Differential Regulation of Phosphoinositide and Phosphatidylcholine Hydrolysis by Protein Kinase C-Dl Overexpression EFFECTS ON STIMULATION BY (u-THROMBIN, GUANOSINE 5’-0-(THIOTRIPHOSPHATE), AND CALCIUM*

fibroblasts that stably overexpress cDNA for the B1 isozyme of protein kinase C (PKC3 cells) were used to determine the effect of protein kinase C (PKC) overexpression on hormonal stimulation of phospho- lipid hydrolysis. In control Rat 6 cells, inositol trisphosphate levels (InsP3) were increased 9-fold in 15 s in response to 10 nM a-thrombin, compared with only a 2-fold increase in PKCS cells. PKC overexpression also inhibited thrombin-stimulated production of 1,2-diacylglycerol, the other product of phosphatidylino- sitol 4,5-bisphosphate hydrolysis, by 73% at 15 s. In permeabilized cells, PKC overexpression greatly re- duced guanosine thiotriphosphate-stimulated InsP3 accumulation, but did not affect InsPS stimulation by increased free calcium concentration. These data suggest that desensitization of thrombin-stimulated phos-phoinositide-phospholipase C is enhanced by PKC-B1 overexpression and may involve modulation of G-pro-tein/phospholipase C coupling. In contrast, thrombin was 4.5-fold


Differential Regulation of Phosphoinositide and Phosphatidylcholine Hydrolysis by Protein Kinase C-Dl Overexpression
Rat 6 fibroblasts that stably overexpress cDNA for the B1 isozyme of protein kinase C (PKC3 cells) were used to determine the effect of protein kinase C (PKC) overexpression on hormonal stimulation of phospholipid hydrolysis. In control Rat 6 cells, inositol trisphosphate levels (InsP3) were increased 9-fold in 15 s in response to 10 nM a-thrombin, compared with only a 2-fold increase in PKCS cells. PKC overexpression also inhibited thrombin-stimulated production of 1,2diacylglycerol, the other product of phosphatidylinositol 4,5-bisphosphate hydrolysis, by 73% at 15 s. In permeabilized cells, PKC overexpression greatly reduced guanosine thiotriphosphate-stimulated InsP3 accumulation, but did not affect InsPS stimulation by increased free calcium concentration. These data suggest that desensitization of thrombin-stimulated phosphoinositide-phospholipase C is enhanced by PKC-B1 overexpression and may involve modulation of G-protein/phospholipase C coupling.
In contrast, thrombin was 4.5-fold more effective in stimulation of phosphatidylcholine-phospholipase D activity in PKC3 cells than in control cells, as determined by phosphatidylethanol formation. In permeabilized cells, guanosine thiotriphosphate also stimulated phospholipase D activity more effectively in PKCS cells than in control cells, suggesting that upregulation of phospholipase D activity by PKC overexpression occurs distal to the thrombin receptor. These results suggest that PKC may act as a switch to up-regulate phosphatidylcholine-phospholipase D and down-regulate phosphoinositide-phospholipase C stimulations.
Many extracellular hormone and neurotransmitter receptors transduce signals into cells through activation of phosphatidylinositol-specific phospholipase C (PI-PLC)' (1). PI-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Although the stimulation of protein kinase C by DAG has been well characterized (6), less is known about the feedback regulation of PI-PLC and PC-PLD activities by PKC. Previous studies of this feedback regulation have relied mainly on pharmacological approaches to assess PKC function, such as short term activation of PKC with phorbol esters, inhibition with PKC inhibitors (7)(8)(9)(10)(11), or down-regulation of PKC by chronic exposure of cells to phorbol esters (12,13). However, phorbol esters have been reported to exert some effects that may be independent of PKC stimulation (14,15), and it has been suggested that effects of phorbol esters on phosphoinositide and PC hydrolysis are not mediated by PKC (16,17). In addition, these pharmacological approaches do not allow for investigation of the roles of individual PKC isozymes. We have previously used Rat 6 fibroblasts that stably overexpress the /31 isozyme of PKC (18) to study the feedback regulation of phorbol ester-stimulated phospholipase activity, and we found that PKC overexpression enhances phorbol ester-stimulated PC-PLD activity (19). We now report the effects of PKC-/31 overexpression on thrombin-, guanine nucleotide-, and calcium-stimulated PI-PLC and PC-PLD activities. Our results show that PKC overexpression downregulates agonist-stimulated phosphoinositide-specific phospholipase C activity with concomitant up-regulation of agonist-stimulated PC-PLD activity. These effects appear to occur, at least in part, through regulation of the coupling of regulatory GTP-binding protein(s) (G-proteins) to catalytic activity.  (21).
