Introduction

Bioplastics have received tremendous attention in the recent years due to their high molecular weight, thermoplastic/elastomeric properties, biodegradability, biocompatibility, non-toxicity and its production from renewable carbon sources. Polyhydroxyalkanoates (PHAs) are simple macromolecules synthesized by a wide variety of Gram-positive and Gram-negative bacteria, and members of family Halobacteriaceae of the Archaea (Philip et al. 2007; Hezayen et al. 2002). The synthesis of the polymer is initiated when acetyl-CoA is restricted from entering the tricarboxylic acid cycle due to the nutrient limitation, shunting the acetyl units from the TCA cycle into the production of polyhydroxybutyrate (PHB), which is an ideal carbon and energy storage polymer. Currently more than 150 different hydroxyalkanoic acids have been identified as substrates by PHA synthases, thereby determining the type of PHA produced by the organism.

Biosynthesis of PHB involves the condensation of two molecules of acetyl-CoA to form acetoacetyl CoA, which is reduced to (R)-3-hydroxybutyryl-CoA. This serves as a substrate for PHA synthase and acts as a direct precursor for PHA production (Rehm 2007; Valappil et al. 2007 and Volova et al. 2004).

PHA synthases have been classified into three types. Class I PHA synthases, represented by that of Ralstonia eutropha, utilize CoA thioesters of (R)-3-hydroxy fatty acids comprising 3–5 carbon atoms and consist of PhaC subunits with molecular weights between 61 and 73 kDa. Class II PHA synthases, represented by that of Pseudomonas aeruginosa, consist of PhaC subunits with molecular weights also ranging from 61 to 73 kDa and utilize CoA thioesters of (R)-3-hydroxy fatty acids comprising 6–14 carbon atoms. Class III PHA synthases, represented by that of Allochromatium vinosum, consist of two different types of subunits, PhaC and PhaE, which have a molecular weight of 40 kDa each. These prefer CoA thioesters of (R)-3-hydroxy fatty acids comprising 3–5 carbon atoms (Rehm 2007; Sudesh et al. 2000). Bacillus megaterium possesses a novel PHA synthase similar to class III containing two subunits except that the PhaE typical of class III is replaced by a 20 kDa PhaR subunit (Satoh et al. 2002; McCool and Cannon 2001). To date, PHA synthase enzymes from representatives of the genus Bacillus have only been characterized at a preliminary level (Valappil et al. 2007). Even though the presence of PHA as a storage compound was first discovered in Bacillus megaterium, only a few reports on the structural studies of PHA inclusion bodies in Bacillus cereus and properties of the granule associated PHA synthase from Bacillus megaterium are available (Lenz and Marchessault 2005; Griebel et al. 1968).

This report focuses on the isolation of PHA granules from a marine derived Bacillus sp. NQ-11/A2 and characterization of the granule-bound PHA-synthase (GBPS).

Materials and methods

Collection of sediment sample and bacteriological analysis

A sediment sample from Arabian Sea (Lat 19°33.949′ N, Long 71°22.041′ E), 100 km West from the Indian coast was collected aseptically using a Grab sampler at a depth of 63 m and stored at 4°C. One gram of the sediment sample was suspended in 100 ml of sterile physiological saline, in a 250 ml capacity Erlenmeyer flask and shaken for 30 min at 170 rpm (Orbitek environmental shaker). Dilutions of the sediment suspension were prepared and 100 μl was spread plated on Nutrient Agar (NA) and Luria–Bertani agar prepared with aged sea water (LBS). NA contained in g l−1: beef extract- 10; peptone-10; NaCl- 5; agar agar- 20; pH 6.8. LBS contained in g l−1: tryptone- 10; yeast extract- 5; NaCl- 5; agar agar- 20; pH 7.2. The plates were incubated at 30°C for 48 h. Morphologically dissimilar colonies were picked and restreaked to purify the isolates. Colony characteristics were studied and the isolates were maintained at 4°C. All the chemicals used were procured either from Sigma chemicals (USA), HiMedia, Merck or Qualigens (India).

Screening of bacterial isolates for PHA production

The ability of the bacterial isolates to accumulate PHA was tested using E2 mineral medium which contained [g l−1 double distilled H2O]: KH2PO4·3H2O- 3.7; K2HPO4·3H2O- 7.5; NaNH4HPO4·4H2O- 3.5; MgSO4·7H2O- 10 ml (0.1 M); yeast extract- 0.5; microelement stock solution- 1 ml (Lageveen et al. 1988); glucose- 20; pH 7.0. The isolates were spot inoculated on E2 medium agar plates in triplicates and incubated at 30°C. Accumulation of PHA was monitored every 24 h for 3 days by flooding plates with 0.05% (w/v) Nile blue A in ethanol and incubating in the dark for 20 min (Kitamura and Doi 1994). The stain was decanted and plates were exposed to UV light. Bright orange fluorescence was graded and recorded.

