HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hayakawa, M. Articles by Kitagawa, M. Search for Related Content PubMed Articles by Hayakawa, M. Articles by Kitagawa, M.

¹ School of Pharmacy, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan
² School of Life Science, Tokyo University of Pharmacy and Life Science, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
³ Department of Molecular Pathology, Graduate School of Medicine, University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
⁴ Department of Biochemistry 1, Hamamatsu University School of Medicine, 1-20-1 Handayama, Hamamatsu, 431-3192, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
We previously demonstrated that FGD1, the Cdc42 guanine nucleotide exchange factor (GEF) responsible for faciogenital dysplasia, is targeted by the ubiquitin ligase SCF^FWD1/β-TrCP upon phosphorylation of two serine residues in its DSGIDS motif and subsequently degraded by the proteasome. Here we show that FGD3, which was identified as a homologue of FGD1 but has been poorly characterized, has conserved the same motif and is down-regulated similarly by SCF^FWD1/β-TrCP. Although FGD3 and FGD1 share strikingly similar Dbl homology (DH) domains and adjacent pleckstrin homology (PH) domains, both of which are responsible for guanine nucleotide exchange, there also exist remarkable differences in their structures. Indeed, FGD1 and FGD3 induced significantly different morphological changes in HeLa Tet-Off cells: whereas FGD1 induced long finger-like protrusions, FGD3 induced broad sheet-like protrusions when the level of GTP-bound Cdc42 was significantly increased by the inducible expression of FGD3. Furthermore, FGD1 and FGD3 reciprocally regulated cell motility: when inducibly expressed in HeLa Tet-Off cells, FGD1 stimulated cell migration whereas FGD3 inhibited it. Thus we demonstrate that the highly homologous GEFs, FGD1 and FGD3 play different roles to regulate cellular functions but that their intracellular levels are tightly controlled by the same destruction pathway through SCF^FWD1/β-TrCP.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
The Rho family small GTP-binding proteins (G-proteins or GTPases), consisting mainly of the Rho, Rac and Cdc42 subfamilies, regulates various actin cytoskeleton-dependent cell functions, including cell shape change, cell migration, cell adhesion and cytokinesis (Van Aelst & D'Souza-Schorey 1997; Etienne-Manneville & Hall 2002; Malliri & Collard 2003). Like all GTPases, Rho proteins act as binary switches by cycling between an inactive (GDP-bound) and an active (GTP-bound) conformational state.

Guanine nucleotide exchange factors (GEFs) stimulate the exchange of GDP for GTP to generate active forms of Rho proteins, whereas GTPase-activating proteins (GAPs) accelerate the intrinsic GTPase activity of Rho proteins to inactivate the switch. In theory, the active state of Rho proteins may be obtained through stimulation of a GEF or inhibition of a GAP; however, most evidence indicates that GEFs are the critical mediators for the activation of Rho proteins (Schmidt & Hall 2002). A family of candidate GEFs for Rho proteins has been identified based on the homology with CDC24, a protein identified genetically as an upstream activator of CDC42 in yeast (Bender & Pringle 1989). Dbl, isolated as an oncogene in a focus formation assay using NIH 3T3 cells transfected with DNA from a human diffuse B-cell lymphoma, was identified as the first mammalian Rho GEF showing significant sequence similarity to CDC24 (Eva & Aaronson 1985). The domain conserved in both Dbl and CDC24 is now recognized as the Dbl homology (DH) domain and almost 60 DH-domain-containing proteins have since been found (Schmidt & Hall 2002). Although the primary structures of these proteins are known, their physiological roles have been poorly characterized.

The activities of GEFs must be tightly regulated and each member of the GEF family is likely to have a unique mechanism for activation and inactivation. Many GEFs contain a regulatory domain that blocks activity through an intramolecular interaction. For example, the removal of N-terminal region from certain GEFs, such as Dbl, Vav, Asef and Tiam, leads to constitutive activation when the protein is expressed in cells (Ron et al. 1989; Katzav et al. 1991; Miki et al. 1993; van Leeuwen et al. 1995). It is assumed that activation of full-length GEF occurs through the relief of autoinhibition by phosphorylation or by binding to other proteins, although the mechanism is still unclear in most cases (Schmidt & Hall 2002).

How GEFs are inactivated is largely unknown, however, in our previous report, we presented a unique mechanism for the inactivation of FGD1, a GEF for Cdc42 and known as the faciogenital dysplasia gene product, by the ubiquitin/proteasome system (Hayakawa et al. 2005). Ubiquitin-dependent proteolysis by the proteasome plays an essential role in a number of key biological processes, including cell cycle progression, transcription and signal transduction (Karin & Ben-Neriah 2000; Glickman & Ciechanover 2002). Protein ubiquitination usually requires three processes, those involving an ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzymes (E2s) and ubiquitin ligases (E3s). Because substrate recognition is usually governed by E3s, the central determinants of specificity in ubiquitination are E3s (Glickman & Ciechanover 2002). E3s are classified into two groups: the HECT (homologous to E6-AP carboxy terminus) domain-containing E3s and the RING finger-containing E3s.

Among the RING finger-type E3s, one of the better-defined ones is SCF^FWD1/β-TrCP (Karin & Ben-Neriah 2000). SCF-type E3s are assemblages of several common (Skp1, Cul1 and Roc1/Rbx/Hrt1) and single-variable (F-box protein) protein components. The F-box protein FWD1/β-TrCP has been implicated in the ubiquitination of important signaling molecules, such as IBs and β-catenin (Karin & Ben-Neriah 2000). These molecules share similar phosphorylation sites with the consensus sequence of DSGXS ( represents a hydrophobic residue and X represents any amino acid) (Karin & Ben-Neriah 2000; Mantovani & Banks 2003). Earlier we found that FGD1 is recognized by SCF^FWD1/β-TrCP through the phosphorylation of two serine residues in its DS²⁸³GIDS²⁸⁷-sequence, leading to degradation of the GEF by the proteasome (Hayakawa et al. 2005).

