Role of the CDC25 Homology Domain of Phospholipase Cε in Amplification of Rap1-dependent Signaling

Phospholipase Cε (PLCε) is a novel class of phosphoinositide-specific PLC characterized by possession of CDC25 homology and Ras/Rap1-associating domains. We and others have shown that human PLCε is translocated from the cytoplasm to the plasma membrane and activated by direct association with Ras at its Ras/Rap1-associating domain. In addition, translocation to the perinuclear region was induced upon association with Rap1·GTP. However, the function of the CDC25 homology domain remains to be clarified. Here we show that the CDC25 homology domain of PLCε functions as a guanine nucleotide exchange factor for Rap1 but not for any other Ras family GTPases examined including Rap2 and Ha-Ras. Consistent with this, coexpression of full-length PLCε or its N-terminal fragment carrying the CDC25 homology domain causes an increase of the intracellular level of Rap1·GTP. Concurrently, stimulation of the downstream kinases B-Raf and extracellular signal-regulated kinase is observed, whereas the intracellular level of Ras·GTP and Raf-1 kinase activity are unaffected. In wild-type Rap1-overexpressing cells, epidermal growth factor induces translocation of PLCε to the perinuclear compartments such as the Golgi apparatus, which is sustained for at least 20 min. In contrast, PLCε lacking the CDC25 domain translocates to the perinuclear compartments only transiently. Further, the formation of Rap1·GTP upon epidermal growth factor stimulation exhibits a prolonged time course in cells expressing full-length PLCε compared with those expressing PLCε lacking the CDC25 homology domain. These results suggest a pivotal role of the CDC25 homology domain in amplifying Rap1-dependent signal transduction, including the activation of PLCε itself, at specific subcellular locations such as the Golgi apparatus.

Ras family small GTPases direct a wide variety of intracellular signaling pathways (1)(2)(3). In mammalian cells, ϳ20 proteins, such as Ras, Rap, R-Ras, Ral, Rit, Rin, and Rheb, belong to the Ras family. The best characterized members of the family are Ha-Ras, Ki-Ras, and N-Ras, which have been implicated in the regulation of cell proliferation and differentiation downstream of diverse cell surface receptors. Localization at the plasma membrane following post-translational modifications is crucial for the function of Ras proteins. On the other hand, Rap1, which was originally isolated as a Ras-related GTPase, can suppress Ki-Ras-induced transformation and thus is also termed Krev-1 (4). Inhibition of Ras-mediated pathways by Rap1 is attributable to tight binding of Rap1 to the second Ras/Rap1-binding sites of Ras effectors, such as Raf-1 and yeast adenylyl cyclase, without stimulating their activities (5)(6)(7). However, it is likely that Rap1 exerts functions other than the inhibition of Ras pathways because Rap1 localizes mainly in the perinuclear compartments including the Golgi apparatus and cytoplasmic vesicles, where Ras proteins do not exist (8 -11). Indeed, Rap1 is involved, for example, in the activation of integrin and subsequent cell aggregation independently of Ras pathways (12)(13)(14)(15).
Ras family GTPases cycle between GDP-bound inactive and GTP-bound active states, serving as a molecular switch of signal transduction (1). In response to extracellular stimuli, transition from the GDP-bound state to the GTP-bound state is facilitated by GEFs 1 specific to an individual member of the Ras family, leading to the accumulation of the GTP-bound active form. Once activated, Ras family GTPases become associated with specific effectors, thereby stimulating downstream signaling pathways. Thereafter, protein-bound GTP is hydrolyzed to GDP and inorganic phosphate, causing dissociation from the effector. Thus, for the understanding of the biological function of the Ras family, it is important to identify effectors for each GTPase and to clarify the molecular mechanism underlying their activation upon binding to GTPases.
To date, various proteins, such as Raf kinases (A-Raf, B-Raf, and Raf-1), Ral GEFs (RalGDS, Rlf, and Rgl), and PI3-K, have been characterized as Ras effectors on the basis of their ability to bind to Ras in a GTP-dependent manner (2). In the case of Raf-1, the RBD (amino acids 51-131) directly interacts with the effector region of Ras (amino acids [32][33][34][35][36][37][38][39][40]. On the other hand, the domain of RalGDS responsible for the binding to Ras⅐GTP is designated the RA domain (16). Interestingly, x-ray crystallography revealed that the overall tertiary structure of the RA domain of RalGDS is similar to that of the Raf-1 RBD, although no extensive sequence similarity between these two domains was found (17)(18)(19). Moreover, the crystal structure of PI3-K␥ was determined, demonstrating that the RBD of PI3-K␥ has the same fold as the Raf-1 RBD and the RA domain of RalGDS (20,21). Collectively, the GTP-dependent interaction between Ras and its effectors is mediated by structurally conserved modules.
