Purified Group X Secretory Phospholipase A2 Induced Prominent Release of Arachidonic Acid from Human Myeloid Leukemia Cells*

Group X secretory phospholipase A2 (sPLA2-X) possesses several structural features characteristic of both group IB and IIA sPLA2s (sPLA2-IB and -IIA) and is postulated to be involved in inflammatory responses owing to its restricted expression in the spleen and thymus. Here, we report the purification of human recombinant COOH-terminal His-tagged sPLA2-X, the preparation of its antibody, and the purification of native sPLA2-X. The affinity-purified sPLA2-X protein migrated as various molecular species of 13–18 kDa on SDS-polyacrylamide gels, andN-glycosidase F treatment caused shifts to the 13- and 14-kDa bands. NH2-terminal amino acid sequencing analysis revealed that the 13-kDa form is a putative mature sPLA2-X and the 14-kDa protein possesses a propeptide of 11 amino acid residues attached at the NH2 termini of the mature protein. Separation with reverse-phase high performance liquid chromatography revealed that N-linked carbohydrates are not required for the enzymatic activity and pro-sPLA2-X has a relatively weak potency compared with the mature protein. The mature sPLA2-X induced the release of arachidonic acid from phosphatidylcholine more efficiently than other human sPLA2groups (IB, IIA, IID, and V) and elicited a prompt and marked release of arachidonic acid from human monocytic THP-1 cells compared with sPLA2-IB and -IIA with concomitant production of prostaglandin E2. A prominent release of arachidonic acid was also observed in sPLA2-X-treated human U937 and HL60 cells. Immunohistochemical analysis of human lung preparations revealed its expression in alveolar epithelial cells. These results indicate that human sPLA2-X is a unique N-glycosylated sPLA2 that releases arachidonic acid from human myeloid leukemia cells more efficiently than sPLA2-IB and -IIA.

Low molecular mass sPLA 2 s (13-18 kDa) have several features including a high disulfide bond content, a requirement for millimolar concentrations of Ca 2ϩ for catalysis, and a broad specificity for phospholipids with different polar head groups and fatty acyl chains (15,16). Mammalian sPLA 2 s are classified into different groups depending on the primary structure characterized by the number and positions of cysteine residues. At present, five types of functional sPLA 2 s (group IB, IIA, IID, V, and X) have been identified in humans (10,15), whereas group IIC sPLA 2 found in rodents is a pseudogene in humans (17). Among them, group IIA sPLA 2 (sPLA 2 -IIA) is thought to play a pivotal role in the progression of inflammatory conditions, since its local and systemic levels are elevated in numerous inflammatory diseases (18,19). However, some inbred mouse strains have a natural frameshift mutation in the sPLA 2 -IIA gene (20,21). The phenotype of these deficient mice is similar to that of sPLA 2 -IIA-expressing mouse strains in their responses to various inflammatory challenges that initiate arthritis (22,23). In addition, we have recently shown that indoxam, one of the potent sPLA 2 inhibitors (24), suppressed the endotoxin-induced elevation of plasma tumor necrosis factor-␣ levels, with a similar potency for sPLA 2 -IIAexpressing and sPLA 2 -IIA-deficient mouse strains (25). Transgenic mice expressing the human sPLA 2 -IIA gene do not develop any overt inflammatory conditions (26). These findings point to the need to reevaluate the contribution of sPLA 2 -IIA in inflammatory diseases and suggest that other types of sPLA 2 may play pivotal roles in place of or in concert with sPLA 2 -IIA.
Group IB sPLA 2 (sPLA 2 -IB) has been thought to act as a digestive enzyme, given its abundance in digestive organs (27). * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
However, a series of our studies have identified a variety of biological responses induced by sPLA 2 -IB via binding to its specific receptor, the PLA 2 -receptor (PLA 2 R) (28 -31). Recent studies with PLA 2 R-deficient mice have demonstrated its potential role in the production of inflammatory cytokines during the progression of endotoxic shock (32). Group V sPLA 2 (sPLA 2 -V) has been reported to be involved in the release of lipid mediators in P388D 1 murine macrophages and mouse bone marrow-derived mast cells based on antisense experiments (33,34). Recent studies have shown that this type of sPLA 2 hydrolyzes phosphatidylcholine (PC) more effectively than sPLA 2 -IIA (35). We have recently cloned a cDNA encoding a novel type of sPLA 2 , termed IID (sPLA 2 -IID) (10). The sPLA 2 -IID is most similar to sPLA 2 -IIA with respect to the number and positions of cysteine residues as well as overall sequence identity (10,11). The expression of its mRNA dramatically changes upon challenge with endotoxin in rats and sPLA 2 -IIAdeficient mice, suggesting its potential role in the progression of inflammatory processes.
Human group X sPLA 2 (sPLA 2 -X) has been cloned from fetal lung based on sPLA 2 -related sequences from DNA data bases (9). The sPLA 2 -X cDNA clone suggests a mature sPLA 2 protein of 123 amino acids and the presence of a signal peptide with 32 amino acids. The presence of an arginine doublet and other polar residues preceding the mature sPLA 2 -X protein indicates that the signal sequence is a prepropeptide, although there is no biochemical evidence for the cleavage sites. sPLA 2 -X is the most acidic (pI 5.3) among human sPLA 2 s thus far identified and contains one potential glycosylation site. This sPLA 2 possesses 16 cysteine residues located at positions characteristic of both sPLA 2 -IB and sPLA 2 -IIA/IID and also has an amino acid COOH-terminal extension that is typical of sPLA 2 -IIA and -IID. A 1.5-kilobase transcript coding for sPLA 2 -X was detected in human spleen and thymus, suggesting its potential role in the immune system and/or inflammation (9,16), although there are no data available with respect to its functional role in physiological and pathological conditions.
