The highly selective production of 2-arachidonoyl lysophosphatidylcholine catalyzed by purified calcium-independent phospholipase A2gamma: identification of a novel enzymatic mediator for the generation of a key branch point intermediate in eicosanoid signaling.

Herein, we report the heterologous expression of the human peroxisomal 63-kDa calcium-independent phospholipase A2gamma (iPLA2gamma) isoform in Sf9 cells, purification of the N-terminal His-tagged enzyme by affinity chromatography, and the identification of its remarkable substrate selectivity that results in the highly selective generation of 2-arachidonoyl lysophosphatidylcholine. Mass spectrometric analyses demonstrated that purified iPLA2gamma hydrolyzed saturated or monounsaturated aliphatic groups readily from either the sn-1 or sn-2 positions of phospholipids. In addition, purified iPLA2gamma effectively liberated arachidonic acid from the sn-2 position of plasmenylcholine substrates. In contrast, incubation of iPLA2gamma with 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine resulted in the rapid release of palmitic acid and the selective accumulation of 2-arachidonoyl lysophosphatidylcholine (LPC), which was not metabolized further by iPLA2gamma. The putative regiospecificity of the 2-arachidonoyl LPC product was authenticated by its diagnostic fragmentation pattern during tandem mass spectrometric analysis. To identify the physiological relevance of iPLA2gamma-mediated 2-arachidonoyl LPC production utilizing naturally occurring membranes, we incubated purified rat hepatic peroxisomes with iPLA2gamma and similarly identified the selective accumulation of 2-arachidonoyl LPC. Furthermore, tandem mass spectrometric analysis demonstrated that 2-arachidonoyl LPC is a natural product in human myocardium, a tissue in which iPLA2gamma expression is robust. Because 2-arachidonoyl LPC represents a key branch point intermediate that can potentially lead to a variety of bioactive molecules in eicosanoid signaling (e.g. arachidonic acid, 2-arachidonoylglycerol), these results have uncovered a novel eicosanoid selective pathway through iPLA2gamma-mediated 2-arachidonoyl LPC production to amplify and diversify the repertoire of biologic lipid second messengers in response to cellular stimulation.

Phospholipases A 2 (PLA 2 ) 1 constitute a large family of enzymes that catalyze the hydrolysis of the sn-2 ester bond of phospholipids to produce free fatty acid and lysophospholipid (1,2). Previously, a novel human calcium-independent phospholipase A 2 (GenBank TM accession no. AF263613), termed iPLA 2 ␥ (also known as group VIB phospholipase A 2 ), has been cloned from the human genome and expressed in Sf9 cells (3,4). In addition to its dual signature lipase and nucleotide binding motifs that define the iPLA 2 subfamily of enzymes, the iPLA 2 ␥ polypeptide also contains a C-terminal peroxisomal tripeptide localization sequence (-SKL). The first identification of the iPLA 2 ␥ protein from non-recombinant tissues demonstrated that it was present in the peroxisomal fraction of rat liver predominantly as a 63-kDa polypeptide (5). Direct mass spectrometric analysis of the lipidome of hepatic peroxisomes demonstrated the high content of arachidonate-containing choline glycerophospholipid species (5). Thus, these findings suggested that iPLA 2 ␥ might participate in the generation of lipid second messengers through the mobilization of arachidonic acid in response to cellular stimuli. However, because iPLA 2 ␥ did not selectively cleave arachidonate directly from the sn-2 position (3), the molecular mechanism(s) potentially responsible for iPLA 2 ␥-mediated release of arachidonic acid were not clear.
Historically, the ability of PLA 2 s to selectively mobilize arachidonic acid from endogenous phospholipid storage depots has served as an important and defining characteristic in identifying enzymes contributing to eicosanoid-mediated signaling processes. It is well established that cPLA 2 ␣ possesses high hydrolytic selectivity toward lipids containing arachidonic acid in the sn-2 position (6 -9), making it an important and intensely investigated enzymatic candidate for intracellular eicosanoid signaling studies (10,11). In addition, calcium-independent PLA 2 ␤ (iPLA 2 ␤) has also been shown to participate in the release of arachidonic acid upon cellular stimulation (12)(13)(14)(15)(16)(17)(18). Previous consideration of arachidonate selectivity has largely focused on the PLA 2 -catalyzed release of arachidonic acid directly from the sn-2 position of glycerophospholipids. This represents the simplest and most direct mechanism for the selective generation of arachidonate and initiation of eicosanoid cascades by signaling phospholipases. However, other less well characterized multistep processes for eicosanoid generation are also possible and have been reported in the literature (19). Detailed lipid analyses using mass spectrometry and/or reversed phase high performance liquid chromatography have identified the choline and ethanolamine phospholipid pools as the major contributors to arachidonic acid mass release in most well studied systems (20 -24).
We considered the alternative possibility that "arachidonate selectivity" and subsequent eicosanoid second messenger generation may be partially manifest through the selective production of 2-arachidonoyl lysophosphatidylcholine during cellular signaling. This possibility was particularly attractive because a lysophospholipase activity that selectively deacylates arachidonoyl LPC at a rate of 70 mol⅐min Ϫ1 ⅐mg Ϫ1 was purified from bovine brain by Exton and colleagues (25). Furthermore, it appeared that, at least in bovine brain, a large portion of arachidonate was produced through sequential deacylation of diacyl-PC by phospholipase A 1 and subsequent lysophospholipase activities (26). The high specific activity lysophospholipase from bovine brain was immunologically identical to human cPLA 2 ␣ (25), which is known to possess much greater lysophospholipase activity than its PLA 2 activity (27). Thus, the sequential deacylation of arachidonoyl-containing phospholipids by PLA 1 and lysophospholipase activity is very likely an important and effective pathway contributing to the in vivo release of arachidonic acid.
