Heterotrimeric G Protein βγ Subunits Stimulate FLJ00018, a Guanine Nucleotide Exchange Factor for Rac1 and Cdc42*

We previously reported that Gβγ signaling regulates cell spreading or cell shape change through activation of a Rho family small GTPase, suggesting the existence of a Gβγ-regulated Rho guanine-nucleotide exchange factor (RhoGEF). In this study we examined various RhoGEF clones, found FLJ00018 to beaGβγ-activated RhoGEF, and investigated the molecular mechanism of Gβγ-induced activation of Rho family GTPases. Co-expression of the genes for FLJ00018 and Gβγ enhanced serum response element-mediated gene transcription in HEK-293 cells. Combined expression of Gβγ and FLJ00018 significantly induced activation of Rac and Cdc42 but not RhoA. FLJ00018 also enhanced gene transcription induced by carbachol-stimulated m2 muscarinic acetylcholine receptor, and this enhancement was blocked by pertussis toxin. Furthermore, we demonstrated Gβγ to interact directly with the N-terminal region of FLJ00018 and the N-terminal fragment of this molecule to inhibit serum response element-dependent transcription induced by Gβγ/FLJ00018 and carbachol. In NIH3T3 cells, FLJ00018 enhanced lysophosphatidic acid-induced cell spreading, which was also blocked by the N-terminal fragment of FLJ00018. These results provide evidence for a signaling pathway by which Gi-coupled receptor specifically induces Rac and Cdc42 activation through direct interaction of Gβγ with FLJ00018.

GTPases is controlled by three distinct classes of regulatory proteins, namely (i) guanine-nucleotide dissociation inhibitors, which stabilize the inactive form (3), (ii) guanine-nucleotide exchange factors (GEFs), 2 which catalyze GDP/GTP exchange (4,5), and (iii) GTPases-activating proteins, which stimulate low, intrinsic GTPase activity of Rho GTPases (6). In particular, 60 different GEFs for Rho family members (RhoGEFs) have been described so far (4). A common feature of RhoGEFs is the Dbl homology (DH) domain responsible for exchange activity followed by a pleckstrin homology (PH) domain considered to be involved in subcellular localization. Besides this tandem motif, RhoGEFs often contain one or more additional signal transduction domains, such as PDZ, Src homology (SH) 2, SH3, and RGS (regulator of G protein signaling), which can function as molecular bridges between different signal transduction pathways.
It is well established that a large variety of G protein-coupled receptors (GPCRs), particularly those coupling to the G 12/13 type of heterotrimeric G proteins, are upstream regulators of Rho proteins (7,8). A family of RhoA-specific GEFs consisting of p115RhoGEF, PDZ-RhoGEF, and leukemia-associated RhoGEF (LARG), which mediate this activation process, has been identified (7)(8)(9)(10). All these proteins contain in addition to the DH/PH tandem motif an RGS (regulator of G protein signaling) homology domain for direct interaction with and activation by G 12 type G proteins. More than a dozen Rac-specific RhoGEFs have been described so far, including for example Vav, Sos, Tiam, and P-Rex (4). Some of these GEFs are specific for Rac, and others have more broad targets, but all have activities that are tightly regulated, usually by protein kinases (11)(12)(13), phosphatidylinositol 3,4,5-trisphosphate, a lipid second messenger produced by class I phosphatidylinositol 3-kinase (14 -16), and interaction with other proteins containing G␤␥ subunits or adenomatous polyposis coli protein (16 -18).
We have previously reported that G␤␥ induces actin stress fiber and focal adhesion formation in a Rho-dependent manner in HeLa cells (19) and that G i signals through both G␣ i2 and G␤␥ regulate Rac and Cdc42 during lysophosphatidic acid (LPA)-induced cell spreading in NIH3T3 fibroblasts (20). Others have described G i -coupled receptor signaling to induce cell migration and neurite outgrowth (21,22). These results suggest the presence of G protein-regulated RhoGEFs, but the details of the underlying molecular mechanisms have yet to be clarified.
In the present study we searched for G␤␥-regulated RhoGEFs in various RhoGEF clones in the Kazusa HUGE data base and the NEDO data base and found FLJ00018/PLEKHG2 (23) to link specifically the G i -coupled receptor to Rac and Cdc42 by directly interacting with G␤␥ subunits. This RhoGEF, therefore, represents a hitherto unknown G␤␥ effector molecule.

