G Protein-mediated Inhibition of Phospholipase C Activity in a Solubilized Membrane Preparation*

In solubilized bovine brain membrane preparations AlF; (20 pM AICls plus 10 mM NaF) and 50 nM guano- sine 5-0-(2-thiotriphosphate) (GTPyS) promoted a rapid but transient inhibition of phospholipase C (PLC) activity. Maximal inhibition was evident within 7 min of incubation, followed by reversal of inhibition. In contrast, 10 p M GTPyS did not induce inhibition of PLC activity but rather produced a time-dependent stimulation of PLC activity. GTPyS-dependent inhi- bition of PLC activity was concentration-dependent with half-maximal inhibition at 1 nM. Inhibition was antagonized by guanosine 5-0-(2-thiodiphosphate (GDPBS). Pertussis toxin delayed the onset of inhibition by GTPyS but did not prevent the inhibitory ef- fect. a,-GTPyS or a,-GDP had little effect on PLC activity. ai-GTPyS and ai-GDP produced a 15% inhi- bition of PLC activity. By subunits did not inhibit basal PLC activity but did attenuate the net degree of inhi- bition due to GTPyS. Inhibition was associated with a decrease in the Ca2+ sensitivity of PLC. Preincubation of membranes with anti-PLC-BI antibody, but not anti- PLC-yl or anti-PLC-31, prevented the GTPyS-me-diated inhibition of PLC. These studies implicate PLC- B1 as an effector system that is under negative modulation by a G protein-dependent mechanism.


G Protein-mediated Inhibition of Phospholipase C Activity in a Solubilized Membrane Preparation*
(Received for publication, November 5, 1992) Irene LitoschS, Inna Sulkholutskaya, and Cong Weng In solubilized bovine brain membrane preparations AlF; (20 pM AICls plus 10 mM NaF) and 50 nM guanosine 5-0-(2-thiotriphosphate) (GTPyS) promoted a rapid but transient inhibition of phospholipase C (PLC) activity. Maximal inhibition was evident within 7 min of incubation, followed by reversal of inhibition. In contrast, 10 p M GTPyS did not induce inhibition of PLC activity but rather produced a time-dependent stimulation of PLC activity. GTPyS-dependent inhibition of PLC activity was concentration-dependent with half-maximal inhibition at 1 nM. Inhibition was antagonized by guanosine 5-0-(2-thiodiphosphate (GDPBS). Pertussis toxin delayed the onset of inhibition by GTPyS but did not prevent the inhibitory effect. a,-GTPyS or a,-GDP had little effect on PLC activity. ai-GTPyS and ai-GDP produced a 15% inhibition of PLC activity. By subunits did not inhibit basal PLC activity but did attenuate the net degree of inhibition due to GTPyS. Inhibition was associated with a decrease in the Ca2+ sensitivity of PLC. Preincubation of membranes with anti-PLC-BI antibody, but not anti-PLC-yl or anti-PLC-31, prevented the GTPyS-mediated inhibition of PLC. These studies implicate PLC-B1 as an effector system that is under negative modulation by a G protein-dependent mechanism.