In this study, we show that FGD3, a homologue of FGD1, is also down-regulated through the polyubiquitination by SCF.^FWD1/β-TrCP FGD3 contains adjacent DH and pleckstrin homology (PH) domains, both of which are highly homologous to those of FGD1, suggesting that FGD3, like FGD1, may act as a GEF for Cdc42. Indeed, Pasteris et al. reported that over-expression of DH + PH domains of FGD3 induced a dramatic and rapid induction of a large number of striking filopodia extensions in Swiss 3T3cells, a finding consistent with the selective activation of Cdc42 (Pasteris et al. 2000). However, in this study, we demonstrate that the inducible expression of full-length FGD3 led to the formation of remarkable lamellipodia structures in HeLa Tet-Off cells, whereas that of the full-length FGD1 induced striking filopodia structures in the same parental HeLa Tet-Off cells. Furthermore, the cell motility of FGD3-expressing cells is significantly reduced whereas that of FGD1-expressing cells was remarkably increased. These results indicate that FGD3 and FGD1 play different roles in cells but undergo the same destruction pathway regulated by ubiquitin ligase SCF^{FWD1/β-TrCP.}

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
FGD3 is recognized by FWD1/β-TrCP, a subunit of ubiquitin ligase SCF^FWD1/β-TrCP

In our previous study, we showed that FGD1, known as a GEF for Cdc42, is recognized by ubiquitin ligase SCF^FWD1/β-TrCP through the phosphorylation of serine residues in the conserved motif, that is, DS²⁸³GIDS²⁸⁷ (Hayakawa et al. 2005). We found that FGD3, earlier identified as a homologue of FGD1 (Pasteris et al. 2000), also conserves this same motif, that is, DS⁷²GIDS⁷⁶ (Fig. 1a). Sequence similarity between FGD1 and FGD3 is also observed in their DH and adjacent PH domain (i.e. amino acid residue identity is 70.0% and 60.6%, respectively) (Pasteris et al. 2000). In contrast, there are striking differences in their sequence in other regions; for example, most prominently, FGD3 lacks the N-terminal proline-rich domain conserved in FGD1 (Fig. 1b). However, compared with FGD1, FGD3 has been very poorly characterized.

View larger version (36K): [in this window] [in a new window]   Figure 1  Establishment of stable cell lines that inducibly express the wild-type and mutant FGD3 proteins. (a) Alignment of amino acid sequences required for association with FWD1/β-TrCP in IB, IBβ, IB, β-catenin, hDlg, FGD1 and FGD3. Black shading indicates sequence identity. (b) Schematic representation of FGD1 and FGD3. Domains shown are "Proline-rich domain" (pro-rich), "Dbl homology" (DH), "pleckstrin homology" (PH domain), "domain present in Fab1, YOTB, Vac1 and EEA1" (FYVE). (c) Hygromycin-resistant clonal cell lines were isolated from HeLa Tet-Off cells transfected either with pTRE-FLAG-FGD3(WT) or with pTRE-FLAG-FGD3(SA), as described in "Experimental procedures" and designated as HeLa Tet-Off-FGD3(WT) or HeLa Tet-Off-FGD3(SA), respectively. FGD3(SA) bears Ser to Ala amino acid substitutions at positions 72 and 76 within the FWD1/β-TrCP putative consensus binding site. The cells were incubated with or without 2 µg/mL tetracycline (Tc) for 72 h, and then lysed in SDS-PAGE sample buffer. The resultant whole cell lysates were subjected to immunoblotting analysis using anti-FLAG antibody or anti-GAPDH antibody.

Using these cells, we investigated whether or not FGD3(WT) was associated with FWD1/β-TrCP. As shown in Fig. 2a, FGD3(WT) was co-immunoprecipitated with FWD1/β-TrCP whereas FGD3(SA) was not (lane 3 vs. lane 6). Interestingly, the commercially available anti-phospho-β-catenin (Ser33/37/Thr41) antibody (#9561, Cell signaling), which recognizes the SCF^FWD1/β-TrCP recognition motif [DS³³GIHS³⁷] of β-catenin when phosphorylated at S³³, S³⁷ (and additional T⁴¹) by GSK-3β, detected a 110-kDa band in immunoprecipitates of FGD3(WT) but not in those of FGD3(SA) (Fig. 2b, lane 2 vs. lane 6). This band was confirmed as the FLAG-tagged FGD3(WT) band after reprobing the same PVDF membrane with anti-FLAG antibody. When cells were pretreated with the proteasome inhibitor MG132, the density of the FGD3(WT) band was significantly increased (Fig. 2b, lane 4). Similar results were obtained by using whole cell extracts instead of FLAG-immunoprecipitates (Fig. 2c). These results suggest that this anti-phospho-β-catenin antibody could detect FGD3(WT) when phosphorylated at S⁷² and S⁷⁶. Furthermore, the treatment of cells with a GSK-3β inhibitor, 3-(3-carboxy-4-chloroanilino)-4-(3-nitrophenyl) maleimide, clearly reduced the densities of the bands of FLAG-FGD3(WT) detected by anti-phospho-β-catenin antibody (Fig. 2c, lanes 3 and 5), suggesting that GSK-3β is the kinase responsible for phosphorylating the sequence of DSGIDS in FGD3(WT).

View larger version (48K): [in this window] [in a new window]   Figure 2  FGD3(WT), but not FGD3(SA), is associated with FWD1/β-TrCP through the phosphorylation at Ser72 and Ser76. (a) HeLa Tet-Off-FGD3(WT) and HeLa Tet-Off-FGD3(SA) cells transfected with the expression plasmid encoding HA-FWD1/β-TrCP were incubated with or without 2 µg/mL Tc for 72 h. Cell extracts were immunoprecipitated with anti-HA antibodies followed by immunoblotting to detect the FLAG epitope. Separately, the cell extracts were directly subjected to immunoblotting analyses to detect HA or FLAG epitope (whole cell lysates, WCL). n.s. denotes non-specific bands. (b) HeLa Tet-Off-FGD3(WT) and HeLa Tet-Off-FGD3(SA) cells were incubated with or without 2 µg/mL Tc for 72 h. To inhibit the proteasome activity, MG132 (10 µM) was added to the culture during the last 2 h of incubation. Cell extracts prepared in RIPA buffer were immunoprecipitated with anti-FLAG antibodies followed by the immunoblotting using anti-phospho-β-catenin antibody, which recognizes the SCFFWD1/β-TrCP-recognition motif [DS33GIHS37] of β-catenin when phosphorylated at S33, S37 (and additional T41) by GSK-3β. Then PVDF membrane was stripped and reprobed with anti-FLAG antibody. (c) Effect of a GSK-3β inhibitor on the phosphorylation of FGD(WT) or FGD3(SA). HeLa Tet-Off-FGD3(WT) and HeLa Tet-Off-FGD3(SA) cells were incubated with or without 2 µg/mL Tc for 72 h. MG132 (10 µM) or GSK-3β inhibitor, 3-(3-carboxy-4-chloroanilino)-4-(3-nitrophenyl) maleimide (15 µM) was added to the culture during the last 2 h of incubation. Cell extracts were directly subjected to immunoblotting analysis to detect the phosphorylation of FGD3(WT) or β-catenin using anti-phospho-β-catenin antibody or to detect GAPDH levels as a loading control.

