Molecular Mechanism of the Inhibition of Phospholipase C β3 by Protein Kinase C*

Activation of protein kinase C (PKC) can result from stimulation of the receptor-G protein-phospholipase C (PLCβ) pathway. In turn, phosphorylation of PLCβ by PKC may play a role in the regulation of receptor-mediated phosphatidylinositide (PI) turnover and intracellular Ca2+ release. Activation of endogenous PKC by phorbol 12-myristate 13-acetate inhibited both Gαq-coupled (oxytocin and M1 muscarinic) and Gαi-coupled (formyl-Met-Leu-Phe) receptor-stimulated PI turnover by 50–100% in PHM1, HeLa, COSM6, and RBL-2H3 cells expressing PLCβ3. Activation of conventional PKCs with thymeleatoxin similarly inhibited oxytocin or formyl-Met-Leu-Phe receptor-stimulated PI turnover. The PKC inhibitory effect was also observed when PLCβ3 was stimulated directly by Gαq or Gβγ in overexpression assays. PKC phosphorylated PLCβ3 at the same predominant sitein vivo and in vitro. Peptide sequencing ofin vitro phosphorylated recombinant PLCβ3 and site-directed mutagenesis identified Ser1105 as the predominant phosphorylation site. Ser1105 is also phosphorylated by protein kinase A (PKA; Yue, C., Dodge, K. L., Weber, G., and Sanborn, B. M. (1998) J. Biol. Chem. 273, 18023–18027). Similar to PKA, the inhibition by PKC of Gαq-stimulated PLCβ3 activity was completely abolished by mutation of Ser1105 to Ala. In contrast, mutation of Ser1105 or Ser26, another putative phosphorylation target, to Ala had no effect on inhibition of Gβγ-stimulated PLCβ3activity by PKC or PKA. These data indicate that PKC and PKA act similarly in that they inhibit Gαq-stimulated PLCβ3 as a result of phosphorylation of Ser1105. Moreover, PKC and PKA both inhibit Gβγ-stimulated activity by mechanisms that do not involve Ser1105.

Stimulation of seven transmembrane receptors coupled to the G␣ q or G␣ i subunits of heterotrimeric G proteins results in activation of PLC␤ 1 isoforms that hydrolyze phosphatidylinositol 4,5-bisphosphate to generate the second messengers inositol 1,4,5-trisphosphate (IP 3 ) and diacylglycerol (1,2). IP 3 binds to a receptor in endoplasmic reticulum and releases intracellular calcium from its stores. Diacylglycerol, alone or in conjunction with elevated intracellular calcium, activates PKC and initiates additional cellular responses (3). Currently, four isoforms of mammalian PLC␤ have been identified and characterized (4 -10). Significantly, PLC␤ 3 is ubiquitously expressed and activated by all known PLC␤ activators (G␣ q , G␤␥, and calcium) (2). Regulation of PLC␤ 3 may be of great importance in many cellular processes (11)(12)(13)(14)(15). Insufficient expression of PLC␤ 3 has been correlated with increased sensitivity to tumor formation (15,16), whereas overexpression of PLC␤ 3 seems to suppress tumor growth (17). PLC␤ 3 knockout mice exhibit altered response to -opioids (11) or early embryonic lethality (18).
Phosphorylation appears to play an important role in regulating the activity of PLC␤ isoforms. Phosphorylation of PLC␤ 3 or PLC␤ 2 by PKA inhibits their activity and establishes a mechanism for cross-talk between G␣ q -or G␣ i -coupled and G␣ s -coupled receptors (12,19). The inhibition of G proteincoupled receptor-mediated PI turnover or intracellular calcium release by protein kinase C has been reported (20 -25). Protein kinase C is comprised of three subfamilies, the conventional (␣, ␤ 1 , ␤ 2 , and ␥), novel (␦, ⑀, , , and ), and atypical ( and ) PKCs (3). The conventional and novel PKCs are activated subsequent to the stimulation of G␣ q -or G␣ i -coupled receptors (3,26). The inhibition of PI turnover by PKC may present a feedback for determining the frequency and amplitude of signals being transmitted.
