The Ca2+-sensing Receptor Activates Cytosolic Phospholipase A2 via a Gqα-dependent ERK-independent Pathway*

The Ca2+-sensing receptor (CaR) stimulates a number of phospholipase activities, but the specific phospholipases and the mechanisms by which the CaR activates them are not defined. We investigated regulation of phospholipase A2(PLA2) by the Ca2+-sensing receptor (CaR) in human embryonic kidney 293 cells that express either the wild-type receptor or a nonfunctional mutant (R796W) CaR. The PLA2activity was attributable to cytosolic PLA2(cPLA2) based on its inhibition by arachidonyl trifluoromethyl ketone, lack of inhibition by bromoenol lactone, and enhancement of the CaR-stimulated phospholipase activity by coexpression of a cDNA encoding the 85-kDa human cPLA2. No CaR-stimulated cPLA2 activity was found in the cells that expressed the mutant CaR. Pertussis toxin treatment had a minimal effect on CaR-stimulated arachidonic acid release and the CaR-stimulated rise in intracellular Ca2+(Ca2+ i ), whereas inhibition of phospholipase C (PLC) with U73122 completely inhibited CaR-stimulated PLC and cPLA2 activities. CaR-stimulated PLC activity was inhibited by expression of RGS4, an RGS (Regulator of Gprotein Signaling) protein that inhibits Gαqactivity. CaR-stimulated cPLA2 activity was inhibited 80% by chelation of extracellular Ca2+ and depletion of intracellular Ca2+ with EGTA and inhibited 90% by treatment with W7, a calmodulin inhibitor, or with KN-93, an inhibitor of Ca2+, calmodulin-dependent protein kinases. Chemical inhibitors of the ERK activator, MEK, and a dominant negative MEK, MEKK97R, had no effect on CaR-stimulated cPLA2 activity but inhibited CaR-stimulated ERK activity. These results demonstrate that the CaR activates cPLA2 via a Gαq, PLC, Ca2+-CaM, and calmodulin-dependent protein kinase-dependent pathway that is independent the ERK pathway.

The extracellular Ca 2ϩ -sensing receptor (CaR) 1 is a G pro-tein-coupled receptor that is expressed in the parathyroid and kidney and senses extracellular Ca 2ϩ in the millimolar range. This receptor acts through at least two G proteins, G␣ i and G␣ q , to regulate multiple intracellular enzymes that control production of second messengers including cAMP, inositol trisphosphate (IP 3 ), diacylglycerol (DAG), intracellular Ca 2ϩ (Ca 2ϩ i ), and arachidonic acid (AA) metabolites (1). In the parathyroid, the CaR inhibits parathyroid hormone production and secretion in response to elevated extracellular Ca 2ϩ (Ca 2ϩ o ) levels, and in the kidney, activation of the CaR inhibits NaK2Cl cotransporter activity, Ca 2ϩ reabsorption, and the action of vasopressin leading to a Na ϩ , Cl Ϫ , Ca 2ϩ , and H 2 O diuresis. CaR-stimulated production of AA metabolites, possibly products of 12-and 15-lipoxygenase, contributes to inhibition of parathyroid hormone secretion in parathyroid cells (2)(3)(4). In cells from the thick ascending limb of Henle, activation of phospholipase A 2 (PLA 2 ) by the CaR results in the production of 20-hydroxyeicosatetraenoic acid, a cytochrome P450 metabolite, that inhibits the apical 70-picosiemens potassium channel activity that would reduce NaK2Cl transporter activity (5). In these tissues and others including the brain, pancreas, stomach, colon, and skin, the CaR may also sense Ca 2ϩ extrusion by adjacent cells and function in cell-cell communication (6 -8).
