Involvement of group VI Ca2+-independent phospholipase A2 in protein kinase C-dependent arachidonic acid liberation in zymosan-stimulated macrophage-like P388D1 cells.

We investigated the possible involvement of group VI Ca2+-independent phospholipase A2 (iPLA2) in arachidonic acid (AA) liberation in zymosan-stimulated macrophage-like P388D1 cells. Zymosan-induced AA liberation was markedly inhibited by methyl arachidonoyl fluorophosphonate, a dual inhibitor of group IV cytosolic phospholipase A2 (cPLA2) and iPLA2. We found that a relatively specific iPLA2 inhibitor, bromoenol lactone, significantly decreased the zymosan-induced AA liberation in parallel with the decrease in iPLA2 activity, without an effect on diacylglycerol formation. Consistent with this, attenuation of iPLA2 activity by a group VI iPLA2 antisense oligonucleotide resulted in a decrease in zymosan-induced prostaglandin D2 generation. These findings suggest that zymosan-induced AA liberation may be, at least in part, mediated by iPLA2. A protein kinase C (PKC) inhibitor diminished zymosan-induced AA liberation, while a PKC activator, phorbol 12-myristate 13-acetate (PMA), enhanced the liberation. Bromoenol lactone suppressed the PMA-enhanced AA liberation without any effect on PMA-induced PKC activation. Down-regulation of PKCalpha on prolonged exposure to PMA also decreased zymosan-induced AA liberation. Under these conditions, the remaining AA liberation was insensitive to bromoenol lactone. Furthermore, the PKC depletion suppressed increases in iPLA2 proteins and the activity in the membrane fraction of zymosan-stimulated cells. In contrast, the zymosan-induced increases in iPLA2 proteins and the activity in the fraction were facilitated by simultaneous addition of PMA. Although intracellular Ca2+ depletion prevented zymosan-induced AA liberation, the translocation of PKCalpha to membranes was also inhibited. Taken together, we propose that zymosan may stimulate iPLA2-mediated AA liberation, probably through a PKC-dependent mechanism.

The liberation of arachidonic acid (AA) 1 upon stimulation is an important event leading to the generation of biologically active lipid mediators, such as prostaglandins and leukotrienes, and is mainly dependent on the hydrolysis of membrane glycerophospholipids catalyzed by phospholipase A 2 (PLA 2 ) (1, 2). Numerous types of mammalian PLA 2 s have been identified and classified into several groups (3). The mammalian PLA 2 s include at least two types of intracellular PLA 2 s, i.e. Ca 2ϩ -dependent and -independent enzymes. It is widely accepted that the intracellular PLA 2 responsible for stimulus-induced AA liberation is group IV Ca 2ϩ -dependent cytosolic PLA 2 (cPLA 2 ), which preferentially hydrolyzes glycerophospholipids with an arachidonoyl residue at the sn-2 position (4,5). The activation of cPLA 2 upon stimulation is mediated by Ca 2ϩ -dependent translocation to membranes (6,7), and by mitogen-activated protein kinase-catalyzed phosphorylation (8,9). On the other hand, Ca 2ϩ -independent PLA 2 (iPLA 2 ) has been detected in a variety of cells and tissues (reviewed in Ref. 10). Among several types of iPLA 2 s, which have been purified, sequenced, and well characterized (11)(12)(13)(14)(15), group VI iPLA 2 in mouse macrophagelike P388D 1 cells has been proposed to participate in phospholipid remodeling rather than stimulus-induced AA liberation (16,17).
