Introduction

Top Summary Introduction Experimental Procedures Results Discussion References

Various receptors transduce signals through heterotrimeric G proteins of the G_q family, resulting in activation of phospholipase C (PLC)- isoenzymes and subsequent cleavage of membrane phosphatidylinositol(4,5)P₂ [PtdIns(4,5)P₂] to the second messengers inositol(1,4,5)P₃ [Ins(1,4,5)P₃] and diacylglycerol (Berridge and Irvine, 1987). Ins(1,4,5)P₃ initiates release of calcium from intracellular stores, and diacylglycerol, in conjunction with calcium and phospholipids, activates protein kinase C (PKC).

Agonist-induced desensitization is an important regulatory process in inositol lipid signaling (Fisher, 1995), although the precise mechanisms involved are unclear. Receptor-promoted desensitization is mimicked in several cell types by activation of protein kinase C (Orellana et al., 1985; Rittenhouse and Sasson, 1985) and is prevented by down-regulation of PKC by chronic exposure to 4-phorbol-12-myristate-13-acetate (PMA; Hepler et al., 1988) or by PKC inhibitors (Galas and Harden, 1995). These observations strongly suggest that PKC plays a negative regulatory role in phosphoinositide hydrolysis, probably by causing a phosphorylation-dependent change in the activity of another component of the pathway. Potential targets of PKC-catalyzed phosphorylation in this putative negative-feedback system include the receptor, the G protein, the effector enzyme PLC-, or other less well-defined membrane-signaling proteins. GTPS-stimulated PLC- activity is attenuated in membranes isolated from PMA-treated cells (Orellana et al., 1987), suggesting that PKC acts on a protein downstream of the receptor.

The turkey erythrocyte is a well-characterized model of receptor-promoted inositol phospholipid signaling and is particularly useful because the three primary proteins in the pathway, i.e., the turkey P2Y₁ receptor (Boyer et al., 1989; Filtz et al., 1994), the G protein G₁₁ (Waldo et al., 1991; Maurice et al., 1993), and the effector enzyme PLC-t (Morris et al., 1990a,b; Waldo et al., 1996), have been identified and cloned. Desensitization of phosphoinositide hydrolysis has been described in the turkey erythrocyte. Both indirect activation of PKC by P2Y-receptor agonists and direct PKC activation by PMA attenuate agonist and GTPS-stimulated PLC-t activity in turkey erythrocyte membranes (Martin and Harden, 1989; Galas and Harden, 1995), suggesting that regulation of inositol phospholipid signaling occurs downstream of the turkey P2Y₁ receptor. Previous studies of desensitization in the turkey erythrocyte membrane assessed receptor and G protein-stimulated inositol lipid hydrolysis catalyzed by endogenous PLC-t activity, making it difficult to distinguish between effects of PKC on PLC-t and those on G₁₁ or other membrane-signaling proteins. Similarly, previous studies with mammalian cell preparations have failed to localize desensitizing effects of PKC activation to PLC-, to the involved G_q family G protein, or to other less well-defined components of the inositol lipid-signaling pathway. In this study, we have established a turkey erythrocyte membrane assay with reconstituted, purified PLC-1 that provides direct quantification of the activity of G₁₁. We have used this assay to demonstrate for the first time that G₁₁-stimulated activation of a PLC- isoenzyme is inhibited in membranes from cells treated with PMA.

	Experimental Procedures

Top Summary Introduction Experimental Procedures Results Discussion References

Materials. Recombinant PLC-1 was purified after baculovirus-promoted expression in Sf9 insect cells as previously described (Paterson et al., 1997). PtdIns(4)P was purified from bovine brain Folch fraction as described previously (Morris et al., 1990a). [³H]PtdIns(4)P was purified from [³H]myo-inositol-labeled turkey erythrocytes as described previously (Waldo et al., 1994). Phosphatidylethanolamine (bovine heart), 1,2-dioleoyl-sn-glycerol, and phosphatidylserine (brain) were purchased from Avanti Polar Lipids (Alabaster, AL). GTPS, protein A-agarose, and PKC purified from rat brain were purchased from Boehringer Mannheim (Indianapolis, IN). PMA, 4-phorbol-12,13-didecanoate, and phorbol-12,13-dibutyrate were obtained from Sigma (St. Louis, MO). 2-Methylthioadenosine triphosphate (2-MeSATP) was purchased from Research Biochemicals Inc. (Natick, MA). [³²P]Orthophosphate and [-³²P]ATP were purchased from DuPont NEN (Boston, MA).

