A novel phospholipase B from Streptomyces sp. NA684 – purification, characterization, gene cloning, extracellular production and prediction of the catalytic residues

Authors

  • Yusaku Matsumoto,

    1. Department of Symbiotic Systems Science and Technology, Graduate School of Symbiotic Systems Science and Technology, Fukushima University, Japan
    Search for more papers by this author
  • Shingo Mineta,

    1. Department of Symbiotic Systems Science and Technology, Graduate School of Symbiotic Systems Science and Technology, Fukushima University, Japan
    Search for more papers by this author
  • Kazutaka Murayama,

    1. Division of Biomedical Measurements and Diagnostics, Graduate School of Biomedical Engineering, Tohoku University, Sendai, Japan
    Search for more papers by this author
  • Daisuke Sugimori

    Corresponding author
    1. Department of Symbiotic Systems Science and Technology, Graduate School of Symbiotic Systems Science and Technology, Fukushima University, Japan
    • Correspondence

      D. Sugimori, Department of Symbiotic Systems Science and Technology, Graduate School of Symbiotic Systems Science and Technology, Fukushima University, 1 Kanayagawa, Fukushima 960-1296, Japan

      Fax: +81 24 548 8206

      Tel.: +81 24 548 8206

      E-mail: sugimori@sss.fukushima-u.ac.jp

    Search for more papers by this author

Abstract

A novel metal ion-independent phospholipase B (PLB684) from Streptomyces sp. strain NA684 was purified 264-fold from the culture supernatant with 2.85% recovery (6330 U·mg protein−1). The enzyme functions as a monomer with a molecular mass of 38.9 kDa. Maximum activity was found at pH 8.4 and 50 °C. The substrate specificity was in the order: phosphatidylcholine ≥ phosphatidic acid ≥ lysophosphatidylcholine > phosphatidylserine > phosphatidylinositol > phosphatidylglycerol. The enzyme did not hydrolyze phosphatidylethanolamine, tristearin and dipalmitin. PLB684 hydrolyzed lysophosphatidylcholine and diacylphosphatidylcholine, and lysophosphatidylcholine was primarily produced during the early stages of phosphatidylcholine hydrolysis. The apparent Km, Vmax and kcat for hydrolysis of dimyristoyl phosphatidic acid were 14.5 mm, 15.8 mmol·min−1·mg protein−1 and 1.02 × 104 s−1, respectively. The positional specificity of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine hydrolysis was investigated using GC. In the reaction equilibrium, the molar ratio of released fatty acids (sn-1 : sn-2) was 45 : 55. The ORF of the gene is 1239 bp in length and codes for a 30-amino acid signal peptide and a 382-amino acid mature enzyme. The deduced amino acid sequence of PLB684 shows 60% identity to a uncharacterized protein of Streptomyces auratus AGR0001 (UniProt accession number: J1RQY0). The extracellular production of PLB684 was achieved using a pUC702 expression vector and Streptomyces lividans as the host. Mutagenesis analysis showed that Ser12 is essential for the catalytic function of PLB684 and that the active site may include residues Ser330 and His332.

Abbreviations
AP

alkaline phosphatase

DLS

dynamic light scattering

DMPA

1,2-dimyristoyl-sn-glycero-3-phosphate, monosodium salt

DP

dipalmitin

DPPC

1,2-dipalmitoyl-sn-glycero-3-phosphocholine

DTT

dithiothreitol

FFA

free fatty acid

GPC

glycerol-3-phosphocholine

GPCP

glycerol-3-phosphocholine phosphodiesterase

LPC

1-palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine

PcPLB

PLB from P. chrysogenum

PI

l-α-phosphatidylinositol

PLA1

phospholipase A1

PLA2

phospholipase A2

PLB

phospholipase B

POPA

1-palmitoyl-2-oleoyl-sn-glycero-3-phosphate, monosodium salt

POPC

1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine

POPE

1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine

POPG

1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol), monosodium salt

PS

l-α-phosphatidyl-l-serine

SaPLA1

PLA1 from Streptomyces albidoflavus

ScPLA2

PLA2 from Streptomyces coelicolor

SMC

sphingomyelinase C

SsEst

esterase from Streptomyces scabies

TSB

tryptic soy broth

TS

tristearin

Introduction

Phospholipases play a central role in cellular processes such as signal transduction and inflammation through their effects on the metabolism of phospholipids and lysophospholipids. Microbial phospholipases further contribute to pathogenesis and virulence through the release or breakdown of bioactive compounds that affect host cell function [1-4]. Substrates for phospholipases are either phospholipids or lysophospholipids, comprising a polar head group (e.g. ethanolamine, choline, inositol or serine esterified to phosphoric acid) and one or two nonpolar fatty acyl chains esterified to a glycerol backbone. The substrate specificity of phospholipases is determined by the phospho-head groups, the acyl chain length and saturation of the fatty acyl side-chains. Some phospholipases have broader substrate specificities than others. Phospholipase A1 (PLA1) (EC 3.1.1.32) and phospholipase A2 (PLA2) (EC 3.1.1.4) hydrolyze the ester bonds at the sn-1 and sn-2 positions of the glycerol moiety, respectively, yielding free fatty acids (FFAs) and 2-acyl or 1-acyl lysophospholipids. The two phosphodiester bonds found in the polar head group of an amphipathic phospholipid are cleaved by phospholipase C (first bond), which releases the phospho-head group, and phospholipase D (second bond), which releases only the head group.

Phospholipase B (PLB) (EC 3.1.1.5) is an enzyme that catalyzes hydrolytic cleavage of both the sn-1 and sn-2 acyl ester bonds of glycerophospholipids, forming FFAs and lysophospholipids or glycerol-3-phosphodiesters. PLBs from fungi have been reported to prefer lysophospholipids to diacylphospholipids as substrates and to hydrolyze diacylphospholipids without forming lysophospholipids as intermediates during hydrolysis. Additionally, lysophospholipase–transacylase activity is associated with some PLB enzymes, allowing these enzymes to transfer a FFA to a lysophospholipid and produce a diacylphospholipid. Hydrolase and acyltransferase activities have been detected in several fungi, including Saccharomyces cerevisiae [5], Candida albicans [6], Candida utilis [7], Penicillium chrysogenum [8] and Cryptococcus neoformans [9]. PLBs from Escherichia coli [10] and Bacillus subtilis [11] are known; however, there has been no investigation of bacterial PLBs with respect to whether they produce lysophospholipids during diacylglycerophospholipid hydrolysis and possess lysophospholipase–transacylase activity. Moreover, the substrate recognitions and the catalytic mechanisms of bacterial PLBs have not been revealed. Therefore, we have embarked on a study of bacterial PLBs. We found that a novel PLB was produced by Streptomyces sp.

In the present study, we report the purification, characterization and gene cloning of PLB684 from Streptomyces sp. NA684. We also demonstrate the kinetic parameters, positional specific hydrolysis and the hydrolytic intermediates formed. Moreover, the recombinant PLB684 was produced extracellularly using an expression vector pUC702 and Streptomyces lividans as the host, and a predictive active site is discussed on the basis of a mutagenesis analysis.

Results

Identification of strain NA684

Strain NA684, isolated from a soil sample of Fukushima, Japan, was identified as Streptomyces sp., a near relative of Streptomyces chattanoogensis NBRC12754, by morphological, physiological and biochemical characterization, as well as 16S rDNA sequence analysis (data not shown). Streptomyces sp. strain NA684 was deposited as NITE BP-1015 in the NITE Patent Microorganisms Depositary (NPMD, Chiba, Japan). The 16S rDNA sequence of strain NA684 was deposited in the DDBJ database under accession number AB738936.

