Structures, enzymatic properties, and expression of novel human and mouse secretory phospholipase A(2)s.

Mammalian secretory phospholipase A(2)s (sPLA(2)s) form a family of structurally related enzymes that are involved in a variety of physiological and pathological processes via the release of arachidonic acid from membrane phospholipids or the binding to specific membrane receptors. Here, we report the cloning and characterization of a novel sPLA(2) that is the sixth isoform of the sPLA(2) family found in humans. The novel human mature sPLA(2) consists of 123 amino acids (M(r) = 14,000) and is most similar to group IIA sPLA(2) (sPLA(2)-IIA) with respect to the number and positions of cysteine residues as well as overall identity (51%). Therefore, this novel sPLA(2) should be categorized into group II and called group IIE (sPLA(2)-IIE) following the recently identified group IID sPLA(2) (sPLA(2)-IID). The enzymatic properties of recombinant human sPLA(2)-IIE were almost identical to those of sPLA(2)-IIA and IID in terms of Ca(2+) requirement, optimal pH, substrate specificity, as well as high susceptibility to the sPLA(2) inhibitor indoxam. Along with the biochemical properties of proteins, genetic and evolutional similarities were also observed among these three types of group II sPLA(2)s as to the chromosomal location of the human gene (1p36) and the exon/intron organization. The expression of sPLA(2)-IIE transcripts in humans was restricted to the brain, heart, lung, and placenta in contrast to broad expression profiles for sPLA(2)-IIA and -IID. In sPLA(2)-IIA-deficient mice, the expression of sPLA(2)-IIE was markedly enhanced in the lung and small intestine upon endotoxin challenge, which contrasted with the reduced expression of sPLA(2)-IID mRNA. In situ hybridization analysis revealed elevation of sPLA(2)-IIE mRNA at alveolar macrophage-like cells in the lung of endotoxin-treated mice. These findings suggest a distinct functional role of novel sPLA(2)-IIE in the progression of inflammatory processes.

Phospholipase A 2 (PLA 2 ) 1 comprises a diverse family of en-zymes that catalyzes the hydrolysis of glycerophospholipids at the sn-2 position to produce free fatty acid and lysophospholipids (1,2). PLA 2 s participate in pathophysiological processes by releasing arachidonic acid from membrane phospholipids leading to the production of various types of proinflammatory lipid mediators, such as prostaglandins and leukotrienes (3,4). To date, several mammalian intracellular and secretory PLA 2 s (sPLA 2 s) have been characterized and classified into different families according to their biochemical features (5,6). Intracellular PLA 2 s comprise the Ca 2ϩ -sensitive arachidonoyl-specific 85-kDa cytosolic PLA 2 and a number of Ca 2ϩ -independent PLA 2 s (5). In contrast, sPLA 2 s have several common characteristics including a relatively low molecular mass (13)(14)(15)(16)(17)(18), the presence of 6 -8 disulfide bridges, an absolute catalytic requirement for millimolar concentrations of Ca 2ϩ , and a broad specificity for phospholipids with different polar head groups and fatty acid chains (4,7). At present, five distinct sPLA 2 s have been identified in humans and classified into different groups (group IB, IIA, IID, V, and X) depending on the primary structure characterized by the number and positions of cysteine residues (7), while group IIC sPLA 2 found in rodents is a pseudogene in humans (8). Among them, group IIA sPLA 2 (sPLA 2 -IIA) is thought to be one of the key enzymes in the pathogenesis of inflammatory diseases, since its local and systemic levels are elevated in diseases, such as septic shock, acute pancreatitis, and rheumatoid arthritis (9 -11). However, some inbred mouse strains have a natural frameshift mutation in the sPLA 2 -IIA gene (12,13) and are susceptible to arthritis in the antigen-induced model similar to sPLA 2 -IIA-expressing mouse strains (14,15). In addition, a potent sPLA 2 -specific inhibitor, indoxam (16), was reported to suppress endotoxininduced lethal effects as well as inflammatory cytokine production with a similar potency for sPLA 2 -IIA-expressing and sPLA 2 -IIA-deficient mouse strains (16). Transgenic mice expressing the human sPLA 2 -IIA gene do not develop any overt inflammatory conditions (17). These findings point to the need to reassess the role of sPLA 2 -IIA in inflammatory diseases and suggest that other types of sPLA 2 including some unidentified isoforms may play a compensatory role. For example, the most classical sPLA 2 , group IB sPLA 2 (sPLA 2 -IB), has long been thought to act as a digestive enzyme because of its abundance in digestive organs (18). However, recent studies have identified sPLA 2 -IB as a signaling molecule that induces the lipid mediator releases via binding to its specific receptor, the PLA 2receptor (19 -25). Although sPLA 2 -IIA acts as a ligand for murine PLA 2 receptor (4), our studies with mice deficient in both PLA 2 receptor and sPLA 2 -IIA have demonstrated a potential role of sPLA 2 -IB/PLA 2 receptor-mediated responses in the pro-* 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.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AF189279 and AF112984.
