HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 22, 20051-20058, May 31, 2002

This Article Abstract Full Text (PDF) An addition or correction has been published All Versions of this Article: 277/22/20051    most recent M200264200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sugiyama, M. Articles by Misaki, H. Search for Related Content PubMed PubMed Citation Articles by Sugiyama, M. Articles by Misaki, H. Social Bookmarking                      What's this?

A Novel Prokaryotic Phospholipase A₂

CHARACTERIZATION, GENE CLONING, AND SOLUTION STRUCTURE^*

Masanori Sugiyama^§, Kazuhiro Ohtani, Miho Izuhara, Tohru Koike, Koji Suzuki^¶, Shigeyuki Imamura^¶, and Hideo Misaki^¶

From the  Institute of Pharmaceutical Sciences, Faculty of Medicine, Hiroshima University, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8551, and ^¶ Diagnostic Research and Development, Fine Chemicals and Diagnostics Division, Asahi Kasei Corp., 632-1, Mifuku, Ohito, Shizuoka 410-2321, Japan

Received for publication, January 10, 2002, and in revised form, February 27, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Until now, phospholipase A₂ (PLA₂; EC 3.1.14) has been found only from eukaryotic sources. In the present study, we found a secreted PLA₂, which is produced by a soil bacterium, Streptomyces violaceoruber A-2688, demonstrating that the enzyme is the first phospholipase A₂ identified in prokaryote. After characterization of the novel PLA₂, a gene encoding the enzyme was cloned, sequenced, and overexpressed using a Streptomyces host-vector system. The amino acid sequence showed that the prokaryotic PLA₂ has only four cysteines and less homology to the eukaryotic ones, which have 12-16 cysteines. The solution structures of the prokaryotic PLA₂, bound and unbound with calcium(II) ion, were determined by using the NMR technique and structure calculation. The overall structure of the S. violaceoruber PLA₂, which is composed of only five -helices, is completely different from those of eukaryotic PLA₂s, which consist of -sheets and -helices. The structure of the calcium-binding domain is obviously distinct from that without the ion; the ligands for the calcium(II) ion are the two carboxylates of Asp⁴³ (monodentate) and Asp⁶⁵ (bidentate), the carbonyl oxygen of Leu⁴⁴, and three water molecules. A calcium-binding experiment showed that the calcium dissociation constant (~5 mM) for the prokaryotic PLA₂ is much larger than those of eukaryotic ones.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Phospholipase A₂ (PLA₂¹; EC 3.1.1.4), which hydrolyzes the 2-acyl ester bonds of 1,2-diacylglycero-3-phospholipids (Fig. 1), is intra- or extracellularly produced by eukaryotic cells (1). Intracellular PLA₂s are divided into calcium-dependent and calcium-independent types. Since a certain calcium-dependent intracellular PLA₂ catalyzes the release of arachidonic acid and participates in the inflammatory cascade, the development of drugs inactivating the human PLA₂ is of great pharmacological interest (2-4).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Hydrolysis of the 2-acyl ester bond of 1,2-diacylglycero-3-phosphocholine by PLA2.

All secreted PLA₂s require calcium(II) ion for the expression of catalytic activity. These PLA₂s have been grouped into three classes based on their primary structures (5). Class I enzymes are found in mammalian pancreatic juice and cobra venom. Class II enzymes are contained in synovial fluid and rattlesnake venom. The PLA₂s included in both classes have a molecular mass of 14 kDa and mutually exhibit high similarities on their primary structures. A PLA₂ contained in bee venom has a molecular mass of 12 kDa and is classified into class III.

With respect to the physiological functions of PLA₂, for example, snake venom PLA₂s have evolved into extremely potent toxins displaying neurotoxic, myotoxic, anticoagulant, and proinflammatory effects (6, 7). Pancreatic PLA₂, secreted from the pancreatic acinar cells as an inactive zymogen, functions as a digestive enzyme after proteolytic cleavage of the pro form by trypsin to the mature form (8). We have shown that the mature form of human pancreatic PLA₂, but not its precursor form, stimulates the proliferation of a human pancreatic cancer cell line (9). We have found that the proliferation is mediated via the activation of mitogen-activated protein kinase after binding of the mature PLA₂ to its specific receptor (10) but not via its catalytic property (11). In the proliferation experiment, since the bacterial PLA₂ was ineffective, we examined the tertiary structural difference for the binding of PLA₂ to the specific receptor between the human pancreatic and bacterial PLA₂s.

Several structural studies of PLA₂ in solution (12) and in crystal (13-20) have been reported. The crystal structural studies of secreted PLA₂s have shown that the overall structures of class I PLA₂s are similar to those of class II enzymes. Although the primary structures of class I and class II enzymes are distinct from that of class III, x-ray crystallographic analyses (20) show that the catalytic and calcium-binding sites in class I and class II are conserved in class III.

In recent years, many additional forms of secreted PLA₂s have been discovered. These additional forms are clearly related to classes I, II, and III PLA₂s but do not easily fit into those groups. This has led to the establishment of classes V, IX, X, XI (21), and XII (22). According to this classification, cytosolic PLA₂s are divided into classes IV, VI, VII, and VIII (21). For example, a PLA₂, conodipine-M, isolated from the venom of the marine snail Conus magus, has been classified as class IX (21, 23).