Turkey Erythrocyte Treatment and Membrane Preparation. Turkey erythrocytes were collected and washed as previously described (Boyer et al., 1989). For treatment with phorbol esters, 1 ml of packed erythrocytes was resuspended in 4 ml HEPES-Dulbecco's modified Eagle's medium (DMEM), pH 7.4, and equilibrated at 37°C for 5 min. Phorbol esters were added as stocks in 100% ethanol, and an equal volume (~0.1% of total volume) of this solvent was always used as a vehicle control. Drug treatments proceeded at 37°C for 20 min unless otherwise indicated. Treated erythrocytes were centrifuged at 500g, the medium was aspirated, and the cells were resuspended in ice-cold lysis buffer [20 mM HEPES, pH 7.0, 145 mM NaCl, 5 mM MgCl₂, 50 mM NaF, 1 mM EGTA, 200 µM phenylmethylsulfonyl fluoride (PMSF), and 200 µM benzamidine] at a 1:1 ratio (v/v). Resuspended cells (2 ml) were vortexed vigorously in the presence of 1 g of glass beads (0.45 mm) for four rounds of 30 s each. Erythrocytes were cooled on ice between vortexing, and all subsequent steps were carried out at 4°C. Samples were centrifuged at 500g for 3 min to sediment the glass beads and unlysed cells. The supernatant was collected in a fresh tube, and the glass beads were resuspended in 1 ml of lysis buffer. The vortex and centrifugation steps were repeated to lyse the remaining cells. Glass beads were washed a final time with 1 ml of lysis buffer, vortexed, and allowed to settle for 5 min without centrifugation. The resultant supernatant was added to the previous supernatant fractions. The supernatant pool was centrifuged at 35,000g for 10 min, and the membrane pellet was resuspended in 1 ml of lysis buffer. After two washes in lysis buffer, the membrane pellet was resuspended in 1 ml of membrane buffer [20 mM HEPES, pH 7.4, 1 mM MgCl₂, 100 mM NaCl, 40 mM -glycerophosphate, 2 mM dithiothreitol, 200 µM benzamidine, 200 µM PMSF, and 200 nM calyculin A] and homogenized with DNase I (45 mU/ml) to degrade any residual DNA. Membranes were used immediately in a reconstitution assay or frozen at 80°C for later use.

Reconstitution of Purified PLC-1 with Turkey Erythrocyte Membranes. Receptor and G protein-regulated PLC-1 activity was quantitated with turkey erythrocyte membranes reconstituted with purified PLC-1 and phospholipids. Erythrocyte membranes were prepared as described above, assayed for protein concentration, and diluted in membrane buffer to 25 µg/µl, unless otherwise indicated. Phospholipid substrate was prepared as a mixture of PtdIns(4)P (5 nmol/assay), phosphatidylethanolamine (25 nmol/assay), and [³H]PtdIns(4)P (~10,000 cpm/assay), dried under nitrogen, and resuspended by sonication in membrane buffer (10 µl/assay). Membranes and phospholipids were mixed at a ratio of 3:2 (v/v) and incubated on ice for 30 min. Twenty-five microliters each of radiolabeled membranes, DB buffer, and 20 mM HEPES, pH 7.0; GTPS; or GTPS with 2-MeSATP was mixed and preincubated at 30°C for 2 min. Drugs were prepared as 4X stocks in 20 mM HEPES, pH 7.0. Unless otherwise indicated, the assay was initiated by the addition of 25 µl of purified PLC-1 (3 ng) in 4X enzyme buffer (80 mM HEPES, pH 7.2, 3.2 mM MgCl₂, 12 mM EGTA, 60 mM NaCl, 120 mM KCl, 0.8 mM EDTA, 10.6 mM CaCl₂, and 4 mM dithiothreitol or 4X enzyme buffer alone. The assay was at 30°C for 20 min, or as indicated in the figure legends. The reaction was terminated by the addition of 375 µl of CHCl₃/MeOH/concentrated HCl (40:80:1). CHCl₃ (125 µl) and 0.1 M HCl (125 µl) were added, and the samples were centrifuged at 1800g. Three hundred fifty microliters of the aqueous phase was counted by liquid scintillation spectrometry to quantitate [³H]Ins(1,4)P production. All data are reported as means ± S.D. of triplicate determinations and are representative of two or three experiments.

