J Biol Chem, Vol. 274, Issue 49, 34691-34698, December 3, 1999

Phospholipase D Stimulation by Receptor Tyrosine Kinases Mediated by Protein Kinase C and a Ras/Ral Signaling Cascade^*

Matthias Voß, Paschal A. Oude Weernink, Stephan Haupenthal, Ursula Möller, Robbert H. Cool^§, Bettina Bauer^§, Jacques H. Camonis^¶, Karl H. Jakobs, and Martina Schmidt

From the  Institut für Pharmakologie, Universitätsklinikum Essen, D-45122 Essen, Germany, the ^§ Max-Planck-Institut für Molekulare Physiologie, D-44227 Dortmund, Germany, and the ^¶ INSERM U-248, F-75248 Paris Cedex 05, France

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Stimulation of phospholipase D (PLD) in HEK-293 cells expressing the M₃ muscarinic receptor by phorbol ester-activated protein kinase C (PKC) apparently involves Ral GTPases. We report here that PKC, but not muscarinic receptor-induced PLD stimulation in these cells, is strongly and specifically reduced by expression of dominant-negative RalA, G26A RalA, as well as dominant-negative Ras, S17N Ras. In contrast, overexpression of the Ras-activated Ral-specific guanine nucleotide exchange factor, Ral-GDS, specifically enhanced PKC-induced PLD stimulation. Moreover, recombinant Ral-GDS potentiated Ral-dependent PKC-induced PLD stimulation in membranes. Epidermal growth factor, platelet-derived growth factor, and insulin, ligands for receptor tyrosine kinases (RTKs) endogenously expressed in HEK-293 cells, apparently use the PKC- and Ras/Ral-dependent pathway for PLD stimulation. First, PLD stimulation by the RTK agonists was prevented by PKC inhibition and PKC down-regulation. Second, expression of dominant-negative RalA and Ras mutants strongly reduced RTK-induced PLD stimulation. Third, overexpression of Ral-GDS largely potentiated PLD stimulation by the RTK agonists. Finally, using the Ral binding domain of the Ral effector RLIP as an activation-specific probe for Ral proteins, it is demonstrated that endogenous RalA is activated by phorbol ester and RTK agonists. Taken together, strong evidence is provided that RTK-induced PLD stimulation in HEK-293 cells is mediated by PKC and a Ras/Ral signaling cascade.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Phospholipase D (PLD)¹ catalyzes formation of phosphatidic acid from the major membrane phospholipid, phosphatidylcholine (PtdCho), and this reaction is implicated in the regulation of diverse cellular processes, such as vesicular trafficking and cell growth and differentiation. A large variety of receptor tyrosine kinases (RTKs) and receptors coupled to heterotrimeric G proteins in a wide range of cell types has been reported to mediate PLD stimulation in response to their specific agonists. However, the mechanisms of receptor signaling to PLD in intact cells are only poorly understood and seem to involve distinct signal transduction components, in particular different small GTPases and protein kinase C (PKC) isoforms (for reviews, see Refs. 1-5).

In HEK-293 cells, stably expressing the M₃ muscarinic acetylcholine receptor (mAChR), GTPases of distinct families, such as ADP-ribosylation factor (ARF), Rho, and Ral, as well as various protein kinases, such as PKC, Rho kinase, and tyrosine kinases, apparently mediate PLD activation. Specifically, stimulation of PLD by the G protein-coupled M₃ mAChR is dependent on ARF and Rho GTPases and apparently involves a tyrosine kinase and Rho kinase but not PKC. On the other hand, PLD stimulation by phorbol ester-activated PKC, which is phosphorylation-dependent in HEK-293 cells, is apparently ARF-independent and only poorly inhibited by inactivation of Rho proteins (6-11). Recently, evidence has been provided that Ral GTPases are involved in PKC-induced PLD stimulation in HEK-293 cells (12).

Previously, Ral GTPases have been reported to be involved in Ras-mediated PLD activation in v-Src-transformed Balb/c- and NIH-3T3 fibroblasts (13, 14). Moreover, RalA has recently been shown to interact directly with PLD1 and to enhance ARF-stimulated PLD1 activity (15-17). As Ral proteins are activated by Ras-controlled Ral-specific guanine nucleotide exchange factors (Ral-GEFs), such as Ral-GDS, Rgl, and Rlf (for review, see Ref. 18), Ral-induced PLD stimulation has been suggested to involve a Ras/Ral-GEF/Ral signaling cascade (14, 18-21). Thus, the aim of this study was to examine the existence of a Ras/Ral signaling cascade in Ral-mediated PLD stimulation by PKC in HEK-293 cells and to identify receptors stimulating PLD by such a signaling pathway. We report here that the Ral-dependent PLD stimulation by PKC involves Ras and a Ral-GEF. Most important, it is demonstrated that PLD stimulation in HEK-293 cells by endogenously expressed RTKs is mediated by PKC and a Ras/Ral signaling pathway.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- [³H]Oleic acid (5 Ci/mmol) and 1-palmitoyl-2-[9,10-³H]palmitoylglycerophosphocholine ([³H]PtdCho, 89 Ci/mmol) were obtained from NEN Life Science Products. Unlabeled PtdCho, phorbol 12-myristate 13-acetate (PMA), and insulin were from Sigma, epidermal growth factor (EGF) and staurosporine were from Biomol, and Gö 6976 and PD98059 were from Calbiochem. Glutathione-Sepharose was from Amersham Pharmacia Biotech, phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P₂) was from Roche Molecular Biochemicals, and platelet-derived growth factor (PDGF) was from Falcon. The antibodies against RalA and Ras were obtained from Transduction Laboratories, and those against ERK1 and phospho-p44/p42 mitogen-activated protein (MAP) kinase were from Santa Cruz and New England Biolabs, respectively. Clostridium difficile toxin B-1470 was purified as described (12).