EXPERIMENTAL PROCEDURES
pCMV5-FLAG-Rac1T17N, and pCMV5-FLAG-Cdc42T17N were generous gifts from Dr. H. Itoh (Nara Institute of Science and Technology). pcDNA3.1-G␣ i2 QL, pcDNA3.1-G␣ s QL, pcDNA3.1-G␣ q QL, pcDNA3.1-G␣ 11 QL, pcDNA3.1-G␣ 12 QL, pcDNA3.1-G␣ 13 QL, pcDNA3.1-G␤ 1 , pcDNA3.1-G␥ 2 , and pcDNA3.1-m2 muscarinic acetylcholine receptor (m2 receptor) were from UMR cDNA Resource Center. Complementary DNA clones for RhoGEF genes were isolated during the Kazusa human cDNA project, which aimed to accumulate information on the coding sequences of long cDNAs for unidentified human genes (24). To construct native-form protein expression clones, the open reading frames of all RhoGEF cDNAs were subcloned into the pcDNA-DEST-47 vector used by the Gateway system (Invitrogen). cDNAs encoding deletion mutants of FLJ00018, as indicated in the corresponding figures, were generated by restriction enzyme digestion or polymerase chain reaction amplification using pcDNA-DEST-47-FLJ00018 as a template. Plasmids encoding Myc-tagged FLJ00018 and its mutants were subcloned into the pCMV5 vector. The pSRE.Lluciferase reporter plasmid was purchased from Stratagene, and pRL-SV40 was from Nippon Gene. LPA was obtained from Avanti Polar Lipids Inc., pertussis toxin was from Calbiochem-Novabiochem, glutathione-coupled Sepharose 4B beads were from GE Healthcare, and fibronectin and isopropyl-␤-D-thiogalactopyranoside were from Wako Pure Chemical Industries. The rabbit polyclonal antibody against the G␤ subunit generated by ourselves has been described previously (25). Rabbit polyclonal antibodies against G␣ q/11 , G␣ 12 , G␣ 13 , G␣ s , and RhoA were purchased from Santa Cruz Biotechnology. A mouse monoclonal antibody against Myc epitope (9E10) was purchased from Roche Applied Science, and mouse monoclonal antibodies against human Rac1 and Cdc42 were from BD Transduction Laboratories.
Cell Culture and Transfection-HEK-293 and NIH3T3 cells were grown in Dulbecco's modified essential medium (DMEM) supplemented with 10% fetal bovine serum (HEK-293) or 10% calf serum (NIH3T3) at 37°C. Transient transfection was performed using Lipofectamine Plus according to the manufacturer's instructions (Invitrogen). To examine the effect of m2 muscarinic stimulus on serum response element (SRE)-dependent gene transcription, cells were co-transfected with the m2 receptor and FLJ00018 or other plasmids and cultured for 16 h in DMEM supplemented with 1ϫ insulin-transferrin-selenium-X supplement (Invitrogen). Then cells were washed twice with DMEM, incubated for 2 h in DMEM, and stimulated with 1 mM carbachol for 6 h.
Assay of SRE-dependent Gene Transcription-HEK-293 cells seeded on 24-well plates were co-transfected with the indicated expression plasmids (400 ng of total DNA/well) together with the pSRE.L-luciferase reporter plasmid and the pRL-SV40 control reporter vector. After transfection, cells were washed once with phosphate-buffered saline and lysed with passive lysis buffer (Nippon Gene). Luciferase activities were determined with the dual-luciferase reporter assay system (PicaGene Dual SeaPansy Luminescence kit, Nippon Gene). The activity of the experimental reporter was normalized against the activity of control vector.
Pulldown Assays of Activated Rho Family GTP-binding Proteins-The cellular level of GTP-loaded Rac, Cdc42, and RhoA was determined using a GST fusion protein containing the Rac/Cdc42 binding domain of PAK (GST-CRIB) expressed in and purified from Escherichia coli DH5␣ (20,26) and the RhoA binding domain of Rhotekin (GST-RBD), purchased from Upstate. In brief, subconfluent HEK-293 cells were transfected with the indicated amounts of plasmid DNA or the corresponding empty vectors (2 g/6-cm dishes). Thereafter, the cells were lysed in buffer containing 1% Nonidet P-40, and the particular fraction was removed by centrifugation. The supernatant was then incubated for 1 h at 4°C with GST-CRIB bound to glutathione-Sepharose beads. After three washes of beads, bound proteins were eluted with sample buffer and separated by SDS-PAGE (27). Rac, Cdc42, and RhoA were then detected by immunoblotting with specific monoclonal antibodies.