In Vitro Kinase Reaction. Native G₁₁ from turkey erythrocytes (9 pmol) was incubated with 20 µU of PKC for 30 min at 30°C in a reaction containing 20 mM Tris, pH 7.5, 10 mM MgCl₂, 500 µM CaCl₂, 100 µg/ml phosphatidylserine, 20 µg/ml 1,2-dioleoyl-sn-glycerol, and 200 µM [-³²P]ATP (~1500 cpm/pmol) in a volume of 20 µl. Reactions were terminated by the addition of 20 µl of 2X Laemmli sample buffer. Samples were separated by SDS-polyacrylamide gel electrophoresis through 12.5% acrylamide according to the method of Laemmli (Laemmli, 1970), and the protein bands were visualized by Coomassie stain. The gel was dried and exposed to autoradiography film to detect radioactive bands.

In Vivo [³²P]Orthophosphate Labeling and G₁₁ Immunoprecipitation. Washed turkey erythrocytes were radiolabeled for 4 h in phosphate-free DMEM (4 ml/ml of packed erythrocytes) supplemented with [³²P]P_i (~2 mCi/ml of packed cells) at 37°C in 5% CO₂. Radiolabeled erythrocytes were centrifuged at 500g for 5 min and resuspended in HEPES-DMEM, pH 7.4 (4 ml/ml of packed erythrocytes). Erythrocytes were treated, and membranes were prepared with glass beads as described above. Crude membrane pellets were isolated by centrifugation at 35,000g for 10 min. Membranes were resuspended in cholate extraction buffer (50 mM Tris, pH 7.5, 50 µM AlCl₃, 10 mM MgCl₂, 25 mM NaF, 40 mM -glycerophosphate, 30 µM GDP, 200 µM benzamidine, 200 µM PMSF, and 1.2% cholate) at ~2 mg of protein/ml of extraction buffer and incubated for 2 h at 4°C with mixing. Samples were centrifuged at 100,000g for 30 min, and the supernatants (500 µl) were diluted 1:1 in immunoprecipitation buffer (50 mM Tris, pH 8.0, 2 mM MgCl₂, 100 mM NaCl, 25 mM NaF, 40 mM -glycerophosphate, 200 µM benzamidine, and 200 µM PMSF) and incubated with preimmune serum or anti-G_q/11 antiserum (117, antiserum to a synthetic peptide corresponding to the COOH-terminal sequence of both G_q and G₁₁; 1:100 dilution; Maurice et al., 1993) overnight at 4°C. Immunoprecipitated samples were incubated with protein A-agarose beads for 2 h at 4°C and then centrifuged at 14,000g for 20 s. Immunoprecipitated pellets were washed twice with immunoprecipitation buffer. The washed pellets were resuspended in 70 µl Laemmli sample buffer and incubated for 10 min at 85°C.