To define the biochemical properties of human sPLA 2 -X, we prepared the antibody (Ab) and purified the protein. We found that human sPLA 2 -X releases arachidonic acid from PC more efficiently than other sPLA 2 groups and also induces a rapid and marked release of arachidonic acid from several human myeloid leukemia cells compared with sPLA 2 -IB and -IIA. We also present evidence for the expression of sPLA 2 -X in human alveolar epithelial cells by immunohistochemical analysis.
Preparation of Human sPLA 2 -X, sPLA 2 -X-HisTag, sPLA 2 -V, and sPLA 2 -IID cDNAs-The coding region of human sPLA 2 -X cDNA was obtained by polymerase chain reaction (PCR) using human placenta cDNA as a template. The sequences of the upstream and downstream primers were: hGX-S, (5Ј-CTGTGTACGCGTCCACCATGCTGCTCCT-GCTACTGCCGTC-3Ј; and hGX-AS, 5Ј-TCAAGTGCGGCCGCTCAGTC-ACACTTGGGCGAGTCC-3Ј, respectively. hGX-S had a Kozak sequence and an MluI recognition site. hGX-AS possessed a NotI recognition site. The PCR conditions were 94°C for 30 s, 50°C for 30 s, and 72°C for 2 min, for 35 cycles. The PCR-amplified fragment was digested with MluI and NotI followed by the insertion into the modified pBS-SK(Ϫ). After sequencing confirmation, COOH-terminal His-tagged sPLA 2 -X (sPLA-2 -X-HisTag) was constructed by PCR using hGX-S primer and hGX-H6AS primer: 5Ј-TCAAGTGCGGCCGCTCAATGGTGATGGTGATGA-TGGTCACACTTGGGCGAGTCCGGCT-3Ј (the underlined sequence corresponds to the His tag). The PCR-amplified fragment was digested with NotI and XhoI, which had a recognition site in the coding region of sPLA 2 -X followed by exchange of the corresponding region in the native sPLA 2 -X plasmid. The sequence of the PCR-amplified region was confirmed, and the cDNAs were inserted into the mammalian cell expression vector under SR-␣ promoter.
The sPLA 2 -V cDNA was amplified by PCR from human heart cDNA as a template using the primers 5Ј-AAAGAACGCGTCCACCATGAAA-GGCCTCCTCCCACTGGCT-3Ј and 5Ј-CTCGCTGCGGCCGCCTAGGA-GCAGAGGATGTTGGGAAA-3Ј. The upstream primer had a Kozak sequence and an MluI recognition site. The downstream primer possessed a NotI recognition site. The amplified fragment was digested with MluI and NotI followed by subcloning in the modified pBS-SK(Ϫ). The sPLA 2 -V expression plasmid was constructed by the same method as described above. The preparation of sPLA 2 -IID cDNA and its expression plasmid were performed as described previously (10).
Chromogenic PLA 2 Assay-Spectrometric PLA 2 assay was performed according to the method of Reynolds et al. (36). Briefly, 180 l of the reaction mixture solution containing 1 mM diheptanoylthio-PC, 0.3 mM Triton X-100, 0.12 mM 5,5Ј-dithiobis(2-nitrobenzoic acid), 10 mM CaCl 2 , 0.1 M KCl, and 0.1% BSA in 25 mM Tris-HCl buffer (pH 7.5) was preincubated in a 96-well plate for 15 min at 37°C. The reactions were initiated by the addition of 20 l of the enzyme preparation and continued for an appropriate time at 37°C. The reaction was monitored by the absorbance at 405 nm.
Recombinant Expression, Purification, and Characterization of Human sPLA 2 -X-HisTag Protein-Recombinant plasmid containing sPLA 2 -X-HisTag and 0.5 g of hygromycin B-resistant gene were cotransfected into SV40-transformed human embryonic kidney 293 (293T) cells with LipofectAMINE reagent (Life Technologies, Inc.), and the stably expressing clones were generated by selection against hygromycin B (250 g/ml). Expression of the recombinant protein was determined with the chromogenic PLA 2 assay. The conditioned medium (10% fetal calf serum (FCS)) was applied to a Ni 2ϩ -charged chelating Sepharose fast flow column (Amersham Pharmacia Biotech) equilibrated with 10 mM imidazole, 0.5 M NaCl, 20 mM sodium phosphate buffer (pH 7.4), and the bound materials were eluted with 500 mM imidazole, 0.5 M NaCl, 20 mM sodium phosphate buffer (pH 7.4). Peak fractions of PLA 2 activity were dialyzed against 1 mM phenylmethylsulfonyl fluoride, 20 mM Tris-HCl (pH 7.4) and loaded on a HiTrap Q column (Amersham Pharmacia Biotech). The bound proteins were eluted with a gradient of NaCl from 0 to 0.5 M. The sPLA 2 -X-HisTag fractions was then applied to a reverse-phase HPLC column (Cosmosil, 5C18 300AR, 4.6 ϫ 150 mm, Nacarai Tesque, Japan) with a gradient of acetonitrile from 20 to 95% in 0.05% trifluoroacetic acid. The chromogenic PLA 2 assay revealed that sPLA 2 -X-HisTag was eluted at 43% acetonitrile. The eluted materials were analyzed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) in 15-25% acrylamide gradient gels (Daiichi Chemicals, Co.). In separate experiments, the purified protein was treated with 0.2 unit of N-glycosidase F (E-5003; Oxford GlycoSystems) in 0.5% SDS, 5% 2-mercaptoethanol, 3% Nonidet P-40 for 18 h at 37°C. After SDS-PAGE, electroblotting of proteins to an Immobilion-P membrane (Millipore Co.) was performed as described previously (37). After staining with Coomassie Brilliant Blue, the protein bands were excised, and NH 2 -terminal amino acid sequencing was performed with an Applied Biosystems Procise Sequencer.