Purification of PLB684

PLB684 was induced by the addition of 1% (w/v) lecithin to the culture medium. PLB684 adsorbed to poly(vinylidene difluoride) and polyethersulfone membrane filters (25 mm GS/X syringe filter, 0.45 μm; GE Healthcare UK Ltd, Little Chalfont, Buckinghamshire, UK) and a TPX sample tube (Hi-tech Inc., Tokyo, Japan), suggesting that the enzyme is highly hydrophobic. To prevent the adsorption onto containers such as sample tubes, 0.5% (w/v) Triton X-100 was added to the enzyme sample solution after Resource Phe column chromatography (GE Healthcare UK Ltd.). An extracellular PLB684 from Streptomyces sp. NA684 was purified to electrophoretic homogeneity from the culture supernatant (Fig. 1). The purification steps of PLB684 are summarized in Table 1. Purified PLB684 with a specific activity of 6330 U·mg protein−1 was obtained and the amount of total pure protein was 15.4 μg from 250 mL of the culture supernatant.

Table 1. Purification of PLB from Streptomyces sp. NA684. PLB activity was assayed using the reaction mixture containing 50 mm Tris-HCl (pH 8.4), 0.5% (w/v) DMPA, 0.5% (w/v) Triton X-100 and 10 mm EDTA at 45 °C
Purification stepActivity (U·mL−1)Total activity (U)Protein (mg·mL−1)Total protein (mg)Specific activity (U·mg−1)Yield (%)Fold
108-h culture supernatant13.734310.57314324.01001.00
80% ammonium sulfate33.533460.82882.840.497.51.68
DEAE-650M16.310620.75649.121.630.90.90
HiTrap Q HP47.85350.5656.3384.515.63.52
Resource Phe67.83050.01260.056753688.90224
Mono S97.697.60.01540.015463302.85264
Figure 1.

SDS/PAGE analysis of purified PLB from Streptomyces sp. NA684. Lane M, molecular marker; lane 1, purified PLB (3 μg).

Protein analysis

SDS/PAGE analysis showed that the purified enzyme appeared as a single band of 38.9 kDa (Fig. 1). Gel filtration, native PAGE and dynamic light scattering (DLS) analyses suggested that PLB684 behaves as a monomeric protein (data not shown). The N-terminal amino acid sequence was determined as GKPTAVVSLGDS. The determined internal amino acid sequences were FLASPVGQYGCWK, AGDTLTHPLPSR and GEWPQADQLEDLAR.

Effect of pH, temperature and chemicals on PLB684 activity

As shown in Fig. 2A,B, the highest enzyme activity for 1,2-dimyristoyl-sn-glycero-3-phosphate, monosodium salt (DMPA) hydrolysis was found at 50 °C and pH 8.4. In addition, the enzyme hydrolyzed 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol), monosodium salt (POPG) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) with maximal activity at pH 8.4 and pH 8.4–8.8, respectively (Figs S1 and S2). The activation energy (Ea) was 41.1 kJ·mol−1 (Fig. 2C). Enzyme activity was maintained between pH 4.0 and 10.5 at 4 °C for 3 h (Fig. 3A). The enzyme was stable between 4 and 45 °C for 30 min at pH 8.4 (Fig. 3B). The substrate specificity towards various head-groups was in the order: 1,2-dipalmitoyl-sn-glycero-3- phosphocholine (DPPC) ≥ DMPA ≥ POPC ≥ 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine (LPC) > 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphate, monosodium salt (POPA) > l-α-phosphatidyl-l-serine (PS) > l-α-phosphatidylinositol (PI) > POPG (Table 2). By contrast, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), tristearin (TS) and dipalmitin (DP) were not hydrolyzed under the experimental conditions employed. In addition, PLB684 scarcely hydrolyzed liposomal POPC during the 2-h enzyme reaction (Fig. S3). PLB684 was active in the presence of 10 mm EDTA and all tested metal ions, except for Mg2+ and Fe2+ (Table 3). In the presence of Ca2+, PLB activity was inhibited. Enzyme activity was also inhibited by dithiothreitol (DTT) but not by mercaptoethanol and iodoacetamide (Table 3). As shown in Fig. 4, the enzyme exhibited high activity in the presence of 0.25–1.05% (w/v) Triton X-100.

Table 2. Substrate specificity of PLB. PLB activity was assayed using the reaction mixture containing 50 mm Tris-HCl (pH 8.4), 0.5% (w/v) of each substrate, 0.5% (w/v) Triton X-100 and 10 mm EDTA at 50 °C. Relative activity was determined by defining the activity for DMPA as 100%. Data represent the mean ± SD of experiments performed in triplicate
SubstrateRelative activity (%)
1,2-Dimyristoyl-sn-glycero-3-phosphate (DMPA)100 ± 2
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphate (POPA)73.3 ± 1.1
1,2-Dipalmityl-sn-glycero-3-phos-phocholline (DPPC)104 ± 1
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC)98.7 ± 0.5
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE)0
1-Palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (POPG)22.5 ± 2.2
L-α-Phosphatidylinositol (PI)26.9 ± 3.5
L-α-Phosphatidyl-l-serine (PS)30.2 ± 4.2
1-Palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine (LPC)98.5 ± 1.1
Tristearin0
Dipalmitin0
Table 3. Effect of various chemicals on PLB activity for DMPA hydrolysis. PLB activity was assayed under standard assay conditions: 50 mm Tris-HCl (pH 8.4), 0.5% (w/v) DMPA, 0.55% (w/v) Triton X-100 and 10 mm EDTA. The enzyme was pre-incubated in the standard assay condition without 0.5% (w/v) DMPA in the presence of each chemical at 50 °C for 5 min, and then assayed by incubation at 50 °C for 5 min. The relative activity was determined by defining the activity in the presence of 10 mm EDTA and 0.5% (w/v) Triton X-100 as 100%. Data represent the mean ± SD of experiments performed in triplicate
ChemicalsRelative activity (%)
EDTA free127 ± 2
10 mm EDTA100 ± 2
10 mm CaCl287.3 ± 3.1
10 mm MgCl20
10 mm MnCl21.53 ± 1.92
10 mm CoCl23.07 ± 1.62
10 mm ZnCl27.37 ± 1.66
10 mm FeCl20
10 mm FeCl328.3 ± 1.9
2 mm Iodoacetoamide95.8 ± 2.8
2 mm 2-Mercaptoethanol96.8 ± 5.5
2 mm Dithiothreitol57.5 ± 1.9
2 mm Phenylmethanesulfonyl fluoride101 ± 6
2 mm SDS97.7 ± 3.8
Figure 2.

Effect of pH and temperature on PLB activity. (A) The enzyme activity towards DMPA was assayed at 37 °C for 5 min with 0.5% (w/v) DMPA in 50 mm of each buffer containing 10 mm EDTA and 0.5% (w/v) Triton X-100. The buffers were: sodium acetate (pH 4.1–5.6; ●), bisTris-HCl (pH 5.6–7.2; ■), Tris-HCl (pH 7.2–8.8; ▲) and glycine-NaOH (pH 8.8–10.5; ○). (B) The enzyme activity for DMPA was assayed at each temperature in 50 mm Tris-HCl (pH 8.4). Data represent the mean ± SD of the experiments performed in triplicate. (C) Arrhenius plot of DMPA hydrolysis.