A recent study involving P388D 1 cells has demonstrated that platelet-activating factor (PAF)-induced AA liberation is suppressed by an inhibitor of group IIA secretory PLA 2 or cPLA 2 , but not by a relatively specific iPLA 2 inhibitor, bromoenol lactone (BEL) (18). Furthermore, blockage of group V secretory PLA 2 by antisense oligonucleotides partially inhibits PAF-induced prostaglandin E 2 generation (19), while a group VI iPLA 2 antisense oligonucleotide has no effect (20). These findings clearly indicate that PAF-induced AA liberation may be mediated by group V secretory PLA 2 and cPLA 2 , but not by iPLA 2 . Moreover, a recent report suggested that activation of cPLA 2 is required for the onset of secretory PLA 2 -catalyzed hydrolysis of membrane phospholipids (21). Therefore, in PAFstimulated P388D 1 cells, cPLA 2 activation may be a predominant step in the induction of AA liberation. However, while in mouse peritoneal macrophages, zymosan stimulates cPLA 2 activation in parallel with AA liberation (22,23), zymosan-induced AA liberation in mouse macrophage-like RAW 264.7 cells has been shown to be sensitive to BEL (24). Therefore, it is possible that iPLA 2 , in addition to cPLA 2 , might be involved in AA liberation induced by zymosan but not by PAF. Similar inhibitory effects of BEL on AA liberation have been reported in several types of cells (25)(26)(27)(28)(29). However, BEL is reported to inhibit phosphatidic acid phosphatase, leading to the suppression of stimulus-induced diacylglycerol formation (30,31). This inhibitory effect may explain the inhibition of AA liberation by BEL, because diacylglycerol may contribute to AA liberation through direct and/or indirect modulation of cPLA 2 activity (31) or through the hydrolytic action of lipases toward diacylglycerol (32). It has been shown that zymosan-induced AA liberation is decreased by a protein kinase C (PKC) inhibitor (22,33,34) or intracellular Ca 2ϩ depletion (23,34,35) in mouse peritoneal macrophages. These findings may support the con-cept that cPLA 2 contributes to the AA liberation, although it has been suggested that cPLA 2 activation in P388D 1 cells is not mediated by PKC (36). Thus, the participation of iPLA 2 in stimulus-induced AA liberation remains to be elucidated.
In the present study, to clarify the role of iPLA 2 upon stimulation, we explored the possible involvement of group VI iPLA 2 in AA liberation in zymosan-stimulated P388D 1 cells by evaluating the effects of BEL and a group VI iPLA 2 antisense oligonucleotide, which have been shown to attenuate iPLA 2 activity in P388D 1 cells (16,20). We further examined the role of PKC in iPLA 2 -mediated AA liberation.
Measurement of iPLA 2 Activity-P388D 1 cells in 35-mm dishes were treated with BEL and then washed three times. The cells were lysed by adding a mixture of 800 M Triton X-100, 340 mM sucrose, 2 mM EGTA, 100 M p-(amidinophenyl)methanesulfonyl fluoride, 100 M leupeptin, and 10 mM Hepes (pH 7.4). The resulting lysate was treated with 5 mM dithiothreitol and used as the enzyme source. The substrate consisting of phosphatidylcholine was prepared in the presence of Triton X-100 according to the method of Ackermann et al. (11). The activity of iPLA 2 in the lysate (30 g of protein) was determined by incubation with a mixture of 1,2-dipalmitoyl-sn-glycero-3-[choline-methyl-14 C]phosphocholine and the unlabeled compound (2.5 mCi/mmol, 100 M) at 37°C for 30 min in the presence of 5 mM EDTA, 800 M Triton X-100, and 50 mM Hepes (pH 7.5) in a final volume of 200 l. The reaction was terminated by adding chloroform/methanol/HCl (200:200:1, v/v/v). After lipid extraction, the [ 14 C]lysophosphatidylcholine generated was determined by thin layer chromatography on a Silica Gel G plate using chloroform/methanol/H 2 O (65:35:6, v/v/v) as the development system. The radioactivity was determined and the enzyme activity was calculated. Under these conditions, the addition of CaCl 2 to this assay system did not enhance the increase in lysophosphatidylcholine, indicating that cPLA 2 in the enzyme source is inactive, probably due to the presence of Triton X-100 (37), and therefore the increase in lysophosphatidylcholine detected in this system reflects iPLA 2 activity.