Recombinant Expression of Human sPLA 2 -V and sPLA 2 -IID-Recombinant plasmid containing sPLA 2 -V or sPLA 2 -IID was transfected into CHO cells with LipofectAMINE reagent, and stably expressing clones were generated by selection against G418 (1 mg/ml; Life Technologies, Inc.). In the case of sPLA 2 -IID, the recombinant enzyme was partially purified with a heparin-Sepharose column (Amersham Pharmacia Biotech), as described previously (10).
Preparation and Characterization of Anti-sPLA 2 -X Ab-After immunization of rabbits with purified sPLA 2 -X-HisTag (200 g), the antiserum was prepared, and the IgG fractions were purified with a HiTrap protein G-column (Amersham Pharmacia Biotech). The specificity and titer of the anti-serum and purified Ab were determined by enzymelinked immunosorbent assay (ELISA) as follows. The 96-well plates (Nunc, Immuno MaxiSorp plate) coated with purified sPLA 2 proteins (100 ng) were blocked with Block-Ace solution (Dainippon Pharmaceuticals Co.) and incubated with anti-sPLA 2 -X Ab, anti-human sPLA 2 -IB Ab (Seikagaku Corp.), and anti-human sPLA 2 -IIA serum (Cayman Chemical Co.) for 2 h. After washing with PBS, 0.05% Tween 20, the plates were incubated with peroxidase-conjugated goat anti-rabbit IgG (Immunotech) followed by the addition of 0.5 mg/ml o-phenylenediamine in 50 mM sodium citrate buffer (pH 5.3) containing 0.01% (v/v) H 2 O 2 . The reaction was stopped by addition of 4 N H 2 SO 4 , and the plates were read at 492 nm. In separate experiments, the plates were coated with sPLA 2 -X-HisTag or Axl-HisTag proteins and assayed as described above.
The inhibitory effects of anti-sPLA 2 -X Ab on PLA 2 activity were examined with the chromogenic assay. Purified sPLA 2 proteins (50 ng), partially purified sPLA 2 -IID, and the conditioned medium from the sPLA 2 -V-expressing CHO cells were incubated with anti-sPLA 2 -X Ab for 2 h and then added to the reaction mixture solution, as described above. The inhibitory potency of anti-sPLA 2 -X Ab was examined at the times when each sPLA 2 reaction reached the sub-maximum at 37°C.
Recombinant Expression, Purification, and Characterization of Human sPLA 2 -X Protein-Transfection with sPLA 2 -X expression plasmid and preparation of stably expressing clones were performed as described above, and the established CHO cells were grown in 10% FCS, Dulbecco's modified Eagle's medium. After reaching confluence, the medium was changed to serum-free PM-1000 medium (Eiken Co., Japan), and the cells were incubated for 3 days. The conditioned medium was then applied to the anti-sPLA 2 -X Ab-conjugated affinity column, which had been prepared by coupling of anti-sPLA 2 -X IgG (30 mg) to HiTrap N-hydroxysuccinimide-activated column (5 ml; Amersham Pharmacia Biotech) according to the manufacturer's instructions. After washing with 40 mM sodium phosphate buffer (pH 7.4), the bound protein was eluted with 0.1 M glycine-HCl (pH 1.7). For further separation, the affinity-purified materials were applied to a reverse-phase HPLC column with a gradient of 20 to 40% acetonitrile in 0.1% trifluoroacetic acid. The three major fractions were collected, dried, and dissolved in 50 mM Tris-HCl (pH 7.4) containing 0.3 M NaCl.
PLA 2 Assay for Substrate Specificity-Human sPLA 2 s were subjected to individual reactions with 12 types of commercially available phospholipids as a substrate. These sPLA 2 materials were also subjected to reactions with the mixed phospholipids composed of four types of PCs or three types of phosphatidylethanolamines (PEs) as the substrates. The enzymatic activity was measured using mixed micelles of 3 mM sodium deoxycholate (DOC) and 1 mM of each substrate or micelles of 3 mM DOC and mixed PCs or PEs consisting of 0.25 mM concentrations of each substrate in a total volume of 100 l. The assay mixture contained 10 mM CaCl 2 , 1 mg/ml BSA, 150 mM NaCl, and 100 mM Tris-HCl (pH 8.0). After the incubation at 40°C, the reaction was stopped by the addition of Dole's reagent (heptane, 2-propanol, 2 N sulfuric acid ϭ 10:40:1, v/v/v), and the released fatty acids were quantified according to the method of Tojo et al. (38).
