Cell-bound lipases from Burkholderia sp. ZYB002: gene sequence analysis, expression, enzymatic characterization, and 3D structural model

Background The whole-cell lipase from Burkholderia cepacia has been used as a biocatalyst in organic synthesis. However, there is no report in the literature on the component or the gene sequence of the cell-bound lipase from this species. Qualitative analysis of the cell-bound lipase would help to illuminate the regulation mechanism of gene expression and further improve the yield of the cell-bound lipase by gene engineering. Results Three predictive cell-bound lipases, lipA, lipC21 and lipC24, from Burkholderia sp. ZYB002 were cloned and expressed in E. coli. Both LipA and LipC24 displayed the lipase activity. LipC24 was a novel mesophilic enzyme and displayed preference for medium-chain-length acyl groups (C10-C14). The 3D structural model of LipC24 revealed the open Y-type active site. LipA displayed 96 % amino acid sequence identity with the known extracellular lipase. lipA-inactivation and lipC24-inactivation decreased the total cell-bound lipase activity of Burkholderia sp. ZYB002 by 42 % and 14 %, respectively. Conclusions The cell-bound lipase activity from Burkholderia sp. ZYB002 originated from a multi-enzyme mixture with LipA as the main component. LipC24 was a novel lipase and displayed different enzymatic characteristics and structural model with LipA. Besides LipA and LipC24, other type of the cell-bound lipases (or esterases) should exist. Electronic supplementary material The online version of this article (doi:10.1186/s12896-016-0269-6) contains supplementary material, which is available to authorized users.


Background
Microbial lipase (triacylglycerol lipase, EC 3.1.1.3) catalyze hydrolysis of the long chain triglycerides, or the reverse reaction. Besides hydrolysis activity, lipases also displayed alcoholysis, aminolysis, interesterification, and esterification activity, etc. with rigorous regioselectivity, stereoselectivity, and chemoselectivity [1]. As one kind of non-aqueous enzymes, lipases kept high catalysis efficiency in organic solvent systems or micro-aqueous systems, and were widely used in many industrial fields [2].
Microbial strains can produce multiple types of lipases. Candida rugosa produced more than five types of lipase isoenzymes (Lip1-Lip5), which shared high sequence identity, but displayed significantly different enzymatic characteristics [3]. Different types of lipases produced by a specific microbial strain always were distributed to different cell compartments, respectively. Lipase LipA from Pseudomonas aeruginosa either was secreted into the culture medium, or interacted with the polysaccharide alginate and then anchored on the cell surface [4][5][6], while Esterase EstA from P. aeruginosa was located in the outer membrane [7,8].
Cell-bound lipase could be directly used as a whole cell biocatalyst. Compared with the extracellular enzyme, whole cell biocatalysts displayed many advantages, including high stability in the long-term, inexpensive preparation, independence of the exogenous co-factor for redox reaction, etc. [9]. In previous research, Burkholderia sp. ZYB002 produced both extracellular lipase and cell-bound lipase [10,11]. The cell-bound lipase from Burkholderia cepacia displayed excellent interesterification activity for biodiesel production and highly enantioselective hydrolysis activity for L-menthol synthesis [12,13]. However, there wasn't any report on the type or the gene sequence of the cell-bound lipase from B. cepacia.
In this article, three predictive cell-bound lipase genes from Burkholderia sp. ZYB002, lipA, lipC21 and lipC24, were cloned and expressed in E. coli, respectively. Furthermore, the component of the cell-bound lipase from Burkholderia sp. ZYB002 was analyzed.

Bacterial strains and plasmids
The bacterial strains and plasmids used in this study are listed in Table 1. Briefly, E. coli DH5α was used as the host strain for plasmid amplification, and E. coli BL21(DE3) and E. coli Origami2 (DE3) were used as the expression host strain for three lipase genes, lipA, lipC21 and lipC24, respectively. Burkholderia sp. ZYB002 was the lipase-producing strain, which was isolated and identified in our lab [14]. Antibiotics were added as required to the final concentrations of 60 μg/mL ampicillin, 35 μg/mL chloramphenicol, 50 μg/mL kanamycin, 100 μg/mL trimethoprim, 50 μg/mL gentamicin.

