Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The adrenergic transmitter norepinephrine produces a wide variety of biological actions, including AA release for prostaglandin synthesis, via activation of distinct types of AR, e.g., alpha ARs in the kidney, spleen and blood vessels (Malik, 1988), beta-1 ARs in the heart (Shaffer and Malik, 1982) and beta-2 ARs in the lung (bronchial smooth muscle) (Lew et al., 1992). Pharmacological, radioligand binding and molecular cloning studies have led to the further characterization of subtypes of alpha-1, alpha-2 and beta ARs (Bylund, 1992; Minneman and Esbenshade, 1994; Graham et al., 1996; Lands et al., 1967; Frielle et al., 1987; Kobilka et al., 1987; Emorine et al., 1989). Alpha-1 ARs initially were subclassified into alpha-1A and alpha-1B ARs based on differences in their binding profiles of alpha AR antagonists and the ability to selectively block various biological responses mediated via the activation of alpha-1 ARs (Morrow and Creese, 1986). However, molecular cloning studies have identified cDNAs for three distinct alpha-1 AR subtypes encoding one subtype of alpha AR not previously characterized by pharmacological or radioligand binding studies (Graham et al., 1995). One of these clones, the alpha-1b AR (original terminology), exhibits characteristics identical with the pharmacologically defined alpha-1B AR (Hieble et al., 1995). Another clone, the alpha-1c AR (Schwinn et al., 1990) initially was believed to be a distinct alpha-1 AR, but has now been shown to be a homolog of the pharmacologically identified alpha-1A AR (Hieble et al., 1995, Perez et al., 1994). The cloned alpha-1a/d AR initially was thought to be the alpha-1A AR (Lomasney et al., 1991), but now has been shown to be a distinct receptor expressed in many tissues and has been classified as alpha-1D AR (Perez et al., 1991).

Alpha-1 AR subtypes are coupled to a wide variety of effector systems via distinct G proteins (Minneman, 1988; Exton, 1996; Wu et al., 1992; Schwinn et al., 1991; Nebigil and Malik, 1992; Perez et al., 1993). Although all alpha-1 AR subtypes are coupled predominantly to pertussis toxin-insensitive G proteins of the Gq/₁₁ family, there is evidence that alpha-1 AR subtypes can also couple to other effector systems, including the phospholipase A₂ through pertussis toxin-sensitive G_i or G_o family of G proteins (Minneman, 1988; Exton, 1996; Wu et al., 1992). Activation of all alpha-1 AR subtypes increases levels of cytosolic Ca⁺⁺ by increasing the influx of extracellular Ca⁺⁺ through voltage-operated Ca⁺⁺ channels and/or releasing intracellular Ca⁺⁺ by activating PLC and generating myoinositol trisphosphate (Minneman, 1988; Exton, 1996; Wu et al., 1992; Schwinn et al., 1991; Nebigil and Malik, 1992, 1993; Perez et al., 1993). Moreover, stimulation of cloned alpha-1B and alpha-1D ARs in COS and CHO cells promotes PLA₂ activation and AA release (Perez et al., 1993). Activation of alpha-1 ARs has also been reported to increase PLD activity in rat brain slices (Llahi and Fain, 1992) and rat tail arteries (Gu et al., 1992). PLD promotes the breakdown of phosphatidylcholine into PA and choline. PA is hydrolyzed by PPH into DAG (Exton, 1990; Billah and Anthes, 1990). DAG is phosphorylated into PA by DAG kinase or metabolized by DAG lipase into AA and MAG (Billah and Anthes, 1990; Balsinde et al., 1991). Whether activation of one or more subtypes of alpha ARs stimulates AA release via activation of the PLD pathway is not known. Stimulation of alpha-1 ARs has also been reported to increase cAMP accumulation and to potentiate cAMP levels elicited through other receptors (Schults and Daly, 1973; Johnson and Minneman, 1986). Activation of alpha-1A ARs in COS-7 and HeLa cells (Schwinn et al., 1991) or alpha-1B or alpha-1D ARs in COS-1 and CHO cells was also shown to increase cAMP levels (Perez et al., 1993). In this study, we demonstrate that PHE-induced activation of alpha-1 AR subtypes stably expressed in R-1Fs results in increases in AA release, PLD activity and cAMP accumulation. Furthermore, we report that PHE-induced activation of alpha-1A AR in these cells releases AA through the selective stimulation of PLD via a pertussis toxin-sensitive G protein and that cAMP generated by alpha-1A AR stimulation exerts an inhibitory effect on PLD activity by a mechanism dependent on PKA. Part of this work has been published as a preliminary communication (Ruan et al., 1996).

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials

The drugs purchased for use in this study were: phenylephrine, endothelin-1, calcium ionophore A-23187, IBMX, cpt-cAMP, PMSF, propranolol, cholera toxin, pertussis toxin, EGTA and polyethyleneimine from Sigma (St. Louis, MO); forskolin from Research Biochemicals International (Natick, MA); leupeptin and aprotinin from Calbiochem-Novabiochem (San Diego, CA); SQ 22536, KT 5720, RHC 80267, C₂-ceramide and C₂-dihydroceramide from Biomol (Plymouth Meeting, PA); [³H]prazosin (79.8 Ci/mmol) and [³H]AA (100 Ci/mmol) from Du Pont Corp. (Boston, MA); and [³H]oleic acid (50 Ci/mmol) and phosphatidylcholine, L-a-1-palmitoyl-2-arachidonyl[¹⁴C] (57 Ci/mmol) from American Radiolabeled Chemicals (St. Louis, MO). Forskolin, SQ 22536, KT 5720, RHC 80267 and C₂-ceramide were dissolved in 10 µl of dimethyl sulfoxide and then diluted up to 1 ml with distilled water; 10 µl of this solution was added to 1 ml of medium containing cells to obtain the final concentration. Other drugs were dissolved in 1 ml of distilled water to make stock solutions, and 10 µl of the solution was added to 1 ml of the medium containing cells to obtain the final concentration.