Mechanisms underlying the activation of effectors subsequent to Ras binding are complicated, but a primary role for Ras binding is considered to be the recruitment of effectors to the plasma membrane. For instance, Ras induces subcellular translocation of Raf-1 to the plasma membrane, where Raf-1 undergoes a variety of modifications such as serine/threonine and tyrosine phosphorylation to become fully activated (22). For full activation of Raf-1 kinase activity, however, binding of the RBD to the Ras effector region is insufficient, and an additional interaction between the cysteine-rich region of Raf-1 (amino acids 139 -184) and the activator region of Ras (amino acids 26 -31 and 41-53) is required (6,23).
Recently, we and other groups identified a novel type of phosphoinositide-specific PLC, named PLC⑀, which possesses an RA domain that is responsible for high affinity binding to GTP-bound forms of Ras and Rap1 (24 -27). Upon binding to Ras and Rap1, PLC⑀ translocates to spatially distinct sites, the plasma membrane and the perinuclear region, respectively, where PIPs exist as substrates (25). Ras-dependent subcellular translocation and activation of PLC⑀ were also reconstituted in vitro by the use of a liposome carrying recombinant Ras and phosphatidylinositol 4,5-bisphosphate (25). The activation of PLC⑀ in vivo following cotransfection with constitutively activated Ras was reported as well (27). Taken together, PLC⑀ is involved in Ras-mediated and Rap1-mediated signaling pathways, triggering the hydrolysis of PIPs at different subcellular regions depending on cellular contexts. However, it is unclear whether Ras binding exerts an allosteric effect on enzymatic activity as proposed for PI3-K activation. Also, modifications of PLC⑀, such as serine/threonine phosphorylation, after membrane translocation have not been reported. Thus, precise mechanisms whereby the enzymatic activity of PLC⑀ is regulated at specific subcellular sites remain to be clarified.
In addition to the RA domain, a CDC25 homology domain was found in the N-terminal portion of PLC⑀. The CDC25 homology domain was originally identified in the yeast Saccharomyces cerevisiae CDC25 protein, which acts as a GEF for yeast Ras proteins. Afterward, an array of mammalian Ras GEFs, such as Sos1, Sos2, Ras-GRF-1, Ras-GRF-2, and Ras-GRP, were isolated, revealing that the CDC25 homology domain was conserved among all of these proteins and responsible for their GEF activity. In addition, GEFs for Rap1, such as C3G (28), Epac/cAMP-GEF (29,30), CalDAGGEF1 (31), and RA-GEF-1/PDZ-GEF-1/nRapGEP/CNrasGEF (32)(33)(34)(35), possess the CDC25 homology domain. Hence, the CDC25 homology domain of PLC⑀ is predicted to show GEF activity toward Ras family GTPases. The Ras exchanger motif, which is conserved among an array of Ras GEFs but is not directly implicated in catalysis (36), was not found in PLC⑀.
Here, we show that the CDC25 homology domain of PLC⑀ exhibits GEF activity toward Rap1 but not toward any other Ras family GTPases examined, including Rap2 and Ha-Ras. We further demonstrate that deletion of the CDC25 homology domain diminishes EGF-dependent sustained translocation of PLC⑀ to the perinuclear region in wild-type Rap1-expressing cells. In addition, sustained increase in the Rap1⅐GTP level was observed in PLC⑀-expressing cells but not in cells expressing a PLC⑀ mutant lacking the CDC25 homology domain. Based on these observations, a novel self-amplifying mechanism of the Rap1/PLC⑀ pathway leading to prolonged activation is proposed.
Cell Culture and Transfection-COS-7 cells were cultured in DMEM supplemented with 10% fetal calf serum. Expression plasmids were introduced into COS-7 cells by using GenePORTER (Gene Therapy System) or Superfect (Qiagen) according to the manufacturer's protocol.