	Results

Top Summary Introduction Experimental Procedures Results Discussion References

Activation of PKC by acute PMA treatment has been shown to inhibit agonist-and guanine nucleotide-stimulated inositol phosphate production in various intact cell and membrane assays (Orellana et al., 1985, 1987; Rittenhouse and Sasson, 1985; Hepler et al., 1988; Galas and Harden, 1995). Inositol lipid hydrolysis in these studies was catalyzed by endogenous PLC-, and the relative effects of PMA on G protein versus effector enzyme activity therefore could not be distinguished. To resolve potential PMA effects on the G protein or other membrane-signaling molecules from those on PLC-, we used the turkey erythrocyte model of inositol phospholipid signaling to develop an assay that directly measures G₁₁ activity in situ. In this assay, turkey erythrocyte membranes are depleted of endogenous PLC-t by successive washes, [³H]PtdIns(4)P [or [³H]PtdIns(4,5)P₂] substrate is incorporated into the membranes, and after 2 min incubation with GTPS, purified PLC-1 is reconstituted with the membranes. PLC-1 was selected for this assay because it is much less sensitive to -subunit activation than is PLC-t (Boyer et al., 1992, 1994) and therefore provides a measure of G₁₁-stimulated PLC- activation. [³H]Ins(1,4)P₂ [or [³H]Ins(1,4,5)P₃] production is quantitated after various times of incubation at 30°C with PLC-1.

Washed membranes did not exhibit significant inositol phosphate production in the presence of GTPS or GTPS with 2-MeSATP (Fig. 1A), demonstrating that endogenous PLC-t was removed by the wash procedure. Addition of exogenous purified PLC-1 to membranes resulted in a marked increase in basal phosphoinositide hydrolysis (Fig. 1A). Moreover, whereas GTPS had no effect in washed membranes, significant GTPS-promoted inositol lipid hydrolysis was observed after reconstitution of PLC-1. The small activation observed with GTPS and 2-MeSATP in washed membranes was also markedly augmented after reconstitution of PLC-1 (Fig. 1A). Higher concentrations of membrane protein resulted in increased inositol phosphate production (Fig. 1A).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Characterization of the membrane reconstitution assay. Membrane reconstitution assays were performed as described in Experimental Procedures. A, indicated amounts of membrane protein from vehicle-treated turkey erythrocytes were challenged with 100 µM GTPS or 100 µM GTPS plus 100 µM 2-MeSATP and incubated with 3 ng of purified PLC-1 or enzyme buffer for 20 min at 30°C. B, membranes from control () or 1 µM PMA-treated () turkey erythrocytes were challenged with 100 µM GTPS and incubated with varying concentrations of purified PLC-1. C, membranes from control erythrocytes were stimulated in the presence of 100 µM GTPS and varying concentrations of 2-MeSATP and assayed as described. The data are means ± S.D. of triplicate determinations and are representative of at least three experiments. The levels of [3H]Ins(1,4)P2 production in the absence of PLC-1 were subtracted from the values presented in B.

To establish optimal conditions for assay of G₁₁-stimulated activities, PLC-1 concentration-response and time course experiments were performed. GTPS-stimulated inositol phosphate production increased essentially linearly up to ~5 ng of added PLC-1 (Fig. 1B). After an initial lag of 2 to 3 min, GTPS-stimulated inositol lipid hydrolysis proceeded linearly for up to 30 min (Fig. 2). Incubation with the P2Y₁-receptor agonist 2-MeSATP and GTPS resulted in an increase in inositol phosphate production that was dependent on the concentration of 2-MeSATP (Fig. 1C; EC₅₀ = 25 ± 1 nM; mean ± S.E.; n = 3) and was comparable in agonist potency to that previously observed with turkey erythrocyte ghosts (Boyer et al., 1989). The experiments presented in Figs. 1 and 2 demonstrate that this reconstitution system can be used effectively to assess the capacity of G₁₁ to activate PLC- under conditions in which the enzyme is reconstituted with membranes depleted of endogenous PLC-t.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Time course of GTPS-stimulated inositol phosphate accumulation. Membranes from vehicle () or 1 µM PMA-treated () turkey erythrocytes were assayed for GTPS-stimulated [3H]Ins(1,4)P2 production as described in Experimental Procedures. Membranes were challenged with 100 µM GTPS and incubated with 3 ng of purified PLC-1 for the indicated times. Values are means ± S.D. of triplicate determinations and are representative of three experiments.

