High-level soluble expression and enzymatic characterization of Burkholderia sp. lipase with steryl ester hydrolysis activity in E. coli

Burkholderia cepacia lipase is an important industrial biocatalyst for biodiesel production and chiral pharmaceutical synthesis. Heterologous soluble expression of lipase lipA gene from B. cepacia in Escherichia coli highly depends on co-expression of its cognate foldase gene, lipB. However, the interaction between recombinant lipase LipA and chaperonin LipB is rather complicated and confusing. In this research, various systems of lipA/lipB co-expression combinations are investigated to obtain high-level soluble expression of lipA, respectively. The best co-expression combination system for lipA and lipB is E. coli Origami 2 (DE3)/pETDuet-lipB(MCS1)/lipA(MCS2). The soluble expression level of lipA is 100.4 U/OD600 towards 4-nitrophenyl laurate hydrolysis. The recombinant LipA can be rapidly isolated from cell-free supernatant of recombinant E. coli lysate using HisTrap HP affinity chromatography column, and the lipA/LipB complex is obtained. Enzymatic characterization analysis shows that the purified LipA is a mesothermal and alkaline enzyme. LipA displays preference for medium-chain-length acyl groups (C10-C12) and sn-1,3 regioselectivity. Besides triacylglycerol hydrolase activity (EC. 3.1.1.3), LipA also displays steryl ester hydrolase activity (EC. 3.1.1.13). The specific activity of LipA towards 4-nitrophenyl decanoate and cholesterol linoleate are 638.9 U/mg and 1111.5 mU/mg, respectively. dual

Among the known microbial lipase resources, Burkholderia cepacia lipase is known as an enantioselective biocatalyst for synthesis reactions because of its' distinctive structural characterization of the funnel-like active site [4,5,6]. Moreover, B. cepacia lipase also exhibits excellent organic solvent tolerance and is a promising biocatalyst for biodiesel production [7,8]. Disulfide bond and calcium ion in the 3D molecular structure contributes to the stability of B. cepacia lipase, respectively [9,10]. According to the classification standard for lipolytic enzyme family, B.
cepacia lipase belongs to I.2 subfamily of true lipases [11,12]. The functional expression and secretion in an active form of I.2 subfamily lipases highly depends on a chaperone protein (also named as lipase-specific foldase) [13,14].
Because most B. cepacia strains are human opportunistic pathogens, it is severely regulated to construct recombinant B. cepacia strains using homologous expression system for lipase production in industry [8,[15][16][17]. To obtain high-yield functional recombinant lipase from Burkholderia sp., cell-free protein synthesis systems and foldase-assisted refolding of recombinant lipase in vitro are used to produce the functional recombinant lipase at the early stage, respectively [18,19]. In recent years, two-plasmid co-expression systems or dual expression cassette plasmid systems are used to co-express lipase gene (lipA) with its cognate foldase gene (lipB),, and a method which enabled to obtain high yields of functional recombinant lipase LipA is obtained [20][21][22][23].