In the present study, we investigated the enzymatic properties of the Streptomyces violaceoruber PLA₂, which is the first phospholipase A₂ discovered in prokaryote. Furthermore, we cloned and sequenced a gene encoding the bacterial PLA₂. The overproduction of the S. violaceoruber PLA₂ using a Streptomyces host-vector system allowed us to determine the three-dimensional structure of the enzyme in solution by NMR as well as the x-ray crystal structure, which is described the accompanying paper (38).

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

General Information-- All reagents and solvents used were of analytical reagent grade and were used without further purification. All aqueous solutions were prepared with deionized and distilled water. Good's buffers, PIPES, and EPPS from Dojindo, L--phosphatidyl choline (egg yolk) from Nakalai Tesque, and 1,2-dipalmitoyl-L--phosphatidylcholine (1,2-dipalmitoyl-glycero-3-phosphocholine) and palmitoyl-L--lysophosphatidylcholine (1-palmitoyl-glycero-3-phosphocholine) from Sigma were commercially available. The pH electrode system (DKK Corp. IOL-40 with a Ross combination pH electrode 8102BN) was calibrated by using standard pH buffers (pH 4 and 7; Horiba Ltd.) at 25.0 and 35.0 °C with stirring. Proton NMR (500-MHz) spectra were determined on a JEOL Lambda-500 spectrometer, where the solvent d₄-methanol ( 3.30) was used as the standard signal: 1,2-dipalmitoyl-L--phosphatidylcholine,  0.89 (6H, CCH₃), 1.28 (48H, CCH₂C), 1.59 (4H, CH₂CCOO), 2.31 (4H, CH₂COO), 3.22 (9H, NCH₃), 3.64 (2H, CCH₂N), 3.99 (2H, CCCH₂OP), 4.17 (1H, CHCC), 4.26 (2H, POCH₂CN), 4.43 (1H, CHCC), and 5.24 (1H, CCHC); 1-palmitoyl-L--lysophosphatidylcholine,  0.89 (³H, CCH₃), 1.28 (24H, CCH₂C), 1.61 (4H, CH₂CCOO), 2.35 (4H, CH₂COO), 3.22 (9H, NCH₃), 3.63 (2H, CCH₂N), 3.89 (2H, CCCH₂OP), 3.97 (1H, CCHC), 4.11 (1H, CHCC), 4.17 (1H, CHCC), and 4.28 (2H, POCH₂CN).

Screening of PLA₂-producing Microorganisms-- Many strains of the genus Streptomyces having phospholipase activity, isolated from a soil sample in Japan, are stocked in the Diagnostic Research and Development Department, Fine Chemicals and Diagnostics Division, Asahi Kasei Co., Japan. To screen candidates of PLA₂-producing microorganisms, we examined whether PLA₂ activity was observed in the supernatant fluid obtained by growing the stock strains in a liquid medium consisting of Tryptic soy broth (Difco). The supernatant was incubated with a PLA₂ assay medium (1.0 ml) consisting of a 0.10 M PIPES buffer (pH 7.5 adjusted with NaOH), 6.0 mM 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (Avanti Polar Lipids, Inc.), and 10 mM CaCl₂. The mixture was incubated at 40 °C for 1 h with gentle shaking, and then 0.10 ml of 5.0 M HCl and 1.5 ml of hexane were added. After a portion (150 µl) of the hexane phase was dried up, the residue was dissolved in 0.20 ml of hexane-saturated methanol containing 50 µl of trimethylsilyldiazomethane as a methylating reagent. The obtained oleic methyl ester was subjected to a general gas chromatography. The bee venom PLA₂ (Sigma) was used for a control experiment.

Medium and Culture Conditions-- S. violaceoruber A-2688 was grown in 3% (w/v) tryptic soy broth at 28 °C for 2 days for seed culture. The seed culture was transferred into 20 liters of a liquid medium containing 2% (w/v) dextrin, 1%(w/v) glucose, 0.5% (w/v) soy bean powder, 1% (w/v) polypeptone, 0.3% (w/v) KCl, 0.1% (w/v) MgSO₄, 0.1% (w/v) K₂HPO₄, and a 0.1% (v/v) trace element solution (40 mg of ZnCl₂, 0.20 g of FeCl₃·6H₂O, 10 mg of CuCl₂·2H₂O, 10 mg of MnCl₂·4H₂O, 10 mg of Na₂B₄O₇·4H₂O, and 10 mg (NH₄)₆Mo₇O₂₄·4H₂O in 1 liter of H₂O) and grown at 28 °C for 4 days. Acetone (0.45 volume) was added to the supernatant fluid by centrifuging the culture broth at 10,000 × g for 10 min. The resulting supernatant, obtained by the same speed centrifugation, was again added with 2.8 volumes of acetone and centrifuged to collect the precipitate. The precipitate was dissolved into a 10 mM Tris-HCl buffer (pH 8.0) containing 20 mM CaCl₂ and subjected to a Sephadex G-25 column ( 4.5 × 50 cm) equilibrated with a 10 mM citrate-NaOH buffer (pH 5.0). The active fraction was applied on a CM-Sepharose CL-6B column ( 0.5 × 30 cm) equilibrated with the 10 mM citrate buffer and eluted with a linear gradient of the 10 mM citrate buffer containing 0-0.5 M KCl. After removing the salt using the 10 mM Tris-HCl buffer, the enzyme fraction was applied on a DEAE-Sepharose CL-6B column ( 2.6 × 30 cm) equilibrated with the 10 mM Tris-HCl buffer. The active fraction was obtained by a gradient elution with the 10 mM Tris-HCl buffer containing 0-1 M KCl. The active fraction was dialyzed against the 10 mM Tris-HCl buffer at 4 °C overnight, lyophilized, and stored in a refrigerator until use.

