Screening and identification of a highly lipolytic bacterial strain from barbecue sites in Hainan and characterization of its lipase

Abstract

A cheap and rapid screening method for isolation of lipolytic bacteria was established. A total of 145 lipolytic strains were isolated from oil-contaminated soil samples at barbeque sites in Haikou, China. Highly lipolytic strains were screened based on the formation of clearance zones on turbid solid media supplemented with emulsified peanut oil. One strain, C737-11, had the highest lipolytic activity and was further analyzed. This strain had physiological and biochemical characteristics that were similar to the genus Burkholderia. Phylogenetic analysis based on the ferric uptake regulator (fur) gene sequences located the strain to the clade of Burkholderia cepacia, Genomovar I, while resolution of the phylogeny based on 16S rRNA sequences was not good enough to distinguish Burkholderia species. The fermentation conditions were optimized and the optimal medium inducing or supporting lipase production was 0.5% glucose, 2% peanut oil, 2% peptone, 0.1% (NH₄)₂SO₄, 0.1% K₂HPO₄·3H₂O and 0.05% MgSO₄·7H₂O at pH 8.0. The optimal fermentation temperature was 37°C, and the best fermentation time was 72 h. Under these optimal conditions, lipolytic activity reached 10.5 U mL⁻¹. The lipase produced by C737-11 was thermally stable and had an optimal reaction pH of 8.0 and an optimal reaction temperature of 37°C.

Introduction

Lipases (EC3.1.1.3) form a family of triglyceride ester hydrolases widely distributed among plants, animals, and microorganisms. They act on the ester bonds in triacylglycerol at the ester–water interface, resulting in the reversible hydrolysis of glycerol ester bonds and, thereby also catalyzing the synthesis of glycerol esters and alkyl esters (transesterification; Jaeger et al. 1994, 1999; Arpigny and Jaeger 1999). Lipases are well known in many industrial applications including their addition to detergents, the manufacture of food ingredients, pitch control in the pulp and paper industry, and biocatalysis of stereoselective transformations (Jaeger and Reetz 1998). The potential utilization of lipases in biodiesel production from oils and fats has received much attention (Hsu et al. 2002; Noureddini et al. 2005; Ying and Chen 2007). Lipases from different sources may have different catalytic characteristics and enantio-, chemo-, and stereo-selectivities towards their substrates (Holmquist 1998; Gupta et al. 2003), so that the characterization of these lipases may provide choices for different experimental and industrial applications.

In this paper, we describe direct screening of a highly lipolytic bacterial strain isolated from barbeque sites in Hainan and the properties of its lipase.

Materials and methods

Sample preparation

Six soil samples were collected from barbecue sites in Haikou, Hainan Province, China, where the ground soil has been contaminated continuously by oils and fats over several years and is rich in lipase-active bacteria and fungi. Soil suspensions were prepared by vigorously shaking approximately 10 g soil sample in 100 mL sterile water in a flask containing glass beads.

Preparation of emulsified oil

One volume of peanut oil or palm oil was mixed with four volumes of 5% polyvinyl alcohol solution (w/v), and the mixture was emulsified by shaking at a speed of 250 rpm overnight. The emulsion was autoclaved for 20 min at 121°C. The emulsion is stable for at least half a year at room temperature.

Enrichment of lipase-producing bacteria

The soil samples contained numerous microorganisms, both lipolytic and non-lipolytic, thus enrichment of the lipolytic microorganisms was necessary before screening. A series of media using oil as the main carbon source were tested. A medium including 0.05% peptone (w/v), 0.15% KH₂PO₄ (w/v), 0.1% Na₂HPO₄ (w/v), 0.05% MgSO₄ (w/v), 0.5% (NH₄)₂SO₄ (w/v), 0.05% NaCl (w/v) and 12% sterilized peanut oil emulsion (v/v), was found to be efficient. The pH was adjusted to 6.0, 7.0 or 7.8 for enrichment of different strains. The sterilized emulsified oil was added to the medium after autoclaving at 121°C for 20 min. A 2 mL sample of each soil suspension was added to 50 mL enrichment medium in 150 mL flasks. The flasks were incubated on a shaker at 200 rpm and 28°C or 37°C for 3 days. A 0.5 mL aliquot of the suspension was subcultured in 50 mL fresh liquid medium for a second 3 days of enrichment.

Screening of lipase-producing bacteria

Screening of lipolytic bacteria was based on the formation of clearance zones around the colonies grown on semi-solid turbid media. The medium contained 0.4% peptone, 0.1% (NH₄)₂SO₄, 0.15% KH₂PO₄, 0.05% MgSO₄·7H₂O, 0.05% NaCl, 12% peanut oil emulsion, and 2% agar. The pH was adjusted to 6.0, 7.0 or 7.8. The screening medium was autoclaved at 121°C for 20 min. The plates were poured so that the medium was 2 mm thick.

