Identification of Calcium-independent Phospholipase A2 (iPLA2) β, and Not iPLA2γ, as the Mediator of Arginine Vasopressin-induced Arachidonic Acid Release in A-10 Smooth Muscle Cells

The agonist-stimulated release of arachidonic acid (AA) from cellular phospholipids in many cell types (e.g. myocytes, β-cells, and neurons) has been demonstrated to be primarily mediated by calcium-independent phospholipases A2 (iPLA2s) that are inhibited by the mechanism-based inhibitor (E)-6-(bromomethylene)-3-(1-naphthalenyl)-2H-tetrahydropyran-2-one (BEL). Recently, the family of mammalian iPLA2s has been extended to include iPLA2γ, which previously could not be pharmacologically distinguished from iPLA2β. To determine whether iPLA2β or iPLA2γ (or both) were the enzymes responsible for arginine vasopressin (AVP)-induced AA release from A-10 cells, it became necessary to inhibit selectively iPLA2β and iPLA2γ in intact cells. We hypothesized that the R- and S-enantiomers of BEL would possess different inhibitory potencies for iPLA2β and iPLA2γ. Accordingly, racemic BEL was separated into its enantiomeric constituents by chiral high pressure liquid chromatography. Remarkably, (S)-BEL was approximately an order of magnitude more selective for iPLA2β in comparison to iPLA2γ. Conversely, (R)-BEL was approximately an order of magnitude more selective for iPLA2γ than iPLA2β. The AVP-induced liberation of AA from A-10 cells was selectively inhibited by (S)-BEL (IC50 ∼2 μm) but not (R)-BEL, demonstrating that the overwhelming majority of AA release is because of iPLA2β and not iPLA2γ activity. Furthermore, pretreatment of A-10 cells with (S)-BEL did not prevent AVP-induced MAPK phosphorylation or protein kinase C translocation. Finally, two different cell-permeable protein kinase C activators (phorbol-12-myristate-13-acetate and 1,2-dioctanoyl-sn-glycerol) could not restore the ability of A-10 cells to release AA after exposure to (S)-BEL, thus supporting the downstream role of iPLA2β in AVP-induced AA release.

At present, three distinct subclasses of iPLA 2 have been identified at the genetic level (with subsequent confirmation of iPLA 2 catalytic activity by recombinant technologies) and have been designated iPLA 2 ␣, iPLA 2 ␤, and iPLA 2 ␥, in order of their discovery (22)(23)(24). The iPLA 2 s have been categorized based upon their strict conservation of nucleotide-binding (GXGXXG) and lipase (GXSTG) consensus sequences (Fig. 1). Two of the iPLA 2 subclasses, iPLA 2 ␤ and iPLA 2 ␥, have been cloned from mammalian cDNA libraries, whereas the ortholog of iPLA 2 ␣ (patatin), at the time of writing this paper (with 97.8% of the human genome sequenced), has not been identified in mammals. Calcium-independent phospholipase A 2 ␤ contains eight ankyrin-repeat domains that are believed to facilitate intracellular sorting (23,25,26) and a CaM-binding domain near the C terminus which binds calcium-activated CaM and regulates enzyme activity (27) (Fig. 1). The binding of CaM to iPLA 2 ␤ results in inhibition of iPLA 2 ␤ activity which is reversible * This work was supported by National Institutes of Health Grants 2PO1HL57278-06A1, 2RO1HL41250-10, and 5RO1AA11094-05. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Many prior studies of iPLA 2 have utilized the mechanismbased suicide inhibitor rac-BEL as a pharmacologic tool to identify the type of intracellular phospholipase A 2 involved in many diverse cellular processes. Because rac-BEL inhibits both iPLA 2 ␤ and iPLA 2 ␥ at low micromolar concentrations (24,25,31,32), it is impossible to assign rac-BEL-mediated inhibition of AA release to iPLA 2 ␤ or iPLA 2 ␥ activities. Accordingly, it became necessary to develop pharmacologic approaches that could discriminate between iPLA 2 ␤ and iPLA 2 ␥ to facilitate identification of their biologic roles. In addition, it has been reported previously (33,34) that high concentrations of BEL (25 M) partially inhibit the magnesium-dependent cytosolic phosphatidate phosphohydrolase, PAP-1, which converts phosphatidic acid to diacylglycerol (DAG). In those investigations, it was proposed that PAP-1 inhibition by BEL would prevent activation of protein kinase C leading to attenuated AA release. However, "rescue" experiments in which PKC was exogenously activated by phorbol esters or diacylglycerol analogs after BEL treatment were not reported by the authors to address their hypothesis (33,34).