Measurement of cPLA 2 Activity-P388D 1 cells in 100-mm dishes were treated with BEL and then scraped off. The cells were washed and sonicated in a buffer consisting of 100 mM NaCl, 2 mM EGTA, 100 M p-(amidinophenyl)methanesulfonyl fluoride, 100 M leupeptin, 20 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , and 10 mM Tris-HCl (pH 7.4). After the lysate had been centrifuged at 100,000 ϫ g for 30 min at 4°C, the resulting supernatant was treated with 5 mM dithiothreitol. The activity of cPLA 2 in the supernatant (20 g of protein) was determined by incubation with a mixture of 1-stearoyl-2-[ 3 H]arachidonoyl-sn-glycero-3-phosphocholine and the unlabeled compound (25 Ci/mol, 10 M) at 37°C for 10 min in the presence of 5 mM CaCl 2 and 100 mM Tris-HCl (pH 8.5) in a final volume of 200 l. After lipid extraction, the [ 3 H]AA liberated was determined by thin layer chromatography using petroleum ether/diethyl ether/acetic acid (40:40:1, v/v/v) as the development system. The radioactivity was determined, and the enzyme activity was calculated.
Measurement of Lipid Metabolism-P388D 1 cells were plated in 35-mm dishes at 1 ϫ 10 6 cells in 1 ml of RPMI 1640 containing 10% fetal bovine serum, and then incubated with [ 3 H]AA (0.5 Ci/ml) or [ 3 H]palmitic acid (1 Ci/ml) for 24 h. The labeled cells were washed and placed in 1 ml of RPMI 1640 containing 0.01% bovine serum albumin. The cells were treated with various reagents in the presence of 10 M BW755C (a cyclooxygenase and lipoxygenase inhibitor) (38,39), and then stimulated as described in the figure legends. When BEL was used, the labeled cells were washed three times after treatment with BEL and then placed in fresh RPMI 1640 containing 0.01% bovine serum albumin and 10 M BW755C. The reaction was terminated by transferring the medium and cell lysate, which was prepared by adding 0.1 M NaOH, to ice-cold chloroform/methanol/HCl (200:200:1, v/v/v). Lipids in the medium and lysate were extracted and separated by thin layer chromatography on a Silica Gel G plate with the following development systems: for the analysis of arachidonic acid and diacylglycerol, petroleum ether/diethyl ether/acetic acid (40:40:1, v/v/v); and for the analysis of phosphatidic acid, the combination of chloroform/methanol/7 M NH 4 OH (65:35:7.3, v/v/v) for the first dimension and chloroform/ methanol/acetic acid/H 2 O (60:30:8: The area corresponding to each lipid was scraped off, and the radioactivity was determined by liquid scintillation counting. Transfection of Oligonucleotides into P388D 1 Cells-The antisense or sense oligonucleotides were preincubated with Transome (Wako Pure Chemical Industries), a liposome, at room temperature for 15 min. The mixture (50 l) was added to P388D 1 cells that had been plated at 1 ϫ 10 6 cells in 35-mm dishes and then placed in 0.95 ml of RPMI 1640 containing 0.01% bovine serum albumin. The final concentrations of each oligonucleotide and Transome were 1 and 10 M, respectively. After incubation for 24 h, the cells were washed, lysed, and then subjected to the assay for iPLA 2 activity as described above. For the measurement of prostaglandin D 2 generation, the oligonucleotidetreated cells were placed in 1 ml of RPMI 1640 containing 0.01% bovine serum albumin, and then stimulated with zymosan. After centrifugation of the extracellular medium, the supernatant was subjected to determination with a commercial assay kit. The remaining cells were washed and then subjected to the protein assay. In preliminary experiments, various concentrations of the antisense oligonucleotides (0.1-10 M) and Transome (1-50 M) were examined. When the combination of 1 M oligonucleotide and 10 M Transome was used, maximal inhibition of iPLA 2 activity was observed. These experimental conditions did not affect cell viability, as estimated by trypan blue dye exclusion.
Measurement of Prostaglandin D 2 Generation by Exogenous AA-P388D 1 cells were treated with the antisense and sense oligonucleotides as above, and then placed in 1 ml of RPMI 1640 containing 0.01% bovine serum albumin. The cells were further incubated with a mixture of [ 3 H]AA (0.5 Ci/ml) and the unlabeled compound (5 M) at 37°C for 15 min. Lipids in the medium and cells were extracted and separated by thin layer chromatography using an upper phase of ethyl acetate/ isooctane/acetic acid/H 2 O (9:5:2:10, v/v/v/v) as the development system. The area corresponding to prostaglandin D 2 was scraped off, and the radioactivity was determined by liquid scintillation counting.
