Secretory Group IIA Phospholipase A2 Generates Anti-apoptotic Survival Signals in Kidney Fibroblasts*

Mammalian group IIA phospholipase A2 (PLA2) is believed to play important roles in inflammation, cell injury, and tumor resistance. However, the cellular site of action has not been clearly defined as it has long been recognized that group IIA PLA2 is both a secretory and mitochondrial protein. The purpose of this study was to determine the subcellular target of the group IIA PLA2 and its role in apoptosis stimulated by growth factor withdrawal. Cloning of the rat liver group IIA PLA2 demonstrated a typical secretory signal and no alternative splicing of the primary transcript. When a sequence including the signal peptide and first 8 residues in the mature enzyme or the entire PLA2 (including the signal peptide) was fused to enhanced green fluorescent protein, the fusion protein was directed to the secretory pathway rather than mitochondria in baby hamster kidney (BHK) cells. To examine the role of group IIA PLA2 in cell injury, wild type (wt) rat group IIA PLA2 and a mutant group IIA PLA2 containing a His-47 → Gln mutation (at the catalytic center) were transfected into BHK cells and cells stably expressing these constructs were isolated. After deprivation of growth factors, both normal BHK cells and BHK cells expressing mutant PLA2 underwent massive apoptosis, while BHK cells expressing wt PLA2 showed considerable resistance to growth factor withdrawal-induced apoptosis. The secretory PLA2 inhibitors 12-epi-scalaradial and aristolochic acid abrogated resistance to apoptosis in the wt PLA2 expressing cells. These two inhibitors did not induce cell death in the presence of fetal bovine serum, suggesting that they induce cell death by blocking PLA2 generated survival signals. This study demonstrates that group IIA PLA2generates anti-apoptotic survival signals in BHK cells targeting the secretory pathway, and suggests that high levels of group IIA PLA2 accumulated at inflammatory sites may not only regulate inflammation, but also may protect cells from unnecessary death induced by pro-inflammatory agents.

Mammalian group IIA phospholipase A 2 (PLA 2 ) 1 catalyzes the release of sn-2 fatty acid from membrane phospholipids.
This reaction produces arachidonic acid (AA) and lysophosphatidic acid (LPA) (1)(2)(3)(4)(5), both of which are multifunctional lipid mediators. Group IIA PLA 2 also has anti-bacterial activity and functions as a modifier of tumor multiplicity (6 -8). In addition to being a secretory/plasma membrane-associated protein (9), group IIA PLA 2 has been isolated from purified mitochondria (10 -13), and has been suggested to be localized at inner and outer membrane contact sites (14). Sequence analysis of this mitochondrial group IIA PLA 2 reveals the protein to be nearly identical to the mature secretory enzyme (15), indicating that the pre-protein of mitochondrial group IIA PLA 2 is likely processed at the exactly the same site as the secretory group IIA PLA 2. As discussed in a recent review, mitochondria are thought to be a target of group IIA PLA 2 (2). Using PLA 2 inhibitors and antisense oligonucleotides specific for group IIA PLA 2 , we found that group IIA PLA 2 contributed to cell injury during chemical hypoxia in isolated rat hepatocytes (16,17). Mitochondrial injury is a critical step in cell death during chemical hypoxia (18). Data from our laboratory as well as others have demonstrated that products of PLA 2 catalyzed reactions can induce the mitochondrial permeability transition (MPT), and reagents known to inhibit PLA 2 activity can block induction of the MPT in isolated mitochondria (19,20). Thus, the mitochondria appear to be a logical target of group IIA PLA 2 where it could induce the MPT and onset of cell death.
