Introduction

Poly-3-hydroxybutyrate (PHB), the most common representative of the polyhydroxyalkanoates (PHA), is widespread in different taxonomic groups of prokaryotes as an intracellular storage compound, acting as a carbon reserve and reducing equivalent that facilitates cell survival during stressful conditions [1, 2]. In general, PHB is synthesized by cells under growth-limiting conditions, when the carbon source is in excess and nitrogen, phosphorus, magnesium, oxygen or sulfur is present in a limiting concentration.

PHA are attracting considerable attention because of their potential as renewable and biodegradable plastics in the areas of tissue engineering, environmental friendly packaging materials, and as a chiral hydroxyalkanoate (HA) pool [3]. Poly-3-hydroxybutyrate (PHB), a representative compound of the family of PHA, has many potential applications in medicine, veterinary practice, and agriculture due to its biodegradability and biocompatibility [4].

Gram-positive bacteria, notably Bacillus and Streptomyces, have been used extensively in industry. However, these organisms have not yet been exploited for the production of PHA biodegradable polymers. Gram-negative bacteria, currently the only commercial source of PHA, have lipopolysaccharides (LPS) which copurify with the PHA and cause immunogenic reactions. On the other hand, Gram-positive bacteria lack LPS, a positive feature which justifies intensive investigation into their PHA production [5]. The genus Bacillus seems to be a potential candidate for production of PHB due to its better polymer yields and less stringent fermentation conditions. The novel PHA synthase discovered from Bacillus genus has the ability to incorporate both short chain length (scl) and medium chain length (mcl) PHA, indicating that the genus can be a potential producer of novel and known PHA with different ranges of monomeric compositions [6]. Hence, the accumulation of PHB by the Bacillus genera has distinct features which need extensive investigation.

At present, about 40 PHA synthase genes are known from various microorganisms [7]. The acquisition of a novel PHA synthase gene can facilitate the synthesis of a new type of PHA.

One of the most important factors for the popularization of the use of PHB polymers as conventional plastics is their production cost [8]. Therefore, less expensive substrates, improved cultivation strategies, and easier downstream processing methods are required for cost reduction [9]. It is also essential to identify microorganisms that utilize cheap carbon sources efficiently to produce PHB [10], as the carbon substrate alone accounts for 50% of the total expense of PHB production.

In the present study, production and characterization of PHB from two novel strains of Bacillus spp., which were isolated from soil and activated sludge samples in an earlier investigation, are reported.

Materials and methods

Microorganisms

Bacillus sp. 87I and Bacillus sp. 112A isolated from soil and an activated sludge sample, respectively, collected at Hyderabad, India, have been described elsewhere [11]. These bacterial strains were isolated on Luria–Bertani (LB) medium (Hi media, India).

Strain identification

Both of the selected bacilli were identified by biochemical as well as 16S rRNA sequence analysis and phylogenetic studies. Universal primers 8F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492R (5′-ACGGCTACCTTGTTACGACTT-3′) [12] were used for amplification of 16S rRNA of both isolates. For quantitative analysis of the cellular fatty-acid compositions of the two bacilli, a loop of cell mass from each bacilli isolate was harvested, and fatty-acid methyl esters were prepared and identified following the instructions of the Microbial Identification system (MIDI).

Media and growth conditions

Isolates were grown aerobically (150 rpm) at 30°C in a chemically defined nitrogen-deficient E2 medium [13]. Sugars and mineral salts solutions were autoclaved separately at 110°C for 10 min.

Conditions of PHB production

For the production of PHB, the Bacillus isolates were cultivated in E2 mineral broth with 2% (w/v) glucose as carbon source. Each of these Bacillus isolates was grown in 250-ml Erlenmeyer flask containing 50 ml E2 mineral broth. After inoculation with overnight-grown 4% (v/v) inocula, the flasks were incubated at 30°C for 48 h on an orbital shaker at 150 rpm.

Extraction and quantitative assay of PHB

PHB was extracted from the Bacillus isolates by using the hypochlorite method [14]. Quantitative estimation of PHB extracted from the isolates was done by ultraviolet (UV) spectrophotometer method [15].