Measurement of PKC Activity-P388D 1 cells in 100-mm dishes were treated with BEL, washed three times, and then stimulated with PMA. The cells were scraped off, washed, and then sonicated in a buffer consisting of 340 mM sucrose, 2 mM EGTA, 100 M leupeptin, 100 M p-(amidinophenyl)methanesulfonyl fluoride, and 10 mM Hepes (pH 7.4). After the lysate had been centrifuged at 500 ϫ g for 5 min at 4°C, the resulting supernatant was centrifuged at 100,000 ϫ g for 30 min to separate soluble and membrane fractions. The activity of PKC in the soluble (20 g protein) and membrane (20 g protein) fractions was determined using a commercial assay kit.
Detection of PKC␣-P388D 1 cells in 35-mm dishes were treated with PMA and then lysed as above. P388D 1 cells in 100-mm dishes were treated and stimulated as described in the figure legends, and soluble and membrane fractions were prepared from the cells as above. Samples were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 10% gel. The proteins were transferred to a nitrocel-lulose membrane, and then anti-PKC␣ antibodies were applied. The bound antibodies were visualized using peroxidase-conjugated goat anti-mouse IgG antibodies and enhanced chemiluminescence reagents.
Detection of iPLA 2 and the Activity in the Membrane Fraction-P388D 1 cells in 100-mm dishes were treated and stimulated as described in the figure legend. The cells were sonicated in a buffer consisting of 340 mM sucrose, 2 mM EGTA, 100 M leupeptin, 100 M p-(amidinophenyl)methanesulfonyl fluoride, and 10 mM Hepes (pH 7.4), and then centrifuged at 100,000 ϫ g for 30 min at 4°C to separate supernatant (soluble fraction) and pellets. The pellets were suspended in the same buffer, and solubilized with 0.25% Triton X-100. After the mixture had been centrifuged at 10,000 ϫ g for 5 min at 4°C, the resulting supernatant was used as a membrane fraction. The iPLA 2 activity in the soluble and membrane fractions was determined by incubation with 25 M substrates (22.5 mCi/mmol) as described above. For the detection of iPLA 2 proteins, the membrane fraction was subjected to Western blotting using an antibody against group VI iPLA 2 .

Effect of an iPLA 2 Inhibitor on Zymosan-stimulated Lipid
Metabolism-As shown in Fig. 1, stimulation of [ 3 H]AA-labeled P388D 1 cells with 1 mg/ml zymosan elicited time-dependent AA liberation, of which the time course was similar to that observed in mouse peritoneal macrophages (23). The zymosaninduced AA liberation was almost completely inhibited when the cells were treated with 30 M methyl arachidonyl fluorophosphonate, an inhibitor of cPLA 2 and iPLA 2 (40), suggesting the participation of cPLA 2 and/or iPLA 2 in the liberation. We further examined the effect of BEL, a relatively selective iPLA 2 inhibitor (16,41,42), on iPLA 2 activity and the AA liberation. The iPLA 2 activity in the lysate of BEL-treated P388D 1 cells was inhibited in a BEL dose-dependent manner ( Fig. 2A), while cPLA 2 activity was not affected by 10 M BEL (297 pmol/ min/mg protein for controls and 285 pmol/min/mg protein for BEL treatment, means of two experiments). The results in Fig.  2B show that BEL suppressed zymosan-induced AA liberation by about 40 -45% at 2 or 5 M. Consistent with this, BEL also attenuated zymosan-induced prostaglandin D 2 generation (data not shown). It has been reported that BEL increases the basal level of free AA by itself in P388D 1 cells that have been labeled with [ 3 H]AA (18). However, such an effect was not observed when the P388D 1 cells were washed after the treatment with BEL (Fig. 2B). On the other hand, BEL is known to inhibit phosphatidic acid phosphatase in P388D 1 cells (30), this inhibitory effect being suggested to lead to the inhibition of AA liberation through the suppression of diacylglycerol formation in human amnionic WISH cells (31). However, as shown in Fig.  2B, 2 M BEL did not affect zymosan-induced diacylglycerol formation, while 5 M BEL slightly inhibited the formation, but the inhibition was only about 20%. In addition, BEL at 2 or 5 M did not significantly increase the zymosan-induced phosphatidic acid formation in [ 3 H]AA-or [ 3 H]palmitic acid-labeled P388D 1 cells (data not shown). Thus, under these conditions, BEL (up to 5 M) decreased zymosan-induced AA liberation without any effect on diacylglycerol or phosphatidic acid formation. Previously, we showed that antigen-induced AA liberation is mediated by diacylglycerol lipase in rat peritoneal mast cells (32). However, such a mechanism underlying the liberation is ruled out by the present work, since RHC80267 (up to 100 M), a diacylglycerol lipase inhibitor, did not affect zymosan-induced AA liberation in P388D 1 cells (data not shown), in agreement with a previous result obtained with mouse peritoneal macrophages (33).