Cell Culture

Rat-1 fibroblasts were stably transfected with alpha-1A, alpha-1B and alpha-1D ARs, as described previously (Allen et al., 1991). Cells were maintained under 5% CO₂ at 37°C in DMEM containing 50 units of penicillin, 50 µg of streptomycin per ml and 5% fetal bovine serum. The medium in the clusters or the dishes was changed every 2 days. For AA release and cAMP accumulation studies, cells were plated into a 24-well cluster or 100-mm dishes for PLD and PLA₂ activity measurements. Nine to twelve wells of cells were used for AA release and three dishes of cells for the assay of PLD and PLA₂ activity.

Experimental Protocols

Radioligand binding study. Saturation binding was performed by a modification of the method described previously (Zhang et al., 1992). Cells were washed twice with ice-cold 50 mM Tris-HCl buffer containing 1 mM EGTA (pH 7.4). The cells were scraped from the culture dish and sonicated for 30 sec in an ice bath. Nuclei and cell debris were removed by centrifugation at 1,000 × g for 5 min, and the supernatant, which contained plasma membranes, was used for the binding assay. Binding of [³H]prazosin (79.8 Ci/mmol), 62.5 pM to 8 nM, to the plasma membranes was determined with nonspecific binding measured in the presence of 10 µM phentolamine. The equilibrium binding assay was performed at 30°C for 45 min with 100 µg protein and was terminated by the addition of 3 ml of cold Tris-HCl buffer. Bound and free [³H]prazosin were separated by rapid filtration of the suspension through Whatman GF/C filters presoaked in 0.5% polyethyleneimine. Filters were washed twice with 3 ml ice-cold Tris-HCl buffer and counted by liquid scintillation spectrometry.

Studies on the contribution of different lipases to the release of AA release and its modulation by cAMP. To determine the contribution of different lipases to PHE-induced AA release, R-1Fs expressing different subtypes of alpha-1 AR were labeled with [³H]AA, and the release of AA in response to PHE (10 min) was studied after exposure of the cells to various inhibitors or their respective vehicles for time intervals as shown in diagram 1 

View larger version (7K): [in this window] [in a new window]

View larger version (9K): [in this window] [in a new window]

Release of AA. When cells were 60% confluent, the medium was removed and replaced with 0.5 ml culture medium containing 0.1 µCi [³H]AA (100 Ci/mmol) and incubated for 18 hr at 37°C. Cells were washed twice with HBSS and incubated with DMEM containing 0.2% BSA and various antagonists; cells were then exposed to agonists or their vehicle for 15 min, and the reaction medium was removed and the radioactivity was measured by liquid scintillation spectroscopy. Cells were digested in 1 ml of 1 M NaOH overnight for total radioactivity measurement. Tritium released into the medium was expressed as a percent of the total cellular radioactivity and referred to as fractional release of [³H]AA.

Assay of cAMP accumulation. Cells were washed twice with HBSS and then incubated with 1 ml of DMEM at 37°C containing 10 µM IBMX to minimize cAMP degradation by phosphodiesterases and several antagonists. PHE or the vehicle was added and incubated for 15 min. The reaction was terminated by adding 1 ml of sodium acetate (50 mM, pH 4.2). cAMP was determined by a modification of the method described previously (Bruckner et al., 1985). The cells were frozen at 20°C, then thawed and boiled for 3 min. Samples were centrifuged at 1,000 × g for 5 min, and 0.5 ml of the supernatants were acetylated by the addition of triethylamine and acetic anhydride (3:2) to increase the sensitivity of the assay. Fifty microliters of acetylated sample or standards were added to 50 µl of [¹²⁵I]cAMP [~3000 cpm, ~70 TBq/mmol; prepared by the procedure described (Steiner et al., 1972)], and 200 µl of antibody (gift from Dr. Charles W. Leffler, Department of Physiology, University of Tennessee, Memphis) were added. Both [¹²⁵I]cAMP and antibody were diluted in 50 mM sodium acetate buffer (pH 6.2) containing 5 mg/ml BSA; the mixture was vortexed and incubated at 4°C overnight. Phosphate buffer (150 µl) containing 1% -globulin and 2 ml 25% polyethylene glycol was added to the tubes, which were then vortexed and centrifuged at 1,700 × g for 30 min. The supernatant was removed, and radioactivity was determined in a gamma counter (model 4/200, Micromedic Systems, Horsham, PA).