In our early experiments we employed rac-BEL to demonstrate a 1000-fold selectivity between iPLA 2 versus cPLA 2 and sPLA 2 family members (31,32). Based upon the increasing appreciation of the utility of chiral pharmacologic agents in enhancing the specificity of inhibitors toward targeted biologic processes, we hypothesized that (R)-and (S)-BEL could differentially inhibit iPLA 2 ␤ and iPLA 2 ␥ activities. Moreover, we reasoned that development of chiral mechanism-based inhibitors could provide an increased degree of discrimination between specific targeted enzyme systems and those of "nonspecific inhibition." Here we report the resolution of racemic BEL into its individual enantiomeric constituents by chiral HPLC and the selective inhibition of iPLA 2 ␤ by (S)-BEL and iPLA 2 ␥ by (R)-BEL, and we demonstrate that BEL-mediated inhibition of AA release in A-10 cells is likely mediated by iPLA 2 ␤ and not due to inhibition of iPLA 2 ␥ or the effects of BEL on MAPK or PKC activation.
Recombinant iPLA 2 Enzymes-Recombinant iPLA 2 ␤ was expressed and purified from Sf9 cells as described previously (35). Recombinant full-length iPLA 2 ␥ was expressed in Sf9 cells, and the membrane fraction was washed and isolated as described previously (24).
Separation of BEL Enantiomers-BEL enantiomers were resolved by HPLC utilizing a Chirex column of 3,5-dinitrobenzoyl-(R)-phenylglycine attached to a silica matrix as the stationary chiral phase. The chiral column was equilibrated with hexane/dichloroethane/ethanol (150:15:1), and optical enantiomers were eluted isocratically at a flow rate of 2 ml/min. Elution of BEL racemates from the column was monitored by UV absorbance at 280 nm. Peaks corresponding to the Rand S-enantiomers were collected, dried under N 2 , and stored at Ϫ20°C. The concentration of BEL for each experiment was determined spectrophotometrically based on UV absorbance (⑀ 280 ϭ 6130 cm Ϫ1 M Ϫ1 in acetonitrile).
were incubated with (R)-BEL, (S)-BEL, racemic BEL, or ethanol vehicle for 3 min at 22°C in the presence of 100 mM Tris-HCl, pH 7.0, and 4 mM EGTA (for iPLA 2 ␤) or 100 mM Tris acetate, pH 8.0, and 4 mM EGTA (for iPLA 2 ␥). The concentration of BEL used for each experiment ranged from 0 to 16 M. L-␣-1-Palmitoyl-2-[1-14 C]arachidonylphosphatidylcholine (5 M final concentration) or L-␣-1-palmitoyl-2-[1-14 C]oleoylphosphatidylcholine (5 M final concentration) in ethanol was then added to each sample and incubated at 37°C for 2 min. Reactions were terminated by extraction of radiolabeled products into butanol, and reactants and products were separated by thin layer chromatography using Whatman LK6D 60-Å Silica Gel plates with petroleum ether/ethyl ether/acetic acid (70:30:1) as the mobile phase. Regions corresponding to the migration of a fatty acid standard visualized by iodine staining were scraped into vials, and radioactivity was quantified by scintillation spectrometry.

Quantification of [ 3 H]Arachidonic Acid Liberation from A-10
Phospholipids-Rat aortic smooth muscle A-10 cells, cultured in 60-mm dishes (2.5 ϫ 10 5 cells/dish), were radiolabeled with 0.5 Ci of [ 3 H]arachidonic acid per dish as described previously (29). Cells were washed once with DMEM containing 0.25% fatty acid-free bovine serum albumin followed by two washes with DMEM alone. Cells were then incubated with the indicated concentrations of (R)-, (S)-, rac-BEL, or ethanol vehicle (0.1% final concentration) in DMEM for 20 min. This medium was removed, and the cells were then incubated with DMEM containing 10% heat-inactivated fetal bovine serum in the absence or presence of 1 M AVP. In some experiments, PMA (1 M) or DOG (10 M) was added to the medium containing AVP. After 5 min, 1 ml of this medium was removed; lipids were extracted into 2 ml of chloroform/methanol/acetic acid (25:24:1, v/v) (38), and the remaining cells were scraped into 1 ml of deionized water prior to lipid extraction as described above. The chloroform layer was evaporated under nitrogen, and the extracted lipids were separated by thin layer chromatography (petroleum ether/ ethyl ether/glacial acetic acid, 70:30:1). Regions containing fatty acids and phospholipids were scraped into vials, and radioactivity was quantified by liquid scintillation spectrometry.
