Journal of Cell Science partnership with Dryad

Journal of Cell Science makes data accessibility easy with Dryad

Summary

Dock180, a member of the CDM family of proteins, plays roles in biological processes such as phagocytosis and motility through its association with the signalling adaptor protein Crk. Recently, the complex formation between Dock180 and Elmo1 was reported to function as a bipartite guanine nucleotide exchange factor for Rac. In this study, we demonstrated that the amount of Dock180 increased when Elmo1 was co-expressed. Dock180 was found to be ubiquitylated and Dock180 protein levels could be augmented by treatment with proteasome inhibitor. The ubiquitylation of Dock180 was enhanced by epidermal growth factor (EGF), Crk and adhesion-dependent signals. Furthermore, Elmo1 inhibited ubiquitylation of Dock180, resulting in the increase in Dock180 levels. The Elmo1 mutant Δ531, which encompasses amino acids required for Dock180 binding, preserved the inhibitory effects on ubiquitylation of Dock180. Upon EGF stimulation, both Dock180 and ubiquitin were demonstrated to translocate to the cell periphery by immunofluorescence, and we found ubiquitylation of Dock180 and its inhibition by Elmo1 to occur in cellular membrane fractions by in vivo ubiquitylation assay. These data suggest that Dock180 is ubiquitylated on the plasma membrane, and also that Elmo1 functions as an inhibitor of ubiquitylation of Dock180. Therefore, an ubiquitin-proteasome-dependent protein degradation mechanism might contribute to the local activation of Rac on the plasma membrane.

Introduction

Dock180 was originally identified as a Src-homology 3 (SH3)-domain-binding protein of the signalling adaptor protein Crk (Hasegawa et al., 1996; Tanaka et al., 1993). Subsequently, homologues of Dock180 were identified in Drosophila (Myoblast city) and Caenorhabditis elegans (CED-5), and these were designated as the CDM family of proteins (Cote and Vuori, 2002). CDM proteins are evolutionarily conserved and have been implicated in various biological responses, including cell migration (Cheresh et al., 1999; Kiyokawa et al., 1998a) and phagocytosis (Albert et al., 2000), in mammals, Drosophila (Duchek et al., 2001; Nolan et al., 1998) and C. elegans (Wang et al., 2003; Wu and Horvitz, 1998).

The major SH2 domain targets of Crk are components of focal adhesion p130Cas and paxillin (Feller, 2001), implicating Crk in cytoskeletal reorganisation. Downstream of Crk, Dock180 functions as an activator for Rac and regulates cell motility, filopodia formation and phagocytosis, particularly through stimulation of β1 and αvβ5 integrins (Albert et al., 2000; Gustavsson et al., 2004).

Elmo1 (for `engulfment cell motility 1') was initially identified as a mammalian homologue of C. elegans Ced-12, which is required for cell migration and engulfment of dying cells (Gumienny et al., 2001). Elmo1 functionally cooperates with Crk and Dock180, and promotes phagocytosis and morphological changes (Grimsley et al., 2004; Gumienny et al., 2001; Gustavsson et al., 2004). In addition, Elmo1 binds directly to Dock180 (Gumienny et al., 2001) and functions as an unconventional bipartite guanine nucleotide exchange factor (GEF) for Rac (Brugnera et al., 2002). Recently, it was demonstrated that the small GTPase RhoG interacts directly with Elmo1, and that a tri-molecular complex comprised of RhoG, Elmo1 and Dock180 activates Rac1, which then results in integrin-mediated cell spreading, phagocytosis and nerve growth factor (NGF)-induced neurite outgrowth (deBakker et al., 2004; Katoh and Negishi, 2003).

It is well known that ubiquitylation plays a pivotal role in physiological cellular responses, including growth-factor-mediated signal transduction for cell proliferation and motility. The epidermal growth factor (EGF) receptor is ubiquitylated by a RING-finger-type ubiquitin ligase, Cbl (Galcheva-Gargova et al., 1995; Joazeiro et al., 1999). The event is important for regulation of endocytosis of the receptor (Mosesson et al., 2003; Soubeyran et al., 2002) and for lysosomal degradation (Longva et al., 2002). Furthermore, some of the GEFs, namely Vav and CNrasGEF (for `cyclic nucleotide ras GEF'), are ubiquitylated by Cbl and Nedd4, respectively (Miura-Shimura et al., 2003; Pham and Rotin, 2001). Finally, it has been suggested that ubiquitin-dependent protein degradation regulates actin cytoskeletal reorganisation.

In this study, we present the new findings that Dock180 is ubiquitylated mainly on the plasma membrane; that this is enhanced by EGF, Crk and adhesion-dependent signals; and that its amounts are regulated by an ubiquitin-proteasome-dependent protein degradation mechanism. Furthermore, we demonstrated that endogenous Elmo1 could regulate the amount of Dock180 protein through the inhibition of ubiquitylation of Dock180 by Elmo1.