Figure 3.

pH and thermal stability of PLB. (A) The enzyme was incubated at 4 °C for 3 h in 40 mm of each buffer: sodium acetate (pH 4.1–5.6; ●), bisTris-HCl (pH 5.6–7.2; ■), Tris-HCl (pH 7.2–8.8; ▲) and glycine-NaOH (pH 8.8–10.5; ○). The residual activity for DMPA was assayed at 50 °C for 5 min with 0.5% (w/v) DMPA in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100. (B) The enzyme was incubated at each temperature for 30 min in 50 mm Tris-HCl (pH 8.4). The residual activity for DMPA was assayed at 50 °C for 5 min with 0.5% (w/v) DMPA in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100. Data represent the mean ± SD of experiments performed in triplicate.

Figure 4.

Effect of Triton X-100 concentration in the reaction mixture on enzyme activity. PLB activity was assayed by incubation with 0.5% (w/v) DMPA at 50 °C in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and each percentage (w/v) of Triton X-100.

Steady-state kinetics

Good linear regression analysis was achieved by the Hanes–Woolf plot (Fig. 5). On hydrolysis of DMPA by the purified wild-type enzyme at 50 °C and pH 8.4, the apparent Km, Vmax and kcat values for DMPA were found to be 14.5 mm, 15.8 mmol·min−1·mg protein−1 and 1.02 × 104 s−1, respectively.

Figure 5.

Hanes–Woolf plot of the steady-state kinetics of wild-type PLB activity. The initial rate of DMPA hydrolysis by the purified wild-type PLB was determined at various DMPA concentrations and then plotted in a Hanes–Woolf plot ([DMPA]/v versus [DMPA]). The initial rate of the enzyme reaction was assayed by incubation at 50 °C for 5 min with various DMPA concentrations in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100.

Catalytic properties and positional specificity of PLB684

As shown in Fig. 6A, FFAs released by the purified enzyme were found to increase in a linear fashion as the time of the reaction increased; however, glycerol-3-phosphocholine (GPC) was not detected during the early stages of POPC hydrolysis. The concentration of LPC in the reaction mixture was calculated by [LPC] = [FFA] − [GPC]. As shown in Fig. 6B, LPC was produced by enzymatic hydrolysis of POPC during the early stages of the reaction. Moreover, GC analysis showed that the enzyme releases both FFAs during POPC hydrolysis; palmitic acid from sn-1 and oleic acid from sn-2. In the equilibrium mixture of the enzyme reaction, the molar ratio of the released FFAs comprised an sn-1 : sn-2 ratio of 45 : 55 (Fig. 6C). In addition, PLA1 and PLA2 activity was detected using the EnzCheck® Phospholipase A1 assay kit and the Phospholipase A2 assay kit (Life Technologies Corp., Carlsbad, CA, USA) (data not shown). These results demonstrate that the enzyme is PLB, which catalyzes the hydrolysis of both acyl ester bonds of sn-1 and sn-2 in diacylglycerophospholipids.

Figure 6.

Time course of the reaction products on POPC hydrolysis. The enzyme reaction was carried out by incubation at 37 °C with 0.5% (w/v) POPC in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100. (A) The concentration of the produced FFAs (●), LPC (□) and GPC (▲). (B) Molar ratio of produced GPC (▲) and LPC (□). (C) Positional selectivity of POPC hydrolysis by PLB. ●, palmitic acid (sn-1); ▲, oleic acid (sn-2).

Nucleotide sequence of the gene for PLB684

The nucleotide sequence of the gene encoding PLB684 (plb) was determined from the sequence of the 1.39-kbp PCR product obtained by the genomic PCR. The ORF of plb consisted of 1239 bp encoding a protein of 412 amino acid residues. The N-terminal sequence of the active form of the enzyme starts at Gly31 of the deduced amino acid sequence, indicating that the preceding 30 amino acid residues represent a signal peptide sequence required for secretion (Fig. 7). Homology searches performed with the blast algorithm indicated that the amino acid sequence of the mature PLB684 shows 60% identity to that of SGNH esterase of Streptomyces auratus AGR0001 (UniProt accession number: J1RQY0).

Figure 7.

Nucleotide and deduced amino acid sequences of PLB from Streptomyces sp. NA684. Single underlines represent the N-terminal amino acid sequence (N-1) and the internal amino acid sequences (I-1 to -3) determined by the protein sequencer and nano LC-MS/MS analysis, respectively. The inverse PCR primers were designed based on the amino acid sequence region indicated by the double underline. The deduced ribosome binding domain and stem-loop domain are represented by rbs and the dotted line, respectively.

Expression, purification and characterization of PLB684

High-efficiency extracellular production of PLB684 has been successfully achieved in S. lividans cells transformed with the expression vector pUC702/plb. The specific activity in the culture supernatant (219 U·mg protein−1) was approximately nine-fold higher than that (24.0 U·mg protein−1) of the wild-type strain. The recombinant PLB684 with high specific activity (4344 U·mg protein−1) was purified to electrophoretic homogeneity from 275 mL of the culture supernatant by simple purification steps (Table 4). The recombinant PLB684 exhibited low activities toward 4-nitrophenyl esters of fatty acids (Table 5). The substrate specificity toward 4-nitrophenylesters was in the order: C8 > C10 > C12 ≥ C14 > C16 > C18. PLB684 could not hydrolyze 4-nitrophenyl butyrate. Therefore, PLB684 prefers glycerophospholipids such as DMPA to 4-nitrophenylesters. These results prove that the enzyme is a novel PLB, although not a lipase as well as a carboxylesterase.

Table 4. Purification of the recombinant PLB produced by S. lividans. PLB activity was assayed using the reaction mixture containing 50 mm Tris-HCl (pH 8.4), 0.5% (w/v) DMPA, 0.5% (w/v) Triton X-100 and 10 mm EDTA at 50 °C
Purification stepPLB (U·mL−1)Protein (mg·mL−1)Total protein (mg)Specific activity (U·mg−1)Total activity (U)Yield (%)Fold
36-h culture supernatant1150.52514421931 6251001.00
50% ammonium sulfate11421.4428.779522 84072.13.63
Phenyl-650M1470.06606.60222714 70046.510.2
HiTrap Q HP2650.06101.224344530016.819.8
Table 5. Substrate specificity of the recombinant PLB. PLB activity was assayed using the reaction mixture containing 50 mm Tris-HCl (pH 8.4), 0.05% (w/v) of each substrate, 0.5% (w/v) Triton X-100 and 10 mm EDTA at 50 °C. Relative activity was determined by defining the activity for DMPA as 100%. Data represent the mean ± SD of experiments performed in triplicate
SubstrateRelative activity (%)
DMPA100 ± 13
4-Nitrophenyl butyrate0
4-Nitrophenyl octanoate36.3 ± 3.1
4-Nitrophenyl decanoate25.4 ± 1.7
4-Nitrophenyl laurate12.9 ± 1.8
4-Nitrophenyl myristate12.0 ± 1.8
4-Nitrophenyl palmitate3.37 ± 1.31
4-Nitrophenyl stearate0.684 ± 0.640

Homology modelling and mutant analysis

The homology model of PLB684 based on esterase from Streptomyces scabies (SsEst) (1ESC) is shown in Fig. 8. verify3d (http://nihserver.mbi.ucla.edu/Verify_3D/) was used to assess the reliability of the predicted model and showed that 71.5% of the amino acid residues had an average 3D–1D score of more than 0.2. The putative active site involves a Ser12-His332 dyad and the carbonyl group of Ser330. The mutant Ser12A exhibited no activity (Table 6). The mutants Ser330A and His332A exhibited 67% and 17.2% relative activity, respectively, compared to the activity of the wild-type enzyme.