Recent advances in molecular biology and the growing amount of genetic information have led to the identification of several novel types of sPLA 2 s. Chen et al. (27) discovered group V sPLA 2 (sPLA 2 -V) from a human genomic DNA fragment similar to the sPLA 2 -IIA gene. sPLA 2 -V hydrolyzes phosphatidylcholine (PC) more effectively than sPLA 2 -IIA to induce the release of lipid mediators in several mammalian cells (28). Cupillard et al. (29) have isolated the cDNA of group X sPLA 2 (sPLA 2 -X) which possesses 16 cysteine residues at positions characteristic of both sPLA 2 -IB and -IIA. We have recently shown its strong potency for the release of arachidonic acid from several human myeloid leukemia cells (30). More recently, we and other groups (31,32) have reported the cloning of another sPLA 2 isoform, group IID sPLA 2 (sPLA 2 -IID), which shares structural and enzymatic characteristics with that of sPLA 2 -IIA. The expression of sPLA 2 -IID mRNA is enhanced in the thymus of rats and sPLA 2 -IIA-deficient mice upon endotoxin challenge, suggesting its potential role in the progression of endotoxic shock (32). Identification of the entire set of sPLA 2 family members and elucidation of the relative contribution of each isoform to various cellular responses are critical to understanding the precise roles of the sPLA 2 family in physiological and pathological processes.
In light of the growing molecular diversity of sPLA 2 s, we searched the DNA data base and encountered an expressed sequence tag that could represent part of a new sPLA 2 isoform. Here, we report the cloning of a cDNA encoding a novel human sPLA 2 , which we called group IIE sPLA 2 (sPLA 2 -IIE) based on its structural properties. We characterized the enzymatic properties of recombinant human sPLA 2 -IIE, and we compared its expression profiles in humans as well as in endotoxin-treated mice with those of the other sPLA 2 s. Such comprehensive structural information and the availability of recombinant proteins of six types of human sPLA 2 s should enable assessment of the relative contribution and closer study of the biological roles of each isoform in various disease states.
Molecular Cloning of Mouse sPLA 2 -IIE-Mouse gene fragment of sPLA 2 -IIE was discovered by a tBLASTn search (34) against Gen-Bank TM data base using an 11-amino acid sequence (DRCCVTHDCCY) around the catalytic center of the mouse sPLA 2 -IIA (13). A cDNA fragment corresponding to the identified gene sequence was amplified by polymerase chain reaction (PCR). Reverse-transcribed cDNAs from various tissues of B57BL/6J mice (one of the sPLA 2 -IIA-deficient strains (12,13)) were used as templates. Primers for amplification were 5Ј-cttcaagagangagggaaacctg-3Ј, 5Ј-tcaagagangagggaaacctgcc-3Ј (sense), and 5Ј-agcttgttggggtagtgggc-3Ј, 5Ј-cttgttggggtagtgggcata-3Ј (antisense). Two rounds of amplifications (nested PCR) were carried out with these primers and ExTaq (Takara, Japan). Amplification conditions were 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min for 30 cycles. The PCR products were separated on agarose gel, and DNA of the expected size was isolated. The recombinant plasmid was then constructed with pCRII cloning vector (Invitrogen), purified with GFX Micro Plasmid Prep Kit (Amersham Pharmacia Biotech), and sequenced with Applied Biosystems PRISM 310 genetic analyzer. From the determined DNA sequence, new primers were designed for the isolation of 5Ј and 3Ј portions of the cDNA. The cloning of these remaining parts was performed with the protocol of "rapid amplification of the cDNA end" (RACE) using mouse spleen marathon-ready cDNA (CLONTECH) according to the manufacturer's instructions with a slight modification (ExTaq was used instead of KlenTaq polymerase). The 5Ј-end could not be obtained with single RACE, and successive RACE was performed to cover the initial ATG codon. The full-length cDNA was isolated by PCR with end primers, 5Ј-agaaaagagacctctctca-3Ј, 5Ј-tagacggtgactcagagctgca-3Ј (sense), and 5Ј-ggaaaatagacttctcttattcag-3Ј, 5Ј-agggtattgagatgccagaggc-3Ј (antisense).