On the other hand, the cloned PLA₂ gene was expressed in S. lividans TK24 as a host cell (24). S. lividans carrying the PLA₂ gene-inserted plasmid was grown in 20 liters of 3% (w/v) tryptic soy broth supplemented with 10 µg/ml thiostrepton, which is necessary to maintain the plasmid-harboring thiostrepton resistance gene as a selection marker in the Streptomyces vector pIJ680 (25). After centrifugation at 8,500 × g for 20 min, the supernatant fluid was added with solid (NH₄)₂SO₄ to reach 40% saturation. The resulting precipitate was removed by centrifugation at 8,500 × g for 20 min. The precipitate, formed by the further addition of (NH₄)₂SO₄ to reach 50% saturation to the supernatant, was collected, dissolved in a 20 mM Tris-HCl buffer (pH 8.0), and dialyzed against the same buffer. The dialysate was applied on a Q-Sepharose Fast Flow column ( 5.0 × 15 cm; Amersham Biosciences) equilibrated with 20 mM Tris-HCl buffer, and eluted with a linear gradient of 0-1.0 M NaCl in the 20 mM Tris-HCl buffer. The active fractions were collected and dialyzed against the 20 mM Tris-HCl buffer and then lyophilized. The lyophilized PLA₂ was dispersed in mannitol as a stabilizer and kept below 18 °C.

An Assay Method of PLA₂ Activity-- In order to measure PLA₂ activity, we prepared a reaction mixture consisting of solutions I and II. Solution I (0.50 ml) contained 0.20 ml of a 0.20 M Tris-HCl buffer (pH 8.0), 0.50 ml of 20 mM 1,2-dipalmitoyl-glycero-3-phosphocholine dissolved in 2% (w/v) Triton X-100, 25 µl of 1.0 M CaCl₂, 10 µl of 0.20 M ATP, 20 µl of 0.10 M CoA, 50 µl of acyl-CoA synthetase (Asahi Kasei Co., catalog no. T-16; 10.6 units/ml), and 145 µl of distilled water. Solution II (0.50 ml) consisted of 40 µl of a 0.20 M PIPES buffer (pH 7.5 adjusted with NaOH), 0.10 ml of 0.3% (w/v) 4-aminoantipyrine, 0.10 ml of 0.20% (w/v) phenol, 45 µl of peroxidase (50 units/ml; Sigma, type II P-8250), 50 µl of acyl-CoA oxidase (Asahi Kasei Co., catalog no. T-17; 300 units/ml), 10 µl of 10% (w/v) Triton X-100, 50 µl of 0.20 M ATP, 5.0 µl of 10 mM FAD, and 0.10 ml of distilled water. This assay is based on the increase in absorbance at 500 nm due to quinoneimine dye, which is finally formed when PLA₂ liberates fatty acid as a result of hydrolysis of lecithin as a substrate. The addition of acyl-coenzyme A (acyl-CoA) synthetase with CoA and ATP catalyzes the formation of acyl-CoA in the presence of the fatty acid. The addition of acyl-CoA oxidase contributes to form H₂O₂ in the presence of the resulting acyl-CoA. The dye, 4-quinoneimine, is finally generated from 4-aminoantipyrine as a result of catalysis by peroxidase in the presence of phenol and H₂O₂. An enzyme fraction (0.50 ml) was added to solution I (500 µl) preincubated at 37 °C for 5 min, and incubated at 37 °C for 10 min. The reaction mixture was added in 0.50 ml of 20 mM N-ethylmaleimide and then mixed with solution II (0.50 ml). After incubation at 37 °C for 5 min, 1.5 ml of 5% (w/v) SDS containing 0.10 M EDTA was added to stop the reaction, and then absorbance at 500 nm was measured. One unit of PLA₂ activity was defined as the amount of enzyme that liberates 1.0 µmol of fatty acid from lecithin per min at 37 °C.

Determination of the Amino Acid Sequence of PLA₂-- After the purified PLA₂ (0.50 mg) had been dissolved in a 0.80 M Tris-HCl buffer (pH 8.0) and incubated at 90 °C for 15 min, the sample was digested with endoprotease Asn-N (Roche Molecular Biochemicals) at 37 °C overnight. The digest was mixed with 0.60 ml of 6.0 M guanidine HCl and subjected to HPLC using an ODS column (Cosmoseal C-18, Nakalai Tesque, Japan). On the other hand, the purified PLA₂ (0.50 mg), dissolved in 0.60 ml of 6.0 M guanidine HCl, was adjusted to pH 9.0 with a 0.80 M Tris base. After incubation at 90 °C for 15 min, the sample was cooled down to room temperature, and then lysyl endopeptidase (5.0 µg; Wako, Japan) was added. After incubation at 37 °C for 150 min, the resulting peptide fragments were purified by using HPLC carrying the same ODS column as used above. The N-terminal amino acid sequence of the peptides was determined using an automated protein sequencer (Shimadzu PSQ-1).