The recent discovery of other low molecular PLA 2 s (i.e. group V PLA 2 ) has drawn into question some of the functions previously attributed to group IIA PLA 2 (2). For example, late phase release of AA during mast cell activation was believed to be mediated by group IIA PLA 2, but has recently been found to be normal in a cell line lacking demonstrable group IIA PLA 2 activity (21). Certain reagents used to modulate group IIA PLA 2 activity such as inhibitors, antisense oligonucleotides, and antibodies thought to be specific to group IIA PLA 2 , have now been shown to act nonspecifically via group V PLA 2 . Given these concerns, the most direct way of elucidating group IIA PLA 2 function is to study cellular phenotype changes after expression of group IIA PLA 2 and its mutants. In this study, we generated cells stably expressing rat group IIA PLA 2 and its His-47 (a residue essential for its catalytic activity) to Gln mutant (22). We found that the group IIA PLA 2 generated anti-apoptotic survival signals in BHK cells. This finding sheds new light on the understanding of the function of group IIA PLA 2 in inflammation and other pathological processes.

EXPERIMENTAL PROCEDURES
Vectors and Reagents-Enhanced GFP (EGFP) fusion vector pEGFP-N1, expression vector pIRES1neo, Quick Clone rat liver cDNA, and ApoAlert DNA Fragmentation Assay Kit were purchased from CLON-* 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  Vector Construction-The PCR amplification program used was as follows: 94°C 5 min, 50°C 2 min, 72°C, 3 min (first cycle); 94°C 1 min, 50°C 2 min, 72°C, 3 min (30 cycles), 94°C 1 min, 50°C, 2 min, 72°C 10 min (last cycle). To clone the sequence encoding the mature rat liver group IIA PLA 2 protein, PCR reaction was performed using upstream primer F3B, 5Ј-AATTCAGCTCGAGATGAGCCTTCTGGAGTTTGGGC-AAATG-3Ј and downstream primer R1B, 5Ј-CGAAGATGCTGCAGTA-AGCAACTGGGCGTCTTCCCTTTGCAAAACTTGTTGGGGTAGAA-C-3Ј and rat liver cDNA as template. The fragment was cut with XhoI/PstI and inserted into pEGFP-N1. The resulting plasmid was named pF3B. To clone the complete cDNA sequence of the rat group IIA PLA 2 , PCR reaction was performed using upstream primer F5, 5Ј-GA-GAGCGGCCGCATGAAGGTCCTCCTGTTGCTAGCAGT-3Ј and downstream primer R1D, 5Ј-GATGTCGAATTCAGCAACTGGGCGTCTTCC-CTTTGC-3Ј and rat liver cDNA as template. The PCR fragment was digested with NotI/EcoRI and inserted into pIRES1neo, resulting in pPLA 2 . To create an in-frame fusion between the full-length PLA 2 and EGFP, PCR reaction was performed using upstream primer F4B 5Ј-A-ATTCAGCTCGAGATGAGCCTTCTGGAGTTTGGGCAAATG-3Ј and downstream primer R1B and rat liver cDNA as template. The PCR fragment was digested with PstI/XhoI, then cloned into PstI/XhoI sites upstream of the N terminus of EGFP in pEGFP-N1. The resulting plasmid was named pF4B. The PLA 2 -EGFP coding region was removed by digestion with NdeI (a site in the CMV promoter upstream of the PLA 2 -EGFP region) and NotI, and was cloned into pIRES1neo, resulting in pFPG. To construct MHIS that carries a His-47 3 Gln mutation in PLA 2 (22), we used the upstream primer MHISF 5Ј-CAGAGGATC-CCCCAAGGATGCCACAGATTGGTGCTGTGTGACTCAAGACTGTT-GTTAC-3Ј, which overlaps the single BamHI site in the PLA 2 and carried a CAT (His) to CAA (Gln), and the downstream primer MHISR 5Ј-ATGGTGGCGACCGGTGGATCCCGGGCCCGC-3Ј and pF4B as template in the PCR reaction. The fragment carrying the Gln mutation was digested with BamHI and recovered from agarose gel. The His-47 in pF4B was then replaced with Gln using the following procedure: pF4B was digested with BamHI to remove the fragment carrying His-47. The remaining vector sequence (including part of the PLA 2 sequence) was then ligated to the fragment carrying the mutation at 16°C for 1 h. PCR was then performed using F4B and R1B as primers with the ligation reaction as template. The fragment corresponding to the correct size was recovered from the agarose gel, digested with XhoI/PstI, and cloned into pEGFP-N1. The resulting plasmid was named pMHIS. To clone the signal peptide sequence, PCR was performed using primer SIGF 5Ј-GTGTATCATATGCCAAGTACG-3Ј and primer SIGR 5Ј-CAGAATCTGCAGCCCAAACTCCAGAAGGCTCCCC-TGGACCTGAATTGAG-3Ј and pFPG as template. The fragment was recovered and digested with NdeI/PstI, then cloned into the corresponding sites in pF4B, resulting in in-frame fusion with EGFP coding region. The resulting plasmid was named pSIG/G. To generate the group IIA PLA 2 mutants carrying a His-47 3 Gln mutation, PCR was performed using F5 and R1D primers and pMIHS template. The fragment was cloned as described for pPLA 2 . In all the fusion proteins, the stop codon of PLA 2 (TGA) was changed to TTA which encodes a Leu residue. The sequences of all PCR-derived constructs were confirmed.