Phospholipase C (PLC)' is an effector system utilized by hormones that mediate increases in cytosolic Ca2+ levels and activation of protein kinase C (1). GTP-binding proteins (G proteins) function as key intermediaries in promoting receptor-dependent regulation of phosphoinositide-specific PLC activity. Pertussis toxin has functionally identified a pertussis toxin-sensitive and a pertussis toxin-insensitive G protein linked to the stimulation of PLC activity (2-4). G,, a 42-kDa pertussis toxin-insensitive G protein, and PLC-pl, a cytosolic 150-kDa PLC isozyme, have been identified as two components in the receptor-regulated pertussis toxin-insensitive * This work was supported by Grant DK37007 from the National Institutes of Health, a grant-in-aid from the American Heart Association, Florida Affiliate, an Established Investigatorship Award from the American Heart Association, and by funds contributed by the Florida Affiliate of the American Heart Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertkement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Increasing evidence suggests that PLC activity is negatively modulated by a pertussis toxin-sensitive G protein. Adenosine (11) and dopamine (12) inhibited PLC activation due to thyrotropin-releasing hormone in GH, cells (11) and cultured anterior pituitary cells (12) through a pertussis toxin-sensitive mechanism. In a permeabilized rat thyroid cell line FRTL5, carbachol inhibited both basal and norepinephrine-stimulated PLC activity through a GTP-and pertussis toxin-sensitive mechanism (13). Dual regulation of PLC activity by guanine nucleotides has been demonstrated in rat cerebral cortical membranes (14). Inhibition of PLC activity occurred with nanomolar guanine nucleotide concentrations, whereas stimulation ensued with micromolar concentrations of guanine nucleotides. A similar dual regulation of effector function by guanine nucleotides has been described in the adenylylcyclase system and has been attributed to the temporal activation of inhibitory and stimulatory G proteins (15,16).
At present, little is known concerning the mechanism(s) that mediate the inhibition of PLC activity. The present studies demonstrate that solubilized membrane preparations retain G protein-dependent inhibition of PLC activity. Inhibition of PLC activity occurs through a rapid but transient mechanism and results in a decrease in the Ca2+ sensitivity of PLC.

EXPERIMENTAL PROCEDURES
Preparation of Membranes-Membranes were prepared essentially as described previously (17). The only modification in the procedure was to omit the sucrose gradient purification step. The washed membranes obtained from the 17,000 X g centrifugation were frozen in a dry ice/alcohol bath and stored at -80 "C until use. For membrane studies, 3 ml of frozen membrane was thawed, diluted with 4 ml of buffer consisting of 10 mM Tris-HC1 (pH 8.0), 1 mM EDTA, 1 mM dithiothreitol, and 100 PM phenylmethysulfonyl fluoride. The membranes were pelleted at 16,000 rpm for 30 min. The membranes were resuspended in the same buffer and centrifuged an additional time. The resulting pellet was resuspended in buffer to a final protein concentration of 7 mg/ml. PLC activity in membranes was determined as described under "Measurement of Phospholipase C Activity." Membrane Solubilization-Membranes were thawed and resuspended by vortexing in 4 volumes of buffer containing 0.1% sodium cholate, 50 mM NaCl, 1.0 mM EDTA, 10 mM Tris-HC1 (pH 8.0), 1 mM dithiothreitol, and protease inhibitor mixture (17). The resuspended membranes were maintained on ice for 15 min followed by centrifugation at 32,000 X g for 30 min to pellet the membranes. The supernatant was discarded. This step was repeated twice. Preextracted membranes were brought to a final protein concentration of 10 mg/ml in preextraction buffer, and sufficient sodium cholate was added to stirred membranes to bring the final cholate concentration to 1%. Ethylene glycol was added at 0.1%. The membranes were solubilized for 1.5-2 h at 5 "C. At the end of the solubilization, the ethylene glycol concentration was increased to 10%. The solubilized mixture was centrifuged at 100,000 X g for 90 min. The clear supernatant (extract) was removed, diluted 3-fold in buffer A which con-tained 1% cholate, 1.0 mM EDTA, 10 mM Tris (pH 8.0), 10% ethylene glycol, 1 mM dithiothreitol, and protease inhibitor mixture. The sample was applied to a 1-ml DEAE column equilibrated with buffer A. The column was washed with 2 column volumes of buffer A to remove unbound protein. The bound activity was eluted with 2 ml of buffer A containing 250 mM NaC1. The solubilized preparation was stored frozen in 100-pl aliquots at -80 "C. Each aliquot was thawed once for use in the designated experiments.