Effect of an iPLA 2 Antisense Oligonucleotide-We also examined the effect of a group VI iPLA 2 antisense oligonucleotide, which has been shown to attenuate iPLA 2 activity in P388D 1 cells (20), and obtained the results shown in Fig. 3. Under our experimental conditions, an approximately 50% decrease in iPLA 2 activity was observed in P388D 1 cells exposed to 1 M antisense oligonucleotide compared with in control or sensetreated cells, as shown in Fig. 3A. Furthermore, Fig. 3B reveals that 1 M antisense oligonucleotide was found to decrease 1 mg/ml zymosan-induced prostaglandin D 2 generation by about 45%, whereas 1 M sense oligonucleotide had no effect on the generation. However, the antisense oligonucleotide did not suppress prostaglandin D 2 generation induced by ionomycin (0.5 M, 10 min of stimulation), a Ca 2ϩ ionophore (data not shown). We further examined the effect of the antisense oligonucleotide on the conversion of exogenous AA to prostaglandin D 2 , but the generation of prostaglandin D 2 was not affected by the antisense oligonucleotide (37.4 pmol/dish for controls and 34.1 pmol/dish for antisense treatment, means of two experiments). These results suggest that zymosan-induced AA liberation in P388D 1 cells may be, at least in part, mediated by group VI iPLA 2 .
Involvement of PKC in iPLA 2 -mediated AA Liberation-Several studies involving mouse peritoneal macrophages have demonstrated that zymosan-induced AA liberation is sensitive to PKC inhibitors such as GF109203X (22,33,34). We also confirmed that GF109203X attenuated 1 mg/ml zymosan-induced AA liberation dose-dependently in P388D 1 cells (Fig.  4A). Furthermore, as shown in Fig. 4B, when the cells were stimulated with 1 mg/ml zymosan and 5-20 nM PMA, the synergistic enhancement of zymosan-induced AA liberation by PMA was observed. Stimulation with 20 nM PMA caused about 2-fold enhancement of the AA liberation induced by zymosan alone, although PMA did not cause AA liberation by itself. However, 20 nM 4␣-phorbol 12,13-didecanoate, an inactive phorbol ester, had no effect (data not shown). In addition, the PMA-enhanced AA liberation was completely inhibited by pretreatment with 5 M GF109203X (Fig. 4B). Under these conditions, 20 nM PMA did not exhibit a significant effect on zymosan-induced diacylglycerol or phosphatidic acid formation (data not shown). These findings indicate that PKC may be involved in zymosan-induced AA liberation in P388D 1 cells.
In order to evaluate the involvement of PKC in iPLA 2 -mediated AA liberation, we examined the effect of BEL on the PMA-enhanced AA liberation. As shown in Fig. 5, BEL at 2 or 5 M markedly suppressed the enhancement by 20 nM PMA of 1 mg/ml zymosan-induced AA liberation. At 5 M BEL, the PMA-enhanced AA liberation was reduced to the level of BELinsensitive AA liberation in response to zymosan alone, suggesting that iPLA 2 may contribute to the PKC-dependent AA liberation in response to zymosan. Furthermore, we examined whether or not BEL affects PMA-induced PKC activation, the results being shown in Fig. 6. Stimulation of P388D 1 cells with 20 nM PMA resulted in an increase in PKC activity in the membrane fraction, with a concomitant decrease in the activity in the cytosol fraction. However, pretreatment of the cells with 20 M BEL did not affect the exchange of PKC activity between the fractions. In addition, even on treatment of the membrane fraction of PMA-treated cells with 20 M BEL, the PKC activity in the fraction was not affected (data not shown).
