Erk1/2- and p38 MAP kinase-dependent phosphorylation and activation of cPLA2 by m3 and m2 receptors

Huiping Zhou, Sankar Das, Karnam S. Murthy

Abstract

This study examined the upstream signaling pathways initiated by muscarinic m2 and m3 receptors that mediate sustained ERK1/2- and p38 MAP kinase-dependent phosphorylation and activation of the 85-kDa cytosolic phospholipase (cPL)A2 in smooth muscle. The pathway initiated by m2 receptors involved sequential activation of Gβγi3, phosphatidylinositol (PI)3-kinase, Cdc42, and Rac1, p21-activated kinase (PAK1), p38 mitogen-activated protein (MAP) kinase, and cPLA2, and phosphorylation of cPLA2 at Ser505. cPLA2 activity was inhibited to the same extent (61 ± 5 to 72 ± 4%) by the m2 antagonist methoctramine, Gβ antibody, pertussis toxin, the PI3-kinase inhibitor LY 294002, PAK1 antibody, the p38 MAP kinase inhibitor SB-203580, and a Cdc42/Rac1 GEF (Vav2) antibody and by coexpression of dominant-negative Cdc42 and Rac1 mutants. The pathway initiated by m3 receptors involved sequential activation of Gαq, PLC-β1, PKC, ERK1/2, and cPLA2, and phosphorylation of cPLA2 at Ser505. cPLA2 activity was inhibited to the same extent (35 ± 3 to 41 ± 5%) by the m3 antagonist 4-diphenylacetoxy-N-methylpiperdine (4-DAMP), the phosphoinositide hydrolysis inhibitor U-73122, the PKC inhibitor bisindolylmaleimide, and the ERK1/2 inhibitor PD 98059. cPLA2 activity was not affected in cells coexpressing dominant-negative RhoA and PLC-δ1 mutants, implying that PKC was not derived from phosphatidylcholine hydrolysis. The effects of ERK1/2 and p38 MAP kinase on cPLA2 activity were additive and accounted fully for activation and phosphorylation of cPLA2.

  • cytosolic phospholipase A2
  • cytosolic phospholipase A2 phosphorylation
  • muscarinic receptors.

phospholipase(PL)A2 comprises a large group of enzymes that catalyze the hydrolysis of the sn-2-fatty acyl ester bonds of membrane glycerol-phospholipids to yield fatty acid and lysophospholipid (2, 11, 18). The 85-kDa cytosolic PLA2(cPLA2), a group IV enzyme, possesses several distinctive features including a preference for hydrolysis of arachidonate-containing phospholipids, a dependence on Ca2+for translocation of the enzyme to specific membrane sites, and a susceptibility to regulatory phosphorylation by various protein kinases (10, 19). Stimulatory phosphorylation at Ser505 and Ser727, chiefly by mitogen-activated (extracellular signal-regulated) protein (MAP) kinases, p40/p42 MAP kinases (ERK1/2), and/or p38 MAP kinase, has been described in several cell systems (3, 4, 12, 16, 20, 31). We (25) have recently demonstrated inhibitory phosphorylation of the 85-kDa cPLA2 by cAMP- and cGMP-dependent protein kinases in intestinal smooth muscle.

Binding of two Ca2+ to Asp43 and Asp93 in the NH2-terminal C2 domain of cPLA2 is essential for the enzyme to gain access to phospholipid-containing membranes. Mutation of either one of these residues yields an inactive cPLA2 (5, 10, 31). Although, Ca2+ binding can occur at resting Ca2+ concentrations, the magnitude of the increase in Ca2+ levels determines the intracellular membrane (Golgi, endoplasmic reticulum, or perinuclear membrane) targeted by the enzyme (9).

Dual phosphorylation of Ser505 and Ser727 by ERK1/2 and/or p38 MAP kinase (mainly the 2a isoform) is required for activation of cPLA2 and acts synergistically with the increase in intracellular Ca2+ (1, 3, 4, 13). The enzyme has a consensus motif for phosphorylation by various MAP kinases, which includes Ser505. Mutation of either Ser505 or Ser727 reduces cPLA2activity as much as mutation of both residues (13, 16). Both residues are phosphorylated on treatment of human platelets with thrombin or collagen; phosphorylation is entirely dependent on activation of p38 MAP kinase despite the concurrent activation of ERK1/2. p38 MAP kinase phosphorylates Ser505, and a distinct kinase, MAP kinase-interacting kinase 1 (MNK1), that is downstream of p38 MAP kinase, phosphorylates Ser727(13). Phosphorylation and activation of human platelet cPLA2 by phorbol 12,13-dibutyrate, however, is mediated by ERK1/2 and probably an MNK1-related kinase.

We (21) have previously shown that an early transient activation of cPLA2 and generation of arachidonic acid was confined to intestinal longitudinal muscle in which it caused arachidonic acid-dependent Ca2+ influx that led to Ca2+-induced Ca2+ release via sarcoplasmic ryanodine receptors/Ca2+ channels. Sustained activation of cPLA2 was present also in circular muscle (21). Recent studies (21, 27) on intestinal circular muscle suggested that sustained activation of cPLA2 was partly responsible for regulating capacitative Ca2+ influx. The effect appeared to be mediated by 4,5-epoxyeeicosatrienoic acid, a product of arachidonic acid metabolism via the monooxygenase pathway.