Effects of different co-expression combination systems on the soluble expression of lipA
In the dual expression cassette plasmid system, gene loci of lipA and lipB on the plasmid were the key factors for the functional soluble expression of lipA.  Ihara et al. (1995) reported that it was the recombinant LipA, rather than the recombinant LipB, that was prone to form inclusion body under the control of T7 promoter [25]. Low-level expression of lipA using low-copy number plasmid (pACYCDuet or pETDuet used in this study) contributed to form functional soluble LipA. Among the investigated two-plasmid coexpression systems, plasmid combinations of pACYCDuet with pETDuet showed excellent effect on the functional soluble expression of lipA, in which lipase activity was as high as from 9.9±0.4 U/OD 600 to 12.6±1.0 U/OD 600 (Table 1, E. coli BL21 (DE3) as the expression host).
In previous report, co-expression of an additional copy of chaperone lipB gene resulted in a considerable increase of functional LipA yield in native host strain [14,26]. However, no statistical significance of functional LipA yield was observed in E. coli expression system when an additional copy of lipB was introduced into the co-expression combination system of lipA and lipB. Lipase activity was 6.5±0.2 U/OD 600 in the dual expression cassette plasmid system of pACYCDuet-B1A2.
However, lipase activity still was 6.3±0.1 U/OD 600 in the two-plasmid co-expression system of pACYCDuet-B1A2 and pETDuet-B1, in which an additional copy of lipB was introduced using plasmid pETDuet-B1 (  [10]. Refolding in vitro of LipA from B. cepacia indicated that the functional LipA was highly depended on the correct formation of a disulfide bond between Cys 190 and Cys 270 [27]. The functional soluble expression yield of lipA from Burkholderia sp. was improved by fusing LipA with the N-terminal peptide tags of Dsb-family chaperones [28]. Compared with that in E. coli BL21 (DE3), soluble expression yield of lipA in E. coli Origami 2 (DE3) was increased by 1.3-fold to 7.9-fold (  (Table 3, E. coli Origami 2 (DE3) as the expression host). Same results were obtained when other co-expression combinational systems were induced at 20°C and 30 °C, respectively. Low temperature culture contributed to not only the correct folding of recombinant protein, but also improving stability, which resulted in productivity enhancement of soluble recombinant protein [29,30].
During exponential phase, growth rates of recombinant E. coli BL21(DE3) strains were higher than that of recombinant E. coli Origami 2 (DE3) strains at 20°C (Fig.   1A). It were 15 hours for recombinant E. coli BL21(DE3) strains that were taken to reach stationary phases at 20°C. While for recombinant E. coli Origami 2 (DE3) strains, 25 hours were needed to reach stationary phases at 20°C (Fig. 1A). No ZYB002 [31]. In other reports on co-expression of lipA/lipB from B. cepacia or P.
aeruginosa in E. coli, the purified preparation also simultaneously contained the recombinant LipA and LipB [20,21]. Pauwels and Van Gelder (2008) even developed a rapid affinity-based purification method of a bacterial lipase through steric chaperone interactions [32]. Partial hydrolysis of triolein by the recombinant LipA resulted in three kinds of hydrolysis products, including 1,2-diolein, oleic acid, 1-monoolein (Fig. 5), which showed that LipA cleaved only the 1,3-positioned ester bonds and displayed sn-1,3 regioselectivity. Trace amounts of 1-monoolein in the hydrolyzed products might be related to acyl migration of sn-2 to sn-1, which occurs spontaneously in glycerides [33].
The recombinant LipA showed a preference for medium chain length fatty acid esters (C10-C12) when assayed using 4-nitrophenyl derivatives ( Table 2).  (Table 2). Kontkanen et al. (2004) also reported that some commercial lipase preparations displayed steryl esterase activity. However, the steryl esterase activity from lipase preparations was considerably lower than the native lipase activity [34].

Discussion
In native Burkholderia cepacia strain, lipA and lipB usually form an operon suggesting a 1 to 1 ratio for both lipA and lipB expression [35,36]. However, LipB acted as multi-turnover catalysts to direct lipase folding, and expression yield of lipB was lower than lipA in native host strains [14,37]. While LipA and LipB formed a stable complex, in which LipB acted as single-turnover catalysts to direct lipase folding in heterologous host, E. coli strains21,38]. Both El Khattabi et al. (2000) and Quyen et al. (1999) reported that an excess of LipB was the prerequisite for correct LipA folding in E. coli host [19,38]. Moreover, high-level active LipA could be obtained only when LipB was synthesized first in two-plasmid co-expression systems in E. coli host [24]. The similar result was also observed in this research. High level soluble LipA was obtained only when lipB and lipA was inserted into the MCS1 site and MCS2 site on the dual expression cassette plasmid pETduet, respectively.
DsbA-DsbB disulfide bond formation system existed in native B. cepacia strains [39]. Disulfide bond in Burkholderia sp. lipase played a key role in activating lipolytic activity and stabilizing the 3D structure [9,40]. In heterologous strain, the soluble expression yield of Burkholderia sp. lipase was improved by co-expression of cytoplasmic chaperone GroEL/ES, which could facilitate refolding of disulfide-bond [28,41]. Host strain was another substitution strategy for high-level soluble expression of lipase with disulfide-bond in 3D structure. To produce recombinant When E. coli was selected as host strain for co-expression of lipA and lipB from Burkholderia sp., LipA/LipB complex would be formed and it was difficult to isolate LipA from LipA/LipB complex [43]. In previous research work, a low-Mr compound was purified from cell-free lysate of Pseudomonas sp. and then identified as Glutathione, which could facilitate the dissociation of LipA/LipB complex and liberate free active lipase [44,45]. In this work, E. coli Origami 2 (DE3) was selected as the expression host, which would enhance disulfide bond formation. However, cepacia ST200 was firstly reported by Takeda et al. (2006) and enzymatic characterization for steryl esterase activity was determined. However, lipase activity of this enzyme was not tested [49]. The amino acid sequence identity was over 93% between steryl esterase from B. cepacia ST200 and LipA from Burkholderia sp. ZYB002 [49,50]. In this research, both lipase activity and steryl esterase activity were determined from LipA from Burkholderia sp. ZYB002 (Table 2). Moreover, microbial lipase could act on some non-triglyceride substrates and displayed promiscuous activity [51,52]. In previous reports, a few microbial lipases could catalyze hydrolysis reaction of steryl esters [34]. Because of the structural difference between triglyceride and steryl ester, hydrolysis efficiency of steryl ester catalyzed by lipase was always low, and it was necessary to improve catalysis activity towards steryl esters using protein engineering technology.