Cloning of a PLA₂ Encoding Gene-- To make a genomic library in Escherichia coli TG1 of S. violaceoruber A-2688, the chromosomal DNA was isolated and digested with KpnI and inserted into the KpnI-digested pUC118. To obtain the S. violaceoruber PLA₂ gene, four kinds of mixed probes (probes 1-4), shown in Fig. 2, were designed and synthesized, based on the N-terminal sequences of the S. violaceoruber PLA₂ digested with or without peptidases described above. The library, prepared in E. coli TG1, was screened using the mixed probes labeled with -³²P and by colony hybridization.

Transformation in Streptomyces-- The experiment of the introduction of a plasmid carrying the PLA₂ gene into the S. lividans cells was conducted mainly according to the reported method (25). This method is based on the discovery that protoplasts, prepared from the Streptomyces mycelium, are transformed by a Streptomyces plasmid DNA at a high frequency in the presence of polyethylene glycol. The transformed protoplasts are regenerated to form colonies on a regeneration medium, designated R5.

Determination of the Calcium Dissociation Constant of PLA₂-- The calcium dissociation constant (K_d) was determined by the following kinetic procedure. An aqueous solution containing 1.7 mg/ml (~2 mM) L--phosphatidylcholine (PC) from egg yolk, a 1.0 mM EPPS buffer, 10 mM Triton X-100, 0.10 M NaCl, and CaCl₂ (1.0, 2.0, 3.0, 5.0, 10, or 20 mM) was prepared as a test solution. After the solution had been sonicated for more than 10 min at 35 °C, the pH was adjusted to 8.00 with 0.10 M NaOH. The PC hydrolysis catalyzed by the S. violaceoruber PLA₂ was followed by the pH decrease that results from the release of fatty acid (pK_a ~5) in the 2-position of PC. After an aqueous solution (10 µl) containing 23 µM PLA₂ and 1.0 mM ethylenediamine tetraacetic acid disodium salt had been rapidly injected in the test solution (10 ml) at 25 and 35 °C under a nitrogen atmosphere (99.999% purity), the pH was immediately recorded. The time (T) for the pH drop from 7.90 to 7.70 was measured in the presence of calcium(II) ion (e.g. T = 120 s at [Ca²⁺] = 10 mM and 35 °C), where the PC hydrolysis rate (V) at pH 7.8 is defined as 1/T (s^-1). Background PC-hydrolysis in the absence of calcium(II) ion was determined under the same condition, where no pH drop was observed within 10 min. The same pH drop corresponds to ~40 µM H⁺ release, which was determined by pH titration with 2.0 mM HCl. All kinetic experiments were run in triplicate, and the obtained hydrolysis rates were reproducible within 10%.

NMR Spectroscopy-- The recombinant PLA₂ dispersed in mannitol was dialyzed against distilled water for 48 h at room temperature. The dialysate was freeze-dried and then subjected to the preparation of a test solution of 3.0 mM calcium-free PLA₂ in a 20 mM phosphate buffer (pD 7.6 in D₂O or pH 7.6 in 10% (v/v) D₂O/H₂O). The same buffer containing 20 mM CaCl₂ was used for the calcium-bound PLA₂. The pD value in D₂O was corrected for the deuterium isotope effect using pD = (pH meter reading) + 0.4. All measurements were run on a JEOL Lambda-500 spectrometer equipped with a field gradient H-X probe at 313 K. The ¹H and ¹³C NMR assignments were obtained by one-dimensional (¹H and ¹³C) and two-dimensional NMR (double quantum-filtered COSY, double quantum-filtered NOESY, and TOCSY) experiments. Double quantum-filtered NOESY spectra (20 ppm sweep width, 1,024 points in F2 and 512 points in F1, 128 scans per F1 block) were obtained with mixing times ranging from 20 to 200 ms and zero-filled to 2,048 × 2,048 complex points before Fourier transformation. The 1% shifted sine squared bell window was used for F2, and the 3% sifted sine squared bell window was used for the F1 dimension. TOCSY spectra (20 ppm spectral width, 1,024 points in F2 and 512 points in F1, 256 scans per F2 block) were recorded with spin-locking times ranging from 50 to 100 ms and treated in the same way as for the NOESY spectra. Phase-sensitive NOESY spectra (20 ppm spectral width, 1,024 points in F2 and 512 points in F2, 384 scans per F2 block) were measured with mixing times ranging from 20 to 200 ms. The 3% sifted sine squared bell for F2 and the 5% sifted sine squared bell were used for window functions. All NMR calculations were performed on an SGI work station O2 with the Lambda program (JEOL).

Constraints-- The obtained nuclear Overhauser effects were classified as strong, medium, weak, and very weak, corresponding to the interproton distance constraints of 2.0-2.5, 2.0-3.0, 2.0-4.0, and 2.0-5.0 Å, respectively. For the upper distance limits for nuclear Overhauser effect distances involving methyl protons, nonstereospecifically assigned methylene protons and aromatic ring protons as well as pseudo-atom corrections were applied of 1.0, 0.9, and 2.0 Å, respectively. The final distance geometry (DG) structures of calcium-free PLA₂ were calculated based on 1,616 constraints (448 for the intraresidue, 535 for the sequential interresidue, 329 for the medium range, and 304 for the long range). For the calculation of the final DG structures of calcium-bound PLA₂, 1,593 constraints (437 for the intraresidue, 437 for the sequential interresidue, 332 for the medium range, and 298 for the long range) were used.

