Stimulation of Phospholipase C-ε by the M3Muscarinic Acetylcholine Receptor Mediated by Cyclic AMP and the GTPase Rap2B*

Stimulation of phospholipase C (PLC) by Gq-coupled receptors such as the M3muscarinic acetylcholine receptor (mAChR) is caused by direct activation of PLC-β enzymes by Gαq proteins. We have recently shown that Gs-coupled receptors can stimulate PLC-ε, apparently via formation of cyclic AMP and activation of the Ras-related GTPase Rap2B. Here we report that PLC stimulation by the M3 mAChR expressed in HEK-293 cells also involves, in part, similar mechanisms. M3 mAChR-mediated PLC stimulation and [Ca2+] i increase were reduced by 2′,5′-dideoxyadenosine (dd-Ado), a direct adenylyl cyclase inhibitor. On the other hand, overexpression of Gαs or Epac1, a cyclic AMP-regulated guanine nucleotide exchange factor for Rap GTPases, enhanced M3 mAChR-mediated PLC stimulation. Inactivation of Ras-related GTPases with clostridial toxins suppressed the M3 mAChR responses. The inhibitory toxin effects were mimicked by expression of inactive Rap2B, but not of other inactive GTPases (Rac1, Ras, RalA, Rap1A, and Rap2A). Activation of the M3 mAChR induced GTP loading of Rap2B, an effect strongly enhanced by overexpression of Gαs and inhibited by dd-Ado. Overexpression of PLC-ε and PLC-β1, but not PLC-γ1 or PLC-δ1, enhanced M3 mAChR-mediated PLC stimulation and [Ca2+] i increase. In contrast, expression of a catalytically inactive PLC-ε mutant reduced PLC stimulation by the M3 mAChR and abrogated the potentiating effect of Gαs. In conclusion, our findings suggest that PLC stimulation by the M3 mAChR is a composite action of PLC-β1 stimulation by Gαq and stimulation of PLC-ε apparently mediated by Gs-dependent cyclic AMP formation and subsequent activation of Rap2B.

Stimulation of phosphatidylinositol 4,5-bisphosphate (PIP 2 ) 1hydrolyzing phospholipase C (PLC) is a major signal transduc-tion system used by a wide variety of membrane receptors and apparently regulates various cellular functions, such as smooth muscle contraction, secretion, neuronal signaling, and cell growth and differentiation (1,2). The eleven PLC isoforms (PLC-␤1-4, PLC-␥1-2, PLC-␦1-4, and PLC-⑀) identified so far differ largely in their structure and regulatory mechanisms. Receptor regulation of PLC-␤ and PLC-␥ isozymes is well established. While tyrosine kinase receptors, such as those for epidermal growth factor and platelet-derived growth factor, activate PLC-␥ enzymes by recruitment to the autophosphorylated receptor and subsequent tyrosine phosphorylation, G protein-coupled receptors (GPCRs) activate PLC-␤ enzymes, either via GTP-liganded ␣ subunits of the G q class of G proteins or by ␤␥ dimers liberated from G i type G proteins (3,4). Regulation of PLC-␦ has been assumed to involve the G protein G h (transglutaminase II) and/or capacitative Ca 2ϩ influx (5,6). The very recently identified PLC-⑀ contains, in addition to the PLC-defining catalytic and calcium-binding domains, two Rasbinding domains and a Ras-specific guanine nucleotide exchange factor (GEF) domain. Initial data suggest that the activity of PLC-⑀ is controlled by G␣ 12 and Ras and Rap GTPases, by yet unresolved mechanisms; thus, PLC-⑀ may link signaling by heterotrimeric G proteins and Ras-related GTPases (7)(8)(9)(10).
The M 3 muscarinic acetylcholine receptor (mAChR) is a prototypical GPCR known to stimulate PLC via pertussis toxin (PTX)-insensitive G␣ q type G proteins (11)(12)(13)(14). We very recently observed that G s -coupled receptors, i.e. the ␤ 2 -adrenoreceptor expressed in HEK-293 cells and the prostaglandin E 1 receptor endogenously expressed in N1E-115 neuroblastoma cells, can induce PLC stimulation and PLC-dependent calcium signaling (15). These receptor responses were apparently dependent on G s -mediated cyclic AMP formation and under control of the Ras-related GTPase Rap2B. As activation of the M 3 mAChR expressed in HEK-293 can increase cyclic AMP levels (11), we examined whether M 3 mAChR signaling to PLC may involve cyclic AMP and Rap GTPases as well. We report here that the M 3 mAChR can stimulate PLC-⑀ and that this PLC stimulation is apparently mediated by G s -dependent formation of cyclic AMP and subsequent activation of the GTPase Rap2B.

