Selective inhibition of cytosolic phospholipase A2 in activated human monocytes. Regulation of superoxide anion production and low density lipoprotein oxidation.

Our previous studies have shown that monocyte activation and release of O2 are required for monocyte-mediated low density lipoprotein (LDL) lipid oxidation. We have also found that intracellular Ca2+ levels and protein kinase C activity are requisite participants in this potentially pathogenic process. In these studies, we further investigated the mechanisms involved in the oxidation of LDL lipids by activated human monocytes, particularly the potential contributions of the cytosolic phospholipase A2 (cPLA2) signaling pathway. The most well-studied cPLA2, has a molecular mass of 85 kDa and has been reported to be regulated by both Ca2+ and phosphorylation. We found that cPLA2 protein levels and cPLA2 enzymatic activity were induced upon activation of human monocytes by opsonized zymosan. Pharmacologic inhibition of cPLA2 activity by AACOCF3, which has been reported to be a specific inhibitor of cPLA2 as compared with sPLA2, caused a dose-dependent inhibition of cPLA2 enzymatic activity and LDL lipid oxidation by activated human monocytes, whereas sPLA2 activity was not affected. To corroborate these findings, we used specific antisense oligonucleotides to inhibit cPLA2. We observed that treatment with antisense oligonucleotides caused suppression of both cPLA2 protein expression and enzymatic activity as well as monocyte-mediated LDL lipid oxidation. Furthermore, antisense oligonucleotide treatment caused a substantial inhibition of O2 production by activated human monocytes. In parallel experimental groups, cPLA2 sense oligonucleotides did not affect cPLA2 protein expression, cPLA2 enzymatic activity, O2 production, or monoctye-mediated LDL lipid oxidation. These studies support the proposal that cPLA2 activity is required for activated monocytes to oxidize LDL lipids.

diated low density lipoprotein (LDL) lipid oxidation. We have also found that intracellular Ca 2؉ levels and protein kinase C activity are requisite participants in this potentially pathogenic process. In these studies, we further investigated the mechanisms involved in the oxidation of LDL lipids by activated human monocytes, particularly the potential contributions of the cytosolic phospholipase A 2 (cPLA 2 ) signaling pathway. The most well-studied cPLA 2 , has a molecular mass of 85 kDa and has been reported to be regulated by both Ca 2؉ and phosphorylation. We found that cPLA 2 protein levels and cPLA 2 enzymatic activity were induced upon activation of human monocytes by opsonized zymosan. Pharmacologic inhibition of cPLA 2 activity by AA-COCF 3 , which has been reported to be a specific inhibitor of cPLA 2 as compared with sPLA 2 , caused a dosedependent inhibition of cPLA 2 enzymatic activity and LDL lipid oxidation by activated human monocytes, whereas sPLA 2 activity was not affected. To corroborate these findings, we used specific antisense oligonucleotides to inhibit cPLA 2 . We observed that treatment with antisense oligonucleotides caused suppression of both cPLA 2 protein expression and enzymatic activity as well as monocyte-mediated LDL lipid oxidation. Furthermore, antisense oligonucleotide treatment caused a substantial inhibition of O 2 . production by activated human monocytes. In parallel experimental groups, cPLA 2 sense oligonucleotides did not affect cPLA 2 protein expression, cPLA 2 enzymatic activity, O 2 . production, or monoctye-mediated LDL lipid oxidation. These studies support the proposal that cPLA 2 activity is required for activated monocytes to oxidize LDL lipids.
Human native low density lipoprotein (LDL) 1 can be oxidized by activated human monocytes, neutrophils, and cells of the monocytoid cell line U937 (1,2) as well as endothelial cells and smooth muscle cells (3). Once oxidized, LDL is chemotactic for monocytes (4), serves as a cytotoxin for target cells (1,(5)(6)(7), and hinders the movement of macrophages (8). It is recognized by scavenger and oxidized LDL receptors on macrophages and is taken up by these receptors in an unregulated fashion (9 -12). Oxidized LDL has been detected in atherosclerotic lesions (13,14). Macrophages trapped in the artery wall may take up oxidized LDL, thus contributing to the formation of foam cells and fatty streak lesions. Cell-mediated oxidation of LDL has therefore been suggested to be a key event in atherogenesis as well as in inflammatory tissue injury (15). In our culture system, human monocyte oxidation of LDL is dependent on monocyte activation. Since activation of monocytes is a complex process, one that involves a series of secondary messengers that mediate signal transduction and alter cell function, we have begun to identify several key signaling pathways that are required for converting a blood monocyte to an activated monocyte that can mediate LDL lipid oxidation. We have found that superoxide anion (O 2 . ) production is required for this process and that intracellular Ca 2ϩ levels are integrally involved in oxidation of LDL lipids by activated human monocytes. Both the influx of extracellular Ca 2ϩ and the release of intracellular Ca 2ϩ are involved (16). Recently, we demonstrated that a Ca 2ϩ -regulated, intracellular signaling pathway, protein kinase C (PKC), was required. Our experimental results showed that depletion of PKC activity by phorbol 12myristate 13-acetate, inhibition of PKC activity by pharmacologic inhibitors, or suppression of PKC levels by antisense oligonucleotides caused an inhibition of LDL lipid oxidation by activated human monocytes (17). The isoenzyme of PKC, required for oxidation of LDL by activated monocytes, was shown to be a member of the cPKC group of isoenzymes. The rise of intracellular Ca 2ϩ levels and activation of PKC elicit a variety of cellular responses including phosphorylation of target proteins which are located throughout the cell, on the plasma membrane, in the cytosol, and in the nucleus. This can initiate a cascade of other second messengers to transmit intracellular signals that ultimately alter cell function (18). Ca 2ϩ -and PKC-dependent signaling, therefore, provide exceptionally versatile signaling mechanisms. Downstream effects of Ca 2ϩ and PKC have been reported to be related to the induction of several other intracellular signal transduction pathways, one of these pathways involves phospholipase A 2 (PLA 2 ) which hydrolyzes the sn-2 fatty acid on phospholipids producing free fatty acid and lysophospholipid (18,19). Both free fatty acid and lysophospholipid serve as lipid mediators to regulate cell functions. PLA 2 also plays a critical role in providing substrate for the biosynthesis of prostaglandins and leukotrienes by releasing arachidonic acid (AA) from membrane phospholipids.