Measurement of Cytosolic and Membrane-bound Phosphatidate Phosphohydrolase Activities-A-10 cells were grown to confluency, washed twice in ice-cold phosphate-buffered saline, and harvested in lysis buffer (50 mM Tris-HCl, pH 7.4, containing 0.25 M sucrose and 0.2 mM dithiothreitol). After brief sonication utilizing a Vibra-Cell VC40 sonicator (5 times with 1-s pulses at 30% power), the lysed cell suspension was centrifuged at 100,000 ϫ g for 1 h to separate cytosolic and membrane fractions. In some experiments, A-10 cells were washed and pretreated with BEL or ethanol vehicle (as described above for experiments examining [ 3 H]AA liberation) before isolation of cell homogenates. Each fraction (40 l) was preincubated with BEL (up to 200 M) or ethanol vehicle at 22°C for 5 min in the presence of 50 mM Tris-HCl, pH 7.2, containing 10 mM ␤-mercaptoethanol, 2 mM MgCl 2 , and 1 mM EGTA (90 l final volume). Dipalmitoylphosphatidic acid (100 M final concentration containing 0.05 Ci of L-␣-dipalmitoyl[U-glycerol- 14 C]phosphatidic acid per reaction in the presence of 1 mM Triton X-100) was added to each reaction and incubated at 37°C for 5-10 min. Reactions were terminated with 900 l of 5% acetic acid and extracted into chloroform by the method of Bligh and Dyer (38) prior to separating dipalmitoylglycerol by TLC utilizing chloroform/methanol/water (65: 35:2) as the mobile phase prior to quantification by scintillation spectrometry.
Determination of Phosphorylated MAPK-Confluent A-10 cells in 10-cm dishes were incubated overnight in the presence of DMEM containing 1% fetal bovine serum to reduce background phosphorylation of ERK1 and ERK2. Cells were washed twice with DMEM without serum and preincubated with 5 M (R)-BEL, (S)-BEL, rac-BEL, or ethanol vehicle in DMEM without serum for 15 min at 37°C. This medium was then removed, and the cells were incubated with 1 M AVP for 5 min at 37°C. After washing once with ice-cold phosphate-buffered saline (PBS), cells were scraped into RIPA buffer (PBS, pH 7.4, containing 1% Igepal CA-630, 0.5% sodium deoxycholate, 0.1% SDS, 0.5 mM 4-(2aminoethyl)benzenesulfonyl fluoride, 10 g/ml aprotinin, and 1 mM sodium orthovanadate), incubated on ice for 30 min, and then centrifuged at 10,000 ϫ g for 10 min. The protein concentrations of the sample supernatants were determined utilizing the bicinchoninic acid (BCA) assay (Pierce) with bovine serum albumin (BSA) as a standard. Samples were electrophoresed according to the method of Laemmli (39) and transferred to a polyvinylidene difluoride membrane by electroelution in 10 mM CAPS, pH 11, for ECL Western analysis. After blocking with Tris-buffered saline containing 0.1% Tween 20 (TBS-T) and 5% non-fat dry milk for 2 h, primary rabbit polyclonal antibodies against phosphorylated (pTEpY) and dephosphorylated MAPK diluted 1:5000 in PBS containing 5% BSA were incubated with the blot for 1 h. After washing with TBS-T, the blots were incubated with ice-cold PBS containing 0.25% glutaraldehyde for 15 min as described previously (40), washed, and incubated with a protein A-peroxidase conjugate diluted (1:5000) in TBS-T containing 5% BSA for 1 h. Immunoreactive bands were visualized by ECL as described by the manufacturer (Amersham Biosciences).