The suspensions enriched in lipolytic organisms were diluted with sterilized water and plated on the screening media. The plates were incubated at 28°C or 37°C for 3 days. Lipase-producing colonies were indicated by the presence of clearance zones. Colonies with large clearance zones were selected for further purification by subculturing on the same medium several times until the morphology of colonies was uniform.

Taxonomical identification of lipolytic bacteria

Representative strains were identified according to their morphological, physiological and biochemical characteristics using standard procedures as described in the literature (Bergey et al. 1984; Boone et al. 2001)

The 16S rDNA of representative strains was amplified from genomic DNA by polymerase chain reaction (PCR) with the primers 5′-AGA GTT TGA TCC TGG CTC AG-3′ and 5′-AAG GAG GTG ATC CAG CCG CA-3′ (Wu and Shi 2008).The fragments were sequenced at Shanghai Sangon Biological Engineering Technology & Services (Shanghai, PR China). The sequences were analyzed with MacVector software (Oxford Molecular, Oxford, UK). BLAST searches using the 16S rDNA sequences as queries were performed against GenBank databases.

To further genotype the Burkholderia strain, 5′ ATG ACC AAT CCG ACC GAT CTC AA 3′ and 5′ TCA GTG CTT GCG ITN IGG GCA GTT 3′ were used to amplify the ferric uptake regulator (fur) gene fragment in bacteria (Lynch and Dennis 2008).

For phylogenetic analysis, 16S rRNA or the fur gene sequences of related taxa were obtained from GenBank databases and aligned with ClustalX 2.0 (Chenna et al. 2003). The alignment result was exported to Mega3.1 (Kumar et al. 2004). Phylogenetic trees were generated with Neighbor-Joining (NJ), Minimum Evolution (ME), and Maximum Parsimony (MP) methods with 1,000 bootstrap replicates. The trees were rooted with a 16S rDNA sequence from Burkholderia gladioli (X67038) for the 16S phylogeny, or with two fur gene sequences from B. gladioli (EU090888 and EU090889).

Optimization of fermentation conditions

The bacterial strain was inoculated in the growth medium overnight and subcultured in the lipase-induction medium supplemented with different combinations of carbon sources and nitrogen sources. The cultures were inoculated at 37°C for 72 h with shaking at 300 rpm. The growth medium contained 2% peptone, 0.1% (NH₄)₂SO₄, 0.1% K₂HPO₄·3H₂O, and 0.05% MgSO₄·7H₂O, at pH 7.0. The medium was autoclaved at 121°C for 20 min. The basal lipase-induction medium contained 2% peptone, 0.1% (NH₄)₂SO₄, 0.1% K₂HPO₄·3H₂O, 0.05% MgSO₄·7H₂O and 2% emulsified peanut oil, at pH 7.0. The medium was autoclaved at 121°C for 20 min.

Lipase assay

Lipolytic activity was measured as described previously (Hamid et al. 2003), with minor modifications. Emulsified peanut oil mixture (4 mL) and 4 mL 0.1 M Tris-HCl (pH 7.8) were added to a 150 mL flask and mixed thoroughly; the mixture was incubated in a water bath at 37°C for 5 min, then 2 mL crude lipase (supernatant of culture medium) was added to the reaction mixture and incubated at 37°C for 20 min; 10 mL 95% ethanol was added to terminate the reaction. Phenolphthalein (1%) was then added to the mixture, which was titrated with 0.05 M NaOH until the end point was reached. One unit of lipase activity (U) is defined as the release of 1 μmol fatty acid per minute under the above conditions. The amount of fatty acid liberated and lipase activity were calculated accordingly.

Optimization of lipase assay conditions

Lipase activity was analyzed over a pH range of 5.0–11.0 and a temperature range of 20°C –80°C. The thermal stability of the lipase was assayed by incubating the supernatant of the culture medium containing lipase secreted by C737–11 at 50°C, 60°C and/or 70°C for 1 h; an aliquot was removed every 20 min to measure the remaining enzyme activity. pH stability experiments were carried out by incubating the supernatant of the culture medium in 0.1 M Tris-HCl (pH 6.0–11.0) at 37°C for 1 h, followed by measuring the residual lipase activity as described above at pH 7.8 and 37°C.