Phospholipase A 2 , the rate-limiting enzyme in AA metabolism, hydrolyzes cellular phospholipids to form lysophospholipids that lead to the production of platelet-activating factor and liberation of polyunsaturated fatty acids including AA that are the precursors for prostaglandins, thromboxanes, leukotrienes, and a variety of other metabolites (eicosanoids) (1,9). Eight different groups of PLA 2 enzymes have been described. The majority of these enzymes are extracellular and not hormonally regulated, whereas two groups, group IV and group VI, are intracellular and are subject to regulation by extracellular signals. The hormone-regulated enzymes are the 85-kDa Ca 2ϩsensitive cytosolic PLA 2 (cPLA 2 , a group IV enzyme) and the 80 -88-kDa Ca 2ϩ -insensitive PLA 2 (iPLA 2 , a group VI enzyme). Both enzymes are subject to activation by G protein-dependent signaling systems (10,11). cPLA 2 is expressed in most tissues and is activated by many G protein-coupled receptors including those for angiotensin II, ATP, bradykinin, endothelin, lysophosphatidic acid, and thrombin (1). The AA products of cPLA 2 appear to be involved primarily in signaling functions. cPLA 2 activity is inhibited by the substrate analogue arachidonyl trifluoromethyl ketone (AACOCF 3 ) but not bromoenol lactone (BEL) (12). The precise mechanism of activation of cPLA 2 is variable and depends on the receptor. Activation of cPLA 2

requires a rise in Ca 2ϩ
i which leads to translocation of the enzyme from the cytosol to the plasma membrane. In different cell types, pathways that involve pertussis toxin-sensitive or -insensitive G proteins, phospholipase C (PLC), protein kinase C (PKC), MAP kinases (extracellular signal-regulated kinases, ERK), calmodulin (CaM), and calmodulin-dependent protein kinase (CaMK) have been described (13). cPLA 2 is a substrate for ERKs in many cell types, and phosphorylation of cPLA 2 by ERK appears to be required for activation of the enzyme (14).
iPLA 2 also appears to be ubiquitously expressed and is activated by receptors for ATP, ␣ 2 agonists, vasopressin, and Fas (15)(16)(17). The proposed functions of iPLA 2 include cellular lipid remodeling, generation of substrates for leukotriene biosynthesis, and generation of second messengers that regulate ion channel activity (11,18). iPLA 2 activity is inhibited by both AACOCF 3 and BEL (12). iPLA 2 is activated by Ca 2ϩ store depletion and PKC and is inactivated by association with calmodulin (16).
The form of PLA 2 that is activated by the CaR and the mechanism by which the CaR stimulates PLA 2 activity have not been defined in any cell type. To determine which PLA 2 , cPLA 2 or iPLA 2 , is activated by the CaR and which G protein, second messenger, and kinase pathway(s) are involved, we expressed the CaR in HEK-293 cells and defined the early components of the signaling pathway that lead to activation of PLA 2 . We find that the CaR activates cPLA 2 via a G␣ q , PLC, Ca 2ϩ i , CaM, and CaMK-dependent but ERK-independent signaling pathway.
Sources and Construction of cDNAs-The cDNAs encoding human wild-type and nonfunctional mutant CaR (R796W) in pcDNA3 were generous gifts from Drs. E. M. Brown, S. C. Hebert, and M. Bai in the Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, and Vanderbilt University School of Medicine. To add the HA epitope tag to the C termini of CaR, we synthesized a 57-bp oligonucleotide coding for the last 6 amino acids of CaR followed by 9 amino acids of the HA epitope, a stop codon, an XbaI restriction site, and a 3-bp tail (5Ј CGC TCT AGA CTA AGC GTA GTC TGG GAC GTC GTA TGG GTA TGA ATT CAC TAC GTT TTC 3Ј). The cDNA was amplified by polymerase chain reaction using an oligonucleotide that is 5Ј to a unique BamHI site in the C-terminal region of receptor (5Ј ACC TTT ACC TGT CCC CTG AA 3Ј) and the 57-bp oligonucleotide coding for C terminus of receptor and the HA epitope. The polymerase chain reaction products were digested with BamHI and XbaI, purified, and ligated into the expression vector (pcDNA3) that contained the remainder of the cDNA for the CaR. The human cPLA 2 cDNA was obtained from Dr. Harry Nick, University of Florida, and subcloned into pcDNA3 (19). MEK K97R and the HA-tagged ERK K53R were generous gifts from Dr. Melanie Cobb, University of Texas, Southwestern Medical School. The HEK-293 cells that stably express RGS4 were described previously (20).