Determination of PKC Translocation-Confluent A-10 cells in 10-cm dishes were washed twice with DMEM without serum, followed by incubation with either 5 M (S)-BEL or ethanol vehicle in DMEM for 15 min at 37°C. This medium was then removed, and DMEM with or without 1 M AVP was incubated with the cells for 5 min. After washing with ice-cold PBS, the cells were collected by scraping into 20 mM Tris-HCl, pH 7.4, containing 0.33 M sucrose, 5 mM EDTA, 0.5 mM 4-(2-aminoethyl)benzenesulfonyl fluoride, and 5 g/ml leupeptin and were lysed by three cycles of flash freezing with liquid nitrogen and thawing. Each sample was then further homogenized utilizing a Teflon homogenizer before isolating the low speed pellet (1,000 ϫ g), membrane (100,000 ϫ g pellet), and cytosol (100,000 ϫ g supernatant) fractions. The protein concentrations of the fractions were determined utilizing the bicinchoninic acid (BCA) assay (Pierce) with BSA as a standard. Samples were electrophoresed and subjected to ECL Western analysis utilizing rabbit polyclonal antibodies against PKC␣ and PKC⑀ as described above for MAPK phosphorylation. For blots incubated with mouse monoclonal antibodies against PKC␦ and PKC, an anti-mouse IgG (Fab-specific)-peroxidase conjugate was utilized in place of the protein A-peroxidase conjugate.

RESULTS
Separation of BEL Enantiomers-Because BEL contains a chiral center at C-2 ( Fig. 2), we hypothesized that (R)-BEL and (S)-BEL might have different potencies and/or selectivities for iPLA 2 ␤ and iPLA 2 ␥ so that individual enantiomers of BEL could be exploited to identify the roles of iPLA 2 ␤ and iPLA 2 ␥ in agonist-stimulated AA release in intact cells. Accordingly, a chiral HPLC column was used to separate (R)-BEL and (S)-BEL from rac-BEL (Fig. 2). Separation of the BEL enantiomers under the conditions employed resulted in resolution of two major UV-absorbing peaks with an R T difference of ϳ2 min. The first peak eluted at 18.8 min (Peak A), and the second peak eluted at 20.5 min (Peak B) (relative retention time (R RT ) Peak A/Peak B ϭ 0.917) (Fig. 2). Integration of the areas of each peak was identical (within 1%), suggesting that the enantiomers of BEL had been separated. Proton NMR data demonstrated peaks with the anticipated chemical shifts and coupling constants as described previously (data not shown) (41). Moreover, electrospray ionization-mass spectrometric analysis of the ma- Re-injection of Peak A or Peak B onto the chiral column demonstrated that each purified moiety eluted at its previous retention time with negligible amounts of contaminating material demonstrating that, as expected, no equilibration had occurred. Finally, both peaks co-eluted utilizing a non-chiral C18 HPLC column (data not shown).
Identification of the Absolute Chirality of the BEL Enantiomers-To determine the absolute chirality of the resolved BEL enantiomers, synthetic enantiomers of ␣NpI6 of known chirality were chromatographed separately and together on the chiral stationary phase. Under these conditions, (S)-␣NpI6 (R T ϭ 19.5 min) eluted prior to (R)-␣NpI6 (R T ϭ 21.4 min) with approximately the same degree of separation (R RT ϭ 0.911) as the BEL enantiomers. Chymotrypsin has been identified previously as a suitable target for aromatic haloenol lactones resulting in its mechanism-based inhibition as detailed by careful kinetic analyses by Katzenellenbogen (41)(42)(43)(44). In prior studies, (R)-BEL was determined to be a more efficient inhibitor of chymotrypsin than (S)-BEL in comparison to its chiral counterpart (36). To substantiate further the absolute stereochemistry of the BEL enantiomers resolved by chiral HPLC and to confirm the ability of the resolved enantiomers to inhibit selectively chymotrypsin activity, increasing concentrations of (R)-BEL, (S)-BEL, or rac-BEL were incubated with chymotrypsin and diluted in buffer as described under "Experimental Procedures." Activity assays were performed using the fluorogenic chymotrypsin substrate N-succinyl-Ala-Ala-Pro-Phe-7-amido-4methylcoumarin as described under "Experimental Procedures." Under the conditions employed, (R)-BEL stoichiometrically and irreversibly inhibited chymotrypsin, whereas(S)-BEL was considerably less potent (Fig. 3). These results confirm that both peaks are distinct enantiomers of BEL and substantiate, by independent criteria, the assigned absolute stereochemistry of the peaks eluting from the chiral HPLC column.
