Q column, Pharmacia) for such purpose. To determine quaternary structure of α-amylase, SDS-PAGE and Native PAGE were employed, together with their zymogram to observe the α-amylase bands.
SDS-PAGE result under non reducing conditions showed three protein bands with estimated molecular masses of 97, 55 and 45 kDa, respectively. All of the protein bands displaying amylolytic activities were detected from their zymogram (Fig. 3).
However, the zymogram of native-PAGE result identified only a single amylolytic protein with a molecular mass of 97 kDa (Fig. 4). Our findings suggested that the quaternary structure of the α-amylase of the B. amyloliquefaciens strain ABBD (hereafter called BAA) was a heterodimeric complex consisting of two different subunits: 55 and 45 kDa. Apparently, BAA molecules under denaturing conditions such as in SDS-PAGE were partially separated into its subunits. Most interestingly, after incubation in the non denaturating buffer, BAA refolded and showed amylolytic activities for both, in each of the subunits and in the complex. In terms of heterodimeric complex formation of a quaternary structure, the BAA was a novel α-amylase that differed from all BAA so far known, as most were monomeric (Table 1) The α-amylase properties: The BAA exhibited an optimum pH of 7.0 (Fig. 5A). The enzyme was relatively stable in acidic conditions with at least 40% of the activity retained at pH 4.0-5.0; and more than 50% of the activity was observed in the range of pH 6.0-9.0. Such characteristics may be important for industrial processes in which a broad pH-range is needed. It was noteworthy to state that the total loss of activity occurred at pH 10.0. The effect of NaCl on the BAA is depicted in Fig. 5B. The enzyme retained more than 80% of its activity in NaCl up to 20%. As the bacterium was isolated from the surface of a hard coral from the sea, the halophilic properties of the enzyme matched our expectations. The enzyme activity that could adapt to a wide range of salt concentrations might be important for industrial application.
Optimal temperature of the BAA occurred at 60°C (Fig. 5C). At least 50% of the activity was detectable at temperatures between 40 and 70°C (the enzyme was kept under such temperatures for 10 min). However, more than 80% loss of activity was observed at temperatures above 80°C. Based on temperature characteristic, the BAA can be considered as a mesophilic enzyme.
Thermal stability of the BAA at pH 7.0 is given in Fig. 5D. The enzyme retained more than 60% activity at 50°C for 1 h. Consistent with its mesophilic character, only half of the BAA activity was obtained when the enzyme was incubated at 60°C for 1 h and its activity completely lost at 80°C for 30 min.  performed. An early reaction product of starch digestion at 70°C yielded maltose (G2) to maltoheptaose (G7), whereas small amounts of glucose (G1) were observed (Fig. 6A). The end products of starch digestion at 70°C were predominantly G2, G3 and G4, with minor amount of G1. Interestingly, starch hydrolysis by the enzyme at room temperature yielded predominantly G1 and G2 (Fig. 6B). With respect to the major products of hydrolysis, the BAA action is useful for starch liquefying, short-chain-oligosaccharide forming, or starch saccharifying. Degradation patterns towards various raw starch granules by BAA at room temperature were observed by the SEM (Fig. 7). BAA degraded corn-and sago granules as digging means (from the surface to the center) producing holes on the granules. The enzyme hydrolyzed rice-and cassava granules as peeling ways (scattered over the surface). Adsorption of enzymes on starch granules remains unclear; however, some studies of amylases showed the important role of the C-terminal domain of α-amylase for starch binding (Machovic and Janecek, 2006). For industrial use, the capability of BAA to degrade raw starch granules at lower temperatures may give advantages owing to it will reduce the cost of energy. The relative activity of the BAA on raw starch granules from corn, rice, cassava and sago was 100, 34, 30 and 18%, respectively.

CONCLUSION
The present study has yielded a raw starch degrading α-amylase produced by a hard coral associated bacterium, B. amyloliquifaciens strain ABBD. In terms of heterodimeric complex formation, the BAA from strain ABBD is a novel α-amylase differing from the published BAA. The α-amylase of the B. amyloliquifaciens strain ABBD showing activities in a broad pH-range, in high salt concentrations, at lower temperatures as well as on raw starch granules may be a promising candidate for starch processing industries. To understand the mechanism of the enzyme activity, further research will focus on cloning and overexpressing gene(s) encoding BAA, as well as on structural studies of the enzyme.