EXPERIMENTAL PROCEDURES
Materials-myo-[ 3 H]Inositol (10 -25 Ci/mmol) and D-myo-[ 3 H]inositol 1,4,5-trisphosphate ([ 3 H]IP 3 ; 21 Ci/mmol) were from PerkinElmer Life Sciences. Unlabeled IP 3 was from Biomol, 2Ј5Ј-dideoxyadenosine (dd-Ado) was from Calbiochem-Novabiochem, and Fura-2/AM was from Molecular Probes. The antibodies against Rac1, Rap1, Rap2, PLC-␤1, PLC-␥1, and PLC-␦1 were from Santa Cruz. The antibodies against RalA and Ras were from Transduction Laboratories. The antibody against HA-tagged proteins (12CA5) was a kind gift of Dr. J. L. Bos. The polyclonal rabbit anti-PLC-⑀ antibody raised against the unique N-* 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. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
terminal 600 amino acids of PLC-⑀ did not recognize other PLC isoforms (not shown). Clostridium difficile toxin B-1470 and Clostridium sordellii lethal toxin (strains 82 and 1522, respectively) were kind gifts of Dr. C. von Eichel-Streiber.
Measurement of PLC Activity-For measurement of inositol phosphate formation, cellular phospholipids were labeled by incubating cells for 24 h with myo-[ 3 H]inositol (0.5 Ci/ml) in growth medium. Thereafter, the adherent cells were first treated for 10 min at 37°C in Hanks' balanced salt solution, containing 118 mM NaCl, 5 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , and 5 mM D-glucose, buffered at pH 7.4 with 15 mM HEPES, plus 10 mM LiCl, followed then by further incubation for 30 min at 37°C in the presence of stimulatory agents and determination of [ 3 H]inositol phosphate accumulation. For measurement of IP 3 formation, unlabeled cells were incubated for the indicated periods of time at 37°C with and without carbachol. Stop of the PLC assays, extraction, and analysis of [ 3 H]inositol phosphates or IP 3 mass were performed as described previously (16). To study the effects of dd-Ado, the cells were pretreated for 30 min with the agent or its solvent, dimethyl sulfoxide (0.1%); dd-Ado was also present during the PLC assays. To study the effects of clostridial toxins, the cells were treated for 24 h without and with the toxins at the indicated concentrations, followed by PLC activity assays.
Calcium Measurements-Intracellular free Ca 2ϩ concentration ([Ca 2ϩ ] i ) was determined in cell suspensions with the fluorescent Ca 2ϩ indicator dye Fura-2 in a Hitachi spectrofluorometer as described previously (17).
Activation of Rap2B-Cells were stimulated without and with carbachol for 5 min at 37°C, followed by two washes with ice-cold phosphate-buffered saline and lysis in a buffer containing 10% glycerol, 1% Nonidet P-40, 50 mM Tris/HCl, pH 7. trypsin inhibitor, 10 mM NaF, and 1 mM Na 3 VO 4 . After centrifugation, the supernatants were incubated with 15 g of purified glutathione S-transferase-tagged RalGDS-RBD (Rap-binding domain of the Ral guanine nucleotide dissociation stimulator) bound to glutathione-Sepharose beads for 1 h at 4°C. Then, the beads were washed three times with lysis buffer and finally incubated in Laemmli buffer for 10 min at 95°C. Bound Rap2-GTP was determined by immunoblotting with an anti-Rap2 antibody (15,18). Densitometric analysis of the bands was performed with ImageQuant software (Molecular Dynamics).
Data Presentation-Data shown in figures are means Ϯ S.E. of n independent experiments, each performed in triplicate. Comparisons between means were either with the Student's paired t test or one-way analysis of variance test, and a difference was regarded significant at p Ͻ 0.05. Curves were analyzed by fitting iterative nonlinear regression analysis to the experimental data with the GraphPad Prism program (Version 2.0, 1995).