Chemicals and biochemistry reagents
High-fidelity DNA polymerases, restriction enzymes, T 4 -DNA ligases, PCR purification kits, the DNA Gel-Extraction Kits, DNA markers, and protein markers etc. were purchased from Takara Biotechnology Co. Ltd (Dalian, China). Primers synthesis and DNA sequencing was completed by Sangon Biotechnology Co. Ltd (Shanghai, Beijing). All antibiotics were purchased Gene cloning and sequence alignment of lipA, lipC21 and lipC24 The full lengths of three different lipase genes, lipA/ lipB, lipC21 and lipC24 were amplified by PCR using the genomic DNA from Burkholderia sp. ZYB002 as the templates. The primer pairs for PCR were listed in the Table 2. All PCR conditions and PCR procedures used in this research were given in the Additional file 1. PCR products were ligated into pMD18-T simple vector to construct the cloning plasmid pMD18T-lipAB, pMD18T-lipC21, and pMD18T-lipC24, respectively. The three lipase genes were sequenced in full length. The nucleotide sequences of lipA/lipB, lipC21 and lipC24 have been deposited in the GenBank database. To construct the phylogenetic tree, the deduced amino acid sequences of LipA, LipC21 and LipC24, were submitted to BLAST at the NCBI web site, respectively. The retrieved-sequences displaying over 30 % sequence identity to LipA, LipC21 or LipC24 were selected and then aligned using BioEdit editor (Version 7.0.1). The phylogenetic tree was constructed using the software MEGA4.
Expression of E. coli BL21(DE3)-pEDSF-lipC21/pGro7 was induced by IPTG and L-arabinose, respectively, as described by Pérez et al. [15]. In brief, 0.5 mg/mL L-arabinose was initially added into the culture medium to induce expression of the chaperone gene of groES-groEL. IPTG (1 mmol/L final concentration) was not added into culture medium until the cell density (OD 600 ) reached 0.6. Induction incubation of IPTG was lasted for 16 h at 25°C and the cells were then collected by centrifugation.
Purification of LipA, LipC21, and LipC24 E. coli cells were lysed using sonication and the supernatant was collected by centrifugation. The recombinant protein carrying a (His) 6 -tag was purified from the supernatant using immobilized metal-affinity chromatography (HisTrap HP, 1 mL, GE Healthcare) followed by anion exchange chromatography (HiTrap DEAE F. F., 1 mL, GE Healthcare). Before loaded onto the HisTrap chromatography column, the supernatant was incubated with 2 mmol/L ATP for 10 min at 37°C to dissociate the recombinant protein/chaperone complex.
The loading buffer for affinity chromatography column consisted of the following components, 20 mmol/L Na 2 HPO 4 -NaH 2 PO 4 buffer (pH7.5),20 mmol/L imidazole,500 mmol/L NaCl. The recombinant protein was eluted using a linear concentration gradient from 20 mmol/L to 1 mol/L imidazole in the same buffer, and the active fractions (or the target fractions identified by anti-His Western-blot) were pooled and dialyzed against 50 mmol/L Tris-HCl buffer (pH7.5). The desalted protein solution was loaded on anion exchange chromatography column using 50 mmol/L Tris-HCl buffer (pH7.5) and eluted using a linear concentration gradient from 0 mol/L to 1 mol/L NaCl in the same buffer.
The homogeneity of the purified protein was determined by SDS-PAGE on a 12 % separating gel in the presence of 0.1 % SDS. The protein concentration was analyzed using the Bradford method, with bovine serum albumin as standard.