Assay of PLD activity. PLD activity was measured by a modification of the method described previously (Liu et al., 1992). Cells were incubated for 18 hr in 100-mm culture dishes with culture medium containing [³H]oleic acid (1 µCi/ml; 50 Ci/mmol) or [³H]AA (1 µCi/ml; 100 Ci/mmol). The labeled cells were washed twice with HBSS and fresh serum-free DMEM containing various antagonists, and 200 mM ethanol was added. After 10 min, cells were stimulated with PHE for an additional 15 min, and the reaction was terminated by adding 1 ml of ice-cold methanol/HCl (2 M; 9:1 v/v). The cells were scraped from the culture dish and washed once with 0.25 M HCl. The cells and wash solutions were then transferred to 15-ml centrifuge tubes and sonicated for 1 min in an ice bath. One milliliter of chloroform was added to each tube, mixed and centrifuged at 1,000 × g for 10 min. The top aqueous layer was discarded, and 0.8 ml from the chloroform layer was transferred to a glass tube. A 40-µl aliquot was removed to estimate radioactivity in the total lipid fraction. The remaining chloroform phase was evaporated under a nitrogen stream and the residue resuspended in 50 µl of methanol/chloroform (1:9) containing 10 µg of nonradiolabeled PEt standard (Biomol, Plymouth Meeting, PA). Samples were spotted on a channeled silica gel TLC plate (Analtech, Newark, DE) and the plate developed using chloroform/acetone/methanol/acetic acid/H₂0 (100:40:25:20:10). Lipids were visualized with iodine and identified by comigration with standards. Lanes containing PEt were moistened with water and scraped, and radioactivity was measured by liquid scintillation spectroscopy; [³H]PEt production was expressed as the fraction of total [³H]lipids.

Assay of PLA₂ activity. Cells grown in 100-mm dishes were washed twice with HBSS and incubated with DMEM containing an agonist or its vehicle for 15 min at 37°C. PLA₂ activity was measured as described previously (de Carvalho et al., 1995). The cells were scraped and sonicated in lysis buffer (pH 7.4) containing (in mM): sucrose, 340; EGTA, 1; PMSF, 0.3; leupeptin, 0.04; aprotinin, 0.02; and 20 µg/ml soybean trypsin inhibitor. The concentration of protein in the lysate was determined by the method of Lowry et al. (1951) and adjusted to 1 mg/ml. PLA₂ activity was measured with use of phosphatidylcholine, L-a-1-palmitoyl-2-arachidonyl[¹⁴C] (57 mCi/mmol) as substrate (50,000 cpm/5 µl/assay tube), cosonicated with 9 µM dioleoylglycerol, 1 mg/ml BSA, 150 mM NaCl, 5 mM CaCl₂ and 25 mM HEPES, pH 7.4. Cell lysates containing 25 µg of protein were added, and the reaction mixture was incubated for 30 min at 37°C. The reaction was terminated with 2.5 ml of Dole's reagent [2-propanol/heptane/0.5 M H₂SO₄ (20:5:1)] to which 1 ml of heptane and 1 ml of water containing 20 µg of nonradiolabeled AA as carrier were added. The contents of the tube were mixed, and the heptane layer was applied to a Sep-Pak 3cc silica column (Waters; Milford, MA) for separation of radiolabeled fatty acid. Each column was washed further with 1 ml heptane, the eluates air-dried and radioactivity measured by liquid scintillation spectroscopy.

Assay of MAG and DAG. MAG and DAG levels were measured with a modification of the method described previously (Weis and Malik, 1989). Cells were incubated for 18 hr in 100-mm culture dishes with culture medium containing [³H]AA (1 µCi/ml; 100 Ci/mmol). The labeled cells were washed twice with HBSS, and fresh serum-free DMEM containing various antagonists (RHC80267, propranolol) was added. After 30 min, cells were exposed to PHE for an additional 15 min, and the reaction terminated by adding 2 ml of ice-cold methanol/2 M HCl (50:50; v/v). The cells were scraped from the culture dish and 1 ml of chloroform was added to each tube, mixed and centrifuged at 1,000 × g for 10 min. After lipid extraction, a 40-µl aliquot was removed to estimate radioactivity in the total lipid fraction. The lipid classes were separated by TLC in two solvent systems (S1, pentane/ethyl ether/methanol/acetic acid, 110:20:10:1, developed to 8 cm and then dried under nitrogen; S2, petroleum ether/ethyl ether/acetic acid, 168:30:1, developed to 15 cm). MAG and DAG zones on TLC were determined by authentic standards (Sigma) and the corresponding areas eluted with methanol. Radioactivity was measured by liquid scintillation spectroscopy; [³H]MAG and [³H]DAG production was expressed as the fraction of total cellular ³H-labeled lipids.

PKA assay. Cells for PKA activity measurement were lysed with lysis buffer (pH 7.4) containing (in mM): Tris-HCl, 25; NaCl, 150; ethylenediaminetetraacetic acid, 1.0; EGTA, 1.0; PMSF, 0.3; leupeptin, 0.04; aprotinin, 0.02; and 0.05% BSA. Activation of PKA was measured with a nonradioactive detection kit (Promega Corp.; Madison, WI). The density of the bands was measured with an IS-1000 Digital Imaging System (Alpha Innotech Corp.; San Leandro, CA).

Data analysis. The results are expressed as means ± S.E. The data were analyzed by one-way analysis of variance; the Newman-Keuls multiple range test was applied to determine the difference among multiple groups; and the unpaired Student's t-test was applied to determine the difference between two groups. The null hypothesis was rejected at P < .05. Although the basal release of [³H]AA was variable in different batches of cells, the effect of agonists to increase release of [³H]AA was consistent within each batch of cells. Therefore, the increase in fractional [³H]AA release elicited by agonists was expressed as percentage above basal level. PLD and PLA₂ activity were expressed as the fold increase from basal.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Binding of [³H]prazosin to alpha-1A, alpha-1B and alpha-1D ARs in R-1Fs. Binding studies with [³H]prazosin showed a single saturable binding site for each of the alpha-1 AR subtypes (alpha-1A, alpha-1B and alpha-1D) in R-1Fs. The values of the maximum number of binding sites (B_max) and the affinity K_d of [³H]prazosin, determined by Scatchard analysis, respectively, were: alpha-10 AR (transfected with plasmid containing vector alone for negative control), 25.8 ± 3.7 fmol/mg protein and 0.31 ± 0.09 nM; alpha-1A AR, 288 ± 2 fmol/mg protein and 0.18 ± 0.07 nM; alpha-1B AR, 799 ± 123 fmol/mg protein and 0.34 ± 0.05 nM; alpha-1D AR, 138 ± 13 fmol/mg protein and 0.089 ± 0.03 nM.