View larger version (44K): [in this window] [in a new window]   Figure 3  FGD3 is down-regulated through SCFFWD1/β-TrCP-mediated polyubiquitination. (a) HeLa Tet-Off-FGD3(WT) and HeLa Tet-Off-FGD3(SA) cells transfected with the expression plasmid encoding HA-tagged ubiquitin were incubated with or without 2 µg/mL Tc for 72 h. To inhibit the proteasome activity, MG132 (10 µM) was added to the culture during the last 2 h of incubation. Cell extracts were immunoprecipitated with anti-FLAG antibodies and resultant immunoprecipitates were boiled in SDS-PAGE sample buffer containing 2% SDS and 5% 2-mercaptoethanol, and the supernatant was diluted with the wash buffer to reduce SDS concentration to < 0.1%. These samples were then immunoprecipitated with anti-FLAG antibodies followed by immunoblotting to detect the HA epitope. Then PVDF membrane was stripped and reprobed with anti-FLAG antibody. (b) Ubiquitination of endogenous FGD3 in L929 cells. Cells were treated with MG132 (10 µM), Epox (1 µM), or LiCl (20 mM) for 20 h. Cell extracts prepared in RIPA buffer were immunoprecipitated with anti-FGD3 antibodies followed by immunoblotting using anti-ubiquitin antibody. Separately, cell extracts were directly subjected to immunoblotting analyses to detect GAPDH levels as a loading control. (c) L929 cells were treated with MG132, Epox or LiCl as described in "B". Cell extracts were prepared in RIPA buffer and endogenous FGD3 was immunoprecipitated with anti-FGD3 antibodies followed by immunoblotting using anti-FGD3 antibodies. Separately, cell extracts were directly subjected to immunoblotting analyses to detect GAPDH levels as a loading control. Quantification of the FGD3 bands was done by densitometric analysis (NIH Image 1.63). (d) L929 cells were transfected with control siRNA (#1022076 from Qiagen) or with mouse FWD1/β-TrCP siRNAs (stealth-349, -533, or -1176 from Invitrogen Life Technology) for 48 h. Total RNA was prepared and subjected to real-time PCR analysis using TaqMan Gene Expression Assay kit. FWD1 mRNA levels were normalized to the level of β-actin mRNA as an internal control. The amount of each FWD1 mRNA is shown as a percentage of the value of control siRNA-treated cells. Results are expressed as average of four independent experiments. Error bars represent SDs. (e) L929 cells were transfected with control siRNA or FWD1/β-TrCP siRNAs for 72 h. Cell extracts were prepared in RIPA buffer and endogenous FGD3 was immunoprecipitated with anti-FGD3 antibodies followed by immunoblotting using anti-FGD3 antibodies. Separately, cell extracts were directly subjected to immunoblotting analyses to detect GAPDH levels as a loading control. Quantification of the FGD3 bands was done by densitometric analysis (NIH Image 1.63). (f) L929 cells were transfected with control siRNA or FWD1/β-TrCP siRNA (stealth-533) for 72 h. MG132 was added to the culture during the last 20 h of incubation. Then the medium was removed and cells were incubated for the indicated times in the presence of cycloheximide (10 µg/mL). Cell extracts prepared in RIPA buffer were precleared with protein G-Sepharose and then subjected to immunoprecipitation using anti-FGD3 antibodies followed by immunoblotting to detect endogenous FGD3. Separately, precleared supernatants were directly subjected to immunoblotting analyses to detect β-catenin and GAPDH.

Because FGD3 and FGD1 share a high degree of sequence identity in the DH domain and in the PH domain, which is immediately adjacent to the DH domain, we predicted FGD3 to be, like FGD1, a GEF for Cdc42. Indeed, over-expression of the DH + PH region of FGD3 was reported to induce striking filopodial extensions (Pasteris et al. 2000). However, the sequence similarity between FGD3 and FGD1 does not extend to the amino or carboxy termini, suggesting that the full-length FGD3 and FGD1 are regulated in different manners in intact cells. Thus, we next examined the morphological changes induced by full-length FGD3. When cells were analyzed by immunostaining, FGD3(WT)-expressing cells as well as FGD3(SA)-expressing ones showed thin protrusive sheet structures, known as lamellipodia (Fig. 4a,b). Over-expression of SCF^FWD1/β-TrCP resulted in the reduced immunofluorescence signal from the FLAG epitope in FGD3(WT)-expressing cells but not in FGD3(SA)-expressing ones (data not shown).

View larger version (43K): [in this window] [in a new window]   Figure 4  Similar morphological changes were induced by FGD3(WT) and FGD3(SA). (a) HeLa Tet-Off-FGD3(WT)cells seeded on coverslips were incubated with or without 2 µg/mL Tc for 72 h. The cells were then analyzed by immunostaining to detect FLAG epitope (green) as described in "Experimental procedures". The nuclei were counterstained with Hoechst 33258 (blue). Separately, differential interference contrast (DIC) images were captured. (b) HeLa Tet-Off-FGD3(SA)cells seeded on coverslips were incubated with or without 2 µg/mL Tc for 72 h. The cells were then subjected to immunostaining analysis as described in "a".

View larger version (34K): [in this window] [in a new window]   Figure 5  FGD3 activates Cdc42 in HeLa Tet-Off cells. (a) HeLa Tet-Off-FGD3(WT) cells seeded on coverslips were transfected with the expression plasmid encoding Myc-tagged N17Rac and then cultured for 72 h in the absence of Tc. The cells were then analyzed by immunostaining to detect FLAG epitope (green) and Myc epitope (red), respectively. The nuclei were counterstained with Hoechst 33258 (blue). Separately, differential interference contrast (DIC) images were captured. (b) HeLa Tet-Off-FGD3(WT) cells seeded on coverslips were transfected with the expression plasmid encoding Myc-tagged N17Cdc42 and then cultured for 72 h in the absence of Tc. The cells were then subjected to immunostaining analysis as described in (a). (c) HeLa Tet-Off-FGD3(WT)cells or HeLa Tet-Off-FGD3(SA) cells were transfected with the expression plasmid encoding Myc-tagged Cdc42 and then incubated with or without 2 µg/mL Tc for 72 h. Cell extracts were then prepared and incubated with GST–CRIB as described in "Experimental procedures". Complexes were collected with glutathione–Sepharose and subjected to SDS-PAGE followed by immunoblotting to detect Myc-Cdc42. (d) Extracts of HeLa Tet-Off-FGD3(SA) cells were prepared and then incubated with GST-fusion Rho proteins as described in "Experimental procedures". Complexes were collected with glutathione–Sepharose and subjected to SDS-PAGE followed by immunoblotting to detect FLAG-FGD3. CBB, Coommassie Brilliant Blue R250-stained recombinant GST-Cdc42, GST-Rac1 and GST-RhoA used in the assay.