Identification of iPLA 2 ␤ and Not iPLA 2 ␥ as the Mediator of AVP-induced AA Release in A-10 Cells-Upon stimulation with AVP, A-10 smooth muscle cells rapidly released a relatively large percentage (ϳ5%) of their esterified arachidonic acid (10). Previous studies (10,29) have demonstrated that pretreatment of the cells with 2-5 M BEL inhibits Ϸ60 -80%, respectively, of AVP-inducible arachidonic acid release. Moreover, the absence of extracellular calcium ion (incubations performed in the presence of EGTA in the media) or the presence of intracellular calcium ion chelators (e.g. BAPTA) does not affect AVP-induced AA release in A-10 cells (29). Because combined incubations with EGTA and BAPTA completely ablated FURA-2-observable increases in intracellular calcium ions, these results further implicated the involvement of a calcium-independent phospholipase A 2 in this process. However, since both iPLA 2 ␤ and the newly identified iPLA 2 ␥ are both inhibited by rac-BEL (and are calcium-independent), the identity of the iPLA 2 -mediating AA release in AVP-stimulated A-10 cells was unknown. To address this issue, we exploited the selectivity of (S)-BEL and (R)-BEL for inhibition of iPLA 2 ␤ and iPLA 2 ␥, respectively, to determine the type of iPLA 2 -catalyzing AVP-induced release of [ 3 H]arachidonic acid from A-10 cells. As demonstrated previously, A-10 cells stimulated with AVP resulted in a substantial increase in the amount of nonesterified [ 3 H]arachidonic acid relative to control cells incubated with vehicle alone (Fig.  5). This AVP-induced increase in non-esterified [ 3 H]arachidonic acid was significantly reduced in the presence of low concentrations of rac-BEL (1 and 5 M, p Ͻ 0.01 and p Ͻ 0.001, respectively) (Fig. 5). Importantly, (S)-BEL (1 M) significantly inhibited AA release (40% inhibition, p Ͻ 0.01) and 5 M (S)-BEL largely attenuated AVP-induced AA release (80% inhibition, p Ͻ 0.001). In sharp contrast, (R)-BEL is virtually ineffective in inhibiting AVP-induced AA release from A-10 cells under similar conditions (Fig. 5). Thus, iPLA 2 ␤, and not iPLA 2 ␥, is the likely mediator of AA release in this system.

Confirmation of the Lack of Effects of BEL on Processes
Typically Associated with cPLA 2 ␣ Activation-Activation of cPLA 2 ␣ in most systems depends on the concomitant activation of MAPK, PKC, and increases in intracellular [Ca 2ϩ ] (45)(46)(47). In previous studies (29) we demonstrated that BEL does not inhibit AVP-induced increases in [Ca 2ϩ ] i and that ablation of changes in [Ca 2ϩ ] i by BAPTA does not attenuate AA release. Recently, Dennis and co-workers (33,34) have suggested that cytosolic phosphatidate phosphohydrolase (PAP-1) in some cell types may be a target for BEL and that the resulting inhibition of PAP-1 would result in diminished levels of diacylglycerol produced from phosphatidic acid, thereby attenuating PKC activation precluding cPLA 2 ␣ activation and AA release. To address this possibility, we first examined the effects of rac-BEL on A-10 cell PAP activities in cytosol and membrane fractions (Fig. 6A) as well as in intact cells (Fig. 6B). These experiments consistently demonstrated the lack of any effect of BEL on either the cytosolic (PAP-1) or the membrane-bound (PAP-2) forms of A-10 cell phosphatidate phosphohydrolase at concentrations up to 200 M BEL (Fig. 6A). Furthermore, homogenates from intact A-10 cells previously exposed to up to 100 M racemic BEL did not inhibit total phosphatidate phosphohydrolase activity in comparison to ethanol-treated controls (Fig. 6B). Next, we examined whether activation of PKC by exogenous addition of either PMA or DOG could rescue AA release after BEL pretreatment. Neither PMA nor DOG could restore the ability of (S)-BEL treated A-10 cells to release arachidonic acid, thereby demonstrating that BEL is likely inhibiting arachidonic acid release in a manner that is no longer responsive to PKC activation (i.e. irreversible covalent modification of iPLA 2 ) (Fig. 7). A-10 