Assay for the Release of Fatty Acids, PGE 2 , and LTC 4 from THP-1 Cells-Human THP-1 cells were grown in RPMI 1640, 10% FCS and pretreated with 1.3% dimethyl sulfoxide for 2 days before the assay (39). After washing with PBS, the cells were suspended in Hanks' buffered saline (pH 7.6) containing 0.1% BSA at a density of 5.56 ϫ 10 6 cells/ml. Aliquots of cell suspension (0.45 ml) were preincubated for 15 min at 37°C and stimulated with sPLA 2 proteins (0.05 ml). To analyze the released fatty acids, the reaction was stopped with 2 ml of Dole's reagent, and 3 nmol of margaric acid (Nu-Chek-Prep, Inc.) was added as an internal standard. After the addition of 1.2 ml of heptane and 0.8 ml of water, the heptane phase was removed and dried in vacuo, and the fatty acids in this fraction were labeled with 9-anthryldiazomethane (Funakoshi Co.), as described by Tojo et al. (38). The sample was analyzed by reverse-phase HPLC on a LiChroCART 125-4 Superspher 100 RP-18 column (Merck) with acetonitrile/water (97:3, v/v) as the solvent. The column oven temperature was 26°C, and the column effluents were monitored by the fluorescence intensity at the wavelengths of excitation and emission of 365 and 412 nm, respectively. To measure PGE 2 and LTC 4 , the reaction was stopped by cooling on ice. The supernatant was then collected after centrifugation at 1,000 ϫ g for 5 min at 4°C, and the released PGE 2 and LTC 4 were quantified with a radioimmunoassay kit (NEN Life Science Products) and an ELISA kit (Cayman Chemicals Co.), respectively. Assay for [ 3 H]Arachidonic Acid Release from U937 and HL60 Cells-Human U937 and HL60 cells (5 ϫ 10 5 cells/ml) grown in 10% FCS, RPMI 1640 were labeled with [ 3 H]arachidonic acid (0.5 Ci/ml; Amersham Pharmacia Biotech) for 18 h. After washing three times, the cells were resuspended in 10% FCS, RPMI 1640. Aliquots of cell suspension (5 ϫ 10 5 cells) were stimulated with sPLA 2 proteins in a total volume of 400 l at 37°C. After the reaction, the supernatant was collected by centrifugation at 3,000 rpm for 2 min at 4°C, and the released radioactivity in the supernatant (100 l) was counted. Total incorporated radioactivity into the cells used for one assay was 1.04 ϫ 10 6 and 1.02 ϫ 10 6 dpm in U937 and HL60 cells, respectively.
Immunostaining of Human Lung Preparations with Anti-sPLA 2 -X Ab-Preparations of normal adult human lung tissue were purchased from Novagen Inc. (Madison, WI). The tissue slides were dewaxed, incubated in methanol containing 0.3% H 2 O 2 for 30 min, and then treated with 5% normal rabbit serum for 20 min. The slides were incubated with anti-sPLA 2 -X Ab (6 g/ml) in PBS containing 0.1% BSA for 14 h at 4°C. After washing with PBS, they were incubated with biotin-conjugated goat anti-rabbit IgG for 30 min followed by treatment with horseradish peroxidase avidin-biotin complex reagent (Vector Laboratories). After washing, the peroxidase activity was visualized with 10-min incubation in 50 mM Tris-HCl (pH 7.6) containing 200 g/ml diaminobenzidine and 0.006% H 2 O 2 . After the nuclei were counterstained with 1% methyl green in 0.1 M sodium acetate (pH 4.0), the coverslips were placed on the preparations in aqueous mounting medium. The sPLA 2 -X positive signals were detected as diaminobenzidine deposits of dark brownish color. Neutralization of sPLA 2 -X specific signals was performed by incubating anti-sPLA 2 -X Ab with affinitypurified sPLA 2 -X protein (60 g/ml) for 2 h before the addition to the slides.

RESULTS
Preparation and Characterization of Anti-sPLA 2 -X Ab-As a prompt approach to preparing the antigen, human sPLA 2 -X was expressed in 293T cells as a COOH-terminal His-tagged form that retained PLA 2 activity. The sPLA 2 -X-HisTag protein in the conditioned medium was purified with three steps of Ni 2ϩ -affinity column, HiTrap Q column, and a reverse-phase HPLC. As shown in Fig. 1A, SDS-PAGE analysis of the final preparations revealed the presence of a major component with an apparent molecular mass of 16 kDa as well as a minor component of 14 kDa. Treatment with N-glycosidase F, which releases N-linked oligosaccharides, shifted the 16-kDa band to a 14-kDa single band. The amino acid sequence analysis of the 14-kDa polypeptide revealed that the NH 2 -terminal sequence (GILELAGT) precisely matched the putative mature form of human sPLA 2 -X (9). In the sequence of human sPLA 2 -X, there is one potential N-glycosylation site at position 71 (9). Thus, these findings demonstrate that the sPLA 2 -X-HisTag protein is largely expressed as an N-glycosylated mature form in 293T cells.
Using purified sPLA 2 -X-HisTag protein as an antigen, rabbits were immunized, and IgG fractions in the anti-serum were purified. As shown in Fig. 1B, the purified Ab recognized human sPLA 2 -X-HisTag protein but did not react with the purified sPLA 2 -IB and IIA. Conversely, the commercial Abs specific for human sPLA 2 -IB or IIA were not reactive with sPLA 2 -X-HisTag protein (data not shown). The prepared Ab did not react with unrelated His-tagged proteins including Axl-HisTag (data not shown), demonstrating that the COOH-terminal Histagged portion is not an epitope for this Ab.