Data analysis

Single factor ANOVA was used to compare the effects of different carbon and nitrogen sources. To test the effects of different carbon sources on lipase induction from C737–11, 0.5% (w/v) sucrose, glucose, maltose, galactose, and mannitol were used as supplements to the basal lipase-induction medium. Nine replicates were performed. Peptone, soybean powder, yeast extract, beef extract, and corn powder were used to test the effects of different nitrogen sources on lipase induction: 2% (w/v) of each nitrogen source was used; four replicates were performed. Data were analyzed using SAS (Release 9.1.3). Duncan’s test was performed to determine the significance of the differences between the group means.

Results and discussion

Isolation of lipase-producing bacteria

The method most commonly used to screen lipolytic bacteria is based on the formation of clear halos around colonies grown on tributyrin- or triolein-containing agar plates (Jaeger et al. 1994; Bora and Kalita 2007; Ertugrul et al. 2007). We developed a simple procedure that requires only household peanut oil or palm oil. Oils and fats are insoluble in water, so that a fine and stable oil emulsion is important to the subsequent medium preparation and enzyme activity assay. Some formulations and protocols have been tried for preparing an oil emulsion of good quality (San Clemente and Vadehra 1967; Sorokin and Jones 2009). The best formulation was achieved by suspending one volume of analytical pure olive oil in four volumes of 5% gellan gum (an emulsifier and stabilizer) by sonication on ice (Sorokin and Jones 2009). We used polyvinyl alcohol, which is more commonly available, as the emulsifier instead. One volume of peanut oil or palm oil was emulsified in four volumes of 5% polyvinyl alcohol solution (w/v) by shaking at a speed of 250 rpm overnight. The autoclaved peanut oil emulsion is stable for at least half a year at room temperature.

The turbid agar plates plus the enrichment medium supplemented with emulsified peanut oil or palm oil were effective in the enriching and screening of lipase-producing bacterial. After enrichment, most bacteria clones formed clearance zones on the turbid agar plates (Fig. 1a). The lipolytic bacterial strains were further purified by subculturing on turbid solid medium until the morphology of the clones was uniform. In total, 145 lipolytic bacterial strains were isolated from oil-contaminated soil samples.

Fig. 1a–d

Screening of lipolytic strains and morphology of C737-11. a A representative plate showing the screening method for lipolytic bacteria. b Clearance rings of strain C737-11 on solid turbid medium containing peanut oil. c Colonial morphology of C737-11. d Cell morphology of C737-11 after Gram staining

The lipolytic activity of the purified strains was determined by dotting the strains on solid turbid plates. Strain C737-11 produced the largest clearance zones among the strains tested (Fig. 1b). Representative strains were chosen for preliminary lipase production. All of the cultures showed good lipolytic activities (Table 1). Strain C737-11 exhibited a lipolytic activity as high as 3.01 U mL⁻¹, followed by strain H728-4 with an activity of 1.75 U mL⁻¹. Strains H728-11, E7837-2, C7828-1, and F7828-1 had moderate lipolytic activities of 1.30, 1.20, 1.19, and 1.07 U mL⁻¹, respectively. Strain C737-11 was selected for further analysis.

Table 1 Comparison of lipolytic activity of the representative bacteria strains

Classical and phylogenetic identification of strain C737–11

Colonies of C737-11 grown on Luria-Bertani agar plates were round, yellow and opaque with a wet, convex and smooth surface (Fig. 1c). No secreted pigments were observed in the medium. Conventional physiological and biochemical examinations showed that the cells were motile, short rods, Gram negative (Fig. 1d), aerobic, oxidase-negative, catalase-positive, methylred-positive, V-P-negative, indole-negative, denitrification-negative and positive to assay of gelatin liquefaction. C737-11 utilized glucose, sucrose, maltose, mannitol, soluble starch, or galactose as the sole carbon source for cell growth and lipase production. It also utilized various organic and inorganic nitrogen sources including ammonium sulfate, beef extract, peptone, yeast extract, soybean powder, and corn powder as the sole nitrogen source for growth and lipase production. It grew well at temperatures between 20 and 40°C. Based on these features, strain C737-11 was identified as Burkholderia cepacia complex species.

Burkholderia cepacia complex (BCC) is a group of Gram-negative bacteria composed of at least nine different species, including B. cepacia, Burkholderia multivorans, Burkholderia cenocepacia, Burkholderia vietnamiensis, Burkholderia stabilis, Burkholderia ambifaria, Burkholderia dolosa, Burkholderia anthina, and Burkholderia pyrrocinia (Lipuma 2005). Although BCC species are capable of causing life-threatening respiratory tract infections in persons with cystic fibrosis (Coenye et al. 2001; Mahenthiralingam et al. 2001; Magalhaes et al. 2002), studies from several countries reiterate that two species, B. cenocepacia and B. multivorans, account for most BCC infections in cystic fibrosis (Smalley et al. 2003; Lipuma 2005), while B. dolosa and B. vietnamiensis were occasionally isolated in cystic fibrosis (Coenye et al. 2001; Lipuma 2005). The BCC species are also of ecological importance and some species have been isolated from diverse environments including soil, water, and plants (Singh et al. 2006; Bartholdson et al. 2008; Jacobs et al. 2008).