Phenotypic characterization of the selected bacterial strain

The isolate NQ-11/A2 was phenotypically characterized using the methods described in Bergey’s Manual of Systematic Bacteriology (Sneath 1986; Priest et al. 1988). The data was analyzed numerically, using the simple matching coefficient (SSM). Clustering was achieved by unweighted pair group average linkage (UPGMA). The computations were performed using the Probiosys program.

Genotypic characterization of the selected bacterial strain

Genomic DNA was extracted using the protocol described in Sambrook et al. (1989). The 16S rRNA gene was amplified using isolated chromosomal DNA as the template. The primers used for the amplification were S-D-Bact-0011-a-S-17 5′-GTTTGATCCTGGCTCAG-3′ as forward and S-*-Univ-1392-b-A-15 5′- ACGGGCGGTGTGTTC-3′ as reverse primer (Alm et al. 1996). All PCR reactions were performed in a reaction mixture of 100 μl total volume using 1 μg template DNA. The reaction mixture was heated at 94°C for 3 min followed by 40 cycles of 94°C for 1 min, 52°C for 1 min and 72°C for 1 min. The final elongation step was carried out at 72°C for 5 min. The PCR product obtained was gel purified and sent for sequencing at Macrogen Inc, Korea. Sequence was obtained using the forward primer and has been deposited in GenBank under Accession number FJ392860. The sequence of the strain was compared with similar sequences of reference organisms by BLASTn 2.2.18+ and EzTaxon (Chun et al. 2007; Zhang et al. 2000). Partial sequences of the 16S rRNA gene (873 bp) including the variable region were analyzed for comparison with those of other Bacillus species according to Goto et al. (2000). ClustalX version 2.0.7 was used to generate multiple alignments between the selected sequences using the IUB matrix (Larkin et al. 2007). A Neighbor-joining (NJ) tree was obtained with 500 seeds and 3,000 bootstraps. The final tree obtained was rooted and drawn using MEGA4 (Tamura et al. 2007).

Further analysis of the 16S rRNA gene sequence (1,419 bp) was also carried out with the help of Chromous Biotech Pvt. Ltd (Bangalore, India). The forward (5′-AGAGTRTGATCMTYGCTWAC-3′) and reverse primers (5′ CGYTAMCTTWTTACGRCT-3′) used are specific for prokaryotic 16S rRNA genes. The sequence obtained has been deposited in GenBank under Accession number FJ392860.

Culture growth and submerged PHA production

A single colony from a 24 h culture was inoculated in a 250 ml Erlenmeyer flask containing 100 ml E2 mineral medium supplemented with glucose (20 g l−1) as sole carbon source. The flask was incubated on an Orbitek environmental shaker (170 rpm) for 24 h at 30°C. This culture broth was either used as inoculum or the cells were harvested at 10,000 g for 10 min, washed twice with sodium phosphate buffer (0.1 M, pH 7.0) and used for further experiments.

Erlenmeyer flasks containing 500 ml E2 mineral medium with glucose (20 g l−1) were inoculated with 5% of the inoculum. 50 ml culture broth was withdrawn every 6 h from the flask and processed for determination of biomass and PHA content.

Determination of carbon chain length of monomers using Nile red staining

The monomer chain length incorporated into the polymer was determined using Nile red O (Wu et al. 2003). Cell suspensions of the colonies growing on E2 medium agar plates containing different carbon substrates were prepared, centrifuged and used for staining with 10 μl of Nile red O solution (1 mg ml−1 in acetone). The acetone was allowed to evaporate and the pellets were suspended in 1 ml of distilled water. The suspensions were scanned for maximal emission wavelength from 530 to 650 nm using an excitation wavelength of 488 nm.

Determination of the monomeric composition of PHA using gas chromatograph- mass spectrometer (GC-MS)

Production of PHA using selected carbon substrates such as xylose, glucose, sucrose, lactic acid, succinic acid and citric acid was achieved by growing the isolate NQ-11/A2 under submerged conditions. These substrates were added as sole carbon source in E2 mineral medium at a final concentration equivalent to 20 g l−1 glucose. Inoculated flasks were incubated at 30°C (170 rpm). The polymer accumulated using the above carbon substrates was used for further analysis.