Molecular Cloning of Human sPLA 2 -IIE-Human genomic DNA and yeast strain harboring YAC (957 F 12) was used as a template for the PCR to isolate the genomic fragments of novel human sPLA 2 -IIE genes. Samples of 200 ng of human genomic DNA or yeast colony suspension were subjected to PCR with two primers, 5Ј-cta(ct)ggctg(ct)(ct)a(ct)tg(ct)gg-3Ј and 5Ј-g(ctg)(ct)c(ga)tagca(ga)cagtcatg-3Ј and ExTaq. Thermal conditions were 94°C for 1 min, 50°C for 1 min, and 72°C for 1 min for 30 cycles. All of the amplified bands were isolated, and their sequences were determined. Three of them were identified as corresponding to sPLA 2 -IIA, -IID, and the novel type of sPLA 2 genes (ϳ350, 800, and 300 base pairs, respectively), all of which contained one intron at the same position. Based on the determined sequence, isolation of human sPLA 2 -IIE cDNA was attempted with the strategy used to clone the mouse sPLA 2 -IIE cDNA with marathon-ready cDNA from human small intestine (CLONTECH). Two sets of primers were designed to isolate the corresponding cDNA. For 5Ј-extension, 5Ј-tggcagcaccagtcagtctg-3Ј and 5Ј-agtctggtccaccggccagtg-3Ј were successively used with fixed primers corresponding to the adapter sequence attached to the 5Ј terminus of the cDNA. For 3Ј-extension, four primers were used, 5Јggcatcggtggtccccactgg-3Ј and 5Ј-tcccactggccggtggaccag-3Ј, 5Ј-ggggtggctatgggagccgagc-3Ј and 5Ј-gctatggagccgagcagggcc-3Ј. The latter two primers were derived from the mouse sPLA 2 -IIE cDNA sequence at the 3Ј-noncoding region.
The existence and positions of the intron of human sPLA 2 -IIE gene were analyzed using several primers. The sequence of the amplified DNA from human genomic DNA was compared with that of the cDNA. Separately amplified genomic portions were arranged to cover the whole region of the open reading frame. An identical amplification pattern of the genomic fragment was observed using the YAC (957 F 12) DNA as a template.
Recombinant Expression of Human sPLA 2 -IIE-Two successive PCRs were performed to amplify human sPLA 2 -IIE cDNA from small intestine cDNA. The first PCR was done with 5Ј-atgaaatctccccacgtgctgg-3Ј and 5Ј-tcagcagggcggggtggg-3Ј, and the second with 5Ј-agtagttgatgcggccgccaccatgaaatctccccacgtgctggtgttc-3Ј and 5Ј-taagcttttctagatcagcagggcggggtgggcccggtgcacag-3Ј. The upstream primer has a NotI recognition site and Kozak sequence (in italic). The downstream primer has a XbaI recognition site. The sPLA 2 -IIE cDNA thus amplified was digested with NotI and XbaI and inserted into pcDNA3.1(ϩ) (Invitrogen) to construct the expression plasmids. After sequencing confirmation, 18 g of recombinant plasmid was transfected into 50% confluent COS-7 cells grown in 148-cm 2 Petri dishes with FuGENE 6 transfection reagent (Roche Molecular Biochemicals), and the culture medium was collected after 3 days. PLA 2 activities in the culture medium and cell fractions were then measured with [ 3 H]oleate-labeled Escherichia coli membranes as a substrate (35). The recombinant human sPLA 2 -IIE was partially purified from the culture supernatant by heparin-Sepharose affinity chromatography (Amersham Pharmacia Biotech; the sPLA 2 activity was eluted with 1 M NaCl) and characterized with respect to the Ca 2ϩ dependence and pH optimum as described previously (32). PLA 2 Assay for Substrate Specificity and Evaluation of Inhibitory Potency of Indoxam-Partially purified human sPLA 2 -IIE was subjected to individual reactions with 13 types of commercially available phospholipids as a substrate, as described previously (32). The hydrolysis rates within the linear range of the enzymatic assay of each sPLA 2 were determined. In separate experiments, human sPLA 2 -IIE was subjected to reactions with mixed phospholipids composed of four types of PCs or three types of PEs with different fatty acid chains at the sn-2 position (oleic acid, linoleic acid, and arachidonic acid in PC and PE or docosahexaenoic acid in PC), as described in our previous paper (32).
The released fatty acids were quantified according to the method of Tojo et al. (36).
Analysis of Expression of sPLA 2 -IIE and -IID mRNAs in LPS-treated Mice-LPS (S. typhosa 0901) was injected intraperitoneally into C57BL/6J mice at a dosage of 10 mg/kg, and total RNA was extracted from various tissues after 24 h with RNeasy Mini Kit (Qiagen). mRNA was then purified with QuickPrep Micro mRNA Purification Kit (Amersham Pharmacia Biotech) and subjected to Northern analysis (5 g). The coding region of mouse sPLA 2 -IIE was cloned into pCRII plasmid, purified after restriction enzyme digestion and subjected to 32 P labeling using Megaprime DNA Labeling System (Amersham Pharmacia Biotech). Northern analysis was also performed with mouse sPLA 2 -IID cDNA probe. The intensity of the signals was quantified with BAS 2000 image analyzer (Fuji Photo Film) and normalized with cyclophilin as a control.