To identify which species strain C737-11 belongs to, the 16S rDNA fragment was amplified from its genomic DNA and sequenced. BLAST searches in the GenBank databases revealed 99.9% identities with BCC Genomovar III, B. cenocepacia strain LMG 12614 (AF311970.1), and BCC Genomovar I, B. cenocepacia strains ATCC700070 (AY741360) and ATCC17759 (AY741334). Phylogenetic analysis based on the 16S rRNA sequences showed that C737-11 was closely related to B. cepacia and B. cenocepacia (Fig. 2). Unfortunately, the resolution of the phylogeny was not good enough to distinguish B. cepacia and B. cenocepacia, and both B. cepacia and B. cenocepacia were dispersed in several clades in the phylogenetic tree, with very low bootstrap supports, similar to previous reports (LiPuma et al. 1999; Mahenthiralingam et al. 2000; Brown and Govan 2007; Lynch and Dennis 2008).

Fig. 2

Phylogenetic analysis of C737-11 based on 16S rDNA sequences. The tree was rooted with a 16S rDNA sequence of Burkholderia gladioli (X67038). The sequences were aligned with ClustalX 2.0, using the Blosum 30 matrix. The alignment result was exported to Mega3.1. Phylogenetic trees were calculated with Neighbour-Joining (NJ), Minimum Evolution (ME) and Maximum Parsimony (MP) methods. Only the phylogeny calculated with the NJ method is presented in this figure. The bootstrap values provided on the branches were calculated using the NJ method with 1,000 replicates

The fur gene sequence was reported to be more effective in rapid differentiation of BCC species (Lynch and Dennis 2008). Therefore, the fur gene fragment from C737-11 was amplified and sequenced. The fur gene in the Burkholderia genus is more divergent than the 16S gene. BLAST searches with the fur gene sequence against the GenBank databases resulted in the highest identities of 99% with B. cepacia type strain ATCC 25416. Phylogenetic trees calculated with NJ, ME and MP methods clearly separated the BCC species into nine clades (Fig. 3), and C737-11 was located in the clade of B. cepacia (Genomovar I, BCC1) with bootstrap support of 93%, 97% and 72% as calculated with NJ, ME and MP methods, respectively. Thus, we concluded C737-11 is a Burkholder cepacia strain.

Fig. 3

Phylogenetic analysis of C737-11 based on ferric uptake regulator (fur) gene sequences. The tree was rooted with (fur) gene sequences from Burkholderia gladioli (EU090888, EU090889). The sequences were aligned with ClustalX 2.0, using the Blosum 30 matrix. The alignment result was exported to Mega3.1. Phylogenetic trees were calculated with NJ, ME and MP methods. The topologies of trees generated with different methods were similar. The topology presented in this figure was generated with the NJ method. The bootstrap values provided on the branches were calculated using the NJ, ME, and MP methods with 1,000 replicates, respectively

Burkholderia cepacia strains have been isolated from a variety of habitats, including naturally growing surface-sterilized roots (Singh et al. 2006; Mendes et al. 2007), stems (Mendes et al. 2007), field soil (Jacobs et al. 2008), water (Koenig et al. 1995), and even machines (Vaisanen et al. 1998). They are not only conditionally pathogenic to plants and animals, but also promote biomass accumulation and grain yield in greenhouse-grown rice plants inoculated with some isolates (Singh et al. 2006), and/or serve as antagonistic bacteria (Mu et al. 2008).

Optimization of lipase-induction conditions for C737-11

Lipases from different B. cepacia strains may have different enzymatic properties that are useful for different applications (Yang et al. 2002; Noureddini et al. 2005; Turner et al. 2006). To induce lipase production from C737-11 for enzymatic analysis, the fermentation conditions were optimized.

The production of extracellular lipases is often promoted by the presence of lipids in the growth medium (Gupta et al. 2004; Kiran et al. 2008; Yan and Yan 2008). When 2% peanut oil was added to the basal growth medium, the lipolytic activity of the culture medium increased by 284%. A readily available carbon source in the medium may repress lipase production (Kiran et al. 2008). However, when 0.5% sucrose was added to the basal lipase-induction medium containing 2% peanut oil, the lipolytic activity of C737-11 increased by 154%. The addition of 0.5% maltose, glucose, manitol, or galactose enhanced lipase production by 350%, 342%, 186%, or 221%, respectively (Table 2). Glucose was then used to promote lipase production in further experiments.