Consequently, PLA 2 s have been implicated in many cellular processes and disease states, such as maintenance of cellular phospholipid pools, participation in inflammatory reactions and host defense, and involvement in myocardial ischemia. Furthermore, AA has been reported to induce O 2 . production in human neutrophils (20) and monocytes (21) by activation of NADPH oxidase or by its metabolism via lipoxygenase pathways (22). Our laboratory has previously shown that both O 2 .
production and lipoxygenase are involved in monocyte-mediated LDL lipid oxidation (5,16,17), suggesting that PLA 2 might participate in this process. Phospholipases A 2 are a diverse family of enzymes with a growing number of members. Among the mammalian enzymes, the most well-characterized are the 14-kDa secretory PLA 2 (sPLA 2 ) and the 85-kDa cytosolic PLA 2 (cPLA 2 ) (19). The sPLA 2 is Ca 2ϩ -dependent and requires mM levels of Ca 2ϩ for activity. It also has seven disulfide bonds that are required for activity and therefore is sensitive to treatment with reducing agents such as dithiothreitol (DTT). In contrast, the 85-kDa cPLA 2 requires only M levels of Ca 2ϩ , levels that can be reached intracellularly, and it does not have disulfide bonds so its activity is not susceptible to reducing agents. Although cPLA 2 is 85 kDa, it migrates as an 110-kDa protein in SDS gels. Unlike the sPLA 2 , cPLA 2 shows a preference for arachidonic acid in the sn-2 position of substrate phospholipid. The activity of this latter enzyme is induced by protein phosphorylation and Ca 2ϩ -dependent translocation to membranes from the cytosol. In addition to these two enzymes several Ca 2ϩindependent PLA 2 (iPLA 2 ) have been described, including the canine myocardial 40-kDa iPLA 2 (23), the murine macrophagelike cell line P388D 1 80-kDa iPLA 2 (24), bovine brain 100-kDa iPLA 2 (25), and an 80-kDa iPLA 2 from CHO cells (46). These iPLA 2 are Ca 2ϩ -independent and activated by ATP or detergent. To date, only the latter iPLA 2 has been cloned (46).
The 85-kDa cPLA 2 is believed to be an important regulator of arachidonic acid availability and thereby controls the production of potent lipid mediators. We were particularly interested in this enzyme because its activity has been shown to be regulated by PKC phosphorylation and by Ca 2ϩ levels, and both PKC and Ca 2ϩ have been shown to be key participants in monocyte oxidation of LDL. We therefore designed a series of experiments to test the hypothesis that cPLA 2 participates in regulating the activation-dependent oxidation of LDL lipids by human monocytes.
Zymosan, obtained from ICN Biochemicals (Cleveland, OH), was opsonized (26) and used at a concentration of 2 mg/ml to activate human monocytes and U937 cells. Opsonized zymosan (ZOP) was suspended in phosphate-buffered saline as a 20-fold stock solution and stored at Ϫ70°C prior to use.
Lipoprotein Preparation-Low density lipoprotein (LDL) was prepared according to previously described methods which minimize oxidation and exposure to endotoxin (2). All reagents used for LDL isolation were prepared with Chelex-treated MilliQ water. Each batch of LDL was assayed for endotoxin contamination by the limulus amebocyte lysate assay (kit QCL-1000, Whittaker Bioproducts Inc., Walkersville, MD). Final endotoxin contamination was always Ͻ0.03 unit/mg LDL cholesterol. LDL was stored in 0.5 mM EDTA. Immediately before use, LDL was dialyzed at 4°C against phosphate-buffered saline with-out calcium or magnesium (Life Technologies, Inc.) in the dark. 1 g/liter Chelex was added to the dialysis buffer. LDL was used at a final concentration of 0.5 mg of cholesterol/ml.