In Situ Hybridization Analysis of sPLA 2 -IIE mRNA Expression in Mouse Lung-Antisense and sense single-strand cRNAs were synthesized from sPLA 2 -IIE cDNA fragments subcloned in pCRII vector. The template was linearized with BamHI (antisense probe) or EcoRV (sense probe), and the labeled RNA probes were synthesized with T7 (antisense) and Sp6 (sense) RNA polymerase in the presence of digoxigenin (DIG)-labeled UTP (DIG Labeling Kit, Roche Molecular Biochemicals).
After treatment with LPS or saline for 24 h, mice were flushed with 0.1 M phosphate-buffered saline (PBS, pH 7.4) through the abdominal aorta followed by perfusion with 4% paraformaldehyde in PBS. Pieces from the removed lungs were immersed in the same fixative at 4°C overnight and dehydrated by a diluted alcohol followed by xylene. The preparations were then passed into paraffin at 56°C, and transversal tissue sections (4-m thick) were mounted on slides freshly coated with 3-aminopropyltriethoxysilane. After dewaxing, the sections were incubated with 0.2 M HCl followed by 1% Triton X-100 and permeabilized with 10 g/ml proteinase K (Roche Molecular Biochemicals). The sections were fixed with 4% paraformaldehyde and acetylated with 0.1 M triethanolamine containing 0.25% acetic anhydride. In situ hybridization was performed essentially according to the method of Lan et al. (38) with some modifications. Briefly, the sections were hybridized with 200 ng/ml DIG-labeled probes for 16 h at 52°C, and then free labeled cRNA was digested by 10 g/ml RNase A. The samples were immersed in 1.5% blocking reagent, preincubated in normal mouse serum, and then incubated with sheep anti-DIG antibody. After washing, the samples were incubated with biotinylated mouse anti-sheep antibody and treated with avidin-biotinylated horseradish peroxidase complex solution (Vector Laboratories Inc.). The samples were then processed with 0.1% 3,3Ј-diaminobenzidine hydrochloride substrate dissolved in 50 mM Tris-HCl (pH 7.4) containing 0.05% H 2 O 2 , and the reaction was stopped by washing with 10 mM Tris-HCl (pH 7.6), containing 1 mM EDTA. The sections were counterstained with 1% methyl green dye, dehydrated, and mounted in Entellan new (Merck).
Immunohistochemical staining for mouse alveolar macrophages was performed as follows. The prepared slides of lung sections were dewaxed, incubated in methanol containing 0.3% H 2 O 2 , and treated with 5% normal rabbit serum for 20 min. The slides were incubated with rat anti-mouse macrophage F4/80 antibody in PBS containing 0.1% bovine serum albumin and then incubated with biotin-conjugated rabbit antirat antibody followed by the treatment with peroxidase avidin-biotin complex regent. After washing, the peroxidase activity was visualized by incubation with 50 mM Tris-HCl (pH 7.4) containing 0.1% 3,3Јdiaminobenzidine and 0.05% H 2 O 2 . The slides were then counterstained and mounted as described above.

Molecular Cloning of Novel sPLA 2 and Characterization of
Its Genomic Organization-In searching for novel sPLA 2 s in the data base, we identified a cDNA fragment (GenBank TM accession number AF046275) that was derived from the sequence library created by the newly developed exon-trapping method designed for retrieval of the coding region distributed in the mouse genome (39). This fragment is composed of 320 bases including several sequence ambiguities. Translation of this tagged sequence presumed the existence of a novel sPLA 2 with a characteristic of the sPLA 2 -IIA/IID subfamily, although the raw sequence in the data base required frameshifts to code functional protein. The corresponding cDNA was amplified by PCR from reverse-transcribed cDNA isolated from several mouse tissues, and the sequence determined could encode a part of the functional sPLA 2 . The remaining parts were isolated from mouse spleen cDNA, and the full-length cDNA (883 bp) thus identified encoded a novel sPLA 2 consisting of 142 amino acids.
To isolate its human counterpart, we initially designed degenerate primers based on the sequence in the conserved region of the mouse novel sPLA 2 . However, extensive search by the PCR using cDNA derived from several tissues with various primer combinations was unsuccessful. Therefore, we attempted to isolate the corresponding genomic clone using primers that are capable of detecting the sPLA 2 -IIA/IID subfamily based on their structural similarities with mouse novel sPLA 2 . Three clones were amplified from human genomic DNA by PCR and identified as corresponding to IIA, IID, and the novel type of sPLA 2 genes, all of which contain one intron at the same position. The presumed exon part of the novel sPLA 2 sequence (encoding 29 amino acids) has 85.1% identity with the corresponding part of the mouse sPLA 2 cDNA. Because the same genomic fragments (sPLA 2 -IIA, -IID, and the novel type) were amplified by PCR with the same pair of primers from the YAC strain (957 F 12) which harbors a part (1.55 megabases) of the human chromosome around 1p36, these three types of sPLA 2 genes are closely located in the vicinity of this chromosome position. The 5Ј and 3Ј parts of the cDNA connected to the isolated genomic sequence of the novel sPLA 2 were then isolated from a human small intestine cDNA library by PCRbased protocol. The 3Ј-noncoding region could not be isolated with the standard strategy using the fixed sequence for the adaptor attached to the poly(A) tail of the mRNA. We utilized the 3Ј-noncoding sequence of the mouse novel sPLA 2 sequence to isolate human cDNA under the assumption that the noncoding part is also made up with a similar sequence in these two animals. A presumed open reading frame of the novel human sPLA 2 cDNA clone thus obtained was composed of 426 bases that encoded 142 amino acids with 89% sequence identity against the mouse counterpart.