Cell Culture and Expression of cDNAs-HEK-293 cells were obtained from the American Type Culture Collection and cultured in DMEM supplemented with 10% calf serum, and 25 mM Hepes (pH 7.4). The cDNAs, pcDNA3, CaR-HApcDNA3, or CaR(R796W)-HApcDNA3 and cPLA 2 -pcDNA3 were introduced into cells using the calcium phosphate precipitation method. Studies using transiently transfected cells were performed 48 h after transfection. In these transfections, the total amount of DNA transfected was held constant by addition of pBluescript so that the total amount used was 3 g. For stable expression of cDNAs, the cells from each individual well were plated in 100-or 150-mm dishes or 96-well plates and cultured in medium containing 500 g/ml G418 24 h after transfection. G418-resistant clones were isolated after 3-4 weeks and screened by immunoblotting for the expressed protein. Cells were used for experiments between passages 5 and 15. Similar results were obtained with multiple clones.
Antibodies and Immunoblotting-To verify expression of proteins, membranes were immunoblotted with the anti-HA antibody 12CA5 (CaR), the anti-Myc antibody 9E10 (Myc-RGS4) (both antisera were from the UF hybridoma core), and anti-human cPLA 2 (Santa Cruz BioTechnology, Santa Cruz, CA), and to measure ERK activity, antiphospho-ERK (Promega, Madison, WI) and anti-ERK (Promega). HEK-293 cell membranes were prepared by homogenization in a buffer containing 10 mM Tris-HCl (pH 7.8), 1 mM EDTA, and protease inhibitors and brought to a final concentration of 30 mM NaCl and 2 mM MgCl 2 and centrifuged at 250 ϫ g for 2 min. The supernatant was centrifuged at 23,000 ϫ g for 10 min; the pellet was resuspended in the same buffer, and the protein concentration was measured. Samples containing equal amounts of protein were subjected to SDS-PAGE and processed for immunoblotting. The immunoreaction signals were detected by enhanced chemiluminescence.
Fluorescent Measurement of Ca 2ϩ i -Cells were released from dishes with phosphate-buffered saline containing 0.5 mM EDTA and rinsed twice with calcium measurement solution (CaMS) containing 140 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl 2 , 10 mM Hepes (pH 7.4), 1 mg/ml bovine serum albumin (BSA), and 2 mg/ml glucose. The cells were resuspended and incubated in 2 ml of uptake medium containing 2.5 M Fura-2-AM in CaMS for 30 min at 37°C, washed twice with CaMS, and diluted to a concentration of ϳ10 6 cells in 2 ml of CaMS. Ca 2ϩ i was measured fluorometrically in Fura-2-loaded cells in suspension by a dual excitation microfluorometer (SLM Fluoromax) at 37°C with constant stirring. Excitation was at 340 and 380 nm, and emission was at 510 nm. Experiments were started when a steady fluorescent base line was obtained. Each tracing was calibrated by lysing the cells with digitonin (75 g/ml) in the presence of 2 mM Ca 2ϩ o to determine F max , and then 20 mM EDTA (pH 8.6) was added to obtain F min . The Ca 2ϩ i concentration was calculated from the expression [ Measurement of Arachidonic Acid Release-The HEK-293 cells were grown in collagen-coated 24-well plates and were prelabeled with 0.1 Ci/well [ 3 H]AA in serum-free DMEM at 37°C for 6 h. After removal of the labeling medium, the cells were rinsed with 1.5 ml of buffer A containing 130 mM NaCl, 5.4 mM KCl, 0.8 mM MgCl 2 , 0.5 mM CaCl 2 , 10 mM glucose, and 10 mM Hepes (pH 7.4) and incubated with buffer A containing 2 mg/ml BSA for 30 min in the presence or absence of inhibitor. In the PKC down-regulation experiments, cells were incubated with or without 100 nM PMA for 20 -24 h and prelabeled with [ 3 H]AA for the 6 h preceding the experiment. For pertussis toxin treatment, cells were pretreated with or without 100 ng/ml pertussis toxin for 12-15 h and labeled with [ 3 H]AA during the last 6 h. After equilibration, cells were incubated at 37°C for 15 min with 0.5 ml of the same buffer in the presence or absence of 5 mM CaCl 2 or 0.5 mM neomycin. LPS (lipopolysaccharide)-stimulated [ 3 H]AA release was measured in RAW 264.7 cells in a similar manner except that the cells were grown and labeled in DMEM (high glucose) with 10% FCS in 24-well trays without collagen coating. The BEL (10 M) was added 15 min before the LPS in buffer A without BSA, and the cells were exposed to LPS (5 g/ml) for 1 h with BSA present (22). Equal volumes of medium were removed from the wells and centrifuged at 13,000 rpm for 5 min. The radioactivity in the cell-free medium was quantitated by liquid scintillation spectrometry and normalized for total radioactivity incorporated into cellular phospholipids that were solubilized with 0.2 N NaOH, 0.2% SDS. [ 3 H]AA release was linear for 60 min.