Isolation of Human Monocytes and Cell Culture-Human monocytes were isolated from heparinized whole blood by sequential centrifugation over a Ficoll-Paque density solution and adherence to serum-coated cell culture flasks (5). Nonadherent cells were then removed. The adherent cells were released with 5 mM EDTA and plated into multiwell tissue culture plates at 1.0 ϫ 10 6 cells/ml. This cell population consisted of greater than 95% monocytes (5). The isolated human monocytes were then cultured overnight in Dulbecco's modified Eagle's medium with 10% serum before use in experiments. In the experiments, monocytes were washed twice with RPMI 1640 (Whittaker, Walkersville, MD) and incubated with LDL (0.5 mg of cholesterol/ml) and ZOP (2 mg/ml) in the presence or absence of different reagents. After 24 h incubation, cell supernatants were collected, and LDL lipid oxidation was determined.
U937 cells, obtained from the American Type Culture Collection (Rockville, MD), were cultured in 150-cm 2 flasks (Corning, Corning, NY) in RPMI 1640 supplemented with 10% bovine calf serum (HyClone, Logan, UT), 100 units/ml penicillin, and 100 g/ml streptomycin (Life Technologies, Inc.) at 37°C in a humidified atmosphere of 90% air, 10% CO 2 . U937 cells were maintained in log phase (cell number was kept between 0.1 and 1.0 ϫ 10 6 cells/ml). For experiments, U937 cells (in log phase between 3 to 6 ϫ 10 5 cells/ml) were washed twice with RPMI 1640. 5 ϫ 10 5 cells/ml were plated into multiwell tissue culture plates (Costar, Cambridge, MA) and incubated with LDL (0.5 mg of cholesterol/ml) and ZOP (2 mg/ml) in the presence or absence of different reagents. After 24 h incubation, cell supernatants were collected, and lipid oxidation of LDL was determined.
Measurement of Lipid Oxidation-The oxidation of LDL lipids was measured by both the thiobarbituric acid (TBA) assay and the lipid peroxide (LPO) assay.
Thiobarbituric Acid Assay-The oxidation of LDL lipids was measured by the TBA assay, a modification of the assay described by Schuh et al. (27). The thiobarbituric acid assay detects malondialdehyde (MDA) and MDA-like compounds reacting with TBA. Compounds that react with TBA are referred to as TBA-reactive substances. Data are expressed in terms of MDA equivalents (nmol of MDA/ml).
Lipid Peroxide Assay-The lipid peroxide levels on LDL were determined by a modification of the assay described by El-Saadani et al. (28). This assay measures the oxidative capacity of lipid peroxides to convert iodide to iodine, which can be measured spectrophotometrically at 365 nm. Data are expressed in nanomoles of lipid peroxide/ml (nmol LPO/ml). cPLA 2 Activity Assay-Human monocytes (2.5 ϫ 10 6 cells/ml) in RPMI without serum were incubated with ZOP (2 mg/ml) in the presence or absence of a variety of reagents as indicated. After incubation, cells were harvested and resuspended in Buffer A (50 mM Hepes, pH 7.5, 250 mM sucrose, 1 mM EDTA, 1 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 0.1 mM sodium vanadate, 0.5 mM phosphoserine, and 0.5 mM phosphothreonine) at a concentration of 2.5 ϫ 10 7 cells/ml. Cells in Buffer A were lysed by sonication, and then cell lysates were centrifuged at 1000 ϫ g for 15 min. Supernatants were collected and 100 g of total protein, as determined by the Lowry assay, were used in the PLA 2 assay.
For PLA 2 assays, the reaction was initiated by adding cell lysate (100 g of protein) to the substrate (100 nCi, 1.8 nmol) followed by incubation at 37°C for 30 min. The total volume was 100 l. Then, lipids were extracted by the method of Bligh and Dyer (30). Free fatty acid and phospholipids were separated by TLC using the solvent system, chloroform/acetone/methanol/acetic acid/H 2 O, 6:8:2:2:1 (v/v). The free fatty acid and phosphatidylcholine were scraped from the TLC plate, and the radioactivity was counted on a ␤-counter. In some cases the DTT was not included. The DTT-sensitive activity (the activity without DTT minus that with DTT) is referred to as sPLA 2 activity, and the DTTresistant activity is referred to as cPLA 2 activity. This latter activity may include that mediated by Ca 2ϩ -requiring cPLA 2 and Ca 2ϩ -independent iPLA 2 , but for simplicity we refer to it as cPLA 2 activity. Western Blotting Analysis-Human monocytes (2.5 ϫ 10 6 cells/ml) were incubated with ZOP (2 mg/ml) in the presence or absence of different reagents for 24 h as indicated in figure legends. After incuba-tion, cells were harvested and resuspended in 200 l of hypotonic lysis buffer (50 mM Tris-HCl, pH 7.5, 5 mM MgSO 4 , 0.5 mM EGTA, 0.1% 2-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 20 g/ml leupeptin, and 0.5% Nonidet P-40). The cells were vortexed for 15 s, and cellular debris and nuclei were removed by centrifugation in a microcentrifuge at 1000 ϫ g for 10 min. The supernatants were collected, and 100 g of cell lysate protein was prepared for 7% SDS-PAGE (31). The SDS-PAGE gel was transferred to a polyvinylidene difluoride membrane by the semi-dry method (32). After blocking the nonspecific binding sites with 10% milk in Tris buffer (20 mM Tris-base, pH 7.4, 1.5 M NaCl, 1% Nonidet P-40) for 1 h at room temperature, cellular cPLA 2 protein was detected with a 1:1000 dilution of rabbit anti-human recombinant cPLA 2 polyclonal antibody (generously provided by Dr. J. Clark, Genetic Institute, Inc., Andover, MA), followed by incubation with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:5000 dilution). The polyvinylidene difluoride membrane was developed using enhanced chemiluminescence (Amersham Corp.).