Biochemical characterization of LipC24
Temperature optimum and temperature stability The optimal temperature was determined by incubating the standard reaction mixture at different temperatures ranging from 30°C to 60°C, and the maximum lipase activity was considered 100 %. To analyze temperature stability, the LipC24 preparation was incubated at 40°C and aliquots were continuously taken at 3-min interval to assay the residual activity. Inactivation process of LipC24 preparation was continued until 80 % of the activity was lost. Half-life of thermal inactivation was calculated using the method as described by Zhao and Arnold [16].

pH optimum and pH stability
The optimal pH for lipase activity was determined by incubating the lipase with substrate in a suitable buffer at various pH ranging from 4 to 9, and the maximum lipase activity was considered 100 %. The corresponding buffers were NaAc/HAc(pH4.0-5.0), Na2HPO4-NaH2PO4 (pH6.0-8.0), and Gly/NaOH (pH9.0), respectively, and the concentrations of all used buffers were 20 mmol/L. To determine the effect of pH on lipase stability at pH ranging from 6.0 to 8.5, aliquots of the concentrated LipC24 preparation were diluted five-fold in the corresponding buffer (pH 6.0, pH6.5, pH7.0, pH7.5, pH8.0 and pH8.5) and then incubated for 24 h at 4°C. The residual lipase activity after incubation was determined and the lipase activity at the start was taken as 100 %.

Positional specificity assay
Positional specificity was determined by analyzing lipolysis products of triolein by thin-layer chromatography (TLC) on silica gel GF254, following the procedure described by Rahman et al. [17]. In brief, the reaction mixtures containing 0.1 mol/L of triolein, 1.3 mL Na 2 HPO 4 -NaH 2 PO 4 (20 mmol/L, pH7.4), and 5 U(500 μL) LipC24 solution were shaken at 200 rpm at 40°C for 3 h. The reaction products were extracted with n-hexane and then analyzed by TLC. The silica gel plate was developed in a mixture of chloroform and acetone (96:4).
The lipase activity of the purified LipC24 was measured using spectrophotometric assay under standard assay conditions, as described by Kordel et al. [18]. The spectrophotometric assay method was used in the whole experiment unless stated otherwise. All reactions were carried out at 40°C and 20 mmol/L of Na 2 HPO 4 -NaH 2 PO 4 buffer (pH7.5). One unit of lipase activity was defined as the amount of LipC24 that liberated 1 μmol 4-nitrophenol from 4-nitrophenyl fatty acid ester per min.
Site-directed mutagenesis of the lipC24 gene Amino acid substitutions (Ser 179 Ala) in LipC24 were performed using the Quickchange site-directed mutagenesis method. The plasmid pEDSF-lipB-lipC24 was used as template and the complementary mutagenic oligonucleotide primers were listed in Table 2. PCR products were firstly hydrolyzed using DpnI restriction endonuclease to remove methylated parental template DNA, and then transformed into E. coli Origami2 (DE3). The desired nucleotide substitutions was confirmed by DNA sequencing.
Induction expression of lipC24-Ser 179 Ala and purification of LipC24-Ser 179 Ala was carried out using the same method as described above.

Construction of the lipA-inactivation mutation strain and the lipC24-inactivation mutation strain
To rapidly inactivate lipA gene or lipC24 gene of Burkholderia sp. ZYB002, the suicide plasmid pBCMB-S3 (for lipA-inactivation) and pBCMB-S5 (for lipC24inactivation) were constructed, respectively. Construction of plasmid pBCMB-S3 and plasmid pBCMB-S5 was adopted the identical method and flow diagram (Additional file 2: Figure S1 and Figure S2). Details of the construction of the lipA-inactivation mutation strain was only given.
The suicide plasmid pBCMB-S3 was constructed as follows. Firstly, the full length of the trimethoprim resistance gene was cloned from plasmid pBBR1TP using tmpF/tmpR primers incorporated in the Bgl II restriction site. The PCR product was digested with Bgl II, and then ligated into the Bgl II-digested plasmid pJQ200SK. The resulting plasmid was designated pBCMB-S1. Secondly, the fragment of the lipA gene was amplified using lipAMF/lipAMR primers, which were designed to add a BamH I site and a Xho I site at the 5'-terminal and the 3'-terminal of the lipA gene fragment, respectively. The PCR product (named as lipA') was digested with the respective enzymes and then ligated to the Xho I/BamH I-digested plasmid pBCMB-S1. The resulting plasmid was designated pBCMB-S2. Thirdly, the gfp gene was cloned from plasmid pEGFP-N1 using gfpF/gfpR primers incorporated the Pst I restriction site. The PCR product was digested with Pst I, and then ligated into the Pst Idigested plasmid pBCMB-S2. The resulting plasmid was designated pBCMB-S3.
The suicide plasmid pBCMB-S3 was delivered to Burkholderia sp. ZYB002 by triparental mating as described previously [20]. Candidate mutants were primarily selected on trimethoprim Luria-Bertani (LB) medium and identified by PCR, and further confirmed by Southern blot hybridization using the gfp gene fragment labeled with digoxigenin as a probe. The corresponding mutant strain was designated Burkholderis sp. ZYB002-ΔlipA.