Table 6. Enzyme activity of the wild-type and the recombinant enzymes. The activity was determined in the reaction mixture consisting of 50 mm Tris-HCl (pH 8.4), 0.5% (w/v) DMPA, 0.5% (w/v) Triton X-100 and 10 mm EDTA at 50 °C. Relative activity was determined by defining the activity of wild-type enzyme (specific activity, 2480 U·mg protein−1) as 100%
EnzymeRelative activity (%)
Wild-type100
Ser12A0
Ser330A67.0
His332A17.2
Figure 8.

Catalytic residues, His73 and a cleft of the PLB predicted by homology modelling. His73 may stabilize the phosphoric group of the bound phospholipid.

Discussion

This is the first report of a PLB from an actinomycete, Streptomyces. Known bacterial PLBs include the Escherichia coli TAP protein (20.5 kDa) and B. subtilis LipC (24.7 kDa), and both proteins function as monomers [11]. TAP and LipC are localized intracellularly and on the spore coat, respectively. The subunit composition of PLB684 from Streptomyces sp. NA684 is the same as these bacterial PLBs. However, the molecular mass and localization of PLB684 differs from these bacterial PLBs, suggesting the possibility of structural and physiological differences between PLB enzymes of actinomycete and other bacteria. Glycosylated PLB from culture supernatant of C. neoformans has been reported to have a specific activity of 559 U·mg protein−1 [9]. PLB684 was isolated to high purity, and a high specific activity (6330 U·mg protein−1) was achieved by employing efficient purification steps. Thus, PLB684 has much higher specific activity than that of C. neoformans. In addition, PLB684 is a metal ion-independent enzyme from a nonpathogenic bacterium. These properties may be advantages for commercial reagent and industrial applications, such as the production of lysophospholipids, natural good emulsifiers [12] and glycero-3-phosphodiesters. It has been reported that GPC has physiological functions, such as enhancing the secretion of growth hormone and the improvement of the cognitive symptom of Alzheimer's disease [13, 14]. Furthermore, we have successfully achieved the efficient extracellular production of the enzyme using S. lividans cells.

The optimum pH of PLB684 is 8.4 and resembles the optimal pH activity of TAP [10] and LipC [11] but differs from fungal PLBs [6, 7, 9, 15-17], which function optimally under acidic pH conditions (pH 2.5–6). The observed optimum temperature of PLB684 activity (approximately 50 °C) was higher than that of lysophospholipase of Mycobacterium leprae (approximately 37 °C) [18]. The activation energy (Ea) of PLA2 from cobra venom was reported as 29.7 kJ·mol−1 for diheptanoyl-PC micelles [19], whereas the Ea values of PLA1 (SaPLA1) from Streptomyces albidoflavus for egg yolk PC at pH 5.6 and 7.2 are 18.8 and 58.3 kJ·mol−1, respectively [20]. There are no reported Ea values of PLBs from other organisms. The Ea value for DMPA hydrolysis by PLB684 is 41.1 kJ·mol−1, suggesting the reaction velocity is highly susceptible to temperature. The high Ea value of PLB684 with high specific activity indicates that the turnover rate (k+2) may be faster than the formation rate (k+1) of the enzyme–substrate complex (ES). Therefore, raising the reaction temperature increases the ratio of the ES concentration to free enzyme concentration in the steady-state, and this probably accelerates the apparent reaction velocity. PLB684 is stable between pH 4.1 and 10.5 for 3 h at 4 °C, and pH 8.4 for 12 months at 4 °C; however, incubation at pH 8.4 and 50 °C for 30 min decreases the relative activity to 80%. Jovel et al. [21] reported that the second PLA2 (ScPLA2) from Streptomyces coelicolor was stable at pH 8.0 and 50 °C for 10 min in the presence of 50 mm CaCl2, indicating that Ca2+ is necessary to stabilize the structure of ScPLA2 [21]. PLB684 is stable at 50 °C in the absence of metal ions, suggesting that the stability of the enzyme structure is independent of metal ions.

PLB684 prefers diacylphosphatidylcholine and lysophosphatidylcholine as substrates; however, fungal PLBs prefer lysophosphatidylcholine to diacylphosphatidylcholine [5, 7, 9, 17, 22]. TAP prefers p-nitrophenyl esters to lysophosphatildylcholine, and LipC shows no selectivity for the polar head groups of phospholipids. By contrast, PLB684 prefers phospholipids to p-nitrophenyl esters. In addition, PLB684 exhibits specificity towards phospholipids with polar head groups. Based on these observations, we suggest that PLB684 contains structures for recognizing phospholipids. Mansfeld et al. [23] reported that PLA2-α from Arabidopsis thaliana showed specificity for the acyl chain length. Correspondingly, the acyl chain length specificity with 4-nitrophenyl esters of PLB684 is the highest toward C8 and decreased when the esters acyl chain length is longer. Consequently, PLB684 preferred DMPA (14 : 0) to POPA (16 : 0, 18 : 1). Further studies are needed to determine the mechanism of acyl length specificity of PLB684. Substrate specificity has been reported to be affected by the pH of the reaction. This pH influence probably results from the ionization state of amino acid residues located in the active site of the enzyme and the ionization state of the head groups of the substrate [20]. The substrate pKa is in the order: PG (2.9) [24] < PS (3.6) [25] < PA (8.5) [26] < PE (9.6) [25]. The substrate specificity of PLB684 appeared to be related to substrate pKa because the optimum pH towards DMPA is 8.4 and PE was not hydrolyzed by PLB684. Therefore, we concluded that the ionization state of the amino acid residues in the active site of the enzyme is most likely to be important in substrate specificity. It was reported that PLB from C. neoformans preferred PC and PG head groups, followed by PE > PS > PI, and showed weak or no hydrolysis of PA. By contrast, PLB684 preferred PC and PA head groups, followed by PS > PI > PG, and showed no activity toward PE. Further studies are needed to reveal the mechanism of head group specificity of PLB684.

Doi and Nojima [27] reported that a lysophospholipase of E. coli was strongly inhibited by 10 mm Fe2+, Fe3+ and Al3+ [27]. Masayama et al. [11] reported that LipC was inhibited by 1 mm Ca2+, Mn2+ and Hg2+. PLB684 was strongly inhibited by 10 mm Mg2+, Mn2+, Co2+, Zn2+, Fe2+ and Fe3+. The Ca2+ ion slightly inhibited the enzyme. It is well known that divalent cations such as Ca2+ and Zn2+ form complexes with phospholipids affecting the hydration of the phosphate group [28-30]. Thus, the cations would mask the phospholipid negative charge to be recognized by PLB684. PLB684 exhibited activity in the presence of 10 mm EDTA, confirming that the enzyme is a metal ion-independent PLB. PLB684 was inhibited by 2 mm DTT with 57.7% residual activity. DTT has no effect on the hydrolysis activity of the lysophospholipase of E. coli [27]. LipC shows 78% residual activity in the presence of 5 mm DTT [11]. In general, secretory Streptomyces phospholipases such as PLA2 [31] and phospholipase D [32] have several cysteine residues involved in disulfide bonds. SaPLA1 [33] and sphingomyelinase C (SgSMC, PDB: 3WCX) from Streptomyces griseocarneus also have disulfide bonds in the structure. PLB684 has 10 cysteine residues that potentially form disulfide bonds. It is known that fungal PLBs, C. neoformans [9], Cryptococcus gattii [17] and S. cerevisiae [33], are inhibited by Triton X-100. By contrast, the enzyme activity of Streptomyces phospholipases is stimulated by Triton X-100 [20, 34-39]. PLB684 is also stimulated by Triton X-100. The substrate specificities of phospholipase Cs toward liposomal substrates have been reported previously [40, 41]. However, PLB684 scarcely hydrolyzed liposomal substrate of POPC and POPG (Figs S3 and S4). Therefore, we considered that the surface binding model is adaptable for PLB684 [42]. Enzyme activity is likely affected by the form and concentration of the mixed micelle composed of the substrate and detergent. These enzyme properties indicate that PLB684 hardly interacts with lipid bilayer of membrane. We consider that the physiological role of PLB684 might be as a secondary metabolism enzyme acting with biosurfactant, and Streptomyces sp. NA684 probably produces PLB684 to utilize glycerophospholipids as energy and phosphate sources from the biomass in the environment.