Effects of PHE on AA release and cAMP accumulation in R-1Fs transfected with alpha-1A, alpha-1B or alpha-1D ARs. PHE produced a consistent concentration-dependent increase in the release of AA (fig. 1A) and cAMP (fig. 1B) in R-1Fs expressing alpha-1A, alpha-1B or alpha-1D ARs. PHE did not alter either AA release or cAMP accumulation in cells transfected with empty vector (as negative control, alpha-10). The effect of PHE to increase AA release and cAMP accumulation was much greater in cells expressing the alpha-1A AR than in those expressing alpha-1B or alpha-1D ARs. In cells expressing alpha-1D AR, AA release was increased by PHE only at 1 µM, whereas cAMP was increased at 1 and 10 µM PHE. The PHE-induced increase in AA release and cAMP accumulation was abolished by the alpha-1 AR antagonist prazosin (1 µM), but not by the beta AR antagonist timolol (1 µM) in cells expressing alpha-1A, alpha-1B or alpha-1D (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Effects of PHE-stimulated [3H]AA release (A) and cAMP production (B) in different subtypes of alpha-1 ARs expressed in R-1Fs. Cells were incubated in DMEM containing 0.2% BSA at 37°C for [3H]AA release or medium containing IBMX (10 µM) for cAMP determination and exposed to PHE for 15 min as described under "Experimental Protocols." [3H]AA release in the medium and cAMP accumulation in the cells were determined. The total amount of radioactivity ([3H]AA) incorporated in R-1Fs/well expressing alpha-1A, alpha-1B, alpha-1D and alpha-10 was 22.1 to 27.6 × 103 cpm. Data are shown as means ± S.E. of 9 to 12 wells of cells. * Denotes value significantly different from basal (P < .05).

Effects of PHE on PLD and PLA₂ activity in the R-1Fs transfected with different alpha AR subtypes. PHE increased PLD activity in cells transfected with each of the three subtypes of alpha-1 ARs, but not in those transfected with plasmid alone (alpha-10) (fig. 2A). The increase in PLD activity elicited by PHE was greater in cells expressing alpha-1A than in alpha-1B and alpha-1D AR expressing R-1Fs. On the other hand, PHE did not significantly increase PLA₂ activity in these AR-receptor-expressing cells, although endothelin-1 did enhance PLA₂ activity in cells transfected with alpha-1A but not alpha-1B or alpha-1D AR (positive control) (fig. 2B).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effects of PHE and endothelin-1 on the activity of PLD (A) and PLA2 (B) in R-1Fs expressing alpha-1 ARs. Cells were prelabeled with [3H]oleic acid for measurement of PLD. Activity of PLD (from accumulation of [3H]PEt) and PLA2 (from hydrolysis of [14C]arachidonyl-labeled phosphatidylcholine) were measured as described under "Experimental Protocols." Data are shown as means ± S.E. from three dishes of cells. *Denotes value significantly different from the basal (P < .05).