In mouse peritoneal macrophages, zymosan-induced AA liberation has been suggested to be mediated by PKC␣ (33). To study the involvement of PKC␣ in AA liberation in P388D 1 cells, we examined whether or not down-regulation of PKC␣ affects zymosan-induced AA liberation (Fig. 7). The exposure of the cells to PMA (100 or 200 nM) for 10 h, which led to a decrease in the amount of PKC␣ (Fig. 7A), caused the suppres-sion of 1 mg/ml zymosan-induced AA liberation by about 50% (Fig. 7B). Under these conditions, 5 M BEL did not further decrease the remaining AA liberation in PMA-pretreated cells, although it attenuated zymosan-induced liberation in control cells. Furthermore, as shown in Fig. 7C, we confirmed that 1 mg/ml zymosan caused translocation of PKC␣ to the membrane fraction from the cytosol, and that 5 M BEL did not exert any effect on the translocation. However, translocation of PKC␤ to the membranes was not observed in zymosan-stimulated P388D 1 cells (data not shown), as already reported elsewhere (36). These results suggest that iPLA 2 -mediated AA liberation may occur in parallel with PKC␣ activation.
To examine the mechanism underlying PKC-dependent regulation of iPLA 2 -mediated AA liberation, we determined iPLA 2 activity in PKC-depleted P388D 1 cells upon stimulation with zymosan. As shown in Fig. 8A, stimulation with 1 mg/ml zymosan resulted in an increase in iPLA 2 activity in the membrane fraction with a concomitant decrease in the activity in the cytosol fraction. Pretreatment with 100 nM PMA for 10 h significantly suppressed the zymosan-induced increase in iPLA 2 activity (Fig. 8B). Under these conditions, zymosan in-creased iPLA 2 proteins in the membrane fraction, and this increment was inhibited by the PKC depletion (Fig. 8C). We further examined effect of acute treatment with PMA on the responses induced by zymosan, the results being shown in Fig.  9. Simultaneous stimulation with 100 nM PMA facilitated the zymosan (1 mg/ml)-induced increases in iPLA 2 proteins and the activity in the membrane fraction, although PMA alone did not cause such increments. These results suggest that zymosan may induce iPLA 2 translocation to the membrane fraction in a PKC-dependent manner.
Ca 2ϩ -dependent PKC␣ Translocation-It has been demonstrated that zymosan-induced AA liberation is dependent on Ca 2ϩ in mouse peritoneal macrophages (23,34,35). We also observed that zymosan-induced AA liberation was almost completely suppressed when intracellular Ca 2ϩ was depleted by treatment of P388D 1 cells with 100 M BAPTA-AM and 1 mM EGTA, intracellular and extracellular Ca 2ϩ chelators, respectively (Fig. 10A). This finding appears to support the idea that the AA liberation is mediated by Ca 2ϩ -dependent PLA 2 , i.e. cPLA 2 , but does not rule out the possibility that Ca 2ϩ is required for PKC activation, presumably followed by iPLA 2 -mediated AA liberation. To examine this possibility, we determined the effect of intracellular Ca 2ϩ depletion on zymosaninduced PKC␣ translocation. As shown in Fig. 10B, treatment with 100 M BAPTA-AM and 1 mM EGTA inhibited the 1 mg/ml zymosan-induced increase in PKC␣ proteins in the membrane fraction. The results in Fig. 10 suggest that the inhibition of AA liberation by intracellular Ca 2ϩ depletion may be ascribed to, at least in part, suppression of PKC activation. DISCUSSION In the present study, we explored the possible involvement of iPLA 2 in stimulus-induced AA liberation using zymosan-stimulated mouse macrophage-like P388D 1 cells, which possess group VI iPLA 2 , one of the purified and sequenced iPLA 2 s (11, 13). The group VI iPLA 2 in P388D 1 cells has been shown to be inhibited by BEL, a relatively selective iPLA 2 inhibitor (42). Based on the inhibitory effect of BEL on the incorporation of arachidonic acid into phospholipids (16), it has been recognized that group VI iPLA 2 participates in phospholipid remodeling. Furthermore, in PAF-stimulated P388D 1 cells, AA liberation is partially suppressed by a secretory PLA 2 inhibitor or methyl arachidonyl fluorophosphonate, a dual inhibitor of cPLA 2 and iPLA 2 , but not by BEL (18), suggesting that PAF-induced AA liberation is mediated by cPLA 2 and secretory PLA 2 , but not by group VI iPLA 2 . However, as shown in this study, we found that BEL significantly decreased the zymosan-induced AA liberation under the conditions where BEL actually suppressed iPLA 2 activity but not cPLA 2 activity in P388D 1 cells (Fig. 2), suggesting the involvement of group VI iPLA 2 in the response to zymosan.