Upstream signaling pathways initiated by G protein-coupled receptors that eventually result in ERK1/2- and/or p38 MAP kinase-dependent sustained phosphorylation and activation of cPLA2 in smooth muscle, and other cell types have not been fully explored. The source of PKC responsible for ERK1/2 activity, and the relative contributions of ERK1/2 and p38 MAP kinase to cPLA2 phosphorylation and activity appear to be receptor and G protein specific (14, 15,32-34). In this study, we have used freshly dispersed and cultured smooth muscle cells to identify the pathways initiated by muscarinic m2 and m3 receptors that lead to phosphorylation and activation of cPLA2. The pathway initiated by m2 receptors involved sequential activation of Gβγi3→ phosphatidylinositol (PI)3-kinase →Cdc42 and Rac1 →p21-activated kinase (PAK1) →p38 MAP kinase →cPLA2. The pathway initiated by m3 receptors involved sequential activation of PLC-β1 → PKC → ERK1/2 → cPLA2. PKC derived from phosphatidylcholine hydrolysis by PLC or by RhoA-dependent phospholipase D (PLD) did not contribute to activation of cPLA2. The effects of ERK1/2 and p38 MAP kinase on cPLA2 activity were additive and accounted fully for activation and phosphorylation of cPLA2.

MATERIALS AND METHODS

Dispersion and culture of intestinal smooth muscle cells.

Smooth muscle cells were isolated from circular muscle layer of rabbit intestine by sequential enzymatic digestion, filtration, and centrifugation as described previously (23, 25). For some experiments, the muscle cells were placed in culture in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum until they attained confluence (17).

Expression of dominant-negative RhoA, Cdc42, Rac1, and PLC-δ1 cDNA in cultured smooth muscle cells.

Dominant-negative (DN) RhoA (RhoA-DN), Cdc42-DN, Rac1-DN, or PLC-δ1-DN cDNA was subcloned into the multiple cloning site (EcoRI) of the eukaryotic expression vector pEXV, and a myc tag was incorporated into the NH2 terminus. Recombinant plasmid DNAs (2 μg each) were transiently transfected into smooth muscle cells in primary culture by using Lipofectamine Plus reagent for 48 h. In some experiments, muscle cells were cotransfected with Cdc42-DN and Rac1-DN or with RhoA-DN and PLC-δ1-DN. The cells were cotransfected with 1 μg of pGreen Lantern-1 to monitor expression. Control cells were cotransfected with 2 μg of vector (pEXV) and 1 μg of pGreen Lantern-1 DNA. Transfection efficiency (∼85%) was monitored by the expression of green fluorescent protein by using FITC filters.

cPLA2 assay.

cPLA2 activity in dispersed muscle cells was measured as described previously (21, 25). Twenty microliters of cell suspension (106 cells/ml) were incubated with [3H]arachidonic acid (1 μCi/ml) at 31°C for 3 h. After labeling, the cells were washed three times with HEPES medium to remove unincorporated [3H]arachidonic acid and then incubated in the presence or absence of ACh in 1 ml of medium. In experiments in which antagonist and inhibitors were used, the cells were preincubated for 10 min. At the end of incubation, the cells were centrifuged at 1,500 g for 10 min and the radioactivity in the supernatant was determined by liquid scintillation counting.

In vitro kinase assays for p38 MAP kinase and ERK1/2.

p38 MAPK and ERK1/2 activities were determined on immunoprecipitates from cell extracts as described previously (17). Immunoprecipitates were washed twice with a phosphorylation buffer containing 10 mM MgCl2 and 40 mM HEPES (pH 7.4) and then incubated for 5 min on ice with 5 μg of myelin basic protein. Kinase assays were initiated by the addition of 10 μCi of [γ-32P]ATP (3,000 Ci/mmol) and 20 μM ATP, followed by incubation for 10 min at 37°C. [32P]myelin basic protein was absorbed onto phosphocellulose discs, and free radioactivity was removed by repeated washing with 75 mM phosphoric acid. The extent of phosphorylation was determined from the radioactivity on phosphocellulose discs by liquid scintillation.

PI3-kinase assay.

PI3-kinase was measured in smooth muscle cells as described previously (17, 32). Dispersed muscle cells were treated for 5 min with ACh in the presence or absence of 4-diphenylacetoxy-N-methylpiperdine (4-DAMP) or methoctramine. After centrifugation for 5 min, 1 ml of lysis buffer was added to the cells, and incubation was maintained for 20 min at 4°C. The cell lysates were precleared by centrifugation, and an aliquot was incubated with 5 μl of PI3-kinase antibody for 2 h at 4°C, followed by incubation with 30 μl of protein A/G-Sepharose for 2 h at 4°C. The immunoprecipitates were washed with lysis buffer and Tris · HCl buffer and incubated in a medium containing 1 mg/ml phosphatidylinositol, 20 mM MgCl2, 10 μCi of [γ-32P]ATP (3,000 Ci/mmol) and 20 μM ATP for 10 min at 37°C. The organic phase containing phosphoinositol phosphates was analyzed by thin layer chromatography and the spots were visualized by autoradiography.