The mechanisms by which PKC inhibits agonist-stimulated PI turnover have not been well defined. PKC can phosphorylate certain G protein-coupled receptors (platelet-activating factor receptor, C5A receptor) and thereby inhibit PI turnover or intracellular calcium release (reviewed in Ref. 27). PKC also appears to inhibit agonist-stimulated PI turnover at a postreceptor level (25,28). Although phosphorylation of PLC␤ 1 and PLC␤ 2 by PKC has been reported (23,24,29,30), the physiological relevance of these observations has not been demonstrated. PLC␤ t , a turkey PLC␤ isoform with highest homology to PLC␤ 2 , is phosphorylated by conventional PKCs, and its catalytic activity is inhibited (29). PLC␤ 3 is not phosphorylated by PKC␣ in vitro (23). Nonetheless, a correlation between PLC␤ 3 phosphorylation and PKC inhibition of receptor-initiated PI turnover has been reported (21,31).
To determine the importance of PLC␤ 3 phosphorylation by PKC, we have identified the phosphorylation site on PLC␤ 3 and investigated which PKC subfamily can catalyze the phospho-rylation. We report the identification of Ser 1105 as the predominant PKC phosphorylation site, the involvement of conventional PKCs in this phosphorylation, and the convergence of PKC and PKA on phosphorylation and inhibition of PLC␤ 3 by G␣ q . Furthermore, we find that G␤␥-stimulated PLC␤ 3 activity is inhibited by both PKC and PKA by mechanisms that do not involve Ser 1105 phosphorylation. Cloning, Site-directed Mutagenesis, and Protein Purification-PLC␤ 3 , G␣ q , G␤ 1 , and G␥ 2 plasmids were constructed as described elsewhere (12,32). Site-directed mutation of Ser 26 to Ala was achieved with the mutagenic primer (5Ј-CGGCGCGGGGCTAAGTTCATCAAAT-GG-3Ј) identically as described for the Ser 1105 3 Ala mutation (12) using GeneEditor. All plasmid sequences were confirmed by DNA sequencing. Construction of baculovirus containing PLC␤ 3 Ser 1105 3 Ala-(His) 6 and purification of the recombinant protein from Sf9 cells were carried out as described for PLC␤ 3 (His) 6 (12).

Materials-Thymeleatoxin
In Vivo and in Vitro 32 P Labeling and Isolation of PLC␤ 3 -For in vivo phosphorylation, COSM6 cells seeded in 6-well plates were transfected with PLC␤ 3 (His) 6 plasmid and metabolically labeled with [ 32 P] orthophosphate (0.10 mCi) in 0.5 ml of phosphate-free DMEM for 90 min. After PMA (1 M) treatment for 30 min, cells were lysed in 500 l of M-PER lysis buffer (Pierce) containing a mixture of protease and phosphatase inhibitors (21) and centrifuged at 15,000 ϫ g for 5 min at 4°C. Phosphorylated PLC␤ 3 (His) 6 was isolated with nickel-nitrilotriacetic acid resin, separated on a 7.5% SDS-polyacrylamide gel, stained with Coomassie Blue, and analyzed by autoradiography.