To determine the composition of PHA, the polymer was methanolyzed as per the method described by Shishatskaya and Volova (2004). Approximately 8 mg of PHA was suspended in 2 ml methanol containing 15% (v/v) concentrated sulphuric acid and 2 ml of chloroform. The reaction was carried out at 100°C for 3.5 h. After cooling, 1 ml of distilled water was added to the reaction mixture and shaken vigorously for 1 min. On phase separation, the organic layer was transferred into a vial and used for analysis.

The methanolyzed polymer sample was analyzed using GC-MS (model QP2010, Shimadzu, Japan), equipped with a 30 m × 0.32 mm RTX-5MS (5% diphenyl and 95% dimethyl polysiloxane) capillary column. Chromatographic conditions were as follows: carrier gas helium; flow rate 1 ml/min; sample input temperature 220°C; initial temperature 70°C, programmed to 230°C at a rate of 8°C/min; interphase temperature 250°C; ion source temperature 175°C; electron impact mode 70 eV; scanning range 45–450 amu at 0.5 s/scan.

Isolation of native PHA granules

Bacterial cells (1 g wet weight) were suspended in 5 ml lysis buffer (0.1 M sodium phosphate buffer, pH 7.0 containing 1 mg/ml lysozyme) and incubated for 1 h under shaking conditions at 30°C. The lysate was centrifuged for 30 min at 10,000g. The pellet was suspended in sodium phosphate buffer (0.1 M, pH 7.0), centrifuged and re-suspended in minimum volume of buffer. 0.5 ml of the suspension was loaded on to sucrose density gradient consisting of 2.5 ml each of 40, 55 and 60% sucrose in 11.2 ml polyallomer Quick-seal centrifuge tubes. The sealed tubes were placed in a vertical rotor and centrifuged at 79,000g for 2 h at 4°C using an Ultracentrifuge (Beckman, L8-M, USA). The inclusion bodies, which banded at the interphase between 40 and 55%, were collected and suspended in 10 volumes of buffer. The suspension was centrifuged at 10,000g for 10 min, the pellet was re-suspended in sodium phosphate buffer and stored at 4°C.

Scanning electron microscopy of PHA granules

A drop of the granule suspension was placed on a slide, air dried and fixed with 2% (v/v) glutaraldehyde for 2 h. The slide was washed with sodium carbonate-bicarbonate buffer (0.05 M) and passed through a series of increasing acetone concentrations [50, 70, 90 and 100% (v/v)]. The slide was placed on the stub and dried using a critical point dryer. The sample was coated with a thin film of gold using a spi-module sputter coating device and observed under scanning electron microscope (JEOL, 5800LV, Japan).

Determination of PHA synthase activity

The PHA synthase activity was determined by a discontinuous spectrophotometric assay monitoring the release of CoA at 412 nm from the substrate DL-β-hydroxybutyryl-CoA (Muh et al. 1999). The final volume of the reaction mixture was made up to 1 ml and contained 50 mM NaCl, 1 mM 5, 5′-dithiobis(2-nitrobenzoic acid) (DTNB), 60 μM DL-β-hydroxybutyryl-CoA, sodium phosphate buffer (0.1 M, pH 7.0) and 100 μl of granule suspension. The reaction was carried out at 30°C, unless otherwise specified, for 8 min. The mixture was spun at 10,000 g for 2 min. The effect of varying pH was tested using citrate buffer (0.1 M, pH 4 and 5), sodium phosphate buffer (0.1 M, pH 5–8) and Bicine buffer (0.1 M, pH 8 and 9). The absorbance of the supernatant was measured at 412 nm. The concentration of CoA released was quantified using the extinction coefficient (412 nm) of 13,600 cm−1 M−1. One unit was defined as the amount of enzyme required to convert 1 μmol of substrate in 1 min.

Analytical methods

The polymer was extracted using the sodium hypochlorite method (Rawte and Mavinkurve 2002). Twenty-five ml of the culture broth was centrifuged at 10,000g for 10 min. The cell pellet was suspended in physiological saline and centrifuged again. Sodium hypochlorite solution (2% available chlorine) was added to the cell pellet and incubated on an Orbitek environmental shaker (170 rpm) for 20 min. The suspension was centrifuged for 20 min. The polymer pellet obtained was suspended and washed twice with diethyl ether and dried at 80°C.

Cell biomass was estimated by drying the cell pellet at 80°C until constant dry weight was obtained.

Protein content of the samples was determined according to the method described by Lowry et al. (1951) using bovine serum albumin as a protein standard.