Plasmids-- The dominant-negative RalA mutant, G26A RalA, was subcloned in pRK5 (22). The dominant-negative Ras mutant, S17N Ras, subcloned in pRSV was kindly provided by Dr. R. M. F. Wolthuis (23, 24). Ral-GDS subcloned in pEXVKF was kindly provided by Dr. N. van den Berghe. For protein purification, Ral-GDS (amino acids 312-852, containing the Ral-GEF domain and the Ras binding domain) and the Ral binding domain of RLIP (amino acids 397-518) were subcloned in pGEX-4T3 (25, 26).

Cell Culture and Transfection-- Culture conditions of HEK-293 cells stably expressing the M₃ mAChR were as reported in detail before (10). For experiments, cells subcultured in Dulbecco's modified Eagle's medium/F-12 medium were grown to near confluence (145-mm culture dishes). One day before transfection, cells were supplied with 20 ml of fresh medium. Cells were transfected with the indicated concentrations of G26A RalA DNA, S17N Ras DNA, Ral-GDS DNA, or the corresponding empty vectors using the calcium phosphate method (8, 9). Transfection efficiency ranged between 50 and 80%, as revealed by in situ -galactosidase assay in cells cotransfected with constitutively active pSV-gal (Promega). Expression of the proteins was verified by immunoblotting of cell lysates with specific antibodies.

Cell Treatment and Assay of PLD Activity in Intact Cells-- For PKC down-regulation, cells were treated for 16 h with 100 nM PMA or, for control, with 0.1% dimethyl sulfoxide. For inhibition of PKC, cells were treated for 30 min at 4 °C with the indicated concentrations of Gö 6976 or staurosporine or, for control, with 0.1% dimethyl sulfoxide. Where indicated, cells were incubated for 24 h without and with 300 pg/ml toxin B-1470. Transfected cells were replated 24 h after transfection on 145-mm culture dishes. To evoke responsiveness to RTK agonists, cells were serum-starved for at least 36 h before measurement of PLD activity. For this, cellular phospholipids were labeled by incubation of the cells with [³H]oleic acid (2 µCi/ml) in medium. Afterward, cells were detached from the culture dishes, washed twice in Hanks' balanced salt solution containing 118 mM NaCl, 5 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, and 5 mM D-glucose, buffered at pH 7.4 with 15 mM HEPES, and resuspended at a cell density of 1 × 10⁷ cells/ml. PLD activity was assayed for 60 min at 37 °C in a total volume of 200 µl containing 100 µl of the cell suspension (1 × 10⁶ cells), 2% ethanol, and the indicated stimulatory agents. Isolation of labeled phospholipids and the specific PLD product, [³H]phosphatidylethanol ([³H]PtdEtOH), was performed as previously reported (10). Formation of [³H]PtdEtOH is expressed as percentage of total labeled phospholipids.

Assay of PLD Activity in Membranes-- HEK-293 cells treated for 24 h without or with 300 pg/ml toxin B-1470 were collected, and membranes were prepared as described before (12). To measure PLD activity, [³H]PtdCho was mixed with PtdIns(4,5)P₂ in a molar ratio of 8:1, dried, and resuspended in 50 mM HEPES, pH 7.5, 3 mM EGTA, 80 mM KCl, and 1 mM dithiothreitol, followed by sonication on ice. PLD activity was determined as described (12) with [³H]PtdCho/PtdIns(4,5)P₂ (200 µM, 25 µM) as substrate vesicles and 200 µg of membrane protein for 60 min at 37 °C.

Purification of Recombinant Proteins-- Escherichia coli were transformed with pGEX-4T3 containing Ral-GDS or the Ral binding domain (RalBD) of RLIP and grown overnight in LB at 37 °C or 30 °C, respectively. Expression of fusion proteins containing an NH₂-terminal glutathione S-transferase (GST) domain was induced by adding 0.1 mM isopropyl-1-thio--D-galactopyranoside to the culture medium for 3 h. Purification of the proteins was performed essentially as described before with some modifications (8, 9, 12). In brief, bacteria were sonicated on ice in resuspension buffer containing 50 mM NaCl, 5 mM MgCl₂, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, and 50 mM Tris-HCl, pH 7.5, and the crude membrane fraction was removed by centrifugation. For purification of GST-RalBD, the corresponding supernatant was incubated with glutathione-Sepharose beads overnight at 4 °C. Unbound proteins were removed by several washings with resuspension buffer containing 10% glycerol and 5 mM dithiothreitol. Purified GST-RalBD was stored at 4 °C in a buffer containing 150 mM NaCl, 5 mM MgCl₂, 5 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, 20 µM leupeptin, 10% glycerol, and 50 mM Tris-HCl, pH 7.4. For purification of GST-Ral-GDS, the corresponding supernatant was incubated with glutathione-Sepharose beads for 30 min at 4 °C. Thereafter, the beads were washed three times with resuspension buffer, and the parent GST fusion protein was released from the beads by incubation with thrombin (10 units) overnight at 4 °C in a buffer containing 150 mM NaCl, 5 mM MgCl₂, 2.5 mM CaCl₂, 1 mM dithiothreitol, and 50 mM Tris-HCl, pH 8. Afterward, the beads were removed by centrifugation, and the excess of thrombin was removed by the addition of p-aminobenzamidine beads. The homogeneity of the isolated recombinant proteins was analyzed by Coomassie Blue staining of SDS-PAGE gels. Purified Ral-GDS protein exhibiting specific Ral-GEF activity was stable for 4 weeks. Purification of recombinant RalA (COOH-terminus-truncated sRalA, amino acids 1-177) was reported before (12).