View larger version (55K): [in this window] [in a new window]   Figure 6  FGD1 and FGD3 induce different actin cytoskeleton remodelings. (a) HeLa Tet-Off stable cell line that inducibly express FGD1(SA) was established. Hygromycin-resistant clonal cell line was isolated from HeLa Tet-Off cells transfected with pTRE-FLAG-FGD1(SA) and designated as HeLa Tet-Off-FGD1(SA). The cells were incubated with or without 2 µg/mL Tc for 72 h and then lysed in SDS-PAGE sample buffer. The resultant whole cell lysates were subjected to immunoblotting analysis using anti-FLAG antibody or anti-GAPDH antibody. (b) HeLa Tet-Off-FGD1(SA) cells seeded on a coverslip were cultured in the absence of Tc for 72 h. After fixing cells with 4% paraformaldehyde, cells were permeabilized with 0.2% Triton X-100 instead of methanol as described in "Experimental procedures". The cells were then analyzed by immunostaining to detect the FLAG epitope (green). The actin cytoskeleton was visualized with Alexa Fluor 594-conjugated phalloidin (red). The nuclei were counterstained with Hoechst 33258 (blue). Separately, differential interference contrast (DIC) images were captured. Right panels represent the enlarged images of the boxed areas shown in the center panels (scale bar, 10 µM). (c) HeLa Tet-Off-FGD3(SA) cells seeded on a coverslip were cultured in the absence of Tc for 72 h. FLAG epitope (green), actin cytoskeleton (red) and nuclei (blue) were visualized as described in (b). Separately, differential interference contrast (DIC) images were captured. Right panels represent the enlarged images of the boxed areas indicated in the center panels (scale bar, 10 µM).

View larger version (30K): [in this window] [in a new window]   Figure 7  Reciprocal regulation of cell motility by FGD1 and FGD3 expressed in HeLa Tet-Off cells. (a) Schematic flow chart to assess the wound-healing process. Confluent monolayers of cells were wounded, and the phase-contrast images of four frames (a–d) were collected at the indicated times. Then the widths of the wound were measured at four points per frame. By subtracting the widths at each time point from the corresponding widths at 0 h, the closure distances were calculated. Wound closures were then estimated as the averages of these closure distances. (b) Representative images of the wound-healing process of HeLa Tet-Off-FGD1(SA) cells. Cells were incubated with or without 2 µg/mL Tc for 72 h. Confluent monolayers of cells were wounded, then rinsed twice with DMEM, and incubated in 0.1% FCS/DMEM containing 20 mM Hepes/NaOH pH 7.4 for the indicated times. Phase-contrast images were collected by using a x10 objective (scale bar, 0.1 mm). (c) According to the method shown in (a), motility of HeLa Tet-Off-FGD1(SA) cell was assessed as the wound closure at 8 h after wounding. Similar results were obtained in three separate experiments. (d) Representative images of the wound-healing processes of HeLa Tet-Off-FGD3(SA) cells. Cells were incubated with or without 2 µg/mL Tc for 72 h. Confluent monolayers of cells were wounded, then rinsed twice with DMEM, and incubated in 5% FCS/DMEM containing 20 mM Hepes/NaOH, pH 7.4 for the indicated times. Phase-contrast images were collected by using a x20 objective (scale bar, 0.1 mm). (e) According to the methods shown in (a), motility of HeLa Tet-Off-FGD3(SA) cell was assessed as the wound closure at 6 h after wounding. Similar results were obtained in three separate experiments. (f) Transwell motility assay of HeLa Tet-Off-FGD3(SA) cells. Cells were incubated with or without 2 µg/mL Tc for 72 h. Then the cells were harvested and single cell motility was determined as described in "Experimental procedures". Similar results were obtained in two separate experiments.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
The Rho-family small G-proteins are defined by the presence of a Rho-type GTPases-like domain and to date 22 human genes have been described (Wennerberg & Der 2004). Like all GTPases, Rho proteins act as molecular switches by cycling between an active GTP-bound and inactive GDP-bound state. This cycling is regulated by two distinct families of proteins, that is, GEFs and GAPs. Far outnumbering Rho proteins, more than 60 GEFs and approximately 80 GAPs are encoded in human genome (Schmidt & Hall 2002; Moon & Zheng 2003), suggesting that the activities of individual Rho proteins in cells are regulated in temporally and spatially coordinated manners by these over-abundant GEFs and GAPs. Thus, in order to understand the roles of Rho proteins as molecular switches, it is essential to describe how the activities of GEFs and GAPs are regulated in the intracellular environment. Although many studies on the functions of Rho proteins have been done by using dominant negative or constitutively active mutant forms of Rho proteins, GEFs and GAPs physiologically relevant to their regulation have not been characterized extensively.

In the case of GEFs for Rho proteins, most members have been genetically identified by the criteria as to whether or not they possess the DH domain and the PH domain, which is adjacent and C-terminal to the DH domain. In most cases, the DH–PH module is the minimal structural unit that can promote nucleotide exchange in vivo (Schmidt & Hall 2002). However, it is quite important to characterize the full-length GEFs; because most GEFs also contain regulatory domains that are responsible for turning on or turning off the DH–PH-module-driven GEF activity (Schmidt & Hall 2002).

In this study, we characterized the regulatory mechanism and the function of FGD3, which was originally identified as a homologue of Cdc42 GEF FGD1 but has been poorly investigated. We first focused on the sequence "DSGIDS" conserved in FGD1 and FGD3. A similar sequence is also conserved in several unrelated proteins, such as IBs and β-catenin (Fig. 1a), and now is recognized as a destruction motif responsible for the polyubiquitination by SCF^FWD1/β-TrCP and resultant proteasomal degradation of the target molecule (Karin & Ben-Neriah 2000). We previously showed that FGD1 underwent polyubiquitination-coupled proteasomal degradation through the phosphorylation of the two serine residues in its DSGIDS sequence (Hayakawa et al. 2005). By the experiments using pharmacological inhibitors, GSK-3β was shown to be the kinase responsible for the phosphorylation of FGD1 (Hayakawa et al. 2005). In the present study, we demonstrated that FGD3 was also down-regulated by the ubiquitin/proteasome system through the phosphorylation of the serine residues in the DSGIDS motif (Figs 2 and 3). Similar to FGD1, GSK-3β was shown to be the kinase responsible for the phosphorylation of FGD3.

GSK-3β is also known to be responsible for the constitutive phosphorylation of two serine residues in the DSGIHS sequence in β-catenin. In terms of intracellular localization, β-catenin exists in two different pools within cells: a membrane-associated pool of stable protein that participates in the formation of adherence junctions and a soluble pool of highly unstable protein that functions as the signal transducer/transcription factor to promote proliferation (Willert & Nusse 1998).