Results and discussion

Isolation and identification of the isolate

Heterotrophic bacterial counts in sediment collected from Arabian Sea were 89.2 × 104 and 14.4 × 105 cfu g−1 on NA and LBS, respectively. The higher counts on the medium containing sea water indicates the predominance of marine bacteria in the sediment sample. Seven and five colonially distinct bacterial isolates on NA and LBS were obtained, respectively. Among these, four isolates from NA showed accumulation of PHA. However, none of the isolates obtained on LBS exhibited PHA accumulation. PHA serves as a carbon and energy reserve. The accumulation of PHA offers a selective advantage for survival of bacteria in such sediments (Lopez et al. 1995). One of the isolates namely, NQ-11/A2 accumulated maximum PHA on glucose, detected as bright orange fluorescence when stained with Nile blue A. This isolate was Gram-positive, rod-shaped, facultatively anaerobic and an endospore former. It was able to grow at temperatures as high as 50°C and tolerate up to 10% salt. Biochemical characterization with similarity analysis and UPGMA clustering placed this isolate along with Bacillus megaterium and Bacillus flexus. SSM for NQ-11/A2 were 86.0 and 83.33% with B. megaterium and B. flexus, respectively. However, the phenotypic characteristics of this isolate were in sharp contrast to B. megaterium, a strict aerobe, which grows only up to 37°C and in presence of a maximum of 5% salt (Priest et al. 1988). A BLAST (NCBI) search using the 16S rRNA gene sequence (873 bp) of strain NQ-11/A2 showed 90% and above homology to 30 known taxa of Bacillaceae with maximum homology to B. megaterium and B. flexus. The evolutionary relationship of the isolate with selected species of the Bacillaceae family was analyzed using the Neighbor-joining method. From the phylogenetic tree it was evident that isolate NQ-11/A2 forms a separate group with the closest relatives being B. megaterium, B. flexus and Bacillus simplex. Within this group, it was seen that the sequence of the isolate aligned to B. megaterium.

The nucleotide sequence comprising of 1,419 bp of the 16S rRNA gene was also further analyzed and revealed a S_ab (similarity score) of 1.0 when compared with that of Bacillus megaterium DPBS17 (NCBI accession number EU249559) indicating maximum sequence homology between the two gene sequences. The phylogenetic tree is displayed in Fig. 1. These data suggest that strain NQ-11/A2 is a representative of B. megaterium. Bacillus sp. NQ-11/A2 has been deposited in the National Collection of Industrial Micro-organisms (NCIM, Pune, India) with Accession Number NCIM 5334.

Fig. 1
figure 1

Phylogenetic relationship of isolate NQ-11/A2 with closely related members of Bacillus based on 16S rRNA gene sequence analysis. The tree was generated using Weighbor weighted neighbor-joining method (Bruno et al. 2000). Bootstrap values with 100 replicates are displayed on the nodes

Full size image

Bacillus sp. NQ-11/A2 grew rapidly and reached stationary phase within 24 h. PHA accumulation was initiated at 6 h and continued to be synthesized during stationary phase up to 48 h. The strain accumulated a maximum 61% of its total biomass as PHA (Fig. 2). Usually, PHA accounts for 11–37% of cellular mass in Bacillus spp. (Shamala et al. 2003). Recently, Bacillus thuringiensis R1 and an anoxic Bacillus sp. CL1 have been reported with high PHA content (Bordoloi et al. 2007; Rohini et al. 2006; Full et al. 2006). Bacillus sp. NQ-11/A2 accumulated PHA using diverse carbon substrates (Table 1). Except for a few carbon sources, all served as substrates for PHA production. Nile red O fluorescence spectra of the intracellular PHA indicated that the polymer is of the short chain length (SCL) type.

Fig. 2
figure 2

Accumulation of biomass (plus) and PHA (filled diamond) in shake flask of Bacillus sp. NQ-11/A2

Full size image
Table 1 PHA accumulation in Bacillus sp. NQ-11/A2 using diverse carbon substrates
Full size table

The polymer accumulated by isolate NQ-11/A2 using various carbon substrates such as xylose, glucose, sucrose, lactic acid, succinic acid and citric acid was analyzed by GC-MS. The total ion current chromatogram (TIC) of the methanolyzed products of the polymer accumulated by isolate NQ-11/A2 when grown using xylose is shown in Fig. 3a and the mass spectra of the dimers of beta-hydroxybutyric acid methyl ester is shown in Fig. 3b. GC-MS analysis confirmed that the same profile was obtained when isolate NQ-11/A2 was grown on succinic acid and lactic acid as sole carbon sources (Online Supplementary Figs. 1 and 2). In all the cases, the peak with retention time 7.9 min matched with the mass spectrum of the dimer of beta-hydroxybutyric acid methyl ester from the MS library (WILEY7). This confirmed that beta-hydroxybutyric acid was the major momomeric unit incorporated into the PHA polymer by the isolate grown using the above carbon substrates.