Determination of RalA Activity-- The activity state of RalA was measured with GST-RalBD as an activation-specific probe for Ral proteins as described before (25, 26). In brief, confluent and serum-depleted (36 h) HEK-293 cells on 90-mm culture dishes were treated with the indicated agents for 30 min at 37 °C. Thereafter, the cells were lysed in a buffer containing 15% glycerol, 1% Nonidet P-40, 200 mM NaCl, 2.5 mM MgCl₂, 1 mM phenylmethylsulfonyl fluoride, 0.1 µM aprotinin, 1 µM leupeptin, 10 µg/ml soybean trypsin inhibitor, and 50 mM Tris-HCl, pH 7.4. Lysates were clarified by centrifugation, and the resulting supernatants were incubated with 15 µg of purified GST-RalBD bound to glutathione-Sepharose beads for 1 h at 4 °C. Thereafter, the beads were washed 3 times in lysis buffer and finally incubated in Laemmli buffer for 10 min at 95 °C.

MAP Kinase Activation-- Twenty-four h after transfection, HEK-293 cells were replated on 35-mm dishes, serum-starved for 36 h, and then stimulated with EGF and insulin for 5 min at 37 °C. After lysis of the cells in a buffer containing 1% SDS and 10 mM Tris-HCl, pH 7.4, and 5 passages through a 25-gauge needle, the lysates were clarified by centrifugation. After determination of protein concentration by the BCA method (Pierce), 75 µg of protein of each supernatant sample was incubated in Laemmli buffer for 10 min at 95 °C.

Immunoblot Analysis-- For detection of RalA and Ras, proteins were separated by SDS-PAGE on 12.5% acrylamide gels, transferred to nitrocellulose membranes, and blotted with anti-RalA antibody (dilution 1:5000, 1 h incubation) and anti-Ras antibody (dilution 1:400, 1 h incubation), respectively. For detection of phosphorylated and total amount of MAP kinases, proteins were separated by SDS-PAGE on 10% acrylamide gels, transferred to nitrocellulose membranes, and blotted with anti-phospho-MAP kinase antibody (dilution 1:1000, 16 h incubation at 4 °C) and anti-ERK1 antibody (0.4 µg/ml, 1-h incubation), respectively.

Data Presentation-- Data shown in figures are mean ± S.D. from one experiment performed in triplicate, and repeated as indicated in the figure legends. Data mentioned in the text are the mean ± S.E., with n providing the number of independent experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Participation of Ral, Ras, and Ral-GEF in PLD Stimulation by PKC in HEK-293 Cells-- We have recently reported that PLD stimulation in M₃ mAChR-expressing HEK-293 cells by phorbol ester-activated PKC, but not that induced by the M₃ mAChR, is largely reduced by treatment of the cells with C. difficile toxin B-1470 and Clostridium sordellii lethal toxin (12). These two toxins glucosylated members of the Ras GTPase family, suggesting that these GTPases are involved in PLD stimulation by PKC. Using recombinant proteins, it was demonstrated that specifically Ral (A and B) proteins are required for PLD stimulation by PMA-activated PKC in membranes of HEK-293 cells (12). To study whether a Ras/Ral signaling cascade is involved in PLD stimulation by PKC, we first examined the effects of expression of dominant-negative RalA and Ras mutants on basal and stimulated PLD activities in intact HEK-293 cells. Transfection of the cells with dominant-negative RalA, G26A RalA, resulted in a specific reduction of PLD stimulation by PMA, without altering basal PLD activity (Fig. 1A). In cells transfected with 100 µg of G26A RalA DNA/145-mm culture dish, the stimulatory effect of PMA (100 nM) on PLD activity was reduced to 41 ± 3% (n = 3) compared with that observed in control cells. In contrast, stimulation of PLD by the M₃ mAChR agonist, carbachol (1 mM), was not altered by transfection of HEK-293 cells with dominant-negative RalA (Fig. 1B).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Specific inhibition of PMA-induced PLD stimulation by dominant-negative RalA and Ras. HEK-293 cells on 145-mm culture dishes were transfected with empty vectors (pRK5/pRSV), G26A RalA, or S17N Ras (100 µg of DNA each). After 72 h, PLD activity was measured in the absence (Basal) and presence of 100 nM PMA (A) or 1 mM carbachol (B). Upper panel, immunoblot detection of RalA and Ras in lysates of transfected cells. Data are representative of at least three experiments.