To monitor phospholipase D activity, cellular PC pools were labeled for 2 h with [3H]myristate, and phosphatidylethanol (PEt) formation in the presence of 0.5% ethanol was measured as described (19). PEt levels were expressed as percent of total phospholipid counts/min. In a typical experiment, specific activities of [3H]myristate-labeled phospholipids were 13305 cpm/nmol and 13913 cpm/ nmol for R6 C1 and R6 PKCB cells, respectively.
DAG mass levels of serum-deprived unlabeled cells were determined by conversion to [32P]phosphatidic acid using Escherichia coli DAG kinase as described previously (22) with minor modifications (11). DAG mass was normalized to phospholipid phosphate content of the extracts (23).
Cell Permeabilization-For measurement of effects of GTPrS or free calcium concentration, cells were permeabilized with saponin as described previously (24) with slight modifications. After labeling with [3H]inositol or [3H]myristic acid as described above, cell monolayers were washed twice with serum-free DMEM, incubated at 37 "C for 15 min, and then permeabilized for 5 min at 37 "C with 20 pg/ml saponin in an intracellular buffer containing 110 mM KCl, 10 mM NaCl, 1 mM KH2P04, 3 mM Na2ATP, 8 mM creatine phosphate, 6 units/ml creatine kinase, 20 mM HEPES, pH 7.0, and 4 mM MgC12, 1 mM EGTA, and 0.317 mM CaCl,, which were calculated to yield free calcium and magnesium concentrations of 0.1 pM and 1.25 mM, respectively. The dishes were then washed 3 times with intracellular buffer without saponin before treatment with drug in intracellular buffer as indicated. This permeabilization protocol rendered 86-89% of both R6 C1 and R6 PKCB cells permeable to trypan blue with as little loss of cellular protein as possible (20-30% of cellular lactate dehydrogenase released).
For experiments with varying free calcium concentrations, added CaC12 and MgC1, concentrations were calculated with a computer program to yield 1.25 mM free M$+ and the indicated [Ca2+] in the presence of 1 mM EGTA and 3 mM ATP at pH 7.0.
PHlPhorbol Dibutyrate Binding-To monitor the level of PKC expression during repeated passage of the cells, PKC was partially purified (through DEAE-Sephacel) from confluent cultures of R6 C1 and R6 PKCB cells as described previously (18), and [3H]PDBu binding was measured using the procedure of Sando and Young (25).

RESULTS
Stimulation of Inositol Phosphates by Thrombin-To assess the effects of protein kinase C overexpression on PI-PLC, thrombin-stimulated inositol phosphate levels were measured in control (R6 C1) and PKC-pl-overexpressing (R6 PKC3) cells.  a phospholipase D-mediated transphosphatidylation reaction in which PC-PLD transfers the phosphatidyl group to ethanol rather than water (26). Fig. 3 shows time courses of athrombin-stimulated PEt formation in R6 C1 and R6 PKC3 cells. a-Thrombin (9 nM) was clearly more effective in stimulating PEt formation in R6 PKC3 cells (4.9-fold in 1 min) than in control cells (1.1-fold in 1 rnin).' The stimulation of 1,2-diacylglycerol mass levels, which may be produced through both phosphoinositide and PC Following repeated passage of the R6 PKCB cell line, even in the presence of 50 pg/ml G418, the level of PKC overexpression diminished. This loss of PKC expression correlated with a loss of the suppression of a-thrombin-stimulated InsP3 formation. The enhanced PC-PLD response to @-thrombin in R6 PKCB cells reported here was also lost upon repeated passage and appeared to be lost prior to the loss of InSP3 suppression. We do not know the biochemical basis of this temporal difference. However, when new subcultures of R6 PKC3 cells were obtained at an early passage number, we again observed enhanced PC-PLD stimulation and suppressed PI-PLC stimulation by a-thrombin.

FIG. 4. Time course of inhibition of a-thrombin-stimulated diacylglycerol levels by PKC overexpression.