Phosphorylation of cPLA2 at Ser505.

Phosphorylation of cPLA2 was measured by Western blot by using a phospho-specific antibody for phosphorylation at Ser505. Dispersed smooth muscle cells were treated with ACh for different time periods in the presence or absence of 4-DAMP, methoctramine, bisindolylmaleimide, PD 98059, or SB-203580 and solubilized on ice for 1 h in a medium containing 20 mM Tris · HCl (pH 8.0), 1 mM dithiothreitol, 100 mM NaCl, 0.5% SDS, 0.75% deoxycholate, 1 mM PMSF, 10 μg/ml leupeptin, and 100 μg/ml aprotinin. The proteins were resolved by SDS-PAGE and electrophoretically transferred on to polyvinylidene difluoride membranes, which were incubated for 12 h with the phospho-specific antibody and then incubated for 1 h with a horseradish peroxidase-conjugated secondary antibody. The bands were identified by enhanced chemiluminescence.

Materials.

[3H]arachidonic acid and [γ-32P]ATP were obtained from New England Nuclear Life Science Products; polyclonal antibody to phospho-cPLA2 (Ser505) was obtained from Cell Signaling Technology (Beverly, MA). Antibodies to PAK1, ERK1/2, and P38 MAP kinase were obtained from Santa Cruz Biotechnologies, and all other reagents were from Sigma. RhoA-DN cDNA was a gift of Dr. Andrea Todisco (University of Michigan). Cdc42-DN and Rac1-DN were a gift of Dr. Lee Slice (University of California at Los Angeles). PLC-δ1 was a gift of Dr. Srinivas Pentyala (State University of New York at Stony Brook).

RESULTS

Phosphorylation and activation of cPLA2 by muscarinic m2 and m3 receptors.

ACh induced a time-dependent increase in arachidonic acid release from dispersed intestinal smooth muscle cells, which was sustained for a period of 20 min (Fig. 1 A). Maximal arachidonic acid release was virtually abolished (91 ± 3% inhibition) by the selective cPLA2 inhibitor arachidonyltrifluoromethyl ketone (AACOCF3), implying that arachidonic acid release resulted from cPLA2 activity. cPLA2 activity (arachidonic acid release) measured at 5 min was partly inhibited by the m3 receptor antagonist 4-DAMP (35 ± 3%) and the m2 receptor antagonist methoctramine (61 ± 6%) and abolished by a combination of both antagonists (Fig. 1 B).

Fig. 1.

ACh-induced phosphorylation and activation of cytosolic phospholipase A2 (cPLA2) in smooth muscle by m3 and m2 receptors. A: dispersed intestinal muscle cells were prelabeled with [3H]arachidonic acid and treated with ACh (0.1 μM) for different time periods. cPLA2activity was expressed as the increase in [3H]arachidonic acid release above basal level [522 ± 71 counts · min−1 · mg protein−1(cpm/mg protein)]. B: muscle cells were treated with ACh (0.1 μM) for 10 min in the presence or absence of m2 receptor antagonist, methoctramine (Methoc; 0.1 μM) or the m3 receptor antagonist, 4-diphenylacetoxy-N-methylpiperdine (4-DAMP; 0.1 μM). Phosphorylation of cPLA2 (p-cPLA2) at Ser505 was performed by using a phospho-specific antibody and immunoblots of the cPLA2 bands are shown for a loading control. [3H]arachidonic acid release was virtually abolished by AACOCF3 indicating that it reflected cPLA2 activity. Neither antagonist had any effect on basal levels (range 545 ± 68 to 629 ± 94 cpm/mg protein). Values are means ± SE of 5–6 experiments. ** Significant inhibition of cPLA2 activity; P < 0.01.

ACh induced a parallel time-dependent increase in cPLA2phosphorylation at Ser505 (Fig. 1 A). cPLA2 phosphorylation measured at 5 min was partly inhibited by 4-DAMP (24 ± 3%) and methoctramine (76 ± 5%) and was abolished by a combination of both antagonists (Fig.1 B).

Previous radioligand binding and pharmacological studies in these and other types of smooth muscle cells have shown that m3 and m2 receptors are the two main muscarinic receptor types expressed in visceral and vascular smooth muscle (7, 8, 24, 28, 35). Western blot analysis and RT-PCR (28) confirmed expression of m2 and m3 receptors in intestinal smooth muscle cells. After selective protection of m3 receptors in these cells, 4-DAMP inhibited [3H]scopolamine binding with 104 greater potency than methoctramine, whereas after selective protection of m2 receptors, methoctramine inhibited [3H]scopolamine binding with 2 × 104 greater potency than 4-DAMP (24). Schild analysis confirmed that 4-DAMP was 6 × 103 more potent than methoctramine in inhibiting ACh-induced muscle cell contraction (24). ACh-induced activation of Gq was selectively inhibited by 4-DAMP, whereas activation of Gi3 was selectively inhibited by methoctramine and pertussis toxin (PTx) (24). In the present study, 4-DAMP and methoctramine were used to identify signaling pathways mediated by m3 and m2 receptors, respectively. PTx was used to confirm activation of the pathways initiated by m2 receptors.