For immunoprecipitation of cPLA 2 protein, the total cell lysate from 10 ϫ 10 6 cells was incubated with 30 l of polyclonal antibody prebound to protein A beads for 1 h at 4°C. After incubation, cell lysates were centrifuged at 1000 ϫ g for 10 min. Pellets were collected and prepared for 7% SDS-PAGE. After transfer to a polyvinylidene difluoride membrane, cPLA 2 protein was detected by Western blotting using antihuman recombinant cPLA 2 monoclonal antibody (generously provided by Dr. J. Clark, Genetics Institute, Inc., Andover, MA).
Treatment of Cells with Oligonucleotides-The sense and antisense sequences of cPLA 2 were selected from a unique area of the mRNA that is near the 5Ј end of the message. Prior to selection, the sequences were selected by screening for uniqueness using Blast© and were also tested for lack of secondary structure and oligo pairing using Mulfold© (33).
The antisense oligomer was complementary to nucleotides 219 -238 of cPLA 2 which code for amino acids 27-34 of the protein. The sequence was 5ЈCCC CCT TTG TCA CTT TGG TG3Ј. The sequence of the cPLA 2 sense oligomer was 5ЈCAC CAA AGT GAC AAA GGG GG3Ј. Phosphorothioate-modified oligonucleotides were used for these studies to limit degradation. The oligonucleotides were synthesized and purified by HPLC prior to use (Genosys Biotechnologic, Inc., Woodlands, TX).
For these experiments, human monocytes (1 ϫ 10 5 cells/0.1 ml/well), LDL (0.5 mg of cholesterol/ml), and ZOP (2 mg/ml) were cultured in RPMI 1640 without serum in the presence or absence of different concentrations of sense or antisense oligonucleotides in 96-well flatbottomed culture plates (Costar, Cambridge, MA) for 24 h. After incubation, cell-mediated LDL lipid oxidation was assessed by the TBA assay as described above. In some cultures, sense or antisense oligonucleotides were incubated with cells (2.5 ϫ 10 6 cells/ml) for 24 h, and then cellular cPLA 2 activity was measured by the method described above.
Detection of Toxicity of Test Agents-The toxicity of test agents to human monocytes and U937 cells was determined by the [ 14 C]adenine release assay (34). Briefly, 1 ϫ 10 7 cells in 20 ml of RPMI 1640 were labeled overnight by incubating them with 10 Ci of [ 14 C]adenine (ICN Radiochemical, Irvine, CA). For experiments, the cells were washed twice with RPMI 1640, and human monocytes (1 ϫ 10 6 cells/ml) or U937 cells (0.5 ϫ 10 6 cells/ml), ZOP (2 mg/ml), and LDL (0.5 mg of cholesterol/ ml) were added to 12-well tissue culture plates in the presence or absence of test agents. After a 24 h incubation, the amount of [ 14 C] adenine released by the cells was detected by counting 100 l of the supernatant fluid using a Beckman LS-3801 ␤-counter. [ 14 C]Adenine release from ZOP-activated cells in the absence of test agents was defined as 0% release. [ 14 C]Adenine release by 0.2% SDS was defined as 100% release and interpreted as maximum toxicity. The data are expressed in Equation 1 as Detection of Antioxidant Activity of Test Agents-The following method was used to assess the antioxidant effects of test agents (16,35). 5 M of CuSO 4 and 0.5 mg of cholesterol/ml of LDL were incubated together with the various test agents at 37°C for 24 h in RPMI 1640. After incubation, supernatants were collected, and LDL oxidation was measured by the TBA assay as described above. All experiments were performed in triplicate. If copper-induced LDL oxidation was inhibited in the presence of the test agents, this indicated that the test agent could serve as a general antioxidant.
Measurement of Superoxide Anion Production-Superoxide anion (O 2 . ) production was measured by the cytochrome c reduction assay (36).