Measurement of Phospholipase C Activity-PLC activity was assayed as described previously (17). Five microliters of appropriately diluted solubilized preparation was added to buffer containing (final concentration) 25 mM HEPES (pH 6.75), M e as indicated, Ca2+ as indicated, 12 mM LiCl, and 3.5 p~ phosphatidylinositol 4,5-bisphosphate (-100-150 cpm/pmol) in a final volume of 50 pl. ATP and App(NH)p (final concentration of 0.1 mM) were included in early studies, but later it was found that omission of these nucleotides did not affect the inhibitory response. As indicated, a substrate mixture of 200 p~ phosphatidylethanolamine and 3.5 pM phosphatidylinositol 4,5-bisphosphate was used in the membrane studies and G protein subunit reconstitution studies. Except where indicated, the free Ca2+ concentration was 350 nM and set by a Ca2+-EGTA buffer using 3 mM EGTA. The free Ca2+ concentration was determined from the total concentration by solving equilibrium binding equations (18,19) using published stability constants for EDTA, EGTA, and ATP (20). Incubation was conducted at 24 "C and terminated by the addition of 1.25 ml of acidified methanol/chloroform (2:l v/v), chloroform (0.5 ml), and Hz0 (0.5 ml). Membrane studies were conducted at 30 "C. Inositol phosphates were isolated by chromatography on Dowex formate resin. Radioactivity was determined by liquid scintillation counting. As indicated, the data were analyzed for statistical significance using Student's t test for paired analysis.
Pertussis Toxin Treatment-The solubilized preparation was incubated at room temperature in a 50-pl reaction mixture containing 50 mM HEPES (pH 8.0), 1 mM dithiothreitol, 1 mM ATP, 1 mM thymidine, 100 p M GTP, 2.5 mM NAD, and 3 pg of activated or heatinactivated pertussis toxin. After 3 h, an aliquot of the mixture was taken and assayed for PLC activity in the presence or absence of 10 nM GTPyS.

G Protein Subunit Studies-Purified G protein subunits (by, o(i-
GTPyS, ai-GDP, a,-GTPyS, a,-GDP) were generously given by Dr. Studies with Anti-PLC Isozyme-specific Antibody-Membranes (10 pl at 1.4 pg/pl) were incubated with anti-PLC-b, (mixed monoclonal antibodies specific for PLC-PI), anti-PLC-y, (antiserum for PLC-y,), or anti-PLC-61 (mixed monoclonal specific for PLC-6,) at 4 "C for 2 h. Following incubation, the membranes were diluted with an equal volume of fresh buffer, and 5 p1 of the membrane preparation was taken for PLC assay using mixed phospholipid vesicles.
For immunoblots, protein was separated on 7.5% sodium dodecyl sulfate-polyacrylamide gels. Western blotting was performed as described (21) using mixed monoclonal antibodies to PLC-b, or pooled polyclonal antibodies to PLC-7,. The bound antibodies were detected with horseradish peroxidase-conjugated goat anti-mouse IgG (Promega) and detected with the BLAST amplification system (Du Pont-New England Nuclear) or alkaline phosphatase-conjugated goat antimouse. Protein was determined by the Amido Black assay (22).
Materials-Antibodies to PLC-PI, PLC-yl, and PLC-6, were from Upstate Biotechnology. Bovine calf brains were obtained from Mary's Ranch.

RESULTS
The results shown in Fig. 1 demonstrate that AlFh (20 @M AlCI, plus 10 mM NaF) produced a rapid but transient inhibition of PLC activity in solubilized preparations. Approximately 20% o f the basal PLC activity was inhibited after a 7min incubation. Inhibition was followed by a rapid reversal of inhibition and onset of stimulation at 25 min. 50 nM GTPyS also produced a rapid inhibition of PLC activity within the first 7 min of incubation. As with AlF,, inhibition was followed by a reversal toward basal activity. Stimulatory effects of 50 nM GTPyS were not detected within this 25-min incubation period. In contrast, 10 PM GTPyS did not produce any detectable inhibition of PLC activity but rather promoted a slow time-dependent activation of PLC activity. A 30% increase in basal activity was evident at 25 min. A decline in basal activity because of substrate depletion at 30 min prevented further measurement of the stimulatory response.