Inhibition of ACh-stimulated cPLA2 activity and phosphorylation by PKC, ERK1/2, and p38 MAP kinase inhibitors.

ACh-stimulated cPLA2 activity was inhibited to the same extent (37 ± 4 to 41 ± 5%) by the ERK1/2 inhibitor PD 98059, the PKC inhibitor bisindolylmaleimide, or a combination of both inhibitors (Fig. 2). cPLA2activity was more potently inhibited (72 ± 4%) by the p38 MAP kinase inhibitor SB-203580 and abolished by a combination of SB-203580 with either PD 98059 or bisindolylmaleimide (Fig. 2). cPLA2phosphorylation was affected in similar fashion by these inhibitors with SB-203580 causing more profound inhibition of phosphorylation (65 ± 6%) than either PD 98059 (27 ± 4%) or bisindolylmaleimide (32 ± 6%) and virtually abolishing phosphorylation in combination with either PD 98059 or bisindolylmaleimide (Fig. 2).

Fig. 2.

Effects of PKC, ERK1/2, and p38 mitogen-activated protein (MAP) kinase inhibitors on ACh-stimulated cPLA2phosphorylation and activity in smooth muscle cells. Dispersed intestinal smooth muscle cells were prelabeled with [3H]arachidonic acid and treated for 10 min with ACh (0.1 μM) after the addition of bisindolylmaleimide (bis; 1 μM), PD 98059 (10 μM) or SB-203580 (1 μM), alone or in combination. Phosphorylation at Ser505 (p-cPLA2) was performed by using a phospho-specific antibody and immunoblots of the cPLA2 bands are shown for a loading control. cPLA2 activity ([3H]arachidonic acid release) was measured as described in materials and methods and expressed in cpm/mg protein above basal level (613 ± 84 cpm/mg protein). The p38 MAP kinase inhibitor, SB-203580 abolished cPLA2 activity when present in combination with either the ERK1/2 inhibitor PD 98059 or the PKC inhibitor, bisindolylmaleimide. Values are means ± SE of 4 experiments. ** Significant inhibition of cPLA2 activity; P < 0.01.

The degree of inhibition of cPLA2 activity by PD 98059, bisindolylmaleimide or a combination of both inhibitors was similar to that elicited by 4-DAMP, suggesting that m3 receptors initiated a pathway involving sequential activation of PKC, ERK1/2, and cPLA2. Similarly, the extent of inhibition by SB-203580 was similar to that elicited by methoctramine, suggesting that m2 receptors initiated a pathway involving sequential activation of p38 MAP kinase and cPLA2. In accordance with these notions, ACh-stimulated cPLA2 activity in the presence of methoctramine was abolished by PD 98059 or bisindolylmaleimide but was not affected by SB-203580 (Fig. 3), whereas cPLA2 activity in the presence of 4-DAMP was abolished by SB-203580 but was not affected by PD-98059 or bisindolylmaleimide (Fig.3). Results shown in Figs. 1-3 implied that ERK1/2 and p38 MAP kinase accounted fully for cPLA2 phosphorylation and activity.

Fig. 3.

Effects of PKC, ERK1/2 and p38 MAP kinase inhibitors on ACh-stimulated cPLA2 activity in smooth muscle cells in the presence of m3 and m2 receptor antagonists. Dispersed muscle cells were prelabeled with [3H]arachidonic acid and treated for 10 min with ACh (0.1 μM) after the addition of bisindolylmaleimide (1 μM), PD 98059 (10 μM) or SB-203580 (1 μM) in the presence or absence of methoctramine (Methoc; 0.1 μM) or 4-DAMP (0.1 μM). cPLA2 activity in the presence of methoctramine was abolished by PD 98059 or bisindolylmaleimide, whereas cPLA2activity in the presence of 4-DAMP was abolished by SB-203580. cPLA2 activity was expressed as the increase in [3H]arachidonic acid release above basal level (range: 548 ± 70 to 613 ± 86 cpm/mg protein). Values are means ± SE of 4 experiments.

Activation of ERK1/2 via m3 receptors and p38 MAP kinase via m2 receptors.

ACh-stimulated ERK1/2 activity (97 ± 7% above basal level) was abolished by 4-DAMP, PD 98059, and bisindolylmaleimide, but was not affected by methoctramine or SB-203580 (Fig.4), whereas ACh-stimulated p38 MAP kinase activity (215% ± 21% above basal level) was abolished by methoctramine and SB-203580 but was not affected by 4-DAMP, PD 98059, or bisindolylmaleimide (Fig. 4). Results from direct measurement of the MAP kinases confirmed the conclusions derived from measurement of cPLA2 in the presence of MAP kinase inhibitors and muscarinic receptor antagonists. Inhibition of ERK1/2 activity by bisindolylmaleimide supported the notion that ERK1/2 was downstream of PKC in the pathway leading to activation of cPLA2 by m3 receptors.