The kcat value (1.02 × 104 s−1) of PLB684 is the fastest velocity of known PLBs, as shown in the BRENDA database (http://www.brenda-enzymes.org/). We have previously reported that the kcat values of SgSMC and SaPLA1 are 346 s−1 [43] and 630 s−1 [20], respectively. The kcat value of PLB684 is higher than the kcat values of these enzymes. However, the Km (14.5 mm) value of PLB684 is higher than those of fungal PLBs (0.05–2.9 mm) [6, 9, 15, 33, 44], SMC (0.458 mm) [43] and SaPLA1 (2.38 mm) [20]. PLB684 has a much higher turnover rate and somewhat low affinity toward phospholipid, indicating that the formation of the ES may be the rate limiting step. The kcat/Km value (703 mm−1·s−1) of PLB684 is higher than that of SaPLA1 (265 mm−1·s−1) [20] and is comparable to that of SMC (756 mm−1·s−1) [43].

The catalytic properties of PLB684 were investigated. The results of GC analysis revealed that the positional specificity of the diacylglycerophospholipid hydrolytic reaction of PLB684 was 45 : 55 (sn-1 : sn-2), indicating that the hydrolysis rate of the sn-2 acyl ester bond was somewhat faster than that of the sn-1 acyl ester bond. In addition, the PLA1 and PLA2 activities of PLB684 were examined using the EnzCheck® Phospholipase A1 and Phospholipase A2 assay kits. PLB684 also hydrolyzes LPC as well as diacylglycerophospholipid, suggesting that the transesterificated acyl group may be hydrolyzed. It was reported that acyl migration occurs from the 2-position to the 1(3)-position or the opposite of diacylglycerol [45]. However, no acyl migration would take place in the short reaction time as a result of the low acyl migration rate. We concluded that the enzyme is PLB, which catalyzes the hydrolysis of both acyl groups, sn-1 and sn-2, in diacylphospholipids. Saito and Kate [22] proposed that PLB from P. chrysogenum (PcPLB) may involve an active center with one catalytic site and two substrate-binding sites: site I for diacylphospholipid and site II for lysophospholipid [22]. When diacylphospholipid binds to site I, the acyl group at sn-2 is simultaneously transferred to the catalytic site and then the resultant lysophospholipid is bound to site II. Consequently, no production of lysophospholipid was detected in the reaction mixture of PcPLB. On the other hand, PLB684 mainly produced LPC during the early stages and GPC was gradually produced after LPC accumulation. Accordingly, we propose that PLB684 hydrolyzes diacylphospholipid in no particular order; the enzyme can initially hydrolyze either the sn-1 acyl ester bond or the sn-2 acyl ester bond. In our proposal, the diacylphospholipid would bind only one substrate binding site with two binding patterns: pattern I for hydrolyzing the ester bond of sn-1 and pattern II for hydrolyzing the ester bond of sn-2. After hydrolysis, the LPC produced is released from the enzyme once and binds the other binding pattern. We concluded that the hydrolysis mechanism of PLB684 is different from fungal PLBs because an accumulation of LPC occurs upon diacylphospholipid hydrolysis.

The ORF of PLB684 consists of 1239 bp. The deduced amino acid sequence of mature PLB684 exhibited 60% identity to uncharacterized protein (UniProt accession number: J1RQY0) annotated in the genome of S. auratus AGR0001. ‘Streptomyces auratus AGR0001 uncharacterized protein’ has been integrated into the EMBL/GenBank/DDBJ databases (October 2012); however, hydrolase activity of this protein was only inferred from electric annotation but remains uncharacterized. Moreover, PLB684 exhibited no lipase activity and showed specificity for phospholipids. PLB684 has a GDSL motif near the N-terminus, showing that PLB684 belongs to GDSL hydrolase, which has a sequence motif distinctly different from the G-x-S-x-G motif found in many lipases [46]. A subgroup of this GDSL family is further classified as SGNH-hydrolase as a result of the presence of four strictly conserved residues Ser-Gly-Asn-His in four conserved blocks (blocks I, II, III and V, respectively) [46]. A hypothetical SGNH esterase motif is present at positions Ser12, Gly97, Asn142 and His332 of PLB684. This supports the proposal that one active center is present in one enzyme molecule of PLB684. It is well known that secretory proteins have signal peptides associated with their secretion. The signal peptide sequence of PLB684 is composed of 30 amino acid residues prior to the N-terminus, as determined by the protein sequencer. PLB684 should be secreted via the secretory pathway (Sec-system secretion) because the signal peptide sequence of PLB684 includes no R-R-x-φ-φ ‘twin-arginine’ amino acid motif that represents a distinct motif of secreting proteins via the translocation pathway [47, 48]. The transcription start sites of many Streptomyces genes have been defined [49, 50]. Although the promoter sequences are highly heterogeneous, many of them probably reflect the occurrence of multiple σ factors. Nevertheless, there are promoter sequences that show a high degree of similarity to the consensus sequence for the promoter sequences recognized by the major and essential σ factors of other bacteria, notably E. coli70) and B. subtilisA), and that also are probably recognized by σhrdB, which is essential in S. coelicolor, and its orthologues in other Streptomyces species. In the case of plb, the -10 region sequence, CCGAGTC, is very similar to the consensus sequence, although the -35 region sequence, CTCGGCG, is remarkably different from the consensus sequence. We considered that the sequence differences may be recognized by a minor σ factor because PLB684 is an inducible protein. The transcription of plb will be terminated by a stem-loop structure (GACGGTCTGCCGGT and ACCGGCAGACCGT) that is 3′ downstream of the plb gene. The ribosome binding site of plb would be in the -10 region sequence: CCGAGTC.

The results of the mutagenesis analysis demonstrated that Ser12 is essential for the catalytic function of PLB684, and the active site may comprise Ser330 and His332. This catalytic site resembles that of SsEst (1ESC) with a Ser-His dyad and the carbonyl group of Trp280. The active site of PLB684 may correspond to that of SaPLA1, which was previously reported by Sugimori et al. [20, 33]. Molecular modelling indicated that His73 may stabilize the phosphoric group of the bound phospholipid, thereby imparting the substrate specificity that PLB684 has towards phospholipids over 4-nitrophenyl esters. The catalytic residues of PLB from C. neoformans have been predicted using a developed homology model based on the X-ray structure of the human cytosolic PLA2 catalytic domain [51]. The two proteins share a closely-related fold, with the three catalytic residues located in identical positions as part of a single active site, with Ser146 and Asp392 forming a catalytic dyad. On the other hand, Ser12 and His332 of PLB684 presumably form a catalytic dyad. However, His332A mutant exhibited 17.2% relative activity compatred to the wild-type enzyme. Instead of His332, the other residue possibly deprotonate from Ser12, although this was not found in the structure model. The functional amino acid residues in the active site of PLB probably form a serine hydrolase-like conformation; however, PLBs have different catalytic properties with respect to their specificities to head groups and the acyl chain length. Further investigations are required to confirm these discussions and reveal the mechanisms responsible for substrate binding and ligand recognition by PLB684.