Measurement of Inositol Trisphosphate Formation-Cells at 65% confluence in 12-well plates were prelabeled with 5-10 Ci/ml myo-[ 3 H] inositol in 0.5 ml of inositol-free DMEM containing 10% calf serum for 48 -50 h. For pertussis toxin treatment, cells were incubated with pertussis toxin 100 ng/ml or vehicle for the last 12-15 h of the prelabeling period. Before experiments, cells were equilibrated for 30 min in inositol-free DMEM containing 20 mM Hepes (pH 7.4) and 20 mM LiCl with or without 10 M U73122. Cells were then incubated at 37°C for 5 min in 0.4 ml of equilibration medium with or without 5 mM CaCl 2 . The reactions were terminated by adding 0.4 ml of 10% perchloric acid to each well, and the plates were stored at 4°C for 3-5 h. The samples were transferred to microcentrifuge tubes, neutralized by the addition of 0.32 ml of a solution containing 2 M KOH, 1 mM EDTA, and 1 M Hepes (pH 7.4), and centrifuged at 12,000 ϫ g for 5 min. The supernatants were applied to AG1-X8 anion exchange columns, and tritiated inositolcontaining compounds were separated as described (20).
Statistical Analysis-Concentration-response relationships (EC 50 and Hill coefficient values) were determined using GraphPad prism software, and significance was calculated using the Instat two-tailed, unpaired t test statistics program.

Expression of CaR in Mammalian
Cells-To study regulation of PLA 2 by the CaR, we expressed an HA-tagged wild-type and an HA-tagged nonfunctional mutant receptor (R796W) in HEK-293 cells (23). The expressed proteins were detected by immunoblotting with the monoclonal anti-HA antibody (12CA5) (Fig. 1). Blots with both the transiently and stably transfected cells revealed bands of the appropriate molecular mass, ϳ125 and 140 kDa, and demonstrated similar levels of CaR protein expression. The two bands of the CaR result from differential glycosylation (24,25). Bands were not present in membranes from the G418-resistant cells.
Activation of cPLA 2 by the CaR-Extracellular Ca 2ϩ and neomycin, both ligands for the CaR, stimulated [ 3 H]AA release from cells that express the CaR in a dose-dependent manner but not from cells that express the nonfunctional mutant CaR R796W (Fig. 2). The EC 50 for Ca 2ϩ was 4.2 mM, and the EC 50 for neomycin was 0.15 mM. To determine which form of PLA 2 , cPLA 2 or iPLA 2 , is activated by the CaR, we treated cells with AACOCF 3 , an inhibitor of both enzymes, and BEL, a specific inhibitor of iPLA 2 (1,12). The CaR was activated with neomycin to avoid a possible increase in Ca 2ϩ entry as a result in increased extracellular calcium. As shown in Fig. 3A, AACOCF 3 inhibited the CaR-stimulated [ 3 H]AA release (IC 50 25 M), but BEL had no effect up to a concentration of 10 M, a concentration that inhibits iPLA 2 in other systems (26). In parallel experiments, 10 M BEL inhibited LPS-stimulated [ 3 H]AA release in RAW 264.7 cells, an iPLA 2 -dependent response, by 92% (26). To confirm that the CaR activates cPLA 2 , we transiently coexpressed cDNAs coding for the CaR and cPLA 2 , and we measured [ 3 H]AA release (19). Fig. 3B shows that coexpression of increasing amounts of the cPLA 2 cDNA with the CaR resulted in increased CaR-stimulated [ 3 H]AA release without an increase in basal [ 3 H]AA release. Inhibitor studies and increasing CaR-stimulated PLA 2 activity with increasing levels of expression of cPLA 2 demonstrate that the CaR activates cPLA 2 .