Analogous to the case of β-catenin, the destruction pathway composed of GSK-3β/SCF^FWD1/β-TrCP/the proteasome may allow FGD1 and FGD3 to be stabilized at appropriate intracellular compartments where they are not targeted by the destruction components. On the other hand, molecules recognized by the active destruction components may be constitutively degraded in intracellular regions where the activities of FGD1 or FGD3 are not required. Interestingly, GSK-3β is reported to be inactivated at the leading edge of migrating cells as a result of phosphorylation of its serine-9 by atypical protein kinase C (Etienne-Manneville & Hall 2003). At the leading edge, phosphatidylinositol(3,4,5)-trisphosphate (PIP₃) is known to accumulate (Meili et al. 1999; Haugh et al. 2000; Servant et al. 2000) and Akt/PKB, which is activated downstream of PIP₃, is also recognized as the kinase responsible for the phosphorylation of GSK-3β at its serine-9 (Dobel & Woodgett 2003; Woodgett 2005). Thus, it is likely that FGD1 and FGD3 are stabilized at the PIP₃-enriched lipid microdomains by binding to PIP₃ through their PH domains.

Although FGD1 and FGD3 were down-regulated similarly by the GSK-3β/SCF^FWD1/β-TrCP/the proteasome-regulated pathway, their biological functions are obviously different as discussed below. Pasteris et al. previously reported that the DH + PH module of FGD3 induced a large number of striking filopodia extensions in NIH 3T3 cells whereas no stress fiber or overt lamellipodia were observed (Pasteris et al. 2000). Their data suggest that, like FGD1, FGD3 may act as a GEF for Cdc42. In our study using full-length FGD3, the activation of Cdc42 by FGD3 was also supported by the following two observations: The results of the GST–CRIB pull-down assay demonstrated significant elevation of the GTP-bound form of Cdc42 in HeLa Tet-Off cells that inducibly over-expressed full-length FGD3 (Fig. 5c). Furthermore, the direct interaction of full-length FGD3 with recombinant GST-Cdc42 was observed in vitro (Fig. 5d). However, inducible expression of FGD3 caused remarkable morphological changes that included the formation of thin and sheet-like protrusions, known as lamellipodia, which are known to be induced by the activation of Rac (Fig. 4). Indeed, over-expression of N17Rac1 inhibited the lamellipodia formation, while striking filopodia-like protrusions became obvious in FGD3-expressing cells (Fig. 5a). These observations may reflect the classical hierarchical cascade wherein Cdc42 activates Rac (Van Aelst & D'Souza-Schorey 1997). Unexpectedly, we could not see the FGD3-dependent increase in the level of GTP-Rac1 in FGD3-expressing cells by use of the GST–CRIB pull-down assay. This may reflect the difficulty to minimize the GTPase activity of Rac1 in HeLa Tet-Off-FGD3(SA) cells under the conditions currently used for GST–CRIB pull-down assays. In addition, it should be noted that the pull-down assay using the GST–CRIB fusion protein estimates the entire amount of GTP-Rac1 in cells and is not suitable for the analysis of Rac1 activation that may occur in limited areas in cells. It is likely that FGD3 is recruited to the PIP₃-enriched lipid subdomain through its PH domains, and causes Cdc42 activation that subsequently leads to the local activation of Rac1. However, if the entire GTP-Rac1 level in cells is not negligible before FGD3(SA) induction, it would be difficult to assess the effect of FGD3(SA) expression on the small increase in the GTP-Rac1 level in specific intracellular compartments.

Current data suggest that filopodia arise from the elongation, convergence and bundling of pre-existing actin filaments in the lamellipodium (Svitkina et al. 2003). Interestingly, by staining the F-actin cytoskeleton, the broad sheet-like protrusions recognized as lamellipodia in FGD3(SA)-expressing cells were shown to contain many microspikes that were clearly distinguished from the finger-like long protrusions induced by FGD1(SA). Furthermore, at the tips of several microspikes, FGD3(SA) was clearly stained; however, actin filaments were not stained at all. In contrast, FGD1(SA) seemed to distribute widely in a filopodium where F-actin was concomitantly stained. Svitkina et al. reported that interconversion between microspikes, filopodia and retraction fibers occurs at lamellipodial protrusions (Svitkina et al. 2003). Although the molecular mechanism to regulate this interconversion has not been fully elucidated yet, FGD1 and FGD3 may participate in this process in different manners.

Functional differences between FGD1 and FGD3 should arise from the structural differences in both molecules. The most striking structural difference between the two is the presence (FGD1) or absence (FGD3) of the N-terminal proline-rich domain. Many proline-rich proteins are known to participate in delivering actin monomers to specific cellular locations where actin-rich membrane protrusions are formed (Holt & Koffer 2001). Proline-rich sequences on these proteins serve to bind their docking partner such as EVH1/WH1, SH3 and WW domains on other proteins. Indeed, FGD1 has been already shown to bind to the SH3 domain of cortactin, which is a c-src substrate associated with sites of dynamic actin assembly at the leading edge of migrating cells (Hou et al. 2003; Kim et al. 2004). Kim et al. demonstrated that the amino acid residues 158–165 [KPQVPPKP] in the proline rich domain of FGD1 binds directly to the SH3 domain of cortactin, which promotes actin assembly by actin-related protein Arp2/3 complex (Kim et al. 2004). Besides the SH3-binding motif, the EVH1/WH1-binding motif also plays an important role in the regulation of actin remodeling. Wiskott–Aldrich syndrome protein (WASP) and its homologue N-WASP, both of which are activated downstream of Cdc42 and are responsible for filopodium formation, contain EVH1/WH1 domain at their N-terminal region (Holt & Koffer 2001; Yamazaki et al. 2005). WIP (WASP-interacting protein), which was identified as a protein interacting with WASP, contains a potential EVH1/WH1-binding motif at residues 462–467 (LPPPEP) (Holt & Koffer 2001). Interestingly, FGD1 contains the same motif [LPPPEP] at residues 174–179 located in its proline-rich domain. Thus FGD1, but not FGD3, might directly participate in the filopodium formation by binding to the regulators of actin assembly, such as cortactin or WASPs.

As the most significant functional difference between FGD1 and FGD3, the opposite regulation of cell motility was demonstrated (Fig. 7). It is widely accepted that Cdc42 induces filopodia formation and functions to provide cellular polarity, whereas Rac promotes lamellipodia formation leading to the elevation of cell motility (Ridley 2001). Because FGD3(SA)-expressing cells showed broad sheet-like protrusions, so-called lamellipodia, the cell motility of FGD3(SA)-expressing cells was expected to be higher than that of cells in which the expression of FGD3(SA) was suppressed. However, both the collective cell migration assessed by wound-healing assay and the single cell migration assessed by the assay using chemotactic chambers were rather inhibited by the expression of FGD3(SA). In contrast, FGD1(SA)-expressing cells, in which lamellipodia formation was less prominent, showed elevated cell motility.