Immunoprecipitation-HEK-293 cells were seeded in 6-cm dishes and transfected at 80% confluence with 2 g of the indicated cDNA constructs. After transfection, cells were solubilized in 300 l of immunoprecipitation buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1 mM EDTA, 1 mM Na 3 VO 4 , 0.2 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 g/ml leupeptin, 0.5% Nonidet P-40) and incubated on ice for 10 min. After centrifugation, 2 g of anti-c-Myc antibody was added to the clear supernatant, and the mixture was incubated for 2 h at 4°C, added to protein G-agarose, and gently shaken for 1 h at 4°C. Beads were washed 3 times with washing buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1 mM EDTA, 1 mM Na 3 VO 4 , 0.2 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 g/ml leupeptin, 0.1% Nonidet P-40), and bound proteins were eluted with sample buffer and loaded onto a 7.5% polyacrylamide gel. After SDS-PAGE, immunoprecipitated proteins were transferred to nitrocellulose membranes and detected with anti-Myc and anti-G␤ antibodies.
In Vitro Binding Assays-GST or GST-p7 fusion proteins were expressed and extracted from E. coli strain DH5␣ and bound to glutathione-Sepharose 4B. Three g of purified GST fusion protein (with the glutathione-agarose beads) and 3 g of G␤␥ subunit (purified from bovine brain) (28) were incubated for 3 h at 4°C in buffer A (50 mM Tris-HCl, pH 8.0, 0.5 mM EDTA, 150 mM NaCl, 0.2% Nonidet P-40, 0.5 mM phenylmethylsulfonyl fluoride). Beads were washed three times with buffer A, and bound proteins were separated by SDS-PAGE and detected by Western blotting with an anti-G␤ antibody.
Cell Spreading Assays-NIH3T3 cells, which were transfected with GFP and indicated plasmids, were detached with trypsin-EDTA and washed with DMEM containing 0.3 mg/ml trypsin inhibitor and 1 mg/ml fatty acid-free bovine serum albumin. Cells were then re-suspended in DMEM, re-plated on fibronectin-coated glass coverslips, and incubated for 15 min in the presence or absence of 10 M LPA. Cells were fixed in 4% paraformaldehyde in phosphate-buffered saline for immunocytochemistry. To quantitate cell spreading, images of cells were obtained using a laser-scanning microscope (LSM-510, Carl Zeiss), and the areas of GFP-positive cells were measured using LSM-510 image analysis software. For each experiment, the areas of 50 cells were quantitated.
Statistical Analysis-All results shown are the means Ϯ S.D. of values from at least three independent experiments. The significance of group differences was analyzed by one-way analysis of variance followed by the Bonferroni correction. A p value of Ͻ0.05 was considered significant.

Identification of a RhoGEF Inducing G␤␥-dependent Gene
Transcription-To search for G␤␥-dependent RhoGEFs, we screened various RhoGEF clones in the Kazusa HUGE data base and the NEDO data base by co-expression with G␤ 1 ␥ 2 in HEK-293 cells. To monitor Rho family GTPase activations in intact cells, we measured transcription of an SRE-controlled reporter gene (SRE.L-luciferase), which is known to be induced by Rho family activation. Overexpression of several RhoGEF clones containing KIAA0380 (no. 4), KIAA0424 (no. 5), KIAA0521 (no. 6), and FLJ00018 (no. 10) caused an increase in luciferase expression, and co-expression with G␤ 1 ␥ 2 greatly increased FLJ00018-induced gene transcriptional activity and slightly enhanced KIAA0424-induced activity (Fig. 1). Co-expression of G␤ 1 ␥ 2 with other RhoGEF clones caused a small increase or did not enhance RhoGEF-induced gene transcriptional activity (Fig. 1).

G␤␥/FLJ00018-induced SRE-dependent Gene Transcription
Is Mediated by Rac and Cdc42-To clarify which subtype of Rho family proteins contributes to increase of SRE-dependent gene transcription by FLJ00018, we examined the effects of various Rho family mutants. A dominant negative mutant of Rac1 completely inhibited SRE-dependent gene transcription induced by FLJ00018, whereas that of Cdc42 inhibited it to a lesser extent, and a dominant negative Rho was without effect ( Fig. 2A). To provide further support for FLJ00018-dependent activation of Rac and Cdc42, we carried out pulldown assays using GST-CRIB and GST-RBD as a direct readout of Rho family GTPases activation. Expression of FLJ00018 alone only  slightly increased the amount of active Rac (1.1 times) and Cdc42 (1.2 times), and combined expression of FLJ00018 and G␤ 1 ␥ 2 induced Rac activation by 2.1 times and Cdc42 activation by 1.5 times (Fig. 2B). However, expression of FLJ00018 or FLJ00018/G␤ 1 ␥ 2 did not influence RhoA activation.