Growth pattern

The growth response of Bacillus spp. was studied using 500-ml Erlenmeyer flask containing 100 ml E2 broth with 2% 24-h-aged culture inoculum. Initial cell density was adjusted to A 600 of 0.1, and incubated for 48 h at 30°C and 150 rpm. After inoculation, cell density was recorded for an interval from 4 h up to 48 h.

Effect of time on PHB production

Inoculated E2 broth was incubated for up to 48 h to determine the optimum incubation period. Ten milliliters of culture was drawn at 2-h intervals and centrifuged at 8,000 × g for 15 min to obtain the cell pellet. Quantitative assay was performed using the assay suggested by Aslim et al. [14].

Effect of temperature and pH on PHB production

PHB was accumulated by the two isolates, Bacillus sp. 87I and Bacillus sp. 112A, at various temperatures (25°C, 30°C, 35°C, 40°C, and 45°C) and at different pH values (6.8, 7.2, and 7.5) using E2 media and other standard conditions as described earlier.

PHB production from various carbon substrates

The capacity of these two isolates, Bacillus sp. 87I and Bacillus sp. 112A, to utilize various carbohydrates (lactose, galactose, fructose, glucose, mannitol, sucrose, maltose, and starch) was determined in the biochemical tests. For this, Bacillus sp. 87I and Bacillus sp. 112A were incubated separately in nitrogen-free E2 medium which had the above different carbon sources substituted for glucose. PHB content in the cells was extracted according to the method described earlier and quantified by UV spectrophotometer.

Polymer analysis

The fermentation broth was used for determination of dry cell weight (DCW). The cell concentration was determined by measuring DCW: 5 ml culture broth was centrifuged, washed, and dried at 105°C until the weight did not decrease further. The residual mass was defined as total DCW–PHB weight; PHB (%) was defined as the percentage of the ratio of PHB to DCW. The polyesters content of the cell and the composition of polyesters were determined by using gas chromatography-mass spectrometry (GC–MS) analysis. The structure and mole fractions of HB units in the polymer samples were investigated by 400-MHz H1 nuclear magnetic resonance (NMR) spectra recorded at 27°C in CDCl3 solution of polyester. The average molecular mass of the polymers were estimated by the inherent viscosity method. Infrared (IR) spectra and differential scanning calorimeter (DSC) scans were also recorded for the polymers [16].

Results

Isolation and identification of strains

In an initial study [11], spore-forming bacteria were selectively isolated by heat or ethanol treatment from various environmental samples. A total of 240 colonies, representing many macroscopic differences in colony morphologies, were observed and picked up randomly from plates containing different samples for further analysis. Based on UV spectrophotometer quantifications, 25 Bacillus isolates with maximum PHB accumulation were selected from the total 240 Bacillus isolates as optimal PHB-accumulating Bacillus spp.

Based on their high PHB productivity, the two best strains were selected for further studies. Morphologically, Bacillus sp. 87I (deposited as MTCC 9593) appeared as off-white, circular, smooth, large colonies and Bacillus sp. 112A (deposited as MTCC 9719) formed white, round-shaped, large colonies in nutrient agar. Microscopically, both isolates appeared as rod-shaped Gram-positive bacilli.

Genetic, biochemical, and fatty-acid analysis

16S rRNA sequencing followed by basic local alignment search tool (BLAST) analysis showed that the isolates had close 16S rRNA database similarity with known Bacillus spp. The 16S rRNA analysis revealed that Bacillus sp. 87I (1,428 bp; deposited in the EMBL database under accession number AM903377) had 95% nucleotide base homology with B. mycoides, and Bacillus sp. 112A (1,512 bp; deposited in the EMBL database under accession number AM900681) had only 94% homology with B. flexus.