Effects of PLD, PPH and DAG lipase inhibitors on PHE-induced AA release and PLD activity in R-1Fs transfected with the alpha-1A AR. Because PHE produced a greater increase in PLD activity and cAMP levels in cells expressing the alpha-1A AR, we performed further studies with these cells to determine the possible contribution of PLD to receptor-induced AA release, and its relationship to the increase in cAMP levels elicited by PHE. To evaluate the role of the PLD pathway in receptor-induced AA release, we examined the effect of C₂-ceramide, an inhibitor of PLD activation (Gomez-Munoz et al., 1995) and an inactive analog C₂-dihydroceramide (Hannun, 1994); of propranolol, a beta AR antagonist known to inhibit PPH activity (Pappu and Hauser, 1983); and of the DAG lipase inhibitor RHC 80267 (Sutherland and Amin, 1982) on PHE-stimulated AA release in R-1Fs expressing the alpha-1A AR. All of these inhibitors but not C₂-dihydroceramide, an inactive analog of C₂-ceramide (data not shown), decreased AA release elicited by PHE (fig. 3A). PHE-induced PLD activity also was reduced by C₂-ceramide, but not by C₂-dihydroceramide (data not shown), RHC 80267 or propranolol (fig. 3B). Neither C₂-ceramide, RHC 80267 nor propranolol altered basal AA release or PLD activity (fig. 3, A and B). PHE also increased PLD activity and it also was inhibited by C₂-ceramide (P < .05) but not by C₂-dihydroceramide (P > .05) in cells expressing alpha-1A AR and labeled with [³H]AA instead of [³H]oleic acid for measuring [³H]PEt (basal = 2.81 ± 0.86 [³H]PEt/[³H]total lipid, 5.42 ± 0.57 fold increase in the presence of vehicle; basal = 2.02 ± 0.23 [³H]PEt/[³H]total lipid, 2.85 ± 0.59 fold increase with the presence of C₂-ceramide; and basal = 3.14 ± 0.10, [³H]PEt/[³H]total lipid 5.74 ± 0.37 fold increase in the presence of C₂-dihydroceramide, n = 3 with each agent). In alpha-1A-AR-expressing R-1Fs the uptake of [³H]AA (1 µCi, 100 Ci/mmol) (89.7 ± 0.8%, 334.4 ± 7.4 × 10³ cpm total/well and 300.2 ± 7.5 × 10³ cpm in cells/well) and [³H]oleic acid (1 µCi, 50 Ci/mmol) (90.8 ± 0.8%, 540.5 ± 9.2 × 10³ cpm total/well and 491.3 ± 10.0 × 10³ cpm in cells/well) was not different (n = 6 in each group, P > .05). Moreover, most of the radioactivity was found in phospholipids, primarily in zones on TLC plates corresponding to authentic phosphatidylcholine, in cells labeled with [³H]AA (93%) or [³H]oleic acid (92%) and the rest in triglycerides (n = 2 with each fatty acid).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of inhibitors of PLD (C2-ceramide), PPH (propranolol) and DAG lipase (RHC 80267) on PHE-induced increase in AA release (A) and PLD activity (B) in alpha-1A-AR-expressing R1-Fs. Cells prelabeled with [3H]AA (A) or [3H]oleic acid (B) were preincubated with C2-ceramide for 4 hr or propranolol and RHC 80267 for 30 min at 37°C. Cells were then incubated with DMEM containing PHE for 15 min as described under "Experimental Protocols." Basal values of [3H]AA and PLD are presented beneath each group. The total amount of radioactivity ([3H]AA) incorporated in R-1Fs/well expressing alpha-1A AR was 26.0 ± 2.6 × 103 cpm, 20.6 ± 2.1 × 103 cpm, 23.1 ± 2.7 × 103 cpm and 26.6 ± 2.3 × 103 cpm, in vehicle-, ceramide-, propranolol- and RMC 20867-treated groups, respectively. Data are expressed as mean ± S.E. of nine wells of cells for [3H]AA release, and three dishes of cells for assay of PLD activity. * Denotes value significantly different from the basal;  indicates value significantly different from that obtained in the presence of the vehicle (P < .05).

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effect of inhibitors of PPH (propranolol) and DAG lipase (RHC80267) on PHE-induced increase in DAG content in alpha-1A-AR-expressing R1-Fs. Cells prelabeled with [3H]AA were preincubated with propranolol or RHC80267 for 30 min. Cells were then incubated with DMEM containing PHE for 15 min as described under "Materials and Methods." The cpm values under each column represent the radioactivity corresponding to the authentic DAG standard on TLC plates. Data are presented as the ratio [3H]DAG/[3H]total lipid. Data are expressed as mean ± S.E. of three dishes of cells. *Denotes value significantly different from the vehicle;  indicates value significantly different from that obtained in the presence of PHE alone (P < .05).

Effects of forskolin, cpt-cAMP and the adenylyl cyclase inhibitor SQ 22536 on PHE-induced AA release and PLD activity. Forskolin, an activator of the catalytic subunit of adenylyl cyclase alone, increased the cAMP accumulation and potentiated cAMP production elicited by PHE more than 40-fold; the effect of PHE to increase cAMP and that of forskolin to potentiate its effect on cAMP accumulation were attenuated by SQ 22536, an adenylyl cyclase inhibitor (Fabbri et al., 1991) (fig. 5). PHE-stimulated AA release (fig. 6A) and PLD activity (fig. 6B) were reduced by forskolin and by the nonhydrolyzable cAMP analog cpt-cAMP. Forskolin and cpt-cAMP did not alter basal AA release or PLD activity (fig. 6, A and B). SQ 22536 potentiated the PHE-induced increase in AA release and PLD activity and prevented the inhibitory effect of forskolin but not that of cpt-cAMP on the increase in AA release and PLD activity elicited by PHE (fig. 6, A and B).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Effect of adenylyl cyclase inhibitor, SQ 22536, and its vehicle on PHE-, forskolin- and PHE + forskolin-induced cAMP production in R-1Fs expressing alpha-1A ARs. The cells were incubated in DMEM containing IBMX (10 µM) and SQ 22536 for 30 min at 37°C. PHE, forskolin or PHE + forskolin was then added and incubation continued for 15 min. The reaction was stopped by removing the medium and adding 1 ml sodium acetate (50 mM, pH 4.2), and cAMP was determined by radioimmunoassay as described under "Experimental Protocols." Data are expressed as mean ± S.E. of nine wells of cells. *Denotes value significantly different from the basal;  indicates value significantly different from that obtained in the presence of the vehicle of SQ 22536; § indicates value significantly different from that obtained in the presence of PHE + forskolin alone (P < .05).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Effect of SQ 22536 and its vehicle on PHE-stimulated AA release (A) and PLD activity (B) in the presence of forskolin, cpt-cAMP or their vehicle in R-1Fs expressing the alpha-1A AR. Cells were prelabeled with [3H]AA (A) or [3H]oleic acid (B) and preincubated with SQ 22536 or vehicle, forskolin or cpt-cAMP for 30 min at 37°C. Cells were then stimulated with PHE in DMEM for 15 min as described under "Experimental Protocols." The total amount of radioactivity ([3H]AA) incorporated in R-1Fs/well expressing alpha-1A AR in the absence of SQ 22536 was 16.8 ± 1.8 × 103 cpm, 12.9 ± 1.4 × 103 cpm and 2.1 ± 1.6 × 103 cpm, and in the presence of SQ 22536 was 16.3 ± 1.3 × 103 cpm, 13.9 ± 1.3 × 103 cpm and 22.0 ± 2.0 × 103 cpm, in vehicle-, forskolin- and cpt-cAMP-treated groups, respectively. Basal values of [3H]AA and PLD measured as [3H]PEt/[3H]total lipids are presented beneath each group. Data are expressed as means ± S.E. of nine wells of cells for [3H]AA release and three dishes of cells for the assay of PLD activity. * Denotes value significantly different from the basal;  denotes value significantly different from that obtained in the presence of the vehicle of forskolin and cpt-cAMP; indicates value significantly different from that obtained in the presence of the vehicle of SQ 22536 (P < .05).