The neutralizing effects of anti-sPLA 2 -X Ab on the activities of five types of human sPLA 2 s were examined by the chromogenic assay. The inhibitory potency of the Ab was examined at the times when each reaction reached sub-maximum levels. Pretreatment with anti-sPLA 2 -X Ab led to a dose-dependent and specific inhibition against the sPLA 2 -X-HisTag activity. At 100 g/ml, Ab blocked the enzyme activity by approximately 90% without suppression for the other four types of sPLA 2 activities (data not shown). Taken together, these results demonstrate that the prepared Ab was specific for sPLA 2 -X among the known human sPLA 2 proteins.
Purification and Characterization of Human sPLA 2 -X-Hu-man sPLA 2 -X was stably expressed in CHO cells, and their serum-free conditioned medium was applied to the anti-sPLA 2 -X Ab affinity column. SDS-PAGE analysis of the acideluted materials showed the presence of 15-18-kDa broad bands, 14-kDa and 13-kDa single bands ( Fig. 2A, lane 1). Treatment with N-glycosidase F caused shifts to two bands (14 and 13 kDa) (Fig. 2B, lane 1). The NH 2 -terminal sequence analysis revealed that the 13-kDa major protein (GILELAGT) is a putative mature form of human sPLA 2 -X (9), whereas the 14-kDa minor protein possesses an additional 11 amino acid residues (EASRILRVHRR) at the NH 2 termini of mature protein. Reactivity and susceptibility of affinity-purified sPLA 2 -X with anti-sPLA 2 -X Ab were virtually identical with its His-tagged form (Fig. 1B).
Reverse-phase HPLC of the affinity-purified materials resulted in their separation into three major protein peaks. SDS-PAGE analysis of these fractions without or with N-glycosidase F treatment (Fig. 2, A and B)  Substrate Specificity of Human sPLA 2 -X-The substrate preference was examined with commercially available phospholipids, which possess palmitic acid at the sn-1 position and have different fatty acids at the sn-2 position as well as polar head groups. The hydrolysis rates of sPLA 2 proteins (three HPLC fractions of sPLA 2 -X, sPLA 2 -IB, and sPLA 2 -IIA) were determined in micelle assays containing an individual substrate of 12 types of phospholipids. As shown in Table I, HPLC fraction 1 showed only 30 -40% of the activity compared with fractions 2 and 3. This reduction was conceivably due to the presence of 14-kDa pro-enzyme in fraction 1 that might have no or relatively lower enzymatic activity compared with the mature protein. Absolute activities of fraction 3 were not considerably different from those of fraction 2, and the relative activity toward each phospholipid was essentially identical among the three HPLC fractions, indicating that the N-linked carbohydrates in the mature sPLA 2 -X are not essential for its PLA 2 activity as well as the substrate specificity. Among the tested phospholipids, the mature form of sPLA 2 -X (fractions 2 and 3) possessed the best hydrolysis rates toward PAPC, whereas sPLA 2 -IB and -IIA prefer POPG to other phospholipids as reported previously (40). sPLA 2 -X also possessed higher hydrolysis rates for 2-arachidonyl PE and PA compared with the other two sPLA 2 types, whereas it showed weak hydrolyzing  activity toward POPS and POPA similar to sPLA 2 -IB and -IIA in this assay system. The sPLA 2 -X activity toward POPG was dose-dependently suppressed by the addition of 1-oxamoylindolidine derivative compound, indoxam, with an IC 50 value of 300 nM (data not shown), which also possessed strong inhibitory potency for sPLA 2 -IB and -IIA (IC 50 ϭ 54 and 1.2 nM, respectively) under the same assay conditions (25). Next, five types of human sPLA 2 preparations were subjected to the reactions with the mixed substrates composed of four types of PCs or those of three types of PEs. Since each PLA 2 reaction showed different time dependence and hydrolyzing activity due to differences in substrate specificity and enzyme quantities, the percentages of hydrolyzed PCs and PEs were examined at the time (30 min) when each hydrolysis rate was within the linear range of enzymatic assays. Among the tested PCs, sPLA 2 -X showed a preference for PAPC over the other three PCs, whereas the other four types of sPLA 2 s possessed a lesser preference for this phospholipid species (Fig.  3A). Among the tested PEs, all of the five sPLA 2 groups showed the best preference for PLPE (Fig. 3B). However, sPLA 2 -X showed relatively higher preference for PAPE compared with the other four sPLA 2 s. A similar substrate preference was observed when the N-glycosylated mature form of sPLA 2 -X (fraction 2) was used (data not shown).