G␤␥, but Not G␣ Subunits, Enhances FLJ00018-induced SREdependent Gene Transcription-Because the data obtained so far argued for activation of FLJ00018 by G␤ 1 ␥ 2 , we studied the influence of various G␣ subunits on FLJ00018-induced luciferase expression. GTPase-deficient G␣ mutants induced transcriptional activity in a range similar to that observed with coexpression of FLJ00018, although strong transcriptional activity was observed with several G␣ mutants containing G␣ q QL, G␣ 11 QL, G␣ 12 QL, and G␣ 13 QL (Fig. 3A). Next, we studied the effect of various G␥ subunits, G␥ 2 , G␥ 5 , G␥ 7 , and G␥ 12 , co-expressed with G␤ 1 on FLJ00018-induced luciferase expression, but there were no differences among them (Fig. 3B).
FLJ00018 Enhances m2 Receptor-induced SRE-dependent Gene Transcription-It is generally accepted that G␤␥-mediated signaling mainly acts through G i/o -coupled receptors. To examine whether these might stimulate SRE-dependent transcription through FLJ00018, we co-expressed FLJ00018 together with the m2 receptor in HEK-293 cells. M2 receptors preferentially couple to G i/o proteins but can additionally acti-vate other G proteins. The combined expression of FLJ00018 and the m2 receptor led to a significant enhancement of luciferase expression by carbachol stimulation, whereas such stimulation was not observed in cells expressing FLJ00018 alone (Fig. 4A). The dominant negative mutant of Rac1 strongly inhibited carbachol-induced luciferase production, but those of RhoA and Cdc42 did not (Fig. 4A), consistent with the results obtained for G␤␥-induced transcription ( Fig. 2A). Co-expression of the dominant negative mutant of RhoA increased carbachol-induced luciferase production, although the reason remains unclear. This may raise the possibility that a RhoA dominant negative mutant cancels the attenuation of Rac activation by RhoA.
The ␣ subunits of G i family G proteins were specifically ADP-ribosylated by pertussis toxin, then became unable to couple with GPCRs. Pretreatment with pertussis toxin eliminated carbachol-stimulated luciferase production in cells cotransfected with m2 receptor and FLJ00018 (Fig. 4B).

Wortmannin Does Not Influence G␤␥/FLJ00018-induced SRE-dependent Gene Transcription and Rac Activation-The
RacGEF, P-Rex1, is directly regulated by G␤␥ like FLJ00018, but full activation of P-Rex1 requires the binding of phosphatidylinositol 3,4,5-trisphosphate (16). To examine the requirement of phosphatidylinositol 3,4,5-trisphosphate for FLJ00018 activation, we used an inhibitor for phosphatidylinositol 3-kinase, wortmannin. The treatment of wortmannin did not influence G␤␥/FLJ00018-induced SRE-dependent gene transcription (Fig. 5A). Wortmannin seemed to be functional, because 100 nM wortmannin inhibited 10% fetal bovine serum-stimulated SRE-dependent gene transcription by 40% as compared with control. G␤␥/FLJ00018-induced Rac activation was also unaffected by wortmannin (Fig. 5B). The activation mechanism of FLJ00018 seems to be different from the one of P-Rex1. In the future, however, it is necessary to compare the phosphoinositide dependence of two molecules in detail.