The effects of PMA treatment of turkey erythrocytes on G₁₁ activity were studied. Membranes were prepared from control or 1 µM PMA-treated turkey erythrocytes, activated with GTPS, and reconstituted with varying concentrations of PLC-1. PMA treatment inhibited GTPS-stimulated phosphoinositide hydrolysis at each concentration of PLC-1 (Fig. 1B) and time of incubation (Fig. 2) examined. PMA treatment decreased GTPS-stimulated PLC-1 activity in a concentration-dependent manner, with half-maximal inhibition occurring at ~50 nM PMA and maximal suppression at 1 µM (Fig. 3A). GTPS-stimulated phosphoinositide hydrolysis decreased with time of PMA treatment, and maximal inhibition was observed within 20 min of PMA treatment (Fig. 3B). The concentration and time dependence for the inhibitory effects of PMA on G protein-regulated PLC activity were similar to those previously observed in studies with turkey erythrocytes (Galas and Harden, 1995) and with various mammalian cell lines (Orellana et al., 1985, 1987; Rittenhouse and Sasson, 1985; Hepler et al., 1988).

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Concentration and time dependence of PMA-induced inhibition of GTPS-stimulated PLC-1 activity. Turkey erythrocytes were treated with the indicated concentrations of PMA for 20 min (A) or 1 µM PMA for the indicated times (B) at 37°C. Membranes were challenged with 100 µM GTPS and assayed for [3H]Ins(1,4)P2 production as described in Experimental Procedures. Basal levels of [3H]Ins(1,4)P2 were subtracted from the values presented in A. The data in A are means ± S.D. of triplicate determinations and representative of three experiments. The data in B are means ± S.D. from two experiments performed in triplicate.

Experiments were carried out to determine whether PMA decreased the efficacy of GTPS or the concentration of GTPS needed for half-maximal stimulation. In the absence of 2-MeSATP, PMA treatment suppressed maximal GTPS-stimulated activation of PLC-1 by 35 to 50% compared with that of control membranes (Fig. 4A). The EC₅₀ concentration of GTPS was not significantly changed in membranes from PMA-pretreated erythrocytes (control, 7 ± 3 µM versus PMA treated, 13 ± 12 µM; mean ± S.E.; n = 3). Maximal GTPS-stimulated inositol phosphate production in the presence of 2-MeSATP was inhibited 30 to 40% by PMA treatment, with no significant change in the EC₅₀ concentration of GTPS observed (Fig. 4B; control, 27 ± 10 nM versus PMA treated, 37 ± 13 nM,; mean ± S.E.; n = 2). These data indicate that PMA-induced inhibition of G₁₁-stimulated phosphoinositide hydrolysis results from a decrease in the apparent efficacy of GTPS for activation of PLC-1 rather than as a change in apparent affinity for the guanine nucleotide.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Effect of PMA treatment on the efficacy of GTPS for the stimulation of inositol phosphate production in the presence and absence of agonist. Membranes from vehicle-treated () or 1-µM-PMA-treated () turkey erythrocytes were challenged with the indicated concentrations of GTPS alone (A) or GTPS plus 10 µM 2-MeSATP (B) and assayed as described in Experimental Procedures. Values are means ± S.D. of triplicate determinations and are representative of at least two experiments.

Although PMA is a well characterized PKC activator, the observed suppression of G₁₁ stimulation of PLC-1 could originate from a nonspecific effect of phorbol ester treatment. Treatment with phorbol-12,13-dibutyrate, which also activates PKC, resulted in decreases in GTPS and GTPS and 2-MeSATP-stimulated inositol phosphate production similar to that observed in membranes from PMA-treated cells (Fig. 5). Conversely, membranes prepared from erythrocytes treated with 4-phorbol-12,13-didecanoate, which does not activate PKC, displayed receptor- and G protein-stimulated PLC-1 activities equivalent to those of vehicle-treated erythrocytes (data not shown). To further establish the role of PKC in PMA-induced inhibition of G₁₁-activated phosphoinositide hydrolysis, the capacity of the PKC inhibitor bisindolylmaleimide to reverse the PMA-promoted suppression was examined. Membranes from erythrocytes treated with bisindolylmaleimide in the presence of PMA exhibited GTPS-stimulated PLC-1 activity equivalent to that of control cells (Fig. 5).