Strains, plasmids and reagents
Bacterial strains and plasmids used in this study were listed in Table 3. Briefly, E.

Construction of recombinant plasmids
To discriminate lipase and its cognate lipase-specific foldase, lipase and foldase from Burkholderia sp. ZYB002 were named as LipA and LipB, respectively. The corresponding gene symbols of LipA and LipB were designated as lipA and lipB, respectively. Recombinant plasmids used in this research were listed in Table 3.
Except for the above mentioned recombinant expression plasmids, other recombinant expression plasmids listed in Table 3 were constructed as follows. The DNA fragments of lipA and lipB were amplified by PCR using plasmid pMD18T-lipAB as the template [50]. The oligonucleotide sequences of PCR primers, PCR primer pairs, annealing temperatures, and PCR products were listed in Table 4. The resulting PCR products were digested with the restriction endonucleases (shown in the oligonucleotide sequences of PCR primer in Table 4) and then ligated into the corresponding endonuclease-digested expression vector pACYCDuet, pETDuet, and pET28a, respectively.
All of the resulting recombinant plasmids were transformed into E. coli DH5α, and the reading frames were confirmed by DNA sequencing.

Co-expression of lipA with lipB in E. coli
To obtain high-yield soluble expression of lipA, various systems of lipA/lipB coexpression combinations were screened, including two-plasmid co-expression systems and dual expression cassette plasmid systems. Total eleven co-expression combination systems of lipA with lipB were investigated ( Table 1) The purity of the active fractions was monitored by SDS-PAGE on a 10% separating gel [53]. The fractions with pure LipA was pooled and dialyzed against 20 mmol/L Na 2 HPO 4 -NaH 2 PO 4 (pH7.4) buffer overnight at 4°C. The protein concentration was analyzed using the method of Bradford, with bovine serum albumin as the standard [54].

Lipase activity determination of recombinant LipA
Lipase activity was determined using spectrophotometric assay method [55] with slight modifications. The reaction mixture consisted of 0.4 mmol/L of each pnitrophenol ester in 20 mmol/L Na 2 HPO 4 -NaH 2 PO 4 (pH8.0) buffer and 30 μL the appropriately diluted lipase solution. The kinetics was detected for 5 min at 410 nm.
All reactions were carried out at 40 °C and 20 mmol/L Na 2 HPO 4 -NaH 2 PO 4 (pH8.0) buffer. One unit of lipase activity was defined as the amount of lipase that liberated 1 μmol of p-nitrophenol from p-nitrophenol esters per min. All measurements were carried out three times and the average value was taken.

Cholesterol ester hydrolase activity assay of LipA
Cholesterol ester hydrolase activity of LipA was assayed using spectrophotometric method as described by Stępień and Gonchar (2013) [56], and the instruction Effect of pH on activity and stability The optimal pH for lipase activity was determined by incubating lipase substrates in a suitable buffer at various pH ranging from 6.5 to 9.0, and the maximum lipase activity was considered 100%. To determine the effect of pH on lipase stability at pH ranging from 6.5 to 10.0, aliquots of the concentrated lipase preparation were diluted five-fold in the corresponding buffer and then incubated for 4 h at 4 °C. The residual lipase activity after incubation was determined and lipase activity at the start was taken as 100%. The corresponding buffers were Na 2 HPO 4 -NaH 2 PO 4 (pH6.5-

Effect of temperature on lipase activity and stability
The optimal temperature for lipase activity was determined by incubating the standard reaction mixture at different temperatures ranging from 25°C to 55°C, and the maximum lipase activity was considered 100%. To determine the effect of temperature on lipase stability, the lipase preparation was incubated at 40°C and aliquots were continuously taken at 5-min interval to assay the residual activity.
The lipase activity at the start was taken as 100%. Half-life of thermal inactivation was calculated using the method as described by Zhao and Arnold

Statistical analysis
All experiments were carried out three times independently. Data are presented as the average ± standard deviation. The data were statistically analyzed using SPSS software and groups were compared using Student's t-test with significant differences defined as P<0.05, whereas P<0.01 represented a highly significant difference.

Declarations
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: Figure S1. Restriction maps of the recombinant plasmids containing lipA and (or) lipB in this study.

Competing interests
The authors declare that they have no competing interests.
Authors' contributions ZY Table 3 Strains and plasmids used in the current study