Cell-bound lipase production and activity assay
Cell-bound lipase production was carried out as described by Shu et al. [11]. Cell-bound lipase activity was determined using alkali titration method as described by Saxena et al. [21]. All reactions were carried out at 40°C and 20 mmol/L Na 2 HPO 4 -NaH 2 PO 4 buffer (pH7.5) unless stated otherwise. One unit of cell-bound lipase activity was defined as 1 μmol of fatty acid produced from olive oil per min by the cell culture of 1 OD 600 under the standard assay conditions.

Sequence of LipA, LipC21 and LipC24
Nucleotide sequences of the lipA/B, lipC21 and lipC24 have been deposited in the GenBank database under the accession No. EU768869, No. KF192626, No. KF438175, respectively. Nucleotide sequence analysis revealed that the lipA ORF, lipC21 ORF and lipC24 ORF coded for a putative protein of 364 amino acids, 427 amino acids and 438 amino acids, respectively. Protein sequence analysis revealed that the LipA displayed 96 % identity with the known lipase from Pseudomonas sp. KWI-56 [22], while all homologous protein sequences of LipC21 and LipC24 were the putative and uncharacterized lipases from the whole genomic DNA sequences. Moreover, protein sequence alignment through BLAST did not reveal any sequence identity among LipA, LipC21 and LipC24 (Fig. 1).

Expression and enzymatic characterization of LipA, LipC21 and LipC24
Soluble LipA, LipC21 and LipC24 could be obtained only when lipA, lipC21 and lipC24 were co-expressed with their corresponding chaperone genes (Fig. 2). The soluble expression of lipA and lipC24 required the assistance of the lipase-specific folding gene, lipB. However, co-expression of the chaperone GroEL-GroES gene is the prerequisite for the soluble expression of lipC21. Moreover, the soluble expression level of LipC24 could be significantly increased when E. coli Origami2 (DE3) was used as the expression host strain (Data not shown).
Except for LipC21, both LipA and LipC24 displayed lipase activity. As most reports on the enzymatic characterization of the LipA from B. cepacia, the relative molecular weight of the LipA from Burkholderia sp. ZYB002 was 34 kDa. LipA was an alkaline mesothermalactive lipase [23]. The optimum temperature and pH of LipA for hydrolysis activity were 40°C and 8.0, respectively [23]. The enzymatic characterization of LipC24 was totally different from that of LipA. The LipC24 was purified 17.7-fold from the supernatant of the E. coli cell lysate and yielded 21.49 % of the initial activity. The specific activity of LipC24 was 15.63 U/mg using 4nitrophenyl palmitate as substrate (Table 3), which was far lower than that of LipA (253.82 U/mg for 4nitrophenyl palmitate) [23]. SDS/PAGE analysis of LipC24 displayed a single band, which corresponded to a molecular mass of 45 kDa (Fig. 3). The optimum temperature and pH of LipC24 for hydrolysis activity were found to be 40°C and 7.5, respectively. The LipC24 could be kept stable in the pH range 7.0-8.0 for 24 h at 4°C, while the half-time of the LipC24 was only 16 min at 40°C (Fig. 4). LipC24 was less stable than LipA. LipA displayed excellent thermostability up to 65°C and could keep stability over a broad pH range from 3.0 to 10 [23]. The LipC24 indicated a clear preference for esters with the medium acyl chain length (C10-C14) when assayed using 4-nitrophenyl derivatives ( Table 4). The LipC24 exhibited a simple Michaelis-Menten kinetics for pNPM hydrolysis. The values of K m and V max of LipC24 were 0.37 ± 0.07 mmol/L and 138.8 ± 7.90 μmol · min −1 · mg −1 , respectively (Fig. 5). Michaelis constan K m of LipC24 is less than that of lipase from Bacillus sp.. Accordingly, the maximum reaction velocity V max of LipC24 was higher than that of Bacillus sp. [24]. LipC24 cleaved not only the 3-positioned ester bonds, but also the 2positioned ester bond of triolein (Fig. 6). Thus, LipC24 could nonspecifically hydrolyze the ester bonds of triolein. The same experiment results were verified with other Pseudomonas sp. lipases [17,25].