Structural Features of Novel sPLA 2

IB, a prepropeptide sequence and the C-terminal extension).
Overall sequence comparison revealed that the novel sPLA 2 had the highest identity with sPLA 2 -IIA (51 and 48% at the amino acid level with human and mouse types, respectively). Based on these structural features, the novel sPLA 2 should be categorized into group II based on the traditional grouping criteria proposed by Heinrikson (40) and was thus called group IIE (sPLA 2 -IIE) following the recently identified group IID isoform (31,32).
During the cloning process of human cDNA, we first identified the genomic fragment of sPLA 2 -IIE containing the exon part between the Ca 2ϩ -binding region and the catalytic center, as well as the intron sequence. Together with this clone, genomic fragments of sPLA 2 -IIA and -IID were also isolated harboring an interrupting intron at the identical position (data not shown) (41,42). Further PCR-based analysis revealed that the open reading frame of human sPLA 2 -IIE gene was interrupted with three introns as shown in Fig. 2. The sequences around the exon/intron boundary agreed with the GT-AG consensus rule (43). The positions of intron 1 and 3 were identical to those of the sPLA 2 -IIA gene (41,42). Thus, similar genomic organizations among these group II sPLA 2 s suggest that they are evolutionarily related.
Recombinant Expression of Novel sPLA 2 s and Characterization of sPLA 2 Activity-In order to confirm that the novel cDNAs encode functional proteins, we expressed human and mouse sPLA 2 -IIE using COS-7 cells as the host and characterized the recombinant products from various aspects. As shown in Fig. 3A, significant activity was detected in the supernatants of human sPLA 2 -IIE-expressing cells using radiolabeled E. coli membranes as a substrate, whereas only 2% of the total activity was detected in the cell-associated fraction, indicating that human sPLA 2 -IIE is a secreted enzyme. In contrast, hardly any activity was detected in the supernatants of mouse sPLA 2 -IIEexpressing cells, possibly due to a very low expression level and/or a low specific activity toward this substrate. We next constructed a C-terminal histidine-tagged mouse sPLA 2 -IIE cDNA and expressed it in COS-7 cells in the same manner. We could detect significant PLA 2 activity in the affinity purified materials prepared from the culture medium (data not shown). For further characterization of enzymatic properties of human sPLA 2 -IIE, we partially purified it with heparin-affinity chromatography, because group II sPLA 2 proteins (IIA and IID) are known to have high affinity for heparin (32,44). The PLA 2 activity of recombinant human sPLA 2 -IIE was completely dependent on Ca 2ϩ and required 2 mM Ca 2ϩ for the maximal level (Fig. 3B), and its optimal activity was detected within pH 7-9 (Fig. 3C). These characteristics are compatible with the common features of sPLA 2 proteins thus far identified.
The substrate preference of human sPLA 2 -IIE was then determined individually with 13 types of commercially available phospholipids. The summary of the results (Table I) indicated the absence of a preference for the arachidonic acid-containing phospholipids. Among the phospholipids examined, sPLA 2 -IIE showed preferences in the order of POPG Ͼ PLPE ϭ POPE, which is compatible with the substrate specificity of sPLA 2 -IIA. In the reactions with mixed substrates composed of various types of PCs or PEs with different fatty acid chains at the sn-2 position, sPLA 2 -IIE showed a weak hydrolyzing activity toward 2-arachidonoyl PC or PE and displayed the same preference profiles as sPLA 2 -IIA and -IID (data not shown). The inhibitory potency of sPLA 2 -specific inhibitor was then examined for six types of human sPLA 2 s. In this experiment, we used one of the 1-oxamoylindolidine derivatives, indoxam (33), which has a powerful inhibitory potency for sPLA 2 -IIA activity toward POPG as a substrate (IC 50 ϭ 1.2 nM) with no suppression against the activities of pancreatic lipase nor cytosolic PLA 2 at 50 M (16). The inhibitory activities of indoxam were examined under the optimal conditions of each sPLA 2 reaction with POPC as a substrate. As shown in Fig. 3D, strong inhibition was observed against the activities of three types of group II sPLA 2 s (IIA, IID, and IIE) with IC 50 values within 1-2 nM, whereas the other isoforms (IB, V, and X) were less sensitive to the inhibitor with over 100-fold higher IC 50 values. Taken together, these findings demonstrate that human sPLA 2 -IIE has almost identical features with group II sPLA 2 s (IIA and IID) in terms of both structural and catalytic properties.