Results

Elmo1 has a stabilising effect on Dock180 protein

To analyse the functions of Elmo1 in the regulation of Dock180, we overexpressed Dock180 in the presence or absence of Elmo1. In the presence of Elmo1, the amount of Dock180 was increased (Fig. 1A, left panel). In addition, the amount of the active form of Rac also increased in the presence of Elmo1 (Fig. 1A, left panel). By contrast, Elmo1 did not alter the levels of other co-expressed proteins including Crk (Fig. 1A, right panel).

These data suggested that Elmo1 regulates the stability of Dock180. To test the idea, we performed a pulse-chase analysis. The labelled form of Dock180 decreased by 79% within 24 hours in the absence of Elmo1. However, in the presence of Elmo1, the levels of Dock180 decreased by 68% (Fig. 1B). Three sets of pulse-chase analyses were performed and a significant difference between the levels of Dock180 with and without Elmo1 was confirmed as described, and shown in a bar graph with the standard error indicated (Fig. 1C). To exclude the possibility that Elmo1 activates the transcription of Dock180, we also performed RT-PCR analysis and found the unchanged mRNA levels of Dock180 with Elmo1 (Fig. 1D). In this PCR condition, Dock180 bands were detectable in a dose-dependent manner for templates (Fig. 1D). Primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as a control for unchanged levels of mRNA (Fig. 1D).

Furthermore, we investigated the expression levels of Dock180 when protein expression of Elmo1 was suppressed by short interfering (si)RNA for Elmo1 in the fibrosarcoma cell line HT1080 to examine whether Elmo1 can inhibit or regulate the amounts of endogenous Dock180. Among three siRNAs for Elmo1, designated as Elmo1#1, #2 and #3, Elmo1#3 most-efficiently reduced the expression levels of Elmo1 in HT1080 cells; the assay was performed using a negative control and a scramble control against Elmo1#3, and Elmo1#3. Expression levels of Dock180 were reduced by Elmo1#3, compared with both negative control and scramble control (Fig. 1E). Three sets of this siRNA assay were performed, and significant differences in the endogenous levels of Dock180 between the controls and Elmo1 siRNA were confirmed, as described by a bar graph with standard error (Fig. 1F).

To exclude the possibility that Elmo1 or siRNA for Elmo1 might affect the amount of mRNA levels of Dock180, we performed RT-PCR analysis and found that mRNA levels of endogenous Dock180 were almost equal; by contrast, those of Elmo1 were reduced by siRNA for Elmo1 (Fig. 1G). Moreover, we performed the same assay using siRNA for Elmo1 in HEK293T human embryonic kidney cells. We used Elmo1#2 because it showed the best efficiency for reduction of Elmo1 in HEK293T cells, and found that protein levels of Dock180 decreased in concert with reduction of those of Elmo1 by siRNA (Fig. 1H).

It is noteworthy that the endogenous levels of Dock180 were not altered even with the presence of force-expressed Elmo1 in HEK293T cells (data not shown). This was probably because the physiological levels of Elmo1 were sufficient for stabilisation of endogenous Dock180.

Ubiquitylation of Dock180 in vivo

We next investigated whether Dock180 is ubiquitylated and degraded by the proteasome, because the ubiquitin-proteasome pathway has been recognised as one of the major mechanisms for the regulation of cellular protein levels. By in vivo ubiquitylation assay, an ubiquitylated protein band of over 180 kDa was observed when HA-ubiquitin was co-expressed with Dock180, and this band was significantly enhanced by the treatment of proteasome inhibitor MG-132 (Fig. 2A). To exclude the possibility that we were detecting only the ubiquitylation of a contaminated protein of a size similar to that of Dock180, we performed a urea-reversal immunoprecipitation assay. In support of the idea that the band is specific to Dock180, we found that the anti-Dock180 antibody could precipitate a protein over 180 kDa that was also detectable with the anti-HA tag antibody (Fig. 2B).

We also examined the effect of MG-132 on endogenous levels of Dock180 in MCAS cells as compared with those that are expressing Dock180 at high levels. We treated cells with cyclohexamide, an inhibitor of protein synthesis, and then assayed Dock180 levels. The level of Dock180 was markedly decreased after treatment with cyclohexamide, an effect that can be rescued by treatment with MG-132 (Fig. 2C). In HEK293T cells, almost the same results were obtained (data not shown). These data suggest that the levels of endogenous Dock180 protein are regulated by an ubiquitin-proteasome-dependent protein degradation mechanism.

Elmo1 inhibits the ubiquitylation of Dock180

To examine whether the association of Elmo1 regulates ubiquitylation of Dock180, we examined the effect of wild-type and mutant forms of Elmo1 on Dock180. Two deletion forms of Elmo1 - T625, which does not bind to Dock180, and Δ531, which contains C-terminus binding sites for Dock180 (Fig. 3A,B) (Shimazaki et al., 2005) - were used in an in vivo ubiquitylation assay. In this experiment, we performed immunoprecipitation analysis 24 hours after transfection. After 24 hours, the amounts of Dock180 were still almost equal in each transfectant. The equal amounts of Dock180 enabled us to compare the ubiquitylation levels of Dock180. Both wild-type Elmo1 and the Δ531 mutant inhibited ubiquitylation of Dock180, whereas the T625 mutant form of Elmo1 could not (Fig. 3C). It should be noted that levels of the hyper-ubiquitylated form of Dock180 (≫500 kDa) seemed to decrease in the T625 mutant of Elmo1 (Fig. 3C). The pleckstrin-homology (PH) domain of Elmo1 might have some effect on the poly-ubiquitylation of Dock180, because Elmo1 was known to bind the Docker domain (DHR-2 domain) of Dock180 through the PH domain, which was partially contained in the T625 mutant (Lu et al., 2004).

We also detected ubiquitylated bands around 70 kDa with the anti-HA tag antibody when full-length Elmo1 was co-expressed; because the molecular weight of Elmo1 is around 70 kDa, these bands may be ubiquitylation bands of Elmo1 (Fig. 3C, upper panel, arrow). In fact, we found that Elmo1 was also ubiquitylated during an in vivo ubiquitylation assay using lysates from HEK293T cells expressing both Elmo1 and HA-ubiquitin (data not shown). Further study should be carried out to determine the physiological role of ubiquitylation of Elmo1, since overexpression of proteins is known to induce their ubiquitylation and thereby ensure their quality control.

Fig. 1.

Elmo1 increases the stability of Dock180 protein. (A) Left panels; HEK293T cells were transiently transfected with mammalian expression plasmids for Flag-Dock180 alone or Flag-Dock180 and myc-Elmo1. After 48 hours, cells were lysed and analysed by immunoblotting (IB) with anti-Flag tag, anti-myc tag and anti-actin antibodies. The active form of Rac was detected in a pull-down assay. Right panels: HEK293T cells were transfected with mammalian expression plasmids for Flag-Crk II alone or Flag-Crk II and myc-Elmo1, and cells were analysed in the same way as for the left panels. (B) For pulse-chase analysis for Dock180, HEK293T cells transiently expressing Dock180 alone (-) or Dock180 and Elmo1 (+) were labelled for 1 hour and then chased for the time indicated at the top of the panel. (C) The signal intensity of Dock180 was measured and shown as a bar graph with the average and standard errors of three independent experiments; **P<0.05. (D) mRNA levels of exogenous Dock180. RT-PCR analysis was performed using mRNA extracted from HEK293T cells transiently expressing Dock180 alone or Dock180 and Elmo1. GAPDH is a control for the amounts of PCR templates. T/F, transfection; RT, reverse transcriptase. (E) HT1080 cells were transfected with siRNAs for negative control, scramble control and Elmo1#3. After incubation for 96 hours, cell lysates were subjected to immunoblotting for detection of Dock180, Elmo1 and actin. (F) The signal intensity of Dock180, normalised by the intensity of actin, is shown as a bar graph with the average and standard errors of three independent experiments; **P<0.05. (G) mRNA levels of endogenous Dock180. RT-PCR analysis was performed using the same mRNA extracts as E. GAPDH is a control for the amount of PCR template. (H) HEK293T cells were transfected with negative control and Elmo1#2 and cells were analysed in the same manner as E.

To confirm that the ubiquitylated bands over 180 kDa observed in Fig. 3C are Dock180, we performed urea-reversal immunoprecipitation assay using the same cell lysates as those analysed in the experiment shown in Fig. 3C. We found that the ubiquitylated bands were suppressed to almost the same degree as seen in Fig. 3C by co-expression of full-length Elmo1 and the Δ531 mutant (Fig. 3D). In addition, ubiquitylated bands of around 70 kDa (Fig. 3C, lane 3) disappeared (Fig. 3D). Furthermore, to confirm that the inhibitory effect of Elmo1 for ubiquitylation of Dock180 is not non-specific, we examined an irrelevant protein [glutathione S-transferase (GST)] on Dock180 ubiquitylation. Unlike Elmo1, GST was found to have no effect on the levels of ubiquitylation of Dock180 (Fig. 3E).

Fig. 2.

Ubiquitylation of Dock180 in HEK293T cells. (A) Results of an in vivo ubiquitylation assay. HEK293T cells were transfected with the indicated plasmids, and after treatment of 10 μM MG-132 for 12 hours, cells were lysed and subjected to immunoprecipitation (IP) and immunoblotting (IB). MIG, normal mouse immunoglobulin; TCL, total cell lysate; Ub, ubiquitin. (B) For the urea-reversal immunoprecipitation, the initial precipitates from the anti-Dock180 mAb were treated with 8 M urea buffer for 1 minute and then immunoprecipitated again using the same antibody. (C) MCAS and HEK293T cells were treated with 20 μg/ml cyclohexamide and 10 μg/ml MG-132 for 24 hours as indicated at the top of the panel. Cell lysates were immunoblotted with anti-Dock180 antibody to detect endogenous Dock180.

We employed an alternative approach to confirm that Elmo1, when bound to Dock180, suppressed ubiquitylation. We performed an in vivo ubiquitylation assay using the Dock180Δ357 mutant (which has a 357 amino acid deletion at the N-terminus and does not bind Elmo1) (Fig. 3F) (Brugnera et al., 2002). We found that the Dock180Δ357 mutant was significantly ubiquitylated, much more so than wild-type Dock180, both in the absence and presence of Elmo1 (Fig. 3F). These results suggest that Elmo1 functions as an inhibitor of the ubiquitylation of Dock180 through a mechanism that is dependent on the interaction between Dock180 and Elmo1.

Ubiquitylation of Dock180 on the plasma membrane

Next, we investigated the subcellular localisation for ubiquitylation of Dock180. An in vivo ubiquitylation assay for Dock180 was performed using either the cytosolic or the membrane fraction of cell lysates. We found that Dock180 treated with the membrane fraction was highly ubiquitylated, an effect that was inhibited by Elmo1 (Fig. 4A). However, in the cytosolic fraction, Elmo1 did not alter the level of Dock180 ubiquitylation (Fig. 4A). In contrast to the inhibition of ubiquitylation of Dock180 by Elmo1, protein levels of Dock180 in the membrane fraction were lower when co-expressed with Elmo1 than those without Elmo1 (Fig. 4A). We examined the effect of Elmo1 on the amount of endogenous Dock180 in the membrane fraction in HEK293 cells. The amount of endogenous Dock180 in the membrane fraction was also found to be decreased by Elmo1 (data not shown).

For further analysis, we employed immunofluorescent microscopy and observed the subcellular localisation of ubiquitin and Dock180. Dock180 and HA-ubiquitin were co-expressed in HEK293T cells, and cells were stained with anti-Dock180 and anti-HA tag antibody. We found that, in quiescent cells, both Dock180 and ubiquitin were partially localised at the cell periphery, although most of them were localised mainly in the cytoplasm (Fig. 4B, upper panel, arrowheads). Upon EGF stimulation, both Dock180 and ubiquitin were translocated and colocalised at the edge of the ruffled membrane (Fig. 4B, upper panel, arrows). This EGF-induced colocalisation of Dock180 and ubiquitin were also examined in Cos-7 cells (Fig. 4B, lower panel). The RFP fusion forms of Dock180 and HA-ubiquitin were co-expressed in Cos-7 cells, and HA-ubiquitin was visualised by immunostaining with anti-HA tag antibody. In quiescent cells, most of the detectable Dock180 and ubiquitin were diffusely found in the cytoplasm (Fig. 4B, lower panel). However, upon EGF stimulation, both Dock180 and ubiquitin were translocated to the cell periphery (Fig. 4B, lower panel, arrowheads).

Fig. 3.

Elmo1 suppresses ubiquitylation of Dock180. (A) A schematic representation of wild-type and mutant forms of Elmo1 used in the study. DUF609, domain of unknown function; FL, full length; LZ, leucine zipper; PH, pleckstrin homology; PxxP, polyproline-rich motif. (B) Association between Dock180 and Elmo1 in HEK293T cells. Cells were transiently transfected with the indicated plasmids, and lysates were immunoprecipitated with anti-myc tag antibody, and then probed with anti-myc tag and anti-Flag tag antibodies. TCL, total cell lysate. (C) Ubiquitylation of Dock180 in HEK293T cells in the presence of Elmo1 and its mutants. 24 hours after transfection with the indicated plasmids, cells were lysed and subjected to in vivo ubiquitylation assay. (D) The urea-reversal immunoprecipitation was performed using the same lysates as C. The IP ratio represents the relative signal intensity of immunoprecipitated Dock180 in lane 1 as 1.0 (bottom of the panel). (E) The ubiquitylation of Dock180 in the absence or presence of Elmo1 and GST. The immunoprecipitation was performed using anti-Flag tag antibody. (F) The ubiquitylation of Dock180 and its mutant. Lysates from HEK293T cells expressing the indicated plasmids were immunoprecipitated with anti-Dock180 antibody at in vivo ubiquitylation assay. The IP ratio represents the relative signal intensity of immunoprecipitated Dock180 in lane 1 as 1.0.

Furthermore, we confirmed that ubiquitylation of Dock180 in the membrane fraction was enhanced by EGF stimulation, and found that Elmo1 could inhibit this enhancement (Fig. 4C). The enhancement of the ubiquitylation of Dock180 in whole cell lysates was not detected (data not shown). These results indicate that ubiquitylation of Dock180 occurs mainly on the plasma membrane and Elmo1 inhibits this ubiquitylation.

Fig. 4.

Ubiquitylation of Dock180 on the plasma membrane. (A) Ubiquitylation of Dock180 in the membrane fraction. Cell lysates of HEK293T cells expressing the indicated plasmids were separated into cytosolic and membrane fractions. The samples were subjected to an in vivo ubiquitylation assay. The expression levels of transfected proteins were analysed using cell lysates of the cytosol and membrane fractions. Anti-E-cadherin antibody was used as a marker for the membrane fraction. CL, cell lysate. (B) Localisation of Dock180 and ubiquitin in cells. HEK293T cells (upper panels) and Cos-7 cells (lower panels) were transiently transfected with Flag-Dock180 or Dock180-RFP and HA-ubiquitin expression plasmids. After incubation for 36 hours, cells were stained with anti-HA tag antibody and analysed by confocal microscopy. (C) Ubiquitylation of Dock180 in the membrane fraction under EGF stimulation. HEK293T cells expressing the indicated plasmids were stimulated by EGF or not. Cell lysates were subjected to in vivo ubiquitylation assay.

Enhancement of the ubiquitylation of Dock180 by Crk

Next we investigated whether a main regulator of Dock180 such as Crk was involved in the ubiquitylation of Dock180 on the plasma membrane, and found its enhancement by both Crk I and Crk II in the membrane fraction (Fig. 5). By in vivo ubiquitylation assay using whole cell lysates, no significant Crk-dependent enhancement of ubiquitylation of Dock180 was observed (data not shown). In addition, we found that Dock180 levels in the membrane fraction were significantly increased by co-expression with either Crk I or Crk II (Fig. 5).

Fibronectin-stimulation-enhanced ubiquitylation of Dock180

As Crk and Dock180 are known to function downstream of integrins, especially α5β1 (Iwahara et al., 2004; Kiyokawa et al., 1998b; Sakai et al., 1994; Schaller and Parsons, 1995), we carried out a re-plating assay using a fibronectin-coated dish. An increase in ubiquitylation of Dock180 in the membrane fraction was observable 15 minutes after re-plating and persisted for at least 6 hours (Fig. 6A). The amount of Dock180 in the membrane fraction was not altered when cells were re-plated on a fibronectin-coated dish (Fig. 6A). In these cells, Elmo1 suppressed ubiquitylation of Dock180 even in the cells in suspension (Fig. 6B, lanes 1 and 3), and also inhibited augmentation of the re-plating-induced ubiquitylation of Dock180 (Fig. 6B, lanes 2 and 4).

Fig. 5.

The enhancement of ubiquitylation of Dock180 by Crk. An in vivo ubiquitylation assay was performed using lysates of the membrane fraction of HEK293T cells expressing the indicated plasmids (left panels). The IP ratio represents the relative signal intensity of immunoprecipitated Dock180 in lane 1 as 1.0 (bottom of the left panel). The expression levels of Dock180, Crk I and Crk II in the membrane fraction of cell lysates (membrane) and in total cell lysates (total) were determined by immunoblotting (right upper and middle panels). Anti-E-cadherin antibody was used as a marker for the membrane fraction (right lower panel).

Discussion

We have investigated the physiological function of Dock180, a member of the CDM family of proteins, and its interacting protein Elmo1, and found that Dock180 could be ubiquitylated and its amounts regulated by an ubiquitin-proteasome-dependent protein degradation mechanism. Initial observations, in which the amount of force-expressed Dock180 increased when Elmo1 was co-expressed in HEK293T cells, prompted us to investigate the involvement of ubiquitylation in regulation of the amounts of Dock180. To exclude the possibility of Elmo1-dependent transcriptional regulation of Dock180, we performed RT-PCR and pulse-chase analyses for the amounts of Dock180 with or without Elmo1 in HEK293T cells, and found that Elmo1 stabilised Dock180. In fact, we found that Elmo1 inhibited ubiquitylation of Dock180, and that the levels of endogenous Dock180 were decreased by siRNA for Elmo1. Since a rapid alteration in the level of Dock180 during the process of maturation of dendritic cells has been reported (Akakura et al., 2004), we plan to analyse the involvement of ubiquitylation of Dock180 in this process.

We also showed that the Δ531 mutant that binds to Dock180 preserves an inhibitory activity towards ubiquitylation of Dock180. Even so, the precise mechanism of Elmo1-dependent inhibition of ubiquitylation remains obscure. Elmo1 might block the physical association of E3 ubiquitin ligase to Dock180, or it might mask the ubiquitylation sites of Dock180. Alternatively, Elmo1 has been reported to induce a conformational change of Dock180, which could inhibit its ubiquitylation (Lu et al., 2004; Lu et al., 2005). To define the mechanism of Elmo1 action further, it will be necessary to identify the E3 ubiquitin ligase that is required for Dock180. It can at least be said that co-expression of Cbl with Dock180 does not change the levels of ubiquitylation of Dock180 in HEK293T cells (data not shown).

Fig. 6.

Enhancement of ubiquitylation of Dock180 in HEK293T cells re-plated on fibronectin-coated dishes. (A) HEK293T cells were transiently transfected with the indicated plasmids for Dock180 and HA-ubiquitin and re-plated on fibronectin-coated dishes for the indicated duration (top of the panel); lysates in the membrane fraction were subjected to in vivo ubiquitylation assay. Sus, suspended cells; FN, fibronectin. Ratio represents the relative signal intensity of Dock180 in the membrane fraction in lane 1 as 1.0 (bottom of the panel). (B) Inhibition of ubiquitylation of Dock180 by Elmo1. HEK293T cells were transiently transfected with the indicated plasmids, and the re-plating assay and in vivo ubiquitylation assay were performed as in A.

Recently, ubiquitylation of GEFs as a regulatory mechanism has been reported for other proteins. In the case of Vav, an activated Vav mutant (Y174F) was shown to be more sensitive to Cbl-dependent ubiquitylation, which suggests the regulation of protein degradation by tyrosine phosphorylation (Miura-Shimura et al., 2003). Furthermore, binding to Ras has been shown to be necessary for ubiquitylation of Ras-GRF2 (de Hoog et al., 2001). We demonstrated that Dock180 was mainly ubiquitylated on the plasma membrane and that this was enhanced by both EGF and Crk. Furthermore, in re-plating cells onto fibronectin-coated dishes, ubiquitylation of Dock180 on the plasma membrane was also enhanced. It should be noted that various levels of ubiquitylation of Dock180 were observed even when cells were re-plated on dishes coated with poly-L-lysine and collagen (data not shown). Thus, the enhancement of ubiquitylation of Dock180 observed with fibronectin-coated dishes is likely to be dependent on cell attachment rather than be specific for integrin stimulation. These data suggest that recruitment of Dock180 to the plasma membrane by EGF, Crk and adhesion-dependent signals might contribute to the ubiquitylation of Dock180.

Fig. 7.

Model of Dock180 regulation. (A,B) Following integrin stimulation by their ligands (including fibronectin), tyrosine kinases are activated and phosphorylate several components of focal adhesion such as p130Cas, and both Crk and Dock180 are recruited from the cytoplasm. (C) Recruited Dock180 is known to activate Rac, and we speculate that recruited Dock180 is ubiquitylated near the plasma membrane by an ubiquitin ligase. (D) Furthermore, ubiquitylated Dock180 might be removed from the complex comprising Crk and p130Cas, and is degraded by the proteasome.

It should be noted that despite the fact that ubiquitylation levels of Dock180 were elevated by stimulation with EGF or fibronection, or forced expression of Crk, the protein levels of Dock180 in the membrane fraction did not decrease. We speculate that the reason why Dock180 seemed not to be removed in the membrane fraction is that Dock180 is translocated from the cytoplasm by Crk. As the supplied amount of Dock180 may be more than that degraded by the proteasome, the apparent amount of Dock180 is not decreased after being ubiquitylated. As shown in Fig. 7, by integrin stimulation, Crk recruits Dock180 to the focal adhesion complex and Dock180 then activates Rac. In this process, several modifications of Dock180 including ubiquitylation might occur and a part of Dock180 is removed from the complex comprising Crk and p130Cas; such regulations might modulate local Rac activity. Thus, in the local areas of a cell, Dock180 ubiquitylation, which decreases the amount of Dock180, functions as one of the negative-feedback machineries for the Dock180-dependent activation of Rac, after integrin-dependent signals are turned on.

Such a spatio-temporal alteration of Dock180 levels on the plasma membrane could function in cell migration. We hope that future works both on the identification of ubiquitin ligase for Dock180 and on the analyses of ubiquitylation of Dock180 in living cells might further clarify the mechanism for ubiquitin-dependent Rac-GEF regulation.

Materials and Methods

Cells

HEK293T (human embryonic kidney 293 cells with SV40 T antigen), Cos-7, MCAS (human ovarian mucinous adenocarcinoma) and HT1080 (human fibrosarcoma) cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 100 μg/ml penicillin and streptomycin and 10% fetal bovine serum (Sigma), in 5% CO2 at 37°C.

Antibodies

The following antibodies were used: anti-influenza hemagglutinin (HA) tag mouse monoclonal antibody (mAb; clone 12CA5, Roche Diagnostics); anti-Dock180 mAb (clone H4, Santa Cruz Biotechnology); anti-Dock180 polyclonal antibody (clone H70, Santa Cruz Biotechnology); anti-FLAG tag mAb (clone M2, Sigma); anti-myc tag mAb (clone 9E10, gift from Hiroshi Ariga, Hokkaido University, Sapporo, Japan); anti-Rac mAb (clone 102, BD Transduction Laboratories); anti-actin mAb (clone C4, Chemicon International); anti-Elmo1 polyclonal antibody (clone ab2239, Abcam); and anti-E-cadherin mAb (clone 36, BD Transduction Laboratories). The anti-mouse immunoglobulin Ab conjugated with Alexa Fluor 488 and the anti-rabbit immunoglobulin Ab conjugated with Alexa Fluor 594 were purchased from Molecular Probes.

Expression plasmids

The pCXN2-Flag-Dock180, pCAGGS-myc-CrkI and -Crk II vectors were gifts from M. Matsuda (Osaka University, Osaka, Japan) and the pCGN-HA-Ubiquitin vector was constructed as described previously (Hatakeyama et al., 2001). The red fluorescent protein (RFP) fragment was amplified by PCR and subcloned into XhoI-NotI-digested pCXN2-Flag-Dock180; the resulting plasmid was named pCXN2-Dock180-RFP. pCXN2-Flag-Crk-II was also constructed (by H. Nishihara, Hokkaido University, Sapporo, Japan).

The pEBB-Flag-ELMO1 vector was kindly provided by K. Ravichandran (University of Virginia, VA). PCR fragments of full-length Elmo1, T625, and Δ531 were subcloned into a pMyc-CMV mammalian expression vector (Clontech Laboratories) that had been digested with XhoI and NotI; the resulting plasmid was named pCMV-myc-Elmo1 and contains the following changes: T625 [amino acids (aa.) 1 to 625], and Δ531 (aa. 532 to 727). pGEX-PAK2-RBD was described previously (Nishihara et al., 2002). All PCR fragments were verified by sequencing.

RT-PCR analysis

RT-PCR analysis was performed by the method described previously (Akakura et al., 2004; Shimazaki et al., 2005). Total RNA was prepared using the RNeasy Mini Kit (Qiagen) from HEK293T cells expressing the indicated plasmids and HT1080 cells. The forward primer 5′-TGGAGACAAAGTCACGGAGG-3′ and the reverse primer 5′-GATGAGAGGGAAGAGACAGAGG-3′ for Dock180 yielded a product of 219 bp. The forward primer 5′-CCGGATTGTGCTTGAGAACA-3′ and the reverse primer 5′-CTCACTAGGCAACTCGCCCA-3′ for Elmo1 yielded a product of 121 bp. The forward primer 5′-TTCGTCATGGGTGTGAACCA-3′ and the reverse primer 5′-GGTCATGAGTCCTTCCACGATAC-3′ for GAPDH yielded a product of 138 bp.

Transfection, immunoprecipitation and immunoblot analysis

HEK293T cells were transfected with plasmids using Lipofectamine 2000 (Invitrogen). After incubation for 24-48 hours, the cells were lysed with 1% Tx-100 lysis buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 5 mM EDTA, 10% (v/v) glycerol, 1% (v/v) Triton X-100, complete protease inhibitor cocktail (Roche Diagnostics) and 1 mM phenylmethylsulphonylfluoride (PMSF). Lysates were centrifuged at 20,000 g for 10 minutes at 4°C. The supernatants were incubated with the indicated antibodies and then with protein A beads (Protein A Sepharose 4 Fast Flow; Amersham Pharmacia Biotech) for 1 hour at 4°C. Precipitates or cell lysates were separated by SDS-PAGE, transferred onto PVDF filters (Immobilon), and incubated with primary antibodies. Positive signals were detected by enhanced chemiluminescence (ECL) western blotting reagents (Amersham Pharmacia Biotech) and quantified using a Lumino Image Analyzer (LAS1000; Fuji Film).

siRNA for Elmo1

HT1080 and HEK293T cells were transfected with siRNAs indicated below by Lipofectamine 2000. After incubation for 96 hours, cells were lysed and subjected to immunoblotting. Target sequences of siRNA are GGGUGGUCUCUUGCCAACCAUGAAU for Elmo1#1 (120-144 bp of Elmo1), GGCACUAUCCUUCGAUUAACCACAU for Elmo1#2 (216-240 bp of Elmo1), CCGAGAGGAUGAACCAGGAAGAUUU for Elmo1#3 (1561-1585 bp of Elmo1), and CCGGAAGGUAACCGAAGGAAGAUUU for scramble control against Elmo1#3, respectively. All siRNA duplex oligoribonucleotides including negative control (Stealth RNAi Negative CTL MED GC) were purchased from Invitrogen.