Fig. 4.

Activation of ERK1/2 and p38 MAP kinase by m3 and m2 receptors, respectively. Dispersed muscle cells were treated for 10 min with ACh (0.1 μM) in the presence or absence of methoctramine (0.1 μM), 4-DAMP (0.1 μM), bisindolylmaleimide (1 μM), PD 98059 (10 μM), or SB-203580 (1 μM), and ERK1/2 and p38 MAP kinase activity was measured in immunoprecipitates as described in materials and methods. ACh-induced ERK1/2 activity was inhibited by 4-DAMP, PD 98059, and bisindolylmaleimide, whereas p38 MAK kinase activity was inhibited by methoctramine and SB-203580. ERK1/2 and p38 MAP kinase activities were expressed as the increase in cpm/mg protein above basal level (ERK1/2: 1,942 ± 264 cpm/mg protein; p38 MAP kinase: 6,519 ± 538 cpm/mg protein). None of the inhibitors had any effect on basal levels. Values are means ± SE of 4 experiments. ** Significant inhibition; P < 0.01.

Source of PKC for m3 receptor-dependent activation of ERK1/2 and cPLA2.

Earlier studies on intestinal smooth muscle cells have shown that agonist-stimulated PKC activity is derived initially from phosphoinositide hydrolysis by PLC-β and subsequently from phosphatidylcholine hydrolysis by a RhoA-dependent PLD and a phosphatidylcholine-specific PLC (23, 29). ACh-stimulated, PKC-dependent ERK1/2 activity was not affected by expression of a RhoA-DN in cultured smooth muscle cells, or coexpression of Cdc42-DN and Rac1-DN (Fig. 5). We considered the possibility that in cells expressing RhoA-DN, PKC could be derived from activation of PLC-δ1, which is normally repressed by RhoA and is derepressed in cells expressing RhoA-DN. However, cPLA2activity was not affected in cells expressing RhoA-DN, PLC-δ1-DN, or coexpressing RhoA-DN and PLC-δ1-DN (Fig.6). The functionality of these mutants is supported by the fact that expression of RhoA-DN inhibits ACh-stimulated RhoA activity, whereas expression of Cdc42-DN and/or Rac1-DN inhibits ACh-stimulated PAK1 activity (29, 30). Expression of PLC-δ-DN inhibits Ca2+-stimulated PLC-δ activity (26).

Fig. 5.

ACh-stimulated ERK1/2 and p38 MAP kinase activities in cultured smooth muscle cells expressing RhoA-dominant negative (RhoA-DN) or coexpressing Cdc42-DN and Rac1-DN. Primary cultures of smooth muscle cells expressing RhoA-DN or coexpressing Cdc42-DN/Rac1-DN were treated with ACh (0.1 μM) for 10 min and ERK1/2 and p38 MAP kinase activities were measured in immunoprecipitates. ACh-induced p38 MAP kinase activity was inhibited only in cells coexpressing Cdc42-DN and Rac1-DN (**P < 0.01), but was not affected in cells expressing RhoA-DN. ERK1/2 and p38 MAP kinase activities were expressed as the increase in cpm/mg protein above basal level (ERK1/2 range: 1,959 ± 195 to 2,182 ± 220 cpm/mg protein; p38 MAP kinase range: 6,623 ± 816 to 7,359 ± 921 cpm/mg protein). Expression of dominant negative mutants had no effect on basal levels. Values are means ± SE of 4 experiments.

Fig. 6.

ACh-stimulated cPLA2 activity in cultured smooth muscle cells expressing RhoA-DN and/or PLC-δ1-DN. Primary cultures of smooth muscle cells expressing RhoA-DN and/or PLC-δ1-DN were treated with ACh (0.1 μM) for 10 min and cPLA2activity was measured. cPLA2 activity was expressed as the increase in [3H]arachidonic acid release above basal level (range: 523 ± 71 to 576 ± 62 cpm/mg protein). Expression of dominant negative mutants had no effect on basal level. Values are means ± SE of 6–8 experiments. ** Significant inhibition of activity; P < 0.01.

Finally, we considered the possibility that the PKC responsible for sustained cPLA2 activity was derived from phosphoinositide hydrolysis by PLC-β1. ACh-stimulated phosphoinositide hydrolysis in smooth muscle cells peaks within 30 s coincidentally with maximal Ca2+ release, and declines rapidly to a low suprabasal level (22). ACh-stimulated ERK1/2 activity was strongly inhibited (78 ± 6%) by U-73122 implying that it was mediated by PKC derived from phosphoinositide hydrolysis. ACh-stimulated cPLA2 activity was inhibited 37 ± 3% by U-73122 (Fig. 7), which is to the same extent as inhibition by 4-DAMP, bisindolylmaleimide, or PD 98059 (see Figs. 1 and2). cPLA2 activity was abolished by U-73122 in the presence of methoctramine (Fig. 7). The pattern of inhibition of ERK1/2 and cPLA2 activity by U-73122 implied that m3 receptor-dependent, ERK1/2-induced cPLA2 activity was mediated by PKC derived from phosphoinositide hydrolysis, presumably acting cooperatively with the increase in Ca2+.