Transfection of the cells with dominant-negative Ras, S17N Ras, resulted in a similar decrease in PLD stimulation by PMA as transfection with dominant-negative RalA (Fig. 1A). In cells transfected with 100 µg of S17N Ras DNA/145-mm culture dish, PLD stimulation by PMA was reduced to 47 ± 4% (n = 3) compared with control cells. Again, basal PLD activity and PLD stimulation by carbachol were not altered in cells expressing dominant-negative Ras (Fig. 1B). Together, these data indicated that PLD stimulation in HEK-293 cells by phorbol ester-activated PKC involves both Ral and Ras GTPases.

Activation of Ral is caused by Ral-GEFs such as Ral-GDS, Rgl, and Rlf, which are downstream targets of active Ras (for review, see Ref. 18). To study whether the Ral- and Ras-dependent PLD stimulation by PKC in HEK-293 cells involves a Ras-activated Ral-GEF, we studied the effects of cell transfection with the ubiquitously expressed Ral-GDS on PLD stimulation in intact cells and of recombinant Ral-GDS on PLD activities in HEK-293 cell membranes. As shown in Fig. 2A, overexpression of Ral-GDS markedly increased PMA-induced PLD stimulation. In cells transfected with 50 µg of Ral-GDS DNA/145-mm culture dish, PLD stimulation by PMA was increased by 86 ± 8% (n = 4). In contrast, basal PLD activity and PLD stimulation by carbachol were not altered in Ral-GDS-overexpressing cells (Fig. 2B).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Specific potentiation of PMA-induced PLD stimulation by Ral-GDS. HEK-293 cells were transfected with 50 µg of DNA of pEXVKF (0) or the indicated concentrations of Ral-GDS DNA in pEXVKF. After 72 h, PLD activity was measured in the absence (Basal) and presence of 100 nM PMA (A) or 1 mM carbachol (B). Data are representative of four experiments.

Addition of purified recombinant Ral-GDS to membranes of control HEK-293 cells had no effect on basal PLD activity (Fig. 3A). PLD stimulation by PMA-activated PKC in HEK-293 cell membranes requires MgATP (11, 12). Accordingly, Ral-GDS had also no effect on PLD activity measured in the presence of PMA but without MgATP (data not shown). However, in the presence of MgATP (1 mM) plus PMA (100 nM), which increased basal PLD activity about 2-fold, the addition of Ral-GDS (1.5 µM) caused a marked increase in PKC-stimulated PLD activity (Fig. 3A). As reported before (12), the addition of recombinant RalA had no effect on basal and PMA-stimulated PLD activities in control membranes. Furthermore, in the presence of Ral-GDS, further addition of RalA (10 µM) was without effect on basal and PMA-stimulated PLD activities. These data suggested that the potentiation of PMA-stimulated PLD activity by Ral-GDS is either independent of Ral proteins or that Ral-GDS acts via endogenous Ral proteins present in the membranes in sufficient amounts. To resolve this, the effect of Ral-GDS on PLD activity was studied in membranes of HEK-293 cells pretreated for 24 h with toxin B-1470, which caused inactivation of endogenous Ral proteins (12). As shown in Fig. 3B, PMA-stimulated PLD activity was reduced by 50% in these membranes, and addition of 10 µM recombinant RalA partially restored PMA-induced PLD stimulation. Most important, the addition of Ral-GDS, which strongly increased PMA-stimulated PLD activity in control membranes, was without any effect on PLD activity measured in the presence of MgATP plus PMA in membranes of toxin B-1470-treated cells. However, when RalA was added, the addition of Ral-GDS strongly increased PMA-stimulated PLD activity in membranes of toxin B-1470-treated HEK-293 cells as well (Fig. 3B). The fact that PMA-stimulated PLD activity measured with exogenous RalA and Ral-GDS in these membranes did not reach the same level as that observed in control membranes (in the presence of Ral-GDS) is probably due to a distinct efficiency of endogenous Ral present in control membranes and exogenously added RalA, which is a COOH-terminus-truncated and thus not post-translationally modified version of RalA (amino acids 1-177). The lack of post-translational modification may also explain why rather high concentrations of recombinant RalA were required for the restoration of PMA-stimulated PLD activity in membranes of toxin B-1470-treated cells (12). As reported before, post-translational modifications of both Ras and Ral are important for Ral-GDS-dependent Ral activation (27, 28). Alltogether, these data suggested that the Ras effector, Ral-GDS, potentiates the Ral-dependent PLD stimulation by PKC by an action on the Ral proteins, and thus, that a Ras/Ral-GEF/Ral signaling cascade is involved in PLD stimulation by PKC in HEK-293 cells.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Effects of recombinant Ral-GDS on PMA-stimulated PLD activity in membranes of HEK-293 cells. PLD activity was measured in membranes of HEK-293 cells pretreated for 24 h without (A, Control) and with 300 pg/ml toxin B-1470 (B) using [3H]PtdCho/PtdIns(4,5)P2 vesicles as enzyme substrate in the absence (Basal, open columns) and presence of 100 nM PMA plus 1 mM MgATP (PMA, filled columns) without and with 10 µM recombinant RalA, 1.5 µM recombinant Ral-GDS, or RalA plus Ral-GDS as indicated. Data are representative of three to four experiments.