Effects of cholera toxin and pertussis toxin on the action of PHE and A-23187 on cAMP accumulation, AA release and PLD activity in R-1Fs expressing alpha-1A AR. Cholera toxin, but not pertussis toxin, enhanced PHE-induced cAMP accumulation (>20 times) in R-1Fs expressing the alpha-1A AR. Basal cAMP levels were also increased by cholera toxin (>17 times), but not by pertussis toxin (fig. 7). The calcium ionophore, A-23187, did not alter cAMP levels (data not shown). Cholera toxin and pertussis toxin significantly reduced PHE-stimulated AA release and PLD activity in R-1Fs (fig. 8, A and B). In contrast, A-23187-induced AA release (from basal 2.9 ± 0.2% fractional release to 27.6 ± 1.1% fractional release; n = 9, P < .05) or PLD activity (from basal 4.3 ± 0.4 × 10³ [³H]PEt/[³H]total lipids to 23.5 ± 1.0 × 10³ [³H]PEt/[³H]total lipid; 2.7 ± 0.2 fold increase; n = 3, P < .05) was not altered by either cholera toxin or pertussis toxin.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Effect of PHE and vehicle on cAMP accumulation in R-1Fs expressing the alpha-1A AR and pretreated with pertussis toxin or cholera toxin or their vehicle for 4 hr. After washing twice with HBSS, the medium was replaced with fresh DMEM containing the above toxins or vehicle and the cells exposed to PHE for 15 min as described under "Experimental Protocols." Data are shown as mean ± S.E. of nine wells of cells. * Denotes value significantly different from the corresponding vehicle of PHE,  denotes value significantly different from that obtained in the presence of vehicle of PHE and in the absence of toxins, and § indicates value significantly different from the corresponding value obtained with PHE in the absence of toxins (P < .05).

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Effect of PHE on AA (A) and PLD activity (B) in the presence of cholera toxin, or pertussis toxin or their vehicle in R-1Fs expressing alpha-1A AR. Cells were prelabeled with [3H]AA (A) or [3H]oleic acid (B) and treated with toxins or their vehicle for 4 hr as described under "Experimental Protocols." After washing twice with HBSS, the medium was replaced with fresh DMEM containing the cholera or pertussis toxin or vehicle, and the cells were exposed to PHE for 15 min. Data are shown as means ± S.E. of nine wells of cells for [3H]AA release and three dishes of cells for the assay of PLD activity. The total amount of radioactivity ([3H]AA) incorporated in R-1Fs/well expressing alpha-1A AR: 19.5 ± 2.2 × 103 cpm, 18.8 ± 2.6 × 103 cpm and 19.2 ± 2.2 × 103 cpm, in vehicle, cholera toxin and pertussis toxin group, respectively. * Denotes value significantly different from the basal and  denotes a value significantly different from that obtained in the presence of the vehicle (P < .05).

Effects of the PKA inhibitor KT 5720 on PHE-induced activation of PKA and on the increase in AA release and PLD activity in R-1Fs expressing the alpha-1A AR. PHE increased PKA activity, which was reduced by the PKA inhibitor, KT 5720 (Kase et al., 1992) (fig. 9). KT 5720 also potentiated the PHE-induced increase in AA release and PLD activity, and prevented the inhibitory effect of forskolin or cpt-cAMP on agonist-stimulated AA release and PLD activity without altering basal AA release or PLD activity (fig. 10, A and B).

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Effect of PHE on PKA activity in the presence of KT 5720 or vehicle in R-1Fs expressing the alpha-1A AR. Cells were preincubated in DMEM with KT 5720 at 37°C for 30 min. PHE or its vehicle was then added, and the cells were incubated for an additional 15 min as described under "Experimental Protocols." Data are shown as mean ± S.E. of four dishes of cells. * Denotes value significantly different from the vehicle and  denotes a value significantly different from that obtained in the presence of PHE alone (P < .05).