Transfection-Baby hamster kidney (BHK) cells were cultured at 37°C in 5% CO 2 in Dulbecco's modified Eagle's medium (DMEM) (Life Technologies, Inc., Gaitherburg, MD) containing 10% fetal bovine serum (FBS, Sigma). In some experiments, DMEM/F-12 (Life Technologies, Inc.) was used. For transient expression, cells were grown on coverslips in a 35-mm culture dish to about 50% confluence. In a typical transfection experiment, 2 g of DNA was mixed with 100 l of medium (FBS-and antibiotic-free) and 25 l of SuperFect Transfection Reagent (QIAGEN) and incubated at room temperature for 30 min. 1 ml of medium containing 10% FBS was added and the mixture immediately added to the cells. Following incubation at 37°C for 2 h, the cells were washed with phosphate-buffered saline, then cultured in fresh medium. EGFP fluorescence usually could be observed after overnight incubation. To obtain stable clones, Geneticin (500 g/ml, Life Technologies, Inc.) was added to medium 2-3 days post-transfection. The cells were selected for 2 weeks.
RNA Isolation and Reverse Transcriptase (RT)-PCR-RNA was isolated from cells using High Pure RNA Isolation Kit (Roche Molecular Biochemicals) according to the manufacturer's manual. RT-PCR was performed using Access RT-PCR System (Promega). A 50-l reaction contained 25 l of nuclease-free water, 10 l of avian myeloblastosis virus/Tf1 5 ϫ buffer, 1 l of 10 mM dNTP mixture, 5 l of F5 primer (4 M), 5 l of R1D primer (4 M), 2 l of 25 mM MgSO 4, 1 l of avian myeloblastosis virus reverse transcriptase (5 units/l), 1 l of Tf1 DNA polymerase (5 units/l), and 2 l of RNA. Reverse transcription was carried out at 48°C for 1 h. After incubation at 95°C for 4 min to inactivate reverse transcriptase, PCR amplification was performed using 95°C 1 min, 60°C 1 min, 70°C 1 min for 30 cycles. The DNA was separated on 1% agarose gel.
Fluorescent Microscopy-Final concentrations of the probes were 10 nM for MitoTracker and 15 M for BODIPY TR ceramide. To stain organelles, cells were cultured in medium containing the appropriate probes for 10 -30 min, then washed three times with phosphate-buffered saline. The coverslips were then mounted on slides and sealed with rubber cement. Fluorescence was examined with Olympus IX-70 microscope and recorded with an ORCA100 CCD camera (Hamamatsu). The filter settings for EGFP fluorescence were a 488/20 nm excitation bandpass filter, a 495-nm dichroic mirror, and a 513/30 nm emission filter. The filter settings for red fluorescence (MitoTracker, BODIPY TR ceramide) were a 530 -550 nm excitation band-pass filter, a 570-nm dichroic mirror, and a 590-nm (long pass) emission filter.