Structure Calculations and Iterative Structure Refinements-- All structure calculations were carried out using the InsightII/Discover-2000 software package (MSI, San Diego, CA) on an SGI work station O2. Starting structures for calcium-free and calcium-bound enzymes were generated with distance geometry using the DGII module. The obtained starting structure was initially subjected to 100 steps of steepest descent minimization followed by 1,500 iterations using conjugate gradient minimization and then by restrained molecular dynamics for 6 ps, using a consistent valence force field (26). Finally, another energy minimization step was performed (250 steepest descent steps followed by 2,500 iterations using conjugate-gradient minimization) to give initial structures. The final structures were calculated by using an iterative structure refinement, as described in the literature (27). The atomic coordinates of the NMR structure of the enzyme together with the constraints and the chemical shifts have been deposited in the Protein Data Bank.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Properties of PLA₂ Produced by S. violaceoruber-- The product analysis by gas chromatography showed that an enzyme sample, partially purified from the supernatant from the culture broth of S. violaceoruber A-2688, liberated oleic acid from 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine, suggesting that the enzyme sample, extracellularly produced by the bacterium, contributed to hydrolyze the 2-ester bond of 1,2-diacyl-3-sn-phosphoglycerides. Thus, this enzyme sample seemed to contain PLA₂. We purified the putative S. violaceoruber PLA₂ to homogeneity and confirmed that it certainly was PLA₂ by the same gas chromatographic experiment. The molecular weights of the PLA₂, estimated by SDS-PAGE and Sephadex G-100, were 14,000 and 15,000 daltons (± 1,000), respectively, suggesting that the S. violaceoruber PLA₂ is a monomeric enzyme with almost the same size as PLA₂ from eukaryotic sources. The isoelectric point was 7.5, and the pH optimum for enzyme reaction is 7.3-8.3, when measured using a 20 mM Britton-Robinson buffer. The enzyme was stable at 50 °C for 10 min in a 20 mM Britton-Robinson buffer (pH 8.0) in the presence of 50 mM CaCl₂. The PLA₂ reaction was activated and inhibited by the addition of calcium ion and EDTA, respectively.

Various phospholipids differing in the polar head groups are known. When the relative activity of the S. violaceoruber PLA₂ to PC is defined to be 100%, those of phosphatidylethanolamine and phosphatidic acid were 30 and 10%, respectively. Lysophosphatidylcholine was not useful as a substrate of bacterial PLA₂. However, human synovial PLA₂ hydrolyzes phosphatidylethanolamine more effectively than phosphatidylserine and PC (28).

S. violaceoruber A-2688 produced 0.5 units of PLA₂ per ml at a large scale in a 20-liter liquid medium consisting of tryptic soy broth. Twenty mg of purified PLA₂ was obtained from the 20-liter culture broth.

Cloning and Sequencing of a PLA₂-encoding Gene-- To clone a gene encoding the S. violaceoruber PLA₂, we determined the amino acid sequences of the resulting peptide fragments of PLA₂ digested with and without endoproteinase Asp-N or lysyl endopeptidase. Fig. 2 shows the N-terminal sequences of the undigested protein and the resulting peptide fragments used for the design of mixed probes. Four kinds of mixed probes (probes 1-4) were synthesized, as shown in Fig. 2. The N-terminal sequence of the undigested protein was used for the design of probe 4. A genomic library of the S. violaceoruber, prepared in E. coli vector pUC118, was screened with probe 3, and, by colony hybridization, a candidate clone was obtained from a population of about 10,000 colonies. A partial restriction map of a 2.5-kb insert of the Streptomyces DNA contained in a plasmid having a size of 5.7 kb, designated pPLA51, which had been isolated from this clone, was determined (Fig. 3a). Probes 1, 2, and 4 also hybridized to the BglI-SphI DNA fragment shortened to 0.5 kb, as did probe 3, suggesting that at least a part of the PLA₂ structural gene is present within the 0.5-kb fragment.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   N-terminal sequence of the mature S. violaceoruber PLA2 and the peptide fragments isolated after digestion of the protein with endoproteinase. Asp-N or lysyl endopeptidase and design of the mixed probes to clone a PLA2-encoding gene. The amino acid sequence used for the design of probe 4 shows the N terminus of the mature PLA2.

View larger version (54K): [in this window] [in a new window]   Fig. 3.   a, a partial restriction endonuclease map of the 2. 5-kb cloned Streptomyces DNA fragment carrying the PLA2 gene in the plasmid pPLA51. b, the nucleotide sequence of the 0.5-kb fragment containing the PLA2 gene. The consensus Shine-Dalgarno sequence is indicated by double underlining. The putative signal sequence necessary for the secretion of the enzyme is underlined. The N-terminal amino acid sequence of the PLA2, determined by the Edman degradation method, is shown by a dotted line. Amino acids that were not determined by the Edman degradation method are shown without the dotted line. Cys residues, which form disulfide bonds, are enclosed by a circle. An arrow indicates a putative cleavage site for the Streptomyces signal peptidase. In this figure, the N-terminal Met residue in the precursor PLA2, which contains the signal peptide sequence, was numbered as 1. The nucleotide sequence data reported here have been deposited in the DDBJ/EMBL/GenBankTM databases under accession E08479.