View larger version (25K): [in this window] [in a new window]   Fig. 10.   Effect of PKA inhibitor, KT 5720, and its vehicle on PHE-stimulated AA release (A) and PLD activity (B) in R-1Fs expressing the alpha-1A AR in the presence of forskolin, cpt-cAMP or their vehicle. Cells were prelabeled with [3H]AA (A) or [3H]oleic acid (B) and pretreated with KT 5720 or its vehicle and forskolin, cpt-cAMP or their vehicle for 30 min at 37°C. Cells were then exposed to PHE in DMEM for 15 min as described under "Experimental Protocols." Basal values of [3H]AA and PLD measured as [3H]PEt/[3H]total lipids are presented beneath each group. Data are expressed as means ± S.E. of nine wells of cells for [3H]AA release and three dishes of cells for the assay of PLD activity. * Denotes value significantly different from the basal;  denotes value significantly different from that obtained in the presence of the vehicle of forskolin and cpt-cAMP; denotes value significantly different from that obtained in the presence of the vehicle of KT 5720 (P < .05).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study demonstrates that all alpha-1 AR subtypes (alpha-1A, alpha-1B and alpha-1D) are coupled to PLD and that activation of these receptors increases AA release and cAMP levels in R-1Fs. Alpha-1A ARs seem to be coupled more effectively to AA release, PLD activity and the cAMP system than either alpha-1B or alpha-1D ARs. This conclusion is based on our finding that the alpha-1 AR agonist PHE produced a much greater increase in AA release, PLD activity and cAMP accumulation in R-1Fs expressing the alpha-1A AR than in those expressing alpha-1B or alpha-1D ARs. The order of increase in AA release, PLD activity and cAMP accumulation in response to PHE in R1-Fs expressing the different subtypes of alpha-1 ARs was found: alpha-1A > alpha-1B > alpha-1D AR. The smaller increase in AA release, PLD activity and cAMP in response to PHE in cells expressing alpha-1B ARs was not caused by decreased receptor expression because the density (B_max) of alpha-1B ARs was three times greater than that of alpha-1A AR as determined by binding of the alpha-1 AR antagonist [³H]prazosin. The smaller increase in AA release, PLD activity and cAMP in response to PHE in R-1Fs expressing alpha-1D AR could be caused by decreased expression of these receptors. The alpha-1A AR has also been reported to be coupled more efficiently to PLC than the alpha-1B AR in HeLa or COS-7 cells (Schwinn et al., 1991). On the other hand, in CHO cells expressing alpha-1B and alpha-1D ARs, the activation of PLC (as indicated by the hydrolysis of phosphoinositides) depended on the type of agonist used (Perez et al., 1993). Alpha-1B and alpha-1D ARs expressed in COS-1 and CHO cells have been shown to be coupled effectively to PLA₂ and involved in agonist-induced AA release (Perez et al., 1993). However, in R-1Fs expressing alpha-1A, alpha-1B or alpha-1D ARs, PHE produced only a small but insignificant increase in PLA₂ activity, whereas endothelin-1 was able to increase PLA₂ activity in cells expressing alpha-1A but not alpha-1B or alpha-1D AR. The inability of PHE to activate PLA₂ in alpha-1A receptor subtype-expressing cells could be caused by the lower affinity of PHE for these receptors in R-1Fs, because epinephrine (2 µM) was able to increase PLA₂ activity by only 1- to 1.5-fold vs. 6- to 7-fold increase in PLD activity above basal level. Epinephrine was more potent (ED₅₀, 0.1 µM) than PHE (ED₅₀, 0.5 µM) in stimulating PLD activity (Parmentier, J. H. and Malik, K. U., unpublished work). Although PHE did not increase PLA₂ activity in R-1Fs expressing the alpha-1 AR subtypes, it did increase AA release in these cells. AA released by PHE in R-1Fs could result from the activation of PLC. However, D-609, which inhibited PLC activity, produced only a small decrease in AA release (Parmentier, J. H. and Malik, K. U., unpublished observations). Therefore, the principal pathway mediating the receptor-induced release of AA in these cells seems to be via the activation of PLD. Activation of PLD promotes the breakdown of phosphatidylcholine into PA, which is then metabolized by PPH into DAG and subsequently by DAG lipase into AA (Exton, 1990).

Our demonstration that a PLD inhibitor, C₂-ceramide (Gomez-Munoz et al., 1995), but not inactive analog, C₂-dihydroceramide (Hannun,1994), attenuated both the increase in PLD activity and AA release elicited by PHE suggests the involvement of PLD in alpha-1A-stimulated AA release in R-1Fs. The inhibitory effect of C₂-ceramide on PHE-induced increase in PLD activity was independent of the fatty acid used to label the cells for PLD measurement. For example, in alpha-1A AR expressing cells labeled with [³H]AA, PHE also increased [³H]PEt production which was inhibited by C₂-ceramide but not by the inactive analog C₂-dihydroceramide. Whether C₂-ceramide inhibits PLD activity by activating a specific kinase and/or phosphatase (Liu et al., 1994; Hannun, 1994) or by inhibiting translocation of small GTP-binding proteins to the membrane (Abousalham et al., 1997) required for PLD activation is not known. The longer exposure of cells (4 hr) required to inhibit PLD activity (Gomez-Munoz et al., 1995; Venable et al., 1994) raises the possibility that C₂-ceramide might act by reducing the availability of phosphatidylcholine to PLD by competing with phosphocholine synthesis and promoting the formation of sphingomyelin. That PHE-induced AA release in R-1Fs is mediated via PLD activation is further supported by our findings that propranolol, which inhibits PPH (Pappu and Hauser, 1983), and RHC 80267, which inhibits DAG lipase (Sutherland and Amin, 1982), both attenuated PHE-induced AA release in R-1Fs expressing the alpha-1A AR. The inhibition of PHE-induced AA release by propranolol and RHC-80267 was not the result of an alteration in PLD activity, because these agents did not alter the PHE-induced increase in PLD activity but inhibited the increase in DAG and MAG production, respectively, elicited by PHE.