Role of Cyclic AMP in M 3 mAChR-mediated PLC Stimulation-
We have recently reported that stimulation of cyclic AMP formation by the ␤ 2 -adrenoreceptor expressed in HEK-293 cells or direct activation of adenylyl cyclase by forskolin results in PLC stimulation (15). To examine whether cyclic AMP and cyclic AMP-dependent processes are involved in PLC stimulation by the M 3 mAChR as well, several approaches were used. First, we studied the effect of the P-site adenylyl cyclase inhibitor, dd-Ado (19), on PLC stimulation by the M 3 mAChR stably expressed in HEK-293 cells. Treatment of the cells with dd-Ado (10 M) had no effect on unstimulated inositol phosphate accumulation or IP 3 levels. However, dd-Ado strongly reduced, by 30 -40%, IP 3 formation or inositol phosphate accumulation (see below) stimulated by the mAChR agonist, carbachol (Fig. 1A). In line with this inhibition, dd-Ado significantly (p Ͻ 0.0001) reduced the carbachol (1 M)-induced increase in [Ca 2ϩ ] i from 485 Ϯ 25 nM to 246 Ϯ 35 nM (n ϭ 8 -10). Second, we examined whether overexpression of G␣ s , the adenylyl cyclase-stimulatory G protein, alters PLC stimulation by the M 3 mAChR. In cells overexpressing G␣ s , basal inositol phosphate formation was not altered. However, PLC stimulation induced by carbachol (1 M) was increased by about 50%. This increase in PLC stimulation was almost fully suppressed by treatment of the cells with 10 M dd-Ado (Fig. 1B). Together, these data suggested that PLC stimulation by the M 3 mAChR is, at least in part, dependent on cyclic AMP formation.
As observed with ␤ 2 -adrenoreceptor-induced PLC stimulation (15), treatment of HEK-293 cells with the protein kinase A inhibitor, H-89 (10 M), had no significant effect on PLC stimulation induced by carbachol, neither in control cells nor in cells overexpressing G␣ s (data not shown), suggesting that the action of cyclic AMP is mediated by another effector. Therefore, we overexpressed Epac1, a cyclic AMP-activated GEF for Rap GTPases (18,20). As illustrated in Fig. 2A, similar to G␣ s , overexpression of Epac1 had no effect on basal PLC activity, but strongly enhanced PLC stimulation induced by carbachol (1 M). Overexpression of Epac1 also strongly enhanced the potency of carbachol to increase [Ca 2ϩ ] i in HEK-293 cells. While carbachol increased [Ca 2ϩ ] i in control cells with an EC 50 value of 450 Ϯ 14 nM, this value was reduced by about one order of magnitude, to 26 Ϯ 2 nM, in cells overexpressing Epac1 (Fig.  2A). These data suggested that cyclic AMP/Epac1-controlled Rap GTPases are involved in PLC stimulation by the M 3 mAChR. Involvement of Rap2B in M 3 mAChR-mediated PLC Stimulation-To study whether and which type of Rap GTPases are involved in PLC stimulation by the M 3 mAChR, we first examined the effects of C. difficile toxin B-1470 and C. sordellii lethal toxin, known to inactivate Ras-related GTPases (21,22). Treatment of HEK-293 cells for 24 h with 300 pg/ml toxin B-1470 and 100 ng/ml lethal toxin strongly reduced inositol phosphate accumulation (data not shown) and IP 3 formation induced by carbachol (Fig. 2B). PLC stimulation induced by the direct G protein activators, AlF 4 Ϫ (intact cells) and GTP␥S (permeabilized cells), was also strongly reduced in cells treated with the toxins (data not shown). The reduction in M3 mAChRmediated IP 3 (23). Furthermore, the toxins had no effect on the cellular PIP 2 levels, measured as [ 3 H]PIP 2 or PIP 2 mass in control and toxin-treated cells (24), and did not reduce Ca 2ϩ (1 M)-stimulated PLC activity, measured with exogenous PIP 2 in cell lysates (17) (data not shown). Thus, inhibition of PLC and calcium signaling by the GTPase-inactivating toxins was apparently not due to a loss of cell curface receptors, a defective receptor-G protein coupling, a fall in PLC substrate levels, or a general reduction in PLC activities, suggesting that inhibition of M 3 mAChR signaling to PLC by the toxins is caused by inactivation of small GTPases specifically involved in PLC stimulation.