A phylogenetic tree (Fig. 1) demonstrated that the isolated strains were members of the genus Bacillus and formed a monophyletic lineage. Sequence-similarity calculations after neighbor-joining analysis indicated that the closest relatives of strain 87I were B. mycoides (98.1%), B. thurengenesis (98.4%), and B. weihenstephanensis (98.1%) and those of strain 112A were B. flexus (95.4%), B. megaterium (94.8%), B. simplex (94.8%), and B. nealsonii (94.4%). Low 16S rDNA gene sequence similarity values (<95%) were obtained between the novel strain 112A and all species with valid published names from the genus Bacillus. From the phylogenetic analysis, it was clear that, based on good 16S rDNA (1,512 bp) sequences, the isolate belonged to the genus Bacillus and represented a distinct lineage that could be equated with a separate genomic species [17].

Fig. 1
figure 1

Unrooted phylogram obtained by neighbor-joining of the 16S rDNA sequence from the Bacillus spp. type strains obtained from the GenBank with Bacillus sp. 87I and Bacillus sp. 112A

Full size image

Results of biochemical analysis based on mannitol utilization and acid formation [18, 19] were also in agreement with 16S rRNA analysis that Bacillus sp. 87I and Bacillus sp. 112A were close to B. mycoides and B. flexus, respectively. Apart from this, these were Gram-positive, endospore-forming, and catalase-positive organisms (results not shown).

The fatty-acid profile of Bacillus sp. 87I was characterized by predominance of C15:0 iso and C16:0, accounting for 23.12% and 19.67% of total acids, respectively, along with C13:0 iso (9.89%). Other fatty acids were present at lower concentrations, i.e., C15:0 anteiso (5.98%), C14:0 (4.21%), C13:0 anteiso (2.0%), C12:0 (1.44%), C12:0 iso (0.86%), C13:1 (0.58%), C16:1 w9c (0.55%), and C 10:0 (0.52%). The fatty-acid profile of Bacillus sp. 112A showed predominance of C 15:0 iso and C 15:0 anteiso, accounting for 32.57% and 29.91% of total acids, respectively, along with C16:0 (10.02%). Other fatty acids were present at lower concentrations.

Since the Bacillus sp. 87I fatty-acid profile showed variation from its close neighbor (Table 1), it could be a potent new strain or species of the genera Bacillus; further studies are needed to ascertain its position. On the other hand, the results of Bacillus sp. 112A 16S rDNA sequence and also the variability in its fatty-acid profile (Table 1) make the isolate a unique strain of Bacillus sp., which also needs further investigation to ascertain its position.

Table 1 Comparison of fatty-acid content (%) of Bacillus sp. 87I and Bacillus sp. 112A with their closely related B. cereus and B. flexus, respectively
Full size table

Growth pattern of the selected isolates

Results showed that there was a long lag phase up to 4 and 8 h for 112A and 87I, respectively, whereas log phase was up to 28 h, followed by stationary phase (Fig. 2).

Fig. 2
figure 2

Growth pattern of the isolates

Full size image

Effect of time on PHB production

In the time course of PHB production, maximum PHB accumulation was recorded at around 32 and 36 h for 87I and 112A, respectively (Fig. 3). Accumulation of the polymer begins in the late log phase of growth and becomes maximum during the late stationary phase of growth; it is therefore important to harvest cells at the optimum time (late stationary phase) to obtain a maximum yield of PHB.

Fig. 3
figure 3

Effect of time on PHB production

Full size image

Effect of temperature and pH on PHB production

PHB was accumulated in Bacillus sp. 87I and Bacillus sp. 112A at various temperatures to estimate PHA productivity (Table 2). To compare PHA productivities, Bacillus megaterium MTCC 453 and Ralstonia eutropha MTCC 1472 were used; these are the most common PHB-producing bacteria and have been studied in detail. The PHB productivities of Bacillus sp. 87I and Bacillus sp. 112A were higher than those of B. megaterium at all temperatures. Although B. megaterium and R. eutropha could not accumulate PHB at temperatures above 40°C, strain 112A accumulated PHB even at 45°C. The PHB synthesis-related enzymes of 112A could have moderate thermostability (due to its ability to grow even at 45°C), which is an important feature, useful for industrial application of such strain. Isolate 87I could not grow at 45°C (Table 2).