Fig. 3
figure 3

a The total ion current chromatogram (TIC) of the methanolyzed products of the polymer accumulated by isolate NQ-11/A2 grown on xylose and b mass spectrum of the dimer of beta-hydroxybutyric acid methyl ester from isolate NQ-11/A2

Full size image

The genus Bacillus has been demonstrated to accumulate SCL-PHA (Tajima et al. 2003). The type of polymer synthesised by the organism is dependent on the carbon substrate. The use of some carbon substrates as precursors for production of new types of PHA is mainly limited by their cost and availability (Rehm 2003). Alternatively, screening for bacteria metabolizing cheap carbon sources and synthesizing these precursor substrates endogenously can help to overcome the problem (Steinbuchel and Valentin 1995).

Native PHA granules of Bacillus sp. NQ-11/A2 formed a white band at the interphase between 40 and 55% sucrose concentration on sucrose density centrifugation. PHA granules were spherical and intact in shape (Fig. 4). These granules were 0.51 ± 0.074 μm. There were three to nine granules in the cells. These granules were either single or in budding groups. McCool et al. (1996) have reported the presence of inclusion bodies in numbers of 2–8 which also appeared dividing or budding.

Fig. 4
figure 4

Scanning electron micrograph of native PHA granules

Full size image

The specific activity of the GBPS was 11.625 U mg−1 whereas specific cytosolic PHA synthase was only 0.056 U mg−1. This granule associated specific activity was higher than that reported for Pseudomonas oleovorans but lower than reported in B. megaterium and halophillic archaeon (De Roo et al. 2000; Griebel et al. 1968; Hezayen et al. 2002). The GBPS of this isolate showed a requirement for sodium ions in the reaction mixture (Table 2). Interestingly, potassium ions had a negative effect on the enzyme. This could be due to the marine origin of Bacillus sp. NQ-11/A2. Sodium was also found to be necessary for the PHA synthase activity in a halophillic archaeon (Hezayen et al. 2002). The GBPS was active over a narrow temperature and pH range with optima at 30°C and 7.0, respectively (Fig. 5a, b). Either side of the optima for temperature and pH, the GBPS activity dropped drastically, unlike the GBPS activity reported for the haolophillic archaeon (Hezayen et al. 2002). The substrate saturation constant of the GBPS was 7.1 × 10−5 M for 3-hydroxybutyryl CoA (Fig. 6) and therefore twenty percent better than reported previously for B. megaterium (Griebel et al. 1968).

Table 2 Effect of sodium and potassium ions on granule bound PHA synthase
Full size table
Fig. 5
figure 5

a Effect of temperature on the activity of granule bound PHA synthase. b Effect of pH on the activity of granule bound PHA synthase

Full size image
Fig. 6
figure 6

Effect of substrate concentration (HB-CoA) on the activity of the granule bound PHA synthase. 100 μl of granules (filled square), 50 μl of granules (empty diamond)

Full size image

Detergents adversely affected GBPS of NQ-11/A2 (Fig. 7). This effect was maximal with CTAB buffer and minimal with Triton-X-100. Ren et al. (2000) have reported complete inhibition of GBPS activity with Triton-X-100. These detergents were expected to stabilize the amphiphilic enzymes. However, GBPS was inhibited. The loss in activity could be due to the interference of the detergent with protein–protein interactions of GBPS and other proteins in the granule membrane. GBPS was not affected by MgCl2, similar to the observations made with GBPS from an archeal halophile (Hezayen et al. 2002). This was further confirmed as EDTA did not have any effect on GBPS. Only BSA and glycerol showed positive effects on the enzyme. This could be due to the stabilizing effect of BSA and glycerol as they are known to bind to the PHA granules and interact directly with PHA synthases (De Roo et al. 2000). Lysozyme, a known inhibitor of GBPS (Liebergesell et al. 1992), reduced the activity by 40%. PHA synthase in the granules of Bacillus sp. NQ-11/A2 was apparently covalently linked to the PHA-core and therefore addition of the chaotropic agent urea reduced the activity.

Fig. 7
figure 7

Effect of various additives on granule-bound PHA-synthase activity

Full size image

To our knowledge, this is the first report on marine derived Bacillus sp. growing rapidly and accumulating large amounts of SCL-PHA. Our studies indicate a large amount of polymer accumulation inside the granule by GBPS in vitro with minimum maintenance requirement.