Release of Arachidonic Acid from THP-1 Cells by the Action of Human sPLA 2 -X-The preference of sPLA 2 -X for 2-arachidonyl PC suggests its involvement in the release of arachidonic acid from intact cells, because the extracellular face of the plasma membrane of mammalian cells is largely composed of zwitterionic PC and sphingomyelin (41). As the expression of sPLA 2 -X transcript was restricted in the immune tissues (9), several human myeloid leukemia cell lines were used to evaluate the potency of sPLA 2 -X in the release of arachidonic acid. Human monocytic THP-1 cells were incubated with sPLA 2 -IB, -IIA and the nonglycosylated mature form of sPLA 2 -X (fraction 3), and the released fatty acids were quantified by reversephase HPLC analysis. As shown in Fig. 4A, sPLA 2 -X induced a prompt and marked release of arachidonic acid from THP-1 cells compared with sPLA 2 -IB and -IIA. This reaction reached a steady state level within 5 min, and the amount of arachidonic acid released within 10 min was about 12-and 17-fold more than that induced by sPLA 2 -IB and -IIA, respectively. No release of lactate dehydrogenase activity was detected in THP-1 cells during the incubation with these sPLA 2 s for up to 80 min (data not shown). As shown in Fig. 4B, sPLA 2 -X also induced remarkable release of oleic acid in contrast to slight releases by sPLA 2 -IB and -IIA. The oleic acid release by sPLA 2 -X was rather slow compared with the arachidonic acid release, and the maximum quantity of released oleic acids by sPLA 2 -X (2.0 nmol/ml at 20 min) was larger than the amount of released arachidonic acid (0.60 nmol/ml). In contrast, the palmitic acid release was not detected within 10 min for any of three sPLA 2 s groups (data not shown). Fig. 4C shows the dose dependence of sPLA 2 -X in the release of arachidonic acid from THP-1 cells. The release was significant at 5 ng/ml and almost the same level as that induced by 5 g/ml sPLA 2 -IB and -IIA. The arachidonic acid levels released by 5 g/ml sPLA 2 -X (1.14 nmol/ml) were about 20% that of the maximum level induced by 3 M A23187. The N-glycosylated sPLA 2 -X (fraction 2) showed essentially the same responses as nonglycosylated sPLA 2 -X (data not shown). As shown in Fig. 4D, a sPLA 2 inhibitor, indoxam (25), blocked the sPLA 2 -X-induced arachidonic acid releases in a dose-dependent manner with an IC 50 value of 2.5 M, whereas a known cPLA 2 inhibitor, AACOCF 3 (42), did not suppress the responses at up to 100 M. The sPLA 2 -X-induced release reaction was also blocked by anti-sPLA 2 -X Ab (500 g/ml) (data not shown).
As shown in Fig. 5A, the nonglycosylated mature form of sPLA 2 -X (fraction 3) induced a significant production of PGE 2 , which contrasted with little, if any, production by the action of sPLA 2 -IB and -IIA for up to 40 min. The sPLA 2 -X-induced reaction of PGE 2 production was slower compared with that in the arachidonic acid release (Fig. 4A), and reached the maximum level at 20 min. As shown in Fig. 5B, sPLA 2 -X induced PGE 2 production with a similar dose dependence as in the case of the arachidonic acid release (Fig. 4C), whereas sPLA 2 -IB and -IIA evoked a slight PGE 2 production even at 5 g/ml. The amounts of PGE 2 produced by 5 g/ml sPLA 2 -X (91 fmol/ml) was about 10% that of the maximum level induced by 3 M A23187. The sPLA 2 -X-induced PGE 2 production was blocked by pretreatment with anti-sPLA 2 -X Ab (500 g/ml) and also suppressed by the addition of indoxam, which showed an IC 50 value of 0.4 M (data not shown).

Release of [ 3 H]Arachidonic
Acid from U937 and HL60 Cells by Human sPLA 2 -X-The effects of sPLA 2 -X on the arachidonic acid release were further examined in human myelomonocytic U937 cells and human promyelocytic leukemic HL60 cells after labeling of the cells with [ 3 H]arachidonic acid, since the amounts of released fatty acids from these cell types were below the detection range in HPLC analysis. As shown in Fig.  6, sPLA 2 -X (5 g/ml) induced a prompt release of [ 3 H]arachidonic acid from U937 (A) as well as from HL60 cells (B), which contrasted with a slow and little release by sPLA 2 -IB and -IIA. The dose dependence of sPLA 2 -X for the release of [ 3 H]arachidonic acid was almost the same with that in THP-1 cells, and the sPLA 2 -X-induced responses were completely suppressed by pretreatment with anti-sPLA 2 -X Ab (500 g/ml) or by the addition of indoxam (10 M).
Immunostaining of Human Lung Preparations with Anti-sPLA 2 -X Ab-Since sPLA 2 -X was cloned from human fetal lung cDNAs (9), the expression of sPLA 2 -X protein was examined in human lung preparations by immunostaining with anti-sPLA 2 -X Ab. As shown in Fig. 7A, positive signals of sPLA 2 -X were detected in the cytosol of alveolar type II epithelial cells. In contrast, there were no signals in type I squamous alveolar epithelial cells. These signals were specific for sPLA 2 -X, since pretreatment of Ab with a 10-fold excess of sPLA 2 -X protein resulted in abolishment of the signals (Fig. 7B). In addition, these positive signals could not be detected in parallel control samples with nonimmunized rabbit normal IgG (Fig. 7C). DISCUSSION Among the five groups of human sPLA 2 s thus far cloned, the biochemical properties of sPLA 2 -X have been poorly characterized. The present study describes the characteristics of sPLA 2 -X in terms of the structure and enzymatic activity as well as the expression in human lung. The recombinant sPLA 2 -X was released as a N-glycosylated form in the stably transfected 293T and CHO cells (Figs. 1A and 2). In the sequences of human sPLA 2 s, sPLA 2 -IB and -V do not possess any potential glycosylation site, whereas sPLA 2 -IIA, -IID, and -X have one (5,9,10). However, sPLA 2 -IIA is expressed as a single 14-kDa polypeptide with no sugars in human platelets and rheumatoid synovial fluids (6), and the purified sPLA 2 -IIA from the stably expressed CHO cells did not have the carbohydrates. 2 Thus, sPLA 2 -X is the first example of glycosylated mammalian sPLA 2 . In the PLA 2 assays (Table I), the substrate specificity and the hydrolyzing rates were not considerably different between glycosylated and nonglycosylated sPLA 2 -X (fractions 2 and 3). In addition, these two forms showed similar effects on the fatty acid release in THP-1 cells. These findings demonstrate that N-linked sugar chains in sPLA 2 -X are not essential for the PLA 2 activity. Recently, the specific terminal oligosaccharide sequences of glycoproteins have been demonstrated to participate in various biological events, including the clearance from circulation and cell-cell interactions (43), and several types of mammalian lectins, such as the selectins, have been identified in various cell types (44). Although the precise structures of carbohydrates in sPLA 2 -X have not yet been clarified, some lectins in the specialized cell types might recognize its sugar chains and concentrate this sPLA 2 in the cell surfaces to make it accessible to phospholipids or to make it become internalized into the cells. Since sPLA 2 -X was expressed in the transfected cell systems in the present study, its natural status must be examined to speculate its functional significance in vivo.