Previously, 2-arachidonoylglycerol was isolated from intestine and brain as an endogenous ligand for the cannabinoid receptors (28 -30), thus revealing its biological importance as an endocannabinoid signaling molecule in mammalian tissues. In addition, recent studies have demonstrated that 2-arachidonoylglycerol serves as an excellent substrate for cyclooxygenase-2 oxygenation (31), and the product, prostaglandin H 2 glyceryl ester, was identified as the precursor of a wide variety of prostaglandin glyceryl ester derivatives (32) with signaling potential independent of prostaglandins (33). Oxygenation of 2-arachidonoylglycerol by 15-lipoxygenase yields 15-hydroxyeicosatetraenoic acid glyceryl ester, a possible peroxisome proliferator-activated receptor ␣ agonist (34). It has been shown that multiple pathways contribute to the biosynthesis of 2-arachidonoylglycerol, such as the sn-1-specific lipase-catalyzed hydrolysis of 2-arachidonoyl diglyceride species, derived either from the actions of PLC (30) or phosphatidic acid phosphohydrolase (35). Importantly, prior studies have demonstrated the lysophospholipase C activity in mouse N18TG2 neuroblastoma cells leads to the rapid conversion of 2-arachidonoyl LPC into 2-arachidonoylglycerol (36). Collectively, these studies have identified 2-arachidonoyl LPC as an important branch point metabolite in eicosanoid signaling, although the identity of the phospholipase(s) A 1 catalyzing the highly selective production of 2-arachidonoyl LPC have thus far remained enigmatic.
Herein, we report the expression, isolation, and purification of the human 63-kDa iPLA 2 ␥ from Sf9 cells utilizing metal affinity chromatography and demonstrate its highly selective PLA 1 activity toward 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (PAPC). Remarkably, iPLA 2 ␥ does not substantially hydrolyze 2-arachidonoyl LPC, thereby allowing its accumulation in vitro. Furthermore, we show that 2-arachidonoyl LPC is a natural product in human heart tissue and that 2-arachidonoyl LPC rapidly accumulates after incubation of naturally occurring peroxisomal membranes with purified recombinant iPLA 2 ␥. Collectively, these results implicate iPLA 2 ␥ as a likely proximal mediator of the production of 2-arachidonoyl LPC in vivo, which has the potential to integrate multiple eicosanoid signaling cascades in response to cellular stimulation.

EXPERIMENTAL PROCEDURES
Materials-Talon cobalt-charged affinity resin was purchased from BD Biosciences. 1-Palmitoyl-2-[1-14 C]oleoyl-sn-glycero-3-phosphocholine ([ 14 C]POPC) was purchased from PerkinElmer Life Sciences. Diacyl-phospholipids, plasmenylcholine, and phospholipid internal standards for ESI/MS were obtained from Avanti Polar Lipids (Alabaster, AL). Deuterated fatty acid (i.e. 7,7,8,8-d4 -16:0 fatty acid), used as an internal standard for the quantitation of non-esterified fatty acids, was purchased from Cambridge Isotope Laboratory Inc. (Andover, MA). The ProteoProfile trypsin in-gel digestion kits were purchased from Sigma-Aldrich. High-performance liquid chromatography grade organic solvents and channeled LK6D Silica Gel 60-Å thin layer chromatography plates (Whatman) were obtained from Fisher Scientific. Chloroform and methanol used for lipid extraction were purchased from Burdick & Jackson (Muskegon, MI). Sf9 cell culture media and reagents were obtained from Invitrogen. All other chemicals and reagents were typically obtained from either Fisher Scientific or Sigma-Aldrich.
Cloning, Expression, and Affinity Purification of 63-kDa iPLA 2 ␥(His) 6 from Sf9 Cells-The cDNA encoding the N-terminal His-tagged 63-kDa iPLA 2 ␥ was prepared as follows: sense primer (5Ј-AAAAGTCGACAAT-GCATCACCATCACCATCACTCTCAACAAAAGGAAAATGAAC-3Ј) and antisense primer (5Ј-GGTACCGCATGCTCACAATTTTGAAAAGA-ATGGAAG-3Ј) were paired to amplify a 1.7-kb product from a fulllength iPLA 2 ␥ pFASTBac1 construct (3) for cloning via SalI/SphI sites into vector pFASTBac1 (Invitrogen). The Bac-to-Bac Baculovirus Expression System (Invitrogen) was subsequently utilized for bacmid preparation and CellFECTIN-mediated transfection of Spodoptera frugiperda (Sf9) cells to produce baculovirus, which was amplified and titered by the Neutral Red agar overlay method. Expression of the N-terminal His-tagged iPLA 2 ␥ was initiated by infection of 300 ml of Sf9 cells (1 ϫ 10 6 cells/ml) with recombinant virus at a multiplicity of infection of Ϸ1. After incubation for 48 h at 27°C, cells were pelleted by centrifugation, resuspended in ice-cold Grace's insect medium without serum, and repelleted. The cell pellet was then resuspended in 30 ml of lysis buffer (25 mM sodium phosphate, pH 8.0, containing 0.25 M sucrose, 0.2 mM dithiothreitol, 5 g/ml aprotinin, and 5 g/ml leupeptin), sonicated (20 ϫ 1 s bursts utilizing a Vibra-cell sonicator at 30% output), and centrifuged at 100,000 ϫ g for 1 h. The pellet was resuspended in an equal volume of lysis buffer and saved as the membrane fraction. Recombinant iPLA 2 ␥ was purified from Sf9 cell cytosol by using Talon Co 2ϩ affinity column chromatography (BD Biosciences) employing an imidazole gradient generated using an Amersham Biosciences fast protein liquid chromatography system. Fractions were assayed for activity utilizing [ 14 C]POPC as substrate as described below. Active samples were flash frozen in liquid nitrogen and stored at Ϫ80°C.
Cloning, Expression, and Affinity Purification of iPLA 2 ␤(His) 6 from Sf9 Cells-Sequence encoding iPLA 2 ␤ with a C-terminal His tag was generated by PCR amplification of the native iPLA 2 ␤ cDNA utilizing sense (5Ј-AAAAGTCGACGCCACCATGCAGTTCTTCGGACGCCTTGT-CAACAC-3Ј) and antisense (5Ј-AAAACTCGAGTCAGTGATGGTGATG-GTGATGGGGCGACAGCAGCATTTGGA-3Ј) primers containing SalI and XhoI sites, respectively. The PCR product was subcloned into pFASTBac1 for preparation of recombinant baculovirus as described above for iPLA 2 ␥(His) 6 . Following infection of 300 ml of Sf9 cells (1.5 ϫ 10 6 cells/ml) with baculovirus encoding iPLA 2 ␤(His) 6 for 48 h, cells were harvested by centrifugation (900 ϫ g for 10 min), washed once with Grace's insect medium without serum, and resuspended in 30 ml of 25 mM sodium phosphate, pH 7.8, 20% glycerol, 2 mM ␤-mercaptoethanol, 5 g/ml aprotinin, 5 g/ml leupeptin. After lysing the cells by sonication (30 ϫ 1-s bursts), the homogenate was centrifuged at 100,000 ϫ g for 1 h to obtain the cytosol to which NaCl was added to a final concentration of 250 mM. Recombinant iPLA 2 ␤ in Sf9 cell cytosol was purified by Talon Co 2ϩ affinity column chromatography (BD Biosciences) employing an imidazole gradient generated using an Amersham Biosciences fast protein liquid chromatography system. Column fractions were assayed for iPLA 2 activity utilizing [ 14 C]POPC as substrate as described below, pooled, and dialyzed overnight against Buffer A (25 mM imidazole, pH 7.8 containing 20% glycerol, 1 mM dithiothreitol, and 1 mM EGTA). The dialyzed sample was applied to a 2.5-ml column of ATPagarose equilibrated with Buffer A and washed with Buffer A contain-ing 1 mM AMP and 50 mM NaCl. Bound iPLA 2 ␤(His) 6 was eluted with Buffer A containing 2 mM ATP and 100 mM NaCl, and dialyzed against Buffer A (EGTA concentration was reduced to 0.1 mM) containing 50 mM NaCl. Active samples were flash frozen in liquid nitrogen and stored at Ϫ80°C.