Inhibitory effects of AlF; and GTPyS were often more log GTP--S (M) sustained in stored preparations. As shown in Fig. 2, A and B, the inhibition induced by GTPyS and AlF; was maintained for at least 25 min. The magnitude of the inhibition was greater then that of Fig. 1 and resulted in a 30% decrease in basal activity at 15 min. This increased inhibitory response was due, in part, to the reduced rate of reversal. Some reversal of inhibition was nonetheless evident at 30 min. The next series of experiments was designed to characterize the inhibitory regulation in greater detail using preparations that were stored for at least 1 week. Most studies were conducted within a 15-min incubation period to minimize against reversal of inhibition.
The dose-response curve for GTPyS-mediated inhibition of PLC activity is shown in Fig. 3. Half-maximal inhibition was obtained with approximately 1 nM GTPyS, and maximal inhibition was evident with 1 PM GTPyS. Reversal of inhibition was evident with 10 PM GTPyS.
The effect of GDPpS which antagonizes G protein action is shown in Fig. 4. GDPPS had little effect on basal PLC activity. GDPPS, however, antagonized the inhibition caused by GTPyS.
The effect of pertussis toxin on the GTPyS-mediated inhibition of PLC activity was determined. The soluble preparation was incubated in the absence or presence of pertussis toxin for 3 h followed by measurement of PLC activity. Pertussis toxin treatment (3 pg/ml) did not affect basal PLC activity. In the control, 10 nM GTPyS produced a 20% inhibition of basal PLC activity a t 10 min which was followed by a reversal of inhibition (Fig. 5). In the pertussis toxin-pretreated preparation, GTPyS did not inhibit PLC activity within the first 10 min of incubation. However, at 15 min, GTPyS produced a 15% decrease in PLC activity. These results indicate that pertussis toxin pretreatment delayed but did not prevent GTPyS-mediated inhibition of PLC activity.
The role of G protein a subunits on the basal PLC activity was next examined. Neither a,-GTPyS nor a,-GDP had any major effect on PLC activity (Table I). Both ai-GTPyS and ai-GDP produced a 15% inhibition of PLC activity. The results shown in the right column compare the effects of GTPyS and a subunits within the same experiment. A 7-min incubation with GTPyS produced a 10% inhibition of PLC  activity. PLC activity, as measured in the presence of a,-GDP and a,-GTPyS, was decreased by 7 and 5%. However, in the presence of ai-GDP and ai-GTPyS, the activity was decreased by 13 and 14%, respectively. These results are consistent with the data shown in the center column and indicate that the ai subunits produced a greater degree of inhibition than did the CY, subunits. fly subunits are shared by all G proteins. by subunits modulate a activity and have been shown to regulate some effector systems directly (2)(3)(4). The results shown in Fig. 6A demonstrate that Py subunits did not inhibit basal PLC activity, indicating that the G protein-mediated inhibition of PLC activity was not a sole consequence of the release of By subunits. However, the net inhibitory response measured in the presence of GTPyS was attenuated by the addition of By subunits (Fig. 6B).
Caz+ activates PLC activity in vivo. As shown in Fig. 7, incubation of the preparation with 50 nM GTPyS decreased the Ca2+ sensitivity of PLC. Both the maximal degree of activation as well as the Ca2+ sensitivity of basal PLC activity was reduced. Inhibition of PLC did not require ATP and could not be induced by the addition of protein kinase C activators such as 12-0-tetradecanoylphorbol-13-acetate or the addition of the catalytic subunit of protein kinase A (data not shown). Thus, the observed inhibition was not a result of phosphorylation.
Antibodies to the three major isozymes of PLC have been used to identify PLC effector systems in membranes (7). The results shown in Table I1 demonstrate that pretreatment of membranes with a mixture of monoclonal antibodies to PLC-p1 blocked the ability of GTPyS to inhibit PLC activity.