Pankov et al. reported that a relatively small change in Rac1 activity can serve as a switch that can regulates the overall intrinsic pattern of cell migration of a cell (Pankov et al. 2005). Their findings suggest that Rac1 levels regulate persistence of migration by controlling the number of peripheral lamellae and associated total amount of membrane protrusions that can mediate cell turning. When FGD3(SA) was induced in HeLa Tet-Off cells, several cells showed two broad sheet-like lamellae at the opposite sides of a cell. The "lamellipodium" is sometimes referred as the "leading edge" or the "leading lamella" (Small et al. 2002). In the case of FGD3(SA)-expressing cells having these two broad sheet-like lamellae oriented in opposite directions, it is therefore difficult to know which of them represents the "leading lamella". Thus cell motility in one direction might be disturbed in these cells, resulting in the decreased cell migration shown in Fig. 7. In the case of FGD1(SA)-expressing cells, such broad sheet-like lamellae were not formed. In these cells, FGD1-induced filopodia formation via the activation of Cdc42 might lead to the directional cell migration by activating the downstream signaling mediated by WASPs and Arp2/3 (Holt & Koffer 2001; Yamazaki et al. 2005).

In this study, we characterized two highly homologous GEFs, FGD1 and FGD3. They are subjected to the same destruction pathway through the same motif conserved on both proteins. On the other hand, these homologous but distinct GEFs are shown to play different roles to regulate cell morphology or motility. The studies focusing on the roles of full-length FGD1 and FGD3 enabled us to reveal their unique characteristics. However, we should not simply conclude about FGD1 and FGD3 that the former acts as a stimulatory GEF and the latter acts as an inhibitory GEF in the regulation of cell motility. It is likely that the functions of each GEF vary in different types of cells. Further studies are necessary to reveal how GEFs are temporally and spatially controlled by interacting with various signaling molecules in different cellular environments.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Reagents and antibodies

Antibodies (Abs) specific for GAPDH (clone 6C5), HA epitope (Y-11), Myc epitope (A-14) and ubiquitin (clone P4D1) were purchased from Santa Cruz Biotechnology, Santa Cruz, CA. Anti-phospho-β-catenin (Ser33/37/Thr41) Ab was from Cell Signaling Technology, Danvers, MA. Tetracycline (Tc) and anti-FLAG M2 Ab were obtained from Sigma-Aldrich, St. Louis, MO; and anti-HA (clone 12CA5) Ab and anti-Myc Ab (clone 9E10) from Roche Applied Science, Basel, Switzerland. Proteasome inhibitor MG132 and GSK-3β inhibitor [3-(3-carboxy-4-chloroanilino)-4-(3-nitrophenyl) maleimide] came from Calbiochem, San Diego, CA. FITC-conjugated AffiniPure goat anti-mouse IgG and Texas Red-conjugated AffiniPure anti-rabbit IgG were products of Jackson ImmunoResearch Laboratories, West Grove, PA. Polyclonal antibodies against FGD3 were obtained by immunizing rabbits with a KLH-coupled synthetic peptide corresponding to the amino acid residues K⁵²⁴TRRDKEK⁵³¹ located in the FYVE domain of FGD3. The antibodies were then isolated from the serum by peptide affinity chromatography.

cDNA construction and plasmids

A cDNA encoding full-length mouse FGD3 was generated by RT-PCR based on the reported nucleotide sequence (accession number AF017369). Briefly, poly (A+) RNAs were isolated from mouse L929 cells by using a QuickPrep Micro mRNA Purification Kit (GE Healthcare, Little Chalfont, UK) and then reverse-transcribed by the use of Ready-To-Go-You-Prime First-Strand Beads (GE Healthcare) according to the manufacturer's instructions. FGD3 cDNA was amplified with Takara LA Taq (Takara Co. Ltd, Otsu Japan). A FLAG tag was created by PCR. In order to replace Ser72 and Ser76 with alanines, we carried out site-directed mutagenesis by PCR using the following primers designed in inverted tail-to-tail direction to amplify the vector together with the target sequence: 5'-ATTGACGCCCCATCCTCCAGT-3' (sense) and 5'-TCCAGCGTCTCTGTTGGGGAT-3' (anti-sense). FLAG-FGD3 and its S/A mutant cDNAs were subcloned into the pTRE vector (Clontech, Palo Alto, CA), and the resultant plasmids were termed pTRE-FLAG-FGD3(WT) and pTRE-FLAG-FGD3(SA), respectively. Other plasmids used in this study were previously described (Hayakawa et al. 2003, 2005). All the amplified sequences were verified by dideoxy sequencing.

Transfection and establishment of stable cell lines

Transfection was carried out by using FuGENE 6 (Roche Applied Science) according to the manufacturer's instruction. HeLa Tet-Off cells purchased from Clontech were co-transfected either with pTRE-FLAG-FGD3(WT) and pTK-Hyg or with pTRE-FLAG-FGD3(SA) and pTK-Hyg. The hygromycin-resistant clonal cell lines were tested for Tc-regulated FLAG-tagged FGD3 expression.

RNA interference experiments

To knockdown the endogenous FWD1/β-TrCP expression, siRNAs against mouse FWD1/β-TrCP were designed using "BLOCK-iT RNAi Designer (Invitrogen)". The stealth 349 (#349; 5'-CATAAACCAAGAGACAGTATGTCTA-3'), stealth 533 (#533; 5'-CAGGTGGAATTTGTAGAACACCTTA-3') and stealth 1176 (#1176; 5'-CCACCGTCAGAGTGTGGGAT GTAAA-3') can target the sequences conserved in both β-TrCP-1 and β-TrCP-2 transcript variants in mouse. L929 cells were seeded on 60-mm dishes followed by the transfection with siRNAs (100 nM) using HiPerFect Transfection Reagent (Qiagen). To evaluate the knockdown efficiency of the siRNAs, total RNA was prepared from the siRNA-treated cells and 5 µg of total RNA was used for real-time PCR. PCR analysis was performed using TaqMan Gene Expression Assay kit (Mm00477680 m1; Applied Biosystems).