In vitro phosphorylation by PKC was carried out according to protocols provided by the vendor. Briefly, 0.8 M purified recombinant PLC␤ 3 (His) 6 or PLC␤ 3 Ser 1105 3 Ala(His) 6 was incubated with purified constitutively active PKC fragment (0.04 M) in the presence of 2.5 Ci of [␥-32 P]ATP and 100 M ATP in a total volume of 10 l of PKC buffer (50 mM MES, pH 6.5, 1.25 mM EGTA, 12.5 mM MgCl 2 ) for the times specified at 30°C. Equal amounts of PLC␤ 3 (His) 6 were also incubated for 40 min with purified PKC␤ 1 or PKC␥ (20 ng) in a total volume of 10 l of reaction buffer (20 mM HEPES, pH 7.4, 100 M CaCl 2 , 10 mM MgCl 2 , 100 g/ml phosphatidylserine, 20 g/ml diacylglycerol, 0.03% Triton X-100). Reactions were terminated by adding 10 l of 2ϫ SDS sample buffer and boiling for 5 min. Proteins were separated by 7.5% SDS-polyacrylamide gel electrophoresis and stained with Coomassie Blue. The phosphorylated bands were localized by autoradiography. The stoichiometry of PLC␤ 3 phosphorylation by PKC was determined at 100 min by filter binding assay as described elsewhere (12).
Phosphoamino Acid Analysis, Peptide Mapping, and Sequencing-For two-dimensional tryptic peptide mapping and phosphoamino acid analysis, 32 P-labeled PLC␤ 3 from in vivo or in vitro phosphorylation reactions was separated by SDS-polyacrylamide gel electrophoresis. The gel was stained with Coomassie Blue, dried between two layers of drying membranes, and exposed to Biomax-MS x-ray film (Eastman Kodak Co.). PLC␤ 3 bands were cut out and rehydrated in 50 mM ammonium bicarbonate, pH 8 (buffer A), overnight. After peeling off the drying membrane, each gel slice was boiled for 5 min in 100 l of buffer A containing 5 mM dithiothreitol. The tube was cooled to room temperature, 50 l of 100 mM iodoacetic acid was added, and the tube was incubated for 30 min in the dark at room temperature. The gel slice was washed again in buffer A and ground with a disposable pestle. The residual Coomassie Blue dye was removed by rinsing the gel slurry with 50 l of 50% acetonitrile in buffer A. The tube was centrifuged at 15,000 ϫ g for 5 min, and the supernatant was discarded. The pellet was resuspended in 50 l of acetonitrile and incubated for 5 min. The tube was centrifuged again, and the pellet was dried in a SpeedVac for 10 min after removal of supernatant. The pellet was resuspended in 75 l of buffer A, and 2 g of trypsin was added. The tube was incubated at 37°C for 5 h before the addition of another 2 g of trypsin, and the total incubation time was between 18 and 24 h. The liquid containing the digested peptides was recovered and further prepared for twodimensional peptide mapping with a Hunter thin layer electrophoresis system (C.B.S. Scientific Co., Del Mar, CA) according to the protocol provided by the manufacturer. External markers for each dimension were included in each thin layer plate to facilitate the comparison between samples. For phosphoamino acid analysis, about 100 cpm of total tryptic peptides mixture was used. Peptide sequencing using 32 Plabeled PLC␤ 3 (His) 6 (150 pmol) recombinant protein purified from Sf9 cells was carried out as described elsewhere (12).
Cell Culture, Transfection, and PI Turnover-HeLa, COSM6, and RBL-2H3 cells were cultured as described for PHM1-41 cells (33). HeLa and COSM6 cells (1.8 ϫ 10 5 /well) were seeded in 6-well plates and transfected 16 -24 h later as described (34) with M1 receptor (1 g), G␣ q (0.5 g), G␤ 1 (0.375 g), G␥ 2 (0.375 g), and PLC␤ 3 (0.25 g) as indicated. Empty rcCMV vector was added to bring the total amount of plasmid DNA to 1.25 g per well. For effects of endogenous PKC on agonist-stimulated PI turnover, near confluent PHM1 and RBL-2H3 cells (12- After 15 min, PMA or CPT-cAMP were added, followed by oxytocin 15 min later. For direct stimulation of PLC␤ 3 by G␣ q or G␤ 1 ␥ 2 , transfected COSM6 cells were first treated with 1 M PMA for 30 min in PBSϩ followed by addition of 20 mM LiCl for 30 min. Cells were lysed, and total IPs were determined as described (19).