Incubation of membranes with a polyclonal antiserum specific for PLC-y, or a mixture of monoclonal antibodies to PLC-& did not alter the ability of GTPyS to inhibit PLC activity. In parallel studies with solubilized preparations, Western blotting was used to identify the PLC isozymes present in the solubilized preparation. PLC-P, was the only PLC isozyme that could be detected in a Western blot against 150 pg of loaded sample using standard detection protocols or an amplified protocol (BLAST) using 0.15 pg of protein (data not shown). These results provide evidence that PLC-P1 is an effector in the inhibitory response.

DISCUSSION
The present studies demonstrate that G proteins mediate a rapid inhibitory modulation of PLC activity in vitro. Evidence that inhibition of PLC activity is mediated through a G protein includes guanine nucleotide sensitivity and antagonism by GDPPS. Inhibition of PLC activity is transient, followed by a rapid reversal of inhibition and onset of stimulation (Fig. 1). The magnitude of the inhibition is affected by the rate of reversal. A decrease in the rate of reversal results in both an increased degree of inhibition and a more sustained inhibitory effect (Fig. 2). Since reversal of inhibition was followed by stimulation at 25 min, it is likely that reversal occurs as a consequence of the concurrent activation of a stimulatory G protein, but this has not been established. These results suggest, in turn, that the magnitude of the stimulatory effect as well as the time course for the onset of stimulation may be modulated by the inhibitory component. Thus in a membrane system that contains both components, the net degree of PLC inhibition and PLC stimulation will reflect a balance between the regulatory input provided by the stimulatory and inhibitory mechanisms. The presence of a dual regulatory mechanism in solubilized bovine brain membrane preparations may account for the difference in the magnitude of PLC stimulation which has been observed in the solubilized preparation as compared with studies in a

PC0
reconstituted system using purified G proteins (5, 6). The temporal relationship between the onset of inhibition and stimulation as well as the magnitude of the inhibitory response is similar to that observed with the G proteinregulated adenylylcyclase system. Dual regulation of adenylylcyclase has been attributed to the temporal activation of an inhibitory G (Gi,) and a stimulatory (G,) G protein.
However, the precise mechanisms involved in regulating adenylylcyclase have not been resolved. Controversy exists concerning the relative role of the a subunit and By subunit in mediating the dual regulation of adenylylcyclase (2, 23, 24).
Similar considerations may apply in the PLC signaling system.

6.0
The present studies, however, do provide insight into the components involved in the inhibition of PLC. PLC-P1 is the effector in this mechanism. This conclusion is based on the observation that antibodies to PLC-pI block GTPyS-dependent inhibition of PLC activity (Table 11) and that PLC-p, is the only PLC isozyme that is detected in a Western blot of the solubilized preparation. The properties of the G protein involved in this mechanism have been partially characterized. Half-maximal inhibition is evident at 1 nM GTPyS, indicating that this event is evoked by the activation of a G protein with a high affinity for guanine nucleotides (Fig. 3). a,-GTPyS had little effect (Table I). The studies with ai were inconclusive since both ai-GDP and ai-GTPyS produced an inhibition Effect of anti-PLC antibodies on GTPyS-mediated inhibition of PLC actiuity in membranes Bovine brain membranes (10 p1 at 1.4 pg/pl) were incubated for 2 h in the presence of 0.2 pg of anti-PLC-PI (mixed monoclonal antibodies specific for PLC-&), 4 ~1 of a 1:lO diluted antiserum stock specific for PLC-yl, or 0.2 pg of anti-PLC-61 (mixed monoclonal specific for PLC-6,) at 4 "C for 2 h. After incubation, the membranes were diluted with an equivalent volume of fresh buffer. Five microliters of membrane was used in the PLC assay which was conducted for 10 min at 30 "C using mixed phospholipid vesicles. Results are the mean f S.E. of three experiments done in duplicate.