Fig. 7.

Inhibitory effect of U-73122 on ACh-stimulated cPLA2 activities. Dispersed smooth muscle cells were stimulated with ACh (0.1 μM) after treatment for 10 min with U-73122 (10 μM) alone or in the presence of methoctramine (0.1 μM) or 4-DAMP. cPLA2 activity was expressed as cpm/mg protein above basal level (602 ± 43 cpm/mg protein). U-73122 inhibited cPLA2 to the same extent as 4-DAMP, and abolished cPLA2 activity in the presence of methoctramine. Values are means ± SE of 4 experiments.

Role of Cdc42, Rac1, and PAK1 in m2 receptor-dependent activation of p38 MAP kinase and cPLA2.

Our recent studies (30) have shown that m2 receptors are coupled to sequential activation of Cdc42/Rac1 and PAK1. ACh-stimulated PAK1 activity was partly inhibited in smooth muscle cells expressing Cdc42-DN or Rac1-DN and abolished in cells coexpressing both mutants. PAK1 activity was not affected by SB-203580 implying that PAK1 is upstream of p38 MAP kinase (30). In the present study, ACh-stimulated p38 MAP kinase activity was abolished in cultured smooth muscle cells coexpressing Cdc42-DN and Rac1-DN but was not affected in cells expressing RhoA-DN (Fig. 5). Activation of p38 MAP kinase by Cdc42 and Rac1 was probably mediated by PAK1 (6). Treatment of permeabilized smooth muscle cells for 60 min with PAK1 antibody (5 μg/ml) inhibited m2 receptor-dependent cPLA2activity (85 ± 7% inhibition) but had no effect on m3 receptor-dependent activity (6 ± 8% inhibition).

ACh-stimulated cPLA2 activity was strongly inhibited (64 ± 5%) in cultured smooth muscle cells coexpressing Cdc42-DN and Rac1-DN (Fig. 8) but was not affected in cells expressing RhoA-DN (vector alone, 4976 ± 549; RhoA-DN, 4,798 ± 662 counts/min (cpm)/mg protein above basal level). Residual cPLA2 activity in cells coexpressing Cdc42-DN and Rac1-DN was abolished by 4-DAMP, bisindolylmaleimide, and PD 98059 but was not affected by SB-203580, implying that it reflected the m3 receptor-dependent response mediated by PKC and ERK1/2 (Fig. 8).

Fig. 8.

ACh-stimulated cPLA2 activity in cultured smooth muscle cells coexpressing Cdc42-DN and Rac1-DN. Muscle cells coexpressing Cdc42-DN and Rac1-DN were treated with ACh (0.1 μM) for 10 min in the presence or absence of methoctramine (0.1 μM), 4-DAMP (0.1 μM), bisindolylmaleimide (1 μM), PD 98059 (10 μM), or SB-203580 (1 μM). ACh-induced cPLA2 activity was inhibited in cells coexpressing Cdc42-DN and Rac1-DN. Residual ACh-induced cPLA2 activity was low and was abolished by 4-DAMP, bisindolylmaleimide, and PD 98059, but was not affected by methoctramine or SB-203580. Results were expressed in cpm/mg protein above basal levels (840 ± 109 cpm/mg protein in cells expressing vector alone vs. 804 ± 114 cpm/mg protein in cells coexpressing Cdc42-DN and Rac1-DN). Values are means ± SE of 4 experiments. ** Significant inhibition; P < 0.01.

Participation of Cdc42 and Rac1 in ACh-stimulated cPLA2 was corroborated by pretreatment of permeabilized intestinal smooth muscle cells for 60 min with antibody to Vav2, a Cdc42/Rac1 GTP exchange factor (5 μg/ml). Vav2 antibody had no effect on basal cPLA2 activity but inhibited ACh-stimulated activity by 62 ± 6% (Fig. 9). Inhibition of cPLA2 activity by Vav2 antibody was similar to that elicited by methoctramine and was not additive to it (Fig. 9). In contrast, inhibition of cPLA2 activity by Vav2 antibody was additive to that elicited by 4-DAMP; the combination of Vav2 antibody with 4-DAMP abolished cPLA2 activity (Fig. 9). The extent of inhibition by Vav2 (62 ± 6%) was similar to that elicited in cultured smooth muscle cells coexpressing Cdc42-DN and Rac1-DN (64 ± 5% inhibition; Fig. 8), and in freshly dispersed smooth muscle cells treated with methoctramine (61 ± 4%; Fig. 1) or SB-203580 (72 ± 4%; Fig. 2).

Fig. 9.

Effect of Vav2 antibody on ACh-stimulated cPLA2 activity. Dispersed saponin-permeabilized smooth muscle cells were treated with ACh (0.1 μM) for 10 min after preincubation for 60 min with Vav2 antibody (5 μg/ml). In some experiments muscle cells were treated with ACh (0.1 μM) for 10 min after preincubation for 60 min with Vav2 antibody in the presence or absence of methoctramine (0.1 μM) or 4-DAMP (0.1 μM). cPLA2 activity was expressed as cpm/mg protein above basal level (613 ± 105 and 564 ± 71 cpm/mg protein in the presence and absence of Vav2 antibody). Residual cPLA2activity was abolished in the presence of 4-DAMP. Values are means ± SE of 4 experiments.