For studies employing ESI/MS, phospholipid substrates were cosonicated to homogeneity with 1,2-dioleoylglycerol (DAG) (37) in the assay buffer. In the assay for cPLA 2 ␣, 100 mM Tris acetate containing 5 mM of CaCl 2 , pH 8.0, was used to replace the EGTA-containing buffer. Lipid substrates were then mixed with phospholipase (0.5-1 g of purified iPLA 2 ␥ or iPLA 2 ␤, or 70 g of cytosolic protein of Sf9 cells expressing cPLA 2 ␣) in a total volume of 200 l, and the final concentrations for phospholipid substrate and DAG were 30 and 10 M, respectively. The mixture was incubated at 37°C for the indicated times in the figures. Reactions were terminated by addition of 4 ml of chloroform/methanol (1:1, v/v) containing internal standards, followed by addition of 2 ml of 50 mM aqueous LiCl (38) and extraction of lipid species into the chloroform layer by the Bligh and Dyer method (39). The solvent in the lipid extracts was then evaporated using a nitrogen stream, and the samples were redissolved in chloroform and filtered with Millex-FG 0.20-m filters (Millipore, Bedford, MA). Extracted lipid samples were routinely stored in 200 l of chloroform/methanol (1:1, v/v) under nitrogen at Ϫ20°C. Hydrolysis of peroxisomal phospholipids catalyzed by iPLA 2 ␥ was examined by incubation of 500 ng of purified iPLA 2 ␥ with isolated rat liver peroxisomes (5) at 37°C for the indicated times using identical assay conditions. Lysophospholipase activity was assayed by incubating 1 g of enzyme (iPLA 2 ␥ or iPLA 2 ␤) with 50 M of LPC substrate, injected in 2 l of ethanol, in 100 l of assay buffer at 37°C for 20 min. The reaction was terminated by addition of chloroform/methanol (1:1, v/v), and the products were analyzed by ESI mass spectrometry.
ESI-MS of Phospholipids-ESI-MS analysis was performed using a Thermo Electron TSQ Quantum Ultra spectrometer (San Jose, CA) equipped with an electrospray ion source as previously described in detail (40,41). Samples were diluted with chloroform/methanol (1:1, v/v, ϳ2 pmol/l) prior to direct infusion into the ESI ion source at a flow rate of 2 l/min with a 2-min period of signal averaging. Phosphatidylcholine and lysophosphatidylcholine molecular species were quantitated by multidimensional mass spectrometry (14:1-14:1 PC and 17:0 LPC as internal standards) after correction for 13 C isotopomer differences in the positive ion mode as previously described (38,42,43). After addition of the appropriate amount of LiOH in methanol (38,44), fatty acid species were quantified by comparisons to the deuterated internal standard (7,7,8,8-d4 -16:0 fatty acid) in the negative ion mode. Tandem mass scanning of neutral loss of 59 atomic mass units (loss of trimethylamine) was performed at a collision energy of 24 eV and a collision gas pressure of 1.0 mTorr. Product ion spectra of sodiated LPC species were acquired at a collision energy of 30 eV and a collision gas pressure of 1.0 mTorr. The determination of the regiospecificity of LPC species by product ion analysis was performed as previously described (45).
Other Procedures-Full-length human cPLA 2 ␣ cDNA was cloned into pFASTBac1 via XhoI/HindIII for bacmid and baculovirus preparation utilizing the Bac-to-Bac Baculovirus Expression System (Invitrogen), and cPLA 2 ␣ was expressed in Sf9 cells. The cytosolic fraction of Sf9 cells expressing cPLA 2 ␣ was isolated by sonicating the cells in 50 mM Tris-HCl, pH 7.4, containing 1 mM EGTA, 150 mM NaCl, and 1 mM dithiothreitol, and subsequent centrifugation at 100,000 ϫ g for 1 h. Affinitypurified anti-iPLA 2 ␥ polyclonal antibodies were generated as previously described (3). End-stage human heart tissue was obtained from the tissue bank of the Cardiology Division, Washington University School of Medicine, with appropriate legal and Institutional Board Review authorization. Human myocardial lipids were extracted by the Bligh and Dyer procedure (39) with modification as previously described. Preparation and regiospecific separation of sn-1 and sn-2 arachidonoyl LPC was accomplished following the procedure of Creer and Gross (46). Protein concentration was measured by the Bradford assay (Bio-Rad) utilizing bovine serum albumin as a standard.

Expression of the 63-kDa iPLA 2 ␥ in Sf9
Cells-Previous work demonstrated that expression of the full-length iPLA 2 ␥ utilizing a pFASTBac (pFB) promoter proximal to the ATG translation initiation site coding the 88-kDa isoform resulted in a relatively modest level of expression of only the membrane associated 77-and 63-kDa proteins (3). However, truncation of the nested transcriptional inhibitory domain in exon 5 resulted in high levels of expression of the 63-kDa peroxisomal iPLA 2 ␥ isoform ( Fig. 1) (47). Accordingly, we constructed a baculovirus encoding the 63-kDa iPLA 2 ␥ with an N-terminal His tag utilizing a pFB vector and examined the level of expression in Sf9 cells. Western analysis demonstrated a strong immunoreactive band at 63-kDa in both cytosolic and membrane fractions of cells expressing recombinant 63-kDa iPLA 2 ␥, but not in control cells infected with empty pFB vector ( Fig. 2A). Other immunoreactive bands were also present, with most predominantly in the membrane fraction, indicating proteolytic degradation and/or other post-translational modifications. The cells expressing the 63-kDa isoform of iPLA 2 ␥ showed robust calciumindependent phospholipase A 2 activity in both cytosolic and membrane fractions in comparisons to control cell fractions (Fig. 2B). Collectively, these results indicated that the 63-kDa iPLA 2 ␥(His) 6 was successfully expressed in Sf9 cells as a highly active calcium-independent phospholipase A 2 .