PLC activity Pretreatment
Basal +lo0 nM GTP+ P m l of PLC activity. Since a similar degree of inhibition was obtained with both the GTP+ and GDP form of ai, the observed effect was unlikely to be a result of an activated ai that directly inhibited PLC activity. The mechanism responsible for the inhibition has not been determined.
Pertussis toxin catalyzes the ADP-ribosylation of a cysteine residue located 4 amino acids from the carboxyl terminus of the G protein a subunit and blocks GTP as well as receptormediated interactions in many systems (15, 23). Pertussis toxin sensitivity has been used to identify G protein-mediated events. GTPyS dissociates the ADP-ribosylated G protein heterotrimer to the ADP-ribosylated CY subunit and By (24). Thus, in most systems, pertussis toxin does not block the effects of G proteins that have been activated by GTPyS. In the studies shown on Fig. 5, it was possible to detect a pertussis toxin effect within the first 10 min of incubation. At 15 min, GTPyS initiated an inhibitory response in the pertussis toxin-treated sample. Since subunit dissociation by activating ligands such as GTPyS is temperature-dependent, the use of low temperatures and relatively short incubation times may have allowed maintenance of the pertussis toxininduced modification for a limited time. Although these studies suggest that a pertussis toxin-sensitive G protein is involved in the mediation of the inhibitory effect, final proof will require purification of a pertussis toxin substrate and reconstitution of PLC regulation. These results, for instance, cannot exclude other effects of pertussis toxin which might include potentiation of the stimulatory component or interference in the ability of the PLC to be inhibited by a G protein or a G protein-controlled mechanism.
The data in Fig. 6 demonstrate that the addition of Py subunits attenuated the GTPyS-mediated inhibition of PLC activity. Since GTPyS produces a persistent activation of G protein a subunits, the addition of ,By subunits might not be expected to reverse the GTPyS effect. There are at least two possible explanations for these results. First, the G protein may not be a typical G protein, and the addition of excess py subunits may be sufficient to deactivate the signal. Alternatively, the effects of 0-y may be indirect. There is increasing evidence indicating that By subunits modulate effector systems. P-y modulates adenylylcyclase activity through a direct as well as an indirect mechanism. The indirect mechanism is dependent on the presence of a subunits but is not a result of the ability of Py to associate with a subunits and cause deactivation of the a subunit. Inhibition as well as activation of adenylylcyclase by B r occur and the observed effects are dependent on the type of adenylylcyclase as well as the presence of CY, (25,26). Py subunits activate a liver cytosolic PLC activity (27) and a turkey erythrocyte cytosolic PLC activity (28) with only marginal effects on PLC-pl (27, 28). By activation of basal PLC activity was not observed under the conditions used in the present studies. Thus, the attenuation of GTPyS dependent inhibition of PLC by fly is not likely to be a result of a By-induced stimulation of basal activity. However, other indirect effects that depend on the presence of activated a subunits cannot be excluded.
These studies also demonstrate that PLC activity is negatively modulated in uitro through a G protein-controlled mechanism. Inhibition of PLC activity is associated with a decrease in the Ca2+ sensitivity of PLC as well as the maximal activity of PLC. This effect contrasts with the observed increase in Ca2+ sensitivity which has been demonstrated in response to G protein activation (6, 14, 29,30). Since PLC-pl is the effector for the G,/Gll stimulatory pathway and the inhibitory pathway, these studies suggest that G proteins regulate PLC activity through a shared mechanism that remains to be identified. In the present studies, inhibition of PLC activity could not be induced by activation of protein kinase C or protein kinase A, indicating that the effects were not mediated through phosphorylation.
In summary, these studies have identified a G proteindependent mechanism that functions to negatively modulate PLC activity in solubilized bovine brain membranes. The mechanism involved in promoting G protein-dependent inhibition of PLC activity and the regulation of this inhibitory response remain to be determined. Given the complexity of G protein-effector interactions, it is possible that both direct and indirect mechanisms may be involved in this effect.