cells contain at least four PKC isoforms, PKC␣, PKC␦, PKC⑀, and PKC, by Western blot analysis (Fig. 8A); however, no bands corresponding to PKC␤I, PKC␤II, PKC␥, PKC, or PKC could be visualized in previous work. 3 Accordingly, we treated A-10 cells with AVP and determined whether (S)-BEL could inhibit PKC translocation to the membrane fraction. Stimulation of A-10 cells with AVP causes translocation of PKC␦ and PKC⑀ from the cytosol to the membrane fraction, but neither PKC␣ nor PKC undergo AVPinduced translocation in A-10 cells (Fig. 8A). Pretreatment of A-10 cells with 5 M (S)-BEL, which causes almost complete inhibition of AA release, does not affect the translocation of either PKC␦ or PKC⑀ (Fig. 8A). Finally, AVP-induced phosphorylation of ERK2 is not affected by the presence of 5 M (R)-, (S)-, or rac-BEL (Fig. 8B). Collectively, these results demonstrate the following: 1) neither PAP-1 nor PAP-2 is a target for BEL in A-10 smooth muscle cells; 2) BEL does not appreciably affect PKC␦ and PKC⑀ translocation or MAPK phosphorylation in A-10 cells; and 3) iPLA 2 ␤ is likely responsible for the large majority of arachidonic acid release from A-10 cells.

DISCUSSION
Genetic approaches have now demonstrated the presence of two types of iPLA 2 activities present in the human genome (iPLA 2 ␤ and iPLA 2 ␥) which are both inhibited by rac-BEL (24,31,32). Accordingly, all prior experiments demonstrating inhibition of arachidonic acid release by rac-BEL cannot discriminate between hydrolysis catalyzed by iPLA 2 ␤ or that mediated by iPLA 2 ␥. Virtually nothing is known about the regulation of iPLA 2 ␥ or its potential role in agonist-stimulated eicosanoid release. In this work, we 1) resolve rac-BEL by chiral HPLC; 2) 3 D. A. Ford, C. M. Jenkins, and R. W. Gross, unpublished observations).

FIG. 5. Inhibition of AVP-mediated arachidonic acid liberation in A-10 smooth muscle cells by racemic, (R)-, and (S)-BEL.
A-10 cells (2.5 ϫ 10 5 cells/dish) were radiolabeled with [ 3 H]arachidonic acid (0.5 Ci/dish) for 20 h. After washing to remove unincorporated [ 3 H]arachidonic acid, cells were then incubated with either 1 or 5 M (R)-BEL, (S)-BEL, rac-BEL, or ethanol vehicle (0.1%) in DMEM for 20 min. The medium was removed, and the cells were then incubated with 3 ml of DMEM containing 10% heat-inactivated fetal bovine serum with or without 1 M AVP. After 5 min, 1 ml of this medium was removed, and lipids were extracted into 2 ml of chloroform/methanol/acetic acid (25:24:1). The remaining cells were scraped into 1 ml of deionized water, and total lipids were extracted as described above. The chloroform layer was evaporated under nitrogen, and the extracted lipids were separated by TLC. Regions containing fatty acids and phospholipids were scraped into vials, and radioactivity was quantified by liquid scintillation spectrometry. Results represent the average of four sets of data (Ϯ S.E.). Asterisks indicate comparisons of the inhibition of AA release by 1 M (*, p Ͻ 0.01) and 5 M (**, p Ͻ 0.001) rac-BEL and (S)-BEL relative to incubations containing AVP in the absence of inhibitor. assign the absolute stereochemistry of the resolved enantiomers by two independent techniques; 3) demonstrate a 10-fold selectivity of (S)-BEL for inhibition of iPLA 2 ␤ and a 10-fold selectivity of (R)-BEL for inhibition of iPLA 2 ␥; 4) demonstrate that (S)-BEL inhibits that vast majority of AVP-induced AA liberation in A-10 cells, whereas (R)-BEL does not; 5) provide evidence that BEL-mediated inhibition of AA release in A-10 cells is not mediated through inhibition of either membranebound or cytosolic phosphatidate phosphohydrolases; and 6) demonstrate that treatment of A-10 cells with (S)-BEL does not attenuate PKC translocation or MAPK activation after AVP stimulation. Collectively, these results, in combination with prior work (see below), demonstrate that AVP-stimulated AA release in A-10 cells is likely mediated by iPLA 2 ␤ and not iPLA 2 ␥, cPLA 2 ␣, or chymotrypsin-like proteases.