Affinity Purification of the Cytosolic 63-kDa iPLA 2 ␥ from Sf9 Cells-The soluble recombinant 63-kDa iPLA 2 ␥ in Sf9 cell cytosol was purified by Talon Co 2ϩ affinity chromatography utilizing an imidazole gradient (Fig. 3A), which resulted in a purification of iPLA 2 ␥ of over 400-fold to a final specific activity of 1.2 mol⅐mg protein Ϫ1 ⅐min Ϫ1 as assayed by [ 14 C]oleic acid release from [ 14 C]POPC. Multiple attempts at further chromatographic purification were largely unsuccessful in removing the small residual amounts of contaminating proteins. Approaches utilized included a secondary metal affinity step as well as hydrophobic interaction, ion exchange, hydroxylapatite, and other non-metal affinity columns (e.g. heparin). Although iPLA 2 ␥ contains a nucleotide binding motif homologous to the sequence that greatly facilitates the purification of iPLA 2 ␤(48), iPLA 2 ␥ expressed in Sf9 cells did not bind to ATP-or ADPagarose resins in our hands. SDS-PAGE silver staining revealed the presence of a predominant 63-kDa protein and a minor band at 70-kDa (Fig. 3B). The identities of these proteins were verified by MALDI-TOF/TOF mass spectrometry to be iPLA 2 ␥ and the heat shock protein, HSP70, respectively (see Table I and supplemental material). The presence of mitochondrial import and peroxisomal targeting signal motifs provide for potential dual localization between these organelles. As a member of the iPLA 2 family, iPLA 2 ␥ possesses a patatin homology domain, which contains conserved dual signature nucleotide and lipase consensus sequence motifs encoded by exons 7 and 8.
Inhibition of purified iPLA 2 ␥ by BEL-The mechanismbased inhibitor, (E)-6-(bromomethylene)-3-(1-naphthalenyl)-2H-tetrahydropyran-2-one (BEL), has been previously shown to inhibit phospholipase A 2 activity in crude membrane preparations obtained from Sf9 cells transfected with the fulllength iPLA 2 ␥ coding sequence (3). Subsequent experiments with the R-and S-enantiomers of BEL utilizing crude membrane fractions revealed that (R)-BEL was ϳ5-fold more potent than (S)-BEL (49). We performed similar inhibition experiments with (R)-and (S)-BEL for the purified soluble 63-kDa iPLA 2 ␥. Over 80% inhibition of PLA 2 activity was manifest with 5 M (R)-BEL, whereas only 20% inhibition was manifest with the same concentration of (S)-BEL (Fig. 4). The IC 50 for (R)-BEL was Ϸ1 M. Thus, the enantiomerically selective inhibition of iPLA 2 ␥ by (R)-BEL resulted from the chiral interactions of the BEL molecule with the purified 63-kDa iPLA 2 ␥ protein in solution and did not result from diastereotopic interactions of the chiral inhibitor with the chiral membrane microenvironment (49).
Mass Spectrometric Determination of the Substrate Selectivity of Purified iPLA 2 ␥-Initial rate analyses of 1-palmitoyl-2- All metabolites, as well as PAPC, were readily quantitated by comparisons of their ion peak intensities to those of the corresponding internal standards. Thus, the predominant products from the iPLA 2 ␥-catalyzed PAPC hydrolysis were 20:4 LPC and palmitic acid (Fig. 6), with their molar amounts correlating with the amounts of hydrolyzed PAPC. The initial rate of iPLA 2 ␥-catalyzed PAPC hydrolysis was 2.4 mol⅐min Ϫ1 ⅐mg Ϫ1 based upon the decrease in PAPC peak intensity and the increase in 2-arachidonoyl LPC peak intensity.
The presence of diolein (DAG) in the artificial liposomes was found not to affect the substrate selectivity of iPLA 2 ␥. Multiple experiments utilizing PAPC liposomes in the absence of DAG showed the same selectivity toward 2-arachidonoyl LPC production, but with ϳ3-fold less total phospholipase activity (data not shown).
We also examined the hydrolysis of 1-palmitoyl-2-docosahexaenoyl-sn-glycero-3-phosphocholine (PDPC) catalyzed by iPLA 2 ␥ in a similar manner. Incubation of PDPC liposomes with iPLA 2 ␥ resulted in rapid production of 2-docosahexaenoyl LPC (ϳ1.0 mol⅐min Ϫ1 ⅐mg Ϫ1 ) but only diminutive amounts of palmitoyl LPC were produced. As anticipated, palmitic acid, but not docosahexaenoic acid, was released as the predominant fatty acid, with the amount of palmitic acid released closely corresponding to the amount of 2-docosahexaenoyl LPC produced.