Immunoblotting, immunoprecipitation and immunostaining

Cells were washed with PBS and cell extracts were prepared by using either SDS-PAGE sample buffer or IP buffer as described below. After normalization of protein content by the protein assay, samples were resolved by SDS-PAGE and subjected to immunoblotting analysis. Immunocomplexes on PVDF membranes were visualized by enhanced chemiluminescence detection (GE Healthcare). For immunoprecipitation analyses, cells were lysed in IP buffer (25 mM Hepes/ NaOH [pH 7.4], 0.3 M NaCl, 10% glycerol, 1% NP-40, 2 mM EDTA, 1 mM EGTA, 2 mM DTT, 20 mM β-glycerophosphate, 10 mM p-nitrophenyl phosphate (PNPP), 20 mM NaF, 0.5 mM Na-O-vanadate, 1 mM PMSF, 10 µg/mL leupeptin, 10 µg/mL aprotinin, 3 µg/mL pepstatin and 15 µg/mL bestatin). In experiments to detect the interaction of FGD3 with FWD1/β-TrCP, DTT was excluded from the IP buffer. To detect the phosphorylated forms or polyubiquitinated forms of FGD3, cell extracts were prepared by using RIPA buffer (IP buffer supplemented with 0.5% sodium deoxycholate and 0.1% SDS) to minimize the coimmunoprecipitation of other proteins. The cell extracts were incubated with the relevant antibodies and protein G-Sepharose beads, and the immunoprecipitates were resolved by SDS-PAGE followed by immunoblotting. Immunostaining analyses were carried out as follows: Cells cultured on glass coverslips were fixed with 4% paraformaldehyde, then permeabilized with methanol and blocked with 10% skim milk in PBS. The cells were incubated with primary Abs for 2 h, followed by incubation with secondary Abs (FITC-conjugated goat anti-mouse IgG and Texas Red-conjugated anti-rabbit IgG) for 1 h. Nuclei were counterstained with Hoechst 33258. To stain the actin cytoskeleton, we made the cells permeable with 0.2% Triton X-100 instead of methanol. After blocking with 10% skim milk in PBS, the cells were double stained with Alexa Fluor 594 phalloidin (Molecular Probes) and anti-FLAG M2 Ab and then incubated with the secondary Ab.

Fluorescence images were captured by CCD camera (QICAM FAST1394; QIMAGING) mounted on a Leica DM IRB microscope using IP Laboratory imaging software (Scanalytics, Billerica, MA).

GST–CRIB pull-down assay

CRIB domain was prepared as a GST-fusion protein as described previously (Hayakawa et al. 2005). HeLa Tet-Off-FGD3(SA) cells were treated or left untreated with 2 µg/mL Tc for 24 h and then transfected with the expression plasmid encoding Myc-tagged Cdc42 followed by the incubation for 48 h. The cells were then lysed in modified RIPA buffer (50 mM Hepes/NaOH [pH 7.4], 10 mM MgCl₂, 0.4 M NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM PMSF, 10 µg/mL leupeptin and 10 µg/mL aprotinin) and the resultant cell extracts were incubated for 1 h at 4 °C with GST–CRIB (10 µg) bound to glutathione–Sepharose, and subsequently washed 3 times with the lysis buffer described above. The resultant beads were subjected to SDS-PAGE followed by immunoblotting to detect the Myc epitope.

Pull-down assay for FGD3 by Rho family G proteins

HeLa Tet-Off cells expressing FGD3(SA) grown in two 10-cm dishes were lysed in buffer containing 20 mM Hepes/NaOH [pH 7.4], 0.15 M NaCl, 10% glycerol, 1% NP-40, 5 mM MgCl₂, 2 mM DTT, 20 mM β-glycerophosphate, 10 mM PNPP, 20 mM NaF, 0.5 mM Na-O-vanadate, 1 mM PMSF, 10 µg/mL leupeptin, 10 µg/mL aprotinin, 3 µg/mL pepstatin and 15 µg/mL bestatin. The resultant cell extracts were precleared by incubation with glutathione–Sepharose beads and then incubated for 1 h at 4 °C with 10 µg of GST-Cdc42, GST-Rac1 or GST-RhoA bound to glutathione–Sepharose, and then washed 3 times with the lysis buffer described above. The resultant beads were subjected to SDS-PAGE followed by immunoblotting to detect the FLAG epitope.

Wound-healing assay

Cells seeded on 35-mm dishes were allowed to grow to confluence. Monolayers of confluent cultures were lightly scratched with a 1000-µL disposable plastic pipette tip (#111 purchased from Quality). After having been washed with DMEM, the wound regions were allowed to heal in the medium described in the legend to Fig. 7.

Transwell motility assay

Cells were suspended in 5% FCS/DMEM and then washed twice with DMEM containing 0.25% BSA (BSA/DMEM) and resuspended in BSA/DMEM at a density of 1 x 10⁶ cells/mL. Cell migration was studied in blind-well chemotactic chambers (Neuroprobe) fitted with polycarbonate filters (8-µm pores; Costar, High Wycombe, UK) precoated with 100 µg/mL type IV collagen. Assays were performed with 200 µL of cell suspension in the top wells and 0.1% FCS/DMEM in the bottom wells. After incubation for 4 h, the filters were fixed with methanol and stained with hematoxylin. The number of cells migrating into the filter in a 1-mm² area was counted.

	Acknowledgements

 
We thank Dr Junji Yamauchi, Dr Yuki Miyamoto and Dr Yoshiko Aoki for helpful advice and discussions. We also thank Chinami Haga, Yuko Tsunoda and Yusuke Takahashi for their excellent technical assistance. This work was supported in part by a grant from the Japan Private School Promotion Foundation.

	Footnotes

 
Communicated by: Keiichi I. Nakayama

^* Correspondence: Email: hayakawa{at}ps.toyaku.ac.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Bender, A. & Pringle, J. R. (1989) Multicopy suppression of the cdc24 budding defect in yeast by CDC42 and three newly identified genes including the ras-related gene RSR1. Proc. Natl. Acad. Sci. USA 86, 9976–9980.[Abstract/Free Full Text]

Dobel, B.W. & Woodgett, J.R. (2003) GSK-3: tricks of the trade for a multi-tasking kinase. J. Cell Sci. 116, 1175–1186.[Abstract/Free Full Text]

Etienne-Manneville, S. & Hall, A. (2002) Rho GTPases in cell biology. Nature 420, 629–635.[CrossRef][Medline]

Etienne-Manneville, S. & Hall, A. (2003) Cdc42 regulates GSK-3β and adenomatous polyposis coli to control cell polarity. Nature 421, 753–756.[CrossRef][Medline]

Eva, A. & Aaronson, S.A. (1985) Isolation of a new human oncogene from a diffuse B-cell lymphoma. Nature 316, 273–275.[CrossRef][Medline]

Glickman, M.H. & Ciechanover, A. (2002) The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol. Rev. 82, 373–428.[Abstract/Free Full Text]

Haugh, J.M., Codazzi, F., Teruel, M. & Meyer, T. (2000) Spatial sensing in fibroblasts mediated by 3' phosphoinositides. J. Cell Biol. 151, 1269–1279.[Abstract/Free Full Text]