Role of PLC and G Proteins-We tested the role of PLC-␤ in activation of cPLA 2 by first demonstrating inhibition of CaRstimulated [ 3 H]AA release by U73122, an inhibitor of PLC-␤. Pretreatment of cells that express the CaR with U73122 eliminated the CaR-stimulated cPLA 2 activity demonstrating that the CaR activates cPLA 2 via PLC-␤ (Fig. 4A). To confirm that U73122 inhibits PLC-␤ activity, we measured CaR-stimulated IP 3 production with and without U73122 (Fig. 4B), and we found that it was completely inhibited by U73122. PLC-␤ can be activated by pertussis toxin-sensitive (G␣ i -dependent) or pertussis toxin-insensitive (presumably G␣ q -dependent) sig-naling systems. Treatment of the cells with pertussis toxin had a minimal inhibitory effect on [ 3 H]AA release (Fig. 4A), and IP 3 production (Fig. 4B) indicating that G proteins of the ␣ i family have a minimal role in stimulation of cPLA 2 and PLC-␤ by the CaR. To test specifically for a role for a G␣ q family member in the activation of PLC-␤ and consequently cPLA 2 , we transiently expressed the CaR in HEK-293 cells that stably express RGS4, an RGS protein that interacts preferentially with members of the G␣ q family, accelerates their GTPase activity, and reduces their activation (20,27). Fig. 4C shows that RGS4 eliminated CaR-stimulated PLC-␤ activity. These results indicate that the CaR acts through a pertussis toxin-insensitive, G␣ q -dependent pathway to activate PLC, the products of which stimulate cPLA 2 .
Role of Ca 2ϩ i -PLC-␤ produces both DAG, which activates PKC, and IP 3 , which raises Ca 2ϩ i , by releasing Ca 2ϩ from intracellular stores. A rise in Ca 2ϩ i could activate cPLA 2 by mechanisms involving calmodulin, Ca 2ϩ , calmodulin-dependent protein kinase, Ca 2ϩ -dependent tyrosine kinases, PKC in cooperation with DAG, or the ERK pathway. To document and characterize the CaR-stimulated rise in Ca 2ϩ i in the cells that express the CaR, we measured Ca 2ϩ i in cells that express the wild-type and mutant CaR (CaR R796W ) using Fura-2 fluorescence (Fig. 5). In the cells that express the wild-type CaR, activation of the CaR with 4 mM extracellular Ca 2ϩ (Ca 2ϩ o ) leads to a rapid rise in Ca 2ϩ i followed by a slow fall to a plateau level above base line. This type of Ca 2ϩ signal has two components, release of Ca 2ϩ from intracellular stores and Ca 2ϩ entry across the plasma membrane. Subsequent stimulation of purinergic receptors in these cells with ATP (100 M) also leads to a rapid rise in Ca 2ϩ i and a plateau phase. Cells that express the CaR R796W do not respond to 4 mM extracellular Ca 2ϩ but do respond to ATP. Fig. 5B shows that the effect of pertussis toxin on the CaR-stimulated Ca 2ϩ i signal, like its effect on cPLA 2 and PLC activity, is minimal. Incubation of the cells with EGTA to chelate Ca 2ϩ o and deplete Ca 2ϩ i (Fig. 5C) reduced CaR-stimulated [ 3 H]AA release by ϳ80% when the CaR was activated with neomycin, indicating that Ca 2ϩ o and a rise in Ca 2ϩ i are required for activation of cPLA 2 by the CaR (28). In studies similar to those shown in Fig. 5, A and B, preincubation of cells with 2 mM EGTA for 20 min in nominally Ca 2ϩ -free medium prevented bradykinin and ATP-stimulated rises in

Ca 2ϩ
i (data not shown). Role of ERK-Many receptors act through the ERK pathway to activate cPLA 2 (14). We assessed the role of ERK in the activation of cPLA 2 using chemical inhibitors of MEK, the activator of the ERKs, and a dominant negative form of MEK (MEK K97R ). Preincubation of the cells with either PD98059 or U0126 had no effect on the ability of the CaR to activate cPLA 2 (Fig. 6A) but inhibited activation of the ERKs by the CaR (Fig.  6B). Treatment of the cells with the inactive U0126 analogue, U0124, did not affect cPLA 2 or ERK activation. Similarly, coexpression of the CaR with the dominant negative MEK, MEK K97R , and HA-tagged ERK-1 did not inhibit activation of cPLA 2 (Fig. 6C) but resulted in inhibition of ERK activation by the CaR (Fig. 6D). These results indicate that activation of the ERK pathway and presumably phosphorylation of cPLA 2 by p42/44 ERK is not required for activation of cPLA 2 by the CaR.