Cell Viability and Apoptosis Detection-Cell viability was determined with CELLTiter 96 AQueous One Solution Cell Proliferation Assay (Promega) according the manufacture's instruction. DNA ladders were detected either with Apoptotic DNA Ladder kit (Roche Molecular Biochemicals) or with partial lysis of plasma membrane with 0.5% Triton X-100 with some modification (23) and preparing low molecular DNA. TUNEL assay was performed using ApoAlert DNA fragmentation Assay Kit (CLONTECH).

Cloning and Characterization of Rat Group IIA PLA 2 -
The full-length sequence (including signal peptide) of secretory group IIA PLA 2 from rat platelets (secretory/membrane-associated) is available in GenBank (accession number D00523). Rat liver group IIA PLA 2 has been reported to be localized in mitochondria (12), and the sequence of the mature liver/mitochondrial group IIA PLA 2 is also available in GenBank (accession number X74364). The sequence of these two proteins (rat platelet secretory group IIA PLA 2 versus rat liver group IIA PLA 2 ) differs by only 1 amino acid residue, i.e. Arg-94 in the liver/mitochondrial group IIA PLA 2 versus Ala-94 in the secretory group IIA PLA 2 (15,24,25).
Because the presequence of the mitochondrial group IIA PLA 2 has not been identified, it is unclear whether these two proteins contain the same signal peptide, and whether group IIA PLA 2 can localize to both mitochondria and secretory pathways. Based on the N-and C-terminal sequence of the liver/ mitochondrial group IIA PLA 2 , we cloned the cDNA sequence encoding mature PLA 2 from rat liver by PCR amplification. The sequence we cloned was identical to that of the secretory group IIA PLA 2 including an Ala at position 94. It is likely that the inaccuracy of earlier versions of Taq polymerase caused the previously reported difference in the sequence of mitochindrial versus secretory PLA 2 . Using primers corresponding to the signal peptide sequence of the secretory PLA 2 and the COOH terminus of mitochondrial/secretory group IIA PLA 2 , we amplified the full-length cDNA sequence of the group IIA PLA 2 from rat liver. The sequence, including signal peptide, was again identical to that of the rat secretory group IIA PLA 2 . We then examined whether there were any alternatively spliced forms of group IIA PLA 2 that may bear different presequence/or signal peptide. Using primers corresponding to the 5Ј terminus of the primary transcript (PSTART in Fig. 1A) (26) and to the C terminus of mitochondrial/secretory group IIA PLA 2 (R1B) to amplify any possible alternatively spliced mRNAs, only one band was detected in the reaction product (Fig. 1B). To examine whether this band contained more than one DNA fragment, the band was recovered and subjected to direct sequencing using PSTART and R1B as primers. No heterogeneity was found (Fig. 1C). Therefore, we conclude that there is only one group IIA PLA 2 molecule in rat liver. The data also indicate that before post-translational processing, mitochondrial group IIA PLA 2 bears the same signal peptide as the secretory group IIA PLA 2 does.
To examine whether rat liver group IIA PLA 2 targets to mitochondria, we fused the N-terminal 29 residues (including the 21-residue signal peptide and 8 residues of the mature enzyme) to the N terminus of EGFP (SIG/G in Fig. 2). When expressed in BHK cells, the SIG/G fusion protein localized mainly at one pole of the nucleus (Fig. 3B), which is characteristic for the Golgi body. In contrast, when expressed alone EGFP was evenly distributed in the cytosol (Fig. 3A). In some cells, fusion proteins were also found in typical secretory organelles, including Golgi stacks (Fig. 3C), vesicles detaching from Golgi body (Fig. 3D), endoplasmic reticulum (Fig. 3E), and vesicles (Fig. 3F). This reflects the different stages of secretory protein undergoing the maturation-secretion process.