The DNA fragment was sequenced by the modified dideoxy chain termination method using 7-deaza-dGTP instead of dGTP (29). Fig. 3b shows the nucleotide sequence of the PLA₂-encoding gene, together with the 5'- and 3'-adjacent regions. An open reading frame consisting of 453 bp starts at GTG (G = +1) and ends at TGA (T = +454) and is predicted to encode a protein with a molecular size of 16,463. Instead of ATG, a codon (GTG) for Val may be used as an initiation codon. The amino acid sequence of the protein deduced from the nucleotide sequence was confirmed by determination of the amino acid sequences of the undigested protein and peptide fragments obtained by digestion with two kinds of peptidase. The dotted lines in Fig. 3b show the amino acid sequence confirmed by the peptide analysis, indicating that we had indeed cloned the PLA₂ gene. The nucleotide sequence for four mixed probes is surely present in the cloned DNA fragment. As shown in Fig. 3b, the N-terminal methionine residue of the precursor PLA₂ containing the signal peptide sequence is numbered as 1, and the amino acid sequence of the undigested PLA₂ reveals that the mature PLA₂ consists of Ala³⁰-Gly¹⁵¹ residues and has a molecular size of 13,558. The value is roughly equal to that estimated by SDS-PAGE and gel filtration chromatography using Sephadex G-100. Therefore, the amino acid sequence from Val¹ to Ala²⁹ is considered to be an essential signal sequence for secreted proteins. The Shine-Dalgarno sequence (GGAGG) (30), found in the E. coli gene, is present at 7 bp upstream from the start codon (GTG) of the PLA₂ gene. Surprisingly, amino acid sequence containing the signal peptide of the S. violaceoruber PLA₂ has 100% identity to that of a hypothetical protein deduced from the genome sequence of S. coelicolor, which is determined by the Sanger Centre and deposited in GenomeNet FASTA server (version 3.4t05, August 18, 2001). The data base also indicates that the amino acid sequence of the S. violaceoruber PLA₂ has about 45% identity to that of a fungal PLA₂ having a molecular mass of 23 kDa (31).

Overproduction of PLA₂ in S. lividans-- The plasmid pPLA51, digested with BamHI, was ligated to BamHI-digested pIJ680, which is a vector for Streptomyces. The chimeric plasmid was introduced into protoplasts from S. lividans TK24 and spread on a regeneration medium containing thiostrepton. The recombinant plasmid, isolated from a thiostrepton-resistant transformant, was designated pPLA101. S. livisans TK24 harboring pPLA101 was grown at 28 °C for 3 days at a small scale of 100 ml in Tryptic soy broth supplemented with 10 µg/ml thiostrepton. By centrifugation at 13,000 × g for 10 min, the supernatant fluid and mycelium were separated. The mycelium was washed with a 10 mM Tris-HCl buffer (pH 8.0), dissolved with the same buffer, and sonicated. The high activity (1.04 units/ml) of PLA₂ was observed in the supernatant fluid but scarcely (0.07 units/ml) in the cell-free extract. Judging from the fact that the PLA₂-encoding gene was inserted at the opposite direction into a Streptomyces vector pIJ680, the gene must be transcribed via its own promoter but not via read-through from the vector. We confirmed that S. lividans carrying the cloned-PLA₂ extracellularly produced at least a 20-fold higher PLA₂ than the original S. violaceoruver. In this study, the enzyme produced by S. lividans TK24 harboring pPLA101 is called "recombinant PLA₂ " in the following description. The characteristics of the recombinant PLA₂, such as molecular weight, isoelectric point, thermostability, and substrate specificity, were the same as those of the S. violaceoruber PLA₂.

Properties of the Recombinant PLA₂-- For product determination of the enzyme-catalyzed hydrolysis, we conducted the reaction of 1,2-dipalmitoyl-L--phosphatidylcholine (37 mg) and the recombinant PLA₂ (~0.1 mg) in aqueous 5 mM CaCl₂ (5 ml) at 25 °C and pH 8.0. After 5 h, one equivalent amount of OH (i.e. 0.1 M NaOH, 0.5 ml) was consumed to keep the pH. The solvent was evaporated, and then the residue was dissolved in CD₃OD (2 ml). The methanol-soluble product was identified by ¹H NMR to be 1-palmitoyl-L--lysophosphatidylcholine without any other compound, such as 2-palmitoyl analogue. These facts show that the enzyme promotes selective hydrolysis at the 2-acyl ester bond to form the corresponding lysophosphatidylcholine (see Fig. 1).

We determined the calcium(II) dissociation constant, K_d, of the recombinant PLA₂ using PC as a substrate. The calcium(II) concentration increases as the PC hydrolysis rate (V) increases (see "Experimental Procedures"). A Lineweaver-Burk plot of 1/V against 1/[Ca²⁺] gives a straight line (Fig. 4), indicating 1:1 calcium (II) complexation at the active center of the PLA₂. From the x-intercept (1/K_d) of the line, the K_d values at 25 and at 35 °C were estimated to be 4.5 ± 0.2 and 5.3 ± 0.2 mM, respectively. The calcium(II) dissociation constant of the S. violaceoruber PLA₂ is much larger than those of eukaryotic PLA₂s (e.g. 0.25 mM for the pancreatic PLA₂ (32) and 0.13 mM for cobra venom PLA₂s (33). This fact indicates that the prokaryotic PLA₂ is a novel enzyme in the PLA₂ superfamily.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   A typical plot of 1/V values (s) as a function of 1/[Ca2+] at 35 °C.

NMR Structure of the Calcium-bound Prokaryotic PLA₂-- The primary structure of the Streptomyces mature PLA₂, except the residues from Cys⁶¹ to Thr⁶⁸, was different from those of eukaryotic secreted PLA₂s (5). In order to investigate the three-dimensional structural difference between prokaryotic and eukaryotic PLA₂s, we determined the NMR structure of the calcium-bound prokaryotic PLA₂ in aqueous solution.