Following 24 h in 0.5% serum, confluent cultures were stimulated with 10 nM a-thrombin for the indicated times. Cellular lipids were extracted, and DAG and phospholipid phosphate content were measured. As we reported earlier (19), resting basal DAG mass levels were lower in R6 PKC3 cells (0.20 mol % in this experiment) than in control cells (0.29 mol %). Similar results were obtained in four separate experiments. metabolism (27), was also affected by PKC-Dl overexpression. The time course in Fig. 4 shows that 10 nM thrombin stimulated DAG production in a biphasic manner, as has been reported previously with fibroblasts (28). At early time points (15-30 s) that are thought to reflect stimulation of DAG production predominantly from phosphoinositide metabolism (27, 28), a rise in DAG mass was evident in both R6 C1 and R6 PKC3 cells. As with thrombin-stimulated InsP3 levels, this early DAG increase was clearly suppressed (73% inhibition) in R6 PKC3 cells relative to control cells. A second peak in DAG mass occurred after 5 min of thrombin stimulation in both cell types and was also suppressed (31% inhibition at 5 min) in R6 PKC3 cells.

Stimulations by GTPyS and [Ca'+] in Permeabilized Cells-
To localize the regulatory effects of PKC overexpression within the thrombin-stimulated signal transduction pathways, the effects of PKC-Pl overexpression on stimulations by GTPyS and [Ca"] were measured in saponin-permeabilized cells. In non-permeabilized R6 C1 cells, GTPyS in concentrations up to 100 PM had no effect on inositol polyphosphate levels (not shown). Fig. 5 shows that in saponin-permeabilized R6 C1 and R6 PKC3 cells, GTPyS over a range of 1-100 PM concentration stimulated InsPs levels. As observed with thrombin-stimulated inositol phosphates, GTPyS was less efficacious in R6 PKC3 cells (5-fold stimulation by 100 FM GTPyS at 5 min) than in R6 C1 cells (27-fold stimulation by 100 PM G T P r S at 5 min) in the stimulation of InsP, levels (Fig. 5 ) , and in the stimulation of InsP1, InsP', and InsP, levels (data not shown).
Since Ca2+ has been shown to stimulate isolated PI-PLC directly (29), we measured the effect of PKC overexpression on free calcium-stimulated inositol phosphate production as an index of PI-PLC activity distal to regulation by agonist receptors and G-proteins. Increasing free calcium concentrations stimulated [3H]InsP3 accumulation (Fig. 6) to a similar extent in saponin-permeabilized R6 C1 and R6 PKC3 cells. The dynamic range of stimulatory free calcium concentration (0.1-100 WM) is similar to that observed previously for phosphatidylinositol 4,5-bisphosphate hydrolysis by purified PI-PLC (29) or by membrane preparations (30).
To determine whether the up-regulation of PC-PLD activity by PKC overexpression occurs at a level after the thrombin receptor, the effect of GTPyS on [3H]PEt formation was measured with permeabilized R6 C1 and R6 PKCB cells. GTP+ stimulated PEt formation in both cell types, with maximal stimulation occurring at 30-100 PM GTPyS (Fig. 7). This maximally effective concentration of GTPyS is similar to that required for maximal stimulation of InsP3 accumulation. As observed with thrombin, GTPyS was more effective  in stimulating PEt formation in R6 PKCB cells than in R6 C1 cells.

DISCUSSION
Protein kinase C has been reported to uncouple PI-PLC activity from stimulation by activated hormone receptors or regulatory G-proteins (7,8,10,11,13,31), although the precise mechanism of this uncoupling has been a subject of controversy (8,10,13,31,32). This is the first report of the effects of PKC overexpression on receptor-or guanine nucleotidestimulated phospholipase activity. Taken together with previous reports that used phorbol esters as pharmacological tools to assess regulation by PKC (7)(8)(9)(10)(11)(12)(13)31), the present data demonstrate more directly that stimulations of PI-PLC and PC-PLD are down-regulated and up-regulated, respectively, by PKC.
Since our experiments use cells in which PKC expression is chronically elevated, there are several potential explanations for how stimulation of InsP3 levels may be inhibited. Although we interpret the inhibition of thrombin-stimulated InsP, levels by PKC overexpression as the result of direct PKC-mediated phosphorylation of a constituent(s) of the receptor/G-protein/PI-PLC pathway, an alternative possibility is that the decreased accumulation of InsP, results from effects of PKC on InsPs metabolism subsequent to inositol 1,4,5-trisphosphate formation. The finding that thrombinstimulated InsP1, InsP?, and InsP, levels are also attenuated by PKC overexpression even at early time points ( e g . 20 s) suggests that the effect is more likely to be on InsP, formation than metabolism. Furthermore, the other product of PIPz hydrolysis, DAG, is also suppressed at early time points, suggesting that desensitization by PKC overexpression occurs at the level of InsP, formation. The observed attenuation of thrombin-and GTPyS-stimulated InsPB accumulation could also result, in part, from effects of long term PKC overexpression on the expression of genes encoding the G-protein(s) or PI-PLC isozymes that mediate thrombin-induced phosphoinositide hydrolysis. Presently, there are no data that directly address this possibility.