The utilization of chiral pharmacologic agents instead of racemic mixtures has increasingly been appreciated to enhance the potency of inhibitors toward targeted processes and markedly reduce toxicity and nonspecific inhibition mediated by interactions with non-targeted systems. Because enzymes possess multiple chiral centers, the interaction between a chiral inhibitor and one or more optically active centers at or near the enzyme active site results in diastereotopic interactions that possess different physical properties and spatial relationships for each diastereotopic pair. In the case of mechanism-based inhibitors such as BEL, these diastereotopic interactions are anticipated to 1) alter binding, 2) modify the rate of formation of the acyl-enzyme intermediate, and 3) alter the covalent trapping of the ␣-halomethyl ketone in the acyl enzyme by nucleophiles at or near the active site. In this study, we have exploited diastereotopic interactions between individual enantiomers of BEL and the known mammalian iPLA 2 s (i.e. iPLA 2 ␤ and iPLA 2 ␥) to achieve a remarkable specificity for inhibition of iPLA 2 ␤ by (S)-BEL and iPLA 2 ␥ by (R)-BEL. Moreover, we demonstrated that proteases with similar stereochemical relationships as chymotrypsin are more likely to be inhibited by (R)-BEL than (S)-BEL, thereby further increasing the utility of mechanism-based inhibition to gain insight into the types of phospholipases A 2 mediating AA release in mammalian cells.
We specifically point out that, as with any pharmacologic compound, unanticipated effects on non-targeted systems may occur with increasing likelihood at high concentrations of inhibitor. Mammalian cells have in excess of 30,000 genes which after splicing and post-translational modification give rise to well over 10 5 and perhaps as many as 10 6 different chemical moieties. Of course, it is impossible to test every compound with each of these chemical moieties in each different microenvironment in the cell in which pharmacologic agents might interact. Indeed, at high enough concentrations in aqueous systems, virtually any low molecular weight organic compound will interact with a diverse array of proteins due to hydrophobic effects alone. That is precisely why it is important to examine biologic effects elicited by pharmacologic agents at or near their effective inhibitory concentrations in intact cells as was determined in isolated purified systems. In the case of BEL, some investigators have employed 50 -100 M BEL which exceeds the effective inhibitory concentration of BEL for the known mammalian iPLA 2 s by almost 2 orders of magnitude. Accordingly, these experiments must be interpreted with caution given the IC 50 of rac-BEL for iPLA 2 ␤ and iPLA 2 ␥ is in the 0.5-3 M range. Moreover, the mere exposure of cells to high concentrations of organic compounds (50 -100 M) is likely to perturb the fragile order of the membrane microenvironment and, in the case of investigating membrane-related phenomena, may have effects that are independent of interactions with targeted enzyme systems alone. Indeed, we have observed cell death employing 100 M BEL which is almost certainly independent of the effects of BEL on targeted processes. 4 Dennis and co-workers have contended that high concentrations of rac-BEL (Ϸ 25 M) effectively inhibit magnesium-dependent cytosolic phosphatidate phosphohydrolase (PAP-1) in mouse P388D 1 macrophages (33) and human amnionic WISH cells (34). The authors argue that inhibition of PAP-1 would be expected to cause a deficiency in DAG, thus blunting PKC activation and possibly activation of cPLA 2 ␣ by MAPK. However, BEL has been subsequently shown not to affect PMAinduced translocation of PKC (or PKC catalytic activity) in P388D 1 macrophages (15) or MAPK phosphorylation in WISH cells (34), rat neutrophils (48), and A-10 cells (this work). It should be mentioned that in their investigations, Balsinde and Dennis (33) did not observe any effect of BEL on PAP-2, the phosphatidate phosphohydrolase isoform which is believed to be involved in lipid signal transduction pathways (49 -51). This absence of PAP-2 inhibition by BEL has since been observed in McA-RH7777 rat hepatoma cells (17), pancreatic islet cells (52), and A-10 smooth muscle cells (this work). Furthermore, cytosolic PAP-1 activity in McA-RH7777 cells is not significantly inhibited by 100 M BEL (17). A second possible effect of inhibited PAP-catalyzed DAG production, as described by Balboa et al. (34), is that the phospholipid substrate will be in a suboptimal environment because DAG has been demonstrated to alter membrane bilayers by creating more distance between phospholipid head groups, thereby making the phospholipid ester linkages more susceptible to PLA 2 -mediated hydrolysis. However, we have found that rac-BEL does not inhibit release of inositol bisphosphate or inositol 1,4,5-trisphosphate in A-10 cells (10), and Akiba et al. (15) have found that BEL (up to 5 M) does not significantly affect levels of diacylglycerol or phosphatidic acid formed in P388D 1 macrophages upon stimulation with zymosan.