A total of 30 DG structures for the calcium-bound PLA₂ were calculated without the constraints of calcium(II) ion. After inspection of the 30 DG structures for chirality, distance, and dihedral angle violations, 28 structures were refined. The obtained 28 structures displayed distance violations of less than 0.7 Å or  angle violations of less than 20°. At the refinement stage for the final structure determination, we adopted the constrains of 2.0 - 4.0 Å for the coordination bonds between calcium(II) ion and the carboxylates of Asp⁴³ and Asp⁶⁵. After analysis for low total energies and violation energies, a final set of 15 refined structures for the calcium-bound PLA₂ was obtained. The superposition of backbone atoms of the 15 refined structures is shown in Fig. 5. An average structure of the 15 refined structures is displayed in Fig. 6. The root mean square difference between the 15 conformers and the averaged structure is 0.65 ± 0.08 Å for the backbone atoms and 1.15 ± 0.07 Å for all atoms. When the disordered N- and C-terminal residues, Ala¹, Pro², Ala³, Ile¹²⁰, Phe¹²¹, and Gly¹²², were not considered for the calculation, the root mean square values decreased to 0.48 ± 0.06 Å for the backbone atoms and 0.89 ± 0.04 Å for all atoms. The average of the distance constraint violations for the 15 refined structures was smaller than 0.5 Å. The average of the five  angle torsion constraint violations for the 15 refined structures was smaller than 6°.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Stereoviews showing best fit superpositions of the backbone atoms (N, C, C') for the 15 refined distance geometry structures of calcium-bound PLA2.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Stereoview of the final NMR structure of the calcium-bound S. violaceoruber PLA2. Calcium(II) ion is shown as a red ball. Two disulfide bonds, the Asp85His64 pair, and the Arg69 residue are colored in yellow, cyan, and magenta, respectively. The N-terminal Ala residue in the mature PLA2 is numbered as 1.

The sterochemical structure of the amino acids and the dihedral angles in the 15 refined structures were checked with PROCHECK (34). A Ramachandran plot of the average structure of the calcium-bound PLA₂ is shown in Fig. 7. The backbone dihedral angles, except for Leu⁴⁴, fall within the core regions of the plot. This fact supports the idea that the carbonyl oxygen of Leu⁴⁴ coordinates to calcium(II) ion, and the backbone is thus twisted toward Asp⁶⁵. The structure around the calcium(II) ion in the calcium-bound PLA₂ is displayed in Fig. 8a.

View larger version (27K): [in this window] [in a new window]   Fig. 7.   A Rmamachandran / plot for the final NMR structure of the calcium-bound S. violaceoruber PLA2 by using the program PROCHECK (34). Data for Gly and the other residues are given as open squares and closed circles, respectively.

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Stereoview (a) and geometry change (b) of the calcium-binding site in the S. violaceoruber PLA2. The calcium-free and calcium-bound forms are displayed in yellow and cyan, respectively. The calcium(II) ion is shown as a red ball. The N-terminal Ala residue in the mature PLA2 is numbered as 1.

All class I/II PLA₂s contain about 50% -helix and an anti-parallel -sheet, whereas the prokaryotic PLA₂ consists of only five -helices (₁, ₂, ₃, ₄, and ₅) without a -sheet (see Fig. 6). The structure of the prokaryotic PLA₂ is divided into two parts; one is the N-terminal domain including two -helices (₁ and ₂), and the other is the C-terminal domain including three -helices (₃, ₄, and ₅). A long loop composed of the residues from Glu³⁷ to Phe⁵⁷ connects the two helical domains. The direction of the ₄-helix is anti-parallel to those of the ₃-helix and the ₅-helix. The ₃-helix is perpendicular to the ₂-helix. PLA₂s from eukaryotic sources have 6-8 disulfide bonds, whereas the S. violaceoruber PLA₂ has only four cysteine residues (Fig. 3b), suggesting two disulfide bonds. In fact, the NMR structure of the prokaryotic PLA₂ discloses the two disulfide bonds of Cys⁴¹Cys⁶⁵ and Cys⁹⁶Cys¹⁰⁷ (see Fig. 6). It is noteworthy that two disulfide bonds are enough to maintain the conformational rigidity for the enzyme activity.

Comparison of the Calcium-free Form with the Calcium-bound Form of PLA₂-- An almost identical NMR analysis of the bacterial calcium-free PLA₂ gave an NMR structure similar to that of the calcium-bound PLA₂. Some proton signals assigned to the Asp⁴³, Leu⁴⁴, Cys⁶¹, and Asp⁶⁵ residues in the calcium-free PLA₂ shifted to the higher field compared with those for the calcium-bound form. This observation supports the idea that calcium(II) ion stays in the vicinity of these amino acids, which is consistent with the NMR structure for the calcium-bound PLA₂ (see Fig. 8). A remarkable structural difference between the calcium-free and calcium-bound PLA₂ is seen in a -turn region composed of the Cys⁴⁵Ala⁴⁸ residues. The turn region for the calcium-bound PLA₂ is more rigid compared with the calcium-free form, which is due to the calcium(II) complexation.