To identify the specific GTPase, we expressed inactive mutants of the GTPases serving as toxin substrates. Compared with the expression of the endogenous GTPases, the various GTPase mutants were overexpressed to a comparable level (Fig. 3A). However, the GTPase mutants largely differed in their effects on PLC signaling. Expression of inactive Rac1, H-Ras, RalA, RalB (not shown), Rap1A, and Rap2A did neither change basal PLC activity nor PLC stimulation by the M 3 mAChR. In contrast, in cells expressing S17N Rap2B PLC stimulation induced by carbachol was reduced by 40 -50% (Fig.  3A). In line with their distinct effects on PLC stimulation, expression of S17N Rap2B, but not S17N Ras or S17N Rap1A (not shown), strongly (p Ͻ 0.0001) reduced the carbachol (1 M)-induced [Ca 2ϩ ] i increase, from 535 Ϯ 55 nM in control cells to 310 Ϯ 45 nM (n ϭ 6 -8) in cells expressing S17N Rap2B (Fig. 3B).
Next, we studied whether Rap2B is activated by the M 3 mAChR and whether this activation is affected by agents inhibiting or enhancing receptor-mediated PLC stimulation. Carbachol (1 M) treatment of HEK-293 cells enhanced GTP loading of endogenous and overexpressed Rap2B, as determined by extraction of the GTPase from cell lysates with immobilized RalGDS-RBD (Fig. 4). Expression of constitutively active G␣ 12 (Q229L G␣ 12 ) or overexpression of wild-type G␣ i2 (not shown), both of which did not alter carbachol-induced PLC stimulation (14, not shown), had no effect on M 3 mAChR-induced Rap2B activation. In contrast, overexpression of G␣ s or Epac1 (not shown), which by themselves did not alter the activity state of Rap2B, strongly enhanced the stimulatory effect of carbachol (Fig. 4). Treatment of the cells with dd-Ado (10 M) almost completely (by 90 Ϯ 5%; n ϭ 4) abrogated the potentiating effect of G␣ s (data not shown). Thus, the M 3 mAChR activates Rap2B and, similarly as observed for PLC stimulation, this activation is apparently controlled by cyclic AMP.
Stimulation of PLC-⑀ by the M 3 mAChR-We next sought to determine which PLC isozyme is activated by the M 3 mAChR. For this, we examined the effects of overexpression of PLC-␤1, PLC-␥1, PLC-␦1, and PLC-⑀ on basal and receptor-stimulated PLC activities. As illustrated in Fig. 5A, overexpression of PLC-␥1, which increased PLC stimulation by epidermal growth factor (not shown), did neither change basal PLC activity nor its stimulation by carbachol (1 M). In cells overexpressing PLC-␦1, basal PLC activity was increased by 2.5-fold, whereas carbachol-stimulated PLC activity was not altered. In contrast, overexpression of PLC-␤1 and PLC-⑀ enhanced PLC stimulation by carbachol by 50 and 70%, respectively, without altering basal PLC activity. In line with these data, carbachol (1 M)induced [Ca 2ϩ ] i increase was strongly potentiated in cells overexpressing PLC-␤1 or PLC-⑀, from 470 Ϯ 55 nM in control cells to 700 Ϯ 45 nM in cells overexpressing PLC-␤1 (n ϭ 6 -8; p Ͻ 0.0001) and from 524 Ϯ 35 nM to 940 Ϯ 65 nM in cells overexpressing PLC-⑀ (n ϭ 8 -10; p Ͻ 0.0001) (Fig. 5B). In contrast to wild-type PLC-⑀, expression of the catalytically inactive PLC-⑀ mutant, H1144L PLC-⑀ (7), reduced PLC stimulation by carbachol by 25% (Fig. 6A). Interestingly, expression of H1144L PLC-⑀ almost fully reversed the potentiating effect of co-expressed G␣ s , suggesting that potentiation of PLC stimulation by G␣ s is due to activation of the PLC-⑀ isozyme. This assumption was corroborated by studies with the adenylyl cyclase inhibitor, dd-Ado, in cells overexpressing PLC-⑀ or PLC-␤1. As shown in Fig. 6B, dd-Ado (10 M) strongly reduced the potentiating effect of overexpressed PLC-⑀ on carbachol-stimulated inositol phosphate formation, whereas dd-Ado was without effect in cells overexpressing PLC-␤1, suggesting that cyclic AMP is involved in stimulation of PLC-⑀, but not PLC-␤1, by the M 3 mAChR.