PKC Inhibits Oxytocin, M1 Muscarinic, and fMLP Receptorinitiated PI Turnover-
The effect of activation of endogenous PKC on predominantly G␣ q -coupled oxytocin receptor-initiated PI turnover (35) was studied in PHM1-41 cells, a human myometrial smooth muscle cell line (33). Stimulation of PHM1 cells with 100 nM oxytocin significantly increased the production of total IPs. Pretreating cells with 1 M PMA completely inhibited this increase (Fig. 1A). The PMA effect was not specific to the oxytocin receptor or to PHM1-41 cells. A similar inhibitory effect of PMA was also evident with G␣ q -coupled M1 muscarinic receptor transfected into HeLa (Fig. 1B) or COSM6 (Fig.  1C) cells. In addition, PMA also significantly inhibited G␣ icoupled fMLP receptor-initiated PI turnover (36) in RBL-2H3 cells (Fig. 1D) in which the only PLC␤ form expressed is PLC␤ 3 (21). This occurred under conditions where the fMLP receptor has been shown not to be phosphorylated by PKC (37). These observations, together with those previously reported (21,31), establish that the PKC inhibitory effect on G protein-coupled receptor-initiated PI turnover is a general mechanism and that the PKC effect can occur at a post-receptor level.
To investigate the potential role of specific PKCs in this process, the effect of Tx, a specific activator of conventional PKCs (38), was compared with PMA, which activates both conventional and novel PKCs (38), in PHM1-41 and RBL-2H3 cell lines. In both cases, Tx was as effective as PMA in inhibiting oxytocin or fMLP-stimulated PI turnover at the concentration tested (Fig. 1, A and D). In addition, Gö 6976, an inhibitor of conventional PKC (39), was able to reverse the PMA inhibitory effect by ϳ50% at a concentration of 4 M (data not shown). These data provide evidence that conventional PKCs are capable of inhibiting G␣ q -or G␣ i -coupled receptorinitiated PI turnover. PKC Inhibits the Direct Stimulation of PLC␤ 3 by G␣ q and G␤␥-Because PLC␤ 3 is present in all four cell lines mentioned above and can be phosphorylated by PKC, at least in RBL-2H3 cells (21), it is highly possible that PKC inhibits PI turnover by decreasing PLC␤ 3 activity. If so, PKC should inhibit the direct stimulation of PLC␤ 3 by G␣ q or G␤␥. COSM6 cells transfected with both PLC␤ 3 and G␣ q plasmids exhibited a marked increase in total [ 3 H]IPs compared with transfection with either plasmid alone ( Fig. 2A). Consistent with the prediction, pretreating cells with PMA nearly abolished G␣ q -stimulated PLC␤ 3 activity. Tx elicited a similar inhibitory effect on G␣ qstimulated PLC␤ 3 activity (data not shown). Cotransfection of G␤ 1 ␥ 2 and PLC␤ 3 into COSM6 cells also resulted in marked increase in PI turnover. This increase was significantly reduced by PMA (Fig. 2B), but the reduction was not of the magnitude observed for G␣ q -stimulated PLC␤ 3 . Thus PKC inhibition of PI turnover occurs at a post-receptor level, and this effect may require the phosphorylation of PLC␤ 3 .
Phosphorylation of PLC␤ 3 by PKC in Vivo and in Vitro-PLC␤ 3 overexpressed in COSM6 cells exhibited significant 32 P incorporation under basal conditions. Nonetheless, PMA induced a substantial increase in 32 P incorporation into PLC␤ 3 (Fig. 3A). The phosphorylation of PLC␤ 3 by PKC was investigated further in vitro. Purified recombinant PLC␤ 3 was incubated with catalytically active PKC fragments (a rat brain mixture of multiple PKC isoforms, including ␣, ␤, and ␥) in the presence of [␥-32 P]ATP. As shown in Fig. 3B, PLC␤ 3 was phos- Reactions were terminated at the times indicated in minutes. Coomassie Blue staining of the respective gels is shown below the autoradiographs. C, autoradiography of PLC␤ 3 (His) 6 after incubation without (Ϫ) or with (ϩ) purified PKC␤ 1 and PKC␥ in vitro.