Another structural difference is in the configuration of Asp⁴³ in the calcium-free form; the Asp⁴³ carboxylic group makes an electrostatic interaction with the Arg⁶⁹ guanidine group. When the calcium ion approaches the calcium-binding site, a hydrogen bond between Arg⁶⁹ and Asp⁴³ dissociates, and four oxygen donors (Asp⁴³, Asp⁶⁵, and Leu⁴⁴) simultaneously associate with the calcium(II) ion. The resulting free guanidyl cation of Arg⁶⁹ may assist the binding with the anionic phosphodiester group of PC (a substrate). Fig. 8b shows the geometry change of the calcium-binding site in the PLA₂. The carboxylate coordination modes of Asp⁴³ and Asp⁶⁵ are monodentate and bidentate, respectively. The coordination polyhedron is completed by three water molecules and by the peptide-carbonyl group of Leu⁴⁴, resulting in a coordination number of 7. The estimated bond lengths for CaO(Asp⁴³), CaO(Asp⁶⁵), and CaO(Leu⁴⁴) are 2.3, 2.5, and 3.5 Å, respectively. A charge-relay system, His⁶⁴ and Asp⁸⁵ residues linked by a hydrogen bond, is close to the water coordination site, which would be in the substrate-binding pocket (see Fig. 6). A nitrogen atom of His⁶⁴, which may act as a general base-catalyst, is 5.5 Å apart from the calcium(II) ion. Similar alignments of imidazole and carboxylate groups are well known in the active center of the other calcium-dependent PLA₂s, such as porcine pancreatic PLA₂ (12).

The solution analysis concludes that the S. violaceoruber PLA₂ has a strikingly novel folding topology. Searching for structures with a fold similar to that of the S. violaceoruber PLA₂ in the DALI data base (35) showed that no similar protein structures were found. The evolutionary origin of S. violaceoruber PLA₂ appears to be different from not only that of eukaryotic PLA₂s but also that of all other proteins registered in the Protein Data Bank. Other research groups have proposed that the PLA₂s discovered until now can be divided into twelve groups, designated groups I-XII (21, 22). In agreement with their proposal, we can propose a new group for the S. violaceoruber PLA₂.

The fatty acids released by PLA₂ can be important as stores of energy, but, more importantly, arachidonic acid can also function as a second messenger and the precursor of eicosanoids, which are potent mediators of inflammation and signal transduction. What role does the S. violaceoruber PLA₂, which is the secreted enzyme, play in the pathogenesis of the bacterium? As shown in this study, the most preferred substrate for the secreted bacterial PLA₂ is PC. In addition, the high concentration of the calcium(II) ion (K_d = ~5 mM) is necessary for expression of the enzyme activity. Furthermore, Streptomyces cells do not have PC as the membrane component (36). A human PLA₂, which is contained in platelets and enriched in rheumatoid synovial fluid, shows a striking preference for phospholipids presented in the E. coli membranes (37). These facts suggest that S. violaceoruber cells must be protected from the digestive effect of their own PLA₂. There is a possibility that the Streptomyces cells might produce the secreted PLA₂ to take ecological priority among microorganisms living in a soil. In other words, the bacterial PLA₂ might function as a toxin to kill microorganisms having phosphatidylcholine as a membrane component.

	FOOTNOTES

^* We have applied for a patent for S. violaceoruber and S. lividans carrying the cloned PLA₂ gene with the Patent Office, Japan, as an invention entitled "Essentially Pure Microorganisms That Produce PLA₂."The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number(s) E08479.

The atomic coordinates and the structure factors (code 1lT4 and 1lT5) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

^§ To whom correspondence should be addressed. Tel.: 81-82-257-5280; Fax: 81-82-257-5284; E-mail: sugi@hiroshima-u.ac.jp.

Present address: Dept. of Chemistry and Material Engineering, Ibaraki National College of Technology, Nakane 866, Hitachinaka, Ibaraki 312-8508, Japan.

Published, JBC Papers in Press, March 15, 2002, DOI 10.1074/jbc.M200264200

	ABBREVIATIONS

The abbreviations used are: PLA₂ phospholipase A₂, DG, distance geometry; EPPS, 3-[4-(2-hydroxyethyl)-1-piperazinyl]-propanesulfonic acid; NOESY, nuclear Overhauser effect correlation spectroscopy; PC, phosphatidylcholine; PIPES, piparazine-1,4-bis(2-ethanesulfonic acid); TOCSY, total correlation spectroscopy; HPLC, high pressure liquid chromatography.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. Guillaume, C. Calzada, M. Lagarde, J. Schrevel, and C. Deregnaucourt Interplay between lipoproteins and bee venom phospholipase A2 in relation to their anti-plasmodium toxicity J. Lipid Res., July 1, 2006; 47(7): 1493 - 1506. [Abstract] [Full Text] [PDF]

				M. J. Nagiec, B. Lei, S. K. Parker, M. L. Vasil, M. Matsumoto, R. M. Ireland, S. B. Beres, N. P. Hoe, and J. M. Musser Analysis of a Novel Prophage-encoded Group A Streptococcus Extracellular Phospholipase A2 J. Biol. Chem., October 29, 2004; 279(44): 45909 - 45918. [Abstract] [Full Text] [PDF]

				Y. Matoba, Y. Katsube, and M. Sugiyama The Crystal Structure of Prokaryotic Phospholipase A2 J. Biol. Chem., May 24, 2002; 277(22): 20059 - 20069. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) An addition or correction has been published All Versions of this Article: 277/22/20051    most recent M200264200v1 Alert me when this article is cited