Table 2 PHB produced by Bacillus sp. 87I and Bacillus sp. 112A, when grown at different incubation temperatures and different pH values
Full size table

The two selected isolates grew best at pH 7.2, and hence maximum accumulation of PHB was obtained at pH 7.2. Although the cultures grew at lower and higher pH values, a drastic fall in PHB yield was seen. Lower PHB contents were obtained at pH values of 6.8 and 7.5 (Table 2).

Under limited amounts of nitrogen, potassium, and phosphorous, PHB accumulation of Bacillus cereus SPV was reported to improve [22]. A similar and significant increase in the amount of PHB accumulation was seen in the present study in Bacillus sp. 87I and Bacillus sp. 112A by limiting the nitrogen content to 5 mM, which probably increases the activity of enzymes metabolizing PHB.

PHB production from various carbon substrates

The two isolates (Bacillus sp. 87I and Bacillus sp. 112A) produced maximum PHB with glucose but exhibited varying specificity in utilizing all other sugars studied. Galactose and starch were not included in the study involving Bacillus sp. 87I, whereas lactose was not tested with Bacillus sp. 112A. Bacillus sp. 87I produced a good PHB yield in lactose, followed by maltose, sucrose, fructose, and mannitol. In the case of Bacillus sp. 112A, the highest PHB productivity in glucose was followed by that in starch, sucrose, fructose, galactose, maltose, and mannitol (Table 3).

Table 3 PHB production by Bacillus sp. 87I and Bacillus sp. 112A grown on different carbon sources
Full size table

A Bacillus strain utilized soya molasses to produce 90% cell dry weight of the polymer without any nutrient limitation [23]. Rohini et al. [24] reported synthesis and accumulation of a higher amount of PHB (64.10%) by Bacillus thuringiensis (isolated from natural soils) when grown on glycerol, whereas Bacillus sp. 87I produced 70.04% PHB DCW when grown on glucose, higher than all earlier reports with Bacillus wild-type strains. Bacillus sp. 112A produced 67.63% PHB DCW when grown on glucose as sole carbon source.

Polymer analysis

To study its physical properties, PHB was dissolved in hot chloroform, and the film formed after evaporation of chloroform was used for analysis. PHB with molecular mass around 2 × 105 to 3 × 106 has been reported in the literature for cultures grown on long-chain hydrocarbons, while the molecular mass of the polymers also influences the crystallinity of PHA [25]. Bacillus sp. 87I produced polymer with average molecular mass of 513 kDa, whereas, the average molecular mass of the PHB polymer from Bacillus sp. 112A was of 521 kDa (Table 4).

Table 4 Physical and mechanical properties of PHB extracted from Bacillus sp. 87I and Bacillus sp. 112A grown on glucose as carbon substrate
Full size table

The T m values of the polymers produced by Bacillus sp. 87I and Bacillus sp. 112A were 152.9°C and 147.7°C, respectively (Table 4). Both polymers were found to be melt-stable; when heated above their T m to 190°C, quenched to room temperature, and reheated to 190°C, the peak at their T m was obtained again on second heating, suggesting their stability. However, when heated beyond T m to 210°C and quenched to 50°C, the peak at T m was lost on subsequent heating. The percentage crystallinity of the PHB films was studied by X-ray diffraction patterns. The percentage crystallinity was calculated from diffraction intensity data according to Vonk’s method. Polymer from Bacillus sp. 112A was more crystalline (50.17%) than polymer from Bacillus sp. 87I (45.8%) (Table 4).

Young’s modulus, tensile strength, and elongation to breakage are three important mechanical properties of polymers. The Young’s modulus of the polymers from Bacillus sp. 87I and Bacillus sp. 112A were 5.8 and 3.5 GPa, respectively. The tensile strength of PHB from Bacillus sp. 87I was 38 MPa, similar to polypropylene, suggesting the polymer to be as tough as polypropylene. However, the elongation to breakage was found to be very low for the two studied polymers when compared with polypropylene and polyethylene, suggesting their low elasticity (Table 4).