Cytochrome c can be reduced by O 2 . on a mol/mol basis, and the reduced cytochrome c has an increased absorbance at 550 nm. Human monocytes (1 ϫ 10 6 cells/ml) and antisense oligonucleotides or sense oligonucleotide (5 M) were preincubated for 24 h in Dulbecco's modified Eagle's medium with 10% bovine calf serum. After preincubation, cells and 320 M cytochrome c (Sigma) were incubated in the presence or absence of 150 units/ml superoxide dismutase (from bovine erythrocytes, Sigma) in 96-well cell tissue culture plates (a total volume of 100 l/well) in RPMI 1640 without phenol red and serum (Whittaker, Walkersville, MD) at 37°C in a humidified incubator with 10% CO 2 for 1 h. ZOP (2 mg/ml) and test agents were included during the incubation. After incubation, the absorbance was measured at 550 nm. Equation 2 was used to determine the nmol of O 2 . produced (where SOD is super- Statistical Analyses-The data from experiments were analyzed using the unpaired two-tailed Student's t test. Statistical tests were performed with GraphPAD InStat software (GraphPAD Software Inc., San Diego, CA). Data points with a p Ͻ 0.05 were considered to be significantly different.

RESULTS
We first investigated the induction of cellular cPLA 2 by Western blotting analysis using different anti-human cPLA 2specific antibodies. In these experiments, human monocytes were incubated with the activator (ZOP) for 24 h. After incubation, cell lysates were prepared. In one set of experiments, cPLA 2 protein was immunoprecipitated from cell lysates using anti-human cPLA 2 -specific polyclonal antibody. After immunoprecipitation, cPLA 2 protein was detected by Western blotting using an anti-human cPLA 2 -specific monoclonal antibody as described under "Materials and Methods." The result is shown in Fig. 1A. In a similar experiment, cPLA 2 cell lysates, without immunoprecipitation, were directly detected by Western blot- ting using the same polyclonal antibody. This result is shown in Fig. 1B. Our data indicate that human monocytes had very low levels of endogenous cPLA 2 (as shown in lane 1 of Fig. 1, A and  B). Upon activation, monocyte cPLA 2 was induced as demonstrated in lane 2 of Fig. 1, A and B. The lysate from unstimulated U937 cells, constitutive producers of cPLA 2 , was included as a positive control as shown in lane 3 of Fig. 1A. These data demonstrate the induction of cPLA 2 protein in activated human monocytes, and similar results were obtained regardless of whether cell lysates were first subjected to immunoprecipitation. In this study, therefore, cPLA 2 proteins in cell lysates were analyzed using anti-human cPLA 2 -specific polyclonal antibody without immunoprecipitation.
The time course for induction of cPLA 2 protein levels in activated human monocytes was also examined. In this experiment, human monocytes and ZOP were incubated for different periods. cPLA 2 protein was detected using Western blotting as described under "Materials and Methods." Although total protein levels were not substantially changed in both unactivated (0.2756 -0.3138 mg/10 ϫ 10 6 cells) and activated cells (0.2242-0.3180 mg/10 ϫ 10 6 cells), the cPLA 2 protein levels were substantially induced as shown in Fig. 2A. The cPLA 2 protein levels began to increase after 4 h of activation and reached maximal levels at 12 h of activation and then gradually decreased at 24 h.
In parallel experiments, the induction of cPLA 2 enzymatic activity at various times following human monocyte activation was also investigated. In these experiments, human monocytes were incubated with ZOP for various times between 0 and 24 h as indicated. After incubation, cell lysates were assayed for cPLA 2 activity. During the cPLA 2 activity assay, DTT was included to effect selective inactivation of sPLA 2 -like activity.
After incubation, 14 C-labeled free arachidonic acid and phosphatidylcholine were separated by TLC using the solvent system as described under "Materials and Methods." The data shown in Fig. 2B indicate that there is a basal level of endogenous DTT-resistant, cPLA 2 activity in unactivated human monocytes. The cPLA 2 enzymatic activity began to increase after 4 h of activation. The maximal induction of cPLA 2 enzymatic activity was 12 h after initiation of activation. Although the cPLA 2 enzymatic activity gradually decreased after 12 h of activation, significant induction of cPLA 2 enzymatic activity was still observed at 24 h of activation. Significant induction of cPLA 2 enzymatic activity is indicated by asterisks (* indicates p Ͻ 0.05). These results indicate that human monocyte cPLA 2 enzymatic activity is induced upon activation, and the induction of cPLA 2 enzymatic activity is correlated with the rise in cPLA 2 protein levels. It should be noted that this assay may detect both the activities of the cPLA 2 and the Ca 2ϩ -independent cytosolic PLA 2 (iPLA 2 ).
The involvement of phospholipases A 2 in the process of human monocyte and U937 cell oxidation of LDL was first evaluated using several, structurally unrelated, pharmacologic inhibitors of PLA 2 , including DEDA, ONO-RS-082, aristolochic acid, and 4-BpB. Freshly isolated human monocytes or U937 cells, LDL, and ZOP were incubated together in the presence or absence of the PLA 2 inhibitors for 24 h. After incubation, the lipid oxidation of LDL was assessed by the TBA assay and the LPO assay. The TBA assay is a widely used method to detect malondialdehyde and MDA-like compounds derived from lipid oxidation products (27). The LPO assay detects lipid hydroperoxide which are produced upon lipid oxidation (28). Our experimental results demonstrated that each of these PLA 2 inhibitors showed dose-dependent inhibition of LDL lipid oxidation by activated human monocytes and U937 cells (data not shown). These data, regardless of the fact that most of these pharmacologic PLA 2 inhibitors are nonselective for sPLA 2 versus cPLA 2 , provided the first suggestion that PLA 2 activity was involved in LDL lipid oxidation by activated human monocytes and U937 cells.