In the stably transfected CHO cells, a 14-kDa form was secreted as a minor component that possesses 11 additional amino acid residues at the NH 2 terminus of the mature protein.
SignalP computer analysis (45) for the potential cleavage positions in its signal sequence revealed that the most likely cleavage site is present between position Ϫ12 and Ϫ11 preceding the mature protein, thus demonstrating that the 11 amino acid residues are a propeptide. HPLC fraction 1 composed of the mixtures of pro-and mature forms showed considerably weak PLA 2 activity compared with the mature sPLA 2 -X ( Table I), suggesting that the cleavage of the propeptide at the arginine doublet, which is known to be efficiently catalyzed by subtilisin-like endoproteases (46), is critical for eliciting the maximum activity. This coincided with the case of sPLA 2 -IB, in which the proenzyme is inactive, and the release of propeptide by serine proteases such as trypsin and plasmin is required for the display of its PLA 2 activity (47). Further studies on the identification of proteases related to the cleavage reactions should provide a clue to the precise maturation processing mechanisms of sPLA 2 -X.
The substrate preference of sPLA 2 -X, -IB, and -IIA toward PCs and PEs were similar between the assays with a single substrate (Table I) and the mixed substrates (Fig. 3). However, the specific activities of each sPLA 2 were not identical between the two systems, as the sPLA 2 activity is known to be dramatically changed by the bile salt/phospholipid molar ratio (48) and depend on the physical state of the substrate in the micelles, as demonstrated in the case of sPLA 2 -V (49). sPLA 2 -X was found to hydrolyze 2-arachidonyl PC and PE more efficiently than the other types of sPLA 2 s in the PLA 2 assay with the mixed phospholipids and possesses the best hydrolysis rate for PAPC 2 K. Kawamoto, unpublished data. among the 12 types of phospholipids examined. This substrate preference was quite a contrast with that of other sPLA 2 proteins. sPLA 2 -IB and -IIA had the best hydrolysis rate for POPG, one of the phospholipid components of lung surfactants (40). sPLA 2 -IID showed a substrate specificity similar to sPLA 2 -IIA in mixed phospholipid assays but possessed the highest preference for PLPE in the individual phospholipid assay (10). In contrast, sPLA 2 -V had little, if any, potency for hydrolyzing 2-arachidonyl-PC and -PE (Fig. 3). This is consistent with the findings of recent studies, in which sPLA 2 -V showed preferences in the order of PLPC Ͼ PPPC Ͼ PAPC (49). Among the five types of sPLA 2 s, sPLA 2 -X is the most acidic (pI 5.3) and has the largest numbers of cysteine residues (9). Although the rationale of the substrate specificity of sPLA 2 -X remains unclear at present, the substrate recognition mechanism can be addressed by mutation analysis, especially at the acidic and cysteine residues.
In several human myeloid leukemia cells, sPLA 2 -X induced a rapid and marked release of arachidonic acid. Since the availability of cell membrane phospholipids as the substrates for PLA 2 is limited, the sPLA 2 -X-induced arachidonic acid release could become saturable within 5 to 20 min. Alternatively, a portion of the released arachidonic acid could be esterified in the cell membrane phospholipids, as reported in murine P388D 1 cells (50), and the reaction rate of release and uptake of arachidonic acid might reach a steady state. In THP-1 cells, sPLA 2 -X also elicited a remarkable release of oleic acid. In the assay with PC/DOC mixed micelles (Table I), sPLA 2 -IB is more active than sPLA 2 -X with regard to the hydrolytic efficiency toward POPC. In contrast, the oleate release from THP-1 cells by sPLA 2 -X is much larger than that by sPLA 2 -IB (Fig. 4B). We found that the enzymatic activity of sPLA 2 -IB is negligible toward an aqueous dispersion of PC without DOC but is dramatically increased with PC/DOC mixed micelles. This is quite a contrast to sPLA 2 -X, which showed an enzymatic activity toward POPC even in the absence of DOC at almost the same level of the activity in the DOC mixed micelles. 3 These findings indicate that sPLA 2 -IB activity is markedly affected by the physical state of the substrate compared with sPLA 2 -X, which might explain the differences in their potencies in the intact cell membranes that contain a mixture of phospholipids.