IR spectra were recorded for PHB dissolved in chloroform. Spectra of the standard PHB showed two intense absorption bands at 1,724.3 and 1,280.3 cm−1, corresponding to C=O and C–O stretching groups, respectively (Fig. 4). The polymer from Bacillus sp. 87I exhibited a sharp peak at 1,714 cm−1, corresponding to C=O stretching. The peak at 1,190 cm−1 confirmed C–O stretching in the polymer as ester of HB. The polymer from Bacillus sp. 112A showed sharp peaks at 1,722.43 and 1,276.88 cm−1, suggesting it also as PHB. Rohini et al. [24] reported that the Fourier-transform infrared (FTIR) absorption band at about 1,730 cm−1 is a characteristic of the carbonyl group and that a band at about 1,280–1,053 cm−1 characterizes valance vibration of the carboxyl group. The polymers from both Bacillus sp. 87I and Bacillus sp. 112A showed similar absorption bands, characteristic of the poly-3-hydroxybutyrate (PHB) polymer. The H1 NMR scans of the polymers from Bacillus sp. 87I and Bacillus sp. 112A were recorded. A doublet was recorded at 2.57 ppm, corresponding to the methylene group (–CH2–), while a multiplet signal at 5.27 ppm corresponded to the methyne group (–CH–) (Fig. 5). Another doublet signal at 1.30 ppm corresponded to the methyl group (–CH3–). The methyl esters showed a sharp signal around 3.50 ppm, corresponding to the CH3–O group of the esters. These results suggest the polymers to be poly-3-hydroxybutyrate (PHB).

Fig. 4
figure 4

IR spectrum of the PHB polymer produced by Bacillus sp. 87I

Full size image
Fig. 5
figure 5

NMR scan of polymer produced by Bacillus sp. 112A

Full size image

The methanolysis products of the polymer samples were analyzed using GC–MS. By analyzing the potential fragmentation patterns and the molecular mass of the fragments, the identities of specific peaks in the mass spectra were correlated to the carbonyl and hydroxyl ends of the representative hydroxyalkanoates. The polymer samples from Bacillus sp. 87I and Bacillus sp. 112A showed two fragments in mass spectrum at 45 and 74 m/z (Fig. 6). The peak at 45 m/z represented the hydroxyl end of the molecule, originating from cleavage of the bond between C3 and C4. The peak at 74 m/z represented the carbonyl end of the molecule, originating from McLafferty rearrangement [26] after cleavage of the bond between C3 and C4.

Fig. 6
figure 6

Mass spectra of methyl ester of 3-hydroxybutyrate (HB) synthesized by Bacillus sp. 112A by GC–MS analysis. The numbering of carbon atoms in methyl-3-hydroxyalkanoates is indicated in superscript

Full size image

Discussion

A number of Bacillus spp. have been reported to accumulate 9–44.5% DCW PHB [27]. By comparison, Bacillus sp. 87I produced as little as 20% DCW PHB and as much as 70.4% DCW PHB from mannitol and glucose, respectively. Similarly, Bacillus sp. 112A produced a minimum of 24.2% DCW PHB from mannitol and a maximum of 67.63% DCW PHB from glucose. A relatively high yield of PHB was obtained in these wild strains, hence these strains were considered as potent organisms for industrial application.

Morphological results (not shown) and phylogenetic analyses clearly demonstrated that strains 87I and 112A are members of the genus Bacillus. The optimal temperature for growth was between 28°C and 30°C, and the optimal pH was 7.0. Phylogenetic analysis of 16S rDNA demonstrated that these bacteria, grouped with B. mycoides (B. cereus) and B. flexus, are well-defined taxa. Further characterizations of these bacteria are necessary before proposing them as novel species.

Carbon sources serve three different functions within the organisms: biomass synthesis, cell maintenance, and PHA polymerization. Among the carbon sources tested with the strains 87I and 112A, glucose was found to be the preferred substrate for polymer accumulation. However, strain 87I also showed good polymer yield (55.5% DCW PHB) with lactose. On the other hand, strain 112A yielded 58.9% DCW PHB with sucrose and also a significant amount (50% DCW PHB) with starch as the carbon source. This study was important, since it yielded two strains of Bacillus sp. that were capable of producing polymer not only from glucose but also from carbon substrates such as lactose, sucrose, and starch.

In further studies, these strains of Bacillus spp. could be effectively utilized to produce PHB from substrates such as whey and starch-containing material.