Protein sequence and structural model analysis of LipC24
There were several conserved sequence blocks between the deduced amino acid sequence of LipC24 and the other putative homologous lipases. In block 3, there was a conserved pentapeptide Gly-Tyr-Ser-Gly-Gly, in which the catalytic serine residue was embedded in most lipases. Besides the conserved serine residue, there were three conserved aspartate residues in block 1, block 2 and block 5, and a conserved histidine residue in block 5 (Additional file 2: Figure S3). The 3D homology model of LipC24 presented the characteristics of a canonical α/ β-hydrolase fold, in which parallel or mixed β sheet in the molecular center was surrounded (or connected) by helices (Fig. 7a). A hydrogen bond network was formed among Ser 179 , Asp 336 , and His 367 , which constituted the catalytic triad (Fig. 7b). Mutant of LipC24-Ser 179 Ala lost 100 % lipase activity, which confirmed the function of Ser 179 in the active site. The oxyanion hole consisted of Ala 82 and Gly 180 , which stabilized the transient state of LipC24-ethyl acetate complex (Fig. 7c). The substratebinding pocket of LipC24 displayed the distinct open Ytype structure (Fig. 7d).

Component of the cell-bound lipase
The cell-bound lipase activity of Burkholderis sp. ZYB002-ΔlipA and Burkholderis sp. ZYB002-ΔlipC24 significantly decreased to 58 % and 86 % of its original activity, respectively (Fig. 8a). The cell-bound lipase activity originated from a multi-enzyme mixture in which LipA was the main component. Besides LipA and LipC24, other type of lipases could exist on the cell surface of Burkholderis sp. ZYB002.