A recent report demonstrated that BEL inhibits phosphatidic acid phosphatase activity, resulting in suppression of the conversion of phosphatidic acid to diacylglycerol in P388D 1 cells (30). Furthermore, in human amnionic WISH cells, the inhibition by BEL of stimulus-induced AA liberation has been suggested to be due to the impairment of diacylglycerol-mediated cPLA 2 regulation through the suppression of diacylglycerol formation (31). Although a number of studies have also demonstrated that stimulus-induced AA liberation is inhibited by BEL in a variety of cells including zymosan-stimulated macrophage-like RAW 264.7 cells (24 -29), it is unclear whether or not BEL affects diacylglycerol formation in these cells. However, the present study showed that 2-5 M BEL suppressed zymosan-induced AA liberation by about 45%, without a significant change in diacylglycerol formation (Fig. 2) or phosphatidic acid formation (data not shown). Therefore, we suggest that the attenuation by BEL of zymosaninduced AA liberation may be due to inhibition of group VI iPLA 2 rather than phosphatidic acid phosphatase.
We further found that a group VI iPLA 2 antisense oligonucleotide decreased iPLA 2 activity and zymosan-induced prostaglandin D 2 generation, while a sense oligonucleotide had no effect (Fig. 3). Furthermore, the antisense oligonucleotide did not affect Ca 2ϩ ionophore-induced prostaglandin D 2 generation (data not shown) or the conversion of exogenous AA to prostaglandin D 2 . These results appear to indicate that attenuation by the anti-sense oligonucleotide of zymosan-induced prostaglandin D 2 generation is due to the inhibition of iPLA 2 activity but not cPLA 2 or cyclooxygenase activity. In contrast to zymosan, a recent report has shown that the group VI iPLA 2 antisense oligonucleotide does not affect PAF-induced AA liberation despite the attenuation of iPLA 2 activity in P388D 1 cells (20). Although it is unclear whether or not the different effects of the antisense oligonucleotide on zymosan-and PAF-induced AA liberation are due to differences in the signaling pathways responsible for zymosan and PAF, we propose that group VI iPLA 2 may be, at least in part, involved in zymosan-induced AA liberation in P388D 1 cells.
In mouse peritoneal macrophages, zymosan-induced AA liberation has been shown to occur in parallel with cPLA 2 activation (22,23). Furthermore, a PKC inhibitor has been shown to suppress the AA liberation with concomitant decreases in mitogenactivated protein kinase and cPLA 2 activities (22). These findings suggest that cPLA 2 -mediated AA liberation may be regulated by PKC-dependent activation of mitogen-activated protein kinase in mouse peritoneal macrophages. We showed here that GF109203X, a PKC inhibitor, decreased zymosan-induced AA liberation, while PMA, a PKC activator, potentiated the AA liberation in P388D 1 cells (Fig. 4), suggesting that PKC may be involved in zymosan-induced AA liberation. However, it is possible that the PKC-dependent AA liberation is not mediated by cPLA 2 , since it has been demonstrated that PKC does not participate in cPLA 2 regulation in P388D 1 cells (36). In this study, BEL partially inhibited zymosan-induced AA liberation (Fig. 2), while methyl arachidonyl fluorophosphonate did so almost com- pletely in P388D 1 cells (Fig. 1), thus indicating that cPLA 2 may be involved in the BEL-insensitive AA liberation.