Tissue Expression Profiles of sPLA 2 -IIE and Other sPLA 2 s in Humans-Relative amounts of the transcript of each sPLA 2 isoform were analyzed in various human tissues by reverse transcription PCR (RT-PCR). The expression profiles of the known five types of sPLA 2 s (Fig. 4) were generally compatible with those of the previous reports. For example, the positive expression of sPLA 2 -IID mRNA in various tissues and its absence in the brain, peripheral blood leukocytes, and testis previously observed by Northern analysis (32) were clearly observed in the present analysis. Specific expression of sPLA 2 -IB in the pancreas, lung, and kidney (45), as well as abundant expression of sPLA 2 -V in the heart and placenta (27), was also confirmed. However, in the case of sPLA 2 -X, the expression in the spleen and thymus reported in the previous paper (29) could not be detected, possibly due to the usage of different tissue sources and/or different detection systems. Furthermore, an extra PCR product was observed in addition to the band at the expected size of sPLA 2 -X, which was identified as an immature or improperly spliced transcript containing part of an intron. 2 Compared with these known sPLA 2 s, the expression of sPLA 2 -IIE was quite different and restricted in the brain, heart, lung, and placenta. Since a single round of PCR did not produce visible DNA bands for sPLA 2 -IIE, -IID, and -X in any 2 A. Saiga, N. Suzuki, and K. Hanasaki, unpublished data. of the tissues examined, their expression levels should be lower than those of sPLA 2 -IB, -IIA, and -V.
Enhanced Expression of sPLA 2 -IIE mRNA in Mice upon Endotoxin Challenge-As the expression of group II sPLA 2 isoforms (IIA and IID) is known to be changed in pathological states (9 -11, 32), the expression levels of sPLA 2 -IIE were examined in endotoxin-challenged mice by Northern blot analysis. In this work, we used C57BL6/J strains in which the sPLA 2 -IIA gene is naturally disrupted (12,13). In untreated mice, one major transcript of sPLA 2 -IIE (0.5 kb) was detected in the thymus, small intestine, lung, and spleen (Fig. 5). At 24 h after LPS injection, the expression level of this transcript was elevated in the thymus, small intestine, and kidney. In addition, upon endotoxin challenge, a distinct transcript (0.7 kb) was expressed in those tissues as well as in the lung but not in the spleen. In contrast, sPLA 2 -IIE mRNA was barely detected in the heart, liver, and pancreas with or without LPS stimulation (data not shown). Although the estimated size of the two observed mRNAs was apparently shorter than the isolated cDNA (883 bases), RT-PCR analysis confirmed an increase of the mRNA consisting of the open reading frame upon LPS stimulation in these tissues (data not shown). In the case of sPLA 2 -IID, the expression of two transcripts (1.0 and 2.0 kb) was up-regulated in the thymus but obviously decreased in the spleen, small intestine, and lung of LPS-treated mice. These findings suggest that the expressions of two related group II sPLA 2 s are differently regulated in the inflammatory processes taking place in various tissues of mice deficient in sPLA 2 -IIA.
In order to examine the sPLA 2 -IIE expressing cell types in the lung, in situ hybridization analysis was performed with a specific cRNA probe. In the control mice, few, if any, signals were detected with an antisense probe (Fig. 6A). However, upon LPS challenge, intense positive signals were detected with an antisense probe in some cell types surrounding the type II pneumocyte (Fig. 6B) in contrast to few signals with a sense probe (Fig. 6C). The sPLA 2 -IIE-expressing cells seemed to be alveolar macrophages judging from their morphologies, and some of the sPLA 2 -IIE signals coincided well with the stained location for rat anti-mouse macrophage F4/80 antibody (data not shown).

DISCUSSION
Recent advances in molecular biology as well as accumulating DNA information have led to the notion that secretory and intracellular PLA 2 activities are attributed to the growing number of PLA 2 proteins in mammals (5,29,31,32,46). Therefore, the total set of family members needs to be identified in order to understand the precise role of each isoform in various biological events. In the present study, we cloned and characterized a novel type of sPLA 2 (sPLA 2 -IIE), which represents the sixth isoform of the sPLA 2 family in humans. sPLA 2 -IIE has structural features common to the group II sPLA 2 (sPLA 2 -IIA and IID) with respect to the characteristic distribution of cysteine residues (Fig. 1). In the C-terminal region of sPLA 2 -IIE, there is a relatively high content of basic amino acid residues, which is also characteristic of heparin-binding PLA 2 s including sPLA 2 -IIA, -IID, and -V (32,44). In fact, the heparin-affinity column enabled a successful purification of human sPLA 2 -IIE. In addition to similarities in the primary structures, the genetic locus of sPLA 2 -IIE was identified on human chromosome 1p36 in the vicinity of IIA and IID genes. The sPLA 2 -V gene was also located close by at 1p34 -36, whereas sPLA 2 -IB and -X genes were mapped on chromosomes 12 and 16, respectively (29,47). Close relationships among sPLA 2 -IIA, -IID, and -IIE were also observed in the exon/intron structure. Taken together, these findings suggest that three types of group II sPLA 2 genes constitute a gene cluster that is likely to have emerged from ancient gene duplication events.