Experimental procedures

Materials

DMPA, POPC, DPPC, POPE, POPA, POPG and LPC were purchased from Avanti Polar Lipids Inc. (Alabaster, AL, USA). PI, PS and DP were purchased from Sigma-Aldrich Japan Co., LLC. (Tokyo, Japan). TS and soybean lecithin were purchased from Wako Pure Chemical Industries Ltd. (Osaka, Japan). l-α-phosphatidylcholine from egg yolk was purchased from Nacalai Tesque Inc. (Kyoto, Japan). Tryptic soy broth (TSB) and Bacto-peptone were purchased from Becton, Dickinson and Company (Franklin Lakes, NJ, USA). Beef extract was purchased from Kyokuto Pharmaceutical Co., Ltd. (Tokyo, Japan). Toyopearl DEAE-650M and Phenyl-650M were purchased from Tosoh Co. (Tokyo, Japan). HiTrap Q HP, Resource Phe and Mono S columns were purchased from GE Healthcare UK Ltd. Glycerol-3-phosphocholine phosphodiesterase (GPCP) was obtained from Asahi Kasei Pharma Co. (Tokyo, Japan). Alkaline phosphatase (AP) was purchased from Oriental yeast Co., Ltd. (Tokyo, Japan). BIOMOL Green Reagent was purchased from Enzo Life Sciences Inc. (Farmingdale, NY, USA). All other chemicals were of the highest or analytical grade.

Bacterial strains, plasmids and culture conditions

Approximately 1500 strains of actinomycetes were isolated from soil samples obtained from Fukushima, Japan, using an humic acid-vitamin agar plate [52]. Producers of phospholipases were screened with lecithin-emulsion plates [53]. Strain NA684, which showed the highest PLB activity in a 5-mL cultivation, was selected for further investigation. Strain NA684 was maintained at −80 °C in 10% (v/v) glycerol. Strain NA684 was incubated in 5 mL of 3% (w/v) TSB medium at 28 °C for 48 h with shaking (160 r.p.m.). One milliliter of the 48-h culture was transferred into a 500-mL flask containing 100 mL of lecithin nutrient broth medium (pH 7.2): 1% (w/v) Bacto-peptone, 1% (w/v) beef extract, 0.5% (w/v) NaCl, 1% (w/v) soybean lecithin and 0.1% (w/v) Tween 80, which was then incubated at 28 °C for 108 h with shaking (180 r.p.m.).

E. coli HST08 premium competent cells (Takara Bio Inc., Shiga, Japan) were used as a host for molecular cloning. S. lividans 1326 (NBRC15675) was used as a host for the expression of proteins. The pGEM-T Easy Vector (Promega Corp., Madison, WI, USA) and the pMD20-T vector (Takara Bio Inc.) were used as cloning vectors. pUC702 was used as an expression vector. Recombinant E. coli cells were cultured on LB agar plates at 37 °C; if necessary, ampicillin, isopropyl-β-d-thiogalactopyranoside and X-Gal were used as supplements in the agar. Recombinant S. lividans cells were cultured in 3% (w/v) TSB containing 5 μg·mL−1 thiostrepton at 28 °C.

Purification of PLB684

All procedures were performed at 4 °C. The culture supernatant was obtained by centrifugation (18 800 g for 20 min) after 108 h of culturing. The resulting culture supernatant was placed in an ammonium sulfate solution at 80% saturation containing 20 mm Tris-HCl buffer (pH 8.0) and was centrifuged at 18 800 g for 30 min. The resultant precipitate was suspended in buffer A (20 mm Tris-HCl, pH 9.0) containing 0.5% (w/v) Triton X-100 and 0.5% (w/v) Tween 20 and dissolved by stirring for 3 h, and this was followed by dialysis of the solution against buffer A. The enzyme sample was centrifuged (21 800 g for 20 min) to remove insoluble material. The obtained supernatant was loaded onto a Toyopearl DEAE-650M (2.5 × 5.5 cm) column equilibrated with buffer A. The column was washed with three column volumes (CV) of buffer A at a flow rate of 10 mL·min−1 (2 cm·min−1) and the protein was eluted with a linear gradient (1 CV) of 0–0.5 m NaCl and 0.5 m NaCl (2 CV) in buffer A at a flow rate of 5 mL·min−1 (1 cm·min−1). The active fractions were pooled and the buffer was exchanged to buffer A using Vivaspin 20-10 000 MWCO (Sartorius Co., Inc., Goettingen, Germany) followed by loading onto a HiTrap Q HP column (5 mL) equilibrated with buffer A. The column was washed with three CV of buffer A at a flow rate of 2 mL·min−1 (1 cm·min−1), and the protein was eluted with a linear gradient (10 CV) of 0–1 m NaCl in buffer A at the same flow rate. The active fractions were pooled. The buffer was exchanged to buffer A (pH 8.0) containing 1 m (NH4)2SO4 using Vivaspin 20-10 000 MWCO. The enzyme solution was loaded onto a Resource Phe column (1 mL) equilibrated with the same buffer. The column was washed with three CV of the same buffer at a flow rate of 1 mL·min−1 (3 cm·min−1), and the protein was eluted with a linear gradient (15 CV) of 1–0 m (NH4)2SO4 in buffer A and with buffer A (10 CV) at the same flow rate. The active fractions were pooled and the buffer was exchanged to buffer B (20 mm MES-NaOH, pH 6.0) using Vivaspin 20-10 000 MWCO. The enzyme solution was then loaded onto a Mono S column (1 mL) equilibrated with buffer B. The column was washed with three CV of the same buffer at a flow rate 0.5 mL·min−1 (2.5 cm·min−1), and the protein was eluted with a linear gradient (20 CV) of 0–0.5 m NaCl in buffer B at the same flow rate. Fractions exhibiting high specific activity were pooled and used for further investigations.

PLB activity assays

The standard assay mixture (100 μL) containing 50 mm Tris-HCl (pH 8.4), 10 mm EDTA, 0.5% (w/v) DMPA, 0.5% (w/v) Triton X-100 and 5% (v/v) enzyme solution was incubated at 50 °C for 5 min. The enzyme reaction was stopped by heating at 100 °C for 5 min. After centrifugation (21 800 g for 5 min), the concentrations of FFAs released by the enzyme reaction were determined with a NEFA C Kit (Wako Pure Chemical Industries Ltd) using oleic acid as standard and in accordance with the manufacturer's instructions. To prepare liposomal substrates, POPC, POPG, or POPC : POPE (molar ratio of 2 : 1) were dissolved in chloroform–methanol (2 : 1, v/v) solvent and then the solvent was evaporated thoroughly. Multilamellar vesicles of phospholipid were prepared by intensive vortexing after hydration of the lipid film with distilled water at 65 °C for 1 h. The formation of the liposome was observed using a phase contrast microscope. The assay mixture (100 μL) containing 50 mm Tris-HCl (pH 8.4), 10 mm above the liposomal phospholipid (diameter 1–10 μm) and 5% (v/v) enzyme solution was incubated at 37 °C for various intervals. One unit (U) of enzymatic activity was defined as the amount of enzyme that released 1 μmol of FFAs from glycerophospholipids per min.