View larger version (32K): [in this window] [in a new window]   Fig. 5.   PKC involvement in the PMA-induced suppression of G11-stimulated PLC-1 activity. Membranes from turkey erythrocytes treated with vehicle (black columns), 1 µM PMA (open columns), 1 µM phorbol-12,13-dibutyrate (PDBu, hatched columns), or 1 µM PMA plus 10 µM bisindolylmaleimide (gray columns) were challenged with 10 µM GTPS alone or with 10 µM GTPS in the presence of 10 µM 2-MeSATP and assayed for [3H]Ins(1,4)P2 production as described in Experimental Procedures. Basal levels of [3H]Ins(1,4)P2 production were subtracted from the values presented. The data are means ± S.D. of triplicate determinations and are representative of two experiments.

We previously demonstrated that the PLC- present in turkey erythrocytes is a substrate for PKC (Filtz et al., 1999). However, because the experiments presented in this article were carried out with an assay that uses exogenously reconstituted PLC-1 to assess G₁₁-stimulated PLC- activity, effects of PKC on the endogenous avian PLC- do not contribute to the phenomenon presented in Figs. 1 through 5. Therefore, to determine whether the inhibition of guanine nucleotide-stimulated inositol lipid hydrolysis was caused by a phosphorylation-dependent change in the activity of G₁₁, we performed in vitro kinase reactions with purified PKC and native G₁₁ from turkey erythrocytes. No evidence of G₁₁ phosphorylation by PKC was observed under conditions sufficient for PKC to promote stoichiometric phosphorylation of PLC-t (data not shown).

Although PKC did not directly phosphorylate G₁₁ in vitro, we investigated the possibility that G₁₁ was phosphorylated in intact turkey erythrocytes in response to PMA treatment. G₁₁ was immunoprecipitated from [³²P]orthophosphate-labeled turkey erythrocytes treated with vehicle or 1 µM PMA. PMA promoted large increases in [³²P] incorporation into PLC-t under these conditions (Filtz et al., 1999; data not shown), but no increase in [³²P] incorporation into G₁₁ occurred (data not shown). These results are consistent with the hypothesis that a signaling protein present in washed turkey erythrocyte membranes other than G₁₁ is the target for PKC and that PKC-promoted phosphorylation of this protein results in an activity change that is responsible for the inhibition of inositol phospholipid hydrolysis observed with PMA treatment.

The results presented thus far illustrate that activation of PKC in turkey erythrocytes reduces the capacity of G₁₁ to promote inositol lipid hydrolysis. However, as mentioned above, we previously reported that PKC promotes phosphorylation of PLC-t in vivo and that in vitro phosphorylation of PLC-t by PKC results in a decrease in the basal catalytic activity of the enzyme (Filtz et al., 1999). Therefore, experiments were carried out to determine whether these two effects of PKC activation were additive by quantitating G₁₁-regulated PLC- activity after reconstitution of phosphorylated PLC-t with membranes from PMA-treated erythrocytes. As with PLC-1, reconstitution of PLC-t conferred G protein- and receptor-regulated phosphoinositide hydrolysis to washed turkey erythrocyte membranes. Phosphorylation of PLC-t by PKC in vitro resulted in a decrease in its capacity to be activated by addition of GTPS after reconstitution with control membranes (Table 1), in agreement with our previous results (Filtz et al., 1999). Furthermore, as was observed with reconstitution of PLC-1, a decrease in G protein-promoted inositol lipid hydrolysis was observed in membranes from PMA-treated erythrocytes after reconstitution of exogenous purified PLC-t (Table 1). Moreover, GTPS-stimulated PLC- activity observed after reconstitution of phosphorylated PLC-t with PMA-treated membranes was lower than the GTPS-stimulated activity observed after either reconstitution of unphosphorylated PLC-t with membranes from PMA-treated erythrocytes or after reconstitution of phosphorylated PLC-t with control membranes (Table 1). It is somewhat surprising that a greater reduction in activity was not observed by combining the effects of PKC-promoted phosphorylation of PLC-t with PMA treatment of erythrocytes. This result may be partially explained by the greater sensitivity of PLC-t to activation by G. We have previously reported (Filtz et al., 1999) with purified proteins that activation of PLC-t by G surmounts the decrease in basal activity induced by the phosphorylation of PLC-t, and the contribution of G activation of PLC-t may therefore confound these experiments. Although there was not a strong additive effect of PMA treatment and phosphorylation of PLC-t, these experiments suggest that activation of PKC results in modification of both a membrane-signaling protein and the effector enzyme and that this dual modification may be responsible for desensitization observed with PMA treatment of intact cells.