Rap1 Pull-down Assay-COS-7 cells were transfected with a combination of expression plasmids as described in figure legends. After incubation for 24 h in DMEM supplemented with 10% fetal calf serum, cells were starved for another 24 h in DMEM supplemented with 0.1% fetal calf serum. In some experiments, the cells were stimulated with EGF as described in the figure legends. Thereafter, the cells were harvested and dissolved in lysis buffer A (50 mM Tris-HCl, pH 7.4, 200 mM NaCl, 5 mM MgCl 2 , 10% glycerol, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, and 1 mM leupeptin). Supernatants of centrifugation (15,000 ϫ g) for 10 min at 4°C were used as cell extracts. GST-RalGDS-RID (20 g) immobilized on glutathione-Sepharose beads (Amersham Pharmacia Biotech) was incubated with cell extracts for 1 h at 4°C and washed with lysis buffer A four times. Precipitated Rap1 was detected by immunoblotting using anti-HA antibody (12CA5; Roche Molecular Biochemicals) or anti-Rap1 antibody (sc-65; Santa Cruz Biotechnology).
Detection of ERK Phosphorylation-COS-7 cells were transfected with a combination of expression plasmids as described in the figure legends. After incubation for 24 h in DMEM supplemented with 10% fetal calf serum, the cells were starved for another 24 h in DMEM supplemented with 0.1% fetal calf serum. Thereafter, cells were harvested and dissolved in lysis buffer C (25 mM Hepes-NaOH, pH 7.3, 150 mM NaCl, 10 mM MgCl 2 , 1 mM EDTA, 1 mM Na 3 VO 4 , 1 mM 4-(2aminoethyl)benzenesulfonyl fluoride hydrochloride, 0.8 M aprotinin, 15 M E-64, 20 M leupeptin, 50 M bestatin, and 10 M pepstatin A). Supernatants of centrifugation (15,000 ϫ g) for 10 min at 4°C were used as cell extracts. Cell extracts were subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting using anti-phosphorylated ERK (number 9106; New England Biolabs Inc.) and anti-ERK (number 9102; New England Biolabs Inc.) antibodies.
Fluorescence Microscopy-COS-7 cells cultured in a slide chamber were transfected with a combination of expression plasmids as described in the figure legends. After incubation for 24 h in DMEM supplemented with 10% fetal calf serum, the cells were starved for another 18 h in DMEM supplemented with 0.1% fetal calf serum. In some experiments, cells were stimulated with EGF as described in the figure legends. Following fixation with 3.7% formaldehyde and permeabilization with 0.2% Triton X-100, the cells were stained with anti-HA (12CA5; Roche Molecular Biochemicals) and tetramethylrhodamineconjugated goat anti-mouse IgG (T2762; Molecular Probes) antibodies. Subcellular localization of EGFP-PLC⑀, EGFP-PLC⑀⌬N, HA-Rap1A WT , and HA-Rap1A V12 was analyzed under a confocal laser microscope (MRC-1024; Bio-Rad). For staining the Golgi apparatus, serum-starved cells were incubated with 5 M BODIPY TR ceramide (D-7540; Molecular Probes) (39) for 10 min at 4°C and washed three times with phosphate-buffered saline. Following fixation with 3.7% formaldehyde, subcellular localization of the Golgi apparatus and EGFP-PLC⑀ was analyzed.

RESULTS
The CDC25 Homology Domain Exhibits in Vitro GEF Activity toward Rap1-Structural features of deletion mutants employed in this study are illustrated in Fig. 1. PLC⑀CDC25 contains the CDC25 homology domain (amino acids 532-775) yet lacks the C-terminal portion. In contrast, PLC⑀⌬N lacks the N-terminal portion containing the CDC25 homology domain. These two constructs as well as full-length PLC⑀ were expressed as FLAG-tagged proteins in Sf9 cells and then purified to near homogeneity by using anti-FLAG M2 affinity chromatography. Thereafter, these recombinant proteins were subjected to in vitro GEF assays using Rap1 as substrate (Fig. 2). PLC⑀ exhibited significant GEF activity toward Rap1 as determined by GDP binding and GDP release assays (Fig. 2, A and   FIG. 1. Structures of PLC⑀ and   B). Further, PLC⑀CDC25 showed Rap1 GEF activity similar to that of PLC⑀, whereas PLC⑀⌬N exerted no significant effect (Fig. 2C), implying that the CDC25 homology domain is responsible for GEF activity. A FLAG-tagged construct consisting of the CDC25 homology domain and additional 100-amino acid flanking sequences (amino acids 432-875) did not show GEF activity under similar conditions, which may be ascribed to improper protein folding (data not shown). To further examine the substrate specificity of GEF activity, various Ras family small GTPases were purified and subjected to in vitro GEF assays (Fig. 3). PLC⑀ did not stimulate GDP release from Ras family GTPases examined other than Rap1. Notably, Rap2 was virtually insensitive to the action of PLC⑀, although Rap2 exhibits high sequence similarity to Rap1.