FIG. 2. PMA inhibits G␣ q -stimulated PLC␤ 3 (A) and G␤␥-stimulated PLC␤ 3 (B) activity in COSM6 cells transfected with plasmids expressing G␣ q , G␤␥, and PLC␤ 3 .
Data are presented as the means Ϯ S.E. (n ϭ 3) of 1 of 3 similar experiments and were analyzed by analysis of variance and Duncan's test. Groups with different letters are different from each other at p Ͻ 0.05. phorylated in a time-dependent manner. A stoichiometry of 0.4 mol of phosphate/PLC␤ 3 was achieved after incubation with PKC for 100 min under these conditions. In similar experiments, no phosphorylation was seen in the absence of PKC (data not shown). Purified PKC␤ 1 or PKC␥ also phosphorylated PLC␤ 3 in vitro, whereas no phosphorylation of PLC␤ 3 was observed in the absence of kinase (Fig. 3C).
Ser 1105 Is the Predominant Phosphorylation Site for PKC-As shown by two-dimensional phosphopeptide mapping of in vivo 32 P-labeled PLC␤ 3 , trypsin digestion yielded multiple phosphopeptides in the basal state (Fig. 4A). PMA specifically induced phosphorylation on one predominant site (Fig. 4B, indicated by the arrow). Minor sites increased by PMA were also present (indicated by arrowhead). We cannot exclude the contribution of incomplete digestion by trypsin to this pattern.
In vitro, PKC phosphorylated PLC␤ 3 on one predominant site (Fig. 4C, arrow). The migration of this peptide relative to the standards was identical to those observed in digests after in vivo phosphorylation. The phosphorylation occurred exclusively on serine residues (Fig. 4D). We utilized in vitro phosphorylated recombinant PLC␤ 3 (His) 6 to identify the PKC phosphorylation sites. After isolation by SDS-polyacrylamide gel electrophoresis, 32 P-labeled PLC␤ 3 was digested with Lys C instead of trypsin to achieve more complete cleavage and fewer peptides (12). The digestion mixture was separated by reversephase high pressure liquid chromatography, and fractions were recovered and counted. Fig. 5A shows the 32 P distribution among these fractions. About 8% of 32 P was found in the followthrough (fraction Ϫ1 to Ϫ4) and appeared to be free 32 P as judged by phosphopeptide mapping (data not shown). Nearly 60% of the total 32 P was recovered in fraction 12. This fraction was subjected to peptide sequencing. Although two peptides were identified in this fraction, nearly 90% of the total 32 P was found in the fourth cycle (Fig. 5B). This clearly identified Ser 1105 and not Ser 1107 in the peptide Arg-His-Asn-Ser 1105 -Ile-Ser-Glu-Ala-Lys as the amino acid phosphorylated. Furthermore, mutation of Ser 1105 significantly diminished PLC␤ 3 phosphorylation by PKC in vitro (Fig. 5C). This strongly argues that Ser 1105 is the predominant site for PKC. Residual weak phosphorylation associated with Ser 1105 3 Ala mutant PLC␤ 3 could indicate the presence of other minor sites. Interestingly, Ser 1105 , unique to PLC␤ 3 among the PLC␤ isoforms, is preferentially phosphorylated by PKA as well (12). FIG. 5. A, 32 P distribution among fractions collected after reverse-phase high pressure liquid chromatography separation of Lys C-digested PLC␤ 3 (His) 6 labeled with 32 P in vitro. Fraction 12 has ϳ60% of the total 32 P. B, sequence of peptides in fraction 12 and associated 32 P. The serine residue with more than 90% of total 32 P loaded onto the sequencing membrane is denoted by *. filter represents 32 vivo (A and B) or in vitro (C). Two markers were applied on each TLC plate as migration controls for each dimension. The black markers on the top of each panel indicate the position of one such marker; others outside of the displayed region were also used in lining up the plates. "O" depicts the sample origin. The predominant PKC-stimulated phosphorylation site is indicated by the arrows (B and C) and the minor sites by the arrowheads. Longer exposure of C revealed some minor sites as well. D, two-dimensional phosphoamino acid analysis of PLC␤ 3 (His) 6