View this table: [in this window] [in a new window]   TABLE 1 Reconstitution of PLC-t with turkey erythrocyte membranes PLC-t was incubated in the presence or absence of 10 µU PKC for 10 min at 30°C as described in Experimental Procedures. Reaction mixtures were diluted in 10 mM HEPES/BSA (1 mg/ml) to a final PLC-t concentration of 20 ng/25 µl. After challenge with 100 µM GTPS, 25 µg of membranes from control erythrocytes were reconstituted with 20 ng of unphosphorylated or phosphorylated PLC-t. Similarly, 25 µg of membranes from erythrocytes treated with 1 µM PMA were reconstituted with unphosphorylated or phosphorylated PLC-t. Samples were incubated for 20 min at 30°C and assayed for [3H]Ins(1,4)P2 production. Ins(1,4)P2 production in the absence of reconstituted PLC-t or GTPS stimulation was 0.19 ± 0.09 and 0.27 ± 0.23 µmol · min1 · mg1 for control and PMA-treated membranes, respectively. Values are means ± S.D. of triplicate determinations and are representative of two experiments.

	Discussion

Top Summary Introduction Experimental Procedures Results Discussion References

Receptor- and G protein-promoted PLC- activity is attenuated in response to short-term PMA treatment of turkey erythrocytes and several mammalian cell types (Orellana et al., 1985; Rittenhouse and Sasson, 1985; Galas and Harden, 1995), which has led to the hypothesis that PKC plays a negative regulatory role in phosphoinositide hydrolysis. Previous studies in the turkey erythrocyte have measured endogenous PLC-t activity, making it difficult to distinguish between effects of PKC on PLC-t and those on G₁₁ or other membrane-signaling proteins. We have developed a membrane reconstitution assay that permits investigation of the effects of PKC on membrane-signaling proteins under conditions that make no assumptions about specific membrane targets. This assay was used to show that activation of PKC in erythrocytes results in an attenuation of G₁₁-stimulated activation of purified PLC-1. Because the assay involves use of exogenous phosphoinositide substrate, reconstitution of purified PLC-1 into membranes from control or PKC-activated erythrocytes, and activation of signaling with GTPS, our results point to a modification of a membrane-signaling molecule other than the receptor, the PLC enzyme, or its substrate. Therefore, we conclude that activation of PKC promotes a change in the activity of G₁₁ or of a protein that directly regulates G₁₁ activity.

Previous studies have demonstrated the phosphorylation of certain G protein  subunits. Carlson et al. (1989) and Lounsbury et al. (1991) reported that G_z is phosphorylated in vitro by PKC and in platelets after PMA treatment. Kozasa and Gilman (1996) demonstrated that G₁₂ is phosphorylated by purified PKC and in National Institutes of Health 3T3-G12 cells treated with PMA. Phosphorylation of both G_z (Fields and Casey, 1995) and G₁₂ (Kozasa and Gilman, 1996) by PKC inhibits their interactions with G protein subunits and thereby induces a functional consequence that may alter cellular signaling activity of these G proteins. Despite the precedent for phosphorylation of G protein  subunits, we observed no phosphorylation of G₁₁ in vitro by PKC or in PMA-treated erythrocytes. These results are consistent with the findings of Kozasa and Gilman (1996) and Lounsbury et al. (1993) for G_q. Although our reconstitution assay localizes the effect of PKC to a protein at or near the level of G₁₁, these results support the conclusion that this membrane-signaling protein is not G₁₁. A G₁₁-signaling cohort is potentially the substrate for PKC, and regulators of G protein signaling (RGS proteins) are strong candidates (Dohlman and Thorner, 1997; Berman and Gilman, 1998). RGS2 has been shown to interact with and inhibit GTPS-promoted signaling by  subunits of the G_q family of G proteins (Heximer et al., 1997). Other RGS proteins, e.g., RGS4 and GAIP (Hepler et al., 1997), also promote GTP hydrolysis by G_q family members and block GTPS-stimulated activation of PLC- by G_q, albeit at significantly higher concentrations than with RGS2. Glick et al. (1998) and Wang et al. (1998) recently reported that PKC-promoted phosphorylation of G_z inhibited the GTPase-activation protein (GAP) activity of a G_z-selective RGSZ1 protein RGS_Z1, thereby demonstrating the sensitivity of an RGS-G protein interaction to protein phosphorylation state. Although the effects of PMA treatment on G₁₁-stimulated activation of PLC-1 were studied under conditions where RGS proteins are unable to act as GAPs, the capacity of some RGS proteins to potentially act as effector anatagonists and inhibit GTPS-promoted signaling leaves open the possibility that PMA induces a phosphorylation-dependent change in the activity of an RGS protein. The role of RGS2 and other RGS proteins in regulating inositol lipid signaling in the turkey erythrocyte and the effects of PKC on RGS activity are currently being examined in our laboratory.