To determine the regiospecificity of the lysolipid products, we employed tandem mass spectrometry, which provides diagnostic fragmentation patterns to discriminate between the sn-1 and sn-2 isomers of LPC as previously described (45). First, 1and 2-arachidonoyl LPC standards were each prepared by sPLA 2 and sn-1-specific lipase treatment of diarachidonoyl-PC, respectively, and the resultant products were subsequently purified by high-performance liquid chromatography as previously described (46). The tandem mass spectra of the sodiated sn-1 and sn-2 arachidonoyl LPC standards ([MϩNa] ϩ , m/z 566) demonstrated that the relative intensities of the peaks at m/z FIG. 3. Affinity purification of histidine-tagged iPLA 2 ␥. A, the cytosolic fraction from S9 cells expressing recombinant iPLA 2 ␥ was loaded onto a Talon Co 2ϩ affinity column, washed, and eluted by an imidazole gradient (dashed line) as described in "Experimental Procedures." UV absorbance (solid line) was monitored at 280 nm. Calciumindependent phospholipase A 2 activities (ࡗ) of column fractions were assayed by incubation with 5 M [ 14 C]POPC at 37°C for 0.5 min. The released [1-14 C]oleic acid was quantitated as described in Fig. 2. B, column fractions in A, with peak iPLA 2 activity were loaded onto an 8% SDS-polyacrylamide gel (1 g/lane), electrophoresed, and visualized by silver staining. The predominantly enriched protein at Ϸ63 kDa, and the major copurifying protein at Ϸ70 kDa, were identified as iPLA 2 ␥ isoform and HSP70 by MALDI-TOF/TOF mass spectrometry (detailed in Table I  104 (choline ion, [C 5 H 14 NO] ϩ ) and m/z 147 (sodiated cyclic ethylene phosphate, [C 2 H 5 PO 4 Na] ϩ ) were dramatically different for the two regioisomers (i.e. m/z 104 to m/z 147 ratio is 6:1 for 1-arachidonoyl LPC and 1:7 for 2-arachidonoyl LPC) (Fig. 7,  A and B), in accordance with the fragmentation mechanisms identified in previous studies for oleoyllysophosphatidylcholine (45). In comparison, the tandem mass spectrum (MS/MS) of the sodiated 20:4 LPC produced by iPLA 2 ␥-catalyzed PAPC hydrolysis showed several salient features. First, these tandem mass spectra contained typical peaks characteristic of the phosphocholine head group (i.e. m/z 507,    (50). Second, the relative intensity of the peak at m/z 147 was 5-fold higher than that of the peak m/z 104 (Fig. 7C), indicating that the dominant species of arachidonoyl LPC produced by iPLA 2 ␥-catalyzed PAPC hydrolysis was the 2-acyl isomer during the entire reaction interval studied (45).
To determine whether the observed sn-1 selectivity of iPLA 2 ␥ toward PAPC was unique to iPLA 2 ␥, we performed ESI/MS analyses of the products of two other prominent intracellular PLA 2 enzymes, iPLA 2 ␤ and cPLA 2 ␣, under similar conditions. Hydrolysis of PAPC liposomes (30 M) catalyzed by purified iPLA 2 ␤ resulted in the production of both 1-palmitoyl LPC (0.8 mol⅐min Ϫ1 ⅐mg Ϫ1 ) and 2-arachidonoyl LPC (0.3 mol⅐min Ϫ1 ⅐mg Ϫ1 ) with the subsequent release of palmitic and arachidonic acid. This result is in accordance with the previous results from kinetic studies of iPLA 2 ␤ (48). To examine the substrate selectivity of cPLA 2 ␣, the cytosolic fraction of Sf9 cells expressing cPLA 2 ␣ was assayed employing mass spectrometry in the presence of CaCl 2 utilizing PAPC liposomes. Only 1-palmitoyl LPC, without detectable amounts of 2-arachidonoyl LPC, was produced during incubations of the cytosol of Sf9 cells expressing cPLA 2 ␣, accompanied by stoichiometric amounts of arachidonic acid release (6.7 nmol⅐min Ϫ1 ⅐mg Ϫ1 ), which was not observed at comparable initial rates with pFB control cytosol.
To further substantiate the unique molecular selectivity of iPLA 2 ␥, hydrolysis of liposomes containing equimolar amounts of PAPC and 1-stearoyl-2-oleoyl phosphatidylcholine (SOPC) was examined by mass spectrometry. ESI/MS spectra of PC and LPC were acquired in the positive ion mode as a function of time after addition of purified iPLA 2 ␥ (Fig. 8, A-C). Spectra were also acquired in the negative ion mode to measure the initial rate of fatty acid release (Fig. 8, D-F). SOPC was hydrolyzed at a similar rate as PAPC, as measured by similar overall decreases in the amounts of each substrate as a function of time (Fig. 9A). For SOPC, the intrinsic PLA 1 and PLA 2 activity of iPLA 2 ␥ resulted in the production of both 18:1 and 18:0 LPC. Palmitic, stearic, and oleic acids were all rapidly released with  only diminutive amounts of arachidonic acid produced. The degree of accumulation of LPC species followed the rank order 20:4 Ͼ 18:1 Ϸ 18:0 Ͼ 16:0 (Fig. 9A), which was closely correlated with the observed release of fatty acid species (Fig. 9B). Collectively, these results indicated that iPLA 2 ␥ is capable of hydrolyzing saturated or mono-unsaturated acyl chains from either the sn-1 or sn-2 position of the phosphatidylcholine species (e.g. SOPC hydrolysis). More importantly, the results from incubations of pure (PAPC or PDPC) and mixed (PAPC/SOPC) phosphatidylcholine liposomal substrates with iPLA 2 ␥ also demonstrated that iPLA 2 ␥ possesses a unique capacity to se-lectively catalyze the generation of 2-polyunsaturated LPC species, including 2-arachidonoyl LPC.
Intrinsic PLA 2 and Lysophospholipase Activity of Purified iPLA 2 ␥-To further investigate the substrate selectivity of iPLA 2 ␥ for polyunsaturated acyl chains at the sn-2 position of the substrate, hydrolysis of mixed liposomes containing diacyl and plasmalogen PC molecular species were examined. Liposomes containing equimolar amounts of PAPC and 1-O-1Ј-(Z)hexadecenyl-2-arachidonoyl-sn-glycero-3-phosphocholine (16:0 -20:4, plasmenylcholine) were incubated with iPLA 2 ␥, and the accumulation of reaction products was analyzed by ESI/MS. The robust release of arachidonic acid from 16:0 -20:4 plasmenylcholine was demonstrated in concert with an increase in the lysoplasmenylcholine peak ([MϩNa], ϩ m/z 502), indicating that iPLA 2 ␥ is capable of effectively hydrolyzing the sn-2 arachidonate ester in plasmalogen substrate (Fig. 10). In contrast, hydrolysis of the diacyl substrate, PAPC, resulted in the accumulation of 2-arachidonoyl LPC and only modest amounts of palmitoyl LPC (Fig. 10). The initial rates of the iPLA 2 ␥-catalyzed hydrolysis of mixed liposomes containing equimolar amounts of diacyl PAPC and 16:0 -20:4 plasmenylcholine were similar for both substrates. Thus, these results demonstrated that purified iPLA 2 ␥ can catalyze the hydrolysis of PC plasmalogen species (based on its intrinsic PLA 2 activity) as well as the hydrolysis of diacyl PC substrates (predominantly from its PLA 1 activity). Moreover, when incubated with arachidonate-containing plasmalogen substrate, unlike with diacyl PAPC, iPLA 2 ␥ is capable of effectively releasing arachidonic acid directly from sn-2 position in a bilayer comprised of mixed phospholipid substrates.