To further investigate the requirement for cPLA 2 , we used another inhibitor, AACOCF 3 , which has been reported to be a selective inhibitor of cPLA 2 (37). In these experiments, human monocytes and ZOP were incubated in the presence or absence of different concentrations of AACOCF 3 for 24 h. After incubation, cell lysates were prepared. Then, PLA 2 activities were assessed in the presence or absence of 2 mM DTT as described under "Materials and Methods." The experimental results are summarized in Fig. 3. AACOCF 3 inhibited DTT-resistant PLA 2 activity in a concentration-dependent fashion, indicating that cPLA 2 activity was inhibited by AACOCF 3 as shown in Fig. 3A. In contrast, AACOCF 3 did not inhibit DTT-sensitive PLA 2 activity, indicating that sPLA 2 activity was not inhibited by AA-COCF 3 as shown in Fig. 3B.
In a parallel experiment, we also monitored monocyte-mediated LDL lipid oxidation in the presence or absence of AA-COCF 3 . In these experiments, human monocytes were incubated with LDL and ZOP in the presence or absence of AACOCF 3 for 24 h. After incubation, cell-mediated LDL lipid oxidation was assessed by both the TBA assay and the LPO assay. The experimental results are summarized in Fig. 4. Upon activation, cell-mediated LDL lipid oxidation was substantially increased as detected by the TBA assay (as shown in Fig. 4A) and the LPO assay (as shown in Fig. 4B). AACOCF 3 caused a concentration-dependent inhibition of cell-mediated LDL lipid oxidation. Taken together, these data suggest that an AACOCF 3 -sensitive PLA 2 activity is required for human monocyte-mediated LDL lipid oxidation. To investigate potential nonspecific effects of AACOCF 3 , we also examined its toxicity to human monocytes and its ability to function as a general antioxidant. For the cytotoxicity studies we used an assay measuring [ 14 C]adenine metabolite release. We have shown that the results obtained with this assay correlate well with the chromium release assay of toxicity (16).
The results, presented in Table I, demonstrate that AACOCF 3 , at doses used for these studies, showed less than 5% toxicity for human monocytes (shown in Table I) or U937 cells (data not shown). We also evaluated the general antioxidant activity of AACOCF 3 as measured by its inhibition of copper-induced LDL lipid oxidation as described previously (16,35). The results of these studies are presented in Table II. AACOCF 3 showed no inhibition of copper-mediated LDL lipid oxidation at concentrations from 1 to 50 M, indicating that AACOCF 3 did not exhibit antioxidant activity at concentrations used in the studies.
To corroborate our findings with AACOCF 3 and to determine whether, indeed, cPLA 2 was required, we used another approach to regulate cPLA 2 activity. For these studies, we used cPLA 2 -specific antisense oligonucleotides to inhibit cPLA 2 expression. We also used a cPLA 2 sense oligonucleotide as a control in parallel cultures. In these experiments, human monocytes were incubated with sense or antisense phosphorothioate-modified oligonucleotides (HPLC-purified) in the presence or absence of ZOP and LDL as described under "Materials and Methods." After incubation, both cPLA 2 protein expression and cPLA 2 enzymatic activity were determined. As shown in Fig. 5A, cPLA 2 protein expression in activated human monocytes was inhibited by cPLA 2 -specific antisense oligonucleotide treatment (lane 3 of Fig. 5A) as detected by Western blotting analysis using cPLA 2 -specific antibody. Sense oligonucleotide treatment had no effect on human monocyte cPLA 2 protein expression (lane 4 of Fig. 5A). We also examined the cPLA 2 activity in lysates of monocytes that had been treated with antisense or sense oligonucleotides. These data are shown in Fig. 5B. cPLA 2 activity was increased upon monocyte activation as previously observed (Fig. 2B) and antisense oligonucleotide treatment substantially inhibited cPLA 2 activity. In contrast, treatment with the sense oligonucleotides did not alter cPLA 2 activity. Furthermore, we also monitored whether oligonucleotide treatment had any nonspecific inhibitory effects on the assay. In this experiment, U937 cell lysates and oligonucleotides were included together during the cPLA 2 activity assay. Neither antisense oligonucleotides nor sense oligonucleotides affected the cPLA 2 activity assay itself (data not shown).