Functional Analysis of Phosphorylation of PLC␤ 3 by PKC
Versus PKA-We have previously shown that phosphorylation by PKA of Ser 1105 is required for inhibition of G␣ q -stimulated PLC␤ 3 activity by PKA. The finding that PKC also phosphorylates Ser 1105 suggested that the same mechanism was utilized by PKC. To test this hypothesis, the Ser 1105 3 Ala mutant PLC␤ 3 was cotransfected with G␣ q into COSM6 cells, and the effect of PMA was evaluated. As shown before (12), the Ser 1105 3 Ala mutant PLC␤ 3 was as effective as the wild type enzyme in coupling to G␣ q (Fig. 6). Importantly, PMA inhibited G␣ qstimulated wild type PLC␤ 3 activity but had no effect on G␣ qstimulated Ser 1105 3 Ala PLC␤ 3 activity. These data unequivocally identify phosphorylation of Ser 1105 by PKC as responsible for PKC inhibition of G␣ q -stimulated PLC␤ 3 activity.
We also investigated the effect of mutating Ser 1105 on PKC inhibition of G␤␥-stimulated PLC␤ 3 activity. The Ser 1105 3 Ala mutant PLC␤ 3 was as effective as wild type PLC␤ 3 in coupling to G␤ 1 ␥ 2 (Fig. 7A). However in contrast to G␣ q -stimulated PLC␤ 3 activity, PKC inhibited G␤ 1 ␥ 2 -stimulated Ser 1105 3 Ala mutant PLC␤ 3 activity to the similar degree as it did the wild type PLC␤ 3 . Thus Ser 1105 is not absolutely required for PKC inhibition of G␤ 1 ␥ 2 -stimulated PLC␤ 3 activity. The N-terminal region of PLC␤ 3 appears to contribute to its interaction with G␤␥ (40). We had identified Ser 26 in the peptide Arg-Arg-Gly-Ser-Lys as a potential phosphorylation site in this region. However, there was no effect of mutating Ser 26 to Ala on PKC inhibition of G␤␥-stimulated PLC␤ 3 activity (Fig. 7A).
In the face of the inability of mutation of Ser 1105 and Ser 26 to reverse the effect of PKC on G␤␥-stimulated PLC␤ 3 activity, we examined the effect of mutation of these residues on PKAmediated inhibition as well. As seen in Fig. 7B, PKA inhibited G␤␥-stimulated PLC␤ 3 activity. Mutation of Ser 1105 or Ser 26 also had no effect on inhibition by PKA of G␤␥-stimulated PLC␤ 3 activity.
Inhibition of Oxytocin-stimulated Total IP Production in PHM1-41 Cells by PKC or PKA Represents Independent Pathways-Phosphoryation of Ser 1105 by PKC or PKA suppressed G␣ q -stimulated PLC␤ 3 activity. This fact raised the interesting possibility that PKC activation might lead to PKA activation, resulting in indirect phosphorylation of PLC␤ 3 at the PKA site or vice versa. We addressed this possibility in PHM1-41 cells. As shown in Fig. 8, H-89, a specific PKA inhibitor, reversed the inhibition by cAMP but did not affect the inhibition by PMA of oxytocin-stimulated PI turnover. Similarly, Gö 6976, a specific PKC inhibitor, significantly diminished the inhibitory effect of PMA but not of cAMP on oxytocin-stimulated PI turnover. These data indicate that PKC and PKA exert their inhibitory effects independent of each other. DISCUSSION We have presented evidence that PKC inhibits G␣ q -coupled (oxytocin and M1 muscarinic) and G␣ i -coupled receptor (fMLP) receptor-initiated PI turnover in four different cell lines expressing PLC␤ 3 . The response to endogenous PKC activation by PMA differs in order of magnitude between cell lines and the state of the receptor (endogenous or transfected). This variation may reflect differences in relative membrane permeability of PMA and the localization and abundance of the PKC isoforms responsible or the relative contribution of G␣ q -coupling to PLC␤ 3 to PI turnover. We have also demonstrated in cotransfection assays that the PKC inhibitory effect occurred at the G protein-PLC␤ 3 level, and we have provided direct evidence to support the hypothesis that phosphorylation of PLC␤ 3 is involved in the PKC inhibitory effect on G␣ q -coupled activation.