Measurement of GPC

The standard assay mixture containing 0.5% (w/v) POPC as a substrate was incubated at 37 °C for various intervals. The enzyme reaction was stopped by heating at 100 °C for 5 min. An aliquot (50 μL) of the reaction mixture was added into the GPCP reaction mixture (100 μL), containing 50 mm Tris-HCl (pH 9.0), 15 mm CaCl2 and 0.17 U of GPCP. The GPCP reaction mixture was incubated at 37 °C for 20 min. An aliquot (10 μL) of the reaction mixture was then added to the AP reaction mixture (100 μL), containing 50 mm glycine-NaOH (pH 9.6), 1 mm MgCl2, 0.1 mm ZnCl2 and 2 U of AP, and this solution was then incubated at 56 °C for 10 min. The concentration of inorganic phosphate produced in the AP reaction mixture was determined with BIOMOL Green Reagent. The concentration of GPC produced by PLB684 was equal to the concentration of inorganic phosphate measured.

Effect of pH, temperature and chemicals on PLB activity

Each buffer (sodium acetate, BisTris-HCl, Tris-HCl and glycine-NaOH) was used to investigate the effect of pH on enzyme activity and stability. The optimum pH was examined by incubation at 37 °C for 5 min with 0.5% (w/v) DMPA in 50 mm of each buffer containing 10 mm EDTA and 0.5% (w/v) Triton X-100. The pH stability was assayed by incubating the enzyme at 4 °C for 3 h in 40 mm of each buffer solution. The residual enzyme activity was measured under standard assay conditions: incubation at 50 °C for 5 min with 0.5% (w/v) DMPA in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100. The optimum temperature was determined by measuring the enzyme activity at each temperature using the standard assay mixture. The apparent activation energy (Ea) for DMPA hydrolysis was determined from the slope of the Arrhenius plot. The thermal stability was determined by incubating the enzyme in 50 mm Tris-HCl (pH 8.4) at each temperature for 30 min, and then the residual activity was assayed under standard assay conditions. The effect of chemicals such as metal ions, reductants and inhibitors on enzyme activity was investigated. The enzyme activity was assayed by incubation at 50 °C for 5 min with 0.5% (w/v) DMPA in 50 mm Tris-HCl (pH 8.4) containing each concentration of the tested chemicals. The effect of the concentration of Triton X-100 on enzyme activity was investigated. The enzyme activity was assayed by incubation at 50 °C for 5 min with 0.5% (w/v) DMPA in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and each concentration of Triton X-100 was examined.

Steady-state kinetics

The purified enzyme was used for steady-state kinetic analysis. For DMPA hydrolysis, the initial velocity of the enzymatic reaction was determined at each concentration of DMPA as a substrate under standard assay conditions. The enzyme concentration was constant at 37 ng·mL−1 (951 pm). [DMPA] was calculated using a relative molecular mass of 614.78. The corresponding [DMPA] versus [DMPA]/v plots were treated according to the Michaelis–Menten equation. The kinetic constants, Km, Vmax and kcat, were determined by fitting the data on activity at different DMPA concentrations (3.25–16.3 mm) to a linear regression on a Hans–Woolf plot (KaleidaGraph, Synergy Software, Reading, PA, USA). Km and Vmax were determined from x- and y-intercepts of the regression line, respectively. The turnover rate (kcat) was calculated using a relative molecular mass of 38 900 for the monomeric protein and one catalytic site.

GC analysis

The positional specificity of the hydrolytic reaction was determined by capillary GC analysis. The enzymatic reaction using 0.5% (w/v) POPC as a substrate was performed at 37 °C in 50 mm Tris-HCl (pH 8.4) containing 10 mm EDTA and 0.5% (w/v) Triton X-100 for various intervals. Produced FFAs in the reaction mixture (20 μL) were extracted by 100 μL of a chloroform–methanol (2 : 1, v/v) solvent. The chloroform phase was dried and the residual material was dissolved in 20 μL of the chloroform–methanol (2 : 1, v/v) solvent. FFAs in the extracts were separated using a SUPELCO Nukol capillary column (Sigma-Aldrich Co., LLC., St Louis, MO, USA; 15 m × 0.53 mm ×0.5 μm) and detected by a GC-14B (Shimadzu Co., Kyoto, Japan). The split ratio was 50 : 1. The temperature program ramped from 110 to 220 °C at 8 °C·min−1, and the temperature was held at 220 °C for 15 min. The injector and detector were set at 250 °C. The flow rate of the He carrier gas was 25 mL·min−1.

PLA1 and PLA2 activity assay

The EnzCheck® Phospholipase A1 assay kit and the Phospholipase A2 assay kit were used. The assay kits are a simple, fluorometric method designed for continuous monitoring of PLA1 or PLA2 activity. The substrates are specific for each enzyme and are dye-labelled glycerophosphoethanolamine and glycerophosphocholine with a BODIPY(R)FL dye-labelled acyl chain at the sn-1 or the sn-2 position. The results are a PLA1- and PLA2-dependent increase in BODIPY(R)FL fluorescence emission, which is detected at approximately 515 nm. Specificity is imparted by the placement of an acyl group with an enzymatic resistant (noncleavable) ether linkage in each position. Each activity was determined in accordance with the manufacturer's instructions.

Protein analysis

The protein concentration was determined with a bicinchoninic acid protein assay reagent kit (Thermo Fisher Scientific Inc., Rockford, IL, USA) with BSA as standard. The molecular mass of the purified PLB684 was estimated by gel filtration, DLS analysis and blue native-PAGE (Life Technologies Corp.) analysis. Gel filtration was performed using a Superdex 200 10/300 GL column (1.0 × 30 cm; GE Healthcare, UK Ltd.) at a flow rate of 0.5 mL·min−1 (37.5 cm·h−1) with 20 mm Tris-HCl (pH 8.0) containing 0.15 m NaCl. The column was calibrated using a standard protein gel filtration calibration kit (GE Healthcare, UK Ltd.). Blue native-PAGE was performed in accordance with the manufacturer's instructions. DLS measurements were performed on a Zetasizer NanoZ (Malvern Instruments, Malvern, UK). SDS/PAGE was carried out according to the method of Laemmli [54]. Proteins on the SDS/PAGE gel were electroblotted onto a poly(vinylidene difluoride) membrane (Millipore Co., Billerica, MA, USA). The gels and poly(vinylidene difluoride) membranes were stained with Coomassie brilliant blue R-250. The 38.9-kDa band transferred onto the poly(vinylidene difluoride) membrane was excised and subjected to protein sequencer analysis (Procise 494 HT Protein Sequencing System; Applied Biosystems, Foster City, CA, USA). For internal amino acid sequencing, the proteins from the excised SDS/PAGE gel were digested by trypsin (Sequencing Grade Modified Trypsin; Promega Corp.) and the resulting peptides were extracted in accordance with the method of Shevchenko et al. [55]. The extracted peptides were separated using a nano ACQUITY UPLC® BEH130 C18 column (Waters Corp., Milford, MA, USA; 75 μm × 150 mm × 1.7 μm) and a Xevo QTOF MS as described previously [20]. The collision gas was argon and the collision energy was varied to monitor the product ions of interest. Accurate mass LC-MS/MS data in the data-dependent acquisition mode were obtained. MS survey scans of 0.6 s in duration with an interscan delay of 0.05 s were acquired. MS/MS data were obtained for up to three ions of charge 2, 3, or 4 detected in the survey scan. MS/MS was obtained at a scan rate of 0.6 with an interscan delay of 0.05 s and a collision energy ramp from 15 to 40 eV. Data acquisition was achieved with masslynx, version 4.1 SCN 712 (Waters Corp.). De novo sequencing was performed with the proteinlynx global server, version 2.3 (Waters Corp.).

