Arachidonic Acid Release from Aortic Smooth Muscle Cells Induced by [Ar$]Vasopressin Is Largely Mediated by Calcium-independent Phospholipase A2*

To identify the phospholipase mediating the majority of [Ar~lvasopressin (AVFWnduced release of arachidonic acid in A-10 smooth muscle cells, we exploited the specificity inherent in the mechanism-based inhibitor, (E)-6-(bromomethylene)tetrahydro-3-(l-naphthalenyl)- W-pyran-2-one (HELSS), which possesses a 1,000-fold selectivity for inhibition of calcium-independent ver-sus calcium-dependent phospholipases Az. Utilizing [’Hlarachidonic acid-labeled A-10 smooth muscle cells, one-half of AVP-inducible [‘Hlarachidonic acid release was inhibited by pretreatment with only 1 p~ HELSS and two-thirds of AVP-stimulated [‘Hlarachidonic acid release was inhibited by 5 p~ HELSS. The inhibition of [‘Hlarachidonic acid release by HELSS was saturable (ie. no additional inhibition of [’Hlarachidonic acid re- lease was present at 10 p~ HELSS), specific (ie. the activities of six intracellular enzymes, as well as the rate of glucose oxidation, were not altered by HELSS treat-ment), and nontoxic (i.e. HELSS-treated cells excluded trypan blue dye and did not leak intracellular enzymes into the medium). Collectively, these results demonstrate that HELSS blocks AVP-induced arachidonic acid release by specific and irreversible inhibition of calcium- independent phospholipase Az and underscore the importance of calcium-independent phospholipase Az in agonist-induced arachidonic acid release in at least some cell types.


To identify the phospholipase mediating the majority of [Ar~lvasopressin (AVFWnduced release of arachidonic acid in A-10 smooth muscle cells, we exploited the specificity inherent in the mechanism-based inhibitor, (E)-6-(bromomethylene)tetrahydro-3-(l-naphthalenyl)-W-pyran-2-one (HELSS), which possesses a 1,000-fold selectivity for inhibition of calcium-independent versus calcium-dependent phospholipases Az. Utilizing ['Hlarachidonic acid-labeled A-10 smooth muscle cells, one-half of AVP-inducible ['Hlarachidonic acid release was inhibited by pretreatment with only 1 p~ HELSS and two-thirds of AVP-stimulated ['Hlarachidonic acid release was inhibited by 5 p~ HELSS. The inhibition of ['Hlarachidonic acid release by HELSS was saturable
( i e . no additional inhibition of ['Hlarachidonic acid release was present at 10 p~ HELSS), specific ( i e . the activities of six intracellular enzymes, as well as the rate of glucose oxidation, were not altered by HELSS treatment), and nontoxic (i.e. HELSS-treated cells excluded trypan blue dye and did not leak intracellular enzymes into the medium). Collectively, these results demonstrate that HELSS blocks AVP-induced arachidonic acid release by specific and irreversible inhibition of calciumindependent phospholipase Az and underscore the importance of calcium-independent phospholipase Az in agonist-induced arachidonic acid release in at least some cell types.
The generation of eicosanoid metabolites in the vascular bed is coordinately regulated by the sequential actions of a multiplicity of enzymes which amplify and propagate the flow of biologic information (cf: Ref. 1). Since the rate-determining step in the generation of eicosanoid metabolites in most cell types is the release of arachidonic acid from their endogenous phospho-34839. The costs of publication of this article were defrayed in part by * This research was supported by National Institutes of Health Grant the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  lipid storage pools (see, e.g., Refs. 1 and 21, the identification and characterization of the intracellular phospholipases responsible for the release of arachidonic acid is a prominent issue in vascular biology. The detailed examination of the stoichiometry and specific activities of eicosanoid precursor pools and metabolites has collectively demonstrated the importance of phospholipase A2 as the major mediator of agonist-induced arachidonic acid release in most cell types (see, e.g., Refs. 2-4). During the last decade, several novel families of intracellular phospholipases Az have been identified and characterized (5)(6)(7)(8)(9). Although it was previously assumed that calcium was the predominant activator of intracellular phospholipase A 2 during cellular stimulation, recent studies have demonstrated that calcium is neither necessary nor sufficient for activation of several recently described intracellular phospholipases A 2 (see, e.g., Refs. [10][11][12][13]. Indeed, a novel class of calcium-independent phospholipases Az have been characterized which represent the predominant phospholipase A2 activity present in several agonistresponsive cell types including canine vascular smooth muscle (14)(15)(16)(17). Traditionally, the importance of individual enzymes, receptors, or regulatory proteins has been elucidated through the utilization of inhibitors that possess substantial specificity toward the polypeptides under consideration. However, prior attempts to identify the importance of specific types of intracellular phospholipases A2 have been confounded by the inability of traditional phospholipase inhibitors to distinguish among the different phospholipases Az present in mammalian cells. Thus, the identification of (E)-6-(bromomethylene)tetrahydro-3-(1-naphthalenyl)-2H-pyan-2-one (HELSS)' as a potent, irreversible, mechanism-based inhibitor that possessed over a 1000-fold selectivity for inhibition of calcium-independent uersus calcium-dependent phospholipases A2 (18) was met with enthusiasm, since it was an agent capable of elucidating the biologic role of calcium-independent phospholipases A2 in receptor-mediated arachidonic acid release. The rat thoracic aortic A-10 smooth muscle cell line expresses vasopressin receptors of the VI subtype whose stimulation results in the selective release of arachidonic acid from membrane phospholipids (19,20). Through specific mechanism-based inhibition of calciumindependent phospholipase A2 by HELSS, we now report that the majority of arachidonic acid release induced by vasopressin stimulation of A-10 cells is due to the activation of calciumindependent phospholipase A2.