Calculation of the rates of arachidonic acid release from THP-1 cells revealed that the reaction rate of sPLA 2 -X (526 nmol/min/mg) was ϳ66and ϳ155-fold higher than that of sPLA 2 -IB and -IIA, respectively. From PAPC that is mostly present at the outer leaflet of the plasma membranes (41), sPLA 2 -X released arachidonic acid at ϳ5and ϳ58-fold higher rate than that of sPLA 2 -IB and -IIA (Table I). The relative differences in the reaction rates between these two systems suggest that some specific machinery might be involved in the induction of the prompt release reactions of sPLA 2 -X in the intact cells. The PLA 2 R is known to mediate the sPLA 2 -IBinduced biological responses including PGE 2 production in various cell types (32). Fonteh et al. (51) have recently reported the potential role of PLA 2 R in the selective release of arachidonic acid by sPLA 2 -IB in several inflammatory cells including THP-1 cells. Our preliminary studies revealed that the binding affinity of human sPLA 2 -X for the mouse PLA 2 R is about 100fold lower than human sPLA 2 -IB, although the species differences between the ligand and receptor must be taken into consideration (16). Since a marked arachidonic acid release was also observed in sPLA 2 -X-treated U937 and HL60 cells where the PLA 2 R mRNA was not detectable by reverse tran-scription-PCR analysis 4 (51), the PLA 2 R should not be involved in this process. It has been proposed that the interaction of sPLA 2 -IIA with cell surface heparan sulfate proteoglycan is important for its potency for the fatty acid release (52). However, mutagenesis studies have shown that the basic amino acid residues rather than the heparinoid binding regions are the main determinants for controlling the rate of fatty acid release from sPLA 2 -IIA-treated intact cells (53). Recently, sPLA 2 -V was found to induce fatty acid release more efficiently than sPLA 2 -IIA in several mammalian cells (35), and Trp-31 on the putative interfacial binding surface was suggested to play an important role in its binding to PC vesicles as well as to the outer cell membrane (54). However, sPLA 2 -X is the most acidic sPLA 2 and dose not possess the corresponding Trp residue in its sequence (9). The possible contribution of anionic residues in sPLA 2 -X should be evaluated by the molecular level approach to understand its strong potency for fatty acid release from intact cells.
In the PLA 2 superfamily, cPLA 2 plays a potential role in the release of arachidonic acid during the cell activation process, since it possesses substrate specificity for 2-arachidonyl phospholipids (12,13). In THP-1 cells, we could not detect the elevation of intracellular Ca 2ϩ concentration by sPLA 2 -X (data not shown), and the sPLA 2 -X-induced arachidonic acid release was suppressed by sPLA 2 inhibitor but not blocked by a cPLA 2 inhibitor (Fig. 4D), indicating that fatty acid release by sPLA 2 -X is not dependent on the cPLA 2 activation. sPLA 2 -X also elicited a significant production of PGE 2 in contrast to little production by sPLA 2 -IB and -IIA (Fig. 5), suggesting its potential role in the production of lipid mediators. We also detected a significant production of LTC 4 by the action of sPLA 2 -X. However, the elevated amount of LTC 4 was as little as 0.05% that of the maximum level induced by 3 M A23187, which made a contrast to 10% in the PGE 2 production. These findings suggest that the released arachidonic acid by sPLA 2 -X was efficiently metabolized via cyclooxygenase but not by 5-lipoxygenase, although the precise mechanisms underlying the functional coupling between the arachidonic acid release and eicosanoid generation remain uncertain. The sPLA 2 -X-induced fatty acid releases could be relevant to the production of bioactive lysophospholipids (55) as well as to the asymmetrical arrangement of membrane phospholipids, which might lead to higher susceptibility to interact with other types of sPLA 2 s (56,57). In addition, the accumulated arachidonic acid and oleic acid might activate phospholipase D to induce functional effects including apoptosis, as reported for Jurkat T cells (58).
In human lung preparations, the sPLA 2 -X protein was detected especially in type II alveolar epithelial cells. Since these cell types are known to play a role in the secretion of lung surfactant, the release of various cytokines as well as the production of eicosanoids (59), the released sPLA 2 -X might regulate these cell functions by releasing fatty acids from the cell membranes or by modulating the compositions of cell membrane phospholipids. Alternatively, sPLA 2 -X released from epithelial cells might act on neighboring alveolar macrophages to elicit lipid mediator releases, such as reported for transcellular PG production in fibroblasts by the action of mast cell-derived sPLA 2 -V (60). The restricted expression of sPLA 2 -X transcripts in immunity-related tissues (9) also suggests its involvement in inflammatory responses. Analysis of the expression levels under pathological conditions should offer more information concerning the biological roles of this enzyme.
In conclusion, we have demonstrated here that sPLA 2 -X possesses more powerful potency for releasing arachidonic acid compared with sPLA 2 -IB and -IIA in human myeloid leukemia cells. We think that sPLA 2 -X and cPLA 2 play essential roles in mobilizing arachidonic acid from the outside and the inside of the cells, respectively. Further studies are required to establish the physiological functions of sPLA 2 -X as well as to understand the precise mechanisms underlying the arachidonic acid release from intact cells. Finally, elucidation of the biological roles of each group of sPLA 2 in disease states, especially in inflammatory conditions, should be of great value for the development of subtype-specific inhibitors as therapeutic drugs.