PKC- and Ras/Ral-dependent PLD Stimulation by RTKs in HEK-293 Cells-- The Ras/Ral-dependent pathway of PLD stimulation induced by phorbol ester-activated PKC is obviously not used by the G protein-coupled M₃ mAChR expressed in HEK-293 cells. Therefore, to identify receptors stimulating PLD activity in HEK-293 cells via this pathway, we studied the effects of EGF, PDGF, and insulin on PLD activity. Receptors for these growth factors are endogenously expressed in HEK-293 cells (29, 30). As illustrated in Fig. 4, the three RTK agonists, EGF (100 ng/ml), PDGF (20 ng/ml), and insulin (10 µg/ml), increased PtdEtOH production in HEK-293 cells nearly as efficient as the M₃ mAChR agonist carbachol. PLD stimulation by EGF and PDGF was mediated by their respective specific receptors, as demonstrated with the tyrphostins AG1478 and AG1296, which largely and specifically reduced PLD stimulation by EGF and PDGF, respectively (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Inhibition of RTK-induced PLD stimulation by PKC inhibition and PKC down-regulation. M3 mAChR-expressing HEK-293 cells prelabeled with [3H]oleic acid were pretreated for 30 min at 4 °C with 0.1% dimethyl sulfoxide (Control) or 100 nM Gö 6976 (A) or for 16 h with 0.1% dimethyl sulfoxide (Control) or 100 nM PMA (B). Thereafter, PLD activity was determined without (Basal) and with 1 mM carbachol (Carb), 100 ng/ml EGF, 20 ng/ml PDGF, or 10 µg/ml insulin (Ins). Data are representative of five similar experiments.

Having established that PLD activity is stimulated in HEK-293 cells by RTKs, we first examined whether PKC is involved in this RTK action. For this, cells were treated for 30 min with the PKC inhibitors, Gö 6976 and staurosporine (100 nM each), or alternatively, PKC was down-regulated by long term (16 h) treatment of the cells with PMA (100 nM). As reported before, both types of treatment blocked PLD stimulation by PMA but failed to affect PLD stimulation by the M₃ mAChR (10-12). In contrast, as shown in Fig. 4A, PLD stimulation induced by the RTK agonists, EGF, PDGF, and insulin, was nearly fully prevented in cells pretreated with the PKC inhibitors Gö 6976 and staurosporine (data not shown). Similarly, in HEK-293 cells long term-treated with PMA, the three RTK agonists failed to increase PLD activity (Fig. 4B). PKC down-regulation did not alter the overall tyrosine phosphorylation pattern stimulated by the RTK agonists (data not shown). Thus, activation of EGF, PDGF, and insulin receptors endogenously expressed in HEK-293 cells induces PLD stimulation, and this RTK-mediated response, in contrast to that induced by the G protein-coupled M₃ mAChR, is apparently mediated by PKC.

To further characterize the PLD signaling pathway of RTKs, we examined the effects of C. difficile toxin B-1470, which glucosylates and thereby inactivates Rac, Rap, and Ral GTPases (12). Treatment of HEK-293 cells for 24 h with 300 pg/ml toxin B-1470, previously shown to attenuate PMA-induced PLD stimulation (12), fully blocked stimulation of PLD activity by EGF, PDGF, and insulin (Fig. 5). As reported before (12), PLD stimulation by carbachol was not affected by toxin B-1470. On the other hand, treatment of the cells with C. difficile toxin B (100 pg/ml, 24 h), known to inactivate Rho, Rac, and Cdc42 and previously shown to inhibit carbachol-induced PLD stimulation (7), did not alter PLD stimulation by the RTK agonists (data not shown). These data thus suggested that PLD stimulation by RTKs in HEK-293 cells is mediated by the same set of GTPases as that involved in PMA-induced PLD stimulation. To verify this hypothesis, HEK-293 cells were transfected with dominant-negative mutants of RalA and Ras, followed by measurement of RTK-stimulated PLD activity. As illustrated in Fig. 6A, stimulation of PLD activity by EGF in cells expressing dominant-negative G26A RalA, was strongly reduced. In cells transfected with 100 µg of G26A RalA DNA/145-mm culture dish, the stimulatory effect of EGF (100 ng/ml) on PLD activity was reduced to 35 ± 9% (n = 3) compared with that observed in control cells. Expression of dominant-negative Ras resulted in a similar decrease in EGF-induced PLD stimulation. In cells transfected with 100 µg of S17N Ras DNA/145-mm culture dish, PLD stimulation by EGF was reduced to 44 ± 8% (n = 3). A similar decrease in RTK-mediated PLD stimulation was seen for insulin (10 µg/ml) in G26A RalA-expressing cells (reduction to 36 ± 7%, n = 4) and for EGF in cells cotransfected with G26A RalA and S17N Ras (reduction to 30 ± 8%, n = 3) (data not shown). In contrast to its inhibitory effect on RTK-induced PLD stimulation, expression of dominant-negative RalA did not affect RTK-induced activation of MAP kinases. In cells transfected with 100 µg of G26A RalA DNA/145-mm culture dish, EGF (100 ng/ml) and insulin (10 µg/ml, not shown) induced phosphorylation of p42 and p44 MAP kinases similarly as in control cells (Fig. 6B). On the other hand, treatment of the cells with PD98059 (10 µM, 15 min), which fully prevented RTK-mediated MAP kinase activation, did not alter PLD stimulation by the RTK agonists (data not shown). Thus, RTK-induced PLD stimulation in HEK-293 cells not only involves PKC but also Ral and Ras GTPases.