When assayed with 1-palmitoyl LPC substrate, iPLA 2 ␥ showed a lysophospholipase activity of Ϸ120 nmol⅐min Ϫ1 ⅐mg Ϫ1 (Table II). Remarkably, iPLA 2 ␥ hydrolyzed the same concen- were prepared by sonication, and the lipid substrate was incubated with purified iPLA 2 ␥ (0.5 g) at 37°C for the indicated times as described under "Experimental Procedures." The reaction was terminated by addition of chloroform/methanol (1:1) containing di-14:1 PC, 17:0 LPC, and d4 -16:0 fatty acid as internal standards. All lipids were extracted into chloroform by the method of Bligh and Dyer. Lipid extracts were diluted by chloroform/methanol (1:1) to a total lipid concentration of ϳ2 pmol/l prior to direct infusion into an ESI source using a syringe pump at a flow rate of 2 l/min. Concentrations of PAPC and LPC at indicated time points after addition of iPLA 2 ␥ were calculated from positive ion ESI mass spectra (Fig. 5, A-C) by comparing their sodiated ion peak intensities to those of the corresponding internal standards (di14:1 PC and 17:0 LPC) after correction for 13 C isotopomer differences. Similarly, the concentrations of released fatty acids were calculated from negative ion ESI mass spectra (Fig. 5, D-F) by comparing the [M-H] Ϫ peak intensities of fatty acids to that of the internal standard (d4 -16:0 fatty acid) after correction for 13 C isotopomer differences. Results are representative of three independent experiments. iPLA 2 ␥-generated Arachidonoyl Lysolipids tration of either 1-or 2-arachidonoyl LPC at a much slower rate (Ϸ20 and 5 nmol⅐min Ϫ1 ⅐mg Ϫ1 , respectively, Table II). In contrast, iPLA 2 ␤ showed much higher lysophospholipase activity toward either saturated (i.e. palmitoyl) or polyunsaturated (i.e. 1-or 2-arachidonoyl) LPC species (Ϸ180, 150, and 100 nmol⅐min Ϫ1 ⅐mg Ϫ1 , respectively, Table II). Thus, the nascent 2-arachidonoyl LPC, generated by rapid cleavage of sn-1-palmitoyl chain during iPLA 2 ␥-mediated PAPC hydrolysis, was not significantly further metabolized by iPLA 2 ␥, and the selectivity of the lysophospholipase activity of iPLA 2 ␥ (saturated palmitoyl palmitoyl LPC Ͼ Ͼ polyunsaturated arachidonoyl LPC) contributed to the accumulation of 2-arachidonoyl LPC.
Identification of Arachidonoyl LPC as a Natural Product in Human Heart Tissue-Previously, the mRNA of iPLA 2 ␥ was shown to have the highest abundance in human heart tissue in comparison to multiple other tissues (3). Accordingly, to gain insight into the biological relevance of the unique substrate selectivity of iPLA 2 ␥ to produce 2-polyunsaturated LPC species, we analyzed human myocardial lipids by tandem mass spectrometry. The mass spectrometric analysis of human heart explant tissue clearly indicated arachidonoyl LPC (m/z 566) as a major constituent in the LPC pool (Fig. 11). Thus, arachidonoyl LPC is a natural product in human myocardium, a prominent site of iPLA 2 ␥ gene expression.
Identification of PLA 2 ␥-catalyzed Accumulation of Arachidonoyl LPC in Naturally Occurring Membrane Systems-Because the localization of the 63-kDa iPLA 2 ␥ has previously been documented in hepatic peroxisomes (5), additional experiments were performed to determine if a similar substrate selectivity was present utilizing native peroxisomal membranes. Rat liver peroxisomes were purified by buoyant density gradient centrifugation (5) and were incubated with purified iPLA 2 ␥. Tandem MS spectrometric analyses demonstrated the presence of a small amount of arachidonoyl LPC (m/z 566) in peroxisomes at the beginning of the reaction (Fig. 12A). After the incubation of peroxisomes with iPLA 2 ␥, the arachidonoyl LPC became the predominant LPC species produced during peroxisomal membrane phospholipid hydrolysis (Fig. 12, B and C). Moreover, product ion mass spectrometry of the arachidonoyl LPC product, based on its diagnostic fragmentation patterns, demonstrated that the arachidonoyl moiety was initially present at the sn-2 linkage of the glycerol backbone (data not shown). These results also indicated substantial production of docosahexaenoyl LPC (m/z 590) from the peroxisomes after incubation with iPLA 2 ␥, which was in agreement with the previously determined substrate selectivity of iPLA 2 ␥ to generate polyunsaturated lysolipids from bilayer membranes. Thus, these results confirmed that iPLA 2 ␥ is able to mediate the selective production of 2-polyunsaturated LPC in its natural membrane context. DISCUSSION The present study describes the purification of the 63-kDa peroxisomal iPLA 2 ␥ and identifies its novel substrate selectivity that results in the selective synthesis of a key branch point metabolite in eicosanoid signaling, 2-arachidonoyl LPC. These studies employed ESI/MS and tandem mass spectrometry to unambiguously identify the regiospecific lipid products, which are difficult to observe utilizing conventional assay methods. Detailed kinetic analyses of the substrate selectivity of highly purified iPLA 2 ␥ utilizing both synthetic phospholipids as well as naturally occurring bilayer membrane substrates, which differed in their acyl substituents, phospholipid subclasses, and membrane physical properties, could be achieved by this method. The co-purification of HSP70 and 63-kDa iPLA 2 ␥ (Table I) likely reflects the tight association of these two proteins. HSP70 is known to participate in peroxisomal import by regulating the interaction between the peroxisomal targeting signal type 1 of cargo proteins and the peroxisomal targeting signal type 1 receptor (e.g. Pex5p) (51). Because iPLA 2 ␥ contains a C-terminal peroxisomal targeting signal type 1 sequence (-SKL) and was previously shown to be enriched in peroxisomes (5), it is likely that the import of iPLA 2 ␥ into peroxisomes is regulated in part by the molecular chaperone HSP70 in vivo.