Chromatography of Inositol Phosphates-Cells were labeled with 90 pCi of myo-[2-3Hlinositol in each 35-mm culture dish and incubated for 24 h. Inositol metabolites in control, stimulated, and inhibitor-treated cells were extracted and radiolabeled mono-, bis-, tris-, and tetrakisphosphates were separated by anion exchange chromatography and were quantified by scintillation spectrometry as described previously (21).
Quantification of Intracellular Calcium Concentration, Glucose Oxidation Rates, and Nonrelated Smooth Muscle Cell Enzymic Activities -A-10 cells were plated onto Cell Tak-coated glass coverslips and incubated overnight a t 37 "C. The attached cells were pretreated with either vehicle alone or with 10 PM HELSS for 10 min a t 37 "C. Intracellular calcium in Fura-2-loaded cells was quantified as described previously (21). Glucose oxidation in A-10 cells was assessed in cells grown for 3 days on Cytodex 3 beads (Pharmacia LKB Biotechnology Inc.) after treatment with vehicle alone or treatment with inhibitor (10 p~ HELSS) by trapping released [14ClC02 from l-[*4Clglucose onto filter paper and subsequent scintillation spectrometry as previously described (21)(22)(23). The integrity of intracellular enzymes in A-10 cells treated for 15 min with 10 PM HELSS was assessed after two freezethaw cycles by measuring alkaline phosphatase, creatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, and y-glutamyl transpeptidase employing previously described techniques (22, [25][26][27].

RESULTS AND DISCUSSION
The major measurable phospholipase A2 activity in smooth muscle A-10 cells grown in culture was calcium-independent (specific activities in the microsomal and cytosolic fractions were 250 and 120 pmollmgmin, respectively), selective for plasmalogen substrate, and exquisitely sensitive to inhibition by the mechanism-based inhibitor (E)-6-(bromomethylene)tetrahydro-3-(l-naphthalenyl)-2H-pyran-2-one (>95% inhibition at 5 p~ inhibitor). These characteristics are similar to those of the major phospholipase activity present in adult rat thoracic aortic smooth muscle which had been previously characterized (16).
Incubation of [3H]arachidonic acid-labeled A-10 cells with 1 p~ AVP for 5 min at 37 "C resulted in the release of -320,000 dpm of [3Hlarachidonic acid (equivalent to -4% of the [3Hlarachidonic acid incorporated into cellular phospholipids) into the culture medium (Fig. 2). The only metabolite released was arachidonic acid, without demonstrable amounts of lipoxygenase or cyclooxygenase products present (Fig. 1, lanes 2 and 3). Incubation of prelabeled rat aortic A-10 smooth muscle cells with 10 p~ HELSS for 15 min immediately prior to AVP stimulation resulted in the ablation of the majority of arachidonic acid release into the medium (Fig. 1, compare lunes 2 and 3  with lanes 4 and 5). One-half of [3H]arachidonic acid release into the medium was blocked by preincubation with only 1 PM HELSS, and two-thirds of agonist-induced f3H1arachidonic acid release was inhibited by 5 PM HELSS (Fig. 2). Importantly, pretreatment with 10 PM HELSS did not result in further inhibition of arachidonic acid release, demonstrating saturation of inhibition through HELSS-sensitive pathways at the 70% of release level. The remaining 30% of [3H]arachidonic acid release is likely mediated by other phospholipases including a 30-kDa dimeric calcium-responsive phospholipase A2 (6), a 85k.Da calcium-responsive phospholipase Az (7, 81, the low molecular weight phospholipases Az (281, andor phospholipases C (5, 29-32) and D (33). Measurements of phospholipase A2 activities from control and AVP-treated A-10 cells demonstrated that calcium-independent phospholipase A2 activity in the microsomal fraction increased modestly after AVP stimulation (118% of control value), which was accompanied by a parallel decrease in calcium-independent phospholipase A2 activity in the cytosolic fraction (84% of control activity). As anticipated, treatment ofA-10 cells with HELSS (10 PM) inhibited over 95% of calcium-independent phospholipase A2 activity in both cytosolic and microsomal subcellular fractions in both control and AVP-stimulated cells. Calcium-dependent phospholipase A2 activity represented <lo% of total measurable phospholipase A2 activity employing either plasmenylcholine or phosphatidylcholine substrate, was predominantly present in the microsomal fraction in both control and AVP-treated cells, decreased modestly after AVP-treatment, and was not inhibited by HELSS. Collectively, these results demonstrate that diminutive concentrations of HELSS are capable of inhibiting the majority of agonist-induced arachidonic acid release in a concentration-dependent and saturable manner and identify calcium-independent phospholipase A2 as the enzymic mediator responsible for the majority (-70%) of [3H]arachidonic acid release in this system.