Hayakawa, M., Kitagawa, H., Miyazawa, K., Kitagawa, M. & Kikugawa, K. (2005) The FWD1/β-TrCP-mediated degradation pathway establishes a "turning off switch" of a Cdc42 guanine nucleotide exchange factor, FGD1. Genes Cells 10, 241–251.[Abstract/Free Full Text]

Hayakawa, M., Miyashita, H., Sakamoto, I., Kitagawa, M., Tanaka, H., Yasuda, H., Karin, M. & Kikugawa, K. (2003) Evidence that reactive oxygen species do not meditate NF-B activation. EMBO J. 22, 3356–3366.[CrossRef][Medline]

Holt, M.R. & Koffer, A. (2001) Cell motility: proline-rich proteins promote protrusions. Trends Cell Biol. 11, 38–46.[CrossRef][Medline]

Hou, P., Estrada, L., Kinley, A.W., Parsons, J.T., Vojtek, A.B. & Gorski, J.L. (2003) FGD1, the Cdc42 GEF responsible for Faciogenital Dysplasia, directly interacts with cortactin and mAbp1 to modulate cell shape. Hum. Mol. Genet. 12, 1981–1993.[Abstract/Free Full Text]

Karin, M. & Ben-Neriah, Y. (2000) Phosphorylation meets ubiquitination: the control of NF-B activity. Annu. Rev. Immunol. 18, 621–663.[CrossRef][Medline]

Katzav, S., Cleveland, J.L., Heslop, H.E. & Pulido, D. (1991) Loss of the amino-terminal helix-loop-helix domain of the vav proto-oncogene activates its transforming potential. Mol. Cell. Biol. 11, 1912–1920.[Abstract/Free Full Text]

Kim, K., Hou, P., Gorski, J.L. & Cooper, J.A. (2004) Effect of FGD1 on cortactin in Arp2/3 complex-mediated actin assembly. Biochemistry 43, 2422–2427.[CrossRef][Medline]

van Leeuwen, F.N., van der Kammen, R.A., Habets, G.G. & Collard, J.G. (1995) Oncogenic activity of Tiam 1 and Rac1 in NIH 3T3 cells. Oncogene 11, 2215–2221.[Medline]

Malliri, A. & Collard, J.G. (2003) Role of Rho-family proteins in cell adhesion and cancer. Curr. Opin. Cell Biol. 15, 583–589.[CrossRef][Medline]

Mantovani, F. & Banks, L. (2003) Regulation of the discs large tumor suppressor by a phosphorylation-dependent interaction with the β-TrCP ubiquitin ligase. J. Biol. Chem. 278, 42477–42486.[Abstract/Free Full Text]

Meili, R., Ellsworth, C., Lee, S., Reddy, T.B.K., Ma, H. & Firtel, R.A. (1999) Chemoattractant-mediated transient activation and membrane localization of Akt/PKB is required for efficient chemotaxis to cAMP in Dictyostelium. EMBO J. 18, 2092–2105.[CrossRef][Medline]

Miki, T., Smith, C.L., Long, J.E., Eva, A. & Fleming, T.P. (1993) Oncogene ect2 is related to regulator of small GTP-binding proteins. Nature 362, 462–465.[CrossRef][Medline]

Moon, S.Y. & Zheng, Y. (2003) Rho GTPase-activating proteins in cell regulation. Trends Cell Biol. 13, 13–22.[CrossRef][Medline]

Pankov, R., Endo, Y., Even-Ram, S., Araki, M., Clark, K., Cukierman, E., Matsumoto, K. & Yamada, K.M. (2005) A Rac switch regulates random versus directionally persistent cell migration. J. Cell Biol. 170, 793–802.[Abstract/Free Full Text]

Pasteris, N.G., Nagata, K., Hall, A. & Gorski, J.L. (2000) Isolation, characterization, and mapping of the mouse FGD3 gene, a new faciogenital dysplasia (FGD1; Aarskog syndrome) gene homologue. Gene 242, 237–247.[CrossRef][Medline]

Ridley, A.J. (2001) Rho GTPases and cell migration. J. Cell Sci. 114, 2713–2722.[Abstract/Free Full Text]

Ron, D., Granziani, G., Aaronson, S.A. & Eva, A. (1989) The N-terminal region of proto-Dbl down regulates its transforming activity. Oncogene 4, 1067–1072.[Medline]

Schmidt, A. & Hall, A. (2002) Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev. 16, 1587–1609.[Free Full Text]

Servant, G., Weiner, O.D., Herzmark, P., Balla, T., Sedat, J.W. & Bourne, H.R. (2000) Polarization of chemoattractant receptor signaling during neutrophil chemotaxis. Science 287, 1037–1040.[Abstract/Free Full Text]

Small, J.V., Stradal, T., Vignal, E. & Rottner, K. (2002) The lamellipodium: where motility begins. Trends Cell Biol. 12, 112–120.[CrossRef][Medline]

Svitkina, T.M., Bulanova, E.A., Chaga, O.Y., Vignjevic, D.M., Kojima, S., Vasiliev, J.M. & Borisy, G.G. (2003) Mechanism of filopodia initiation by reorganization of a dendritic network. J. Cell Biol. 160, 409–421.[Abstract/Free Full Text]

Van Aelst, L. & D'Souza-Schorey, C. (1997) Rho GTPases and signaling networks. Genes Dev. 11, 2295–2322.[Free Full Text]

Wennerberg, K. & Der, C.J. (2004) Rho-family GTPases: it's not only Rac and Rho (and I like it). J. Cell Sci. 117, 1301–1312.[Abstract/Free Full Text]

Willert, K. & Nusse, R. (1998) β-catenin: a key mediator of Wnt signaling. Curr. Opin. Genet. Dev. 8, 95–102.[CrossRef][Medline]

Woodgett, J.R. (2005) Recent advances in the protein kinase B signaling pathway. Curr. Opin. Cell Biol. 17, 150–157.[CrossRef][Medline]

Yamazaki, D., Kurisu, S. & Takenawa, T. (2005) Regulation of cancer cell motility through actin reorganization. Cancer Sci. 96, 379–386.[CrossRef][Medline]

Zheng, Y., Fischer, D.J., Santos, M.F., Tigyi, G., Pasteris, N.G., Gorski, J.L. & Xu, Y. (1996) The faciogenital dysplasia gene product FGD1 functions as a Cdc42Hs-specific guanine nucleotide exchange factor. J. Biol. Chem. 271, 33169–33172.[Abstract/Free Full Text]

Received: 20 July 2007
Accepted: 24 December 2007

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hayakawa, M. Articles by Kitagawa, M. Search for Related Content PubMed Articles by Hayakawa, M. Articles by Kitagawa, M.