We have recently reported that overexpression of the regulator of G protein signaling 4 (RGS4), which acts as a GTPaseactivating protein for G␣ q and G␣ i proteins (25, 26), strongly reduces M 3 mAChR-mediated PLC stimulation (14). As shown in Fig. 7A, in cells overexpressing RGS4 M 3 mAChR-mediated PLC stimulation was reduced by about 40%. Most important, the inhibitory effect of dd-Ado (10 M) on PLC stimulation was fully retained in cells overexpressing RGS4. In contrast, in cells expressing S17N Rap2B, in which the M 3 mAChR response was reduced to a similar extent as in cells overexpressing RGS4, treatment with dd-Ado did not cause a further reduction in receptor-mediated PLC stimulation (Fig. 7B). DISCUSSION PTX-insensitive stimulation of PLC by GPCRs is generally assumed to be caused by direct activation of PLC-␤ isozymes by activated ␣-subunits of G q type G proteins (3,4). The M 3 mAChR is a prototypical example of such GPCRs. PLC stimulation by the M 3 mAChR that preferentially couples to G proteins of the G q family is PTX-insensitive (11)(12)(13). Furthermore, we have recently reported that PLC stimulation by the M 3 mAChR is specifically suppressed by RGS proteins inactivating PTX-resistant G␣ q , but not G␣ 12 type G proteins (14). Interestingly, overexpression of the G␣ q -inactivating RGS4 reduced PLC stimulation by the M 3 mAChR only partially, by 60 -70% (14). In studies with G s -coupled receptors, we made very recently the unexpected observation that such receptors, i.e. the ␤ 2 -adrenoreceptor expressed in HEK-293 cells and the prostanoid receptor endogenously expressed in N1E-115 neuroblastoma cells, can also mediate PLC stimulation (15). This PLC stimulation was PTX-insensitive and apparently mediated by G s -dependent formation of cyclic AMP and activation of the Ras-related GTPase Rap2B, finally resulting in stimulation of the PLC-⑀ isozyme (15). As the M 3 mAChR can couple to G s and increase cyclic AMP formation (11,27), we examined in the present study whether PLC stimulation by the M 3 mAChR may involve similar mechanisms. We report here that PLC stimulation by the M 3 mAChR expressed in HEK-293 cells is a composite action on PLC-␤1 and PLC-⑀ isozymes and that stimulation of PLC-⑀ by the M 3 mAChR is apparently mediated by G s -dependent cyclic AMP formation and activation of the GTPase Rap2B.
First, treatment of the cells with the P-site adenylyl cyclase inhibitor, dd-Ado, reduced PLC stimulation by the M 3 mAChR. Second, overexpression of G␣ s , which by itself had no effect on PLC activity, strongly enhanced M 3 mAChR signaling to PLC. Third, a similar enhancement of M 3 mAChR-mediated PLC stimulation was observed in cells overexpressing the cyclic AMP-activated GEF for Rap GTPases, Epac1, while inhibition of cyclic AMP-dependent protein kinase A by H-89 was without effect. Fourth, inactivation of Ras-related GTPases with C. difficile toxin B-1470 and C. sordellii lethal toxin strongly reduced PLC stimulation. Fifth, the inhibitory toxin effects were mimicked by expression of an inactive Rap2B mutant, but not by inactive mutants of other GTPases serving as toxin substrates. Sixth, the M 3 mAChR induced activation of Rap2B, and this activation was enhanced by overexpression of G␣ s or Epac1 and suppressed by dd-Ado. Seventh, PLC stimulation by the M 3 mAChR was enhanced by overexpression of PLC-⑀, similar to overexpression of PLC-␤1, and reduced by expression of a catalytically inactive PLC-⑀ mutant. Finally, using various combinations, i.e. G␣ s with H1144L PLC-⑀ and dd-Ado with PLC-␤1, PLC-⑀, RGS4, and S17N Rap2B, evidence is provided that cyclic AMP-dependent PLC stimulation by the M 3 mAChR involves Rap2B and the PLC-⑀ isozyme and that this stimulation is largely independent of PLC-␤1 stimulation by G␣ q .