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Inhibition of RTK-induced PLD stimulation by toxin B-1470. HEK-293 cells prelabeled with [3H]oleic acid were treated for 24 h without (Control) and with 300 pg/ml toxin B-1470. Thereafter, PLD activity was measured in the absence (Basal) and presence of 1 mM carbachol (Carb), 100 ng/ml EGF, 20 ng/ml PDGF, or 10 µg/ml insulin (Ins).

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Effects of dominant-negative RalA and Ras on EGF-induced PLD stimulation and MAP kinase activation. A, HEK-293 cells were transfected with empty vectors (pRK5/pRSV), G26A RalA, or S17N Ras (100 µg of DNA each). After 72 h, PLD activity was measured in the absence (Basal) and presence of 100 ng/ml EGF. B, HEK-293 cells transfected with empty vector () or G26A RalA (+) as above were stimulated for 5 min at 37 °C without (Basal) and with 100 ng/ml EGF. Phosphorylated p44 and p42 MAP kinases (MAPK) were detected in cell lysates with an anti-phospho-MAP kinase antibody. Shown is a representative immunoblot. The total amount of MAP kinases, detected with an anti-ERK1 antibody, was not affected by expression of G26A RalA (not shown).

Overexpression of Ral-GDS specifically enhanced PLD stimulation by PMA (see Fig. 2). Therefore, we studied whether overexpression of this Ras-activated Ral-GEF may have a similar effect on PLD stimulation by RTK agonists. As illustrated in Fig. 7A, transfection of HEK-293 cells with Ral-GDS induced a large potentiation of RTK-induced PLD stimulation. For example, in cells transfected with 50 µg of Ral-GDS DNA/145-mm culture dish, the stimulatory effect of EGF (100 ng/ml) and insulin (10 µg/ml) on PLD activity was increased by 264 ± 23% (n = 4) and 230 ± 10% (n = 3), respectively.

View larger version (44K): [in this window] [in a new window]   Fig. 7.   Potentiation of EGF- and insulin-induced PLD stimulation by Ral-GDS; activation of RalA by PMA, EGF, and insulin. A, HEK-293 cells were transfected with 50 µg of DNA of pEXVKF (0) or the indicated concentrations of Ral-GDS DNA in pEXVKF. After 72 h, PLD activity was measured in the absence (Basal) and presence of 100 ng/ml EGF or 10 µg/ml insulin. B, HEK-293 cells were stimulated without (Basal) and with 100 nM PMA, 100 ng/ml EGF, or 10 µg/ml insulin. Thereafter, active RalA was extracted from cell lysates with GST-RalBD bound to glutathione-Sepharose beads, followed by SDS-PAGE and immunoblotting with an anti-RalA antibody as described under "Experimental Procedures." Data are representative of at least three similar experiments.

Finally, as PLD stimulation by PMA and RTK agonists involves Ral, we studied whether endogenous RalA is activated by these agents. For this, we applied the recently developed method, which uses the Ral binding domain of the Ral effector RLIP to selectively extract active Ral from cell lysates (25, 26). As shown in Fig. 7B, the amount of RalA, thus of active RalA, extracted with the Ral binding domain of RLIP from lysates of HEK-293 cells treated with either PMA (100 nM), EGF (100 ng/ml), or insulin (10 µg/ml) was strongly increased compared with untreated controls, indicating that these agents activate endogenous RalA in HEK-293 cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study, we provide evidence that a Ras/Ral signaling cascade is involved in stimulation of PLD activity by PKC in HEK-293 cells. Furthermore, and most important, strong evidence is provided that PLD stimulation by RTKs for EGF, PDGF, and insulin, but not by the G protein-coupled M₃ mAChR, expressed in these cells, is mediated by a PKC- and Ras/Ral-dependent signaling pathway. The evidence is based on the following major findings. First, similar to inactivation of Ral proteins by C. difficile toxin B-1470, expression of dominant-negative RalA largely and specifically reduced PLD stimulation by phorbol ester-activated PKC and ligand-activated RTKs. Second, expression of dominant-negative Ras resulted in a similar specific inhibition of PLD stimulation. Third, endogenously expressed RalA was activated by PKC and RTKs. Fourth, overexpression of the Ras-activated Ral-GEF, Ral-GDS, specifically potentiated PLD stimulation by PKC and RTKs in intact HEK-293 cells. Fifth, recombinant Ral-GDS potentiated PKC-induced PLD stimulation in membranes of HEK-293 cells in a Ral-dependent manner. Finally, PLD stimulation by each of the three RTK agonists, i.e. EGF, PDGF, and insulin, was fully prevented by PKC inhibition or PKC down-regulation.