Other putative membrane targets for PKC include G protein  and  subunits. By using PLC-1, which is much less sensitive to G than PLC-t, we have attempted to focus solely on the effects of PKC on G₁₁-promoted PLC- activity. However, phosphorylation of either G or G could alter the affinity of the dimer for G₁₁, perhaps inhibiting signaling by promoting a higher-affinity interaction of G and G₁₁. G₁₂ was previously shown to be phosphorylated in vitro by PKC, whereas G₁, ₂, ₃, and ₇ are apparently not PKC substrates (Morishita et al., 1995; Yasuda et al., 1998). We have not formally investigated the phosphorylation state of G in turkey erythrocytes after PMA treatment. However, no significant change in the apparent affinity for GTPS was observed in concentration-response experiments, suggesting that there is also no change in the affinity of G for G₁₁. Thus, the data from GTPS concentration-response experiments suggest that it is unlikely that G phosphorylation accounts for the attenuation of inositol lipid signaling seen with PKC activation.

PKC-promoted loss of the activity of G₁₁ potentially could be explained by a phorbol ester-induced decrease in membrane association of this protein. For example, Ransnas et al. (1989, 1992) have reported that receptor-promoted activation of G_s in S49 lymphoma cells results in an increase in cytosolic G_s and a corresponding decrease in membrane-associated G protein. Moreover, Mitchell et al. (1993) and Mullaney et al. (1993) previously demonstrated in CHO cells expressing the human m₁ muscarinic receptor that chronic agonist treatment produced a significant loss of membrane G_q/11. In contrast to these findings, Western blot analysis of membranes from PMA-treated turkey erythrocytes revealed no decrease in immunologically detectable G₁₁ compared with the levels observed in control membranes (data not shown). Additionally, equivalent amounts of G₁₁ were immunoprecipitated from identical starting amounts of membrane from control and PMA-treated erythrocytes, further suggesting that the inhibition of G₁₁-mediated activation of PLC-1 in membranes from PKC-activated cells is not the result of changes in membrane-associated G₁₁ protein.

In contrast to the absence of PKC-catalyzed G₁₁ phosphorylation observed in this study, we previously reported that PLC-t is phosphorylated in intact turkey erythrocytes in response to PMA and 2-MeSATP treatment and in vitro by PKC (Filtz et al., 1999). Phosphorylation of PLC-t results in a decrease in basal catalytic activity (Filtz et al., 1999). The results presented here illustrate for the first time that PKC also promotes an alteration of a membrane-signaling protein other than the receptor or G₁₁. Thus, PKC-promoted desensitization of phosphoinositide hydrolysis in avian erythrocytes stems from a dual modification of the effector enzyme and another membrane-signaling protein. The turkey erythrocyte model should continue to be useful in identifying the membrane target for PKC and for investigating further the relative functional consequences of phosphorylation on PLC-t-catalyzed phosphoinositide hydrolysis.