The present study demonstrated that BEL suppressed PMAenhanced AA liberation in response to zymosan without any effect on PMA-induced PKC activation (Figs. 5 and 6), suggesting the possible involvement of iPLA 2 in PKC-dependent AA liberation. We also showed that the depletion of PKC␣ on prolonged exposure to PMA reduced zymosan-induced AA liberation, and further that the treatment of PKC-depleted cells with BEL did not inhibit the remaining, PKC-independent AA liberation (Fig.  7). Thus, BEL seems to affect only PKC-dependent AA liberation in response to zymosan. In addition, we confirmed that BEL had no effect on zymosan-induced PKC␣ translocation (Fig. 7). Therefore, it is conceivable that iPLA 2 -mediated AA liberation may occur downstream of PKC activation in P388D 1 cells. We further demonstrated that zymosan increased iPLA 2 activity in the membrane fraction with a decrease in the activity in the cytosol fraction (Fig. 8). Of interest was the finding that the depletion of PKC␣ inhibited zymosan-induced increases in iPLA 2 activity and the enzyme proteins in the membrane fraction (Fig. 8). Moreover, the increases in iPLA 2 proteins and the activity in response to zymosan were potentiated by simultaneous stimulation with PMA ( Fig. 9), this being consistent with the result that PMA enhanced BEL-sensitive AA liberation in zymosan-stimulated cells (Fig. 5). These findings suggest that PKC may be involved in the zymosan-induced increase in iPLA 2 in the membrane fraction, due to translocation of the enzyme to the membranes. We propose this idea as one of mechanisms underlying PKC-dependent iPLA 2 regulation. It has been reported that iPLA 2 activity in membranes increases upon stimulation in human monocytes (43) and rat ventricular myocytes (44). In the monocytes, the membrane-associated iPLA 2 activity has been suggested to be modulated by phosphorylation of the enzyme. However, although at present we have no evidence suggesting that iPLA 2 may undergo phosphorylation upon stimulation with zymosan or PMA, group VI iPLA 2 of P388D 1 cells were suggested to have no known consensus sequence for phosphorylation sites (13,17). Based on these findings, it is possible that occurrence of iPLA 2 -mediated AA liberation in response to zymosan may be, at least in part, regulated by PKC-dependent iPLA 2 translocation to the membranes.
Stimulus-induced AA liberation in most cases occurs in a Ca 2ϩdependent manner, and therefore it is reasonable to consider that the liberation is catalyzed by Ca 2ϩ -dependent PLA 2 . In mouse peritoneal macrophages, zymosan-induced AA liberation has been shown to be suppressed by intracellular Ca 2ϩ depletion (23,34,35). We also showed that zymosan-induced AA liberation was not observed in the presence of intracellular and extracellular Ca 2ϩ chelators in P388D 1 cells (Fig. 10). These findings may indicate that iPLA 2 is not involved in the mechanism underlying the zymosan-induced AA liberation, because iPLA 2 does not require Ca 2ϩ for its catalytic action. However, we demonstrated that intracellular Ca 2ϩ depletion prevented zymosan-induced PKC␣ translocation (Fig. 10). Since the activation of PKC␣ is supposed to be upstream of iPLA 2 -mediated AA liberation (Fig.  7), it seems likely that the inhibition of zymosan-induced AA liberation by intracellular Ca 2ϩ depletion is probably due to, at least in part, the suppression of PKC activation.
In this study, although PMA enhanced zymosan-induced AA liberation and iPLA 2 translocation to the membranes, PMA by itself did not cause these responses (Figs. 4 and 9). This may suggest that PKC activation is essential for zymosan-induced AA liberation but insufficient for iPLA 2 regulation. It has been demonstrated that a recombinant 80-kDa iPLA 2 associates with calmodulin in a Ca 2ϩ -dependent manner (45). In P388D 1 cells, zymosan-induced AA liberation was sensitive to W-7, a calmodulin antagonist. 2 Therefore, we speculate that PKC might regulate iPLA 2 -mediated AA liberation in cooperation with unknown factors, although it is unclear at present whether or not calmodulin is a candidate for such a factor.
In conclusion, we propose that zymosan may stimulate iPLA 2 -mediated AA liberation, probably through PKC-dependent iPLA 2 translocation to the membranes. Thus, the present work appears to provide one of the aspects concerning the role and regulation of group VI iPLA 2 upon stimulation.