Previous studies in HEK-293 cells stably expressing the M₃ mAChR indicated that signaling to PLD is mediated by at least two distinct pathways. Stimulation of PLD by the G protein-coupled M₃ mAChR is dependent on ARF and Rho GTPases and involves a tyrosine kinase and Rho kinase, acting apparently upstream and downstream of Rho proteins, respectively, but not PKC (6-10). In contrast, PLD stimulation by phorbol ester-activated PKC, which is phosphorylation-dependent in HEK-293 cells, apparently involves the Ras-related Ral GTPases (11, 12). As Ral proteins are activated by Ral-specific GEFs, such as Ral-GDS, Rgl, and Rlf, which are under control of Ras proteins (18), we first wanted to know whether the PKC-induced PLD stimulation is mediated by a Ras/Ral signaling cascade. Expression of either dominant-negative Ras or RalA markedly and specifically reduced PKC-induced PLD stimulation. As dominant-negative G26A RalA, similar to dominant-negative S17N Ras, is constitutively in the GDP-bound form (18, 22-24), its inhibitory effect on PKC-induced PLD stimulation is probably due to sequestration of endogenous Ras-activated Ral-GEFs and, thus, interruption of a Ras/Ral-GEF/Ral signaling pathway. Furthermore, it is shown that overexpression of the ubiquitously expressed Ras-GEF, Ral-GDS, specifically enhanced PKC-induced PLD stimulation. Finally, recombinant Ral-GDS potentiated PKC-induced PLD stimulation in membranes of HEK-293 cells but only in the presence of functional Ral proteins. As Ral-GDS and Ral had no effect on basal PLD activity but required the presence of activated PKC and, on the other hand, efficient PLD stimulation by activated PKC required the presence of Ral (12), it has to be concluded that for productive PKC-induced PLD stimulation in HEK-293 cells both PKC itself and the Ras/Ral-GEF-activated Ral GTPases are required. In line with a cooperative action of Ral and PKC on PLD activity, Del Peso et al. (31) report that phorbol ester-stimulated PLD activity is highly enhanced in NIH-3T3 cells expressing oncogenic Ras, which by itself had no or only a minor effect on basal PLD activity. As PMA induced activation of endogenous RalA in HEK-293 cells, it can additionally be concluded that PMA activates Ras and, as a consequence, Ral proteins. Recently, various mechanisms of Ras activation by phorbol esters have been reported (32-34). Which of these activation mechanisms is induced by PMA in HEK-293 cells remains to be determined.

The second major aim of this study was to identify receptors mediating PLD stimulation by the PKC- and Ras/Ral-dependent pathway. As the G protein-coupled M₃ mAChR obviously did not use this pathway for PLD stimulation, PLD stimulation by RTKs for EGF, PDGF, and insulin endogenously expressed in HEK-293 cells was examined. Here we demonstrate that these RTK agonists efficiently increased PLD activity and that this RTK action was completely blocked by PKC inhibition or PKC down-regulation. In line with this finding, which has also been reported for RTK-mediated PLD stimulation in various other cell types (for reviews, see Refs. 1-5), we observed that the three RTK agonists increased phospholipase C activity in HEK-293 cells and induced rather long-lasting translocation of PKC isoforms, most prominent of which is PKC-, to membrane compartments (data not shown). Most important, similar to PKC-induced PLD stimulation, stimulation of RTK-mediated PLD activity was blocked by inactivation of Ras-related GTPases with C. difficile toxin B-1470 but not affected by inactivation of Rho GTPases with C. difficile toxin B. Furthermore, expression of dominant-negative RalA specifically reduced RTK-mediated PLD stimulation without affecting MAP kinase activation. A similar reduction in RTK-mediated PLD stimulation was observed in cells expressing dominant-negative Ras. Finally, similar to PMA, RTK agonists activated endogenous RalA in HEK-293 cells, and overexpression of the Ras-activated Ral-GEF, Ral-GDS, largely potentiated RTK-mediated PLD stimulation, reaching PLD activity levels similar to that observed in cells stimulated with PMA. Based on these data, it is proposed that RTK-mediated PLD stimulation in HEK-293 cells is induced by two major signaling pathways known to be under control of such receptors (Fig. 8). First, by activation of phospholipase C- isoforms and, consequently, increased diacylglycerol production, PKC isoforms are activated by RTKs. Second, by activation of Ras-specific GEFs, such as SOS, via the adaptor protein Grb2, RTKs induce activation of Ras, which then in turn activates Ral-GEFs and, as a consequence, Ral proteins. Because PLD stimulation by the RTK agonists in control cells was clearly less than that induced by PMA but elevated to this level in cells overexpressing Ral-GDS, it may be concluded that activation of Ral proteins by RTKs is less efficient that than induced by PMA and, thus, is the limiting factor for RTK-mediated PLD stimulation in HEK-293 cells. It remains to be determined which of the ubiquitously expressed Ral-GEFs, Ral-GDS, Rgl, and Rlf (18), specifically mediates RTK-induced activation of Ral proteins and PLD activity.

View larger version (11K): [in this window] [in a new window]   Fig. 8.   A model for RTK-mediated PLD stimulation in HEK-293 cells. Activation of RTKs for EGF, PDGF, and insulin stimulates PLD activity in HEK-293 cells by a concerted action of RTK-activated PKC and Ras/Ral-GEF-activated Ral proteins. PLC, phospholipase C-; DAG, diacylglycerol. For further explanation, see text.