The Elevated Rap1⅐GTP Level and Activation of Downstream Protein Kinases Following Coexpression of the PLC⑀ CDC25
Homology Domain-In vitro GEF activity of the CDC25 homology domain suggests that Rap1 and its downstream pathways are activated within the cell when this domain is expressed. Fig. 4 shows Rap1⅐GTP levels upon coexpression of indicated proteins as determined by pull-down assays using GST-Ral-GDS-RID as a probe. PLC⑀CDC25 as well as full-length PLC⑀ caused elevation of Rap1⅐GTP levels as observed for RA-GEF-1 as a positive control (Fig. 4A). Thus, the CDC25 homology domain, in fact, acts as a GEF toward Rap1 in vivo. PLC⑀⌬N did not induce Rap1⅐GTP formation, suggesting that Rap1 activation following hydrolysis of PIPs is negligible in COS-7 cells (Fig. 4B).
Raf-1 and B-Raf serine/threonine kinases have been characterized as downstream targets of Ras. Rap1 as well can bind to Raf-1 and B-Raf as Rap1 possesses the same effector region sequence compared with Ras. However, Rap1 is unable to activate Raf-1 because the interaction mediated by the cysteinerich second binding region of Raf-1 is too strong (5, 6). On the other hand, Rap1 can activate B-Raf in PC12 and COS-7 cells (6,40,41). Therefore, kinase activities of B-Raf and Raf-1 were measured to assess the in vivo activity of Rap1 (Fig. 5). In parallel with elevated Rap1⅐GTP levels, PLC⑀CDC25 caused a 3-fold increase in B-Raf kinase activity when coexpressed with Rap1, which is comparable with that induced by RA-GEF-1 (Fig. 5A). Full-length PLC⑀ stimulated B-Raf activity more potently than PLC⑀CDC25 and RA-GEF-1 (Fig. 5A), although the Rap1⅐GTP level was similar to the levels in PLC⑀CDC25-or RA-GEF-1-expressing cells (Fig. 4A). A part of the full-length PLC⑀-induced B-Raf activation may be mediated by Rap1-independent pathways stimulated by breakdown products of PIPs. In contrast, neither PLC⑀CDC25 nor full-length PLC⑀ augmented Raf-1 kinase activity in Rap1-coexpressing cells (Fig. 5B). Furthermore, Raf-1 activity in Ha-Ras-expressing cells remained unchanged upon the expression of PLC⑀CDC25 or full-length PLC⑀ (Fig. 5C), which is consistent with the observation that the CDC25 homology domain of PLC⑀ did not show any GEF activity toward Ras (Fig. 3) ERK acts downstream of Raf family kinases to exert a variety of biological functions. An antibody raised against phosphorylated ERK was employed to assess the activation (Fig. 6). PLC⑀CDC25 and full-length PLC⑀, like RA-GEF-1, induced ERK phosphorylation in Rap1-expressing cells, which is presumably mediated by endogenous B-Raf. Taken together, the CDC25 homology domain of PLC⑀ exhibits GEF activity toward Rap1, leading to the formation of Rap1⅐GTP, which in turn activates the downstream B-Raf/MEK/ERK pathway within the cell. Role of the CDC25 Homology Domain in Rap1-dependent Subcellular Translocation of PLC⑀-Subcellular localization of PLC⑀, which is determined through the interaction of the RA domain with Ras or Rap1, is considered to be important for the function of PLC⑀. When coexpressed with an activated mutant of Ras, PLC⑀ was conveyed to the plasma membrane, whereas translocation to the perinuclear region occurred when activated Rap1 was coexpressed (25). Furthermore, following EGFdependent formation of GTP-bound forms of Ras or Rap1, PLC⑀ was distributed to the plasma membrane and the perinuclear compartments, respectively (25). To better understand the role of the CDC25 homology domain in translocation-dependent regulation of PLC⑀, Rap1-mediated subcellular localization of full-length