Role of Gi3 and PI3-kinase in m2 receptor-dependent activation of cPLA2 activity.

ACh stimulated PI3-kinase activity, which was abolished by the selective PI3-kinase inhibitor LY 294002 and by methoctramine but was not affected by 4-DAMP (Fig. 10). ACh-stimulated cPLA2 activity was inhibited to the same extent (59 ± 5 to 62 ± 4%) by pretreatment of dispersed smooth muscle cells for 60 min with PTx (400 ng/ml) or for 10 min with LY 294002 (10 μM), and by pretreatment of saponin-permeabilized smooth muscle cells with a common Gβ antibody (10 μg/ml) (Fig. 11). Inhibition of cPLA2 activity by PTx, Gβ antibody, and LY 294002 was similar to that elicited by methoctramine (Fig.11). The combination of methoctramine with PTx, Gβ antibody, or LY 294002 was not additive, whereas the combination of 4-DAMP with each of the three agents abolished ACh-stimulated cPLA2 activity (Fig. 11). The inhibitory effects of PTx and Gβ antibody implied m2 receptor-dependent cPLA2 activity was mediated by Gβγi3. The inhibitory effect of LY 294002 implied that Gβγ was linked via PI3-kinase to downstream effectors (Cdc42/Rac1 → PAK1 → p38 MAP kinase → cPLA2).

Fig. 10.

Selective activation of PI3-kinase by m2 receptors. Freshly dispersed smooth muscle cells were treated with ACh (0.1 μM) in the presence or absence of methoctramine (0.1 μM), 4-DAMP (0.1 μM), or the selective PI3-kinase inhibitor, LY 294002 (10 μM). PI3-kinase activity was measured by immunokinase assay and expressed as cpm/mg protein above basal level (311 ± 68 cpm/mg protein). PI3-kinase activity was abolished by methoctramine and LY 294002 but was not affected by 4-DAMP. Values are means ± SE for 4 experiments.

Fig. 11.

Inhibition of ACh-stimulated cPLA2 activity by pertussis toxin (PTx), Gβ antibody, and the PI3-kinase inhibitor, LY 294002. A: muscle cell were treated with ACh (0.1 μM) for 10 min after preincubation with PTx (400 ng/ml) or Gβ antibody (10 μg/ml) for 60 min or with LY 294002 for 10 min. B: muscle cells were treated with ACh (0.1 μM) for 10 min after preincubation with PTx, Gβ antibody, or LY 294002 in the presence of methoctramine (0.1 μM) or 4-DAMP (0.1 μM). PTx, Gβ antibody or LY 294002 abolished cPLA2 activity in the presence of 4-DAMP. cPLA2 activity was measured and expressed as cpm/mg protein above basal level (range: 564 ± 65 to 634 ± 75 cpm/mg protein). PTx, Gβ antibody, and LY 294002 had no effect on basal level. Values are means ± SE of 4 experiments.

DISCUSSION

This study identified the upstream signaling pathways initiated by muscarinic m2 and m3 receptors that lead to sustained phosphorylation and activation of cPLA2 in smooth muscle. Our earlier studies (24, 30) have shown that m2 receptors, the predominant species in smooth muscle (∼80% of the total), are coupled via Gαi3 to inhibition of adenylyl cyclase and via Gβγi3 to transient activation of PLC-β3 and sustained, sequential activation of Cdc42/Rac1 and PAK1. The latter mediates inhibitory phosphorylation of myosin light chain kinase (30) and, as shown in the present study, stimulation of p38 MAP kinase and activation of cPLA2. Furthermore, the present study shows that PI3-kinase acts as the link between Gβγi3 and the downstream pathway.

Previous studies have also shown that m3 receptors, like other Gq coupled receptors (30), are coupled via Gαq to activation of PLC-β1 and phosphoinositide hydrolysis, and via G13 and RhoA to sustained activation of Rho kinase and PLD. phosphatidylcholine hydrolysis by PLD yields phosphatidic acid, which is dephosphorylated to diacylglycerol, causing sustained activation of PKC (23). Phosphatidylcholine hydrolysis by PLC also yields diacylglycerol and contributes to sustained activation of PKC (23). Neither pathway, however, was responsible for sustained PKC-dependent, ERK1/2-mediated activation of cPLA2, which appeared to depend on the initial PKC activity derived from phosphoinositide hydrolysis. This pathway contributes only a fraction, about one-third, of ACh-stimulated cPLA2 activity. The remainder is mediated via m2 receptors.