Cloning of the gene for PLB684

Chromosomal DNA of Streptomyces sp. NA684 was purified by salting out in accordance with the method of Pospiech and Neumann [56]. Two oligonucleotide primers (S1; 5′-GGCAAGCCSACSGCSGTSGTS-3′ and A1; 5′-CTTCCAGCAGCCGTACTGGCCSAC-3′) were designed and synthesized based on the N-terminal (GKPTAVVSLGDS) and internal (FLASPVGQYGCWK) amino acid sequences of the enzyme. The first PCR reaction mixture (50 μL) contained 1× Phusion GC buffer, 10 nmol dNTPs, 25 pmol of each primer, 1 U of Phusion DNA polymerase (Thermo Fisher Scientific Inc.) and approximately 50 ng of Streptomyces sp. NA684 chromosomal DNA as template. The two-step PCR was performed with 30 cycles of 98 °C for 10 s and 68 °C for 20 s. The second PCR reaction mixture (50 μL) contained 1× MightyAmp buffer, 15 pmol of each primer, 1.25 U of MightyAmp DNA polymerase (Takara Bio Inc.) and approximately 50 ng of the DNA fragments extracted from the first PCR reaction mixture. The second PCR was carried out with 30 cycles of 98 °C for 10 s and 66 °C for 30 s. The obtained DNA fragment was cloned into the pGEM-T Easy Vector (Promega Corp.), resulting in the recombinant plasmid pPAT. To reveal the complete sequence of the ORF coding PLB684, two primers for inverse PCR (sense primer IS1 5′-CAACTGGAAGACCTCGCACGCACCTACGAC-3′ and antisense primer IA1 5′-GTCCGCTTGGGGCCACTCCCCCTTGAAGAG-3′) were designed based on the partial sequence of the gene for PLB684 in pPAT. The genomic DNA (5 μg) was digested with SalI and self-ligated. The inverse PCR reaction mixture (50 μL) contained 1× MightyAmp buffer, 15 pmol of each primer, 1.25 U of MightyAmp DNA polymerase and approximately 200 ng of the SalI-digested and self-ligated DNA. The inverse PCR was performed with 20 cycles of 98 °C for 10 s and 68 °C for 5 min. The obtained DNA fragment was cloned into the pMD20-T vector using the Mighty TA-cloning kit (Takara Bio Inc.), resulting in the recombinant plasmid pINV. To clone the complete PLB684 gene from the chromosomal DNA of strain NA684, two cloning primers (sense primer SE1 5′-CAAGAGATCCAGCCGTCGAAAG-3′ and antisense primer AN1 5′-TCCGATCGAGCAGCGAAATC-3′) were designed based on the sequence of the gene for PLB684 in pINV. The genomic PCR reaction mixture (50 μL) contained 1× PCR buffer #1, 50 pmol of each primer, 15 nmol of dNTPs, 62.5 nmol of MgCl2, 3% (v/v) dimethylsulfoxide, 2.5 U of KOD DNA polymerase (Toyobo Co., Ltd., Osaka, Japan) and approximately 100 ng of genomic DNA of strain NA684. The obtained DNA fragment was sequenced and cloned into the pMD20-T vector, resulting in the recombinant plasmid pPLB.

Nucleotide and peptide sequence accession number

The nucleotide sequences of the 16S rDNA of Streptomyces sp. strain NA684 and the gene for PLB684, designated plb, were deposited in the DDBJ database under accession numbers AB738936 and AB602789, respectively.

Expression and purification of recombinant PLB684

S. lividans 1326 (NBRC15675) was obtained from the NITE Biological Resource Center (Chiba, Japan). S. lividans 1326 possessing no phospholipid acyl esterase activity was used as a host for PLB684 extracellular production. The PCR was carried out using the primers: 5′-CCGCGCTAGCGGGAAGCCCACG-3′ (Nhe1plb) and 5′-ACAGATCTTCAAGGACGACTGAGAGG-3′ (plbBgl2), containing the N-terminal codon (NheI, single underline; Gly, italics) and the stop codon (BglII, single underline; stop codon, italics) of the mature PLB684, respectively. The PCR reaction mixture (50 μL) contained: 1× PCR buffer #1, 1.25 mm MgCl2, 0.6 μm of each primer, 200 μm dNTPs, 3% (v/v) dimethylsulfoxide, 1.5 U of KOD DNA polymerase and 50 ng of the pPLB plasmid DNA as template. The thermal cycling parameters were 98 °C for 2 min, followed by 30 cycles of 98 °C for 15 s, 68 °C for 2 s and 74 °C for 40 s. The obtained fragment was purified and digested with NheI and BglII, and then subcloned into the NheI and BglII sites of pUC702 carrying the promoter, signal peptide sequence and the terminator region of phospholipase D from Streptoverticillium cinnamoneum [57]. The obtained recombinant plasmid was sequenced and designated as pUC702/plb. The transformation techniques for S. lividans followed the methods of Bibb et al. [58] and Thompson et al. [59]. Transformants were screened using lecithin-emulsified nutrient plates [53]. Clones exhibiting a cloudy halo were collected and the clone exhibiting the highest activity in the 5-mL cultivation was selected. The PLB684 produced by the transformed S. lividans was purified from a 36-h culture supernatant by ammonium sulfate precipitation, as well as hydrophobic interaction and anion exchange chromatography steps (refer to the wild-type enzyme purification process in this paper).

Homology modelling of PLB684

From the sequence information, PLB684 from Streptomyces sp. NA684 appeared to belong to the SGNH-hydrolase family. SsEst also belongs to the SGNH-hydrolase family. The structural features of SsEst have been studied [60]. Based on a template (Protein Data Bank code: 1ESC), the homology model of PLB684 was created using hhpred (http://toolkit.tuebingen.mpg.de/hhpred).

Site-directed mutagenesis of PLB684

The active site amino acids of SsEst are Ser14, Trp280 and His283, and the esterase hydrolyzes specific ester bonds in suberin, a wax-like lipid [60]. Residues predicted to play a catalytic role in the active center of PLB684 were mutated by site-directed mutagenesis with inverse-PCR amplification. The PLB684 variants Ser12A, Ser330A and His332A were generated using the KOD plus mutagenesis kit (Toyobo Co., Ltd) and pUC702/plb as the template. The mutant proteins were produced extracellularly by the transformed S. lividans. Clones exhibiting a cloudy or no halo around their colonies were screened and selected. The production of all variant proteins was verified by SDS/PAGE analysis and the nucleotide sequences of the resulting constructs were determined. Each variant protein was purified and the enzyme activity was assayed using the methods described above.

Acknowledgements

We thank Asahi Kasei Pharma Corp., Japan, for providing choline oxidase and glycerol-3-phosphocholine phosphodiesterase, and also Associate Professor Chiaki Ogino (Kobe University, Japan) for providing the expression vector.

Ancillary