PLC⑀ and PLC⑀⌬N was examined (Fig. 7). Both full-length PLC⑀ and PLC⑀⌬N uniformly distributed in the cytoplasm without ectopically expressed Rap1 in serumstarved COS-7 cells (Fig. 7A). When coexpressed with constitutively activated Rap1 (Rap1 V12 ), PLC⑀⌬N, like full-length PLC⑀, colocalized with Rap1 V12 in the perinuclear region, indicating that RA domain-dependent translocation remained unaffected after the deletion of the CDC25 homology domain (Fig.  7B). PLC⑀ in Rap1 V12 -expressing cells was also costained with the Golgi apparatus (Fig. 8), indicating that PLC⑀ translocates to the perinuclear compartments, particularly to the Golgi apparatus, upon the binding to Rap1. When serum-starved COS-7 cells expressing wild-type Rap1 were stimulated with EGF for 5 min, PLC⑀ localized in the perinuclear region as described previously (25) (Fig. 7C). Similarly, translocation of PLC⑀⌬N to the perinuclear region was observed upon 5 min of treatment with EGF (Fig. 7C). In marked contrast, after 20 min, EGFinduced translocation of PLC⑀⌬N to the perinuclear region was abolished, whereas PLC⑀ remained colocalized with Rap1 in the perinuclear region (Fig. 7D). Collectively, these results suggest that the CDC25 homology domain may play a pivotal role in the prolonged formation of Rap1⅐GTP in the perinuclear region upon EGF stimulation, which results in a prolonged stay of PLC⑀ in this region. To further clarify this point, EGF-dependent increase in the level of the GTP-bound form of endogenous Rap1 was analyzed by pull-down assays in full-length PLC⑀-or PLC⑀⌬N-expressing cells (Fig. 9). In the absence of ectopically expressed PLC⑀, EGF caused a transient increase in the Rap1⅐GTP level, which peaked at 5-10 min and diminished by 30 min (Fig. 9A). When full-length PLC⑀ was expressed, the Rap1⅐GTP level remained elevated after 30 min of treatment. However, expression of PLC⑀⌬N did not significantly affect the time course of EGF-induced alteration in the Rap1⅐GTP level (Fig. 9B). Thus, the CDC25 homology domain is involved in persistent, but not transient, activation of Rap1 following EGF treatment.

DISCUSSION
In addition to the catalytic (X and Y) and regulatory (C2) domains for PLC activity, PLC⑀ possesses CDC25 homology and RA domains, implicating Ras family GTPases in the reg- ulation of PLC⑀. Indeed, we observed that the RA domainmediated association of PLC⑀ with Ras⅐GTP in a liposome containing phosphatidylinositol 4,5-bisphosphate as substrate enhanced the activity of PLC⑀ (25). Translocation of PLC⑀ to the plasma membrane was also observed when coexpressed with activated Ras (25). Therefore, it is likely that translocation of PLC⑀ to the plasma membrane through the binding to Ras⅐GTP is critical for the activation. Furthermore, Kelley et al. (27) reported that PLC activity was augmented in parallel with the binding of activated Ras in COS-7 cells. On the other hand, Rap1 as well associates with the RA domain and causes translocation of PLC⑀ to the perinuclear region (25). The activation of PLC⑀ upon the binding to Rap1⅐GTP has been observed to date neither in vivo nor in vitro. However, PLC⑀ may become activated subsequent to translocation to the perinuclear compartments in a manner similar to the Ras-dependent activation, considering that PIPs exist in the membrane of the Golgi complex and is implicated in the regulation of membrane traffic (42). Taken together, PLC⑀ presumably acts as a downstream effector for Ras or Rap1 at distinct subcellular locations, for which the RA domain plays an essential role.