Purified iPLA 2 ␥ readily released saturated and mono-unsaturated fatty acids at nearly equal rates from either the sn-1 or sn-2 position in diacyl phosphatidylcholine substrates (e.g. SOPC, Figs. 8 and 9). In contrast, iPLA 2 ␥ was largely unable to cleave polyunsaturated fatty acids from the sn-2 position of diacyl phosphatidylcholine, and the rapid hydrolysis of the sn-1-saturated fatty acyl ester resulted in the production of 2-polyunsaturated LPC (Figs. 5 and 6). We point out that this observation is unique to iPLA 2 ␥, because purified iPLA 2 ␤ and the cytosol from Sf9 cells expressing cPLA␣ did not demonstrate the same selectivity. Clearly, iPLA 2 ␤ showed mixed PLA 1 /PLA 2 activities toward PAPC, and the resultant 1-palmitoyl LPC that is produced appears at an initial rate ϳ3-fold greater than that of the 2-arachidonoyl LPC. The predominant FIG. 7. Identification of the nascent arachidonoyl LPC produced by iPLA 2 ␥ as the 2-acyl regioisomer. Tandem MS spectra of the sodiated adduct ion of high-performance liquid chromatography purified 1-acyl and 2-acyl arachidonoyl LPC (A and B, respectively), and that of the arachidonoyl LPC produced after 2 min of iPLA 2 ␥-catalyzed PAPC hydrolysis (C) were acquired under positive ion mode with a collision energy of 30 eV and a collision gas pressure of 1.0 mTorr. The samples were extracted by the method of Bligh and Dyer, and were diluted with chloroform/methanol (1:1) prior to direct infusion into an ESI source using a syringe pump at a flow rate of 2 l/min. The spectra were normally acquired within 3 h after the arachidonoyl LPC was generated. Signals were averaged over 2 min of acquisition time. The m/z values of all peaks were rounded to the nearest integer. products identified from PAPC hydrolysis catalyzed by cytosolic cPLA 2 ␣ in this report are 1-palmitoyl LPC and arachidonic acid, indicating cPLA 2 ␣ is acting exclusively as a PLA 2 enzyme toward PAPC in accordance with previous observations (52). By comparison, iPLA 2 ␥ catalyzed the production of 2-arachidonoyl LPC at least 10-fold faster than that of the 1-palmitoyl LPC (Fig. 6) from PAPC hydrolysis. Cytosolic PLA 2 ␥ (53, 54) also has been shown not to function as a PLA 1 enzyme (37), although its hydrolytic activity toward phospholipids is calcium-independent (37,(53)(54)(55). Multiple PLA 1 activities have been previously reported, including the purified PLA 1 from bovine brain (56), bonito muscle (57), as well as phosphatidylserine-selective (58) and phosphatidic acid-selective (59) PLA 1 enzymes. In contrast to the strict sn-1 regiospecificity and lack of selectivity toward sn-2 acyl groups (56,57) of the reported PLA 1 enzymes, iPLA 2 ␥ is distinct from these enzymes in that it possesses both PLA 1 (Fig. 5) and PLA 2 ( Fig. 10) activities, and that it functions predominantly as a PLA 1 enzyme in the presence of diacyl PC substrates containing an sn-2 polyunsaturated acyl group (Fig. 5).
The lysophospholipase activity of iPLA 2 ␥ toward lysophospholipid substrates (e.g. palmitoyl LPC) with saturated acyl chains was robust (Table II). The substantial lysophospholipase activity of iPLA 2 ␥ using palmitoyl LPC was similar to that manifested by iPLA 2 ␤ (Table II) and cPLA 2 ␣, which hydrolyzes lysolipids rapidly (27). However, although iPLA 2 ␤ (Table II) and cPLA 2 ␣ (25) readily hydrolyze lysolipids containing arachidonic acid at the sn-2 position, iPLA 2 ␥ is inefficient toward these substrates by comparison (Table II). Therefore, cPLA 2 ␣ possesses high selectivity for the hydrolysis of either phospholipids (6,7) or lysophospholipids (25) possessing arachidonic acid at the sn-2 position, whereas iPLA 2 ␥ readily hydrolyzes diacyl phospholipids at the sn-2 position except those that contain arachidonic acid or docosahexaenoic acid (Figs. 5 and 8). Thus, accumulation of 2-arachidonoyl LPC during iPLA 2 ␥-catalyzed PAPC hydrolysis (Fig. 5) is likely the result of the combined effects of rapid and selective 2-arachidonoyl LPC generation (via PLA 1 activity) and its much slower FIG. 8. Hydrolysis of mixed liposomes containing an equal molar ratio of PAPC and SOPC. Liposomes containing PAPC (15 M) and SOPC (15 M) and DAG (10 M) were incubated with purified iPLA 2 ␥ (0.5 g) at 37°C for the indicated times. All lipids were extracted into chloroform by the method of Bligh and Dyer and ESI mass spectrometric analyses were performed under the same conditions for PAPC hydrolysis (Fig. 5). and LPC species at indicated time points after addition of iPLA 2 ␥ were calculated from positive ion ESI mass spectra (Fig. 8, A-C) by comparing their sodiated ion peak intensities to those of the corresponding internal standards (di14:1 PC and 17:0 LPC) after correction for 13 C isotopomer differences. Similarly, concentrations of released fatty acids were calculated from negative ion ESI mass spectra (Fig. 8, D-F) by comparing the [M-H] Ϫ peak intensities of the fatty acids to that of the internal standard (d4 -16:0 fatty acid) after correction for 13 C isotopomer differences. Results are representative of three independent experiments. degradation (via lysophospholipase activity) by iPLA 2 ␥.
Although these results identified unanticipated substrate selectivities in model liposome systems, it was equally important to know their relevance in naturally occurring membranes, especially in the context of the native subcellular locations. First, we employed tandem mass spectrometry to demonstrate the presence of 2-arachidonoyl LPC as one of the major naturally occurring LPC molecular species in human heart (Fig. 11), which is a predominant tissue of iPLA 2 ␥ expression (3). Secondly, the 63kDa iPLA 2 ␥, a peroxisomal enzyme by virtue of its C-terminal -SKL localization motif (5), was shown to catalyze the hydrolysis of endogenous peroxisomal phosphatidylcholine species, which resulted in the robust production of the 2-arachidonoyl LPC, to make it the most abundant LPC metabolite (Fig. 12). Thus, the selective production of 2-arachidonoyl LPC from the hydrolysis of precursor PC species catalyzed by iPLA 2 ␥ observed in this study employing either artificial bilayers or naturally occurring membranes indicated that iPLA 2 ␥ likely mediates the accumulation of 2-arachidonoyl LPC in vivo. As far as we are aware, this is the first report of a phospholipase that has the capacity to selectively generate 2-arachidonoyl LPC.