Ras proteins can signal via several distinct effector pathways (18, 21) and may also thereby regulate PLD activity. First, active Ras can stimulate production of phosphatidylinositol 3,4,5-trisphosphate by phosphoinositide 3-kinase. This polyphosphoinositide in turn has been shown to bind to and activate ARF-specific GEFs such as GRP-1 and ARNO both in vitro and in intact cells (35-37), thus leading to activation of PLD-stimulating ARF proteins. Accordingly, insulin has been reported to translocate ARNO to the plasma membrane in 3T3 L1 adipocytes in a phosphoinositide 3-kinase-dependent manner (37). Furthermore, insulin-induced PLD stimulation in rat adipocytes and Rat-1 fibroblasts overexpressing human insulin receptors has been shown to be mediated by ARF proteins and to involve phosphoinositide 3-kinase (38-40). However, treatment of HEK-293 cells with the phosphoinositide 3-kinase inhibitors LY294002 and wortmannin did not alter PMA- or RTK agonist-induced PLD stimulation (data not shown), whereas M₃ mAChR-mediated PLD stimulation was fully blocked.² Another Ras effector pathway, possibly involved in Ras-dependent PLD activation, is the activation of PLD-stimulating Rho GTPases. Although the details of the putative Ras/Rac/Rho signaling pathway are not yet resolved, studies performed in Swiss 3T3 and Rat-1 fibroblasts indicated that Ras-induced stress fiber formation and transformation, respectively, is dependent on activation of Rho proteins (41, 42). Finally, active Ras stimulates the Raf-dependent MAP kinase pathway. In accordance, Frankel et al. (43) recently reported that in v-Raf-transformed NIH-3T3 cells PLD activity was increased. The increase in PLD activity induced by this Ras effector was blocked by coexpressing dominant-negative RalA or RhoC. Although it is presently unclear how Ral and Rho proteins are activated by v-Raf, the data suggest the existence of novel regulatory pathways involved in PLD activation. Alternatively, autocrine loops may exist and may be activated by v-Raf or Ras. However, in HEK-293 cells inactivation of Rho GTPases with C. difficile toxin B or blockade of MAP kinase activation by PD98059 did not affect PLD stimulation by PMA or RTK agonists, making it highly unlikely that these Ras effector pathways are involved in PKC- and RTK-induced PLD stimulation in these cells.

The M₃ mAChR expressed in HEK-293 cells mediates marked phospholipase C stimulation (10). Nevertheless, PLD stimulation induced by this G protein-coupled receptor was not affected by inhibition or down-regulation of PKC or by expression of dominant-negative Ras and RalA mutants. From these data, it may be concluded that the M₃ mAChR and the RTKs activate distinct PKC isoforms and/or distinct PLD enzymes. As PLD1 can interact with both Ral and PKC (15, 16, 44-46), this PLD isoform could be the species involved in RTK-mediated PLD stimulation in HEK-293 cells. On the other hand, by expressing catalytically inactive variants of PLD1 and PLD2, evidence has recently been provided that the insulin-induced PLD stimulation in Rat-1 fibroblasts, which is mediated by ARF proteins, specifically involves PLD2 (47). Furthermore, by overexpressing PLD1 and PLD2 together with the EGF receptor in HEK-293 cells, Slaaby et al. (30) recently reported that EGF can activate both PLD1 and PLD2 enzymes and that PLD2 associates with the EGF receptor in a ligand-independent manner and becomes tyrosine-phosphorylated upon EGF receptor activation. Thus, both PLD enzymes can apparently be activated following receptor activation. Which of the PLD isoforms, PLD1 and/or PLD2, is activated by the M₃ mAChR and the RTKs in HEK-293 cells is presently under investigation.

In summary, our study indicates the existence of a Ras/Ral-GEF/Ral signaling cascade involved in PLD stimulation by PKC in HEK-293 cells. Furthermore, and most important, strong evidence is provided that PLD stimulation by EGF, PDGF, and insulin receptors is mediated by a concerted action of PKC and Ras/Ral-GEF-activated Ral proteins.

	ACKNOWLEDGEMENTS

We thank M. Hagedorn, K. Rehder, H. Geldermann, and S. Grunz for expert technical assistance and Dr. R. M. F. Wolthuis for advice in the Ral activation experiments.

	FOOTNOTES

^* This work was supported by the Deutsche Forschungsgemeinschaft and the Interne Forschungsförderung Essen.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Institut für Pharmakologie, Universitätsklinikum Essen, Hufelandstrasse 55, D-45122 Essen, Germany. Tel.: 49-201-723-3460; Fax: 49-201-723-5968; E-mail: martina.schmidt@uni-essen.de.

² M. Schmidt M. Voß, S. Bayer, M. Asmus, P. A. Oude Weernink, B. Lohmann, C. Rother, P. Chardin, B. Antonny, M. Amano, K. Kaibuchi, and K. H. Jakobs, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: PLD, phospholipase D; PKC, protein kinase C; RTK, receptor tyrosine kinase; PtdCho, phosphatidylcholine; mAChR, muscarinic acetylcholine receptor; ARF, ADP-ribosylation factor; GEF, guanine nucleotide exchange factor; PMA, phorbol 12-myristate 13-acetate; EGF, epidermal growth factor; PtdIns(4, 5)P₂, phosphatidylinositol 4,5-bisphosphate; PDGF, platelet-derived growth factor; RalBD, Ral binding domain; GST, glutathione S-transferase; MAP kinase, mitogen-activated protein kinase; PAGE, polyacrylamide gel electrophoresis; PtdEtOH, phosphatidylethanol.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES