Cell migration is essential for normal development and many pathological processes. Rho-family small GTPases play important roles in this event. In particular, Rac regulates lamellipodia formation at the leading edge during migration. The small GTPase RhoG activates Rac through its effector ELMO and the ELMO-binding protein Dock180, which functions as a Rac-specific guanine nucleotide exchange factor. Here we investigated the role of RhoG in cell migration. RNA interference-mediated knockdown of RhoG in HeLa cells reduced cell migration in Transwell and scratch-wound migration assays. In RhoG-knockdown cells, activation of Rac1 and formation of lamellipodia at the leading edge in response to wounding were attenuated. By contrast, expression of active RhoG promoted cell migration through ELMO and Dock180. However, the interaction of Dock180 with Crk was dispensable for the activation of Rac1 and promotion of cell migration by RhoG. Taken together, these results suggest that RhoG contributes to the regulation of Rac activity in migrating cells.
Cell migration plays a central role in many biological and pathological processes, including embryonic development, wound repair, the inflammatory response, tumor metastasis and mental retardation. Cell migration can be considered as a highly integrated multistep cycle process (Lauffenburger and Horwitz, 1996; Webb et al., 2002; Ridley et al., 2003). The migration cycle includes cell polarization, extension of protrusions in the direction of migration, formation of stable adhesion near the leading edge of the protrusions, and detachment of the adhesion and retraction at the rear. These steps require spatially regulated changes of the actin cytoskeleton. It is already well known that members of the Rho family of small GTPases are key regulators of the actin cytoskeleton in diverse cellular functions including cell migration (Ridley, 2001; Etlenne-Manneville and Hall, 2002; Raftopoulou and Hall, 2004). Like other small GTPases, Rho-family GTPases serve as molecular switches by cycling between an inactive GDP-bound state and an active GTP-bound state, and activated GTPases can bind to their specific effectors that lead to a variety of biological functions. Activation of Rho-family GTPases requires GDP-GTP exchange catalyzed by various guanine nucleotide exchange factors (GEFs). Of Rho GTPases, Rac is activated at the leading edge of motile cells and induces the formation of actin-rich lamellipodia protrusions, which serve as a major driving force of cell movement (Nobes and Hall, 1999; Kraynov et al., 2000; Small et al., 2002). The major downstream proteins for Rac that mediate actin polymerization in lamellipodia protrusions are the WAVE family proteins - the activators of the Arp2/3 complex (Miki et al., 1998; Yamazaki et al., 2003). Activated Arp2/3 complex induces rapid polymerization of actin and the formation of the branched actin filaments present in lamellipodia (Welch and Mullins, 2002; Pollard and Borisy, 2003). However, precise mechanisms that lead to Rac activation during cell migration are not fully understood.
Dock180 was originally identified as a 180 kDa protein that binds to the Crk family of adaptor proteins (Hasegawa et al., 1996). Recent studies have shown that Dock180 contains a domain that can directly and specifically bind to Rac and mediate GDP-GTP exchange of Rac in vitro, although it contains no Dbl homology-pleckstrin homology (DH-PH) tandem domain that is present in other known GEFs for Rho-family GTPases (Brugnera et al., 2002; Côté and Vuori, 2002). Furthermore, several proteins that possess this novel GEF domain have been identified and constitute an evolutionarily conserved superfamily of Dock180-related proteins (Côté and Vuori, 2002; Meller et al., 2002). Dock180 forms a complex with ELMO, another Dock180-binding protein that is also evolutionarily conserved from Caenorhabditis elegans to mammals (Gumienny et al., 2001; Zhou et al., 2001; Wu et al., 2001), and the ELMO-Dock180 complex serves as a functional GEF for Rac in intact cells (Brugnera et al., 2002; Lu et al., 2004). They functionally cooperate to promote phagocytosis of apoptotic cells and cell migration through activation of Rac (Gumienny et al., 2001; Grimsley et al., 2004).
RhoG is a member of Rho-family GTPases, and several lines of evidence have shown that RhoG is involved in the regulation of neurite outgrowth in neuronal cells through activation of Rac (Negishi and Katoh, 2002; Estrach et al., 2002; May et al., 2002). Trio, which contains two DH-PH domains, has been shown to promote GDP-GTP exchange of RhoG in vitro through its N-terminal DH-PH domain and to function as a GEF for RhoG in neuronal cells during neurite outgrowth (Blangy et al., 2000; Estrach et al., 2002; Skowronek et al., 2004). We have recently identified ELMO as an effector for RhoG (Katoh and Negishi, 2003). The interaction of RhoG with ELMO induces translocation of the ELMO-Dock180 complex from the cytoplasm to the plasma membrane and activates Rac1. The RhoG-ELMO-Dock180 signaling pathway is required for the nerve growth factor-stimulated neurite outgrowth. In addition to the regulation of neurite outgrowth, RhoG also induces other cellular functions, such as regulation of gene expression and cell adhesion in lymphocytes and stimulation of macropinocytosis in fibroblasts (Vigorito et al., 2003; Ellerbroek et al., 2004). However, the role of RhoG in cell migration remains to be elucidated. In this study, we show that RhoG regulates cell migration by activating Rac1. Furthermore, ELMO and Dock180, but not Crk, participate in the regulation of Rac1 activity and cell migration by RhoG.
Knockdown of endogenous RhoG by RNA interference (RNAi) reduces cell migration
To evaluate the role of RhoG in cell migration, we performed RNAi-mediated knockdown of RhoG in HeLa cells using a short hairpin RNA (shRNA) expression vector. We generated two shRNA vectors designed to target two different regions of human RhoG cDNA (pSilencer-RhoGa and pSilencer-RhoGb; for details see Materials and Methods). One vector (pSilencer-RhoGa) effectively reduced the amount of exogenously expressed Myc-tagged human RhoG without affecting the level of Myc-tagged Rac1 expression, whereas the other (pSilencer-RhoGb) had no effect on the amounts of RhoG and Rac1 expression (Fig. 1A, left). Therefore, we used pSilencer-RhoGb as a control shRNA expression vector. Stable HeLa cell clones expressing pSilencer-RhoGa and pSilencer-RhoGb were generated, and the amount of endogenous RhoG expression was examined in these clonal HeLa cells. Because we could not obtain antibodies for detecting specifically endogenous RhoG, we used a monoclonal antibody (mAb) raised against Rac1, which could effectively recognize human RhoG (Fig. 1A, right). Immunoblot analysis using the anti-Rac1 mAb showed that the level of endogenous RhoG was reduced in HeLa cells expressing pSilencer-RhoGa (RhoG RNAi HeLa cells) compared with the level in cells expressing pSilencer-RhoGb (control RNAi HeLa cells) or parental cells (Fig. 1B). We also confirmed that the expression of endogenous RhoG mRNA was reduced in RhoG RNAi HeLa cells by semiquantitative RT-PCR analysis (Fig. 1C). However, the levels of endogenous Dock180 and Rac1, which act in the downstream signaling pathway of RhoG, and Dock180-binding protein CrkII in RhoG RNAi HeLa cells were similar to those in parental and control RNAi HeLa cells (Fig. 1B). Because we could not detect endogenous ELMO protein in HeLa cells using commercial antibodies or those raised in our laboratory, the levels of endogenous ELMO2 mRNA expression were examined by semiquantitative RT-PCR analysis, and were similar among HeLa cell clones (Fig. 1C).
We have previously reported that expression of dominant-negative RhoG suppressed HeLa cell spreading and the formation of membrane ruffling when cells were plated onto fibronectin (Katoh and Negishi, 2003). We also examined the morphology of RhoG RNAi HeLa cells after plating onto fibronectin. Control RNAi HeLa cells began to spread by 15 minutes after plating, and the average of cell area was similar to that of parental cells. However, RhoG RNAi HeLa cells were impaired in their ability to spread onto fibronectin at 15 minutes after plating (Fig. 2A,B). F-actin staining showed that RhoG RNAi HeLa cells failed to induce extensive membrane ruffling at the cell periphery in response to fibronectin. By contrast, RhoG RNAi HeLa cells finally began to spread by 60 minutes after plating onto fibronectin (Fig. 2C). To confirm that the inability of RhoG RNAi HeLa cells to spread onto fibronectin rapidly was due to decreased RhoG protein level in the cells, we performed a rescue experiment. Although the amino acid sequence of mouse RhoG is 100% identical to that of human RhoG, their nucleotide sequences in the RhoG RNAi target region contain three different nucleotides, and a transient transfection experiment with HEK 293T cells showed that mouse RhoG was insensitive to pSilencer-RhoGa (Fig. 2D). Therefore, mouse RhoG was used for our rescue experiment. RhoG RNAi HeLa cells were transiently transfected with GFP or GFP-tagged mouse RhoG (GFP-mRhoG) and plated onto fibronectin. Although expression of GFP in RhoG RNAi HeLa cells had no effect on the cell morphology, expression of GFP-mRhoG induced cell spreading and membrane ruffling at the cell periphery at 15 minutes after plating onto fibronectin (Fig. 2E). These results confirm our previous data that RhoG is involved in the integrin-mediated cell spreading and formation of membrane ruffling.
To investigate the role of RhoG in cell migration, we used a Transwell migration assay with parental, control RNAi and RhoG RNAi HeLa cells. The cells were plated in the upper chamber of the filters that had been coated with fibronectin on the underside, and were allowed to migrate for 4 hours. The number of control RNAi HeLa cells that had migrated to the underside of the filters was similar to that of parental cells, whereas RhoG RNAi HeLa cells showed a significantly reduced motility compared with parental or control RNAi HeLa cells (Fig. 3A). There were no significant differences in the ability to adhere to the filters between control RNAi and RhoG RNAi HeLa cells (data not shown). We confirmed that the reduced motility of RhoG RNAi HeLa cells was due to decreased RhoG protein level in the cells by a rescue experiment with GFP-tagged mouse RhoG. Although expression of GFP in RhoG RNAi HeLa cells had no effect on the cell motility, expression of GFP-mRhoG fully rescued the reduced cell motility, and the number of the cells that had migrated to the underside of the filters was almost the same as that of control RNAi HeLa cells transfected with GFP-mRhoG (Fig. 3B). These results indicate that RhoG is required for the ability of HeLa cells to migrate normally. In the rescue experiment, we found that expression of GFP-mRhoG also increased migration of control RNAi HeLa cells, further supporting the idea that RhoG functions as a positive regulator in cell migration.
By using stable cell lines expressing shRNAs against RhoG, our data showed the involvement of endogenous RhoG in the regulation of cell migration. To confirm further the involvement of RhoG in this process, we used the N-terminal region of ELMO, which specifically binds to and can trap activated RhoG (Katoh and Negishi, 2003). Parental HeLa cells transiently transfected with GFP or GFP-tagged N-terminal region of ELMO (GFP-ELMO-NT) were subjected to the Transwell migration assay, and the result showed that expression of GFP-ELMO-NT significantly reduced migration of parental HeLa cells compared with expression of control GFP (Fig. 3C).
RhoG regulates Rac1 activity and lamellipodia formation during cell migration
To observe more directly the effect of knockdown of RhoG on the morphology of HeLa cells during migration, wound-healing assays were performed using control RNAi, RhoG RNAi and parental HeLa cells. The rates of wound closure in control RNAi and parental cells were similar for a time period of 9 hours after wounding (22.5±0.43 and 22.7±0.63 μm/hour, respectively). By contrast, RhoG RNAi HeLa cells closed the wound at a rate slower than those of control RNAi and parental cells (12.2±1.8 μm/hour; Fig. 4A). F-actin staining showed that control RNAi and parental cells extended broad lamellipodia into the open wound area at 2 hours after wounding; however, RhoG RNAi cells exhibited only minimal lamellipodia at the wound edge (Fig. 4B). We also performed a rescue experiment with GFP-tagged mouse RhoG. Expression of GFP in RhoG RNAi HeLa cells had little effect on the morphology of the cells at the wounding edge. However, expression of GFP-mRhoG rescued the impairment of the cells to extend lamellipodia into the open wound area (Fig. 4C). These results indicate that RhoG is involved in lamellipodia extension during cell migration.
To determine Rac1 activity in migrating control RNAi and RhoG RNAi HeLa cells, multiple scratch wounds were made in confluent cells, and the level of active GTP-bound Rac1 in the cells was measured using a GST-fused CRIB domain of Pak to precipitate active Rac1 from the cell lysates (Benard et al., 1999). When control RNAi HeLa cells were induced to migrate, Rac1 activity was significantly elevated in the cells. By contrast, we could not detect the migration-induced activation of Rac1 in RhoG RNAi HeLa cells (Fig. 4D). However, we observed no significant difference in basal Rac1 activities between control RNAi and RhoG RNAi HeLa cells. Thus, RhoG is required for the activation of Rac1 in response to migration stimuli.
The regulation of cell migration by RhoG requires ELMO- and Dock180-dependent activation of Rac
To obtain insights into the molecular mechanisms underlying the RhoG-mediated regulation of cell migration, we performed a Transwell migration assay with HEK 293T cells, because the motility of the cells was very low but was dramatically increased by expression of active RhoG in this Transwell assay (Fig. 5A). At 24 hours after transfection, cells were plated in the upper chamber of the filters that had been coated with fibronectin on the underside and allowed to migrate for 6 hours. To score the number of transfected cells that had moved to the underside of the filter, cells were cotransfected with GFP. Few GFP-positive cells were observed on the underside of the filter at 6 hours after being plated in the upper chamber. However, coexpression of GFP with constitutively active RhoG (RhoG-V12) dramatically increased the number of migrated cells on the underside of the filter. By contrast, cotransfection of GFP with RhoG-A37, a mutant of RhoG that had no ability to bind to its effector ELMO (Katoh and Negishi, 2003), did not promote HEK 293T cell migration (Fig. 5A). There were no significant differences in the ability of GFP-positive cells to adhere to the filters among different transfection conditions (data not shown). We have recently reported that RhoG activates Rac1 and induces membrane ruffling through the ELMO-Dock180 complex (Katoh and Negishi, 2003). To evaluate whether the promotion of cell migration by RhoG was mediated through ELMO- and Dock180-dependent Rac1 activation, we used mutants of ELMO and Dock180. ELMO-T618 is a mutant of ELMO lacking the C-terminal Dock180-binding region; Dock180-ISP has three amino acid substitutions in the Rac1 activation domain of Dock180, and does not interact with and activate Rac1, both of which act as dominant-negative forms of the proteins (Katoh and Negishi, 2003). Coexpression of ELMO-T618 or Dock180-ISP significantly suppressed the promotion of cell migration by RhoG-V12, whereas wild-type ELMO and Dock180 had no effect (Fig. 5B). These results suggest that activation of RhoG promotes cell migration through ELMO- and Dock180-dependent activation of Rac1.
Using RhoG RNAi HeLa cells, we also examined whether the reduced motility by knockdown of RhoG could be rescued by the overexpression of ELMO and Dock180. Although expression of GFP alone in RhoG RNAi HeLa cells had no effect on the cell motility, coexpression of GFP with ELMO and Dock180 rescued the reduced cell motility (Fig. 3D). This result further supports the idea that RhoG regulates cell migration through ELMO and Dock180.
The interaction of Dock180 with Crk is dispensable for the activation of Rac1 and regulation of cell migration by RhoG
Dock180 was identified as a protein that binds to the first Src-homology 3 (SH3) domain of CrkII (Hasegawa et al., 1996). CrkII has been shown to function upstream of Rac1 activation in many cellular events including cell migration (Feller, 2001). However, the relationship between RhoG and CrkII has not yet been understood. To examine whether the interaction of Dock180 with CrkII could affect the downstream signaling of RhoG, we constructed Dock180 mutants (Fig. 6A). Dock180-T1657 is a mutant lacking the C-terminal proline-rich region necessary for binding to Crk, and a pull-down assay with purified GST-fused CrkII confirmed that GST-CrkII bound to Flag-tagged wild-type Dock180 but not to Dock180-T1657 (Fig. 6B). Because active RhoG forms a ternary complex with ELMO and Dock180 in cells (Katoh and Negishi, 2003), we tested whether active RhoG also forms a complex with ELMO and Dock180-T1657 by a co-immunoprecipitation study in HEK 293T cells cotransfected with Myc-RhoG-V12, HA-ELMO and Flag-Dock180-T1657. Dock180-T1657 and wild-type Dock180 was co-immunoprecipitated with RhoG-V12 when the two proteins were co-expressed with ELMO (Fig. 6C). We next examined the subcellular localization of Dock180-T1657 in the presence and absence of RhoG. We prepared crude membrane and cytosolic fractions from cellular homogenates of HEK 293T cells expressing HA-ELMO and Flag-Dock180 and analyzed their distributions by immunoblot analysis. When cells were transfected with wild-type Dock180 or Dock180-T1657 together with ELMO, they were observed largely in the cytosolic fraction. By contrast, both wild-type Dock180 and Dock180-T1657 were detected in the membrane fraction when they were co-expressed with ELMO and RhoG-V12; however, RhoG-V12A37, a mutant of RhoG-V12 that had no ability to bind to ELMO, had no effect (Fig. 6D). These results indicate that the interaction of Dock180 with CrkII is not required either for the formation of the ternary complex of Dock180, ELMO and active RhoG, or the RhoG-induced translocation of Dock180 to the membrane.
To evaluate whether the interaction of Dock180 with CrkII affects the activity of Dock180 towards Rac1, HEK 293T cells were transfected with wild-type Dock180 or Dock180-T1657 and the Rac1 activity in the cells was measured using a GST-fused CRIB domain of Pak. Expression of wild-type Dock180 induced an increase in Rac1 activity in cells, and expression of Dock180-T1657 also increased Rac1 activity at a level comparable with that induced by expression of wild-type Dock180. In addition, co-expression of RhoG-V12 and ELMO enhanced both the Dock180- and Dock180-T1657-induced activation of Rac1 at similar levels (Fig. 6E). These results indicate that the RhoG-induced stimulation of Dock180 activity towards Rac1 does not require the Crk binding to Dock180.
Finally, we performed a Transwell migration assay to examine the involvement of the Dock180-Crk interaction in the promotion of cell migration by RhoG using two mutants of Dock180: Dock180-ISP and Dock180ΔN-ISP. Dock180-ISP binds to both ELMO and Crk and could suppress RhoG-ELMO-dependent and Crk-dependent activation of Dock180; by contrast, Dock180ΔN-ISP binds to Crk but not to ELMO, and could suppress only Crk-dependent Dock180 activation (Fig. 6A). Indeed, overexpression of CrkII in HEK 293T cells increased the number of migrated cells to the underside of the filter, and Dock180ΔN-ISP as well as Dock180-ISP suppressed the CrkII-promoted cell migration (Fig. 7B). By contrast, although RhoG-V12-induced promotion of cell migration was inhibited by co-expression of Dock180-ISP, Dock180ΔN-ISP had little effect on the increased HEK 293T cell motility by RhoG-V12 (Fig. 7A). Expression of Dock180-ISP or Dock180ΔN-ISP alone had little effect on the motility of HEK 293T cells (data not shown). These results indicate that the interaction of Dock180 with Crk is dispensable for the promotion of cell migration by RhoG.
Migrating cells extend protrusions with broad lamellipodia at the front, which are driven by actin polymerization and are stabilized by adhering to the extracellular matrix. Members of the Rho family of small GTPases are key regulators of the actin cytoskeleton in diverse cellular functions including cell migration. In particular, activation of Rac is required for lamellipodial extension at the leading edge. Here, we have shown that inhibition of RhoG by RNAi-induced knockdown of RhoG reduces cell motility, whereas its activation promotes cell migration. In addition, activation of Rac1 and formation of lamellipodia at the leading edge is attenuated in migrating RhoG-knockdown cells. Thus, RhoG is a key upstream regulator for Rac during cell migration. In the present study, we also have shown that RhoG is required for the ability of HeLa cells to spread rapidly and to induce extensive membrane ruffling at the cell periphery in response to fibronectin.
Dock180 binds to ELMO, and the ELMO-Dock180 complex serves as a functional GEF for Rac in intact cells (Brugnera et al., 2002; Lu et al., 2004). The ELMO-Dock180 complex is implicated in many cellular functions including cell migration by activating Rac (Gumienny et al., 2001; Grimsley et al., 2004). ELMO and Dock180 localize mainly in the cytoplasm, and the recruitment of the ELMO-Dock180 complex to the plasma membrane is an important step for the complex-mediated activation of Rac and reorganization of the actin cytoskeleton (Hasegawa et al., 1996; Kiyokawa et al., 1998; Gumienny et al., 2001). Our previous study demonstrated that activation of RhoG induced the translocation of the ELMO-Dock180 complex to the plasma membrane and activation of Rac1, leading to formation of lamellipodia at the cell periphery (Katoh and Negishi, 2003). In the present study, we show that expression of constitutively active RhoG promotes cell migration through ELMO and Dock180, and that knockdown of RhoG suppresses the lamellipodial extensions during cell migration. Thus, activation of RhoG in response to migration stimuli might recruit the ELMO-Dock180 complex to the plasma membrane at the leading edge and activate Rac, thereby leading to formation of lamellipodial protrusions at the leading edge. In C. elegans, the ELMO homolog CED-12 and the Dock180 homolog CED-5 function as crucial upstream regulators of Rac in phagocytosis of apoptotic cells and cell migration (Gumienny et al., 2001; Wu et al., 2001). Recently, MIG-2 has been shown to bind to CED-12 and regulate engulfment of apoptotic cell corpses upstream of CED-12 and CED-5, indicating that MIG-2 is the homolog of mammalian RhoG (deBakker et al., 2004). Indeed, RhoG regulates the Rac-dependent phagocytosis of apoptotic cells (deBakker et al., 2004). Thus, the small GTPase RhoG is evolutionarily conserved from worms to humans and is involved in the regulation of engulfment upstream of ELMO, Dock180 and Rac. In addition to the involvement of cell corpse engulfment, MIG-2 also plays a crucial role in cell migration during development of C. elegans (Zipkin et al., 1997; Lundquist et al., 2001; Kishore and Sundaram, 2002; Wu et al., 2002). Taken together, these data suggest that the RhoG-ELMO-Dock180 pathway is a conserved upstream signaling pathway for Rac that plays a crucial role in cell migration.
In addition to the interaction with ELMO, Dock180 binds to the SH3 domain of adaptor proteins, such as CrkII and CrkL, through the C-terminal proline-rich region (Hasegawa et al., 1996; Feller, 2001). Integrin stimulation induces the formation of the p130Cas-Crk complex, and the complex regulates several cellular processes, such as phagocytosis of apoptotic cells, cell spreading and cell movement, by activating Rac (Vuori et al., 1996; Klemke et al., 1998; Albert et al., 2000). Co-expression of p130Cas and CrkII with Dock180 enhanced the Dock180-mediated activation of Rac1 (Kiyokawa et al., 1998). For the internalization of apoptotic cells, integrin recruits the p130Cas-CrkII-Dock180 complex, which in turn triggers Rac1 activation and phagosome formation (Albert et al., 2000). Thus, the p130Cas-Crk-dependent pathway is an important upstream signaling pathway for the Dock180-mediated activation of Rac and cellular morphological events. In the present study, the C-terminal Crk-binding region of Dock180 was not essential for the RhoG-induced activation of Rac1 and promotion of cell migration, suggesting that RhoG controls cell migration through ELMO and Dock180 independently of Crk. This finding is consistent with a previous report showing that the N-terminal region of ELMO1, corresponding to the RhoG-binding domain, but not the C-terminal Crk-binding region of Dock180, is crucial for the cell migration mediated by the ELMO1-Dock180 complex (Grimsley et al., 2004). Thus, these two p130Cas-Crk and RhoG-ELMO pathways converge on Dock180 to activate Rac and regulate cell movement. However, we could not exclude the possibility that CrkII is required for the regulation of cell migration by RhoG, because Dock180ΔN-ISP, which is a large protein with multiple regions, might function as a dominant-negative protein towards CrkII for other reasons. The relationship between the p130Cas-Crk-Dock180 and the RhoG-ELMO-Dock180 pathways in the regulation of cell migration remains unclear, but a recent study reported that the C-terminal second SH3 domain of CrkII regulates the interaction between RhoG and ELMO (Akakura et al., 2005). Further studies are required for elucidating the precise mechanisms for the regulation of cell migration mediated by Dock180-dependent activation of Rac. By contrast, Dock2, a hematopoietic-cell-specific Dock180 family protein, can specifically activate Rac and functions as a central regulator in lymphocyte migration (Nishihara et al., 1999; Fukui et al., 2001). Although Dock2 has no obvious Crk-binding motifs, Dock2 can bind to and cooperate with ELMO1 to promote Rac-dependent cell migration (Sanui et al., 2003; Grimsley et al., 2004). Therefore, RhoG might regulate the activity of Dock2 towards Rac and lymphocyte migration.
In the present study, we have shown that RhoG plays an important role in the regulation of cell migration through Elmo- and Dock180-dependent activation of Rac1. However, we cannot rule out the possibility of the involvement of other signaling pathways in the promotion of cell migration by RhoG, because other RhoG-mediated morphological events such as promotion of neurite outgrowth involve activation of Cdc42 as well as Rac1 (Gauthier-Rouvière et al., 1998; Blangy et al., 2000; Katoh et al., 2000; Vigorito et al., 2003). However, the mechanisms of regulation of RhoG activity during cell migration are not yet well understood. Trio, a member of the Dbl family of proteins, contains two DH-PH tandem GEF domains, and the N-terminal GEF domain of Trio catalyzes nucleotide exchange on RhoG in vitro and activates RhoG in intact cells (Blangy et al., 2000; Estrach et al., 2002; Skowronek et al., 2004; deBakker et al., 2004). The N-terminal GEF domain of Trio promotes cell migration, and genetic studies using C. elegans have also shown that the Trio homolog UNC-73 plays a crucial role in cell migration (Steven et al., 1998; Seipel et al., 1999; Kishore and Sundaram, 2002). Thus, Trio might be an upstream activator of RhoG during cell migration. In addition to Trio, SGEF (for `SH3-containing GEF'), another Dbl-family protein, has recently been identified as a RhoG-specific GEF, and overexpression of SGEF in fibroblasts forms RhoG-dependent membrane ruffling (Ellerbroek et al., 2004). Additional studies with overexpression and knockdown of Trio and SGEF will be required to determine the regulatory mechanisms of RhoG activity during cell spreading and migration.
Materials and Methods
Plasmids and antibodies
The shRNA for human RhoG (pSilencer-RhoGa) was designed to target 19 nucleotides of the human RhoG transcript (nucleotides 78-96, 5′-CGCTTTCCCCAAAGAGTAC-3′), and was expressed using an shRNA expression vector, pSilencer (Ambion). The shRNA targeting another region of the human RhoG transcript (nucleotides 534-552, 5′-CCCCACGCCGATCAAGCGT-3′), which had no effect on the RhoG expression, was used as a control shRNA (pSilencer-RhoGb). The expression plasmid encoding Flag-tagged Dock180 (pCXN2) was a gift from M. Matsuda (Osaka University, Osaka, Japan). The coding sequence for mouse RhoG was obtained by reverse transcription-polymerase chain reaction (RT-PCR) from embryonic mouse brain and subcloned into the pEGFP vector (Clontech). The coding sequence for human CrkII was obtained by RT-PCR from HeLa cells and subcloned into pEF-BOS or pGEX-4T-2 (Amersham Biosciences). The N-terminal region of ELMO (amino acids 1-362) was subcloned into the pEGFP vector. Mutants of Dock180 were generated by PCR-mediated mutagenesis and subcloned into the pCXN2-Flag vector. The nucleotide sequences of all constructs were confirmed after construction using the ABI Prism 310 Genetic Analyzer. Plasmids expressing Myc-tagged wild-type RhoG, RhoG-V12, RhoG-A37, RhoG-V12A37, wild-type Rac1, HA-tagged ELMO, ELMO-T618 and Flag-tagged Dock180-ISP were generated as described previously (Katoh et al., 2000; Katoh and Negishi, 2003).
We used the following antibodies: mouse mAbs against Myc, GFP and Dock180 (Santa Cruz Biotechnology): a rat mAb against HA (Roche): a mouse mAb against Flag (Sigma): mouse mAbs against Rac1 and Crk (Transduction Laboratories): secondary antibodies conjugated to horseradish peroxidase (DAKO); and secondary antibodies conjugated to Alexa Fluor 488 and Alexa Fluor 594 (Molecular Probes). F-actin was visualized with Alexa Fluor 594-conjugated phalloidin (Molecular Probes).
HeLa and HEK 293T cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) containing 10% fetal bovine serum, 4 mM glutamine, 100 units/ml penicillin and 0.2 mg/ml streptomycin under humidified conditions in 95% air and 5% CO2 at 37°C. HeLa and HEK 293T cells were transfected with indicated expression vectors using Lipofectamine 2000 and Lipofectamine Plus (Invitrogen), respectively, according to the manufacturer's instructions. HeLa cells stably expressing shRNAs were obtained by selection with 300 μg/ml hygromycin B.
Spreading and migration assay
For cell spreading assays, cells were harvested with phosphate-buffered saline (PBS)/EDTA, washed with serum-free DMEM, and replated on 10 μg/ml fibronectin-coated 35 mm dishes in serum-free medium. After 15 minutes, they were fixed with 4% paraformaldehyde. Images were captured using a Nikon Eclipse TE300 microscope and a Nikon Plan Fluor 20×0.45 objective equipped with a digital camera (DS-L1 and DS-5M; Nikon). The cell size was determined from digital images of nine randomly selected fields using Image-Pro Plus image analysis software (Media Cybernetics).
For Transwell migration assays, harvested cells (2×104 cells) were replated onto the upper chamber of a Transwell filter with 8 μm pores (Costar) coated with 10 μg/ml fibronectin, and the chamber was placed in serum-free DMEM. After 4 or 6 hours, cells were fixed with 4% paraformaldehyde in PBS. Non-migrated cells on the upper side of the filter were removed with a cotton swab, and cells on the underside of the filter were stained with 0.4% crystal violet in 10% ethanol. In parallel, cells were also separately plated to fibronectin-coated plates without Transwell filters for estimating the total number of attached cells. Images were captured using a Nikon Eclipse TE300 microscope and a Nikon Plan Fluor 10×0.30 objective. Relative cell migration was determined by the number of the migrated cells normalized to the total number of the cells adhering to fibronectin. For each experiment, the number of cells in nine random fields on the underside of the filter was counted, and three independent filters were analyzed.
For wound-healing assays, 1×105 cells were seeded on glass coverslips and cultured for 2 days. They were then scratched with micropipette tips, and images were captured at 0 and 9 hours after wounding using a Nikon Eclipse TE300 microscope and a Nikon Plan Fluor 4×0.13 objective.
Cells on coverslips were fixed with 4% paraformaldehyde in PBS for 15 minutes and washed with PBS five times. Cells were permeabilized with 0.2% Triton X-100 in PBS for 10 minutes and incubated with 10% fetal bovine serum in PBS for 30 minutes to block nonspecific antibody binding. Then cells were incubated with Alexa Fluor 594-conjugated phalloidin in PBS for 1 hour, washed with PBS for 30 minutes, and mounted in 90% glycerol containing 0.1% p-phenylenediamine dihydrochloride in PBS. Images were captured using a Nikon Eclipse E800 microscope and a Nikon 40×0.75 objective equipped with a digital camera (Leica DC350F).
Pull-down assay and immunoprecipitation
Purification of recombinant glutathione S-transferase (GST)-fused CrkII from Escherichia coli was performed as described previously (Katoh et al., 2002). For pull-down assay, transfected cells were rinsed once with PBS and lyzed with the ice-cold cell lysis buffer: 50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 2 mM MgCl2, 1 mM dithiothreitol (DTT), 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 μg/ml of aprotinin and 10 μg/ml of leupeptin. Cell lysates were then centrifuged for 10 minutes at 16,000 g at 4°C. The supernatants were incubated for 10 minutes at 4°C with 5 μg of GST-fused CrkII and subsequently incubated with glutathione-Sepharose beads for 1 hour at 4°C. After the beads were washed with the ice-cold cell lysis buffer, the bound proteins were eluted in Laemmli sample buffer and analyzed by SDS-PAGE and immunoblotting. For immunoprecipitation, HEK 293T cells cotransfected with Myc-tagged RhoG, HA-tagged ELMO and Flag-tagged Dock180 were lyzed with ice-cold cell lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 2 mM MgCl2, 1 mM DTT, 1% Triton X-100, 1 mM PMSF, 10 μg/ml of aprotinin and 10 μg/ml of leupeptin). After centrifugation, the supernatants were incubated with anti-Myc antibody for 1 hour and then with protein A Sepharose (Amersham Biosciences) for 1 hour. The beads were washed with the lysis buffer, and bound proteins were analyzed by SDS-PAGE and immunoblotting.
Proteins were separated by 12.5% SDS-PAGE, and were electrophoretically transferred onto a polyvinylidene difluoride membrane (Millipore Corporation). The membrane was blocked with 3% low fat milk in Tris-buffered saline, and then incubated with primary antibodies. The primary antibodies were detected with horseradish peroxidase-conjugated secondary antibodies and a chemiluminescence detection kit (Chemi-Lumi One; Nacalai Tesque).
Separation of membrane and cytosolic fractions
Separation of membrane and cytosolic fractions in transfected HEK 293T cells was performed according to the method described previously (Kobayashi et al., 2001). Transfected HEK 293T cells were suspended in buffer A (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF). After rapid freezing in liquid nitrogen and thawing in a water bath, cells were centrifuged at 16,000 g for 10 minutes at 4°C. The supernatant was removed and used as a cytosolic fraction. After washing the pellet with buffer A, it was lyzed with buffer A containing 1% Triton X-100 and centrifuged at 10,000 g for 10 minutes at 4°C. The supernatant was removed and used as a membrane fraction. Both cytosolic and membrane fractions were separated by SDS-PAGE and analyzed by immunoblotting.
Measurement of Rac1 activity
Measurement of Rac1 activity was performed according to the modified method of Benard et al. (Benard et al., 1999). The CRIB domain of αPak (amino acids 70-150) was expressed in Escherichia coli as a fusion protein with GST, purified on glutathione-Sepharose beads, and isolated from the beads with 16 mM reduced glutathione. The purified proteins were dialyzed with 10 mM Tris-HCl, pH 7.5, 2 mM MgCl2 and 0.1 mM DTT, and stored at -80°C. Protein concentration was determined by comparing with bovine serum albumin standards after SDS-PAGE and by staining with Coomassie Brilliant Blue. To determine Rac1 activity in migrating cells, 20 scratch wounds with micropipette tips were made randomly in confluent HeLa cells in 60 mm culture dishes. For measurement of RhoG-V12-induced Rac1 activation, HEK 293T cells (1×106 cells per 60 mm culture dish) were transfected with Myc-tagged RhoG-V12 and maintained in serum-free DMEM for 16 hours. Cells were then lyzed for 5 minutes with the ice-cold cell lysis buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 2 mM MgCl2, 1% Nonidet P-40, 10% glycerol, 1 mM DTT, 1 mM PMSF, 1 μg/ml aprotinin and 1 μg/ml leupeptin) containing 4 μg of GST-CRIB. Cell lysates were then centrifuged for 5 minutes at 10,000 g at 4°C, and the supernatant was incubated with glutathione-Sepharose beads for 30 minutes at 4°C. The beads were washed with lysis buffer, and bound proteins were analyzed by SDS-PAGE and immunoblotting. Densitometry analysis was performed with NIH Image software, and relative Rac1 activity was determined by the amount of Rac1 bound to GST-CRIB normalized to the amount of Rac1 in cell lysates.
Total RNA was prepared from HeLa cells using an Isogen RNA isolation kit (Nippon-gene). An equal amount of total RNA (1.25 μg) isolated from parental, control RNAi, or RhoG RNAi HeLa cells was reverse transcribed using M-MLV reverse transcriptase (Invitrogen). The PCR was performed using the following primers: RhoG forward (5′-CCTGAACCTGTGGGACACTGCGG-3′) and RhoG reverse (5′-CCAGGGTCACAAGAGGATGCAG-3′), designed to amplify a 427 bp region; ELMO2 forward (5′-ATGCCCCTGGGAGTGGGACC-3′) and ELMO2 reverse (5′-GGCGGTTCCCAATCTTTCGG-3′), designed to amplify a 690 bp region. The following PCR conditions were used: 94°C for 1 minute; 18 cycles of 94°C for 48 seconds, 60°C for 42 seconds and 72°C for 1 minute; and, finally, 72°C for 7 minutes.
This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan, and a grant from the Uehara Memorial Foundation.
- Accepted September 21, 2005.
- © The Company of Biologists Limited 2006