G␤␥ Interact Directly with the N-terminal Region of FLJ00018-To investigate whether FLJ00018 directly interacts with G␤␥, G␤ 1 ␥ 2 and N-terminal Myc-tagged FLJ00018 were co-expressed in HEK-293 cells, and cell lysates were immunoprecipitated with anti-Myc antibody and analyzed for co-precipitated G␤ protein. As shown in Fig. 6A, the G␤ subunit co-precipitated with the wild-type FLJ00018 (WT). To identify the part of FLJ00018 in which this interaction takes place, we performed similar experiments with seven truncated mutants; WT (amino acids (aa) 1-1386), p1 (aa 1-964), p2 (aa 1-464), p3 (aa 1-309), p4 (aa 108 -309), p5 (aa 1-108), p6 (aa 1-149), p7 (aa 1-134). The mutant p4 was immunoprecipitated to a similar extent as WT and other mutants with the anti-Myc antibody but did not form a complex with the G␤ subunit (Fig. 6A). In contrast, the three mutants containing the N-terminal region (amino acids 1-108), p5, p6, and p7, co-precipitated with the G␤ subunit. To examine whether the N-terminal region of FLJ00018 binds directly G␤␥, G␤␥ proteins purified from bovine brain were mixed with GST and GST-p7 fusion protein in vitro. As shown in Fig. 6B, the G␤ subunit bound p7-GST fusion protein but not GST (Fig. 6B).
To further study the properties of these mutants, we investigated the ability of each to activate SRE-dependent transcrip-tion (Fig. 6C). The mutants lacking the PH domain such as p3, p4, and p5 were inactive, suggesting this domain is strictly required. These results resemble those with PDZ-RhoGEF deletion mutants (9), and the binding of RhoGEF to lipids or other molecules through the PH domain seems to be necessary  for the activation of several RhoGEF molecules. In contrast, p1 and p2 enhanced luciferase production more than WT, but the levels of luciferase production induced by co-expression of G␤ 1 ␥ 2 were similar to that induced by the combination of WT and G␤␥ (Fig. 6C). These results suggest that the C-terminal region of FLJ00018 is inhibitory for the GEF activity, and binding of G␤␥ may release this inhibition.
The P7 Deletion Mutant of FLJ00018 Inhibits LPA/FLJ00018indued Cell Spreading-We have previously shown Rac and Cdc42 to be activated by G i during LPA-induced cell spreading in NIH3T3 fibroblasts (20). To examine whether the expression of FLJ00018 and p7 influences LPA-induced cell spreading, we co-transfected these plasmids and GFP into NIH3T3 cells and re-plated in the presence or absence of LPA. FLJ00018 expression significantly enhanced cell spreading in the presence of LPA after 15 min, whereas LPA did not (Fig. 8A). Co-expression of p7 markedly diminished LPA/FLJ00018-induced cell spreading. To examine the requirement of functional domains (PH domain) in cell spreading, we transfected the deletion mutants FIGURE 6. Direct interaction of G␤␥ subunits with FLJ00018. A, HEK-293 cells were co-transfected with wild-type FLJ00018 (WT) or its truncated mutant (p1-p7) plasmid DNAs and expression vectors for G␤ 1 and G␥ 2 as indicated. Structure of the proteins encoded by each expression plasmid: WT, p1, p2, p3, p4, p5, p6, and p7 constructs code for amino acid residues 1-1386, 1-964, 1-464, 1-309, 108 -309, 1-108, 1-149, and 1-134 of FLJ00018, respectively. Cells were lysed 24 h after transfection, and wild-type FLJ00018 and its variants were immunoprecipitated with anti-Myc antibodies. Precipitated proteins were separated by SDS-PAGE and immunoblotted with anti-G␤ and anti-Myc antibodies. IP, immunoprecipitation; WB, Western blot; TCL, total cell lysate. B, purified G␤␥ proteins were added to GST or GST-p7 fusion proteins bound to glutathione-Sepharose 4B and rotated. Beads were washed, and bound proteins were separated by SDS-PAGE and immunoblotted with anti-G␤. C, cells were co-transfected with pSRE.L-luciferase, pRL-SV40 plasmid DNAs, and expression vectors for wild-type FLJ00018, its truncated mutants, G␤ 1 , and G␥ 2 , as indicated. Luciferase activity obtained with Mock was taken as 1.0, and relative activities are shown. Values are means Ϯ S.D. from four experiments.