Role of PKC-The PLC product, DAG, activates PKC in a Ca 2ϩ -dependent manner. To test for a role of the conventional forms of PKC (those that are both DAG-and Ca 2ϩ -dependent), we treated cells with calphostin, a PKC inhibitor, and downregulated PKC by overnight treatment of the cells with 100 nM PMA and then measured CaR-stimulated cPLA 2 activity. As shown in Fig. 7, both calphostin and down-regulation of PKC with PMA pretreatment reduced CaR-stimulated cPLA 2 activity by ϳ50% but did not eliminate it. In control experiments, 15 min of exposure of the HEK-293 cells that express the wild-type and mutant forms of the CaR to 100 nM PMA stimulated [ 3 H]AA release to a similar degree as treatment of the cells that express the wild-type CaR with 5.0 mM Ca 2ϩ . However, pretreatment of the cells overnight with 100 nM PMA prevented significant stimulation of [ 3 H]AA release by 15 min of exposure to 100 nM PMA, demonstrating that our protocol for downregulation of PMA inhibits activation of the conventional forms of PKC. These results indicate that the conventional PKC isoforms (DAG-and Ca 2ϩ -dependent) contribute to activation of cPLA 2 but that PKC-dependent mechanisms cannot account for the majority of its activation by the CaR.
Role of CaM and CaMK-Another mechanism by which a rise in Ca 2ϩ i could activate cPLA 2 is via calmodulin (13). To Where indicated, cells were treated with 100 ng/ml pertussis toxin for the 12-15 h before experiments or U73122 for the 30 min before experiments. Cells were treated with 0.5 mM Ca 2ϩ or 5.0 mM Ca 2ϩ for 5 min, and [ 3 H]IP 3 production was measured by column chromatography. C, effect of RGS4 on CaR-stimulated PLC activity. HEK-293 cells that stably express RGS4 or G418resistant control cells (V) were transiently transfected with the CaR (wild-type). Twelve hours after transfection, they were prelabeled with [ 3 H]inositol for 48 h and exposed to 0.5 or 5.0 mM Ca 2ϩ for 5 min, and [ 3 H]IP 3 production was separated by column chromatography. Each bar represents the mean of triplicate samples Ϯ S.D., and this figure is representative of three experiments. test for a role for calmodulin in CaR-dependent activation of cPLA 2 , we pretreated cells with W7, a calmodulin antagonist that competitively inhibits interaction of Ca 2ϩ -calmodulin with its target proteins. Pretreatment of the cells with W-7 for 45 min before activation of the CaR eliminated ϳ90% of the CaRstimulated cPLA 2 activity (IC 50 7 M), indicating that activation of calmodulin is an essential step in activation of cPLA 2 by the CaR (Fig. 8A). The effects of calmodulin could be mediated by Ca 2ϩ , calmodulin-dependent protein kinase (CaMK), or other proteins. To test for a role for CaMK, we pretreated cells with KN-93, a specific inhibitor of the CaMK enzymes at the concentrations used (29,30). We found that over a concentration range up to 25 M, KN-93 inhibited ϳ90% of the CaRstimulated cPLA 2 activity with an IC 50 of ϳ6.3 M (Fig. 8B). These results demonstrate that in contrast to other G protein-coupled receptors, the CaR stimulates cPLA 2 via a CaM/CaMKdependent pathway that is independent of the ERK pathway.

DISCUSSION
Although different G protein-coupled receptors may act by similar mechanisms such as activation of PLC, raising Ca 2ϩ i , and activation of PKC, each receptor has unique properties that include its location in the cell, its cycling and processing by the cell, and the set of proteins with which it interacts. These unique characteristics allow it to have discrete functions such as activation of specific effector enzymes. For example, in MDCK cells, ␣ 1 -adrenergic and P 2U receptors activate protein kinase C and MAP kinases but activate cPLA 2 by different mechanisms (31,32). Similarly, although the CaR activates PLC, increases Ca 2ϩ i , and activates PKC and MAP kinases, it activates cPLA 2 by mechanisms that are particular to the CaR. To characterize the mechanism by which the CaR activates PLA 2 and determine which type of PLA 2 was activated by it, we studied activation of PLA 2 in HEK-293 cells that stably express the CaR or a nonfunctional mutant form of the receptor, CaR R796W, to control for nonspecific effects of the receptor ligands. We determined that the CaR activated cPLA 2 via a novel pathway that involves G␣ q , PLC, Ca 2ϩ , calmodulin, and CaMK. In contrast to many other receptors, activation of the ERKs was not required for CaR-dependent activation of cPLA 2 .