Next, we evaluated human monocyte-mediated LDL lipid oxidation after treatment with cPLA 2 -specific antisense or sense oligonucleotides. Treatment with cPLA 2 -specific anti-  a Human monocytes (1 ϫ 10 6 /ml) were labeled with [ 14 C]adenine and then incubated together with ZOP (2 mg/ml), LDL (0.5 mg of cholesterol/ml), and different concentrations of AACOCF 3 as indicated for 24 h. After incubation, the supernatants were collected.
b Release of radioactivity was determined as described under "Materials and Methods." All experiments were performed in triplicate. The results are expressed as mean Ϯ S.D. of data obtained in one of three similar experiments.
c Percent release of [ 14 C]adenine was determined by the equation described under "Materials and Methods." sense oligonucleotides caused a dose-dependent decrease in monocyte-mediated LDL lipid oxidation as detected by the TBA assay (Fig. 6). In contrast, treatment with sense oligonucleotides caused no significant inhibition of LDL lipid oxidation (* indicates p Ͻ 0.05). These data suggest that cPLA 2 activity is a critical regulator of the oxidation of LDL by activated human monocytes.
Previously, our laboratory has shown that O 2 . production is required for monocyte-mediated LDL lipid oxidation, and arachidonic acid has been shown to regulate the activity of the NADPH oxidase O 2 . generating complex. We therefore examined whether cPLA 2 activity was essential for O 2 . production. In this experiment, human monocytes were preincubated with either cPLA 2 -specific antisense or sense oligonucleotides for 24 h and then O 2 . production was quantified in response to activation. As expected, O 2 . production was increased upon monocyte activation as shown in Fig. 7A. Antisense oligonucleotide treatment significantly inhibited O 2 . production in activated human monocytes, whereas sense oligonucleotide treatment was without significant effect. Furthermore, we found that the inhibitory effect of antisense treatment could be negated by addition of arachidonic (AA) one product of cPLA 2 activity. AA alone or with ZOP did not alter O 2 . production except to restore levels to normal in antisense-treated, activated monocytes. We then conducted similar experiments to attempt to restore the ability of antisense-treated, activated monocytes to oxidize LDL lipids. In these experiments (see Fig. 7B), addition of AA restored LDL lipid oxidation by 50% in antisense-treated cells. Addition of lysophosphatidyl choline (lyso-PC) or lyso-PC plus AA did not fully restore LDL lipid oxidation in antisensetreated cells (data not shown). Treatment with AA and/or lyso-PC did not alter levels of LDL lipid oxidation in unactivated monocytes nor in ZOP-activated monocytes that were not treated with oligonucleotides or were treated with sense oligonucleotides.

DISCUSSION
In our previous studies, we found that human peripheral blood monocytes could oxidize LDL in an activation-dependent manner (1,2). We also found that O 2 . production (2), increases in intracellular Ca 2ϩ levels (16), and induction of PKC activity were required as well (17). These observations suggested that one or more Ca 2ϩ -and protein phosphorylation-dependent intracellular signaling pathways regulated monocyte function and participated in the process of monocyte-mediated LDL lipid oxidation. A potential candidate for one of these pathways was the high molecular weight cPLA 2 . We hypothesized that cPLA 2 might prove to be an important regulatory pathway in the oxidation of LDL by activated monocytes. We found that low levels of cPLA 2 protein were detectable in   unactivated human monocytes, and upon activation, the cPLA 2 protein levels and enzymatic activity were substantially increased. The time course studies showed a correlation in the increase of cPLA 2 protein and enzymatic activity, with both reaching a maximum at 12 h of activation. Our previous studies documented an increase in intracellular Ca 2ϩ levels and induction of PKC activity occurring within 30 min of monocyte activation, demonstrating that these two events occur early in the course of human monocyte activation (16,17). Our previous studies also demonstrated that LDL lipid oxidation begins to be detectable from 4 to 6 h after monocyte activation and then increases and begins to plateau at 12 h of activation and gradually increases to 24-h levels (5). Taking all of these observations together, our studies demonstrate that increases in intracellular Ca 2ϩ levels and activation of PKC precede the induction of cPLA 2 activity and that the induction of cellular cPLA 2 activity closely correlates with that of monocyte-mediated LDL lipid oxidation (5).
In experiments using several types of functionally and structurally diverse pharmacologic inhibitors of PLA 2 , results indicated that cPLA 2 activity was required for monocyte-mediated LDL lipid oxidation (data not shown). To confirm this observation we used another inhibitor, AACOCF 3 . AACOCF 3 is an analog of arachidonic acid in which the COOH group is replaced with COCF 3 (trifluoromethyl ketone) (37). It has been reported to be a selective inhibitor of cPLA 2 in that it is 500-fold more potent as an inhibitor of cPLA 2 as compared with sPLA 2 (37). Since sPLA 2 has seven disulfide bridges and is inactivated by DTT, we could distinguish between sPLA 2 and cPLA 2 activities. AACOCF 3 only inhibited DTT-resistant PLA 2 activity but not DTT-sensitive PLA 2 activity, and inhibition was concentration-dependent, demonstrating that cPLA 2 activity was selectively inhibited. Importantly, AACOCF 3 also inhibited human monocyte-mediated LDL lipid oxidation in a dose-dependent fashion. In concert, these data supported our hypothesis that cPLA 2 activity was required for monocyte-mediated LDL lipid oxidation; recently, however, it has been reported that AA-COCF 3 can inhibit cytosolic iPLA 2 activity in a mouse macrophage cell line, P388D1 (38). Technically, it is difficult to distinguish cPLA 2 enzymatic activity from iPLA 2 activity in this assay, because both cPLA 2 and iPLA 2 are insensitive to DTT and both activities would be detected in our assay system (19,39).