The linkage between m2 receptors and cPLA2 involved sequential activation of Gβγi3 and PI3-kinase; cPLA2 activity was inhibited to the same extent by methoctramine, PTx, a common Gβ antibody, and the PI3-kinase inhibitor LY 294002 (Figs. 1 and 11). When m3 receptor-mediated responses were blocked with 4-DAMP, treatment with PTx, Gβ antibody, or LY 294002 abolished ACh-stimulated cPLA2 activity. It is worth noting that blockade of every step in the pathway initiated by m2 receptors (Gβγi3→ PI3-kinase →Cdc42/Rac1 →PAK1 →p38 MAP kinase →cPLA2) elicited the same degree of inhibition (61 ± 5 to 72 ± 4%) of cPLA2 activity or phosphorylation, whether the step involved m2 receptor antagonism with methoctramine (Fig. 1), uncoupling of m2 receptors from Gi3 with PTx (Fig. 11 A), neutralization of Gβγ activity with Gβ antibody (Fig.11 A), inhibition of PI3-kinase activity with LY 294002 (Fig.11 A), inactivation of Cdc42/Rac1 by expression of dominant-negative mutants (Fig. 8) or with specific inhibitor (Vav2 antibody) (Fig. 9), immunoneutralization of PAK1, and inhibition of p38 MAP kinase activity with SB-203580 (Fig. 2).

Unexpectedly, the PKC responsible for m3 receptor-dependent sustained activation of ERK1/2 and cPLA2 was generated from phosphoinositide hydrolysis, which in smooth muscle cells occurs predominantly in the first 2 min after receptor activation before reverting to a low suprabasal level (23). Our previous studies in intestinal smooth muscle have shown that sustained PKC activity is predominantly generated via phosphatidylcholine hydrolysis via RhoA-activated PLD (23, 29). Expression of RhoA-DN in these cells abolishes RhoA and PLD activities (29, 30) but had no effect on ACh-stimulated cPLA2 activity. We considered the possibility that inactivation of RhoA could suppress its inhibitory effect on phosphoinositide-specific PLC-δ1 leading to generation of PKC; however, coexpression of RhoA-DN and PLC-δ1-DN had no effect on ACh-stimulated cPLA2 activity. We concluded that inhibition of ERK1/2 activity by U-73122 and bisindolylmaleimide reflected inhibition of PKC derived from phosphoinositide hydrolysis by PLC-β1.

In their studies of iris smooth muscle, Husain and Abdel Latif (14, 15) reported agonist-dependent differences in the contribution of ERK1/2 and p38 MAP kinase to phosphorylation and activation of cPLA2. Both ERK1/2 and p38 MAP kinase were involved in phosphorylation and activation of cPLA2 by the muscarinic agonist carbachol (15). Upstream pathways initiated by m3 and m2 receptors were not examined, except for the observation that both MAP kinases were dependent on PKC, in contrast with the present study in which only ERK1/2 activation was dependent on PKC. With PGF and endothelin (14, 15), however, although ERK1/2 and p38 MAP kinase were activated, only the latter appeared to be involved in phosphorylation and activation of cPLA2. Studies (11, 12, 18, 31) on peritoneal macrophages have demonstrated the ability of Ca2+(Ca2+ ionophores) or MAP kinase-dependent phosphorylation (phorbol esters and phosphatase inhibitors) to activate cPLA2 independently or synergistically. In thrombin- or collagen-stimulated human platelets, activation of cPLA2 by Ca2+ appears to predominate over activation induced by cPLA2 phosphorylation via p38 MAP kinase (3, 4,16). The extent of Ca2+ mobilization determined the specific intracellular membranes targeted by cPLA2(9).

Although other investigators (1) have detected a negative control of ERK1/2 by p38 MAP kinase, we were unable to detect interplay between ERK1/2 and p38 MAP kinase in the present study. Inhibition of either enzyme had no effect on their activation by ACh or on their ability to activate cPLA2.

We have previously shown that an early transient activation of cPLA2 and generation of arachidonic acid was confined to intestinal longitudinal muscle in which it caused arachidonic acid-dependent Ca2+ influx that led to Ca2+-induced Ca2+ release via sarcoplasmic ryanodine receptors/Ca2+ channels. Sustained activation of cPLA2 was present also in circular muscle (21 and present study). Recent studies on intestinal circular muscle suggested that sustained activation of cPLA2 was partly responsible for regulating capacitative Ca2+ influx. The effect appeared to be mediated by 4,5-epoxyeeicosatrienoic acid, a product of arachidonic acid metabolism via the monooxygenase pathway (21, 27).

In summary, we have characterized the upstream signaling pathways initiated by muscarinic m2 and m3 receptors that lead to phosphorylation and activation of cPLA2. The study shows that the pathways initiated by m2 receptors involved sequential activation of Gβγi3→ PI3-kinase →Cdc42 and Rac1 →PAK1 →p38 MAP kinase →cPLA2, whereas the pathways initiated by m3 receptors involved sequential activation of PLC-β1 → PKC → ERK1/2 → cPLA2. Phosphorylation and activation of cPLA2 were independently and additively mediated by m3 receptor-dependent ERK1/2 and m2 receptor-dependent p38 MAP kinase.

Acknowledgments

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-15564.

Footnotes

  • Address for reprint requests and other correspondence: K. S. Murthy, P.O. Box 980711, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, VA 23298 (E-mail: skarnam{at}hsc.vcu.edu).

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • 10.1152/ajpgi.00345.2002

REFERENCES

View Abstract