The identification of chiral specificity of individual enantiomers of BEL to inhibit the known mammalian iPLA 2 s extends the utility of mechanism-based inhibitors in the study of agonist-mediated AA release. The experiments described herein allow, for the first time, assignment of AVP-induced AA release in A-10 cells to iPLA 2 ␤ and not iPLA 2 ␥. The inhibition of AA release by (S)-BEL and not (R)-BEL excludes participation of chymotrypsin or chymotrypsin-like proteases in these processes. Finally, the utilization of chiral mechanism-based inhibitors reduces potential nonspecific complications through comparisons of the effects of specific optical antipodes on the observed end points (i.e. AA release) with their in vitro potency in purified systems. We point out that assignation of specific enzymes as effectors of AA release requires detailed concurrent consideration of biochemical, pharmacologic, and genetic perturbations on the observed process. In the case of AVP-induced AA release from A-10 cells, we have demonstrated the following: 1) that concentrations of (S)-BEL near the IC 50 for iPLA 2 ␤ markedly attenuate AA release in intact A-10 cells whereas(R)-BEL does not (this study); 2) that BEL-mediated inhibition of AA release occurs in the presence of normal increases in [Ca 2ϩ ] i (29) and cannot be rescued by exogenous activation of PKC by PMA or DOG (this study); 3) that AVP-mediated AA release in A-10 cells is not affected by removal of calcium ions from the external media or by effective buffering of internal calcium concentration by BAPTA-AM (entry of extracellular calcium 4 J. J. Lehman and R. W. Gross, unpublished observations. After lysis of the cells, low speed pellets (LSP) obtained by centrifugation at 1,000 ϫ g, membrane (Memb), and cytosolic (Cyto) fractions (obtained by centrifugation at 100,000 ϫ g) were electrophoresed by SDS-PAGE (20 g of protein per lane), and immunoreactive bands corresponding to the indicated PKC isoforms (␣, ␦, ⑀, and ) were visualized by ECL Western analysis as described under "Experimental Procedures." B, A-10 cells were incubated in the presence of ethanol vehicle or 5 M rac-BEL, (R)-BEL, or (S)-BEL in DMEM for 15 min at 37°C. After removal of the media, the cells were incubated for 5 min in the presence or absence of 1 M AVP as indicated in the figure, washed with ice-cold PBS, and then lysed with RIPA buffer. Extracts from the cells were separated by SDS-PAGE (20 g protein per lane), transferred to a polyvinylidene difluoride membrane, and analyzed by ECL immunoblotting for phosphorylated MAPK as described under "Experimental Procedures." and increases in [Ca 2ϩ ] i are each thought to be necessary for cPLA 2 ␣ activation) (53)(54)(55); and 4) that many other enzymes thought to be necessary or associated with AA release are not inhibited by the concentrations of BEL employed. For example, enzymes that participate in signal transduction cascades that are known not to be inhibited by the concentrations of BEL employed include phosphatidylinositol-specific phospholipase C (10), phospholipase D (17), protein kinase A (56), and channels that mediate Ca 2ϩ release from intracellular stores (10). Of course, as with any other process, we cannot rule out the involvement of as yet undiscovered phospholipases or activation cascades. However, we state that in the presence of PKC activation, MAPK activation, and increases in [Ca 2ϩ ] i (processes typically deemed necessary for cPLA 2 ␣-mediated AA release), arachidonic acid release is still inhibited by (S)-BEL with a dose-response profile and chiral selectivity that closely correspond to iPLA 2 ␤. Clearly the next step is utilization of cell-specific inducible knockouts of iPLA 2 ␤ to clarify further its role in agonist-stimulated AA release in mammalian cells.