LPC plays an important regulatory role in cellular signaling cascades as a lipid second messenger, likely functioning through ion channel (60, 61) and G-protein-coupled receptors (62-65), in a broad range of biological processes. More specifi- and DAG (10 M) were incubated with purified iPLA 2 ␥ (0.5 g) at 37°C for the indicated times. All lipids were extracted into chloroform by the method of Bligh and Dyer and the ESI mass spectrometric analyses were performed under the same conditions for PAPC hydrolysis (Fig. 5).   11. Arachidonoyl LPC is a natural product in human myocardium. Lipids from end-stage human heart tissue were extracted by the procedure of Bligh and Dyer with modifications as previously described (40) for ESI mass spectrometric analysis. Lipid extracts were diluted by chloroform/methanol (1:1) prior to direct infusion into an ESI source using a syringe pump at a flow rate of 2 l/min. A 2-min period of signal averaging was employed for spectrum acquisition. Scanning for the neutral loss of 59 atomic mass units in the positive ion mode at a collision energy of 24 eV and a collision gas pressure of 1.0 mTorr was employed to fingerprint LPC species ([MϩNa] ϩ ). Results are representative of three independent human heart samples. cally, evidence from multiple lines of investigation has suggested that 2-arachidonoyl LPC is an important branch point metabolite in cellular eicosanoid metabolism due to the existence of several intracellular catabolic enzymes that can provide a diverse repertoire of biologically active eicosanoid products from this lysolipid (Scheme I). First, the deacylation of 2-arachidonoyl LPC by highly active lysophospholipases (25) is an efficient way of generating non-esterified arachidonic acid (26). Thus, release of arachidonic acid from the sn-2 position of phospholipids (e.g. PAPC) could result from the sequential actions of the PLA 1 activity of iPLA 2 ␥ followed by the lysophospholipase activity of either cPLA 2 ␣ (25, 27), iPLA 2 ␤ (this study), or other lysophospholipases. Whatever the case, the sequential deacylation reactions catalyzed by iPLA 2 ␥ and lysophospholipase in certain tissues (e.g. heart) would lead to the highly selective release of arachidonic acid in a two-step process exploiting the unusual specificity of iPLA 2 ␥ uncovered in this study. Mass spectrometry unambiguously demonstrated that iPLA 2 ␥ can also hydrolyze phospholipids directly from the sn-2 position using plasmalogens containing arachidonic acid at the sn-2 position as substrate (Fig. 10). The results pre-sented herein demonstrate the versatility of iPLA 2 ␥ in mediating direct and indirect arachidonic acid release in a substratedependent manner, i.e. iPLA 2 ␥ hydrolyzes arachidonatecontaining plasmalogens or diacyl lipids through PLA 2 or PLA 1 activities, respectively, to release different forms of the signaling arachidonoyl species (i.e. free arachidonic acid or arachidonoyl LPC).
Previous work by Waku et al. (36) has demonstrated the presence of a mammalian lysophospholipase C activity, which can convert 2-arachidonoyl LPC into 2-arachidonoylglycerol (2-AG). The biological role of 2-AG as an endocannabinoid signaling molecule has been demonstrated by the binding of this endogenous lipid metabolite to cannabinoid receptors (28,29) and by the subsequent initiation of downstream events (30). Moreover, 2-AG is an effective substrate for cyclooxygenase-2 (but not cyclooxygenase-1) (31) and the resultant glycerollinked prostaglandin derivatives (32) have recently been shown by Marnett and colleagues (33) to have unique signaling functions in RAW 264.7 cells. 2-AG can also serve as substrate for 15-lipoxygenases, resulting in the production of 15-hydroperoxy metabolites that can be utilized for specific signaling functions (34). Thus, it seems likely that iPLA 2 ␥ can participate in the generation of endocannabinoid and glycerol-linked oxygenated eicosanoid metabolites through selective production of 2-arachidonoyl LPC and subsequent conversion of this lysolipid into 2-AG. Finally, 2-arachidonoyl LPC can be transported to other cellular destinations for further metabolism and/or signal transduction functions since the "off-rate" for lysolipids leaving their membrane bilayer of generation greatly exceeds that of fatty acids (66). It has been reported that the uptake of liver derived 2-arachidonoyl LPC from blood may be one of the most important arachidonate sources for extrahepatic tissues in rat (67).
Collectively, these results identify multiple biologic products and signaling pathways that can be initiated by iPLA 2 ␥ in mammalian cells through the use of parallel pathways for ligand-stimulated phospholipid hydrolysis for the production of lipid second messengers. The diversity of products generated through this pathway underscore the complexity of possible hydrolytic events contributing to eicosanoid signaling in mammalian cells. Although the direct release of arachidonic acid from the sn-2 position is the most straightforward process, it seems clear that many other competing and complementary pathways contribute to the repertoire of lipid second messen- FIG. 12. Selective accumulation of polyunsaturated LPC during iPLA 2 ␥-mediated hydrolysis of phosphatidylcholine species in isolated rat liver peroxisomes. Rat liver peroxisomes were purified as previously described (5) and were incubated with purified iPLA 2 ␥ (0.5 g) at 37°C. Reactions were terminated by addition of chloroform/methanol (1:1, v/v) containing PC, LPC, and fatty acid internal standards. Extraction of lipids into chloroform by the method of Bligh and Dyer and the ESI mass spectrometric analyses were performed under similar conditions to those for PAPC hydrolysis (Fig. 5). Scanning for the neutral loss of 59 atomic mass units in the positive ion mode at a collision energy of 24 eV and a collision gas pressure of 1.0 mTorr was employed to fingerprint LPC species ([MϩNa] ϩ ) at the start of hydrolysis A, and after 10 min of hydrolysis (B). C, increase in the concentrations of major LPC species during iPLA 2 ␥-catalyzed PC hydrolysis of peroxisomal membranes. Concentrations of LPC species at the indicated time points (0 and 10 min) after the addition of iPLA 2 ␥ were calculated by comparing their sodiated ion peak intensities to that of the internal standard (17:0 LPC) after correction for 13 C isotopomer differences. Values shown are means Ϯ S.D. of separate determinations from three liver samples. SCHEME 1. Signaling pathways of 2-arachidonoyl LPC. iPLA 2 ␥-generated Arachidonoyl Lysolipids gers in mammalian cells that allow the amplification and diversification of phospholipase initiated signaling products.