Cell fractionation

Cell fractions were prepared by the method described previously (Kobayashi et al., 2001) with some modifications. Briefly, cells were scraped and suspended in buffer A containing 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF and complete protease inhibitor cocktail. After freeze and thaw, cell suspensions were centrifugated at 1,000 g for 7 minutes and then at 20,000 g for 10 minutes. The supernatant was removed (cytosolic fraction). The pellet was washed once with buffer A, lysed with buffer B containing 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% (v/v) Triton X-100, 50 mM NaF, 1 mM Na3VO4, 1 mM PMSF and complete protease inhibitor cocktail, and centrifugated at 20,000 g for 10 minutes (membrane fraction). One-fifth volume of buffer A was used in the preparation of the membrane fraction.

Pull-down assay for Rac activity

HEK293T cells were lysed with a lysis buffer composed of 1% NP-40, 25 mM, HEPES (pH 7.4), 150 mM NaCl, 10% (v/v) glycerol, 1 mM EDTA, 10 mM MgCl2, 1 mM PMSF and complete protease inhibitor cocktail. Lysates were centrifuged at 14,000 g at 4°C for 1 minute. The supernatants were incubated with 10 μg of purified GST-PAK2-RBD and then with glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech). The beads were washed three times with lysis buffer. The precipitants were analysed by immunoblotting with anti-Rac antibody to detect GTP-form Rac.

Pulse-chase assay

HEK293T cells expressing the indicated plasmids were metabolically labelled with 100 μCi [35S]methionine for 1 hour at 37°C, washed with phosphate buffered saline (PBS), and cultured in DMEM for 0-24 hours. Cell lysates were subjected to immunoprecipitation with anti-Flag tag mAb, the precipitates were separated by SDS-PAGE, and signals were analysed using the BAS2000 image analyser (Fuji Film).

In vivo ubiquitylation assay

HEK293T cells were transfected with the indicated plasmids. After 12-36 hours, cells were further cultured in the absence or presence of 10 μM proteasome inhibitor MG-132 for 12 hours. The cells were lysed with 1% Tx-100 lysis buffer (see above). Cell lysates were immunoprecipitated with anti-Dock180 antibody (H4), anti-Flag tag antibody, or mouse normal IgG (IgG), and immunoblotted with anti-HA tag and anti-Dock180 antibodies. For the denaturing condition, the immunoprecipitates were incubated with 8 M urea buffer containing 20 mM Tris-HCl (pH 7.4) and 8 M urea for 1 minute, then diluted with 1% Tx-100 lysis buffer and immunoprecipitated again with the anti-Dock180 antibody (urea-reversal immunoprecipitation).

Confocal laser scanning microscopical study

Cos-7 and HEK293T cells expressing the indicated plasmids were fixed with 3% paraformaldehyde for 15 minutes and permeabilised with 0.1% Triton X-100 containing PBS for 4 minutes at room temperature (RT). The cells were washed and incubated with 1% BSA containing PBS, and then with only anti-HA tag antibody, or both anti-HA tag and anti-Dock180 (H70) antibodies at 4°C overnight. The cells were next incubated with Alexa Fluor 488-conjugated anti-mouse immunoglobulin and/or Alexa Fluor 594-conjugated anti-rabbit immunoglobulin antibodies for 1 hour at RT with a light shield protecting the samples from photobleaching. For the negative control, cells were processed the same way but without primary antibody. The cells were observed using a confocal laser-scanning microscope equipped with a computer (MRC-1024; Bio-Rad Microscience Division).

Re-plating assay

HEK293T cells transfected with the indicated plasmids were incubated for 36-48 hours and harvested in 0.25% trypsin-EDTA solution (Sigma). Cells were then put into suspension in Opti-MEM (Invitrogen) containing 1 mg of soybean trypsin inhibitor (suitable for neutralisation of 2.5 mg of trypsin). The cells were held in suspension for 3 hours at 37°C. Suspended cells were then distributed onto cell culture dishes pre-coated with fibronectin (Iwaki) and the cells incubated at 37°C for the indicated times. Cells were rinsed in cold PBS prior to protein extraction.

Acknowledgments

We thank M. Matsuda (Osaka University, Japan) and Kodi S. Ravichandran (University of Virginia, VA) for plasmids and Tadaki Suzuki (Hokkaido University, Japan) for useful discussion. This study was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and from the Ministry of Health, Labor and Welfare, Japan, and by the Yasuda Medical Research Foundation.

Footnotes

  • * Present address: Department of Molecular Pathobiology and 21st Century COE Program for Zoonosis Control, Hokkaido University Research Center for Zoonosis Control, Sapporo 060-8638, Japan

  • Accepted November 15, 2005.

References

View Abstract