We also generated fusion proteins composed of the fulllength rat group IIA PLA 2 (including signal peptide) and EGFP (FPG or MHIS in Fig. 2). MHIS is identical to FPG except it contains a His-47 3 Gln mutation. When expressed in BHK cells, these fusion proteins also displayed a distribution pattern typical for a secretory protein. Fig. 4, A and B (FPG) and C (MHIS), show the fusion protein localized mainly in vesicles. Fig. 4, D-F (FPG), show the localization of FPG fusion protein in the Golgi body. Fig. 4, G-I (FPG), demonstrate that the fusion protein does not localize in mitochondria. From these results, it is concluded that the rat liver group IIA PLA 2 is a secretory protein. Therefore, during cell injury, group IIA PLA 2 cannot directly regulate mitochondrial function and induce the MPT. Thus, all of the physiological effects elicited by rat liver group IIA PLA 2 should be considered as the result of its interaction on the membranes of the secretory pathways, and in particular, the plasma membrane. Group IIA PLA 2 Generates Anti-apoptotic Survival Signals in BHK Cells-To study the biological function of the secretory group IIA PLA 2 , we generated cells stably expressing this gene. For this purpose, we used a bicistronic expression system (pIRES1neo, CLONTECH). In this system, a single cytomegalovirus promoter directs the transcription of mRNA consisting of both the group IIA PLA 2 and the selection marker neo r . The group IIA PLA 2 is located upstream of neo r in the bicistronic transcript. Because the two genes are simultaneously expressed, after transfection and selection nearly all the G418resistant cells express the group IIA PLA 2 . This allowed us to rapidly obtain a pool of PLA 2 expressing cells without cloning. A total of 35 G418-resistant clones were obtained. After harvesting and pooling of these cells (NP), the expression of rat group IIA PLA 2 was confirmed (Fig. 5A). To our surprise, stable expression of group IIA PLA 2 was not associated with cytotoxicity. In fact, when cultured in serum-free medium, normal BHK cells underwent massive apoptosis, while apoptosis in the NP cells was significantly lower (Fig. 5B).
To examine whether resistance to apoptosis of the NP cells was due to expression of the rat group IIA PLA 2 , we generated a mutant PLA 2 , in which His-47, the residue essential for the hydrolytic activity (by binding of an H 2 O molecule to the catalytic center) of PLA 2 , was mutated to Gln (Fig. 2) (22). A pool of cells (HQ) expressing the His-47 mutant was obtained using the same method as described for the NP cells, and the expression of the mutant PLA 2 was confirmed (Fig. 5A). As shown in Fig. 5, C and D, the HQ cells underwent massive apoptosis after withdrawal of FBS.
We then compared the viability of BHK, HQ, and NP cells in medium lacking growth factors. Fig. 6, A-C, show NP cells survived better than BHK and HQ cells in serum-free medium.
Morphological changes observed in serum-free medium is also more substantial for normal BHK and HQ cells compared with NP cells (Fig. 6D).
If resistance to apoptosis of NP cells is caused by rat group IIA PLA 2 expression, then secretory PLA 2 inhibitors should be able to abrogate this effect. 12-epi-Scalaradial (SCA) is a specific inhibitor for secretory PLA 2 (27)(28)(29). A dose-dependent inhibition of cell survival in serum-free medium was found in NP cells in the presence of SCA (Fig. 7A). SCA also inhibited the survival of BHK and HQ cells. A possible explanation is that it might abrogate the endogenous sPLA 2 function. Another widely used PLA 2 inhibitor, aristolochic acid (ArA) (30 -32), also abrogated the survival effect in NP cells (Fig. 7B). In the presence of FBS, these inhibitors did not have significant effects on cell survival (Fig. 7, C and D). Taken together, these results strongly suggest that expression of rat group IIA PLA 2 produces survival signals in this model system.
Group IIA PLA 2 is also known to generate lipid mediators including AA and LPA (5). To study whether they might mediate group IIA PLA 2 survival activity in the NP cells, we treated the NP cells with LPA and AA. At high concentration (30 M), AA itself induced cell death. At lower concentration, AA did not have apparent protective effects (data not shown). Similarly, LPA did not have any protective effect in either BHK or HQ cells, but it did increase cell survival of NP cells (Fig. 7E). Therefore, neither AA nor LPA can replace the signals elicited by rat group IIA PLA 2 .