Discussion
Different types of lipase produced by a specific microbial strain always displayed a totally different 3D structure and enzymatic characterization [26][27][28]. In previous research, cell-bound lipase from B. cepacia displayed excellent catalytic activity for organic synthesis [12,13]. However, there was not any report on the gene sequence Fig. 2 Map of the expression plasmid for lipA, lipC21 and lipC24. a pEDSF-lipB-lipA was derived from pACYCDuet-1, which was inserted lipA gene at the MCS1 site and the chaperone lipB gene at the MCS2 site; b pEDSF-lipC21 was derived from pET28a, which was inserted lipC21 gene at the MCS site. To obtain the soluble expression of lipC21, plasmid pEDSF-lipC21 and plasmid pGro7 must be co-transformed into E. coli BL21(DE3); c pEDSF-lipB-lipC24 was derived from pACYCDuet-1, which was inserted lipC24 gene at the MCS1 site and the chaperone lipB gene at the MCS2 site , were distributed on the cytoplasmic membrane. The genes, lipC21 and lipC24 were situated on the gene cluster for ammonia metabolism and pilus synthesis, respectively (Fig. 8b). In P. aeruginosa, part of the extracellular lipase could be anchored on the cell surface and act as cell-bound lipase [4][5][6]. Three speculative cell-bound lipase genes (lipA, lipC21 and lipC24) from Burkholderia sp. ZYB002 were cloned and expressed in E. coli. Among LipA, LipC21 and LipC24, only LipA displayed a high sequence identity with the known extracellular lipase from Pseudomonas sp. KWI-56 [22], which suggested that LipA was the authentic triacylglycerol lipase. The protein sequences of LipC21 and LipC24 had not any sequence identity with known lipases or esterases, which could lead to speculations that LipC21 and LipC24 were the novel lipases (Fig. 1).
The expression soluble lipases required different chaperone protein genes. It was necessary for lipA and lipC24 to be co-expressed with the lipB gene. Due to the strong hydrophobicity, a 70-amino acid residue fragment at the N-terminal of LipB had to be truncated when lipB was heterogeneously co-expressed in E. coli   [29]. Among lipB, groES-groEL gene, dnaK-dnaJ-grpE gene and tig gene, it was only the chaperone groES-groEL gene that improved the soluble expression level of lipC21. GroEL-GroES was also reported to be necessary for the soluble expression of the family VIII lipase lipBL from Marinobacter lipolyticus [15].
Enzymatic characterization and 3D structure of LipC24 was totally different from that of LipA. LipC24 displayed high activity in the neutral buffer (pH7.0-7.5) and mesothermal reaction conditions. Furthermore, LipC24 would sharply abolish the lipolytic activity when LipC24 was kept at high temperature, alkaline solution, or acid solution, respectively. On the contrary, LipA was thermostable, alkaline-tolerant, and organic solvent-resistant [30,31]. The open Y-type active site of LipC24 was totally different from the funnel-shaped active site of LipA [32]. In the molecular model of LipC24, a predictive intramolecular disulfide bond was formed between Cys 352 and Cys 395 (PredictProtein 2013 server, https://www.predictprotein.org/), which corresponded to the requirement of the host strain, E. coli Origami2 (DE3) for the soluble expression of lipC24.
Although both titrimetric assay method and colorimetric assay method were widely used for lipase activity  determination, there were obvious differences between the two assay methods [33,34]. Olive oil or other triacylglycerol was used as the substrate in the titrimetric assay method, while 4-nitrophenyl esters were always used as the substrate in the colorimetric assay method. However, 4-nitrophenyl esters could be permeated into the cytoplasm [35][36][37], and was hydrolyzed by the intracellular lipolytic enzymes (including lipase and esterase). 4-nitrophenyl esters could not be used as the substrates for activity determination of the cell-bound lipase. In the present work, membrane-impermeable olive oil and the alkali titration assay method was used for the activity determination of the whole cell lipase. The cell-bound lipase activity of Burkholderis sp. ZYB002-ΔlipA decreased by 42 % of the total cell-bound lipase activity Burkholderis sp. ZYB002 (Fig. 8a), which demonstrated that LipA was the main component of cell-bound lipases. It had been reported that microbial strains could simultaneously produce extracellular lipases and various kinds of biosurfactants (rhamnolipid, lipopolysaccharide, polysaccharide alginate, etc.) when various oils or lipids were used as the inducer or carbon source [4,38]. Part of biosurfactants were firmly associated with the outer membrane of the host strain and could interact with lipases by electrostatic interaction, which resulted in cell surface anchoring of the extracellular lipases [5,6]. LipC24 contributed 14 % of the total cell-bound lipase activity Burkholderis sp. ZYB002 (Fig. 8a). Besides LipA and LipC24, other type of the cell-bound lipases (or esterases) could exist. Further analysis of the whole genome DNA sequence of Burkholderia cepacia J2315 predicted several esterase gene sequences, including two novel family VIII esterase genes, estVIII-C11 [gene locus: BCAL2802] and estVIII-C21 [gene locus: BCAM0511] (Fig. 8b).

Conclusions
The cell-bound lipase activity of Burkholderia sp. ZYB002 was shown to be a multi-enzyme mixture, which at least consisted of LipA and LipC24. LipA was the main component of the cell-bound lipase. LipC24 was a novel lipase, which displayed a totally different enzymatic characterization and 3D structure to that of LipA. Besides LipA and LipC24, other type of the cell-bound lipases (or esterases) may exist.

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional files.

Additional files
Additional file 1: All PCR conditions and PCR procedures used in this research. (DOC 54 kb) Additional file 2: Figure S1. The construction flow diagram for the suicide plasmid pBCMB-S3, which was used to construct the lipA-inactivation mutation strain. Figure S2. The construction flow diagram for the suicide plasmid pBCMB-S5, which was used to construct the lipC24-inactivation mutation strain. Figure S3