To confirm that the reduction in arachidonic acid release was due to specific inhibition of calcium-independent phospholipase A2 and not the result of unforeseen effects on other enzymes or overall cellular viability, a multiplicity of independent criteria were evaluated. First, cellular ultrastructure was unchanged in control or HELSS-pretreated A-10 cells. Second, cellular permeability was unaltered after HELSS treatment as assessed by trypan blue exclusion. Similarly, HELSS treatment did not result in the release of lactate dehydrogenase into the medium, underscoring the sustained functional integrity of the A-10 cell plasma membrane after exposure to HELSS. Third, the activities of multiple enzymes (unrelated to phospholipase A2) present in A-10 cells (including alkaline phosphatase, creatine kinase, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, and y-glutamyl transpeptidase) were unaltered by HELSS treatment. Fourth, the rates of glucose oxidation in control and HELSS-pretreated cells were nearly identical, demonstrating the metabolic integrity of a multiplicity of enzymes participating in the sequential oxidation of glucose. Fifth, control and HELSS-treated cells manifested similar amounts of agonist-induced phospholipase C-mediated hydrolysis of inositol phospholipids (i.e. no significant differences in the amounts of [3H]inositol phosphate or [3Hlinositol trisphosphate released after AVP stimulation of [3HIinositol-prelabeled cells). Collectively, these results demonstrate the morphologic, ultrastructural, functional and metabolic integrity ofA-10 cells following inhibitor treatment and underscore the selectivity of the mechanism-based inhibitor HELSS. Since alterations in intracellular calcium concentration constitute an important element in the initiation and propagation of cellular activation which has been temporally and chemically related to the activation of at least some intracellular phospholipases, additional experiments were performed to determine the effects of HELSS treatment on the alterations in cytosolic Ca2+ induced by AVP stimulation ofA-10 cells. First, treatment of A-10 smooth muscle cells with 1 PM AVP in calcium-free buffer resulted in a transient spike in calcium ion (as ascertained by the 3401380 nm fluorescence ratio of Fura-2) and the intensity and duration of the spike was unchanged after treatment with HELSS (Fig. 3, panels A and B ) . Thus, the intracellular machinery involved in the release and reuptake of intracellular Ca2+ was not altered by HELSS treatment. Furthermore, treatment of smooth muscle A-10 cells with AVP in buffer containing 2.5 m M Ca2+ resulted in a calcium transient that was similar to that present in the absence of extracellular Ca2+ ion but that was substantially prolonged due to calcium influx from the extracellular medium (Fig. 3, panel C).

Mediation of AVP-induced Arachidonic Acid Release
In prior studies Thibonnier (34) demonstrated that treatment of A-10 smooth muscle cells with arachidonic acid prior to AVP stimulation shortens the time necessary for Ca2+ levels to return to base line. Accordingly, we anticipated, and found, that inhibitor-treated cells demonstrated an extended prolongation of the elevation in intracellular calcium ion after exposure to agonist in comparison to control cells (compare panels C and D in Fig. 3; see Fig. 4). Since HELSS treatment attenuates the AVP-induced release of arachidonic acid, these results clearly show the presence of one functional sequela ( i e . a change in temporal longevity of the calcium signal) resulting from HELSS-mediated attenuation of arachidonic acid release and underscore the functional importance of calcium-independent phospholipase A2 in regulating the temporal duration of the calcium signal in agonist-stimulated A-10 cells.
Collectively, these results demonstrate the importance of calcium-independent phospholipase A2 in mediating the majority of agonist-induced arachidonic acid release in vascular smooth muscle A-10 cells and identify one functional sequela of the inhibition of arachidonic acid release after receptor stimulation. Since arachidonic acid accelerates phosphorylation of the myosin light chain and attenuates the rate of its dephosphorylation (351, it seems likely that calcium-independent phospholipase A2 is an important modulator of the increase in contractile force of smooth muscle after AVP stimulation. The bio-