Recombinant human sPLA 2 -IIE possesses enzymatic characteristics common to the known sPLA 2 s in terms of extracellular localization, Ca 2ϩ requirement, and optimal pH range. It showed a substrate preference similar to that of sPLA 2 -IIA and -IID, which contrasted with the preferred hydrolysis of 2-arachidonoyl PC by sPLA 2 -X (30). In addition, a susceptibility against the sPLA 2 inhibitor indoxam divided the sPLA 2 family into two subgroups: subgroup IIA, IID, and IIE, and subgroup IB, V, and X (Fig. 3D). Analysis of the crystal structure of the sPLA 2 -IIA complex with a closely related indolizine compound revealed the inhibitor to be located near the active site in sPLA 2 -IIA and bound to Ca 2ϩ (48). Thus, a similar substrate specificity and a high sensitivity toward indoxam strongly suggest that the three group II sPLA 2 s form a similar three-dimensional structure around the catalytic center. In this context, sPLA 2 -IIE might play a compensatory role in the degradation of the endogenous phospholipids together with sPLA 2 -IID especially in sPLA 2 -IIA-deficient mice. Defensive roles against invading bacteria may also be functions of sPLA 2 -IIE like IIA and IID (49), because they preferentially hydrolyze phosphatidylglycerol and PE which are major components of bacterial phospholipids. sPLA 2 -IIA is thought to be one of the key enzymes in the pathogenesis of inflammatory diseases, since its enhanced expression has been observed in various cell types stimulated by inflammatory cytokines and in various inflammatory models including rat endotoxic shock (50). In addition, its potential contribution to arachidonic acid release via binding to cellsurface heparan sulfate proteoglycan was demonstrated by transfection/overexpression experiments (44). In mice, however, the sPLA 2 -IIA gene is naturally disrupted by a single nucleotide insertion in some inbred strains (12,13). Furthermore, even in BALB/c mice that express sPLA 2 -IIA, its expression was limited to the intestine and increased little during endotoxin shock (51,52). In sPLA 2 -IIA-deficient mice, the expression of sPLA 2 -IIE was markedly enhanced in various tissues upon LPS challenge, whereas the expression of another isoform, sPLA 2 -IID, was slightly enhanced in the thymus but remarkably decreased in the spleen, small intestine, and lung (Fig. 5). In this respect, sPLA 2 -IIE might play more pivotal FIG. 4. Tissue distribution of sPLA 2 -IIE and five known sPLA 2 s in humans. Human multiple tissue cDNA panels were used as templates for PCR for each sPLA 2 subtype. sPLA 2 -IB, -IIA, and -V cDNAs were amplified with a single round of PCR (35 cycles), and sPLA 2 -IID, -IIE, and -X cDNAs were amplified by two rounds of PCR (30 cycles each). G3PDH was used as an internal standard. Arrowhead indicates the size expected from the cDNA sequence of human sPLA 2 -X. PBL, peripheral blood leukocyte.
FIG. 5. Expression of sPLA 2 -IIE and IID mRNAs in LPS-treated mice. C57BL/6J mice were injected with S. typhosa LPS or saline. After 24 h, the tissues indicated in the figure were isolated, and mRNAs were prepared. The mRNA (5 g) was analyzed by Northern blotting as described under "Experimental Procedures." Two mice were subjected to each experiment, and the typical result is shown. roles than sPLA 2 -IID in place of sPLA 2 -IIA. We have recently shown that indoxam suppressed murine endotoxic shock through sPLA 2 -IIA-independent mechanisms (16). The potential contributions of indoxam-sensitive sPLA 2 -IIE and -IID should be considered in this process. Since the expression of sPLA 2 -V was also up-regulated in LPS-treated mice (52), these augmented sPLA 2 species might cooperatively play a role in the development of inflammatory conditions. Further analysis such as comparison of the promoter regions of sPLA 2 genes, especially between sPLA 2 -IIE and -IID, could provide a clue to understanding the distinct regulatory mechanisms of expression.