Both cPLA 2 and iPLA 2 could be activated by G protein-dependent signaling systems, and both enzymes could be activated simultaneously (33). To identify the type of PLA 2 that is activated by the CaR, we tested the ability of AACOCF 3 , an arachidonic acid analogue that inhibits both cPLA 2 and iPLA 2 , and BEL, a "suicide substrate" that is specific for iPLA 2 to inhibit CaR-stimulated [ 3 H]AA release from our cells (1). We found that AACOCF 3 inhibited all of the CaR-stimulated activity with an IC 50 of 25 M, which is comparable to its IC 50 in other systems (12). BEL, used at a concentration that was considerably higher than that required to inhibit completely iPLA 2 in other cell types (10 M), had no effect on CaR-stimulated [ 3 H]AA release and inhibited LPS-stimulated [ 3 H]AA release from RAW 264.7 cells (12,34). Additionally, we expressed human cPLA 2 with the CaR and demonstrated that CaR-stimulated [ 3 H]AA release increased in parallel with increasing amounts of expressed cPLA 2 . These results indicate that in our experimental system, CaR-stimulated [ 3 H]AA release is a result of increased cPLA 2 activity. Activation of cPLA 2 by the CaR requires stimulation of PLC activity which could occur through members of the G␣ i and/or G␣ q families. Pretreatment of cells with pertussis toxin had a minimal effect on the ability of the CaR to stimulate PLC and cPLA 2 , indicating that the CaR does not activate cPLA 2 through members of the G␣ i family. Consequently, the CaR must act through a pertussis toxin-insensitive pathway probably involving members of the G␣ q family, but possibly through another mechanism. Chemical inhibitors of the G␣ q family do not exist, so we used a protein that inhibits G␣ q activity, RGS4, to demonstrate specifically that the CaR activates PLC, and consequently cPLA 2 , through G␣ q family proteins. RGS4 inhibits activation of G␣ i and G␣ q family proteins by stimulating their GTPase activity but acts preferentially on the G␣ q family (20,27). The fact that CaR-stimulated PLC activity is almost completely inhibited by expression of the CaR in cells that stably express RGS4 demonstrates that the CaR must act through G␣ q family members to activate PLC and consequently cPLA 2 .
A rise in Ca 2ϩ i is required for activation of cPLA 2 by most receptor-mediated mechanisms. Activation of the CaR results in a significant rise in Ca 2ϩ i via release from intracellular stores and influx across the cell membrane. We attempted to buffer intracellular Ca 2ϩ with BAPTA but surprisingly found that BAPTA treatment resulted in a slight increase in [ 3 H]AA release. This result can be rationalized by the recent finding that BAPTA may lower Ca 2ϩ i to the point that Ca 2ϩ influx is stimulated (35,36). However, by adding EGTA to the medium for 30 min, we were able to inhibit CaR-stimulated cPLA 2 activity by ϳ80%, demonstrating dependence on Ca 2ϩ o . Treatment of the cells for this period with 5-10 mM EGTA presumably also reduced, but did not completely deplete, the Ca 2ϩ in the intracellular stores so that the CaR-stimulated rise in Ca 2ϩ due to release of the intracellular stores was reduced. Conse-quently, we cannot determine whether the 20% of CaR-stimulated cPLA 2 activity that remained was due to incomplete inhibition of the Ca 2ϩ i signal or a Ca 2ϩ -independent mechanism.
In many cell types, cPLA 2 is activated by a p42/44 ERK-dependent pathway. Phosphorylation of cPLA 2 by ERK in vitro at Ser-505 results in a change in its mobility in gels (retardation) and increased activity, whereas phosphorylation by PKA or PKC does not result in a mobility shift or increased activity. Similarly, in studies of cPLA 2 from intact cells, p42/44 ERK reduces the mobility of cPLA 2 in gels and increases its activity. Mutation of Ser-505 results in an enzyme that is not activated by and that does not undergo a mobility shift in response to ERK, PMA, ATP, or thrombin (14). The PKC (phorbol ester)induced activation of cPLA 2 and CaMK-stimulated activation of cPLA 2 appear to occur via activation of the ERKs by PKC or CaMK (13,14,31,32,37).