Table 2 Effect of carbon and nitrogen sources on lipase production. Data presented as mean ± standard derivation of percentage activity compared to the basal lipase-induction medium

A nitrogen source is an indispensable nutritional component for microbial growth. Peptone was used as the nitrogen source in both growth and lipase-induction media generally in this research. Other nitrogen sources were also tested, however, and when peptone was replaced by yeast extract, beef extract, soybean powder, or corn powder, lipase production decreased significantly (Table 2). Therefore, peptone was selected as the optimal nitrogen source.

The pH of the medium is also an important factor influencing lipase production (Papon and Talon 1988; Ghosh et al. 1996; Shariff et al. 2007). The optimal initial pH for C737-11 was 8.0 (Fig. 4), which is relatively high when compared to other lipase producing Pseudomonas sp. (pH 5.5; Sarkar et al. 1998), and low when compared to Burkholderia sp. C20 (pH 9.0; Liu et al. 2006) and HY-10 (pH 8.5; Park et al. 2007).

Fig. 4

Effect of initial pH on lipase production

In summary, the optimal lipase producing medium was 0.5 % glucose, 2% peanut oil, 2% peptone, 0.1% (NH₄)₂SO₄, 0.1% K₂HPO₄·3H₂O, and 0.05% MgSO₄·7H₂O, at pH 8.0. Under optimal conditions, which included a fermentation temperature of 37°C and fermentation time of 72 h, lipase production reached 10.5 U mL⁻¹.

Properties of the lipase produced by C737-11

Different Burkholderia strains produce lipases with different enzymatic properties. The lipase produced by C737-11 had its highest activity at 37°C. It was sensitive to low temperatures; at 30°C it had only 20% of its highest activity. This enzyme was relatively active at higher temperatures, and maintained about 70%, 65%, and 50% of its highest activity at 50°C, 60°C, and 70°C, respectively (Fig. 5a). The optimal reaction temperature for one previously reported Burkholderia strain, G63 was 37°C (Yang et al. 2007), and for strains HY-10 and C20, it was 60°C and 55°C, respectively (Liu et al. 2006; Park et al. 2007), while the lipase from another Burkholderia strain had a low optimal reaction temperature of 25°C (Liu et al. 2009).

Fig. 5a–d

Properties of the lipase produced by C737-11. a Effect of reaction temperature on enzyme activity. b Effect of pH on enzyme activity. c Stability of the enzyme at different pH values. d Stability of the enzyme at different temperatures

Most bacterial lipases have neutral or alkaline pH optima (Gupta et al. 2004). The optimal reaction pH for lipases produced by different Burkholderia strains was similar with slight variations. The optimal pH for the lipase produced by C737-11 was pH 8.0 (Fig. 5b), similar to a previously reported Burkholderia strain GXU56 (Wei and Wu 2008). The optimal pH for Burkholderia sp. strains HY-10, C20, and 34–5 was 8.5, 9.0 and 9.6, respectively (Liu et al. 2006, 2009; Park et al. 2007).

Generally, bacterial lipases are stable over a wide pH range (Yadav et al, 1998; Dong et al. 1999; Gupta et al. 2004; Hasan et al. 2009). Lipase produced by C737-11 was stable between pH 6.0 and 8.0 (Fig. 5c), and the activity dropped sharply after incubation at pH 9.0 and 10.0 (Fig. 5c). The lipase retained 97%, 88%, and 63 % of its original activity after incubation at 50, 60, and 70°C, respectively for 20 min (Fig. 5d), and 86%, 55%, and 23 % of its original lipase activity was retained after being incubated at 50, 60 and 70°C, respectively, for 40 min, suggesting that the lipase produced by C737-11 was moderately thermally stable. Little data on the thermal stability of other Burkholderia strains are available. The lipase from a Burkholderia strain RGP-10 was stable at 50°C (Rathi et al. 2001). More stable lipases were reported from the thermophilic Bacillus sp. J33, which was stable for 12 h at 60°C (Nawani and Kaur 2000) and from Aspergillus terreus, which retained 100% activity after incubating at 60°C for 24 h (Yadav et al. 1998). An extremely thermostable lipase was reported from Bacillus stearothermophilus SB-1, which had a half-life of 25 min at pH 3.0 and 15 min at pH 6.0 at 100°C (Bradoo et al. 1999).