In addition to releasing AA and increasing PLD activity, we also demonstrate that stimulation of alpha-1A ARs with PHE in R-1Fs increased cAMP accumulation. The increase in cAMP levels elicited by the activation of alpha-1A ARs in COS-7 and HeLa cells (Schwinn et al., 1991) and the activation of alpha-1B and alpha-1D ARs in COS-1 cells (Perez et al., 1993) has been reported previously not to be caused by a direct receptor-mediated activation of adenylyl cyclase, but rather by PKC activation. However, in R-1Fs expressing alpha-1A ARs, the PKC inhibitor bisindolylmaleimide did not alter the PHE-induced increase in cAMP levels; similarly, the PKC activator phorbol 12-myristate 13-acetate also did not alter cAMP levels (Ruan, Y., Parmentier, J. H. and Malik, K. U., unpublished observations). Recently, it was reported that norepinephrine activation of alpha-1B ARs expressed in CHO cells increased cAMP levels via the stimulation of G_s (Horie et al., 1995). In the present study, PHE also appears to increase cAMP levels in R-1Fs expressing the alpha-1A AR via the direct receptor activation of adenylyl cyclase because: 1) forskolin, which enhanced cAMP accumulation, also potentiated the effect of PHE to increase cAMP levels, 2) the adenylyl cyclase inhibitor SQ 22536 (Fabbri et al., 1991) attenuated the effect of PHE and minimized the potentiation by forskolin of the PHE-induced increase in cAMP levels; and 3) cholera toxin, which promotes ADP-ribosylation of G_s, markedly increased the cAMP accumulation elicited by PHE. In R-1Fs expressing the alpha-1A AR, pertussis toxin, which inactivates G_i proteins, failed to alter either the basal or PHE-induced increase in cAMP. However, it inhibited the receptor-mediated release of AA and increase in PLD activity elicited by PHE, which indicates the involvement of pertussis toxin-sensitive G_i in the activation of PLD activity in R-1Fs; the coupling of a pertussis toxin-sensitive G protein to PLD was demonstrated previously in HL-60 granulocytes (Pai et al., 1988).

An important finding in the present study is that cholera toxin also inhibited both the release of AA and increase in PLD activity elicited by PHE in R-1Fs expressing alpha-1A AR, probably because of the increased accumulation of cAMP. That the PHE-induced increase in AA is inhibited by the associated rise in cAMP was also suggested by our finding that the adenylyl cyclase inhibitor SQ 22536 (Fabbri et al., 1991) enhanced the PHE-induced increase in AA and minimized the inhibitory effect of forskolin (but not of cpt-cAMP) on AA release elicited by PHE. SQ 22536 also increased PHE-induced PLD activity and prevented the inhibitory effect of forskolin (but not of cpt-cAMP) on the increase in PLD activity elicited by PHE; this suggests that cAMP generated via alpha-1A AR activation inhibits AA release by attenuating the activity of PLD. Whether cAMP also alters the activity of PPH and DAG lipase remains to be determined. Dibutyryl cAMP or agents that increase cAMP accumulation have inhibited the increase in PLD activity elicited by formyl-methionyl-leucyl-phenylalanine in neutrophils (Tyagi et al., 1991). However, in endothelial cells, cAMP has increased thrombin-stimulated PLD activity (Garcia et al., 1992). PKA activation has also increased PLD activity elicited by vasopressin in rat hepatocytes (Gustavsson et al., 1994). On the other hand, agents that alter cAMP levels did not affect the phorbol ester-stimulated increase in PLD activity in neutrophils (Pai et al., 1988; Agwu et al., 1991). Whether these differential effects of cAMP on PLD activity in various cell systems are caused by differences in the type of PLD or in the mechanisms regulating PLD activity is not known (Hammond et al., 1995; Colley et al., 1997). Some reports indicate that PLD in various cell systems differs in terms of its requirements for divalent cations and phosphatidylinositol 4,5-bisphosphate, pH optimum, subcellular localization and degree of activation by small G proteins (Hammond et al., 1995).

In the present study, the mechanism by which cAMP inhibits AA release and PLD activity in R-1Fs expressing the alpha-1A AR seems to depend on PKA activation because: 1) PHE increased PKA activity; an effect inhibited by the PKA inhibitor KT 5720; 2) the PKA inhibitor attenuated the PHE-induced increase in PKA activity and enhanced the effect of PHE to increase AA release and PLD activity; and 3) KT 5720 minimized the inhibitory effect of forskolin and cpt-cAMP on PHE-induced AA release and PLD activity. Whether PKA decreases PLD activity and AA release by promoting phosphorylation of PLD or a G protein coupled to PLD and/or by activating a protein phosphatase that causes dephosphorylation of PLD remains to be determined. Activation of PLD promotes stimulation of PKC via the generation of DAG (Exton, 1996); PKC increases PLD activity (Exton, 1990) and AA release (Emilsson and Sundler, 1986); and PKC activity is inhibited by cAMP in some cell systems (Kroll et al., 1988). It is possible, therefore, that the PHE-induced increase in cAMP might attenuate PLD activity and AA release by inhibiting PKC activity. Alternatively, cAMP could reduce PLD activity by interfering directly or indirectly with the action of one or more of the small G proteins required for PLD activation (Singer et al., 1995; Bowman et al., 1993; Malcolm et al., 1994; Brown et al., 1995; Jiang et al., 1995).

In conclusion, activation with PHE of alpha-1A, alpha-1B or alpha-1D ARs expressed in R-1Fs results in an increase in PLD, but not PLA₂, and promotes AA release and cAMP accumulation. The alpha-1A AR seems to be coupled more effectively to PLD activation, AA release and cAMP accumulation than alpha-1B or alpha-1D ARs in R-1Fs. Moreover, cAMP generated in response to activation of alpha-1A ARs with PHE in R-1Fs leads to an inhibition of PLD activity by inducing the activation of PKA.