DISCUSSION
The major goal of our study was to elucidate the role of rat liver group IIA PLA 2 in apoptosis induced by serum starvation. It is known that increased group IIA PLA 2 expression is asso- ciated with hypoxia/reperfusion injury in rat hepatocytes (16,17), and that the products of PLA 2 -catalyzed reactions can induce the MPT in isolated mitochondria. In addition, a number of papers have proposed that mitochondrial PLA 2 might induce the MPT following a variety of insults (33)(34)(35). These data, in conjunction with the report detailing the biochemical purification of mitochondrial localized group IIA PLA 2 from rat liver 10 years ago (12), support the hypothesis that group IIA FIG. 5. Detection of the rat group IIA PLA 2 expression and apoptosis. A, detection of rat group IIA PLA 2 expression in BHK, HQ, and NP cells by RT-PCR. B, DNA ladder. BHK and NP cells were recovered by trypsin digestion. After washing twice with serum-free medium, 2 ϫ 10 6 cells were plated into 35-mm dishes with DMEM/F12 medium and 0.5% FBS. After 3 days, DNA was isolated using Apoptotic DNA ladder kit and resolved on 1.2% agarose gel. MW1, /Hin-dIII marker. C, DNA ladder. After washing twice with serum-free medium, 10 7 cells were plated into 90-mm dishes in DMEM/F12 medium and 1% FBS. 24 h later, the cells were washed once with serum-free medium, and cultured in serum-free medium for 24 h. DNA ladder was prepared by Triton X-100 lysis. MW2, 123-base pair DNA ladder. D, TUNEL assay. HQ cells were washed twice with medium then plated onto glass coverslip in DMEM/F-12 medium and 1% FBS. 24 h later, the cells were washed once with medium, and cultured in serum-free medium. After overnight culture, DNA fragmentation was detected with ApoAlert DNA Fragmentation Assay Kit (CLON-TECH). a, fluorescent image (show DNA fragmentation). b, transmitted light image (ϫ 40 objective lens). PLA 2 is part of the injury pathway in cells and exerts this injury promoting activity by targeting the mitochondria. However, it has generally been accepted that mammalian group IIA PLA 2 can be both a secretory and mitochondrial protein (2,3,(12)(13)(14), and to date, most work aimed at elucidating the activity of group IIA PLA 2 have focussed on secretory group IIA PLA 2 . Because of this, we decided to visualize the location of this PLA 2 in normal and stressed cells using an EGFP-PLA 2 fusion protein to determine whether mitochondria are a target of group IIA PLA 2 .
We cloned group IIA PLA 2 from rat liver (reportedly localized in mitochondria (12)), and the sequence we cloned was identical to secretory group IIA PLA 2 . Furthermore, we did not find any alternatively spliced variants of the primary transcript of group IIA PLA 2 that might generate a different signal peptide. These results indicate that there is only one type of group IIA PLA 2 molecule in rat liver.
We then studied the subcellular localization of the rat group IIA PLA 2 . To accomplish this, we used EGFP as a tag to follow the maturation and intracellular location of the fusion protein in living cells. This method has been demonstrated to work for both secretory and mitochondrial proteins (36 -38). To our surprise, EGFP fusion protein consisting of either the signal peptide or full-length sequence of group IIA PLA 2 was found exclusively localized to the secretory pathways, but never in mitochondria. This distribution pattern is not cell-type (BHK) specific, as similar results were obtained in the mouse hepatocyte cell line AML12 (data not shown). Taken together, these data indicate that group IIA PLA 2 is not a mitochondrially localized protein. It is therefore unlikely that group IIA PLA 2 has any direct role in regulating mitochondrial function and onset of the MPT. Secretory group IIA PLA 2 is believed to regulate a number of physiological as well as pathological processes by generating AA and other lipid mediators. High levels of group IIA PLA 2 have been found at inflammatory sites and increased serum group IIA PLA 2 levels are thought to be related to the severity of sepsis (39 -41). In isolated hepatocytes, expression of group IIA PLA 2 was up-regulated during chemical hypoxia and its contribution to hypoxic injury has been demonstrated (17). Despite numerous in vitro data that suggest the important role of the group IIA PLA 2 in inflammation and cell injury, it should be noted that in vivo models of injury have not always supported such a destructive role for group IIA PLA 2 . For example, transgenic mice expressing high levels of group IIA PLA 2 do not develop any overt inflammatory conditions. No abnormalities were found in any tissue other than skin, where alopecia, epidermal hyperplasia, adnexal hyerplasia, and hyperkeratosis were observed (42). Several mouse strains have been found to contain a frameshift mutation in group IIA PLA 2 , resulting in loss of the enzyme activity. Infection of group IIA PLA 2 (Ϫ/Ϫ) mice (C57BL/6) with Helicobacter felis, a gastroduodenal pathogen, lead to increased apoptosis and proliferation compared with group IIA PLA 2 (ϩ/ϩ) BALB/c and C3H/HeJ mouse (43). Much more marked inflammatory and proliferative response to H. felis was also observed in another group IIA PLA 2 (Ϫ/Ϫ) mouse SV129 than in PLA 2 (ϩ/ϩ) mice (43). No satisfactory explanation has been proposed for this contradiction. One possible reason is that sPLA 2 contributes to some extent in the destruction of pathogen in PLA 2 (ϩ/ϩ) mice.