However, some receptors may activate cPLA 2 by mechanisms that are independent of, or only partially dependent on, the ERK pathway (32,38,39). P 2U receptors in MDCK-D 1 cells activate cPLA 2 by PKC-and ERK-dependent pathways (32). In MDCK cells, bradykinin stimulates cPLA 2 activity by a mechanism that is independent of both PKC␣ and ERK but that is dependent on tyrosine phosphorylation (39). In renal proximal tubule cells and breast carcinoma cells, ERK can be activated by PLA 2 -dependent AA metabolites rather than ERK-activating PLA 2 , reversing the conventional relationship (40,41).
Our results indicate that the CaR activates cPLA 2 by a pathway that is independent of ERK activation. Inhibition of CaR stimulated ERK activation by three methods; the two D, in parallel plates, cells were stimulated with Ca 2ϩ (5.0 mM) for 5 min, and ERK activity was measured in cell extracts by a gel shift assay with the monoclonal antibody 12CA5 that recognizes the HA epitope tag on the expressed ERK-1. Activation was quantitated by densitometry, and calculation of the ratio of the shifted band to the total ERK was expressed. A representative blot is shown below D. chemical inhibitors of the ERK activator MEK (PD-98059 and U0126) and expression of a dominant negative form of MEK, MEK K97R , inhibited CaR-stimulated ERK activation but had no effect on CaR-stimulated cPLA 2 activation. Stimulation of cPLA 2 by the CaR also appears to be partially dependent on PKC (ϳ50% inhibition with calphostin, a general PKC inhibitor or down-regulation of PKC with PMA pretreatment).
Activation of cPLA 2 by the CaR is inhibited to a similar degree by W-7, a competitive calmodulin inhibitor, and KN-93, a CaMK inhibitor, implicating CaM and one of the CaMK isoforms in the regulatory pathway. The most likely scenario is that a receptor-stimulated rise in Ca 2ϩ i activates CaM which activates CaMK, and that CaMK activates cPLA 2 by a direct or indirect mechanism. CaM has many potential targets, including kinases (CaMK, myosin light chain kinase, and phosphorylase kinase), phosphatases (calcineurin), cytoskeletal proteins (MAP-2, Tau, fodrin, and neuromodulin), cyclic nucleotide phosphodiesterases, adenylyl cyclases, and Ca 2ϩ transporters (pumps and channels), any or all of which could contribute to cPLA 2 activation (29). However, the CaMK inhibitor KN-93 inhibited 90% of CaR-stimulated cPLA 2 activity indicating a significant role for CaMK.
The extent to which KN-93 inhibited CaR-stimulated cPLA 2 activity in our studies is comparable to that found by others (13,30,42) studying CaMK-dependent processes in whole cell systems. In our studies with 3 h of exposure to KN-93, we found an IC 50 of ϳ6.3 M and ϳ90% inhibition of CaR-stimulated cPLA 2 activity at 25 M KN-93. The IC 50 of KN-93 for purified CaMKII is 370 nM, and it is specific up to ϳ30 M (30). In PC12 cells with a 3-day incubation, the IC 50 value for KN-93-dependent inhibition of tyrosine hydroxylase activity was ϳ2 M, and pretreatment of KCl or acetyl choline-stimulated intact PC12 cells with 10 M KN-93 for 1 h resulted in a 60 -80% inhibition of tyrosine hydroxylase activity (30). Consequently, although we cannot exclude the possibility that some other CaM-dependent process participates in activation of cPLA 2 by the CaR, we think that it is most likely that the CaR acts through CaM to activate one of the isoforms of CaMK which then activates cPLA 2 . At this point, we cannot determine which CaMK isoform is involved or if CaMK acts via a direct or indirect mechanism.
Our studies demonstrate that the CaR activates cPLA 2 by a novel pathway in which Ca 2ϩ , CaM, and CaMK are of principal importance, and ERKs are not involved. CaMK is involved in activation of cPLA 2 by other G protein-coupled receptors, but CaMK activates the ERKs which then activate cPLA 2 . Clearly, receptor-dependent mechanisms that do not involve the ERKs can activate cPLA 2 (39,40). The different mechanisms used by various receptors to activate cPLA 2 may reflect selective cellular localization of signaling proteins with a particular receptor.