In the lung of LPS-treated mice, enhanced expression of sPLA 2 -IIE transcripts was detected in the alveolar macrophage-like cells (Fig. 6). Since alveolar macrophages are known to act as the source and target for a variety of inflammatory mediators produced during pulmonary inflammation (53), the secreted sPLA 2 -IIE might regulate their functions in the defense against infectious agents and toxic particles in the airways. In the guinea pig model of LPS-induced acute lung injury, enhanced expression of sPLA 2 -IIA was detected in the interstitial and alveolar macrophages (54). In this model, lysophospholipids exert a major injurious effect on lung tissue membranes, and administration of LY31127, an sPLA 2 inhibitor structurally related to indoxam, could reduce the increase in lysophospholipids in the bronchoalveolar lavage fluid (55). A high sensitivity of sPLA 2 -IIE toward indoxam and its enhanced expression in the alveoli suggest its potential involvement in lung injury in concert with sPLA 2 -IIA. In this context, previous studies on the functional analysis of sPLA 2 -IIA using sPLA 2 inhibitor reported for various inflammatory models should be reevaluated after consideration of novel types of group II sPLA 2 s.
The discovery of specific receptor for mammalian sPLA 2 s has led to the notion that sPLA 2 can exert various biological responses via binding to the receptor in addition to its digestive function (25). In rats and mice, sPLA 2 -IB was identified as an endogenous ligand of the PLA 2 receptor to induce various physiological responses including cell proliferation and lipid mediator releases (24,25). Our recent studies with PLA 2 receptor-deficient mice have demonstrated its potential role in the production of inflammatory mediators during LPS shock (26). Although sPLA 2 -IIA does not act as a natural ligand for rat PLA 2 receptor, it can bind mouse PLA 2 receptor with ϳ5-10fold lower affinity compared with sPLA 2 -IB (4). In contrast, Valentin et al. (31) have recently shown that mouse sPLA 2 -IID does not bind the receptor. Since the receptor binding of sPLA 2 proteins does not depend on their enzymatic activities, mouse PLA 2 receptor could discriminate between sPLA 2 -IIA and -IID despite their structural similarities around the catalytic center. To date, endogenous sPLA 2 ligands for PLA 2 receptor have not been clearly characterized in humans. Thus, the potency of sPLA 2 -IIE as a natural ligand of the receptor needs to be evaluated. In addition to the cloned PLA 2 receptor, several binding proteins recognized by snake venom sPLA 2 s have been reported, although their molecular structures, endogenous ligands, and biological functions remain uncertain (4,56). Fenard et al. (57) have recently reported the putative binding sites that might be involved in the venom sPLA 2 s-evoked suppression of HIV-1 virus entry into host cells. The possibility of sPLA 2 -IIE acting as a ligand for these binding sites also deserves attention in future studies.
We have previously reported the cloning and characterization of mouse and human sPLA 2 -IID utilizing murine-expressed sequence tag as starting information. Isolation and characterization of human enzymes are important in terms of practical applications such as drug development. Human sPLA 2 -IIE shares 89% identity with the mouse counterpart in the mature peptide region, whereas sequence identities of other sPLA 2 s are below 80% with the lowest identity found in sPLA 2 -IIA (68%) (13,41). Indeed, human and mouse sPLA 2 -IIA have a number of distinct biological properties with respect to tissue distribution, induction levels under inflammatory conditions, a potential function as a genetic modifier of colorectal cancer, and binding affinity for the PLA 2 receptor (12,58). In this regard, some specific and important functions assigned to sPLA 2 -IIE might have been pressed pressure to maintain the required structure during evolution. Sequence alignment (Fig. 1) shows the presence of one distinct residue in sPLA 2  specific function(s) has been assigned to this residue at present, further functional analysis of sPLA 2 -IIE might reveal its biological significance. RT-PCR analysis revealed the expression of sPLA 2 -IIE transcripts in several human tissues in contrast to a broad distribution of sPLA 2 -IIA and -IID (Fig. 4). Hybridization analysis with RNA Master blot membranes confirmed the expression of sPLA 2 -IIE in the same tissues, although discrete bands were not observed in any tissues examined by Northern analysis (data not shown). In physiological states, the expression levels of sPLA 2 -IIE were lower than the levels of sPLA 2 -IIA, which agreed well with the fact that no expressed sequence tags corresponding to human sPLA 2 -IIE exist in the public data base. Considering its enhanced expression under endotoxic shock in mice, analysis for the expression of human sPLA 2 -IIE under some pathological conditions should be performed to evaluate its biological functions.
In conclusion, we isolated novel human and mouse sPLA 2 s (group IIE) and characterized the catalytic activities and expression. Abnormally high levels of PLA 2 activity have been detected in association with various human diseases, especially in inflammatory conditions. Further studies on the biological functions of sPLA 2 -IIE are required to establish its potential roles in the progression of disease states. Finally, the discovery of the novel sPLA 2 -IIE should enable us to assign more precise functions to each sPLA 2 family member which should be of great value for the development of sPLA 2 inhibitors as therapeutic drugs.