During the course of these studies, we noted that cells stably expressing group IIA PLA 2 were more resistant to apoptosis induced by serum starvation. This protective role of group IIA PLA 2 in serum withdrawal-induced apoptosis was further confirmed using a number of other approaches including PLA 2 inhibitors and site-directed mutagenesis of group IIA PLA 2 His-47. Because the lipid mediators AA and LPA could not mimic the effects of group IIA PLA 2 , it appears that other mechanisms are responsible for the protective physiological function of group IIA PLA 2 . Interestingly, although LPA had no effect in BHK and HQ cells, it did promote survival in NP cells. As LPA can transduce survival signals via phosphatidylinositol 3-kinase (43,44), it is possible that the group IIA PLA 2 and LPA might synergistically transduce survival signals in fibroblasts.
How group IIA PLA 2 generates survival signals remains to be elucidated. Group I PLA 2 can stimulate cell proliferation by binding to a receptors (45). Whether group IIA PLA 2 has its own receptors or can bind to group I PLA 2 receptors of BHK cells is unknown. Given the similarity between group I and IIA PLA 2 , it is possible that group IIA PLA 2 may activate a survival pathway by a similar mechanism. If so, this might also explain why PLA 2 possesses some functions independent of its hydrolytic activity.
Secretory PLA 2 has been identified as enhancing factor that enhances the binding of epidermal growth factor to cells (47). It is possible that rat group IIA PLA 2 may trigger anti-apoptotic signaling via epidermal growth factor. Alternatively, group IIA PLA 2 may affect the receptor signaling via modifying membrane topology.
The anti-apoptotic activity of group IIA PLA 2 also may explain the pathogenesis of certain clinical diseases. It has been reported that the use of certain Chinese herbs containing aristolochic acid for weight loss may cause severe renal failure (48). As aristolochic acid inhibits group IIA PLA 2 activity, it seems likely that aristolochic acid inhibition of PLA 2 -mediated signaling may contribute to the renal injury seen in these patients. The immunosuppressive drug cyclosporin A has the well known side effect of nephrotoxicity (49). Cyclosporin A has been shown to down-regulate mRNA level of group IIA PLA 2 in mesangial cells (50). If group IIA PLA 2 generates anti-apoptotic signals in kidney, long-term use of cyclosporin A may impair group IIA PLA 2 -medited survival signal transduction.
Group IIA PLA 2 is transcriptionally up-regulated by inflammatory cytokines, partially via NFB (50). NFB is believed to protect against tumor necrosis factor-induced cell death by inducing the expression of anti-apoptotic genes (51). High levels of group IIA PLA 2 is present at inflammatory sites, and the physiological role of the protein in inflammation remains largely unknown. Our results suggest the possibility that group IIA PLA 2 might counteract the effects caused by some proinflammatory cytokines, such as tumor necrosis factor that promotes cell death.