To more specifically address the participation of cPLA 2 in this process, cPLA 2 -specific antisense and sense oligonucleotides were developed. The sequence was carefully chosen from a region lacking substantial homology with other sequenced human genes. The oligonucleotides were phosphorothioatemodified to limit degradation and purified by HPLC prior to use to remove all incomplete synthesis products thereby limiting nonspecific effects. We have found this latter step to be critical in rendering specificity to antisense oligonucleotide regulation in human monocytes. The finding that antisense treatment, but not treatment with sense oligonucleotides, resulted in decreased cPLA 2 protein expression and decreased enzymatic activity leads us to believe that the decrease in monocyte oxidation of LDL was indeed due to inhibition of cPLA 2 . Recent reports have also shown that cPLA 2 protein expression and enzymatic activity can be inhibited by cPLA 2 antisense oligonucleotide treatment in human monocytes (40,41). In these studies, different cPLA 2 antisense oligonucleotide sequences were used including sequences directly recognizing the initiation site of transcription (40) and sequences recognizing cPLA 2 mRNA downstream from the initiation site (41).
An important mechanistic finding of this study is that monocyte-mediated O 2 . production is inhibited by suppression of cPLA 2 activity (Fig. 7A). Although numerous studies report that AA and phospholipase A 2 appear to regulate O 2 . production, this is the first report that specific suppression of cPLA 2 protein expression and enzymatic activity, using cPLA 2 -specific antisense oligonucleotide treatment, inhibited O 2 . production by activated human monocytes. The finding that cPLA 2 regulates LDL lipid oxidation is also novel. Interestingly, both O 2 . production and LDL lipid oxidation were inhibited to the same extent, but as discussed below, neither was inhibited completely (see Fig. 7 production by antisense-treated, activated monocytes (Fig. 7A), whereas AA only partially restored the capacity of antisensetreated, activated monocytes to oxidize LDL (Fig. 7B). We also examined whether the addition of lyso-PC or both lyso-PC ϩ AA could restore the LDL oxidation mediated by antisensetreated, activated monocytes, but some inhibition of LDL oxi- dation remained (data not shown). These data suggest that AA is the cPLA 2 product required for regulating O 2 . production, whereas AA in addition to another product, likely a specific lysophospholipid, both participate in modulating the process of LDL lipid oxidation. The fact that lyso-PC was not restorative for this function even in the presence of AA indicates that a phospholipid other than PC is the essential substrate in regulating this process. Our published studies have shown that inhibition of PKC decreases O 2 . production by activated monocytes as well as inhibiting LDL oxidation (17). cPLA 2 activity is reportedly regulated by both PKC-dependent and PKC-independent pathways (29). In recent studies, we have found that inhibition of PKC activity caused a related inhibition of cPLA 2 activity in activated human monocytes, 3 thus suggesting that PKC-dependent phosphorylation events regulate cPLA 2 activity. It would appear then that cPLA 2 is an intermediary enzyme in the signal transduction pathway involving PKC regulation of O 2 . production and LDL oxidation.
Another observation from these studies was that antisense oligonucleotide treatment almost completely inhibited cPLA 2 protein expression and enzymatic activity, as measured in monocyte lysates; however, monocyte-mediated O 2 . production and LDL lipid oxidation were not completely suppressed. Our data indicate the presence of a constitutive PLA 2 activity that was not inhibited by antisense to cPLA 2 (see Fig. 5 and Fig. 7). This activity might be due to another PLA 2 , which may also participate in regulating monocyte O 2 . production and LDL oxidation. Multiple forms of PLA 2 s in individual cell types have been reported, such as in the mouse macrophage-like cell line P388D 1 (43), canine vascular smooth muscle cells (44), and the rat mast cell line, RBL-2H3 (45). Further investigations are needed to define the involvement of other PLA 2 activities. In addition, the incomplete inhibition of monocyte-mediated LDL lipid oxidation by both AACOCF 3 and cPLA 2 -specific antisense oligonucleotide treatment may also suggest that one or more parallel, cPLA 2 -independent pathways may regulate monocyte activation and be involved in LDL oxidation. In summary, our data demonstrate that cPLA 2 activity plays an important role in both O 2 . production and optimal LDL lipid oxidation by activated human monocytes. Our current working hypothesis regarding the events required for monocyte oxidation of LDL lipids is that after activation, intracellular Ca 2ϩ levels are increased, by both the influx of extracellular Ca 2ϩ and the release of intracellular Ca 2ϩ , thus causing the induction of PKC activity, which together with Ca 2ϩ can regulate cPLA 2 activity (1,2,16,17). The activation of cPLA 2 then, in concert with Ca 2ϩ and PKC, causes the induction of O 2 . production which participates in the oxidation of LDL. These signaling pathways form a network and induce optimal LDL lipid oxidation by activated human monocytes. Knowledge acquired from studies such as these will contribute to the understanding of the mechanisms of monocyte oxidation of LDL lipids and may suggest optimal points for intervening in this process.