Activation of G s and stimulation of adenylyl cyclase is not considered a primary function of the M 3 mAChR, compared with coupling to G q and stimulation PLC-␤ isozymes (12). Therefore, we were surprised to observe that inhibition of adenylyl cyclase by dd-Ado and overexpression of cyclic AMPactivated Epac1 had such marked effects on M 3 mAChR-mediated PLC stimulation. Similarly as reported before by others (11), increases in total cellular cyclic AMP levels in HEK-293 cells expressing the M 3 mAChR were observed only at rather high carbachol concentrations (Ͼ1 M) (data not shown). However, as shown herein, inhibition of M 3 mAChR-mediated PLC stimulation by dd-Ado was largely independent of the carbachol concentration used (Fig. 1A), and overexpression of Epac1 markedly increased PLC stimulation and [Ca 2ϩ ] i increase induced by carbachol at low concentrations. A possible explanation for this finding is that, in contrast to many other GEFs for other small GTPases, Epac1 is a membrane-associated protein even in its basal activity state, i.e. in the absence of cyclic AMP, and is activated by cyclic AMP at the plasma membrane (28). Thus, similarly as recently described for activation of the L-type Ca 2ϩ channel Ca v 1.2 by the G s -and adenylyl cyclase-coupled ␤ 2 -adrenoreceptor (29), Epac1 and the adenylyl cyclase may be assembled into a signaling complex at the plasma membrane, and local increases in cyclic AMP concentration not detected by measuring total cellular cyclic AMP levels may suffice to activate Epac1 and in consequence Rap2B. The existence of such a signaling complex is under investigation.
The magnitude of PLC inhibition caused by expression of S17N Rap2B was in the similar range as the inhibition induced by treatment of the cells with dd-Ado. Taking the transfection efficiency of 50 -60% into consideration, these data suggest that S17N Rap2B, in addition to inhibiting cyclic AMP-dependent stimulation of PLC-⑀ by the M 3 mAChR, may also interfere with receptor stimulation of PLC-␤1 via G q proteins. However, PLC-␤1, in contrast to PLC-⑀, does not contain binding domains for Ras/Rap proteins, making a direct interaction of PLC-␤1 with Rap2B unlikely. Recently, RGS14 has been shown to interact with Rap GTPases via a RBD domain also found in RGS12 (30 -32). However, no data have been reported whether Rap GTPases alter the GTPase-activating activities of these RGS proteins. Furthermore, RGS12 and RGS14 preferentially act on PTX-sensitive G i/o proteins (33,34), whereas PLC stimulation by the M 3 mAChR is fully PTX-resistant. Thus, although an indirect action of Rap2B on G␣ q cannot be excluded, an alternative explanation for the pronounced inhibitory effect of S17N Rap2B is that Rap2B is also activated by G␣ q , independent of G s -dependent cyclic AMP formation, finally resulting in stimulation of PLC-⑀. Activation of Rap GTPases (Rap1 and Rap2) is not only achieved by the cyclic AMP-activated GEFs, Epac1 and Epac2 (18,20,28), but also by other Rapspecific GEFs, such as the CalDAG-GEFs, which are presumably activated by the second messengers, Ca 2ϩ and diacylglycerol (35), and the PDZ-GEFs, for which the activation mechanisms are not yet known (36). Thus, it is conceivable to assume that G␣ q , e.g. by increasing Ca 2ϩ and diacylglycerol levels via stimulation of PLC-␤ isozymes, activates Rap2B and thereby PLC-⑀. Such a reaction may also explain why overexpression of PLC-⑀ enhanced M 3 mAChR-mediated PLC stimulation at least as efficiently as overexpression of PLC-␤1 (see Fig. 5). The involvement of additional Rap-GEFs in the M 3 mAChR response will be addressed in future studies.
In conclusion, we report here that PLC and calcium signaling by the M 3 mAChR is mediated by the two PTX-insensitive G proteins, G q and G s , finally resulting in stimulation of PLC-␤1 and PLC-⑀, respectively. While stimulation of PLC-␤1 is most likely caused by direct interaction with activated G␣ q proteins, stimulation of PLC-⑀ is apparently dependent on cyclic AMP formation and activation of Epac1 and in consequence Rap2B. This GTPase seems also to be involved in overall PLC stimulation by G q , suggesting that Rap2B and its exchange factors play a major role in PLC stimulation by GPCRs.