Effects of PKC overexpression on GTPySor calciumstimulated PI-PLC and PC-PLD were measured to identify the step(s) subject to regulation by PKC. Thrombin is thought to stimulate PI-PLC activity through interaction with a Gprotein-coupled receptor (33,34). The inhibition of thrombin- and GTPyS-stimulated, but not free calcium-stimulated, InsP, formation by PKC overexpression suggests that the main inhibitory effect of PKC is on the G-protein or the coupling of the G-protein to PI-PLC. Because elevated free calcium concentration stimulates purified PI-PLC activity directly (29), the lack of inhibition of calcium-stimulated PI-PLC activity by PKC overexpression indicates that regulation by PKC occurs prior to the PI-PLC enzyme. Similar conclusions were reached by Orellana et al. (31) who reported that phorbol ester treatment of astrocytoma cells significantly inhibited GTPyS-stimulated, but not calcium-stimulated, formation of inositol phosphates. More recently, it was shown that phorbol ester treatment of several cell types stimulated phosphorylation of PI-PLC-p but not phosphorylation of PI-PLC-6 or PI-PLC-y (32). Interestingly, PI-PLC-P is the only PI-PLC isozyme that has been shown to be regulated by a heterotrimeric G-protein (35). The observation that in vitro phosphorylation of PI-PLC-p by PKC does not affect its activity has led Ryu et al. (32) to propose that this phosphorylation may prevent its activation by G-proteins.
Our observation that PKC overexpression enhances stimulation of PEt formation by a-thrombin and by GTPyS to similar extents suggests that stimulation of PC-PLD is also regulated by PKC at a level distal to the hormone receptor. This finding extends previous reports that phorbol ester treatment significantly enhanced PC-PLD activity in the presence of GTPyS but had little potentiating activity in the absence of GTPyS (9). Although PKC overexpression clearly enhanced a-thrombin-stimulated [,H]PEt formation (Fig. 3), a corresponding effect of PKC overexpression on [3H]phosphatidic acid production was not reproducibly observed in the presence or absence of ethanol (not shown). Presumably, [,HI PEt production is easily measurable because it is a stable end product (26), while phosphatidic acid, the physiological product of PC-PLD activity, is rapidly converted to metabolites such as CDP-diacylglycerol (36) and the putative second messenger lysophosphatidic acid (37, 38). In contrast to results in phorbol ester-stimulated fibroblasts (19), phosphatidic acid does not appear to be dephosphorylated to DAG to a measurable extent in thrombin-stimulated fibroblasts, since ethanol, which decreases phosphatidic acid formation by diverting phosphatidyl groups into PEt, has no effect on a-thrombin-stimulated DAG mass level^.^^^ Thus, the second phase of DAG mass stimulated by a-thrombin is probably derived through pathways other than the sequential action of PC-PLD and phosphatidic acid phosphohydrolase. Consistent with this, the second phase of DAG mass is not enhanced by PKC overexpression (Fig. 4), despite the observed enhancement of PC-PLD activity (Fig. 3).
Scheme I shows a working model for the regulation of PI-PLC and PC-PLD activities by PKC. Stimulation of the thrombin receptor by a-thrombin activates PI-PLC via a regulatory G-protein, perhaps G, (35). PI-PLC hydrolyzes PIP, to yield InsPs and DAG, resulting in stimulation of PKC. PKC then feeds back to inhibit further G-protein coupling to PI-PLC and to enhance G-protein coupling to PC-PLD activity. This shift from PI-PLC activity during the initial phase of the cellular response to PC-PLD activity during the later phase may serve to preserve cellular PIP,, which is less abundant than PC (39,40). Alternatively, InsPB and DAG, the products of PI-PLC activity, may be necessary for initiation of cellular responsiveness to hormones such as a-thrombin, while phosphatidic acid and its metabolites, which are derived from PC-PLD activity, may be necessary for sustenance of cellular responses.
providing the R6 C1 and R6 PKC3 cells, Dr. D. Raben for making a