In contrast to the RA domain, the role of the CDC25 homology domain in Ras family GTPase-dependent regulation of PLC⑀ remains obscure. Herein, we provide evidence that the CDC25 homology domain of PLC⑀ exhibits GEF activity toward Rap1 but not Ras in vitro. These results suggest a mechanism whereby PLC⑀ specifically amplifies Rap1 signaling by yielding Rap1⅐GTP, which in turn may activate PLC activity. To address this issue, we assessed the role of the CDC25 homology domain in Rap1⅐GTP formation in vivo and Rap1-dependent translocation to the perinuclear region by employing deletion constructs of PLC⑀. EGF-induced formation of Rap1⅐GTP exhibited a prolonged time course depending on the coexpression of PLC⑀, although the expression of PLC⑀⌬N exerted virtually no effect on the time course of EGF-dependent increase in the Rap1⅐GTP level (Fig. 9). Moreover, EGF-induced translocation of PLC⑀ to the perinuclear region, mediated by Rap1⅐GTP, exhibited a prolonged time course, whereas translocation of PLC⑀⌬N was transient (Fig. 7). Considered as a whole, it is expected that Rap1-dependent PLC activation is sustained at specific subcellular sites such as the Golgi apparatus through the action of the CDC25 homology domain. It is feasible that Rap1 can serve as a regulator of multiple downstream effectors in the perinuclear compartments. Although Rap1⅐GTP formation in response to EGF is transient as described in Fig. 9A, a subset of Rap1 effectors including PLC⑀ may require sustained interaction with Rap1⅐GTP for exerting their function. To accomplish this, a complicated regulatory mechanism involving the CDC25 homology domain as described above may be important.
A similar regulatory mechanism of Rap1 signaling is proposed for RA-GEF-1, a Rap1 GEF containing both the RA and the GEF domains (32). The RA domain of RA-GEF-1 is responsible for binding to Rap1⅐GTP and has a role in translocation to the perinuclear region and amplification of in vivo GEF activity of RA-GEF-1. 2 Further, another Rap1 GEF, GFR/MR-GEF, possesses an RA domain that specifically interacts with M-Ras (43,44). Although the mechanism remains obscure, the MR-GEF-dependent accumulation of Rap1⅐GTP within the cell was compromised upon coexpression of the activated form of M-Ras (44).
Lopez et al. (26) reported the increase in the Ras⅐GTP level following the expression of wild-type PLC⑀ and its mutant deficient in PLC activity. Further, the CDC25 homology domain of PLC⑀ induced the activation of ERK (26). However, we detected no GEF activity of PLC⑀ toward Ras in vitro (Fig. 3), and the Ras/Raf-1 pathway was not stimulated by PLC⑀ (Fig.  5). Additionally, no significant increase in the Ras⅐GTP level was observed in our assays after the expression of the CDC25 homology domain of PLC⑀ that could activate Rap1 (data not shown). Considering our results, the activation of ERK reported by Lopez et al. (26) may possibly be mediated by the endogenous Rap1/B-Raf pathway, yet the reason for the discrepancy regarding Ras activation remains unclear.
The CDC25 homology domain of PLC⑀ shows GEF activity toward Rap1, but not Rap2 in vitro (Fig. 3), whereas other Rap GEFs, including cAMP-activated Epac/cAMP-GEF (45) and RA-GEF-1 (45), 3 show GEF activity toward both Rap1 and Rap2. Hence, signaling involving Rap1, but not Rap2, is enhanced through the action of the CDC25 homology domain of PLC⑀, and therefore PLC⑀ may play a pivotal role exclusively downstream of Rap1, although differences in biological functions of Rap1 and Rap2 remain obscure. Intriguingly, the tuberous sclerosis-2 gene product tuberin, a GTPase-activating protein for Rap1, but not for Rap2, localizes in the Golgi apparatus (9,46). Thus, PLC⑀ in collaboration with tuberin may dynamically regulate the GDP/GTP state of Rap1. Future studies will reveal the physiological significance of Rap1-specific signal amplification by PLC⑀.  9. Effect of PLC⑀ on EGF-dependent increase in the Rap1⅐GTP level. A, increase in the Rap1⅐GTP level upon EGF treatment. Following stimulation with EGF (100 ng/ml) for indicated periods, the GTP-bound form of endogenous Rap1 in COS-7 cells was precipitated by GST-RalGDS-RID and detected by immunoblotting using anti-Rap1 antibody. B, effects of PLC⑀⌬N and PLC⑀ on EGF-dependent increase in the Rap1⅐GTP level. Following stimulation with EGF (100 ng/ml) for indicated periods, the GTP-bound form of endogenous Rap1 in COS-7 cells transfected with pFLAG-CMV2-PLC⑀ or pFLAG-CMV2-PLC⑀⌬N was precipitated by GST-RalGDS-RID and detected by immunoblotting using anti-Rap1 antibody. The expression levels of PLC⑀ and PLC⑀⌬N were virtually identical (data not shown).