The use of in vitro phosphorylated PLC␤ 3 for identifying the PKC phosphorylation site is supported by the demonstration that a similar site was phosphorylated by PKC in vivo and in vitro. PKC phosphorylates predominantly one residue, Ser 1105 , that is also phosphorylated by PKA (12). The marked reduction of in vitro phosphorylation of the Ser 1105 3 Ala PLC␤ 3 mutant further corroborates this finding. However, the remaining weak phosphorylation associated with this mutant indicates that PKC may phosphorylate other minor sites as well.
Mutation of Ser 1105 to Ala reversed completely the inhibition of G␣ q -stimulated PLC␤ 3 activity by PKC. This provides conclusive evidence for the direct inhibition of PLC␤ 3 by PKC, a response identical to that seen previously for PKA (12). We also demonstrated that the inhibitory effect of PKC occurs in the absence of PKA inhibition, suggesting that it is not a consequence of indirect PKA activation. The convergence of PKC and PKA on Ser 1105 underscores the importance of Ser 1105 in the regulation of G␣ q -stimulated PLC␤ 3 activity in diverse cellular processes and suggests possible redundancy for the inhibition of PLC␤ 3 activity by these two kinases. In addition, these data also argue that the effect of PKC or PKA targets PLC␤ 3 and not G protein or proteins involved in the production of substrate phosphatidylinositol 4,5-bisphosphate, as mutation of Ser 1105 can completely reverse the inhibition by PKC or PKA of G␣ qstimulated PLC␤ 3 activity.
In marked contrast, Ser 1105 does not appear to be critical for inhibition of G␤␥-stimulated PLC␤ 3 activity by either PKC or PKA. Ser 26 was also not required, although the N-terminal region of PLC␤ 3 appears to contribute to its interaction with G␤␥ (40). At present the mechanism for the inhibition of G␤␥stimulated PLC␤ 3 activity by PKC or PKA remains unknown. It is unlikely that G␤ 1 ␥ 2 is the direct target for the inhibitory effects of PKA or PKC as these proteins are not phosphorylated by PKC or PKA in vitro. 2 Identification of PKA or PKC minor phosphorylation sites may help to solve this question. Alternatively, the mechanism may involve phosphorylation of other molecules indirectly involved in the coupling (12).
The effects of a conventional PKC-specific activator and an inhibitor indicate that conventional PKCs are capable of phosphorylating PLC␤ 3 . This conclusion is supported by in vitro phosphorylation of PLC␤ 3 by the constitutively active PKC fragment and by purified PKC␤ 1 and PKC␥. The wide distribution of conventional PKCs (26) and PLC␤ 3 (2) in tissues correlates well with the generality of the PKC inhibitory effect on receptor-initiated PI turnover.
We conclude that conventional PKCs phosphorylate PLC␤ 3 and inhibit G␣ q -and G␤␥-stimulated PLC␤ 3 activity. PKC and PKA act similarly in that they inhibit G␣ q -stimulated PLC␤ 3 as a result of phosphorylation of Ser 1105 . Moreover, PKA and PKC both inhibit G␤␥-stimulated activity by mechanisms that do not involve Ser 1105 .