DISCUSSION
It has been well documented that G␤␥ subunits induce activation of Rho family GTPases in a variety of cells and tissues (19,20,29,30). Because there is increasing evidence that such acti-vation occurs independently of the phospholipase C/protein kinase C pathway, it is most likely that hitherto unidentified RhoGEFs are involved. In this report we present several lines of evidence that FLJ00018, a novel member of the Dbl family of GEFs, might represent the specific GEF or at least one of the GEFs, mediating the response. First, FLJ00018 greatly enhanced the transcriptional activity with G␤␥ subunits but not with GTPase-deficient mutants of various G␣ subunits. This increase in transcriptional activity was completely sensitive to the dominant negative mutant of Rac1. Second, FLJ00018 enhanced the transcriptional activity when the primarily G i/o -coupled m2 receptor was stimulated by carbachol. This enhancement was blocked by pertussis toxin. Similarly, FLJ00018 enhanced LPA-induced cell spreading. Third, co-expression of FLJ00018 and G␤␥ subunits synergistically increased the cellular amount of activated Rac and Cdc42 proteins, although Rac was more intensely activated. Finally, FLJ00018 directly interacted with G␤␥ subunits, as detected by means of direct binding in vitro and co-immunoprecipitation as well as by functional inhibition of G␤␥-induced gene transcription and cell spreading by the N-terminal fragment of FLJ00018 including the G␤␥ binding domain (p7). Therefore, our data indicate that FLJ00018 is a novel RhoGEF for Rac and Cdc42 that is regulated by G␤␥ subunits.
So far it has been shown that G␤␥ could activate several RhoGEFs: Ras-GRF1 (31), P-Rex1 (16), P-Rex2 (32,33), and p114RhoGEF (34). Ras-GRF1, P-Rex1, and P-Rex2 are Rac-specific RhoGEFs, and p114RhoGEF is a RhoGEF for RhoA and Rac. Ras-GRF1 has to be tyrosine-phosphorylated to display RacGEF activity (31). p114RhoGEF binds G␤␥ subunits both via its DH/PH tandem and its C terminus, but it is not known whether they regulate the catalytic activity of this GEF (34), identical to KIAA0521 (no. 6 in Fig. 1), which increased SREdependent gene expression but did not further increase it with co-expression of G␤␥ in our experiments. Dimers of the G␤ 1 subunit with different G␥ subunits varied in their ability to stimulate P-Rex1 in vitro. The ␤ 1 ␥ 3 , ␤ 1 ␥ 7 , ␤ 1 ␥ 10 , and ␤ 1 ␥ 13HA dimers all activated P-Rex1 with similar EC 50 values, whereas dimers composed of ␤ 1 ␥ 12 had lower EC 50 values (35). In the present study, co-expression of G␤ 1 and various G␥ subunits including G␥ 12 showed similar effects on SRE-dependent gene transcription through FLJ00018 activation. These results indicate that G␤␥-induced activation of FLJ00018 is clearly different from that of P-Rex family, Ras-GRF, or p114RhoGEF.
Structural analyses of these RhoGEFs further showed FLJ00018 to be a novel type of GEF. FLJ00018 is 71% identical to the mouse orthologue, Clg, overall, with 94% identity in the DH domain, 96% identity in the PH domain, and differing most in C-terminal region (36). Although FLJ00018 contains DH and PH domains in the N-terminal region like P-Rex1, the sequence in other regions is completely different (16). For example, P-Rex1 contains two DEP domains, two PDZ domains, and a C-terminal InsPx4-phosphatase domain, but these are not present in FLJ00018. In PC12 cells, both PDZ and InsPx4-phosphatase domain, but not PH domain, are required for optimal activity of P-Rex1 (37). In contrast, PH domain is required for optimal activity of FLJ00018 (Fig. 6C). The C-terminal region of FLJ00018 is predicted to have a Formin homology 1 domain (38) from Pfam 21.0 analysis, although the sequence homology is low. Several reports demonstrated that the Formin homology 1 domain is a binding site for profilin (38 -41), apart from having potential for binding Src homology 3 or WW domain-containing signaling proteins (42). Because profilin is a G-actinbinding protein that promotes the addition of monomers to the barbed-end of actin filaments, its binding to Formin homology 1-like domains of FLJ00018 would be expected to contribute to efficient formation of actin filaments around the areas where FLJ00018 is recruited and Rac/Cd42 is activated in cells.
Recently, a series of interactions between Rho-GEFs and scaffolding proteins containing PDZ domains have been described (23). PDZ domains are small protein interaction domains that mediate protein targeting and the assembly of multiprotein complexes. FLJ00018 also contains a putative PDZ binding motif at the C terminus. The function of the region is remains uncertain in the regulatory mechanism of FLJ00018 activity.
In summary, the data presented herein define a new signaling pathway for GPCRs, with FLJ00018 serving as a direct G␤␥ effector molecule as shown in Fig. 9. Its postulated linkage of GPCRs to Rac/Cdc42 and Rac/Cdc42-dependent pathways ultimately would be predicted to affect the cytoskeletal structure, nuclear gene expression, and cellular growth.